-------
The relationship between kFieid and mass of compactor is shown in Fig. 4-27. Linear
regression showed that the data are very scattered (r2 = 0.00), and that the regression
line is nearly horizontal. Thus, there is no relationship between mass of compactor and
kpieid-  This observation comes as a surprise because laboratory tests have consistently
shown that the energy of compaction of the soil is a very significant variable, with a
strong trend of decreasing hydraulic conductivity with increasing compactive energy
(e.g., as discussed in Daniel and Koerner, 1995). However, the compactive energy is a
function of the mass of the compactor, the number of passes, and thickness of lifts. The
relationship between number of passes of the compactor per lift and  the mass of the
compactor for the CCLs in the database is shown in Fig. 4-28.  One would expect that
perhaps the heavier the compactor, the fewer the number of passes  of the compactor
on the assumption that fewer passes would be required to achieve the  required dry
density for a heavier compactor. However, Fig. 4-28 shows large scatter and practically
no trend - the data do not suggest that heavier compactors are typically associated with
fewer passes of the compactor over a given lift of soil.
                 1E-06
-0)
E


^   1E-07
"-^
o

T3

O
O
O
                 1 E-08
             I
             ^
             0
                 1E-09
                                           p
                                           00
                                            R

             o

             o
                         Linear Regression,
                         r2 = 0.00
                                                           o
                                                           0
                                                           o
                                                           o
                                                           o
10000       20000        30000

   Mass of Compactor (kg)
                                                                  40000
Figure 4-27. Field hydraulic conductivity versus mass of compactor.

The relationship between kFieid and number of passes of the compactor per lift is shown
in Fig. 4-29.  The expected trend is decreasing hydraulic conductivity with increasing
number of passes, but the data do not support this expectation.  The data are very
scattered, and there are very few data points for more than 20 passes of the compactor
                                      4-36

-------
per lift. As discussed in 4.1.5, the data do indicate that percentage of (w, yd) points lying
on or above the line of optimums (P0) is a critical parameter. The relationship between
P0 and mass of compactor and number of passes of the compactor is shown in Figs.
4-30 and 4-31, respectively. There is a trend for increasing P0 with increasing mass of
the compactor (Fig. 4-30), but the scatter is large.  Similarly, the relationship between P0
and number of passes of the compactor is weak (Fig. 4-31). The trend line in Fig. 4-31
is significantly influenced by a single point at a very large number of passes (80) for Site
65. Despite the very large number of passes, P0 was zero, probably as a result of the
low water content of the soil. Compactor mass and number of passes alone do not
capture all of the critical factors that influence  P0.
            o
            "o
            CD
            Q.
            E
            o
            O
            CD
            w
            V)
            CD
            CL
            0
            .0
            E
               100
                80
60
40
                20
O       Linear Regression,
        r2 = 0.00
                  0         10000       20000       30000       40000

                                Mass of Compactor (kg)

Figure 4-28.  Number of passes of compactor per lift of soil versus mass of
             compactor.
                                      4-37

-------
              1E-06
              1E-07
          "o
          3
          T3
          c
          o
          O
          .o
          T3
          >s
          X

          ^

          
-------
                  100
                             20        40       60       80

                           Number of Passes of Compactor per Lift
100
Figure 4-31.   Percent of (w, yd) points lying on or above the line of optimums (P0)
              versus number of passes of compactor per lift.
4.1.7 Thickness of Liner
The relationship between kFieid and the thickness of the clay liner is shown in Fig. 4-32.
Although the scatter is large (r2=0.13), there appears to be a trend for decreasing
hydraulic conductivity with increasing thickness.  As shown in Fig. 4-33, there is a
relationship between the probability that the kFieid will be < 1  x 10~7 cm/s and the
thickness of the liner.  It appears from the data that liners with a thickness of 1  m or
more have a significantly better chance of attaining a kFieid < 1 x 10~7 cm/s.
There are 21 CCLs in the database with thicknesses > 1 m. These 21 sites are not
uniformly distributed in terms of geography or data type. The thickest liners in the U.S.
tend to be constructed in Wisconsin, whose state regulations require CCLs with
thicknesses of typically > 1.2 m.  Clays in Wisconsin tend to be very wet.  Also, of the
89 CCLs in the database, kFieid was obtained by lysimeters at 8 sites  and by SDRIs at
81 sites.  Of the 8 in-place liners where kpieid was obtained by lysimeter,  7 of these
CCLs had thicknesses > 1 m.
                                      4-39
-------
         1E-06
         1E-07
    o
    O
    .o
    "5
    SE
    T3
    >,
    X
    ^
    CD
1E-08
         1E-09
                                                   Linear Regression,
    0.25      0.5      0.75       1       1.25      1.5

                 Thickness of Clay Liner (m)
                                                                 1.75
Figure 4-32. Field hydraulic conductivity versus thickness of clay liner.
Thus, the tendency seen in Fig. 4-33 for lower kFieid for liners with thicknesses > 1 m
may simply reflect the fact that the thick liners were wetter than average or were
subjected to greater effective overburden stress.  Construction with a high percentage
of data points on or above the line of optimums appears to be more important than
thickness, at least for liners with thicknesses in the range of 0.6 to 1.5 m.

Benson and Daniel (1994) found that kFieid decreases with increasing thickness of the
liner but that little benefit is gained by increasing the thickness beyond 0.6 to 0.9 m. In
the U.S., the minimum thickness of CCLs is typically 0.6 to 0.9 m, although, as
mentioned earlier, some states require thicker liners. The database does indicate a
trend of decreasing kFieid with increasing thickness (Fig. 4-33), but for liners with
thickness of 0.6 to 0.9 m  or more, construction variables (especially P0) appear to be
more important than thickness.  There may, however, be justification for increasing the
thickness to 1 m or more for critical facilities or for situations in which added redundancy
is desired.
                                       4-40
-------
             100
              80
        VI     60
              40
     CL -o
        >,
        X
              20
                   <0.50   0.51 to 0.75 0.76 to 1.00 1.01 to 1.25  1.25 to 1.53

                             Thickness of Clay Liner (m)

Figure 4-33.  Percentage of clay liners with a hydraulic conductivity < 1 x 10~7
             cm/s versus thickness of the CCL.
4.1.8 Field Hydraulic Conductivity Testing Method
Information was collected on the hydraulic conductivity measured with the TSB test,
which is experiencing expanded use for kFieid measurements on test pads because of
the lower cost, faster testing time, and greater information on variability of kFieid provided
by multiple TSB tests compared to a single SDRI test.  As shown in Fig. 4-34, the kFieid
from the TSB test correlates well with  that of the SDRI test. However, for the four test
pads in Fig. 4-34 for which the SDRI test  indicated that the test pad failed to meet the k
< 1  x 10~7 cm/s criterion, the TSB test  indicated that the test pads had met the hydraulic
conductivity criterion, thus providing four false positives (passing hydraulic conductivity's
when, based on SDRI tests, the hydraulic conductivity was not satisfactory).  Also, in
one case,  the TSB test indicated that kFieid was > 1 x 10~7 cm/s, but the SDRI test
indicated it was < 1 x 10~7 cm/s, thus providing a false negative.  It should be
emphasized that both the SDRI test and two-stage borehole test have limitations, and
the natural range of scatter in hydraulic conductivity and the testing methods may
account for much of the differences noted.
                                      4-41
-------
                1E-06

          o .c
          Z!   1 x 10~7 cm/s) in 7  of the 8 test pads.  Thus, the database
supports the use of laboratory tests on 300-mm-diameter samples from the field as an
alternative to SDRI tests.

4.1.9 Case Histories
This section focuses on an analysis of test pads that failed to achieve a hydraulic
conductivity of 1  x 10~7 cm/s or less.  The reasons why the test pads failed to meet
design objectives are identified and compared.
                                       4-42
-------
             1E-05
             1E-06
      31
      -26
      03 O
      .
      05
             1E-08
             1E-09
                1E-09        1E-08        1E-07        1E-06        1E-05

                    Field Hydraulic Conductivity (cm/s) from SDRI Test

Figure 4-35.   Laboratory-measured hydraulic conductivity on 300-mm-diameter
              samples versus field hydraulic conductivity from the sealed double
              ring infiltrometer (SDRI) test.
4.1.9.1  Test Pads at Sites 26 and 27
These test pads were constructed for the purpose of verifying that the hydraulic
conductivity of two compacted glacial tills would be < 1 x 10~7 cm/s.  The specifications
for both test pads were the same:  (1) compact wet-of-optimum, as determined by
modified Proctor, and (2) compact to a minimum dry density equal to 90% of the
maximum dry density based on modified Proctor.

For test pad 26, 75% of field water content and 100% of dry density measurements
were met the specifications.  However, only 17% of the field water content and dry
density measurements fell on or above the line of optimums,  i.e. P0 = 17%.  Thus, the
low P0 value indicates that a large portion of the test pad was compacted dry of the line
of optimums despite the fact that most of the water content-density data met the project
specifications.
                                     4-43
-------
The test pad at Site 27 had a similar problem.  At Site 27, 100% of the field data met the
density criterion and 90% of the field data met the water content criterion. The P0 value
was only 6%, however.  Compaction below the line of optimums was the problem  at
both of these sites.

Analysis of the case histories in the database shows that compaction dry-of-optimum is
the most common cause for field hydraulic conductivity values > 1 x 10~7 cm/s. The
fundamental flaw of many compaction criteria is that they allow compaction to occur dry
of the line of optimums.  However, the recommended procedure, as described by  Daniel
and Benson  (1990) and as illustrated in Fig. 4-16, will prevent the development of a
specification that allows excessive compaction dry of the line of optimums.

4.19.2  Test Pad at Site 21
The SDRI test performed at this site achieved a kFieid of 2.5 x 10~7 cm/s, whereas the
ki_ab was 3.0  x 10~7 cm/s.  The soil was a western loess that contained 90% fines and
36% clay.  The compaction criteria for the test pad were as follows:  the as-compacted
water content must be wet of standard  Proctor optimum and the dry density must  be
greater than  95% of the standard Proctor maximum. The field water content and dry
density measurements indicate that P0 at this site was 80%.

Laboratory compaction curves  for a reduced Proctor sample indicate a four-order-of-
magnitude decrease in hydraulic conductivity between water contents of 15% and 21%.
The reduced Proctor compaction test uses the same equipment as the standard Proctor
compaction test but only fifteen blows from  the hammer are applied to each lift. The
reduced Proctor test was not performed until after the SDRI test failed to assist with
identification of the causes for the high SDRI result.

Maximum reduced Proctor density was approximately 16.5 kN/m3 and maximum
standard Proctor density was 17.2 kN/m3. The density specification allowed compaction
to occur within the reduced Proctor density  range (i.e. 95% of 17.2 = 16.3 kN/m3). Field
density measurements indicate that although all of the field-measured dry density
values met the density specification, approximately 40% of the dry density values  were
less than or equal to the maximum density from reduced Proctor tests.  Therefore, an
inadequate density specification was the major reason this site had a kFieid greater than
1 x 10~7cm/s.

From  the investigation of this test pad, the usefulness of the reduced Proctor
compaction test is evident.  It is especially important to use the reduced Proctor test as
a lower bound when standard Proctor density is a compaction  criterion. In this case, a
density lower than standard Proctor was allowed, but the hydraulic conductivity of the
soil was never determined at that density. The soil at this site was very sensitive to
changes in compactive energy. If the soil had been tested at a density below standard
Proctor, the compaction criteria most likely would have been changed.
                                     4-44
-------
Another problem with this test pad, not evident in the database information, is the lower
boundary condition. The SDRI test requires a known lower boundary condition in order
to calculate kFieid accurately if the wetting front from the infiltrating water penetrates to
the bottom of the test pad. Sometimes a sand drainage layer is placed beneath the
compacted clay test pad to provide free drainage at the lower boundary of the test pad.
At this site, no sand drainage layer was installed and the subbase material had a
hydraulic conductivity of approximately 1x10" cm/s. As a result, after thirty days of
testing, the hydraulic conductivity was 2.5 x 10  cm/s, until it abruptly decreased to 1 x
10" cm/s a short period  later. It is believed that after thirty days the wetting front
reached the subbase, and then the hydraulic conductivity was misinterpreted as 1 x 10"
cm/s. Therefore, the drainage layer can provide a critical boundary condition for the
SDRI test.

4.1.9.3  Test Pads at Sites 55-63
Nine test pads were constructed at one site using four different clays and various water
contents. Three of the four soils were suitable for use as a  liner with field hydraulic
conductivity < 1 x 10"7 cm/s.  For these sites, the test pads  consistently showed that
whether or not the soil met the hydraulic conductivity criterion of 1 x 10~7 cm/s or less
was dependent almost entirely upon  the water content of the soil. When the soil was
compacted near optimum water content, the soil failed to meet the hydraulic
conductivity criterion. When the soil was typically 2 to 4% wet of optimum, the objective
was met.

One of the soils in this series of field  tests, however, was simply unsuitable.  The soil
was compacted 2% to 6% wet of optimum at Sites 61 and 62, and despite the high
water content the soil still failed to achieve a hydraulic conductivity < 1 x 10~7 cm/s.  All
of the soils were kaolin clays obtained from commercial clay pits.  The clay that was not
suitable was from a different geologic formation than the other kaolin clays that were
suitable.

One lesson learned from these tests is that, in retrospect, a comprehensive program of
laboratory hydraulic conductivity testing prior to field testing would have been desirable
for two reasons:  (1) to define the appropriate range of water content for field
compaction,  and (2) to verify that the soil was suitable.  The soil at Sites 61 and 62
probably would have been found to be unsuitable had it first been tested in the
laboratory.

4.1.9.4  Test Pads at Sites 64 and 65
Sites 64 and 65 provide  an excellent example of the futility of conventional compaction
specifications. Test pads were constructed of the same material but with a minimum
dry unit weight of 95% of standard Proctor (Site 64) or 91 % of modified Proctor (Site
65). Both test pads were compacted several percentage points wet of optimum.
                                      4-45
-------
Neither achieved the desired kFieid of 1 x 10~7 cm/s. Site 65 is particularly interesting
because it was compacted 3.7% wet of optimum to a density greater than 91% of the
maximum from modified Proctor using an astonishing 80 passes per lift of a heavy,
footed compactor. Despite all this, the liner failed to achieve the desired kFieid
apparently because  not a single (w, yd) point was on or above the line of optimums.

4.1.9.5 Test Pads at Sites 43 and 44
Although there was no statistically significant correlation between field hydraulic
conductivity and either weight of compactor or number of passes of the compactor,
there were examples in which the number of passes of the compactor was the only
variable between two test pads.  For example, Sites 43 and 44 represent two test pads
that were constructed from the same soil at essentially the same water content, and
compacted with the same compactor.  The only variable was number of passes of the
compactor: 16 passes for Site 43 and 8 passes for Site 44. The kFieid for Sites 43 and
44 were 7 x 10~8 and 2 x 10~7 cm/s, respectively, illustrating that increasing the
compactive effort can have a significant impact on field hydraulic conductivity. The soil
compacted with 16 passes of the compactor achieved 63% of the (w, yd) points on or
above the  line of optimums, whereas P0 was only 47% when 8 passes were used.

4.1.10 Practical Findings from Database
CCLs are almost always constructed with the objective of achieving a hydraulic
conductivity of 1 x 10~7 cm/s or less. All of the 89 liners and test pads in this database
were constructed for the purpose of demonstrating that the kFieid would meet this
requirement.  Despite all that has been written and learned about CCLs in the past 15
years, the  hydraulic  conductivity objective of kFieid < 1 x 10~7 cm/s was not met at  more
than one-fourth (26%) of the sites in the database. Why such poor success?

The soil was unsuitable at a few sites. It appears that in all cases involving unsuitable
soils, test pads were built without the benefit of a comprehensive laboratory testing
program prior to construction. No simple way of identifying unsuitable soils (based on
plasticity information or other index properties) was identified. The database reinforces
the recommendation that kLab tests be performed on representative samples of soil prior
to construction, e.g., following the procedures recommended by Daniel and Benson
(1990).

Perhaps the  single biggest problem identified from the database is failure to recognize
that conventional specification of water content and dry unit weight based on a minimum
percent compaction  often leads to difficulty. The problem is that this procedure does
not guarantee that any of the (w, yd) points will lie on or above the line of optimums.
Despite widespread  publication of procedures that will avoid this problem (e.g., Daniel
and Benson, 1990; Daniel and Koerner,  1995), many design professionals and
specification writers  continue to repeat the mistake. The type of specification that is not
                                      4-46
-------
generally recommended (but which is still commonly used) is shown in Fig. 4-36A; the
more appropriate and recommended approach is shown in Fig. 4-36B.

A conclusion from analysis of the database is that P0 is the single most important CQA
parameter. The definition of P0 is summarized in Fig. 4-24: P0 is the percent of water
content-dry density (w, yd) points lying on or above the line of optimums.  The line of
optimums is the locus of peaks of compaction curves developed  employing different
compaction energies (Fig. 4-23). The key to success in achieving a kFieid < 1 x 10~7 cm/s
is to ensure that a high percentage (the data indicate a minimum of 70% to 80%) of the
field-measured (w, yd) points lie on or above the line of optimums, i.e., in the shaded
zone shown in Fig. 4-36B.  It should be emphasized that it is practical to compact soil in
the shaded zone in Fig. 4-36B, particularly in the shaded area  above the line of
optimums but below the compaction curve (compaction in  this  zone could be achieved
with a smaller compactive effort than used to develop the compaction curve indicated in
the figure). The data in Fig. 4-25 provides confirmation that it is possible and practical
to achieve a large value for P0, i.e., to compact within the shaded zone shown in Fig. 4-
36B.

Although this analysis has focused on hydraulic conductivity, the authors emphasize
that other factors (such as bearing capacity, internal shear strength of a clay liner, and
interfacial shear strength with geosynthetics) are equally important considerations. The
engineer must give proper consideration to these factors and avoid the temptation to
add too much water to the clay (in order to drive down hydraulic conductivity), at the
expense of compromising other critical engineering properties  of the clay liner. As the
value of P0 is increased,  the shear strength of the soil tends to decrease.  The designer
must ensure that all  criteria, and not just hydraulic conductivity, are satisfied.

4.2  Soil-Bentonite Mixtures

Even though the most common type of CCL is one that is  made from natural soils that
contain a significant quantity of clay,  if the soils found near the waste disposal facility
are not sufficiently clayey to be suitable for direct use as a liner material, a common
alternative is to blend natural soils available on or near a site with sodium  bentonite.
Soil-bentonite liners are discussed by Daniel and Koerner (1995).

4.2.1  Database
A database of 12 test pads that had bentonite added to the natural clay soils was
developed. The database is summarized in Tables C-5 through C-8 in Appendix C. All
test pads were constructed for the purpose of demonstrating that kFieid < 1 x 10~7  cm/s.
                                      4-47
-------
D)
(U

•^

D


Q
                                        Specified
                                      Water Content
                                         Range
                      Line of
                      Optimums
                                 Zero Air
                                 Voids Curve
                           Laboraotry   Minimum yd Based
                           Compaction     ...  .
                           o..~.~       on Minimum
                                      Percent Compaction
     Curve
                                    Water Content

                        (A) Typical Type of Specification Currently Used
               D)
Line of
Optimums
                      Minimum Water
                      Content Determined
                      from Lab. Hydraulic
                      Conductivity Tests
                                  Zero Air
                                  Voids Curve
                                   Minimum Yd Based
                                   on Minimum
                                   Percent Compaction
                                    Water Content

                         (B) Recommended Compaction Specification


Figure 4-36.  Water content-density specifications indicating: (A) the traditional
             (but not recommended) type of specification, and (B) the
             recommended type of specification emphasizing compaction to
             water content-density values on or above the line of optimums.
                                      4-48
-------
In all cases, the large-scale hydraulic conductivity tests were the SDRI test. The SDRI
test results are referred to as kFieid- Additionally, the results of kLab tests obtained on thin-
walled sampling tubes were documented at several sites. Finally, one large block
sample test was performed for kLab-

There is a basic limitation of the database associated with that fact that only 12 test
pads comprise the database.  Further, the information is not complete for all 12 tests
pads. Although this is the most complete database of its type assembled to date, there
remain relatively few well-documented cases in which the large-scale performance of
soil-bentonite clay liners is documented.

4.2.2  Hydraulic  Conductivity Results
kpieid covered a comparatively narrow range of 2 x 10~9 cm/s to 1 x 10~7 cm/s.  All liners
in the database met the objective of kFieid < 1 x 10~7 cm/s. The geometric mean kFieid
was 1.9x 10~8cm/s.

The most important parameter in soil-bentonite mixtures is the amount of bentonite
added to the mixture.  kFieid is plotted as a function of percent bentonite in Fig. 4-37. As
expected, there is a tendency for kFieid to decrease when the amount of bentonite is
increased.  However, the scatter in the data is significant (r2 = 0.28), and there are
relatively few data points in the database.
                1E-06
>
"o
T3
O
O
O
"5

-I
X

15
                1 E-°7
                1E-08
                1E-09
                                          Linear Regression,
                                          ft = 0.28
                             8        10        12        14

                                    Percent Bentonite
                                                     16
Figure 4-37.  Field hydraulic conductivity versus percent bentonite in soil.
                                       4-49
-------
The results of kLab tests are compared to the results of kFieid tests in Fig. 4-38. There is
good agreement between kLab and kFieid- This suggests that all liners were reasonably
homogeneous and free of large-scale features that would tend to lead to large
differences between the results of kLab and kFieid tests.

It was found that for natural soil liner materials, the percent wet of optimum was a
significant variable.  For soil-bentonite liners, this was not found to be the case (Fig.
4-39).  Instead, percent compaction (Fig. 4-40) was found to be more significant.
Experience in the laboratory has shown that water content is much less important for
soil-bentonite admixtures than for natural soil materials, perhaps because of the high
swelling that occurs in bentonite regardless of the initial water  content.  It does appear,
however, that the more compact the soil (i.e., the higher the percent compaction), the
lower the kFieid.

The masses of the compactors used to compact the liners in the database were nearly
all identical. However, the number of passes of the compactor did vary.  kFieid was
found to slightly increase with number of passes within the very small range of number
of passes used to construct the liners that comprise the database (Fig. 4-41).
Unfortunately,  the projects that comprise the database were not sufficiently well
documented to permit determination of P0.
                 1E-06
              >
             "§   1E-07
             T3
              C
              O
             O
              o
             "5
              ro
             T3
             Is  1E-08
             o
             .Q
             ro
                 1E-09
                    1E-09          1E-08          1E-07

                              Field Hydraulic Conductivity (cm/s)
1E-06
Figure 4-38.  Laboratory-measured hydraulic conductivity vs. field-measured
             hydraulic conductivity for soil-bentonite liners.
                                      4-50
-------
i ^-uu
E
!> 1 E-°7
"o
T3
O
O
O
^
CD
•£ 1 E-08
I
T3
CD
-ip.nQ

o


o o o
° 0
i o
f
-
I •
Linear Regression,
r2 = 0.00
O
                     -1012345
                              Percentage Points Wet of Optimum

Figure 4-39.   Field hydraulic conductivity versus percentage points wet of
              optimum for soil-bentonite liners.
                  1E-06
~    1E-07
"G
T3
O
O
O
"5
.§    1 E-08
I
"CD
                  1E-09
                                            Linear Regression,
                     94         96         98         100        102
                                    Percent Compaction

Figure 4-40.  Field hydraulic conductivity vs. percent compaction for soil-
             bentonite liners.
                                       4-51
-------
          E
          o
          O
          O
          o
          
-------
 4.3.1  Omega Hills Final Cover Test Plots
 The first detailed information to be presented on the performance of CCLs in landfill
 covers was described by Montgomery and Parsons (1989).  The study involved the
 construction of three large test pads on the top of a closed municipal solid waste landfill,
 the Omega Hills landfill, which is located approximately 30 km northwest of Milwaukee,
 Wisconsin. The test plots were constructed to evaluate the performance of alternative
 final cover designs.

 The cross sections of the three test plots are shown in Fig. 4-42.  Test plot 1, consisting
 of 150 mm of topsoil overlying 1.2 m of CCL, represented the existing final cover system
 design at the time that the study was initiated. Test plot 2 involved the same thickness
 of CCL, but a thicker topsoil layer that was intended to promote better vegetative growth
 and thereby enhance evapotranspiration. Test plot 3 involved the use of a layer of
 coarse-grained  soil (sand) sandwiched between two CCLs.  The idea for the third plot
 was to take advantage of the so-called capillary barrier effect in which the coarse-
 grained soil (sand) remains unsaturated and thereby serves as a  barrier to downward
 infiltration of water.  With test  plot 3, the intention was for the sand layer to promote
 retention of water in the upper CCL, where the water could be returned to the
 atmosphere via evapotranspiration. The use of this alternative design is consistent with
 the research nature of these test plots. All test plots were constructed on 31-1:1 V
 sideslopes of the actual landfill surface.
         Test Plot 1
Test Plot 2
Test Plot 3
150 mm
       i
                      i
  1.2m
                     450 mm
                       1.2m
L
F
L
F


150 mm
j
600 mm
i
300 mm ;
j
600 mm
i







Topsoil
Compacted
Clay Liner
Sand
Compacted
Clay Liner
 Figure 4-42.  Cross-sectional view of test plot arrangement at Omega Hills
               landfill (after Montgomery and Parsons, 1989).
 The CCL material consisted of CL soil with a high silt content.  The soil was placed and
 compacted in 150-mm-thick lifts to a hydraulic conductivity < 1 x 10~7 cm/s, based on
 laboratory hydraulic conductivity tests on "undisturbed" samples of the compacted soil.
 The topsoil was an uncompacted clay loam to silty clay loam.  The intermediate sand in
                                       4-53
-------
test plot 3 was a clean, washed, medium sand. The topsoil was seeded with a mixture
of grasses.

The test plots contain two principal data collection systems. The first was a lysimeter
located beneath the test plot to collect water that percolated through the cover soils and
permit quantification of the rate of percolation. Figure 4-43 shows the plan location of
the lysimeter system, and Fig. 4-44 shows the location in profile.  The lysimeter
consisted, from top to bottom, of a GT filter, a GC drainage layer, and a GM. The
second data collection system was designed  to collect and measure surface runoff
(Figs. 4-43 and 4-44).

The test plots were constructed in 1986.  Data collection and analysis started  in
September, 1986. Measurements were obtained of precipitation, runoff, percolation,
and other parameters such as temperature.  Soil moisture content was monitored with
neutron access probes.

The 12-month period September 1986 through August 1987 was near normal. The
period of September 1987 through August 1988 was dominated by a severe drought in
1988. The summer months in 1988 were characterized by substantially below-average
rainfall and temperatures that averaged 6°C above normal.  The drought reduced the
cover vegetation to a dry, dormant state, and cracking of the surface of the cover soils
was obvious.  The third and final year of data collection saw a return to moist conditions.

At the end of three years, test pits were excavated in each test plot, outside the area of
the lysimeters. The test pits measured 3 m in length, 1.2 m in width, and 2 m  in depth.
A summary of data collected  is presented in Table 4-3.  The key parameter is  the
quantity of percolation, i.e., rate of flow of water into the lysimeter. In test plots 1  and 2,
the percolation in the first year was 2 to 7 mm/year (6 x 10"9 to 2 x 10"8 cm/s,
respectively).  However, by the third year, these values had increased to a range of 56
to 98 mm/year (2 x 10~7 and 3 x 10~7 cm/s, respectively). The test pits showed that the
CCLs in test plots 1 and 2 were in a similar condition after three years:

  • the upper 200 to 250 mm of CCL was weathered and blocky (probably from
    desiccation and/or freeze-thaw;
  • cracks 6 to 12 mm wide extended 0.9 to 1 m into the CCL;
  • roots penetrated 200 to 250 mm into the CCL in a continuous mat, and some roots
    extended into crack planes as deep as 750 mm into the CCL; and
  • the base of the CCL appeared to be undamaged.
                                      4-54
-------
                                15m
           18m
                              6 m
                  12m
                         :::::^f:^:]
Percolation
Collection
Lysi meter
(Beneath
Test Plot)
                                                         ,Test Plot
    Surface Runoff
   ^Isolation Barrier
   "To Quantify
    Surface Runoff
                                                                3H:1V
                                                                Slope
                                     Collection Pipe for Percolation
                                     Collection Lysimeter (Located
                                     Beneath Test Plot)
                                        Percolation Collection Tank
            Runoff
            Collection
            Tanks
Figure 4-43.   Plan view of test plot arrangement at Omega Hills landfill (after
              Montgomery and Parsons, 1989).
The drought conditions in the second year of the study period apparently caused severe
desiccation of the CCL, which led to the significantly increased hydraulic conductivity in
subsequent years. Although the CCL may have initially had a hydraulic conductivity of
1 x 10~7 cm/s or less, after three years, the desiccation damage caused the CCLs in test
plots to no longer have this low level of hydraulic conductivity.

Test plot 3 was designed with the intention of maintaining moisture in the upper CCL.
The percolation rate through test pad 3 remained more consistent and was found to
range from  22 to 41 mm/year (7 x 10"8  to 1.3 x 10"7 cm/s).  At the end of the three-year
study period, the upper 200 to 250 mm of the uppermost CCL was weathered and
blocky, and cracks extended through the entire thickness of the uppermost CCL.
Cracking of the uppermost CCL allowed significant amounts of water to enter the sand
drainage layer.  Discharge of water from the sand layer was found to occur within hours
of the start of precipitation events, suggesting rapid transmission of water through the
                                       4-55
-------
upper CCL due to flow through cracks. Moisture in the sand drainage layer probably
helped to protect the underlying CCL from damage. The multi-component cap in test
plot 3 did not function as anticipated.  It was expected that the sand drainage layer
would help the overlying CCL retain moisture, but the uppermost CCL quickly cracked.
Runoff Collection
Pipe
                                                    Geomembrane
                                      Percolation Collection
                                      Lysimeter
                  Percolation Collection Pipe
Figure 4-44.
Cross section of underdrain system at Omega Hills landfill (after
Montgomery and Parsons, 1989).
Table 4-3.     Summary of information concerning performance of field test plots
              at Omega Hills landfill (data from Corser and Cranston, 1991).
Test Plot
1
2
3
Year
1986-87
1987-88
1988-89
1986-87
1987-88
1988-89
1986-87
1987-88
1988-89
Precipitation
(mm)
896
579
823
896
579
823
896
579
823
Runoff
(mm)
180
38
56
109
38
51
97
38
66
Percolation
(mm)
2
5
56
7
30
98
40
22
41
                                     4-56
-------
The principal lesson learned from the Omega Hills study was that in a fairly short period
of time (3 years), CCLs overlain by 150 to 450 mm of topsoil are subject to major
desiccation, cracking, and increases in hydraulic conductivity.  The CCL was not
"survivable" with a hydraulic conductivity of 1 x 10~7 cm/s or less under these conditions.

4.3.2  Test Plots in Kettleman City, California
Corser and Cranston (1991) and Corser et al. (1992) describe test plots constructed at
a waste disposal facility located in Kettleman City, California. Three test plots were
constructed as shown in Fig. 4-45.  Test plot 1 consisted of a 900-mm-thick CCL
overlain by an exposed HOPE GM. Test plot 2 consisted of the same profile as test plot
1, except that 600 mm of topsoil covered the GM. Test plot 3 contained 450 mm of
topsoil covering the CCL, with no GM covering the CCL. A portion of the test plots was
flat, and a portion that sloped at 3H:1V. The test plots were constructed to study the
factors that influence desiccation in a CCL placed in a final cover system profile.

The CCL was a high-plasticity clay that was expected to be used to construct final cover
systems for approximately 30 ha of landfill at the site.  The clay had an  average liquid
limit of 66% and plasticity index of 48%.  The instrumentation consisted of thermistors to
monitor temperature in the soil and CCL, and tensiometers to measure  soil suction.
Corser and Cranston (1991) summarize the first 6 months of data collection.  At the end
of the six month period, the surfaces of the CCLs were exposed over an area of 1.5 m
by 1.5 m to observe and document cracking patterns.

Test plot 1 did not represent a final cover situation but is representative of a bottom liner
with an exposed HOPE GM during the construction or operations phase. The clay
exhibited some drying and cracking in areas where the HOPE was not in contact with
the CCL. In other areas where the HOPE was in contact with the CCL,  the moisture
content of the CCL at the surface increased.  It appears that the high temperature of the
exposed HOPE GM  caused  heating and drying of the underlying CCL.  In some areas
(e.g.,  around wrinkles in the  GM), moisture could migrate away via vapor transport.  In
other  areas, the moisture could condense during cooler periods, causing moistening of
the soil. In any case, there clearly was desiccation of the CCL beneath some portions
of the exposed GM.

Test plot 3 did not perform well during the summer season.  The CCL dried, and
cracking was observed at the surface when a test pit was excavated in  the fall.  In
contrast, was no evidence of drying or cracking of the CCL at test plot 2.

Although the test plots were  observed for only six months, significant deterioration of the
CCL was observed in test plots 1 and 3.  Only test plot 2, in which the CCL was covered
with a GM and  450 mm of top soil, performed well. The observations from Kettleman
City are consistent with those of Omega Hills and suggest that perhaps the only
practical way to protect a CCL from desiccation damage in typical final cover system
                                      4-57
-------
cross sections is to incorporate a GM and sufficiently thick cover soil over the GM/CCL
composite barrier.
                        Test Plot 1
          1.5 mm HOPE
          Geomembrane"
              ICompacted Clay
  900 mm
                        Test Plot 2
   1.5mm HOPE
   Geomembrane
              ICompacted Clay
                      Topsoil
  600 mm

  900 mm
                        Test Plot 3
            Topsoil
              ICompacted Ciay|
i
  450 mm

  900 mm
Figure 4-45.  Cross sections of test plots at Kettleman City facility (after Corser
            and Cranston, 1991).
                                 4-58
-------
4.3.3  Test Plots in Hamburg, Germany
Melchior et al. (1994) describe what may be the most extensive test plot program of any
constructed to date involving CCLs. Three test plots were constructed, as shown in Fig.
4-46.  All test plots were constructed on top of an existing MSW landfill.  There were two
sections for each test plot.  The upper section was located on the relatively flat portion
near the top of the landfill and sloped at 4%.  The lower half sloped more steeply at an
inclination of 51-1:1 V (20%).  The test plots were underlain with a lysimeter, much like the
Omega Hills facility (Fig. 4-44).

The CCLs at the Hamburg site were constructed in three lifts, each 200 mm thick.  The
material consisted of 17% clay, 26% silt, 52% sand, and 5% gravel. The principal clay
minerals in the clay fraction were (in decreasing abundance) illite, smectite, and
kaolinite. The liquid limit was 20%, and the plasticity index was 9%. The soil was
compacted 2% wet of optimum at an average degree of compaction of 96%. The CCL
at the Hamburg  site was significantly different from that at Omega  Hills and Kettleman
City. At Omega Hills, the low-plasticity (CL) clay contained a large amount of silt, which
can make the material vulnerable to shrinkage cracking.  The Kettleman City clay was a
highly plastic (CH) clay. At Hamburg, the soil contained more than 50% sand- and
gravel-sized particles and would therefore be classified as a clayey sand (SC). Clayey
sands tend to be less vulnerable to shrinkage cracking than  clays (especially highly
plastic clays) that contain relatively little coarse-grained particles.

The percolation  rates through the CCLs and into the lysimeters are summarized in
Table 4-4. Also shown are the drainage rates in the sand drainage layer that overlies
the CCL. The last column in Table 4-4 expresses the leakage through the CCL as a
percentage of the drainage from the sand drainage layer. The leakage as a function of
drainage is plotted vs. time in Fig. 4-47.

Test plots  1 and 3, which did not have a GM overlying the CCL, underwent a very large
increase in leakage in 1992. The summer of 1992 was extremely dry  in Hamburg, and
the subsequent fall season was very wet.  Excavations made in 1993 confirmed that the
clay liner was cracked. Barely visible fissures were observed between soil aggregates
(around 50 mm in diameter). Plant roots were observed to have reached the upper
parts of the CCLs.  Under the conditions of a hydraulic conductivity of  1 x 10"7 cm/s for
the CCL and a unit hydraulic gradient, the calculated percolation rate through the CCL
is approximately 30 mm/year.  The actual leakage rates through the CCLs at test plots 1
and 3 exceeded 30 mm/year in 1992. The apparent problem was gradual deterioration
of the CCL caused by desiccation during a particularly dry summer.

Test plot 2, on the other hand,  has  maintained a very low leakage rate. This test plot
contains a GM overlying the CCL (Fig. 4-46).
                                      4-59
-------
          Test Plot 1
                      Test Plot 2
                           Test Plot 3
    750
    250
    600
    200
  Topsoil
[Drainage)
I  Sand  III
IllCompactedl
I Clay Liner ||
 ((Drainage!
 I  Sand  II
750
250
   Topsoil
 ([Drainage!
 I  Sand  III
750
250
600
                         200
IllCompactedl
I Clay Liner I
     IDrainagel
     I  Sand  I
                      Thickness of
                      Layer (mm)
                                              400
                                              600
                                     250
 Topsoil
(Drainage!
I!  Sand  I!
                          [Compacted))
                          [ Clay Liner ||
                                           Fine Sand
                           IDrainagel
                           I   Sand  I
      Note: Geotextile Separation and
            Filtration Layers Not Shown.
Figure 4-46.  Cross sections of test plots in Hamburg, Germany (after Melchior et
            a\., 1994).
                                 4-60
-------
Table 4-4.     Summary of information concerning performance of field test plots
              at Hamburg, Germany (data from Melchior et al., 1994).
Test Plot
1
2
3
Year
1988
1989
1990
1991
1992
1988
1989
1990
1991
1992
1988
1989
1990
1991
1992
Drainage
(mm)
371
181
291
184
225
296
155
269
164
311
390
233
321
198
278
CCL Leakage
(mm)
7
8
18
9
103
3
0.6
0.4
0.5
0.8
8
14
31
32
116
Leakage /
Drainage (%)
2
4
5
5
31
1
0.4
0.1
0.3
0.3
2
6
10
16
42
The results from Hamburg are consistent with those from Omega Hills and Kettleman
City, even though the CCL material was very different.  It appears that a CCL placed in
a final cover system without a GM and soil covering the GM is likely to fail to maintain a
hydraulic conductivity < 1 x 10~7 cm/s.  If the CCL is to have a chance of maintaining this
level of hydraulic conductivity for extended periods, it appears that the CCL must be
protected with both a GM and a sufficiently thick layer of cover soil above the GM.

Melchior (1997) describes additional work performed at the Hamburg site in which two
additional test covers were constructed and  monitored. The two additional test covers
both consisted of 300 mm of topsoil underlain by 150 mm of drainage sand, which in
turn was underlain by a GCL. Two different  GT-encased, needlepunched GCLs were
used for the two different test plots. As with  the first three test covers, the two additional
test covers with GCLs were underlain by drainage sand and a GM to collect any water
that percolated through the test cover. Both GCLs performed well for about a year,  with
almost no liquid appearing in the drainage layers beneath the test covers.  However,
about a year after construction (in the fall,  following a dry summer), percolation began to
occur and was closely linked with rainfall events. Peak percolation rates were on the
order of 0.4 mm per hour (about 1 x 10~5 cm/s).  Melchior (1997) states that research on
the causes for high percolation rate is on-going, but indications are that the causes for
the increase in hydraulic conductivity of the GCLs may have been related to: (1)
penetration of the GCL by plant roots; (2) desiccation of the GCL, leading to high initial
seepage rates following major rainfall events; and (3) ion exchange (calcium was
                                      4-61
-------
apparently leached from cover soils, and replacement of sodium in the bentonite with
calcium is expected to cause an increase in hydraulic conductivity).
       CD
       O)
       ^    50
Q

2
3    40
CO
_l
"5
CD
O)
       .2

       I
       CD
       Q.
       to
       CO
       _l
       O
       O
       .c
       O)
       D
       o
       CD
       O)
       CO
       _*:
       co
       CD
      30
     20
      10
                            Test Plot 1

                            Test Plot 2

                            Test Plot 3
      0
      1987
1988
1989
1990

Year
1991
1992
1993
Figure 4-47.   Leakage through CCL as a percentage of the drainage from the
              overlying sand drainage layer plotted vs. time.
4.3.4 Final Covers in Maine
The Maine Bureau of Remediation and Waste Management (1997) reported the results
of field measurements of percolation rates through four CCLs in actual municipal solid
waste landfill covers.  All liners appear to have been constructed using methods of
construction and construction quality assurance practices that are typical of the landfill
industry.

4.3.4.1  Cumberland Site
The 2-ha Cumberland Municipal Solid Waste Landfill was closed in 1992 with a cover
system that consisted of 150 mm of vegetated topsoil that was underlain by 450 mm of
compacted silty clay,  which in turn was underlain by sand-filled trenches that served to
collect gases.  kLab tests were performed during construction and in post-construction
                                      4-62
-------
investigation programs conducted in 1994 and 1996. An SDRI test was performed in
1994.

At the time of construction, the average kLab was 5 x 10~8 cm/s. In the 1994
investigation, kLab was 1 to 2 x 10~7 cm/s. kFieid, measured with the SDRI, was 6 x 10~6
cm/s. It is not certain whether the liner originally had a kFieid < 1 x 10"7 cm/s (since no
field testing was performed at the time of construction).

4.3.4.2  Vassalboro Site
The Vassalboro Municipal Solid Waste Landfill consists of one 1.6 ha site that was
closed in 1990.  The final cover consists of 150 mm of vegetated cover (sludge
amended topsoil) overlying 450 mm of compacted glacial till CCL, which in turn was
underlain by a gas collection layer. kLab tests were performed at the time of construction
(1990) and again in 1994 and 1996. An SDRI test was performed in 1994.

ki_ab at time of construction averaged 2 x 10"7 cm/s.  In 1994, kLab ranged from 9 x 10"7 to
5 x 10~6 cm/s. The kFieid measured by SDRI test in 1994 was 2 x 10~6 cm/s.  It appears
that the hydraulic conductivity of the CCL increased about an order of magnitude from
1990 to 1994.

4.3.4.3  Yarmouth Site
The Yarmouth Municipal Solid Waste Landfill covers 2.5 ha and was closed in 1990
using a cover system consisting of 150 mm of sludge-amended topsoil overlying 450
mm of compacted silty clay, which was underlain by a gas collection  layer.

ki_ab tests performed at the time of construction (1990) indicated an average kLab of 8 x
10~8 cm/s. In a 1994 investigation,  kLab was approximately 3 x 10~7 cm/s, and in 1996,
k|_ab was found to be 2 x 10"6 to 2 x 10"5 cm/s, or about 20 to 100 times larger than  in
1990. kpieid was measured with the SDRI in 1994 and again in 1996. kFieid was 2 x 10~7
cm/s in 1994 and 2 x 10~6  cm/s in 1996. There is a clear trend of increasing hydraulic
conductivity over time, with the magnitude of increase being one to two orders of
magnitude over the six-year study period.

4.3.4.4  Waldoboro Site
The Waldoboro Municipal  Solid Waste Landfill covers 1.6 ha and was closed in 1991
with a cover system consisting of 150 mm of sludge-amended topsoil overlying 450 mm
of compacted silty clay, which in turn was underlain by a gas collection layer.

kLab tests indicated that kLab increased over time from an initial average value of about 5
x 10"8 cm/s (1991) to 1 x 10"6 cm/s  (1993) and to 3 x 10"6 cm/s (1996).  kFieid measured
with SDRI tests was 1 x 10~6 cm/s in 1993 and 4 x 10~6 cm/s in 1996. Thus, the data
indicate that the hydraulic conductivity increased about two orders of magnitude over a
five year period.
                                      4-63
-------
4.3.4.5  Discussion
The observations from these four actual cover systems are consistent with those of the
other sites mentioned previously in this section of the report.  All of the available field
performance data indicate that a CCL overlain by a relatively thin layer of topsoil (150 to
450 mm thick), and without a GM above the CCL, cannot maintain a hydraulic
conductivity of 1 x 10~7 cm/s or less.  From analysis of the condition of the four CCLs at
these sites, it appeared that desiccation was the  most significant factor leading to an
increase in kFieid. Freeze/thaw may also have contributed significantly to damage.
Penetration of plant roots into the CCL was also observed.

4.3.5 Alternative Cover Demonstration at Sandia National Laboratory
A major field demonstration project, initiated  in the mid 1990s, is underway at Sandia
National Laboratories and, although only preliminary data were available at the time of
preparation of this report, the project bears mentioning here.  The project is known as
the Alternative Landfill Cover Demonstration (ALCD), and is a large-scale field test
conducted at  Sandia National Laboratories, located on Kirtland Air Force Base in
Albuquerque, New Mexico.  The climate at the test site is semi-arid. The goal of the
ALCD is to field test, compare, and document the performance of alternative landfill
cover technologies, of various complexities and costs, with emphasis on arid and semi-
arid environments (Dwyer, 1997). The purpose of the ALCD  is to provide information on
cost, construction, and performance, so that  design engineers and regulatory agency
officials will have data on alternatives to conventional cover design.

The test plots are each 13 m wide by 100 m  long (Dwyer, 1997). All covers were
constructed with a 5% slope in all layers. Slope lengths are 50m (the test covers are
crowned at the middle half of the length). The western slopes are maintained and
monitored under natural conditions while a sprinkler system was installed on the eastern
slopes to facilitate stress testing of the covers. Two conventional covers and four
alternative covers comprise the six test covers, with cross sections as follows:

   1.   Baseline Test Cover 1 is a "RCRA Subtitle D" conventional cover, consisting of
       150 mm of topsoil underlain by 450 mm of compacted "barrier layer soil" with a
       maximum hydraulic conductivity of 1  x 10~5 cm/s (actual hydraulic conductivity
       measured on laboratory samples recovered from the  constructed barrier layer
       were in the range of 5 x 10~7 cm/s to 6 x 10~6 cm/s, and an in situ hydraulic
       conductivity test yielded a hydraulic conductivity of 5 x 10~7 cm/s).
   2.   Baseline Test Cover 2 is a "RCRA Subtitle C" conventional cover, consisting
       (from  top to bottom) of 600 mm of topsoil, a GT separator/filter, 300 mm of sand
       drainage material, a 1-mm-thick linear low density polyethylene GM, and 600
       mm of compacted clay with a design hydraulic conductivity less than or equal to
       1 x 10~7 cm/s (although an in situ hydraulic conductivity test indicated a hydraulic
                                      4-64
-------
       conductivity of 8 x 10~7 cm/s, with the comparatively large hydraulic conductivity
       thought to have been caused by desiccation cracking during construction).
  3.   Alternative Test Cover 1 is essentially identical to the RCRA Subtitle C cover,
       except that it incorporates a GCL rather than CCL and (very significantly), the
       GM component was punctured with 8 holes, each measuring 1 cm2, to simulate
       defects in the GM.
  4.   Alternative Test Cover 2 is a capillary barrier, which makes use of a clean,
       granular layer below a topsoil layer to provide a capillary break between the
       topsoil and underlying soils, thus promoting moisture  retention in the topsoil
       layer.  So long as the granular layer beneath the topsoil remains relatively dry,
       the downward movement of moisture should be minimal.  The capillary barrier
       test cover consists (from top to bottom) of 300 mm of topsoil, an upper lateral
       drainage layer comprised of 80 mm of sand underlain by 220 mm of clean pea
       gravel (the sand serves as a filter that prevents the overlying topsoil from
       migrating downward into the gravel), a barrier layer consisting of 450 mm of
       compacted soil, and a lower drainage layer comprised of 300 mm of sand.  The
       "barrier layer" was compacted dry of optimum and was not intended to have a
       hydraulic conductivity comparable to a traditional CCL.
  5.   Alternative Test Cover 3 is referred to as the anisotropic barrier and attempts to
       limit downward movement of water with  a layering of capillary barriers. The
       various layers are enhanced by varying  soil properties and techniques that  lead
       to the anisotropic properties of the cover. The anisotropic barrier consists (from
       top to bottom) of 150 mm of vegetative material (a mixture of 75% topsoil and
       25% pea gravel by weight), 600 mm of native soil to allow for water storage, a
       150-mm-thick interface layer consisting of fine sand that serves as a filter
       between the overlying native soil and underlying gravel, and 150 mm of pea
       gravel.  The fine sand layer was intended to create one capillary break, and the
       gravel was intended to create a second  capillary break.
  6.   Alternative Test Cover 4 is referred to as the evapotranspiration (ET) cover.
       The ET cover consists of a single, 900-mm-thick layer of native soil.  The bottom
       750 mm of soil was placed in lifts and compacted, while the top 150 mm was not
       compacted.  The cover material was seeded with native species that contained
       a mix of cool and warm weather plants (primarily native grasses).

Preliminary results have indicated that all six test covers are performing well, although
there are significant differences in percolation rates. Dwyer (personal communication)
provided a summary of the first year of percolation, as indicated in Table 4-5.  Cost data
are also summarized in Table 4-5.  The test cover program will provide valuable insights
into conventional and alternative cover designs as data are developed, analyzed, and
published.
                                      4-65
-------
Table 4-5.  Summary of Preliminary Data from ALCD Project.
Test Cover
RCRA Subtitle D Cover
RCRA Subtitle C Cover
Alternative RCRA Subtitle C Cover
with GCL (GM with 8 Defects)
Capillary Barrier
Anisotropic Barrier
Evapotranspiration Cover
Construction Cost
($/m2)
$51
$158
$90
$93
$75
$74
Percolation (L)
after One Year
6724
46
572
804
63
80
4.3.6 Test Covers in East Wenatchee, Washington
Khire et al. (1997) describe a project at the Greater Wenatchee Regional Landfill in East
Wenatchee,  Washington (a semi-arid region), in which two test covers were constructed
and monitored.  The test covers measured 30 m by 30 and were constructed on a
2.71-1:1 V slope.  The test covers were instrumented to measure runoff and percolation,
as well as to monitor moisture conditions within the various layers of the test covers.

Test Cover 1 was referred to as a "resistive barrier" and was a RCRA Subtitle D type
cap. Test Cover 1 consisted of 150 of topsoil underlain by a 450-mm-thick barrier layer
constructed from low plasticity silty clay that was compacted to achieve a hydraulic
conductivity of 2 x 10~7 cm/s.  The low-permeability barrier layer was intended to provide
resistance to infiltration of water, and thus the use of the term "resistive barrier." Test
Cover 2 was a capillary barrier consisting of 150 mm of vegetated silt  topsoil, underlain
by a 750-mm-thick layer of medium,  uniformly graded sand that served as the capillary
break layer.

Performance of the test covers has been documented for a 3-year period (Khire et al.,
1997).  For the first three years, Test Cover 1  allowed percolation of a total of 33 mm of
water (equal to 5.1% of precipitation) through the cover, while Test Cover 2 allowed  only
5 mm (equal to 0.8% of precipitation) to percolate through the cover.  Significant
percolation through the capillary barrier occurred only during the winter of 1993, when
record snow fall occurred.  If the surface layer of the capillary barrier cover had been
increased, it is anticipated that percolation through the capillary barrier would have been
nearly zero.  In the resistive barrier cap, percolation occurred only when the wetting
front reached the base of the low-permeability barrier layer. Percolation increased
significantly in 1995. The primary reason for this increase appeared to be preferential
flow through vertical cracks in the barrier layer, which apparently formed from
desiccation during the previous summer.  Animal burrows, found during field
reconnaissance in the spring of 1995, may have also contributed to increase in
percolation.
                                      4-66
-------
The data indicating an increase in percolation through the low-permeability clay layer is
consistent with observations at Omega Hills, Kettleman City, and Hamburg.

4.3.7  Test Covers at Los Alamos National Laboratory
Nyhan et al. (1997) describe the performance of four test covers constructed at Los
Alamos  National Laboratory for the Protective Barrier Landfill Cover Demonstration.
The four test plots were each constructed on slopes of 5, 10, 15, and 25%, making a
total of 16 test plots. None of the plots was vegetated, apparently to simulate extreme
conditions in which plants provided no evapotranspiration.  Precipitation, runoff,
drainage, and percolation were measured for each plot.  The moisture content of the
soils was also monitored.  Performance for the first four years is documented by Nyhan
etal. (1997).

The four test plots contained the following cross sections:

   1.   Test Cover 1 was termed the "conventional design" for Los Alamos, and
       consisted of 150 mm of loam topsoil underlain by 760  mm of crushed Los
       Alamos tuff (an angular, silty sand), underlain by 300 mm of gravel.
   2.   Test Cover 2 was termed the "EPA design" and consisted (from top to bottom)
       of 610 mm of loam topsoil, a GT separator/filter,  300 mm of sand drainage
       material, and 610 mm of low-permeability clay-sand material. The GM
       component that usually overlies compacted clay in "EPA designs" was
       intentionally omitted because it was thought when the design was conceived in
       the late 1980s  that the GM would not have a  sufficiently long service life for
       radioactive waste disposal units.
   3.   Test Cover 3 was termed the "loam capillary  barrier design" and consisted of
       610 mm of loam topsoil  underlain by 760 mm of fine sand, which served as the
       capillary break.
   4.   Test Cover 4 was termed the "clay loam capillary barrier design" and consisted
       of 610 mm of clay loam topsoil underlain by 760  mm of fine sand.

Performance data showed that 86% to 91 %  of all precipitation that fell on the covers
was evaporated from the unvegetated test covers, which was  not unexpected in the
semi-arid climate of Los Alamos. Of the four test covers, the EPA design provided the
least amount of percolation through  the test plots (zero percolation on all four test plots
employing the EPA design). The bentonitic clay mixed in with sand to form  the barrier
layer apparently helped with the water balance at this semi-arid site.  Test Cover 1
(conventional design for Los Alamos) allowed the greatest amount of seepage, varying
from 174 mm  of percolation for the 5% slope to 31 mm for the 25% slope over a 4.5
year period.  Test Cover 3 (loam capillary barrier design) allowed 76 mm of percolation
for the 5% slope, 36 mm of percolation for the 10% slope, and no percolation for the
15% and 25% slopes over the same 4.5 year period.  Test Cover 4 (clay loam capillary
                                      4-67
-------
barrier design) allowed 48 mm of percolation for the 5% slope, and no percolation for
test plots on steeper slopes for the 4.5-year observation period.

The Los Alamos test plots appear to be the only documented cases in which a
compacted clay placed in a test cover without a protective GM worked well over a
period of several years of observation.

4.3.8  Other Studies
Other papers have been  published on the performance of CCLs in final covers, but
none is as comprehensive as the studies discussed in the preceding sections.
Questions have been raised about the long-term survivability of CCLs, even if the CCL
is covered by a GM.  Suter et al. (1993) discuss the factors that might cause long-term
degradation of CCLs. The primary mechanisms of concern are desiccation, freeze-
thaw,  thermally induced moisture movement leading to desiccation,  root penetration,
subsidence, and animal intrusion.

It appears that the best way to document the field performance of CCLs in landfill final
cover systems is with the use of lysimeters installed at the base of the cover system.
Lysimeters consist of a barrier (typically GM) overlain by a drainage material (typically
sand or gravel, but possibly GN or other geosynthetic drainage material), and drained
by gravity at the low point. Few, if any, such systems have been installed, except in test
plots.  Until performance data are  collected over a period of many years on actual
covers, the long-term performance of CCLs will remain the subject of speculation.
However, the admittedly  sparse data that are available points to the likelihood (if not
certainty) of desiccation and subsequent flow rates well in excess of those associated
with a CCL having a hydraulic conductivity of 1 x 10~7 cm/s or less for cover systems
employing a layer of topsoil overlying a CCL (with no GM separating the CCL and
topsoil).   Although the monitoring of leachate production rates can be very useful in
indicating whether or not the cover is working reasonably well, a careful analysis of
actual flow rates through the cover system may be difficult with such a global
measurement as leachate production rate. Scientists and engineers are encouraged to
collect percolation data with lysimeters whenever possible.

4.4 Summary and Conclusions
The objective of this  component of the study was to document the field performance of
CCLs, and particularly to address  the question of whether CCLs are meeting the
objective of having a hydraulic conductivity of 1 x  10~7 cm/s or less.  Field performance
data on CCLs employed  as liners for actual landfills are very limited (only 8 such cases
are documented),  and even fewer data are available on CCLs in landfill cover systems.
Therefore, the  approach  taken was: (1) document large-scale hydraulic conductivity
measurements obtained  from test pads (with the 8 cases of documented field
performance), and (2) document information available on test covers used to evaluate
CCL performance in landfill cover  systems.
                                      4-68
-------
A database consisting of 89 CCLs (81 test pads plus 8 actual bottom liners) constructed
from natural soils was assembled and analyzed for large-scale field hydraulic
conductivity. The cases covered a broad range of soil types, construction methods, and
regions. All CCLs were constructed using construction and CQA procedures that
appear to be consistent with the current state of practice, and only those CCLs that
were constructed for the explicit purpose of achieving a field hydraulic conductivity
(kpieid) < 1 x 10~7 cm/s were included in the database. Conclusions may be summarized
as follows: (1) 25% of the 89 CCLs failed to achieve the desired  large-scale hydraulic
conductivity of 1  x 10~7 cm/s or less, confirming the difficulty that is often encountered in
achieving the required low hydraulic conductivity and indicating that achieving this level
of impermeability requires careful planning, use of appropriate materials, specification of
suitable compaction requirements, and thorough CQA; (2) a few of the CCLs failed to
meet the desired hydraulic conductivity because the soil materials turned out not to be
suitable; (3) for soils that were found to be unsuitable, a comprehensive laboratory
testing program had not preceded construction of the test pad - thorough laboratory
testing is recommended for all CCLs prior to construction to verify the suitability of the
soil and the proposed compaction specification; (4) the single most common problem in
achieving the desired low level of hydraulic conductivity was failure to compact the soil
in the zone of moisture and dry density that will yield low hydraulic conductivity; (5) the
most significant control parameter was not found to be water content or density, but
rather a parameter denoted "P0", which represents the percentage of field-measured
water content-density points that lie on or above the line of optimums - when P0 was
high (80% to 100%) nearly all the CCLs achieved the desired field hydraulic
conductivity, but when P0  was low (0 to 40%), fewer than half the CCLs achieved the
desired field hydraulic conductivity; (6) practically  no correlation was found between
hydraulic conductivity and frequently measured  soil characterization parameters, such
as liquid limit, plasticity index, percentage of clay,  percentage of fine material, indicating
that natural soil CCLs can be constructed with a relatively broad range of soil materials;
(7) good agreement was obtained between kFieid and kLab on small samples for well
constructed  liners with kFieid < 1 x 10~7 cm/s, but poor agreement was found for poorly
constructed  liners with kFieid > 1 x 10"7 cm/s (with laboratory measurements often
yielding significantly lower hydraulic conductivities); and (8) hydraulic conductivity
decreased with increasing thickness of CCLs, up to a thickness of about 1 m, at which
point all CCLs in the database achieved kFieid < 1 x 10~7 cm/s.

A database on soil-bentonite CCLs was also assembled, but only contained  12 field-
measured hydraulic conductivities on test pads. Relatively little information could be
gleaned from the database for soil-bentonite liners.  The desired hydraulic conductivity
of < 1 x 10~7 cm/s was achieved in all 12 cases.  However, all the CCLs in the database
contained a relatively large amount of bentonite (more than 6%).  The data suggest that
there is justification for focusing attention on a high percent compaction for soil-
                                      4-69
-------
bentonite liners rather than a high water content. More data are needed to be able to
draw more definitive conclusions about soil-bentonite liners.

Finally, data were assembled from the literature on performance of CCLs in landfill final
cover systems. Most of the field data indicate that CCLs tend to desiccate over time,
and if not covered with a GM overlain by soil, are likely to undergo significant increases
in hydraulic conductivity within five years after construction as a result of desiccation
cracking. If the CCL is protected from desiccation by a GM and covering soil,  it appears
that water percolation through the composite barrier will be extremely small, and that
the CCL probably be protected from desiccation for at least several years, if not longer.
Of the data analyzed from final cover systems, only the Omega Hills and  Hamburg test
sites were situated on an actual landfill cover, where differential settlement was a
possibility.  MSW landfills are known to undergo significant settlement, which can
produce stress-induced cracking that increases hydraulic conductivity. Although the
discussion herein focused primarily on the desiccation issue, this was because the test
covers themselves were impacted far more by desiccation than by settlement, due to
the nature of the  test arrangements. Settlement-induced cracking may be a far more
significant effect than indicated by this collection of information from performance of
cover system test sections.

4.5 References
Benson, C.H. and Boutwell, G. (1992). "Compaction Control and Scale-Dependent
   Hydraulic Conductivity of Clay Liners," Proceedings of the 15th Annual Madison
   Waste Conference, University of Wisconsin, Madison, Wisconsin, 62-83.
Benson, C.H., Zhai, H., and Rashad, S.M. (1992), "Assessment of Construction Quality
   Control Measurements and Sampling Frequencies for Compacted Soil Liners,"
   Environmental Geotechnics Report No. 92-6, Univ. of Wisconsin, Dept. of Civil and
   Environmental Engineering Madison, Wisconsin.
Benson, C.H., Hardianto, F.S., and Motan, E.S. (1994), "Representative Specimen Size
   for Hydraulic Conductivity Assessment of Compacted Soil Liners," Hydraulic
   Conductivity and Waste  Contaminant Transport in Soils, ASTM STP 1142, D.E.
   Daniel and S.J. Trautwein (Eds.), American Society for Testing and Materials,
   Philadelphia, 3-29.
Benson, C.H., and Daniel, D.E. (1994), "Minimum Thickness of Compacted Soil Liners:
   II. Analysis and Case Histories." Journal  of Geotechnical Engineering, 120 (1): 153-
   172.
Corser, P., and Cranston, P. (1991), "Observations on Long-Term Performance of
   Composite Clay Liners and Covers," Proceedings, Geosynthetic Design and
   Performance, Vancouver Geotechnical Society, Vancouver, BC.
Corser, P., Pellicer, J., and Cranston, M. (1992), "Observation on Long-Term
   Performance of Composite Clay Liners and Covers," Geotechnical Fabrics Report,
   November, pp. 6-16.
                                      4-70
-------
Daniel, D.E. (1989), "In Situ Hydraulic Conductivity Tests for Compacted Clays," Journal
  of Geotechnical Engineering, 115(9): 1205-1227.
Daniel, D.E., and Benson, C.H. (1990), "Water Content-Density Criteria for Compacted
  Soil Liners," Journal of Geotechnical Engineering, 116(12): 1811-1830.
Daniel, D.E., and Koerner, R.M. (1995), Waste Containment Systems: Guidance for
  Construction, Quality Assurance, and Quality Control of Liner and Cover Systems,
  ASCE Press, New York, 354 p.
Dwyer, S.F. (1997), "Large-Scale Field Study of Landfill Covers at Sandia National
  Laboratories," Proceedings, Landfill Capping in the Semi-Arid West: Problems,
  Perspectives, and Solutions, Environmental Science and Research Foundation,
  Idaho, Falls, ID, ESRF-019, 87-108.
Gordon, M.E., Huebner, P.M., and Mitchell, G.R.  (1990), "Regulation, Construction and
  Performance of Clay Lined Landfills in Wisconsin," Waste Containment Systems, R.
  Bonaparte (Ed.), American Society of Civil Engineers, New York, 14-27.
Khire, M.V., Benson, C.H., and Bosscher, P.J. (1997), "Water Balance of Two Earthen
  Landfill Caps in a Semi-Arid Climate," Proceedings, International Containment
  Technology Conference, St. Petersburg, Florida, 252-261.
Maine Bureau of Remediation and Waste Management (1997), An Assessment of
  Landfill Cover System Barrier Layer Hydraulic Performance, Augusta, Maine.
Melchior, S., Berger, K., Vielhaver, B., and Miehlich, G. (1994), "Multilayered Landfill
  Covers: Field Data on  the Water Balance and  Liner Performance," In-Situ
  Remediation: Scientific Basis for Current and Future Technologies, G.W. Gee and
  N.R. Wing (Eds.), Battelle Press, Columbus, Ohio, Part 1, pp. 4111-425.
Melchior, S. (1997), "In-Situ Studies of the Performance of Landfill Caps," Proceedings,
  International Containment Technology Conference, St.  Petersburg, Florida, 365-373.
Montgomery, R.J., and Parsons, L.J. (1989), "The Omega Hills Final Cover Test Plot
  Study: Three-Year Data Summary," Presented at the Annual Meeting of the National
  Solid Waste Management Association, Washington, DC.
Nyhan, J.W., Schofield, T.G., and Salazar, J.A. (1997), "A Water Balance Study of Four
  Landfill Cover Designs Varying in Slope for Semiarid Regions," Proceedings,
  International Containment Technology Conference, St.  Petersburg, Florida, 262-269.
Reades, D.W., Lahti, L.R., Quigley, R.M., and Bacopoulos, A. (1990), "Detailed Case
  History of Clay Liner Performance,"  Waste Containment Systems:  Construction,
  Regulation, and Performance, R. Bonaparte (Ed.), American Society of Civil
  Engineers, New York,  156-174.
Sai, J.O., and Anderson,  D.C. (1990), "Field Hydraulic Conductivity Tests for
  Compacted Soil Liners," Geotechnical Testing Journal,  13(3): 215-225.
Suter, G.W., Luxmoore, R.J., and Smith, E.D. (1993), "Compacted Soil Barriers at
  Abandoned Landfill Sites Are Likely to Fail in the Long Term," Journal of
  Environmental Quality, Vol. 22, pp. 217-226.
Trautwein, S.J., and Boutwell, G.P. (1994),  "In-Situ Hydraulic Conductivity Tests for
  Compacted Soil Liners and Caps," Hydraulic Conductivity and Waste Contaminant
                                     4-71
-------
Transport in Soils, ASTM STP 1142, D.E. Daniel and S.J. Trautwein (Eds.), American
Society for Testing and Materials, Philadelphia, 184-223.
                                   4-72
-------
                                  Chapter 5
              Detailed Summary of Field Performance Tasks

5.1  Introduction

5.1.1  Scope of Work
This portion of the project involved four tasks designed to evaluate the field
performance of liner systems and final cover systems (referred to as cover systems in
this chapter) for modern landfills in the U.S.  The term "modern landfill" refers to a
landfill designed with components substantially meeting current federal regulations and
constructed and operated to the U.S. state of practice from the mid-1980's forward.  The
four tasks are:

   •    evaluation of available published information on the field  performance of modern
       landfills;
   •    collection and analysis of liquids management data for double-lined landfills;
   •    evaluation of problems that have occurred in waste containment systems (i.e.,
       liner systems and cover systems) for waste management facilities; and
   •    assessment of the adequacy of the EPA HELP computer model as a tool for
       LCRS design.

The purpose of performing these tasks is to  develop an improved understanding of the
actual field performance of modern landfill liner systems and cover systems  and, to the
extent possible, provide data that allow answers to be developed for the following
questions:

   1.   What conclusions can be drawn from available LCRS and LDS data  regarding
       leakage rates through primary liners  and hydraulic efficiencies of liners?
   2.   How much leachate is generated in modern landfills, both during  active
       operations and after closure, and what is the effect of site location (climatic
       region) and waste type on leachate generation rates?
   3.   What is the chemistry of modern  landfill leachate and what  is the effect of site
       location, waste type,  and operation conditions on leachate chemistry?
   4.   What is the effect of the federal solid waste regulations of the 1980's and early
       1990's, which limit the disposal of certain types of wastes to HW facilities and
       prohibit the disposal of certain types  of waste in any facility, on landfill leachate
       chemistry?
   5.   How do leachate generation  rates estimated using the EPA HELP computer
       model compare to actual leachate generation rates at modern operating
       facilities?
   6.   Do the HELP model simulations predict the same effects of site location and
       waste type on leachate generation rates as observed from the actual data?
   7.   What is the nature, frequency, and significance of identified problems in liner
       systems and cover systems for modern waste management facilities?
   8.   How can the identified problems be prevented in the future?

                                      5-1
-------
   9.   What overall conclusions can be drawn regarding the likely long-term
       performance of landfills?

Complete results from the first three tasks were incorporated into two appendices to this
report. These appendices are:

   •   "Appendix E:  Evaluation of Liquids Management Data for Double-Lined
       Landfills"; and
   •   "Appendix F: Waste Containment System Problems and Lessons Learned".

Summaries of Appendices E and F are presented below in Sections 5.2 and 5.3,
respectively, of this report.  Section 5.4 presents the results of the fourth task, an
assessment of the appropriateness of the HELP model as a design tool.  The HELP
model is evaluated by comparing LCRS flow rate data for six landfill cells to leachate
generation rates predicted for these cells from HELP model simulations with typical
input parameters.  Sections 5.2 through 5.4 also present the findings of the project with
respect to the questions listed  above.  References are presented in Section 5.5.

5.1.2  Terminology
Waste containment systems for landfills consist of liner systems that underlay the
wastes placed in them and cover systems constructed over the wastes (Figure 5-1). A
liner system consists of a combination of one or more drainage layers and low-
permeability barriers (liners). The functions of liners and drainage layers are
complementary. Liners impede leachate percolation and gas migration out of a landfill
and improve the collection capability of overlying drainage layers.  Drainage layers
collect and convey liquids on underlying liners to controlled collection points (sumps)
and limit the buildup of hydraulic head on the liners. Most liner systems installed
beneath modern landfills are classified as single-composite  liner systems or double-liner
systems and include the components illustrated in  Figure 5-1.  A single-composite liner
system consists of a composite liner overlain by an LCRS drainage layer.  A double-
liner system consists of a primary liner and a secondary liner with an LDS drainage
layer between the two liners and an LCRS drainage layer above the primary liner. The
LCRS and LDS may also contain networks of perforated pipes, sumps, pumps,
flowmeters, and other flow conveyance and monitoring components.  A liner system
may also include a protection layer over the LCRS drainage layer to further isolate the
liner from the environment (e.g., freezing temperature, stresses from equipment).

Once  an area of a landfill is filled to final grade, a cover system is constructed over the
area to contain the waste, minimize the infiltration of water into the waste, and control
the emissions of gases produced by waste decomposition or other mechanisms. A
cover system consists of up to six basic components, from top to bottom: (i) surface
layer;  (ii)  protection layer; (iii) drainage layer; (iv) barrier; (v) gas collection layer; and (vi)
foundation layer.  In some cases, the functions of several adjacent components can be
                                      5-2
-------
provided by one soil layer. For example, a sand gas collection layer may also serve as
a foundation layer. Many modern landfills have a cover system consisting of a soil
surface and protection layer, drainage layer, barrier, and gas collection layer.
         Cover
        System
                      J
                JL
Surface and Protection Layer

    Drainage Layer
                   Gas Collection Layer
                     GM Barrier
  Single-Composite
     Liner System
        (a)
                      Solid Waste
                    LCRS Drainage Layer
                                            Double-Liner
                                              System
                     Composite Liner
                                                            Solid Waste
                                                          LCRS Drainage Layer
                                          LDS Drainage Layer
                                       Composite Secondary Liner
                               (b)
Figure 5-1.  Typical waste containment system components for landfills: (a)
            single-composite liner system and cover system for a closed landfill;
            and (b) double-liner system for an active landfill.
In general, the materials used to construct liners and barriers in modern landfills are
GMs alone and composites consisting of GMs overlying CCLs or GCLs (i.e., GM/CCL or
GM/GCL composites).  Drainage layers and gas collection layers are typically
constructed with sand, gravel, GNs, or GCs.  Protection layers typically consist of soil or
thick GTs. The protection layer over the LCRS drainage layer sometimes consists of
select waste. Surface layers for cover systems are typically constructed with vegetated
topsoil.

Liner systems for modern MSW landfills and nonhazardous MSWcombustor ash (MSW
ash) landfills must, based on state-specific implementation of RCRA Subtitle D
requirements, meet federal  minimum design criteria or performance-based design
requirements (40 CFR 258.40) as described in Section 1.2. Federal minimum design
criteria require a single-composite liner system for new MSW landfills and MSW ash
landfills. While the federal minimum design criteria were adopted by many states, a few
states require that MSW landfills or MSW ash landfills have a double-liner system. For
RCRA HW landfills, federal  regulations (40 CFR 264.301) require a double-liner system
with at least a GM primary liner and a GM/CCL secondary liner, as described in Section
1.2.
                                       5-3
-------
Cover systems for modern lined MSW landfills and MSW ash landfills (40 CFR 258.40)
must meet federal minimum design criteria or performance-based design requirements
(40 CFR 258.60), as described in Section 1.2. The cover system meeting federal
minimum design criteria consists of a soil surface layer over a composite barrier.  Cover
systems for RCRA HW landfills must meet federal performance-based design
requirements (40 CFR 264.310).  There is not a federal minimum design criteria cover
system for HW landfills; however, EPA guidance (EPA, 1989) recommends that the
cover system for these landfills contain a soil surface and protection layer, drainage
layer, and composite barrier, as described in Section 1.2.

There are currently no federal  minimum design requirements for liner systems or cover
systems for ISW landfills.  ISW landfills contain such wastes as papermill sludge, coal
ash, and construction and demolition waste (C&DW).

5.1.3  Data Collection Methodology
The landfill performance data presented in Appendices E and F and summarized in this
chapter were obtained from the technical literature, engineering drawings, project
specifications, as-built records, and operation records, and from discussions with facility
owners, facility operators, design  engineers, and federal and state regulators throughout
the U.S. The data were collected in accordance with a quality assurance plan, which
was reviewed and approved by the EPA. Efforts were made to obtain data from a wide
variety of facilities with different waste types (i.e., MSW, MSW ash, HW,  and  ISW), site
conditions, and waste containment system components. The study focused on landfills,
and only information on landfills is summarized in this chapter. Based on the broad-
based method of data collection for this study, it is believed that the data in this report
are generally representative of landfills nationwide.

5.2 Evaluation of Liquids Management Data for Double-Lined Landfills

5.2.1  Scope of Work
The scope of work for this portion of the project consisted of the collection and analysis
of liquids management data for 187 active or closed double-lined cells at 54 modern
landfills located throughout the U.S. These data are typically  required to be collected
and reported to regulatory agencies as part of the permit conditions for a landfill. The
data were used to evaluate: (i) leakage rates and hydraulic efficiencies of landfill
primary liners; (ii) landfill  leachate generation rates (LCRS flow rates), including how
these flow rates vary with waste type, site location, and presence of cover system; and
(iii) landfill leachate chemistry (LCRS flow chemistry), including how leachate chemistry
varies with waste type, site location, and operation conditions, and whether federal solid
waste regulations promulgated in the 1980's and early 1990's have had an effect on the
quantity of potentially-toxic trace chemicals found in leachate.
                                       5-4
-------
5.2.2 Description of Database
The liquids management data and related data collected for the 187 landfill cells
include: (i) general facility information (including location, average annual rainfall,
subsurface soil types, groundwater separation distance from bottom of landfill); (ii)
general cell information (including cell area, type of waste, height of waste, dates of
construction, operation, and closure); (iii) double-liner system and cover system design
details (including type, thickness, and hydraulic conductivity of each layer); (iv) LCRS
flow rate and chemical constituent data; and (v)  LDS flow rate and chemical constituent
data.  For comparison purposes, the data are sorted according to liner system type,
waste type, and site geographic location (which  is indicative of site climate).  The full
database is presented in Appendix E. The reader is referred to the following figures and
tables in Appendix E for specific information:

   •   double-liner system types:  Table E-1.1 and Figure E-1.1;
   •   geographic regions and site locations: Figure E-1.3;
   •   general facility information: Table E-3.1;
   •   general cell information: Table E-3.2;
   •   double-liner system design details: Table E-3.3;
   •   cover system design details:  Table E-3.4;
   •   LCRS flow rate data: Table E-3.5;
   •   LDS flow rate data: Table E-3.6; and
   •   LCRS and LDS flow chemistry data: Table E-3.7.

The distributions of the landfill facilities and cells in the database by waste type and
geographic region and by primary liner and LDS types are shown in Tables 5-1 (a) and
(b), respectively. From  Table  5-1 (a), most of the landfills in the database are located  in
the northeast (NE).  This is not surprising because: (i) the NE has a relatively dense
population; and (ii) double-liner systems are required for MSW landfills in several states
in the NE. In addition, the majority of the landfills in the database are used for disposal
of MSW. Based on the extent of the database and comparisons of these data with
published data, discussed in Section E-2 of Appendix E, the database appears to
adequately characterize conditions for MSW landfills in the NE and southeast (SE), HW
landfills in the NE and SE, and MSW ash landfills in the NE. The database is quite
sparse for landfills in the west (W), coal ash landfills, and C&DW landfills. Additional
data from these facilities should be collected and evaluated.

From Table 5-1 (b), most of the cells at most of the landfills have either a GM primary
liner (37% of all cells) or GM/CCL or GM/GCL/CCL primary liner (48%).  Fewer cells
(15%) have a GM/GCL  primary liner. About 48% of the cells have a sand or gravel LDS
and 52% have a GN LDS. Based on the distribution of the data, the database appears
to be representative of typical  double-liner system designs in landfills.
                                       5-5
-------
Table 5-1 (a). Distribution of Database by Waste Type and Geographic Region.

Waste Type
MSW
HW
MSW Ash
Coal Ash
C&DW
Geographic Region
Northeast U.S.
24 landfills
71 cells
5 landfills
26 cells
5 landfills
12 cells
1 landfills
1 cell
2 landfills
4 cells
Southeast U.S.
8 landfills
26 cells
5 landfills
31 cells
2 landfills
4 cells
0 landfills
0 landfills
West U.S.
1 landfill
2 cells
3 landfills
10 cells
0 landfills
0 landfills
0 landfills
Table 5-1 (b). Distribution of Database by Primary Liner and LDS Types

Primary Liner Type
GM
GM/GCL Composite
GM/CCL or
GM/GCL/CCL
Composlte
LDS Type
Sand or Gravel
13 landfills
41 cells
3 landfills
19 cells
13 landfills
31 cells
GN
11 landfills
28 cells
4 landfills
9 cells
16 landfills
57 cells
Most of the liquids management data are for open cells; only about 23% of the cells in
the database had received a cover system.

5.2.3  Data Interpretation

5.2.3.1  Landfill Development Stages
In evaluating  LCRS and LDS flow rate and chemical constituent data for this report,
three distinct  landfill development stages were considered: (i) the "initial period of
operation"; (ii) the "active period of operation"; and (iii) the "post-closure period".  These
stages are defined by the waste filling and capping rates of a landfill cell and are
described below.  The initial period of operation occurs during the first few months after
the start of waste disposal in a cell. During this stage, there is not sufficient waste in a
cell to significantly impede the flow of rainfall into the LCRS.  To the extent rainfall
occurs during this stage, it will rapidly find its way into the LCRS. LCRS flow rates
during this stage are  usually controlled by rainfall and can be directly correlated to local
climatic conditions. LCRS flow rates are higher at landfills in wetter climates than at
those in arid climates. During the active period of operation, the cell is progressively
                                       5-6
-------
filled with waste and daily and intermediate layers of cover soil.  As waste placement
continues, more of the rainfall occurring during this stage falls onto the waste and cover
soils rather than directly onto the liner system. As a consequence, the LCRS flow rates
decrease and eventually stabilize.  LCRS flow rates during this stage are generally
dependent on rainfall, waste thickness, waste properties (i.e., initial moisture content,
field capacity, and permeability), and storm-water management practices. During the
post-closure period,  the cell has been closed with a cover system  that further reduces
infiltration  of rainfall into the waste, resulting in further reduction in LCRS flow rates.
LCRS and LDS flows associated with these three development stages are illustrated for
a MSW landfill in Pennsylvania in Figure 5-2.

5.2.3.2 Primary Liner Leakage Rates and Hydraulic Efficiencies
LCRS and LDS flow data were interpreted to assess primary liner leakage rates and/or
apparent efficiencies for the following:

   •   GM primary liners by development stage, LDS type, and use of CQA;
   •   GM/GCL primary liners by development stage and LDS type; and
   •   GM/CCL primary liners by development stage.

The data were first assessed using a methodology presented by Gross et al. (1990) for
using LCRS and LDS flow data to evaluate the performance of primary liners in terms of
primary liner leakage. The basic approach involves the evaluation of LCRS  and LDS
flow rate and chemical constituent data to quantify that portion of LDS flow that is
attributable to primary liner leakage as opposed to other sources.  Other sources of LDS
flow include: (i) water (mostly rainwater) that infiltrates the LDS during construction and
continues  to drain to the  LDS sump after the start of facility operation ("construction
water"); (ii) water that infiltrates the LDS during construction, is held in  the LDS by
capillary tension, and is expelled from the LDS during waste placement as a result of
LDS compression under  the weight of the waste ("compression water"); (iii) water
expelled into the LDS from any CCL and/or GCL components of a composite primary
liner as a result of clay consolidation  under the weight of the waste ("consolidation
water"); and (iv) water that percolates through the secondary liner and  infiltrates the
LDS ("infiltration water"). The sources of LDS flow are illustrated in Figure 5-3.
Evaluation of the potential flow rates  and times of occurrence of each of these potential
sources of flow were made using the calculations procedures contained in Gross et al.
(1990). That portion of LDS flow attributable to leakage through the composite primary
liner of a double-liner system would be leakage into the ground for a single-composite
liner system if the two composite liners have similar characteristics.  Extrapolation of
primary liner performance levels to the secondary liner of a double-liner system enables
inferences to be drawn regarding performance of the entire double-liner system.
                                      5-7
-------
                Initial Period
                of Operation
                            Active Period of Operation
   160
   140
                                       o>

                                       —>
                                            CM
                                            O>
CM
O>
CO
O)
CO
O>
                o>
                c
                ro
                o>
                "5
                -3
LO
O)
LO
O)
                                           DATE
Figure 5-2. LCRS and LDS flow rates over time at a MSW landfill in Pennsylvania.
            (Flow rates are given in liters/hectare/day (Iphd).)
                                         5-8
-------
                           GM
           '. _ .QRQUNPWATER.TABLE_ . _
        Q = TOTAL FLOW
        Q=A+B+C+D

        SOURCES:
        A = PRIMARY LINER LEAKAGE
        B = CONSTRUCTION WATER AND COMPRESSION WATER
        C = CONSOLIDATION WATER
        D = INFILTRATION WATER
    Figure 5-3. Sources of flow from LDSs (from Bonaparte and Gross, 1990).

The relative performances of the different types of primary liners were then evaluated
using the "apparent liner hydraulic efficiency" parameter, Ea, introduced by Bonaparte et
al. (1996) and defined as:
              Ea(%) = (1 - LDS Flow Rate/LCRS Flow Rate) x 100
(Eq.5-1)
The higher the value of Ea, the smaller the flow rate from an LDS compared to the flow
rate from an LCRS. The value of Ea may range from 0 to 100%, with a value of zero
corresponding to an LDS flow rate equal to the LCRS flow rate and a value of 100%
indicating no flow from the LDS.  The parameter Ea is referred to as an "apparent"
hydraulic efficiency because, as described above, flow into the LDS sump of a landfill
may be due to sources other than primary liner leakage (Figure 5-3). The value of Ea is
calculated using total flow into the LDS, regardless of source. If the only source of flow
into the LDS sump is primary liner leakage, then Equation  5-1 provides the "true" liner
hydraulic efficiency, Et. True liner efficiency provides a measure of the effectiveness of
a particular liner in limiting or preventing advective transport across the liner.  For
example, if a liner is estimated to have an Et value of 99%, the rate of leakage through
the primary liner would be assumed to be 1 % of the LCRS flow rate. The true efficiency
of a liner is not constant but rather a function of the hydraulic head in the LCRS and size
of the area over which LCRS flow is occurring (the area is  larger at high flow rates
                                      5-9
-------
compared to low flow rates). The true efficiency of a liner is also a function of design:
identical liners overlain by different LCRSs or placed on different slopes will exhibit
different Et values.  Also, the efficiency of a liner for a given set of hydraulic conditions
could change over time if the physical condition of the liner changes. For example,
time-dependent changes in GMs could result from chemical degradation or stress
cracking under certain conditions. Time-dependent changes in CCLs or GCLs could
result from chemical degradation, consolidation, or other factors.  Notwithstanding all of
these limitations, the hydraulic efficiency concept  has been found useful in
characterizing liner hydraulic performance.

The methodology described above was used to evaluate the hydraulic performance of
GM primary liners and GM/GCL composite primary liners.  Chemical constituent data
were not utilized in  the evaluation of these types of liners because the initial hydraulic
assessment (i.e., comparing LCRS and LDS flow  rates) yielded significant insight into
these liners' true hydraulic efficiencies. However,  the situation was found to be more
complicated for GM/CCL and GM/GCL/CCL composite primary liners due the
generation of consolidation water by these liners not only during the initial period of
operation, but also  during the active and post-closure periods.  The performance
evaluation of these liners included the additional step of comparing the chemistry of
LCRS and LDS liquids to assess whether the liquids had different primary sources (i.e.,
leachate for LCRS  liquids and  CCL pore water for LDS liquids). The concentrations of
five key chemical constituents  (i.e., the inorganic anions sulfate and chloride and the
aromatic hydrocarbons benzene, toluene, and xylene) in the LCRS and LDS flows were
compared in more detail to further assess whether primary liner leakage had contributed
to LDS flows.

It is noted that the presence of chemical constituents in the LDS was evaluated
empirically. Therefore, the concentrations of chemicals collected in the LDS were
directly compared to concentrations of the same chemicals collected in the LCRS. No
fate and transport analysis was performed that accounts for attenuation of the LCRS
chemicals migrating through the primary liner CCL.  However, to overcome the need to
perform such an analysis, the five key chemical constituents were  selected based on
their high solubility  in water, low octanol-water coefficient, high  resistance to
hydrolization, and high resistance to anaerobic biodegradation  in soil.

5.2.3.3 Leachate Generation Rates
LCRS flow rate  data were interpreted in terms of average and peak monthly leachate
generation rates for the following:

   •   MSW landfills by geographic region and development stage;
   •   HW landfills by geographic region and development stage;
   •   ash (i.e., coal ash and  MSW ash) landfills  by geographic region and
       development stage; and
                                      5-10
-------
   •   C&DW landfills by development stage.

The data were also used to evaluate the ratios of average LCRS flow rates to historical
average annual rainfalls by waste type, geographic region, and development stage.

5.2.3.4  Leachate Chemistry
Data on leachate chemical constituents were interpreted in terms of the average
concentrations and detection frequencies (i.e., were the chemicals detected in 50% or
less of the samples or more than 50% of the samples) of 30 representative chemical
parameters. These data were then used to assess the following:

   •   effect of waste type on leachate chemistry; and
   •   effect of federal solid waste regulations of the 1980's and early 1990's on
       leachate chemistry (i.e., has the amount of trace toxic inorganic and synthetic
       organic chemicals in leachate decreased).

The 30 representative chemical parameters consist of water quality indicator
parameters (e.g., pH, specific conductance, total dissolved solids (TDS), etc.), major
inorganic cations and anions (e.g., calcium, chloride, sulfate, etc.), trace metals (e.g.,
arsenic, chromium, lead, etc.), and volatile organic compounds (VOCs) (e.g., benzene,
methylene chloride, trichloroethylene, etc.). The specific trace metals and VOCs were
chosen for this study because these metals and VOCs are sometimes found in
leachates from MSW, HW, and ISW landfills. They were also selected based on
availability of parameters between landfills, frequency of detection, and concentration.

It is recognized that the leachate chemistry database is limited in terms of completeness
and duration of monitoring.  In addition, key MSW and HW leachate constituents, such
as alcohols and ketones, are poorly  represented in the database and, thus, could not be
included in the list of select parameters.  It is important that these additional data be
collected so that our understanding of leachate chemistry can continue to  improve. The
chemical data presented herein are  intended to be representative, not comprehensive.
The data should not be considered complete for purposes of evaluating potential human
health or ecological impacts.

5.2.4  Evaluation Results

5.2.4.1  Primary Liner Leakage Rates and Hydraulic Efficiencies

GM Liners
The performance of 31 of the 69 cells with GM primary liners was assessed. The
remaining 38 cells with GM primary liners were excluded from the assessment primarily
because they do not have continuous LCRS and LDS flow rate data available for an
individual cell from the start of operation and for a significant monitoring period. Flow
rate data are available for the considered 31 cells at 14 landfills with monitoring periods
                                      5-11
-------
of up to 114 months. Twenty-five cells have a HOPE GM primary liner, and the
remaining six cells have a chlorosulfonated polyethylene (CSPE) GM primary liner. A
formal CQA program was used in the construction of 23 of the 25 cells with an HOPE
GM primary liner. To the best of the authors' knowledge, none of the cells received
electrical leak location surveys or ponding tests as part of the CQA program.  The
remaining two cells with an HOPE GM primary liner (i.e., F1 and K1) and the six cells
with a CSPE GM primary liner (i.e., B1, B2, and E1 to E4) were all constructed without
CQA.  A summary of the LCRS and LDS flow rate data for the 31  cells is provided in
Table 5-2, and detailed results are given in Figure E-4.1 and Tables E-4.3 through E-4.5
of Appendix E. The major findings from the data evaluation of cells with GM primary
liners are given below:

   •   LDS flows during the initial period of operation are attributed primarily to
       construction water and primary liner leakage.  LDS flows during the active and
       post-closure periods are attributed primarily to primary liner leakage.
   •   Average monthly LDS flow rates for cells constructed with a formal CQA
       program ranged from about 5 to 440 Iphd during the initial period of operation, 1
       to 360 Iphd  during the active period, and 2 to 60 Iphd during the post-closure
       period. Peak monthly flow rates for these cells were typically below 500 Iphd
       and exceeded 1,000 Iphd in only two of the 23 cells.
   •   Based on an analysis  of the available data, average monthly active-period LDS
       flow rates for cells with HOPE GM primary liners constructed with CQA will often
       be less than 50 Iphd, but occasionally in excess of 200 Iphd. These flows are
       attributable  primarily to liner leakage and, for cells with sand LDSs, possible
       construction water.
   •   The eight cells constructed without a formal CQA program exhibited average
       monthly LDS flow rates about one to two orders of magnitude greater than LDS
       flow rates for cells constructed with CQA. The average flow rates from the eight
       cells ranged from 120 to 2,140 during the initial period of operation, 70 to 1,600
       Iphd during  the active  period, and, for the two cells for which post-closure data
       are  available, 210 to 240 Iphd during the post-closure period. The large
       differences  in LDS flow rates between cells constructed with CQA and cells
       constructed without CQA are partly attributed to the benefits of CQA and partly
       due to differences in the GM materials and construction (i.e., seaming) methods.
       The two cells that had HOPE GM primary liners and no formal CQA had average
       LDS flow rates that are about two to seven times greater than the mean LDS
       flow rate for all cells constructed with a formal CQA program. In contrast, the
       cells with CSPE  GM primary liners and no formal CQA exhibited average LDS
       flow rates that are about one to two orders of magnitude greater than the mean
       LDS flow rate for all cells that had CQA. There are not sufficient data in this
       appendix, however, to accurately separate the effects of CQA and GM type (i.e.,
       HOPE vs. CSPE) and construction methods on leakage rates through GM liners.
   •   Based on an analysis  of the available data, GM liners can be constructed to
       achieve very good hydraulic performance (i.e. Et values greater than 99%).
       However, even when constructed with a CQA program, GM liners sometimes
                                      5-12
-------
        Table 5-2. Summary of LCRS and LDS Flow Rate Data for Landfill Cells with GM Top Liners.
Cell
No.
B1
B2
C1
C2
C3
C4
C5
D1
D3
D4
Cell
Area
(hectare)
3.3
3.5
3.2
3.7
3.6
3.7
2.6
0.4
0.3
0.4
Start of
Waste
Place.
(month-
year)
5-84
5-84
5-90
4-91
8-91
2-92
11-92
10-85
7-87
1-89
End of
Final
Closure
(month-
year)
11-88
11-88
NA(5)
NA
NA
NA
NA
5-86
NA
NA
Initial Period of O
Time
Period
(months)
1-19
1-19
1-9
1-12
1-8
1-4
1-12
1-7
1-12
1-11
LCRS Flow<4)
Avg.
(Iphd)
ND(5)
ND
ND
1,475
3,417
14,828
6,419
ND
20,292
31,281
Peak
(Iphd)
ND
ND
ND
2,585
9,558
41 ,331
12,528
ND
51 ,265
120,527
Deration
LDS Flow
Avg.
(Iphd)
ND
ND
ND
92
63
178
23
32
12
233
Peak
(Iphd)
ND
ND
ND
398
268
265
40
80
56
801
^am
(%)



93.74
98.16
98.80
99.64

99.94
99.25
Active Period of Operation
Time
Period
(months)
20-31
32-43
44-54
20-31
32-43
44-54
10-21
22-33
34-45
46-56
13-24
25-36
37-45
9-20
21-32
33-41
5-16
17-28
29-35
13-26
NA
13-24
25-28
NA
LCRS Flow
Avg.
(Iphd)
2,245
5,223
3,975
2,732
3,740
2,337
789
259
159
103
435
300
161
311
314
268
937
438
407
2,513
NA
13,003
1,010
NA
Peak
(Iphd)
5,754
6,845
7,464
5,393
5,707
3,982
1,419
780
286
200
859
610
464
671
752
987
2,055
622
686
10,440
NA
44,895
2,413
NA
LDS Flow
Avg.
(Iphd)
266
424
892
404
996
665
123
89
27
40
9
22
7
2
33
16
70
51
26
28
NA
7
283
NA
Peak
(Iphd)
499
808
1,426
605
1,690
1,102
304
170
128
227
31
125
14
9
276
103
147
92
29
115
NA
73
341
NA
^am
(%)
88.14
91.87
77.55
85.20
73.36
71.54
84.40
65.52
83.08
61.27
98.03
92.71
95.40
99.49
89.56
94.02
92.52
88.39
93.71
98.88

99.95
71.97

/3\
Post-Closure Period^ '
Time
Period
;months)
55-66
67-78
79-90
91-102
103-114
55-66
67-78
NA
NA
NA
NA
NA
8-19
20-26
27-38
39-50
NA
NA
LCRS Flow
Avg.
(Iphd)
317
703
1,146
1,306
510
493
337
NA
NA
NA
NA
NA
ND
ND
376
715
NA
NA
Peak
(Iphd)
670
1,877
1,956
1,943
718
1,040
654
NA
NA
NA
NA
NA
ND
ND
1,455
1,352
NA
NA
LDS Flow
Avg.
(Iphd)
106
267
279
326
74
154
328
NA
NA
NA
NA
NA
102
1
5
64
NA
NA
Peak
(Iphd)
222
1,134
451
612
97
393
514
NA
NA
NA
NA
NA
886
10
70
156
NA
NA
^am
(%)
66.48
62.02
75.64
75.01
85.41
68.83
2.80





98.58
91.05


en
CO
-------
Table 5-2.  Summary of LCRS and LDS Flow Rate Data for Landfill Cells with GM Top Liners (cont).
Cell
No.



E1


E2


E3
E4
F1

G1





G2


M(6)







|2<6)






Cell
Area


(hectare)
2.4


2.4


1.2
1.2
1.8

3.0





1.6


3.2/2.7<7)







4.2/2.3<7)






Start of
Waste
Place.
(month-
year)
3-88


10-87


5-90
7-90
7-92

6-89





6-89


8-87







10-87






End of
Final
Closure
(month-
year)
NA


NA


NA
NA
NA

NA





NA


10-94







10-94






Initial Period of O
Time
Period

(months)
1-7


1-12


1-12
1-12
1-12

1-12





1-12


1-5
6-8






1-7






LCRS Flow<4)
Avg.

(Iphd)
ND


ND


9,425
20,148
14,472

22,371





22,371


ND
ND






6,627






Peak

(Iphd)
ND


ND


25,394
55,785
45,010

46,120





46,120


ND
ND






13,959






Deration
LDS Flow
Avg.

(Iphd)
2,144


483


1,595
996
124

ND





197


234
ND






31






Peak

(Iphd)
5,026


3,518


1,951
2,362
479

ND





645


508
ND






77






Earn
(%)








83.08
95.06
99.14







99.12










99.53






Active Period of Operation
Time
Period

(months)
8-19
20-31
32-40
13-24
25-36
37-45
13-14
NA
13-24
25-30
13-24
25-36
37-42
43-51
52-63
64-67
13-24
25-36
37-42
9-15
16-32
33-44
45-48
49-54
55-66
67-78
79-84
8-24
25-36
37-40
41-46
47-58
59-70
71-76
LCRS Flow
Avg.

(Iphd)
8,432
11,521
6,525
5,821
4,547
4,434
6,062
NA
9,000
7,826
12,893
3,438
8,356
ND
ND
ND
12,893
3,438
8,356
16,224
ND
7,167
231
ND
624
541
904
ND
1,030
427
ND
624
541
904
Peak

(Iphd)
19,614
36,164
13,075
10,445
11,014
6,830
9,038
NA
25,450
10,932
23,485
1 1 ,652
10,303
ND
ND
ND
23,485
1 1 ,652
10,303
48,932
ND
22,020
332
ND
1,580
752
1,827
ND
3,241
1,054
ND
1,580
752
1,827
LDS Flow
Avg.

(Iphd)
1,436
1,051
743
802
685
596
1,603
NA
66
67
ND
156
101
121
74
49
37
35
60
5
ND
10
4
ND
2
13
79
ND
5
6
ND
8
8
5
Peak

(Iphd)
3,069
1,915
1,015
2,447
1,404
999
1,758
NA
83
77
ND
238
116
384
139
64
65
42
100
18
ND
44
10
ND
5
42
157
ND
35
11
ND
37
23
6
Earn
(%)


82.97
90.88
88.61
86.22
84.93
86.56
73.56

99.27
99.15

95.47
98.79



99.71
98.98
99.28
99.97

99.86
98.49

99.68
97.60
91.26

99.52
98.67

98.67
98.54
99.49
Post-Closure Period'3'
Time
Period

'months;
NA


NA


NA
NA
NA

NA





NA


85-93







77-85






LCRS Flow
Avg.

(Iphd)
NA


NA


NA
NA
NA

NA





NA


800







800






Peak

(Iphd)
NA


NA


NA
NA
NA

NA





NA


1,794







1,794






LDS Flow
Avg.

(Iphd)
NA


NA


NA
NA
NA

NA





NA


62







2






Peak

(Iphd)
NA


NA


NA
NA
NA

NA





NA


119







4






Earn
(%)





















92.25







99.71






-------
        Table 5-2. Summary of LCRS and LDS Flow Rate Data for Landfill Cells with GM Top Liners (cont).
Cell
No.



I3(6)






K1




N2



01(8)




02(8)



S1



S2


Cell
Area


(hectare)
3.4/1 .8(7)






2.7




6.3



4.2




4.9



2.0



1.6


Start of
Waste
Place.
(month-
year)
4-88






12-89




1-92



9-88




3-89



9-90



8-90


End of
Final
Closure
(month-
year)
10-94






NA




NA



NA




NA



NA



NA


Initial Period of O
Time
Period

(months)
1-7






1-12




1-12



1-6




1-12



1-10



1-9


LCRS Flow<4)
Avg.

(Iphd)
1 1 ,559






17,808




ND



ND




4,407



2,226



2,185


Peak

(Iphd)
21,081






24,832




ND



ND




9,826



5,081



4,650


Deration
LDS Flow
Avg.

(Iphd)
37






122




ND



293




6



12



5


Peak

(Iphd)
87






163




ND



620




24



39



24


^am
(%)


99.68






99.31













99.86



99.45



99.78


Active Period of Operation
Time
Period

(months)
8-24
25-36
37-40
41-46
47-58
59-70
71-76
13-24
25-36
37-48
49-60
61-66
13-19
20-31
32-34
35-39
7-18
19-30
31-42
43-54
55-64
13-24
25-36
37-48
49-59
11-22
23-28
29-40
41-45
10-17
18-33
34-46
LCRS Flow
Avg.

(Iphd)
ND
1 1 ,684
2,464
ND
624
541
904
12,929
10,879
6,155
5,952
9,494
4,547
2,561
6,399
2,741
4,407
4,023
7,089
6,201
8,661
4,023
7,089
6,201
8,661
653
ND
1,571
1,086
654
ND
1,255
Peak

(Iphd)
ND
26,339
4,666
ND
1,580
752
1,827
27,663
17,683
11,331
8,024
12,245
5,741
3,460
7,274
3,170
9,826
13,231
16,467
12,561
15,327
13,231
16,467
12,561
15,327
1,220
ND
4,074
2,067
1,135
ND
3,638
LDS Flow
Avg.

(Iphd)
ND
7
5
ND
4
13
17
88
76
514
349
282
113
203
786
201
0
3
0
1
3
2
1
3
1
38
ND
8
4
5
ND
5
Peak

(Iphd)
ND
23
8
ND
17
55
53
180
104
892
495
378
468
669
1,058
406
3
7
5
6
9
5
4
11
5
68
ND
26
7
24
ND
8
^am
(%)



99.94
99.80

99.39
97.64
98.14
99.32
99.30
91.64
94.14
97.03
97.52
92.08
87.72
92.65
99.99
99.93
99.99
99.98
99.97
99.95
99.98
99.96
99.99
94.18

99.51
99.64
99.20

99.63
/3\
Post-Closure Period^ '
Time
Period

;months;
77-85






NA




NA



NA




NA



NA



NA


LCRS Flow
Avg.

(Iphd)
800






NA




NA



NA




NA



NA



NA


Peak

(Iphd)
1,794






NA




NA



NA




NA



NA



NA


LDS Flow
Avg.

(Iphd)
3






NA




NA



NA




NA



NA



NA


Peak

(Iphd)
12






NA




NA



NA




NA



NA



NA


^am
(%)


99.57































en
01
-------
Table 5-2. Summary of LCRS and LDS Flow Rate Data for Landfill Cells with GM Top Liners (cont).
Cell
No.
V1(8)
V2(8)
W1
W2
X1
Cell
Area
(hectare)
4.2
3.9
15.4
15.4
3.0
Start of
Waste
Place.
(month-
year)
1-90
1-90
5-92
5-92
8-92
End of
Final
Closure
(month-
year)
NA
NA
NA
NA
NA
Initial Period of O
Time
Period
(months)
1-10
1-10
1-8
9-12
1-8
1
2-7
LCRS Flow(4)
Avg.
(Iphd)
13,622
13,622
ND
7,492
ND
1 1 1 ,031
32,469
Peak
(Iphd)
49,828
49,828
ND
8,799
ND
111,031
104,645
Deration
LDS Flow
Avg.
(Iphd)
117
135
ND
439
ND
364
4
Peak
(Iphd)
153
256
ND
765
ND
364
25
^am
(%)
99.14
99.01
94.14

99.67
99.99
Active Period of Operation
Time
Period
(months)
NA
NA
13-24
25-35
9-20
21-32
33-35
8-19
20-33
LCRS Flow
Avg.
(Iphd)
NA
NA
2,693
943
4,288
4,813
719
5,926
2,188
Peak
(Iphd)
NA
NA
6,365
1,572
9,389
10,524
2,141
14,315
5,376
LDS Flow
Avg.
(Iphd)
NA
NA
34
19
594
204
32
5
0
Peak
(Iphd)
NA
NA
109
44
1,826
1,217
52
45
2
^am
(%)


98.72
97.98
86.15
95.76
95.50
99.92
99.99
/3\
Post-Closure Period^ '
Time
Period
'months)
NA
NA
NA
NA
NA
LCRS Flow
Avg.
(Iphd)
NA
NA
NA
NA
NA
Peak
(Iphd)
NA
NA
NA
NA
NA
LDS Flow
Avg.
(Iphd)
NA
NA
NA
NA
NA
Peak
(Iphd)
NA
NA
NA
NA
NA
^am
(%)





Notes:
(1) "Initial Period of Operation" represents period after waste placement has started and not more than a few lifts of waste and daily cover have been
    placed in the cell (i.e., no intermediate cover).
(2) "Active Period of Operation" represents period when waste thickness in cell is significant and/or an effective intermediate cover is placed on the waste.
(3) "Post-Closure Period" represents period after final cover system has been placed on the entire cell.
(4) Flow rates are given in liter/hectare/day.
(5) NA =  not applicable; ND = not determined.
(6) LCRS for Cells 11, 12, and I3are combined after February 1992. The measure average flow rates are assumed to represent flow rates for the three cells.
(7) Values given represent LCRS and LDS areas, respectively.
(8) LCRS flows are combined for Cells O1 and O2 and for Cells V1 and V2. Measured average flow rates are assumed to represent flows for the two cells at each landfill.
-------
       will not achieve this performance level and lower Et values, in the range of about
       90 to 99%, will occur. This relatively broad range of Et values is a consequence
       of the potential for even appropriately installed GMs to have an occasional small
       hole, typically due to an imperfect seam, but also potentially due to a
       manufacturing or construction-induced defect not identified by the CQA
       program. Leakage can occur, relatively unimpeded, through a GM hole if the
       GM is not underlain by a low-permeability material such as a CCL or GCL. If a
       hole occurs at a critical location where a hydraulic head exists, such as in a
       landfill sump, the leakage rate through the hole can be significant. In contrast,
       the GCL or CCL component of a  composite liner can impede flow through a GM
       hole, even if it occurs at a critical location.
   •   The conclusion to be drawn from the above data evaluation is that single liner
       systems with GM liners (installed on top of a relatively permeable subgrade)
       should not be used in applications where Et values as low as 90% would be
       unacceptable, even if a thorough CQA program is employed. In these cases,
       single-composite liner systems or double-liner systems should be utilized.  An
       exception to this conclusion may  be made for certain facilities, such as surface
       impoundments or small, shallow landfill cells, with GM primary liners that can be
       field tested over the GM sheet and  seams using electrical leak location  surveys,
       ponding tests, or other methods.  For these facilities, higher efficiencies (i.e.,
       greater than 99%) may be achieved with GM liners by identifying and repairing
       the GM holes during construction and, especially for surface impoundments,
       during operation.  In  all cases, GM  liners should be manufactured and installed
       using formal quality assurance programs.

GM/GCL Composite Liners
The performance of all 28 cells with GM/GCL composite primary liners was assessed.
Flow rate data are available  for the 28 cells at seven landfills with monitoring periods of
up to 83 months. All of these cells were constructed with formal CQA programs.  A
summary of the LCRS and LDS flow rate data for the cells is provided in Table  5-3 and
detailed results are given in Table E-4.10 of Appendix E. The major findings from  the
data evaluation of cells with  GM/GCL primary liners (excluding cell 14, which may have
surface-water infiltration into the LDS at the anchor trench) are given below:

   •   LDS flows during the initial  period of operation are attributed primarily to
       construction water. LDS flows during the active and post-closure periods are
       attributed primarily to primary liner leakage and compression water.
   •   Average monthly LDS flow  rates ranged from about 0 to 290 Iphd during the
       initial period of operation, 0 to 11  Iphd during the active period, and 0 to 2 Iphd
       (with many values reported as zero) during the post-closure period.  Peak
       monthly flow rates were typically below 200 Iphd and exceeded 500 Iphd in only
       four of the 28 cells.
   •   Based on the above  data, average  monthly active-period LDS flow rates
       attributable to leakage through  GM/GCL primary liners constructed with CQA will
       often be less than 2 Iphd, but occasionally in excess of 10 Iphd.
                                      5-17
-------
         Table 5-3.  Summary of LCRS and LDS Flow Rate Data for Landfill  Cells with GM/GCL Composite Primary Liners.
Cell
No.

C6
14
15
AW1
AW2
AX1
AX2
AX3
AX4
AX5
AX6
AX7
AX8
AX9
AX10
AX11
AX12
AX13
AX14
AX15
AX16
AY1
AY2
AYS
AZ1
BB1
BB2
BBS
Cell
Area
(ha)
3.6
4.7
4.7
2.4
2.4
2.0
2.0
1.7
1.7
2.8
3.9
2.6
3.8
3.3
3.9
3.0
4.0
3.0
2.8
2.8
4.5
1.3
1.0
1.0
3.8
4.0
2.4
2.8
Waste
Place.
Start
Date
Aug-93
May-92
Jul-92
May-93
Aug-93
Jul-88
Jul-88
Sep-88
Sep-88
Oct-88
Dec-88
Feb-89
Jul-89
Dec-89
Jul-90
Feb-90
Oct-90
Jan-91
Apr-91
May-92
Jan-93
Oct-94
Aug-94
Aug-94
Dec-92
Feb-91
Jan-93
Jan-93
Final
Closure
Date

NA
Jul-94
May-94
NA
NA
Feb-91
Feb-91
Apr-93
Apr-93
NA(5)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Initial Period of Operation'1'
Time
Period(6)
(months)
1-10
1-12
1-12
1-12
1-10
1-2
1-5
1-5
1-12
1-11
1-9
1-10
1-14
1-9
1-7
1-16
1-12
1-7
1-11
1-12
1-10
1-9
1-11
1-11
2-12
1-6
1-11
1-11
LCRS Flow(4)
Avg.
(Iphd)
3,273
4,494
3,938
6,358
3,553
16,718
15,521
3,366
2,534
1,384
3,759
5,376
4,881
1,047
2,786
4,675
3,494
6,683
2,777
5,573
8,601
6,803
10,964
12,198
4,093
10,378
ND(5)
ND
Peak
(Iphd)
12,155
17,251
7,985
20,570
7,480
19,738
58,671
7,985
12,688
3,394
7,171
12,155
21,038
3,478
13,698
14,586
8,836
14,343
6,582
1 1 ,809
17,756
12,439
23,914
32,326
5,219
22,130
ND
ND
LDS Flow
Avg.
(Iphd)
178
24
2
131
290
0
15
35
101
37
53
34
48
1
0
0
0
0
0
0
0
0
3
6
0
15
1
0
Peak
(Iphd)
823
70
11
524
514
0
45
151
860
92
93
47
189
7
0
0
0
0
0
0
0
0
12
28
0
65
12
0
Ea

94.57
99.47
99.95
97.94
91.84
100
99.90
98.97
96.01
97.30
98.58
99.37
99.02
99.91
100
100
100
100
100
100
100
100
99.97
99.95
100
99.86

Active Period of Operation'2'
Time
Period*6'
(months)
11-17
13-26
13-21


3-33
6-33
6-56
13-56
12-80
10-80
11-76
15-71
10-65
8-59
17-62
13-56
8-53
12-38
13-37
11-29



13-31
7-47
12-23
12-23
LCRS Flow
Avg.
(Iphd)
393
2,041
3,108


540
281
307
75
56
168
234
439
41
374
150
803
1,408
281
299
819



3,473
2,494
5,422
2,284
Peak
(Iphd)
1,403
4,282
1 1 ,669


2,383
570
1,075
187
191
655
851
1,384
159
645
337
3,029
9,294
449
561
5,096



5,054
8,983
14,042
7,945
LDS Flow
Avg.
(Iphd)
3
26
11


0
2
4
1
2
0
2
0
0
0
0
0
0
0
0
0



2
6
0
0
Peak
(Iphd)
15
142
54


0
21
47
13
37
0
9
0
0
0
0
0
0
0
0
0



22
25
0
1
Ea

99.29
98.73
99.65


100
99.33
98.78
98.75
96.67
100
99.20
100
100
100
100
100
100
100
100
100



99.94
99.78
100
100
Post-Closure Period13'
Time
Period*6'
(months)

27-36
22-34


34-83
34-83
57-81
57-81

















LCRS Flow
Avg.
(Iphd)

567
189


66
178
206
47

















Peak
(Iphd)

1389
779


94
421
458
84

















LDS Flow
Avg.
(Iphd)

59
2


0
0
1
0

















Peak
(Iphd)

133
8


0
0
10
0

















Ea


89.59
98.78


100
100
99.55
100

















Ul
oo
         Notes:
         (1) "Initial Period of Operation" represents period after waste placement has started and only a small amount of waste has been placed in the cell.
         (2) "Active Period of Operation" represents period when waste thickness in cell is significant and/or an effective intermediate cover is placed on the waste.
         (3) "Post Closure Period" represents period after final cover system has been placed on the entire cell.
         (4) Flow rates are given in liter/hectare/day.
         (5) NA = not applicable; ND = not determined.
         (6) Breakthrough time for steady-state saturated flow through GCL component of composite liner is estimated to be 2 months based on a calculation using Darcy's equation and a saturated hydraulic conductivity of
            5 x 10~11 m/s, hydraulic gradient of 5, and effective porosity of 0.2.  For this calculation, it is assumed that flow through the GM component of the composite liner occurs through small holes and is instantaneous.
-------
The above data indicate that GM/GCL composite liners can be constructed to achieve
Et values of 99.9% or more.  However, Et values in the range of 99 to 99.9% will also
occur.  These high efficiencies demonstrate that the GCL component of a GM/GCL
composite liner is effective in impeding leakage through holes in the GM component of
the liner.

GM/CCL  and GM/GCL/CCL Composite Liners
The performance of 13 of the 88 cells with GM/CCL or GM/GCL/CCL primary liners was
assessed. The remaining 75 cells with GM/CCL or GM/GCL/CCL primary liners were
generally excluded from the assessment because: (i) they did not have continuous
LCRS and LDS flow rate data available for an individual cell from the start of operation;
or (ii) there were insufficient LCRS and LDS chemical constituent data to evaluate
whether primary liner leakage did or did not occur. Flow rate data are available for 13
cells at nine landfills with monitoring periods of up to 121 months.  All of these cells
were constructed with CQA.  A summary of the LCRS and LDS flow rate data for the
cells is provided in Table 5-4. The main findings from the evaluation of flow rate data
for cells with GM/CCL or GM/GCL/CCL primary liners are given below:

   •   LDS flows during the initial period of operation are attributed primarily to
       construction water. LDS flows during the active and post-closure periods are
       attributed primarily to consolidation water.
   •   Average monthly LDS flow rates ranged from about 10 to 1,400  Iphd during the
       initial period of operation, 0 to 370  Iphd during the active period,  and 5 to 210
       Iphd during the post-closure period.

Given the "masking" effects of consolidation water, chemical constituent data must be
used to assess the hydraulic performance of composite primary liners having a CCL or
GCL/CCL lower component,  as described  in Section 5.2.3.2. This approach was
applied to the 13 landfill cells. Concentrations of chemical constituents in LDS liquids
were compared to concentrations of the same constituents in LCRS liquids.  These
chemical  data are reported in Table E-4.9 of Appendix E. The general water quality
characteristics of LDS liquids were found to be different than the corresponding
characteristics for the LCRS  liquids.  This is due to the different origins of the primary
sources of the  two liquids: leachate for LCRS  liquids and CCL pore water for LDS
liquids. The different origins  of the two liquids are reflected in different major ion
chemistries, as well as differences in chemical oxygen demand (COD), biological
oxygen demand (BOD), and total organic carbon (TOC) concentrations.

To further evaluate whether primary liner leakage had contributed to the LDS flows, the
concentrations of the previously mentioned five key chemical constituents (i.e., sulfate,
chloride, benzene, toluene, and xylene) in LCRS and LDS liquids were investigated.
The results of the comparison of key constituents are presented in Tables 5-5 and 5-6.
Table 5-5 presents the concentrations of the five key constituents as a function of time
                                     5-19
-------
       Table 5-4. Summary of LCRS and LDS Flow Rate Data for Landfill Cells with GM/CCL or GM/GCL/CCL
                Composite Primary Liners.
Cell
No.


B3(b)







Y2



AD1







AD7




AK1
AL1

AM1




Cell
Area

(ha)
6.4







3.0



0.6







1.5




1.4
14.9

3.2/2.4(7)




Waste
Place.
Start
Date
Jul-87







Jan-91



May-85







Sep-87




Oct-93
1990

Oct-90




Final
Closure
Date

NA(6)







NA



Jul-88







Oct-93




NA
NA

NA




Initial Period of Operation01
Time
Period
(months)
1-4







1-10



1-12







1-12




1-12
1-29

1-9




LCRS Flow(4)
Avg.
(Iphd)
15,304







23,368



ND(6)







12,597




9,867
ND

ND




Peak
(Iphd)
24,858







36,791



ND







26,492




17,986
ND

ND




LDS Flow
Avg.
(Iphd)
1,394







655



ND







135




206
ND

ND




Peak
(Iphd)
4,250







1,768



ND







1,101




804
ND

ND




Active Period of Operation (2)
Time
Period
(months)
5-16
17-28
29-40
41-52
53-64
65-76
77-88
89-93
11-22
23-34
35-46
47-54
13-20
21-32






13-24
25-36
37-48
49-60
61-69

30-41
42-54
10-21
22-33
34-45
46-57
58-69
LCRS Flow
Avg.
(Iphd)
5,700
9,272
7,575
2,859
1,189
403
560
578
10,353
11,344
4,404
4,397
ND
373






2,212
1,539
1,429
249
480

934
1,349
270
236
111
20
18
Peak
(Iphd)
8,935
22,444
13,978
6,043
2,280
490
919
648
19,204
25,309
6,380
5,199
ND
892






2,857
2,755
2,813
629
614

2,085
5,885
533
329
283
77
21
LDS Flow
Avg.
(Iphd)
124
101
262
231
45
92
102
98
370
90
70
48
ND
107






71
96
17
33
64

231
103
15
10
3
1
1
Peak
(Iphd)
266
168
803
713
152
133
193
109
1,993
168
248
56
ND
603






291
393
21
74
112

367
183
64
15
14
1
1
Ea


(%)
97.8
98.9
96.5
91.9
96.2
77.3
81.8
83.0
96.4
99.2
98.4
98.9

71.4






96.8
93.8
98.8
87.0
86.6

75.3
92.4
94.4
95.8
97.3
95.0
94.4
Post-Closure Period'3'
Time
Period
(months)












33-44
45-51
52-63
64-75
76-87
88-99
100-111
112-121
70-81
82-87











LCRS Flow
Avg.
(Iphd)












145
85
3
3
3
1
1
2
375
165











Peak
(Iphd)












652
130
22
42
21
4
2
9
533
334











LDS Flow
Avg.
(Iphd)












24
26
28
42
23
8
5
6
73
105











Peak
(Iphd)












42
31
45
103
68
46
43
24
157
172











Ea


(%)












83.4
69.5
-833
-1300
-667
-700
-400
-200
80.5
36.3











CJ1

o
-------
          Table 5-4.  Summary of LCRS and  LDS Flow Rate Data for Landfill Cells with GM/CCL or GM/GCL/CCL
                         Composite Primary Liners (Continued).
Cell
No.


AM1
AM2





AO1


AO2

AQ1



AQ10


AR1

Cell
Area

(ha)

4.8/2.4(7)





1.8


1.8

0.6



0.9


9.7

Waste
Place.
Start
Date

Oct-90





Jan-92


Jul-92

Mar-86



Jan-89


Mar-92

Final
Closure
Date


NA





NA


NA

early 90



mid 91


NA

Initial Period of Operation (1)
Time
Period
(months)

1-9





1-5


1-5

1-6



1-9


1-11

LCRS Flow(4)
Avg.
(Iphd)

ND





ND


15,881

10,203



ND


27,042

Peak
(Iphd)

ND





ND


24,541

18,944



ND


65,871

LDS Flow
Avg.
(Iphd)

ND





ND


149

352



14


292

Peak
(Iphd)

ND





ND


191

569



32


705

Active Period of Operation (2)
Time
Period
(months)
70-81
10-21
22-33
34-45
46-57
58-69
70-81
6-17
18-29
30-37
6-17
18-31
7-25
26-34
35-46
47-58
10-14
15-26

12-23
24-36
LCRS Flow
Avg.
(Iphd)
11
32
35
17
67
64
112
1,984
1,299
1,144
3,027
1,688
ND
ND
ND
4,530
ND
15,933

11,251
9,668
Peak
(Iphd)
18
154
51
45
274
181
136
4,130
1,577
1,371
5,266
2,383
ND
ND
ND
10,531
ND
38,751

23,384
26,274
LDS Flow
Avg.
(Iphd)
5
9
9
3
0
8
9
184
96
60
110
33
255
ND
197
116
26
48

181
155
Peak
(Iphd)
8
42
29
26
0
13
13
353
126
102
158
64
1239
ND
435
143
32
250

470
442
Ea


(%)
54.4
71.9
74.3
82.4
100
87.5
92.0
90.7
92.6
94.8
96.4
98.1



97.4

99.7

98.4
98.4
Post-Closure Period13'
Time
Period
(months)












59-65
66-77
78-89
90-97
27-38
39-50
51-63


LCRS Flow
Avg.
(Iphd)












5,835
644
1,367
1,615
682
300
852


Peak
(Iphd)












1 1 ,244
1,011
3,264
3,575
2,251
1,709
1,588


LDS Flow
Avg.
(Iphd)












215
117
98
51
29
18
24


Peak
(Iphd)












246
165
132
118
48
63
75


Ea


(%)












96.3
81.8
92.8
96.8
95.7
94.0
97.2


en
ro
          Notes:
          (1) "Initial Period of Operation" represents period after waste placement has started and only a small amount of waste has been placed in the cell.
          (2) "Active Period of Operation" represents period when waste thickness in cell is significant and/or an effective intermediate cover is placed on the waste.
          (3) "Post-Closure Period" represents period after final cover system has been placed on the entire cell.
          (4) Flow rates are given in liter/hectare/day.
          (5) 65 percent of Cell B3 received final cover at 60 months after start of waste placement.
          (6) NA = not applicable;  ND = not determined.
          (7) Values given represent LCRS and LDS areas, respectively.
          (8) Breakthrough times for steady-state saturated flow through CCL or GCL/CCL component of composite liners are estimated to be 2 to 145 months based on a
             calculation using Darcy's equation and specified hydraulic conductivities, hydraulic gradient of 5 forGCLs and 1 forCCLs, and effective porosity of 0.2.
             For this calculation, it is assumed that flow through the GM component of the composite liner occurs through  small holes and is instantaneous.
-------
            Table 5-5. Average Concentrations of Five Key Chemicals in LCRS and LDS Flows from

                     Landfill Cells with GM/CCL and GM/GCL/CCL Composite Primary Liners.
Ul

ro
ro

Cell
No.
B3








Y2




AD1










Time
Period
(months)
1-4
5-16
17-28
29-40
41-52
53-64
65-76
77-88
89-93
1-10
11-22
23-34
35-46
47-54
1-12
13-20
21-32
33-44
45-51
52-63
64-75
76-87
88-99
100-111
112-121
Chemical'1'
Sulfate (mg/l)
LCRS
282
105
348
104
47
28
<127
<6

108
108
52


6,353
5,830
5,470
4,455
2,223


1,785
4,488
3,633
3,870
LDS
95
1,286

500
14
123
90
301
48
231
299
326




480
498
308
443
338
456
339
296
369
Chloride (mg/l)
LCRS
25
207
352
580
355
899
203
998
1,383
349
590
876


3,930
24,300
10,763
11,590
13,960


13,900
14,550
14,075
14,800
LDS
19
173
241
118

59
58


60
25
89



289
210
185
217
240
131
377
137
114
138
Benzene (ug/l)
LCRS
<11(2)
<1
<1
<5
8
7
<5
<5
<2

10



492
429
33
5
35



26
18
<25
LDS
<25
<1

<5
6
<1
<5
<5
<5






<4
<4
<4
<4
<4
<4
<4
<4
<1
<1
Toluene (ug/l)
LCRS
150
<1
<1
354
233
101
14
<5
<1

720



305
292
133
60
108
<300


186
<30
<25
LDS
<25
<1

<6
24
<1
<6
<4
<1

7




<6
<6
<6
<6
<6
<6
<6
<6
<1
<1
Xylene (ug/l)
LCRS

























LDS

























-------
           Table 5-5. Average Concentrations of Five Key Chemicals in LCRS and LDS Flows

                     from Landfill Cells with GM/CCL and GM/GCL/CCL Composite Top Liners (Continued).
en

*>
CO

Cell
No.
AD7







AK1
AL1


AM1





AM2





A01



Time
Period
(months)
1-12
13-24
25-36
37-48
49-60
61-69
70-81
82-87
1-12
1-29
30-41
42-54
1-9
10-21
22-33
34-45
46-57
58-69
1-9
10-21
22-33
34-45
46-57
58-69
1-5
6-17
18-29
30-37
Chemical
Sulfate (mg/l)
LCRS
2,818
3,620
7,361
8,213
6,867
5,740
6,998
7,480
47
300
225
247
51
<2
<16
<27
<12
<3
96
<2
<13
<7
<2
<3

49
35
69
LDS
340
683
586
954
1,050
1,148
1,168
1,132
16
1,030
900
1,375


1,341
1,200




1,032
1,300
1,730
2,405

88
41
2
Chloride (mg/l)
LCRS
3,214
9,550
10,720
11,535
14,400
15,775
12,875
14,267
104
330
273
400
77
120
159
219
265
240
140
290
353
326
368
262

930
988
570
LDS
109
216
219
469
418
387
387
357
2
89
203
215


2,260
2,600




2,175
2,600
2,700
2,635

58
27
46
Benzene (ug/l)
LCRS
140
612
1168
644
778
687
540
<240
<5
<4
1
2
<21
18
<19
18
14
13
11
17
19
18
15
7

6
10
<5
LDS
<4
<4
<4
<4
<4

<4
<1
<1
<2
<1
<1

<1
<1
<1



<1
<1
<1
<1
<1

<1
<1
<5
Toluene (ug/l)
LCRS
317
892
1,859
2,960
1,660
1,288
906
450
88
133
5
2
219
160
336
290
199
56
22
89
266
286
148
65

230
288
77
LDS
<6
<6
<6
<6
<6
<6
<1
<1
<1
<1(3)
<1
<1

2
<1
<1



<1
<1
<1
<1
<1

<1
44
<5
Xylene (ug/l)
LCRS








30
540
<3
<3
150
90
121
122
90
82
34
57
95
94
115
50

59
45
21
LDS








<3
<1

<1

<1
<1
<3



<2
<1
<3
<2
<2

<3
<3
<10
-------
en
                Table 5-5.  Average Concentrations of Five Key Chemicals in LCRS and LDS Flows
                            from Landfill Cells with GM/CCL and GM/GCL/CCL Composite Top Liners (Continued).

Cell
No.
A02


AR1


AQ1




AQ10




Time
Period
(months)
1-5
6-17
18-31
1-11
12-23
24-36
1-58
59-65
66-77
78-89
90-97
1-15
15-26
27-38
39-50
51-63
Chemical
Sulfate (mg/l)
LCRS

49
35
180
440
520










IDS

89
93
600
170
265










Chloride (mg/l)
LCRS

930
988
1,000
2,200
1,650










IDS

16
24
49
8
41










Benzene (u,g/l)
LCRS

6
10


<100

<5
<8
<12
<5

<10
<10
<6
<5
LDS

<1
<1

<1
<50

<4
<4
<4
<4

<4
<4
<4
<4
Toluene (u,g/l)
LCRS

230
288


<100

<5
<5
<5
<5

<12
<30
<6
5
LDS

<1
<1

<2
<50

<10
<6
<14
<5

<5
<5
<5
<5
Xylene (ug/l)
LCRS

59
45


<100


<5
<8
<5


190
<7
10
LDS

<3
<4

<4
<50










                Notes:
                (1)  Reported concentrations represent average of 1 to 17 individual analysis results (typically on the order of 5) during incremental reporting period.
                (2)  Data preceded by "<" indicates more than half of analysis results for parameter were reported as non-detects; in calculating
                   average values, half of the test detection limit was conservatively used for all results reported as non-detects.
                (3)  For Cell AL1, toluene was not detected in nine often LDS flow samples obtained during the 1-41 months time period.  Toluene was detected
                   at a concentration of 91 |ig/l in month 30. This one detection is attributed to sampling or analysis error and is not included in the average.
-------
      Table 5-6.  Evaluation of Chemical Constituent Migration Through Landfill GM/CCL and
                GM/GCL/CCL Composite Primary Liners.

Cell
No.



B3







Y2









Monitor.
Period
(months)


93







54
(no key
chemical
data
after 34
months)



Estimated
Advective
Breakthr.
Time for
GCL/CCL
(months)'1'
46







35








Chemical
Sulfate




not diagnostic due
to fluctuating Co(2)
in both LCRS and
LDS




not diagnostic due
to high C0 in LDS
consolidation
water





Chloride




lower C0 in LDS
than in LCRS and
trend of
decreasing LDS
C0 with time not
indicative of
chloride
breakthrough
in LCRS, C0 =
170to1,160mg/l
with m(2) = 628
mg/l; in LDS, C0 =
8 to 1 40 mg/l with
m = 58 mg/l; no
indication of
chloride
breakthrough
Benzene




not diagnostic
due to very low
C0 in both LCRS
and LDS (i.e., C0
almost always
below DL(3) of 5
H9/I)

no LDS data
available







Toluene




in LCRS, C0 up to
700 |ig/l; in LDS,
C0 typically below
DLof 1 to 10
|ig/l; no indication
of toluene
breakthrough

only one C0
available from
each system (at
11 -22 months):
LCRS C0 = 720
|ig/l and LDS C0
= 7 ug/l
f-^y

Xylene




no data
available






no data
available








Summary of
Observations for Five
Key Constituents


no evidence of
significant leachate
migration into LDS
after almost 8 years of
cell operation, twice
the estimated CCL
breakthrough time

data are insufficient to
draw conclusions;
monitoring period is
about equal to
estimated CCL
breakthrough time;
more chemical data
are needed

en
en
-------
      Table 5-6.  Evaluation of Chemical Constituent Migration Through Landfill GM/CCL and
                GM/GCL/CCL Composite Primary Liners.
Cell
No.
AD1









AD7













Monitor.
Period
(months)
121









87













Estimated
Advective
Breakthr.
Time for
GCL/CCL
(months)'1'
70









70













Chemical
Sulfate
in LCRS, C0 =
1,785 to 6,353
mg/l with m =
4,234 mg/l; in
LDS, C0 = 296 to
498 mg/l with m =
392 mg/l; no
indication of
sulfate
breakthrough
in LCRS,
C0=2,818to8,213
mg/l with m=6,137
mg/l; in LDS,
C0=340to 1,168
mg/l with m=882
mg/l; increasing
LDS C0 after 36
months attributed
to decreasing
dilution of consol.
water by
construct, water

Chloride
in LCRS, C0 =
3,930 to 24,300
mg/l, with m =
13,450 mg/l; in
LDS, C0 = 114 to
337 mg/l, with m =
204 mg/l; no
indication of
chloride
breakthrough
in LCRS,
C0=3,214to
15,775 mg/l with
m=1 1,547 mg/l; in
LDS, C0=109to
469 mg/l with
m=320 mg/l;
increasing LDS 0,
after 36 months
attributed to
decreasing dilutior
of consol. water
by construct.
water
Benzene
in LCRS, C0 =
<25to492|ig/l;
in LDS, C0 below
DLof 1 to4|ig/l;
no indication of
benzene
breakthrough



in LCRS, C0 =
<240to 1,168
|ig/l; in LDS, C0
below DL of 1 to
4 |ig/l; no
indication of
benzene
breakthrough






Toluene
in LCRS, C0 =
<25to305|ig/l;
in LDS, C0 below
DLof 1 to6|ig/l;
no indication of
toluene
breakthrough



in LCRS, C0 =
317to2,960|ig/l;
in LDS, C0 below
DLof 1 to6|ig/l;
no indication of
toluene
breakthrough







Xylene
no data
available








no data
available












Summary of
Observations for Five
Key Constituents
no evidence of
significant leachate
migration into LDS
after 10 years of cell
operation and closure,
1 .7 times more than
the estimated CCL
breakthrough time


evidence of possible
breakthrough for
sulfate & chloride at 12
36 months; authors
attribute trend to
decreased dilution of
consolidation water by
construction water; no
evidence of organic
constituent
breakthrough; more
chemical data are
needed

en
en
-------
      Table 5-6.  Evaluation of Chemical Constituent Migration Through Landfill GM/CCL and
                GM/GCL/CCL Composite Primary Liners.

Cell
No.



AK1









AL1









Monitor.
Period
(months)


12









54








Estimated
Advective
Breakthr.
Time for
GCL/CCL
(months)'1'
48









70








Chemical
Sulfate




in LCRS, C0 = 7
to 1 1 0 mg/l with m
= 47 mg/l; in
LDS, C0= 10 to
51 mg/l with m =
16 mg/l; no
indication of
sulfate
breakthrough

not diagnostic due
to high C0 in LDS
consolidation
water





Chloride




in LCRS, C0 = 2
to 230 mg/l with m
= 104 mg/l; in
LDS, C0 = 2 to 6
mg/l with m = 4
mg/l; no indication
of chloride
breakthrough


increasing LDS Q,
with time likely
due to decreasing
dilution of
consolidation
water by
construction water


Benzene




not diagnostic
because C0 is
below DL in both
LCRS and LDS






not diagnostic
because C0 is
below DL in both
LCRS and LDS





Toluene




in LCRS, C0 = 5
to 300 |ig/l with m
= 88 |ig/l; in
LDS, C0 below
DL of 1 |ig/l; no
indication of
toluene
breakthrough


in LCRS, C0 = up
to 600 |ig/l; in
LDS, toluene
below DL of 1
|ig/l; no indication
of toluene
breakthrough


Xylene




in LCRS, xylene
detected in half
of sampling
events at C0 up
to 79 |ig/l; in
LDS, C0 below
DL of 3 |ig/l; no
indication of
xylene
breakthrough
not diagnostic
because C0 is
below DL in
LDS and, after
29 months, also
in LCRS




Summary of
Observations for Five
Key Constituents


no evidence of
significant leachate
migration into LDS;
however, monitoring
period is only about
1/4th of the estimated
GCL/CCL
breakthrough time;
more chemical data
are needed
no evidence of
significant leachate
migration in to LDS;
monitoring period
somewhat less than
estimated CCL
breakthrough time;
more chemical data
are needed
CJl
ro
-------
      Table 5-6.  Evaluation of Chemical Constituent Migration Through Landfill GM/CCL and

                GM/GCL/CCL Composite Primary Liners.
Cell
No.
AM1








AM2








Monitor.
Period
(months)
58








58








Estimated
Advective
Breakthr.
Time for
GCL/CCL
(months)'1'
4








4








Chemical
Sulfate
not diagnostic due
to high C0 in LDS
consolidation
water





not diagnostic due
to high C0 in LDS
consolidation
water





Chloride
not diagnostic due
to high C0 in LDS
consolidation
water





not diagnostic due
to high C0 in LDS
consolidation
water





Benzene
inLCRS, C0= 12
to 20 |ig/l; in
LDS, C0 below
DLof 1 |ig/l; no
indication of
benzene
breakthrough


in LCRS, C0 = 5
to 20 |ig/l; in
LDS, C0 below
DL of 1 |ig/l; no
indication of
benzene
breakthrough


Toluene
in LCRS, C0 = 40
to 420 |ig/l with m
= 267 |ig/l; in
LDS, C0 below
DL of 1 |ig/l; no
indication of
toluene
breakthrough

inLCRS, C0 = 10
to 400 |ig/l with m
= 146|ig/l; in
LDS, C0 below
DL of 1 |ig/l; no
indication of
toluene
breakthrough

Xylene
in LCRS, C0 =
71 to 150|ig/l
with m = 1 22
|ig/l; in LDS, C0
below DL of 1 to
3 |ig/l; no
indication of
xylene
breakthrough

in LCRS, C0 = 2
to 1 30 |ig/l with
m = 71 mg/l; in
LDS, C0 below
DL of 1 to 3
mg/l; no
indication of
xylene
breakthrough

Summary of
Observations for Five
Key Constituents
no evidence of
significant leachate
migration into LDS
after almost 5 years of
cell operation;
monitoring period
more than 12 times
longer than estimated
CCL breakthrough
time
no evidence of
significant leachate
migration into LDS
after almost 5 years of
cell operation;
monitoring period
more than 12 times
longer than estimated
CCL breakthrough
time
CJ1

ro
oo
-------
      Table 5-6.  Evaluation of Chemical Constituent Migration Through Landfill GM/CCL and
                GM/GCL/CCL Composite Primary Liners.
Cell
No.
A01(4)









A02(4)










Monitor.
Period
(months)
37









31










Estimated
Advective
Breakthr.
Time for
GCL/CCL
(months)'1'
140









145










Chemical
Sulfate
not diagnostic due
to similar LCRS
and LDS C0
ranges






not diagnostic due
to similar LCRS
and LDS C0
ranges







Chloride
in LCRS, C0 =
320 to 1 300 mg/l
with m = 860 mg/l;
in LDS, C0 = 7 to
1 00 mg/l with m =
40 mg/l; no
indication of
chloride
breakthrough


in LCRS, C0 =
320 to 1 ,300 mg/l
with m = 862 mg/l;
in LDS, C0 = 3 to
34 mg/l with m =
24 mg/l; no
indication of
chloride
breakthrough


Benzene
in LCRS,
benzene
detected in half ol
the sampling
events at C0 = 7
to 12|ig/l; in
LDS, C0 below
DL of 1 |ig/l; no
indication of
benzene
breakthrough
in LCRS,
benzene
detected in half o
the sampling
events at C0 = 7
to 12|ig/l; in
LDS, C0 below
DL of 1 |ig/l; no
indication of
benzene
breakthrough
Toluene
in LCRS, C0 = 10
to 550 |ig/l with m
= 167|ig/l; in
LDS, C0 below
DL of 1 |ig/l in 2/3
of sampling
events; no
indication of
toluene
breakthrough

in LCRS, C0 = 10
to 550 |ig/l with m
= 167|ig/l; in
LDS, C0 below
DL of 1 |ig/l; no
indication of
toluene
breakthrough



Xylene
in LCRS, C0 =
12to76|ig/l
with m = 34|ig/l;
in LDS, C0
below DL of 3
|ig/l; no
indication of
xylene
breakthrough


in LCRS, C0 =
12to76|ig/l
with m = 34|ig/l;
in LDS, C0
below DL of 3
|ig/l; no
indication of
xylene
breakthrough


Summary of
Observations for Five
Key Constituents
no evidence of
significant leachate
migration into LDS
after 3 years of cell
operation; however,
monitoring period is
only 1/4th of estimated
CCL breakthrough
time; more chemical
data are needed

no evidence of
significant leachate
migration into LDS
after 3 years of cell
operation; however,
monitoring period is
only 1/4th of estimated
CCL breakthrough
time; more chemical
data are needed

en

CO
-------
        Table 5-6.  Evaluation of Chemical Constituent Migration Through Landfill GM/CCL and
                      GM/GCL/CCL Composite Primary Liners.

Cell
No.



AQ1



AQ10



AR1









Monitor.
Period
(months)


97



63



36








Estimated
Advective
Breakthr.
Time for
GCL/CCL
(months)'1'
35



35



2








Chemical
Sulfate




no data available



no data available



not diagnostic due
to similar LCRS
and LDS C0
ranges





Chloride




no data available



no data available



in LCRS, C0 =
600 to 2700 mg/l
with m = 1 625
mg/l; in LDS, C0
= 8 to 74 mg/l with
m = 35 mg/l; no
indication of
chloride
breakthrough
Benzene




not diagnostic
because C0 is
below DL in both
LCRS and LDS
not diagnostic
because C0 is
below DL in both
LCRS and LDS
not diagnostic
because C0 is
below DL in both
LCRS and LDS





Toluene




not diagnostic
because C0 is
below DL in both
LCRS and LDS
not diagnostic
because C0 is
below DL in both
LCRS and LDS
not diagnostic
because C0 is
below DL in both
LCRS and LDS





Xylene




no LDS data
available


no LDS data
available


not diagnostic
because C0 is
below DL in
both LCRS and
LDS





Summary of
Observations for Five
Key Constituents


data are insufficient to
draw conclusions;
more data are needed

data are insufficient to
draw conclusions;
more data are needed

no evidence of
significant leachate
migration into LDS;
monitoring period is
more than 10 times the
estimated GCL/CCL
breakthr. time; data
are not diagnostic;
more data are needed
CO
o
        Notes:
        (1) Advective breakthrough times for steady-state saturated flow through CCL or GCL/CCL component of composite liners were calculated
           using Darcy's equation and specified hydraulic conductivities, hydraulic gradient of 5 for GCLs and 1  for CCLs, and effective porosity of 0.2.
           For this calculation, it is assumed that flow through the GM component of the composite liner occurs  through small holes and is instantaneous.
        (2) C0 = average concentration during incremental reporting period; m = mean concentration for the entire reporting period.
        (3) DL = detection limit.
        (4) Composite liquid quality samples from the LCRSs of Cells AO1 and AO2 were  assumed to represent average conditions at the two cells.
-------
period after the start of landfill cell operation.  Table 5-6 presents the results of the
authors' assessment of the occurrence of key constituent migration through the
composite primary liners.  This assessment is based on a qualitative comparison of the
five key chemical constituents. Table 5-6 also presents an estimate of the advective
breakthrough time for the CCL or GCL/CCL component of each composite primary liner.
The estimated breakthrough times were calculated assuming that the  GM component of
the composite primary liner has one or more holes through which leachate
instantaneously migrates and that leachate migration through the CCL or GCL/CCL
component of the composite liner is governed by Darcy's equation assuming one-
dimensional steady-state saturated flow.   Other assumptions used in the calculations
are given in the table. The effect of chemical retardation was not considered in
calculating the advective breakthrough times. Retardation of chloride  and sulfate should
be negligible. Retardation characteristics for benzene, toluene, and xylene will depend
on the organic carbon content of the CCL or GCL, redox conditions, and other factors.
It is expected, however, that the effective retardation coefficient for these constituents
would have been 2 or more. These organic compounds were chosen  for analysis
notwithstanding their retardation characteristics for a combination of reasons, including
relatively widespread occurrence in leachate, and relatively higher concentrations in
leachate than other organic compounds.   In addition, these three constituents are not
known as laboratory contaminants, in contrast to methylene chloride, a constituent that
is more  mobile but is also a common laboratory contaminant.

The current database is not sufficient to draw definitive  conclusions on the performance
of GM/CCL and GM/GCL/CCL composite primary liners. However, using the data and
comparisons in Tables 5-5 and 5-6, the following observations can be  offered with
respect  to key chemical constituent migration through the composite primary liners of
the 13 considered cells:

   •   There were insufficient data for three  cells (i.e.,  Y2, AQ1, and AQ2) to draw any
       conclusions on primary liner leakage rates based on key chemical constituent
       data.
   •   For the remaining ten cells, key chemical constituent data did not reveal  obvious
       indications of primary liner leakage.
   •   One  of the ten cells (i.e., AD7) exhibited a potential indication of primary  liner
       leakage when sulfate and chloride concentrations in LDS flows increased
       between 12 and 36 months after  construction.  However, the concentrations of
       other chemicals did not increase  over time. The estimated breakthrough time
       for the composite primary liner in this  cell is 70 months, several times greater
       than  the time when sulfate and chloride concentrations increased. The reason
       for the increase in the anion concentrations in the LDS flow from Cell AD7 is
       unclear.
   •   Five  of the ten cells (i.e., B3, AD1, AD7, AM1, and AM2) have  key chemical
       constituent data of sufficient completeness and  duration to conclude that
       leachate migration into the LDS at a rate of any engineering significance has not

                                      5-31
-------
       occurred for a time period exceeding the estimated breakthrough time for the
       CCL component of the composite liner.
   •   Et values were estimated for cells B3, AD1, AD7, AM1, and AM2 using Equation
       5-1, presented in Section 5.2.3, with constituent mass fluxes from the LCRS and
       LDS.  Mass fluxes were calculated using average flow rates and chemical
       concentrations for benzene, toluene, and xylene during the active operation and
       post-closure periods.  With this approach, Et values for these cells were found to
       range from 99.1 to more than 99.9%.
   •   Based on the above data and similar to GM/GCL composite liners, GM/CCL and
       GM/GCL/CCL composite liners of the type evaluated in this study can be
       constructed to achieve Et values of 99.9% or more.  However, Et values in the
       range of 99 to 99.9% will also occur.
   •   Available leakage rate calculation methods for composite liners give leakage
       rates in the same range as the rates estimated from the data for composite
       primary liners presented in Appendix E.  Notwithstanding the uncertainties in
       both the assumptions used in the calculations and the estimated leakage rates,
       this is a useful finding.
   •   In the U.S., landfill cells are typically operated for periods of one to five years,
       occasionally longer, and they are promptly covered with a GM or other low-
       permeability barrier after filling.  This operations sequence defines the timeframe
       for significant leachate generation in a landfill cell that does not contain liquid
       wastes or sludges and that does not undergo leachate recirculation or moisture
       addition.  For the cells in this study, estimated advective breakthrough times
       through CCLs, assuming no chemical retardation, were generally calculated to
       range from about 3 to 12 years (see Table 5-6).  It thus appears that GM/CCL
       and GM/GCL/CCL composite liners are capable  of substantially preventing
       leachate migration over the entire period of significant leachate generation for
       typical modern landfills.
   •   The conclusions given above for GM/CCL composite liners should be
       considered  preliminary. Additional analyses are  recommended using a larger
       database representing a larger time period of operation to confirm or modify
       these preliminary conclusions. The additional analyses should include a more
       thorough analysis of the transport characteristics of a wider array of key
       chemical constituents than considered in this study.

5.2.4.2  Leachate Generation Rates
Average and peak monthly LCRS flow rate data were evaluated for 73 MSW cells at 32
landfills, 56 HW cells at 12 landfills, eight MSW ash or coal ash cells at six landfills, and
three C&DW cells at two landfills.   Most of these landfills are located in the northeast
and southeast; only four of the landfills are located in the west. The LCRS flow rate
data are presented in Table E-3.5 of Appendix E for monitoring periods up to about ten
years.  Almost half of the cells have more than four years of LCRS flow rate data
available. Post-closure data are available for eleven MSW cells at three landfills and 22
HW cells at five landfills. Detailed  results of the data analysis are given in Tables E-5.1
to E-5.5 of Appendix E and are summarized in Tables 5-7 and 5-8 below.  The range of
                                      5-32
-------
average LCRS flow rates for the cells and the mean average LCRS flow rates are
presented in Table 5-7 as a function of waste type, landfill operational stage, and
geographical region of the U.S.  Figure 5-4 illustrates the effects of geographic region
and waste type on LCRS flow rate. Table 5-8 presents the range of average rainfall
fractions (RFs) for the cells and the mean average RFs as a function of the same
variables.  In this report, RF (in percent) is the ratio of average LCRS flow rate to
historical average annual rainfall.
Table 5-7.  Summary of Average LCRS Flow Rates (in Iphd).
Waste
Type
MSW
HW
Ash
C&DW
U.S.
Region
NE
SE
W
NE
SE
W
NE
SE
NE
Initial Period of Operation
Range
1,050-39,900
1,480-43,700
4,980-18,800
480-31,300
42-3,090
2,190-28,600
15,600-19,600
Mean
10,200
10,400
11,000
15,500
480
18,700
17,600
Active Period of
Operation
Range
41-17,700
300-10,900
55-110
1,050-21,300
270-37,100
1-4,280
1,030-35,300
8,940-24,490
3,570-16,200
Mean
3,530
2,930
83(1)
5,380
4,890
990
17,700
17,800
10,600
Post-Closure
Period
Range
55-680
340-1,130
36-1,580
56(1)


Mean
400
780
370
56(1)


Notes:  (1) Values are based on only one or two cells from one landfill.
The major findings from the evaluation of leachate generation rates are given below:

   •   LCRS flow rates during operations (i.e., the initial and active periods of
       operation) can vary significantly between landfills located in the same
       geographic region and accepting similar wastes. Large variations in flow rates
       (e.g., one order of magnitude difference) can even occur between cells at the
       same landfill. Differences in waste placement practices  may be responsible for
       these significant variations.  Limiting the size of the active disposal area and
       using effective measures to minimize rainfall  infiltration into the waste and to
       divert surface-water runoff away from the waste will significantly decrease
       leachate generation rates compared to the rates observed under less controlled
       conditions.
                                       5-33
-------
Table 5-8. Summary of Average Rainfall Fractions (in percent).
Waste
Type
MSW
HW
Ash
C&DW
U.S.
Region
NE
SE
W
NE
SE
W
NE
SE
NE
Initial Period of
Operation
Range
4-160
5-157
21-87
1-73
1-30
8-84
50-63
Mean
39
33
46
33
5
58
56(1)
Active Period of
Operation
Range
0.1-54
1-23
0.5-1
4-81
1-74
0.01-41
4-104
22-60
12-52
Mean
13
8
0.7(1)
21
11
10
55
43
34
Post-Closure
Period
Range
0.2-3
1-4
0.08-3
0.3(1)


Mean
1
3
0.8
0.3(1)


Notes:  (1) Values are based on only one or two cells from one landfill.
       The MSW cells produced, on average, less leachate than the HWand ISW cells.
       Average LCRS flow rates for MSW cells located in the NE and SE ranged from
       1,000 to 44,000 Iphd during the initial period of operation and 40 to 18,000 Iphd
       during the active period of operation.  For this group of cells during the initial
       period of operation, 60%  exhibited average LCRS flow rates less than 10,000
       Iphd and 87% had rates less than 20,000 Iphd.  For the same group of cells
       during the active period of operation, 52% had average LCRS flow rates less
       than 5,000 Iphd and 95% had flow rates less than 10,000 Iphd.  Only two MSW
       cells are located in the W. These two MSW cells had very low average LCRS
       flow rates (i.e., 55 and 110 Iphd).
       RF values calculated for the MSW cells in the NE (means of 39% and 13% for
       the initial and active periods of operation, respectively) were higher than RF
       values for the SE cells (means of 33% and 8% for the  initial  and active periods
       of operation, respectively).  It is possible that the higher water evaporation rates
       and the higher runoff occurring with shorter duration, more intense rainfalls
       associated with the SE offset any potential increases in leachate generation
       rates caused by the higher total amount of rainfall in the SE  as compared to the
       NE. RF values for the two MSW cells that are located at an arid site (average
       annual rainfall of about 430 mm)  in the W were less  than 1%.
       Average LCRS flow rates for HW cells located in the NE and SE ranged from
       500 to 31,000 Iphd during the initial period of operation and  300 to 37,000 Iphd
                                      5-34
-------
en

CO
en
40,000 -,
35,000


— 30,000
T3
.C
LU
^ 25,000
O
^ 20,000
w
or
o
_i
g 15,000
<
or
LU
< 10,000
5,000
n
c

AMSW
• HW "
+ ASH +
u&uw ^ ^
4 NE »
* W fc
+
•
+ + +
A •
• *
+•
X
• A *
S '
m A
A A A A
• HA^A A • E
' A AB "|A * X AA 1
• .it- !;4^ . . - ; |
) 200 400 600 800 1,000 1,200 1,400 1,600 1,800 2,0
                                               AVERAGE ANNUAL RAINFALL (mm)
      Figure 5-4.  Average LCRS flow rate versus average annual rainfall during the active period of operation.
-------
during the active period of operation. About 69% of these cells exhibited
average LCRS flow rates greater than 10,000 Iphd during the initial period of
operation and 21% exhibited average LCRS flow rates greater than 5,000 Iphd
during the active period of operation. Average LCRS flow rates from HW cells
during the active period of operation were 50 to 70% higher than flow rates from
MSW cells. The reason for the higher leachate generation rates at the HW cells
in this study is unclear,  but may, in part, be due to differences in waste
characteristics (e.g., initial moisture content, porosity, and permeability) and
operational practices (e.g., waste placement and covering procedures). The ten
HW cells located in the  W had low  average flow rates, ranging from about 1  to
4,000 Iphd during operations.
RF values calculated for the HW cells in the NE (means of 46% and 21 % for the
initial and active periods of operation, respectively) were higher than RF values
for the SE cells  (means of 33% and 11% for the initial and active periods of
operation, respectively). Similar to the  MSW cells, the HW cells  in the SE had
lower RF values than cells in the NE. For most of the HW cells in the W, RF
values were less than 10% during operations.
Average flow rates during operations ranged from 1,000 to 35,000 Iphd for ash
cells (1,000 to 25,000 Iphd for the seven MSW ash cells and 35,000 Iphd for the
coal ash cell) and from 4,000 to 20,000 Iphd for the C&DW cells.  The limited
number of MSW ash, coal ash,  and C&DW cells considered in this study
exhibited average LCRS flow rates during the active period of operation that
were 300 to 600% higher than average LCRS flow rates from MSW cells during
the same period. It is possible that the higher leachate generation rates at the
MSW ash, coal  ash, and C&D waste landfills may, in part, be due to differences
in waste characteristics and operational practices.
Mean RF values were 53% for ash cells and 43% for C&DW cells.
Peak monthly LCRS flow rates were typically two to three times the average
monthly flow rates for all types of waste and regions of the U.S.
Landfill geographic region has a major  impact on LCRS flow rates. For landfill
sites with historical average annual rainfall less than 500 mm, average LCRS
flow rates were  low, typically less than 2,000 Iphd. LCRS flow rates increased
with increasing rainfall up to a point.  In general, for landfills with historical
average annual rainfall greater than 1,100 to 1,200  Iphd, an increase in rainfall
did not appear to cause a corresponding increase in leachate generation rate.
LCRS flow rates were typically two to three times smaller during the active
period of operation than during  the initial period of operation.
Leachate generation  rates for the closed landfills in this study typically
decreased by a factor of four within one year after closure and by one order of
magnitude within two to four years after closure, as shown in Figure 5-5. Six
years after closure, LCRS flow rates were between  5 and 1,200 Iphd (mean of
180 Iphd). Nine years after closure, LCRS flow rates were negligible. These
data show that well designed and constructed cover systems can be very
effective in minimizing infiltration of rainfall into the waste, thus reducing
leachate generation rates to near-zero values.
                               5-36
-------
en
[
[
~ 10000 [
%
< 1000
1
3
u- 100
C/)
a
O
_l
g 10
<
tt
LJJ
^ 1


0



! a
1 a
a g o
MM

n n


0





1—1

01234
OMSW 1
DHW !

:
-
0
0 |
O g
D °

a n :
B B n a i
n
n n :
n :
n
a a D
-
D :
n
1—1 n

567891
                                          YEARS SINCE FINAL CLOSURE
       Figure 5-5. Average LCRS flow rates after closure for eleven MSW cells and 22 HW cells.
-------
5.2.4.3  Leachate Chemistry
Select leachate chemistry data for 59 cells at 50 double-lined landfills were evaluated in
terms of average constituent concentrations and relative detection frequencies.  The
distribution of leachate chemistry data by waste type and start of operation date is
presented in Table 5-9. For the purposes of the discussions on leachate chemistry in
this chapter, MSWash landfill leachate is grouped with leachate from ISW landfills. This
grouping is considered appropriate because MSWash landfill leachate is typically
nonhazardous and has chemical  characteristics more similar to leachate from ISW
landfills than to leachates from MSW or HW landfills. The MSW leachate chemistry
data are from 36 landfills located  in all geographic regions of the U.S.  Based on the
extent of the leachate chemistry data, the data are believed to be representative of
modern MSW landfills in the U.S  operated without leachate recirculation or other special
activities (e.g., special waste disposal, induced aerobic degradation). About 70% of
these landfills began operating in the 1990's. While the data for modern MSW landfills
are extensive, they should not be considered to reflect the full range of leachate
chemistry associated with the anaerobic decomposition process, from the acid stage to
the methane fermentation stage.  Moreover, differences will exist from facility to facility
based on a variety of climate, site, waste,  and operational factors. Additional data are
needed from more facilities over a longer time period to better identify the potential
range of leachate chemistry characteristics throughout the initial, active, and post-
closure  operational periods of a facility.

Table 5-9. Distribution of LCRS Chemistry Database by Waste Type and Start of
           Operation Date.
Waste Type
MSW
HW
MSW Ash
Coal Ash
C&DW
Pre-1990
Start of Operation
11 landfills
13 cells
3 landfills
5 cells
1 landfill
1 cell
1 landfill
1 cell
1 landfill
2 cells
Post-1990
Start of Operation
25 landfills
28 cells
1 landfill
1 cell
6 landfills
6 cells
1 landfill
1 cell
1 landfill
1 cell
Fewer data are available for HW and ISW landfills than for MSW landfills. In addition,
the types of wastes placed in HWand ISW landfills are generally more variable between
landfills than wastes placed in MSW landfills. With the exception of the leachate
chemistry data for MSWash landfills, it is likely that the data presented  in this report do
not characterize the variation in leachate chemistry for HW and ISW landfills.  The
                                       5-38
-------
chemistry data for MSWash landfill leachate may be representative of modern MSW
ash landfills in the U.S. because seven landfills are included in the database and the
chemistry of MSW ash is less variable than that of HW.

The leachate  chemistry data are presented in Table E-3.7 of Appendix E and
summarized in Table 5-10.  Federal MCLs, which are available for two of the heavy
metals and ten of the VOCs considered in this study, are also listed in Table 5-10. The
distributions of select chemistry data for MSW, HW,  and MSWash cells are shown in
Figures E-6.1  to E-6.3 of Appendix E.  For  MSW landfills, the chemical data for older
landfills that started operating before 1990  (pre-1990 cells) and newer landfills that
started operating during 1990 or later (post-1990 cells) are compared (Figure E-6.4 of
Appendix E).  The major findings from  the evaluation of leachate chemistry data are
given below:
   •   For a  given waste type,  many of the leachate constituents exhibited significant
       concentration variations (e.g., several orders of magnitude difference) between
       landfill cells and, sometimes, for a given cell.
   •   For the leachate types for which data are available for more than two landfills,
       the average value of pH (pH units), specific conductance (jimhos), COD (mg/l),
       BOD5  (mg/l), TOC (mg/l), and chloride (mg/l) were, respectively:
             o  MSW leachate: 6.7, 4,470, 2,500,  1,440, 380,  and 560;
             o  HW leachate: 8.2, 22,100, not available,  not available, 1,620, and
                7,760; and
             o  MSW ash leachate: 7.1, 22,100, 1,670, 55, 62, and 10,400.
       The MSW landfill leachates were mineralized, biologically-active liquids with
       relatively low concentrations of heavy metals and VOCs. On average, the
       leachates were slightly acidic (i.e., average pH of 6.7), which is expected
       because carbon dioxide and organic acids are the primary by-products of the
       first stage (i.e., the acid stage) of anaerobic  degradation of organic compounds
       in MSW landfills. The chemistry of these leachates changed with time as the
       organic compounds degraded (see for example, Table E-6.2 of Appendix E).  In
       general, the leachate characteristics for cells receiving waste were more
       indicative of the acid phase of degradation than the second stage (i.e., the
       methane fermentation phase) of anaerobic degradation.  For closed cells, the
       leachate pH typically increased with time and the BOD/COD ratio decreased
       with time, which is expected as the landfill is more fully in the methane
       fermentation phase of degradation.  Of the heavy metals and VOCs considered
       in Table 5-10, chromium, nickel, methylene chloride, and toluene were detected
       at the  highest concentrations in MSW leachates. Average concentrations of
       cadmium, benzene, 1,2-dichloroethane, trichloroethylene, and vinyl chloride in
       MSW landfills leachates exceeded  federal maximum contaminant levels (MCLs)
       (40 CFR 141.11, 141.61, and 141.62) for community drinking water systems.
       None  of the landfills had leachate with average chemical concentrations
       exceeding the MCLs for ethylbenzene, toluene, or xylenes.
                                      5-39
-------
                        Table 5-10. Summary of Landfill Leachate Chemistry Data.
Ol

-k
o
Waste Type
Number of Landfills
Parameter Units
pH pH units
Specific conductance nmhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic ng/l
Cadmium ng/l
Chromium ng/l
Lead ng/l
Nickel ng/l
Benzene ng/l
1,1-Dichloroethane ng/l
1 ,2-Dichloroethane ng/l
cis-1 ,2-Dichloroethylene ng/l
trans-1 ,2-Dichloroethylene ng/l
Ethylbenzene ng/l
Methylene chloride ng/l
1,1,1-Trichloroethane ng/l
Trichloroethylene ng/l
Toluene ng/l
Vinyl chloride ng/l
Xylenes ng/l
MCLs












50
5
100


5

5
70
100
700

200
5
1,000
2
10,000
MSW
10 Pre-1 990
Average
6.62
6,588
5,487
3,878
2,281
1,509
2,295
801
274
444
153
532
19
< 8
68
36
56
< 17
88
< 33
< 64
< 51
40
435
< 68
< 56
491
< 49
117
Minimum
6.30
3,438
2,740
804
< 2
4
1,508
199
< 23
261
84
225
< 4
< 1
5
1
27
< 3
< 5
< 4
< 53
< 32
< 5
< 5
< 5
< 5
< 5
< 7
< 5
Maximum
7.20
8,983
8,640
8,267
4,510
2,852
3,278
2,263
1,943
610
279
1,115
78
< 17
320
90
98
< 36
294
< 100
< 75
< 100
87
1,303
100
114
959
< 100
277
No. of
Landfills
8
8
9
9
10
8
7
10
10
6
6
8
10
8
10
7
9
7
8
6
2
4
7
8
6
7
7
6
6
26 Post-1 990
Average
6.79
3,693
2,758
1,939
976
527
1,536
463
205
398
83
282
23
< 7
38
15
82
< 19
66
< 16
< 57
< 18
35
334
< 55
< 24
228
< 34
83
Minimum
5.90
597
480
< 10
< 2
24
203
5
< 7
66
10
3
< 2
< 1
3
1
10
< 2
< 2
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 3
< 5
Maximum
8.09
13,548
8,621
6,800
4,700
2,609
5,800
1,625
1,376
1,994
191
1,219
236
< 20
90
50
220
< 100
260
< 100
436
< 110
118
4,150
270
100
740
< 300
220
No. of
Landfills
22
22
21
22
18
21
22
25
24
22
21
23
21
22
21
22
20
21
22
20
13
16
22
22
20
19
22
20
20
                        Note: (1)" " = not analyzed; < = more than 50% of measurements reported as non-detect.
-------
                        Table 5-10. Summary of Landfill Leachate Chemistry Data (Continued).
Ol
Waste Type
Number of Landfills
Parameter Units
pH pH units
Specific conductance nmhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic ng/l
Cadmium ng/l
Chromium ng/l
Lead ng/l
Nickel ng/l
Benzene ng/l
1,1-Dichloroethane ng/l
1 ,2-Dichloroethane ng/l
cis-1 ,2-Dichloroethylene ng/l
trans-1 ,2-Dichloroethylene ng/l
Ethylbenzene ng/l
Methylene chloride ng/l
1,1,1-Trichloroethane ng/l
Trichloroethylene ng/l
Toluene ng/l
Vinyl chloride ng/l
Xylenes ng/l
MCLs












50
5
100


5

5
70
100
700

200
5
1,000
2
10,000
HW
4
Average
8.17
22,096



1,623

7,758
2,985


5,243
26,710
< 119
124
109
738
< 131
123
< 382

< 79
< 133
161
< 99
< 76
< 173
< 1,475
14
Minimum
7.55
12,302



7

3,783
704


2,514
30
< 5
22
24
285
< 7
< 14
5

< 14
< 5
4
8
33
< 9
< 10
9
Maximum
9.36
39,598



3,239

1 1 ,734
5,267


7,972
79,912
< 233
226
249
1,190
370
< 371
< 1,124

< 143
< 512
< 447
< 347
< 146
616
< 4,405
18
No. of
Landfills
3
3



2

2
2


2
3
2
2
3
2
3
4
3

2
4
4
4
3
4
3
2
MSW ASH
7
Average
7.06
22,083
24,493
1,670
55
62
1,942
10,426
881
900
267
1,181
9
< 12
< 30
23
< 40
< 3
< 12
< 3
< 2
< 3
< 4
< 3
< 7
< 3
< 10
< 5
< 2
Minimum
6.54
10,732
6,067
304
15
39
99
2,940
85
96
113
684
5
< 2
< 1
3
< 24
< 1
< 1
< 1
< 1
< 1
< 2
< 1
< 1
< 1
< 1
< 1
< 1
Maximum
7.44
43,383
46,733
5,607
84
109
5,010
22,400
3,430
1,332
420
1,994
17
49
84
74
48
< 5
< 33
< 5
< 3
< 5
< 7
< 6
< 16
< 5
< 25
< 10
< 3
No. of
Landfills
5
4
6
4
4
3
4
4
5
3
2
5
6
6
6
6
4
3
3
3
2
3
3
3
3
3
3
3
2
                        Note: (1)" " = not analyzed; < = more than 50% of measurements reported as non-detect.
-------
                        Table 5-10. Summary of Landfill Leachate Chemistry Data (Continued).
Ol
Waste Type
Number of Landfills
Parameter Units
pH pH units
Specific conductance nmhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic ng/l
Cadmium ng/l
Chromium ng/l
Lead ng/l
Nickel ng/l
Benzene ng/l
1,1-Dichloroethane ng/l
1 ,2-Dichloroethane ng/l
cis-1 ,2-Dichloroethylene ng/l
trans-1 ,2-Dichloroethylene ng/l
Ethylbenzene ng/l
Methylene chloride ng/l
1,1,1-Trichloroethane ng/l
Trichloroethylene ng/l
Toluene ng/l
Vinyl chloride ng/l
Xylenes ng/l
MCLs












50
5
100


5

5
70
100
700

200
5
1,000
2
10,000
COAL ASH
2
Average
7.70
884
723
11
< 3
6
190
21
383
190
22
46
36
< 7
< 16
< 19
38
< 4
< 4
< 4

< 1
< 3
< 4
< 4
< 4
< 2
< 7
< 4
Minimum
7.66
623
347
11
< 3
6
160
21
178
190
15
46
< 9
< 5
< 9
< 4
38
< 4
< 4
< 4

< 1
< 3
< 4
< 4
< 4
< 2
< 7
< 4
Maximum
7.74
1144
1098
11
< 3
6
220
21
587
190
30
46
62
< 9
22
< 34
38
< 4
< 4
< 4

< 1
< 3
< 4
< 4
< 4
< 2
< 7
< 4
No. of
Landfills
2
2
2
1
1
1
2
1
2
1
2
1
2
2
2
2
1
1
1
1

1
1
1
1
1
1
1
1
C&DW
2
Average
6.43
4815
3553
2414
1126
839
2450
681
255
292
202
304
15
< 3
39
7
< 56
17
92
3


66
417
51
< 11
613
8
210
Minimum
6.43
4815
2880
1139
1126
443
2450
671
48
203
202
284
15
< 1
39
3
< 56
17
92
3


66
417
51
< 11
613
8
210
Maximum
6.43
4815
4225
3688
1126
1235
2450
690
463
382
202
324
15
< 5
39
10
< 56
17
92
3


66
417
51
< 11
613
8
210
No. of
Landfills
1
1
2
2
1
2
1
2
2
2
1
2
1
2
1
2
1
1
1
1


1
1
1
1
1
1
1
                        Note: (1)" " = not analyzed; < = more than 50% of measurements reported as non-detect.
-------
The HW landfill leachates were more mineralized and had a higher organic
content than MSW leachates.  All of the HW leachates were alkaline, with pH
values ranging from 7.5 to 9.4. One possible explanation for the alkaline pH
values is the relatively common practice of solidifying hazardous waste with
pozzolonic additives prior to disposal. These relatively high pHs decrease the
mobility of metals. Even so, the average heavy metals concentrations were
generally several times to several orders of magnitude higher in HW leachates
as compared to MSW leachates. The HW leachates also had higher average
concentrations of all VOCs, except methylene chloride, toluene, and xylenes. Of
the heavy metals and VOCs considered in Table 5-10, arsenic, nickel, 1,2-
dichloroethane, and vinyl chloride were detected at the highest concentrations in
HW leachates. Average concentrations of arsenic, cadmium, chromium,
benzene, 1,2-dichloroethane,  trichloroethylene,  and vinyl chloride in HW landfill
leachates exceeded MCLs. None of the  landfills had leachate with average
chemical concentrations exceeding the MCLs for ethylbenzene, toluene, or
xylenes.
The chemistry of the ISW landfill leachates was highly variable due to the wide
variety of wastes disposed in  ISW landfills. The pH values for these leachates
ranged from 6.4 to 7.7.  The MSW ash leachates, the most mineralized of the
ISW landfill leachates, were even more mineralized than the MSW leachates in
this study, as evidenced by the high specific conductance, TDS, sulfate, and
chloride levels of the MSW ash leachates. Coal ash leachates were the least
mineralized. Both the MSW ash and coal ash leachates had low BOD values
that were several orders of magnitude less than the BOD values for MSW
leachate because  most of the organic materials originally in the MSW and coal
had been combusted. The average BOD value for C&DW leachate, however,
was within range of values reported for MSW leachate.  Heavy metals
concentrations in MSW ash and C&DW leachates were similar to those for MSW
leachates.  Metals concentrations in coal ash leachate were lower, generally at
the lower end of the concentration range for MSW leachates. As expected, the
MSW ash and coal ash leachates did not contain VOCs. However,  published
data show that MSW ash leachates can contain trace amounts of base neutral
extractables (BNAs), polychlorinated dibenzo-p-dioxins (PCDDs), and
polychlorinated dibenzo-furans (PCDFs).  The one C&DW landfill for which
organic chemistry  data are available produced leachate containing VOCs.
Average concentrations of cadmium in MSW ash and coal ash landfill leachates
and benzene, trichloroethylene, and vinyl chloride concentrations in C&DW
landfill leachates exceeded MCLs.
In general, the leachate chemistry data collected for the study fall within the
range of published data.
With the federal solid waste regulations promulgated in the 1980's and early
1990's (e.g., 1980 RCRA Subtitle C regulations for HW in 40 CFR 261 and Land
Disposal Restrictions in  40 CFR 268), it is expected that the quality of MSW and
HW landfill  leachates would have improved over time.  No statistically significant
differences  in concentrations of the considered  trace metals or VOCs in
leachates from older modern MSW landfills constructed prior to 1990 (pre-1990
                               5-43
-------
       landfills) and leachates from newer MSW landfills constructed after 1990 (post-
       1990 landfills) were observed at the 90% confidence level.  However, average
       VOC concentrations were generally lower in leachate from the post-1990
       landfills (Table 5-10). The statistical analysis findings were limited by the data.
       The limited number of landfills contributing to each dataset and the wide range
       of chemical concentrations led to large confidence intervals for each parameter
       in the datasets. To further evaluate the differences in leachate chemistry
       between older and newer MSW landfills, the data for the post-1990 MSW
       landfills were compared to published leachate chemistry data for 61 older MSW
       landfills (i.e., pre-1980 landfills in NUS (1988) and pre-1985 landfills in Gibbons
       et al. (1992)).  The distributions of the leachate chemistry data for the older
       MSW landfills were not known,  so the two data sets could not be compared
       statistically. However, the average concentrations of trace metals and VOCs in
       leachate from the newer landfills were almost always less than the average
       concentrations in leachate from the  older landfills.  Based on the above, it
       appears that the solid waste regulations have resulted in improved MSW landfill
       leachate quality. However, more data are needed to quantify this improvement.
       From the published information summarized in this report, the regulations may
       have also reduced the occurrence of certain chemicals.  For example,
       acetonitrile, cyanide, and naphthalene were detected more frequently in
       leachate from older landfills than in leachate from newer landfills.
   •   Published leachate chemistry data for 33 older HW landfills (i.e., pre-1984
       landfills in Bramlett et al. (1987), pre-1983 landfills in NUS (1988), and pre-1987
       landfills in Gibbons et al. (1992)) were compared to the data presented for  HW
       landfills in this  report (i.e., newer HW landfills).  The dataset for newer HW
       landfills is small; only leachate chemistry data for four landfills are available.
       The concentrations of chemicals in leachate from the newer landfills were found
       to be within the range of published values for the older landfills. The distribution
       of the leachate chemistry data for the older  HW landfills was not known, so the
       two datasets could not be compared statistically.  However, on average, most
       heavy metal concentrations and almost all VOC concentrations were lower in
       leachate from the newer landfills.  This reduction in  leachate strength is likely a
       result of the Subtitle C regulations and the Land Disposal Restrictions.

5.3 Lessons Learned from Waste Containment System Problems at Landfills

5.3.1  Scope of Work
The scope of work for this portion of the project involved an investigation into problems that
have occurred in waste containment systems for 69 modern landfill facilities and five
modern surface impoundment facilities located throughout the U.S. The investigation
focused on landfills, and only landfill-related problems are discussed in this section. The
purpose of the investigation is twofold: (i) to better understand the nature, frequency,  and
significance of identified problems; and (ii) to  develop recommendations to reduce the
future occurrence of problems.
                                      5-44
-------
The scope of work specifically excluded consideration of problems in older waste
containment systems not designed and constructed to current standards and practices.
These problems include, for example, the LCRS and cover system internal drainage
layer failures described by Bass (1986),  Ghassemi et al. (1986), and Kmet et al. (1988).
The scope of work also excluded foundation stability problems at older landfills, such as
the problems described by Oweis (1985), Dvirnoff and Munion (1986), Richardson and
Reynolds (1991), Kenter et al. (1997), Stark and Evans (1997),  and Schmucker and
Hendron (1997). Problems at older facilities are often not relevant to current standards
and practices.

5.3.2  Description of Database
The 80 landfill problems identified during the investigation for this report are categorized
on the basis of two criteria. The first criterion addresses the component or attribute of
the landfill liner  system or cover system  affected by the problem. The specific landfill
components and attributes considered in this study are: (i) liner construction; (ii) liner
degradation; (iii) LCRS or LDS construction; (iv) LCRS or  LDS degradation; (v) LCRS
or LDS malfunction; (vi) LCRS or LDS operation; (vii) liner system stability; (viii) liner
system displacement;  (ix) cover system  as built; (x) cover system degradation;  (xi)
cover system stability; and (xii) cover system displacement.  Specific problems that may
affect these components and attributes are  listed in Tables F-4.1 to F-4.3 in Appendix F.
Other components or attributes not specifically associated with  landfill integrity were not
considered in the investigation.  These include landfill daily and intermediate cover
components (except for cracking of the soil  intermediate cover during the Northridge
earthquake), leachate transmission and  treatment components  beyond the leachate
collection sumps or manholes, and landfill gas extraction and management
components.

The second criterion used to categorize  the problem addresses the principal human
factor contributing to the problem. The principal human factors considered are: (i)
design; (ii) construction; and (iii) operation.  While a principal human factor has been
assigned to each problem, it should be recognized that most problems have complex
causes and several contributing factors.  Hereafter, the problem classifications  are
shown as "component or attribute criterion'Tprincipal human factor criterion" (e.g., liner
system stability/design).

Detailed case histories of the problems are presented in Attachment F-A of Appendix F.
The information  sources for the problems are listed in Table F-2.1 of Appendix F.
Summaries of the identified problems are presented in Table F-2.2 and are repeated in
Table 5-11 below.  The problems are grouped according to the above two criteria  in Table
F-2.3 of Appendix F.
                                      5-45
-------
Table 5-11. Summary of Identified Problems at Landfills.
Problem
Classification*1'
liner construction/
construction
liner construction/
construction
liner construction/
construction
liner construction/
operation
liner construction/
operation
liner construction/
design
liner construction/
construction
liner construction/
construction
liner construction/
construction
liner construction/
construction
liner construction/
construction
liner construction/
construction
liner construction/
construction
liner construction/
construction
liner degradation/design
landfill liner
degradation/operation
liner degradation/
design
liner degradation/
construction
liner degradation/
construction
liner degradation/
construction
liner degradation/
design
Facility Designation/
Appendix Section
L-1/F-A.2.1
L-3/F-A.2.2
L-5/F-A.2.3
L-6/F-A.2.4
L-7/F-A.2.5
L-8/F-A.2.6
L-9/F-A.2.7
L-11/F-A.2.8
L-11/F-A.2.9
L-15/F-A.2.10
L-17/F-A.2.11
L-19/F-A.2.12
L-19/F-A.2.13
L-29/F-A.2.14
L-2/F-A.3.1
L-4/F-A.3.2
L-12/F-A.3.3
L-14/F-A.3.4
L-20/F-A.3.5
L-43/F-A.3.6
L-44/F-A.3.7
Problem Summary
leakage through holes in HOPE GM primary liner
leakage through holes in HOPE GM liners
leakage through holes in HOPE GM primary liner
leakage through holes in HOPE GM primary liner
leakage though HOPE GM/CCL composite primary
liner at pipe penetration
landfill gas migrated beyond liner system and into
vadose zone resulting in groundwater contamination
leakage though HOPE GM primary liner at pipe
penetration
construction debris in CCL with initially smooth
surface protruded from CCL after CCL was left
exposed and subsequently eroded
leakage though HOPE GM primary liner at pipe
penetration
sand bag under installed GM liner approved by
CQA consultant
leakage through holes in HOPE GM primary liner
wind uplifted and tore HOPE GM liner during
construction
severe wrinkling of HOPE GM due to thermal
expansion during construction
large folded wrinkles in HOPE GM primary liner at
two exhumed leachate sumps
desiccation cracking of CCL in exposed HOPE
GM/CCL composite liner
HOPE GM/CCL composite liner damaged by waste
fire
leachate extraction well installed in landfill appeared to
puncture GM primary liner
HOPE GM liner damaged by fire believed to be started
by lightning strike
saturation of GCL beneath GM liner when rainwater
ponded on tack-seamed patch over GM hole
water ponded between HOPE GM and CCL
components of composite secondary liner and was
contaminated from a source other than the landfill
landfill gas well punctured GM component of
composite liner and extended into CCL
                                     5-46
-------
Table 5-11. Summary of Identified Problems at Landfills (Continued).
Problem Classification
LCRS or LDS
construction/construction
LCRS or LDS
construction/construction
LCRS or LDS
construction/construction
LCRS or LDS
construction/construction
LCRS or LDS
construction/construction
LCRS or LDS
construction/construction
LCRS or LDS degradation/
design
LCRS or LDS degradation/
design
LCRS or LDS degradation/
construction
LCRS or LDS degradation/
construction
LCRS or LDS degradation/
construction
LCRS or LDS malfunction/
operation
LCRS or LDS malfunction/
design
LCRS or LDS malfunction/
design
LCRS or LDS malfunction/
operation
LCRS or LDS operation/
operation
LCRS or LDS operation/
operation
LCRS or LDS operation/
operation
LCRS or LDS operation/
design
Facility Designation/
Appendix Section
L-10/F-A.4.1
L-15/F-A.4.2
L-16/F-A.4.3
L-28/F-A.4.4
L-32/F-A.4.5
L-33/F-A.4.6
L-9/F-A.5.1
L-11/F-A.5.2
L-13/F-A.5.3
L-18/F-A.5.4
L-30/F-A.5.5
L-12/F-A.6.1
L-22/F-A.6.2
L-36/F-A.6.3
L-37/F-A.6.4
L-5/F-A.7.1
L-23/F-A.7.2
L-34/F-A.7.3
L-35/F-A.7.4
Problem Summary
rainwater entered LDS through anchor trench
sand bags in LCRS drainage layer and debris in
LCRS pipe trench approved by CQA consultant
rainwater entered LDS through anchor trench
excessive needle fragments in manufactured
needlepunched nonwoven GT
HOPE LCRS pipe separated at joints
HOPE LCRS pipe separated at joints
erosion of sand LCRS drainage layer on liner
system side slopes
erosion of sand protection layer on liner system
side slopes
polypropylene continuous filament nonwoven GT
filter degraded due to outdoor exposure
polypropylene staple-fiber needlepunched
nonwoven GT filter degraded due to outdoor
exposure
HOPE LCRS pipe crushed during construction
LCRS pipes were not regularly cleaned and
became clogged, and LCRS drainage layer may
be partially clogged
waste fines clogged needlepunched nonwoven
GT filter wrapped around perforated LCRS pipes
waste fines clogged needlepunched nonwoven
GT filter around LCRS pipe bedding gravel
leachate seeped out landfill side slopes in the
vicinity of chipped tire layers
overestimation of LDS flow quantities due to
problems (e.g., clogging) with automated LDS
flow measuring and removal equipment
valves on LCRS pipes were not opened and
leachate could not drain, and waste and leachate
flowed over a berm into a new unapproved cell
LCRS leachate pump moved air and liquid
causing pump airlock and underestimation of
leachate quantities
LCRS leachate pumps and flowmeters continually
clogged and LDS leachate pumps turned on too
frequently and burned out prematurely
                                    5-47
-------
Table 5-11. Summary of Identified Problems at Landfills (Continued).
Problem Classification
liner system stability/
design
liner system stability/
operation
liner system stability/
design
liner system stability/
design
liner system stability/
design
liner system stability/
design
liner system stability/
design
liner system stability/
design
liner system stability/
design
liner system stability/
operation
liner system stability/
operation
liner system stability/
design
liner system
displacement/design
liner system
displacement/design
liner system
displacement/design
liner system
displacement/design
cover system construction/
construction
cover system construction/
construction
cover system degradation/
design
cover system degradation/
design
Facility Designation/
Appendix Section
L-21/F-A.8.1
L-24/F-A.8.2
L-25/F-A.8.3
L-26/F-A.8.4
L-27/F-A.8.5
L-38/F-A.8.6
L-39/F-A.8.7
L-40/F-A.8.8
L-41/F-A.8.9
L-42/F-A.8.10
L-45/F-A.8.11
L-46/F-A.8.12
L-9/F-A.9.1
L-11/F-A.9.2
L-25/F-A.9.3
L-31/F-A.9.4
C-2/F-A.10.1
C-16/F-A.10.2
C-1/F-A.11.1
C-12/F-A.11.2
Problem Summary
sliding along PVC GM/CCL interface during
construction
sliding along GN/GCL (HOPE GM side) and
GCL(bentonite side)/CCL interfaces during
operation
sliding along HOPE GM/ polyester needlepunched
nonwoven GT and HOPE GM/CCL interfaces
during operation
two tears in HOPE GM liner and cracks in soil
intermediate cover from Northridge earthquake
extensive cracks in soil intermediate cover and
further tearing of GT cushion from Northridge
earthquake
sliding along needlepunched nonwoven GT/HDPE
GM primary liner interface after rainfall
sliding along needlepunched nonwoven GT/HDPE
GM liner interface after rainfall
sliding along gravel/HDPE GM liner interface after
rainfall
sliding along very flexible GM liner/needlepunched
nonwoven GT interface after rainfall
sliding along needlepunched nonwoven GT/PVC
GM liner interface after a thaw
sliding along needlepunched nonwoven GT/HDPE
GM liner interface after erosion of soil anchoring
geosynthetics
sliding along GN/HDPE GM primary liner interface
during construction
uplift of GM by landfill gas after erosion of
overlying sand LCRS drainage layer
uplift of geosynthetics by landfill gas after erosion
of overlying sand protection layer
uplift of composite liner by surface-water
infiltration during construction
uplift of composite liner by surface-water
infiltration during construction
portion of topsoil from off-site source was
contaminated with chemicals
high failure rate of HOPE GM seam samples
during destructive testing
failure of geosynthetic erosion mat-lined
downchute on 3H:1 V side slope
erosion of topsoil layer on 60 m long, 3H:1Vside
slope
                                    5-48
-------
Table 5-11.  Summary of Identified Problems at Landfills (Continued).
Problem Classification
cover system stability/
construction
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
construction
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system
displacement/design
cover system
displacement/construction
Facility Designation/
Appendix Section
C-3/F-A.12.1
C-4/F-A.12.2
C-5/F-A.12.3
C-6/F-A.12.4
C-7/F-A.12.5
C-8/F-A.12.6
C-9/F-A.12.7
C-10/F-A.12.8
C-11/F-A.12.9
C-13/F-A.12.10
C-14/F-A.12.11
C-17/F-A.12.12
C-18/F-A.12.13
C-19/F-A.12.14
C-20/F-A.12.15
C-21/F-A.12.16
C-22/F-A.12.17
C-23/F-A.12.18
C-12/F-A.13.1
C-15/F-A.13.2
Problem Summary
sliding along nonwoven GT/GM interface during
construction
sliding along topsoil/GCL interface after rainfall
sliding along sand/woven GT interface after
rainfall
sliding along sand/GM interface after rainfall
sliding along gap-graded sand/GM interface after
rainfall
sliding along gravel/GT interface during
construction
sliding along sand/calendered nonwoven GT
interface after rainfall
sliding along sand/GM interface after rainfall
sliding along topsoil/nonwoven GT interface
during construction
sliding along PVC GM/CCL interface after a thaw
sliding along geogrid/HDPE GM interface during
construction
sliding along sand/CCL interface during rainfall
sliding along sand/CCL interface immediately after
rainfall
sliding along sand/CCL interface after rainfall
sliding along sand/CCL interface after rainfall
minor cracks in soil intermediate cover from
Northridge Earthquake
215-m long cracks in soil intermediate cover from
Northridge Earthquake
minor cracks in soil intermediate cover from
Northridge Earthquake
cover system settlement caused tearing of HOPE
GM boots around gas well penetrations of GM
barrier
localized cover system settlement during
construction stretched, but did not damage, PVC
GM barrier and opened GCL joints
Note: (1) Each problem classification has two
experienced the problem; and (ii) the principal
parts: (i) the component or attribute of the landfill that
human factor contributing to the problem.
                                          5-49
-------
5.3.3  Study Findings
Based on the results of the investigation into waste containment system problems
presented in Appendix F, the following conclusions are drawn:

   •   This investigation identified 69 modern landfill facilities that had experienced a
       total of 80 waste containment system problems. This number of facilities is
       relatively small in comparison to the over 1,000 modern landfills nationwide.
       The search for problem facilities for this study was not exhaustive, and it is
       certain that there are other facilities with problems similar to those described in
       this report.
   •   About 72% of the landfill problems were liner system related and 28% were
       cover system related.  The ratio of liner system problems to cover system
       problems is probably exaggerated by the fact that a number of the facilities
       surveyed were active and did not have a cover system.
   •   Based on the waste containment system component or attribute criterion, the
       identified landfill  problems were classified as follows, in order of decreasing
       frequency:
            o cover system stability: 23%;
            o  liner construction:  18%;
            o  liner system stability: 15%;
            o  liner degradation:  9%;
            o  LCRS or LDS construction:  8%;
            o  LCRS or LDS degradation:  6%;
            o  LCRS or LDS malfunction: 5%;
            o  LCRS or LDS operation: 5%;
            o  liner system displacement: 5%;
            o cover system construction: 2%;
            o cover system degradation: 2%; and
            o cover system displacement: 2%.
   •   Using this criterion, these problems can also be grouped into the following
       general categories (Figure 5-6):
            o  liner system or cover system slope stability or displacement: 45%
            o  liner, LCRS or LDS, or cover system construction: 28%;
            o  liner, LCRS or LDS, or cover system degradation:  17%; and
            o  LCRS or LDS malfunction or operation:  10%.
   •   Based on the principal human factor contributing to the problem criterion, the
       identified landfill  problems were classified as follows  (Figure 5-7):
            o design: 51%;
            o construction: 35%; and
            o operation: 14%.
                                     5-50
-------
                              Degradation
                                 17%
 Construction
     28%
LCRS or LDS
 Malfunction/
  Operation
    10%
                                               Stability or
                                             Displacement
                                                 45%

Figure 5-6.  General distribution of problems by waste containment system
           component or attribute criterion.
         Construction
            35%
Operation
  14%
                                            Design
                                             51%
Figure 5-7.  Distribution of problems by principal human factor contributing to the
           problem criterion.
                                   5-51
-------
Problems that occurred at two or more landfills and the number of landfills at
which they occurred are as follows, in order of presentation in Chapter F-3 of
Appendix F:
      o  leakage through holes (construction- or operation-related) in an HOPE
         GM primary liner (5 landfills);
      o  leakage through an HOPE GM primary liner or HOPE GM/CCL
         composite primary liner at the LCRS pipe penetration of the liner (3
         landfills);
      o  severe wrinkling of an HOPE GM liner during construction (2 landfills);
      o  liner damage by fire (2 landfills);
      o  liner damage during well installation (2 landfills);
      o  rainwater entered the LDS through the anchor trench (2 landfills);
      o  HOPE LCRS pipe was separated at joints (2 landfills);
      o  erosion of the sand layer on liner system side slopes (2 landfills);
      o  degradation of polypropylene nonwoven GT filters due to outdoor
         exposure (2 landfills);
      o  waste fines clogged the needlepunched nonwoven GT filter in the
         LCRS piping system  (2 landfills);
      o  clogging  and other problems with the leachate pump or flow rate
         measuring system (3 landfills);
      o  liner system slope failure due to static loading (10 landfills);
      o  liner system damage due to earthquakes (2 landfills);
      o  uplift of liner system geosynthetics by landfill gas after erosion of the
         overlying sand layer (2 landfills);
      o  uplift of composite liner by surface-water infiltration during
         construction (2  landfills);
      o  cover system slope failure during construction (4 landfills);
      o  cover system slope failure after rainfall or a thaw (11 landfills); and
      o  soil cover damage due to earthquakes (3 landfills).
For problems that occurred at three or more landfills, the principal human factor
contributing to the problem criterion, detection of the problem, causes of the
problem, and remedy of the problem are described below:
      o  Leakage through holes in an HOPE GM primary liner occurred at five
         landfills.  In each case, the holes were attributed to construction or, at
         one landfill, possibly operation factors. At two of the landfills, leakage
         was first  detected during electrical leak location surveys performed as
         part of CQA and by the relatively high LDS flow rates that occurred
         after rainwater ponded in a landfill. At the remaining three landfills,
         leakage was first detected during operation by the relatively high LDS
         flow rates and the color of and chemical constituents in the LDS liquid.
         The cause of the leakage was attributed to construction-related holes
         in the GM. However, at one landfill, where waste was placed directly
         on liner system geosynthetics (i.e., there is no soil protection layer),
         the GM may have been damaged during waste placement.  The
         leakage problem was resolved at four landfills by repairing the GM
         holes; at the remaining landfill, the problem, clearly identified only
                                5-52
-------
   after the cell had been covered with waste, was partially remedied by
   lowering the "pump on" liquid level in the LCRS sump.
o  Leakage through an HOPE GM primary liner or HOPE GM/CCL
   composite primary liner at the LCRS pipe penetration of the liner
   occurred at three landfills. This leakage was attributed to construction
   factors at two of the landfills and operation factors at the third landfill.
   At two of the landfills, leakage at the pipe penetration was detected
   during construction after rainwater ponded over the penetration and
   LDS flow rates increased. The cause of the leakage was construction
   defects in the pipe penetration; it is difficult to construct a defect-free
   pipe penetration, even when extra measures are taken to enhance the
   integrity of the connection.  At the remaining landfill, leakage was
   detected during operation when the average LDS flow rate increased
   significantly. For this landfill, the pipe penetration was damaged
   during operation when a rubber-tired loader trafficked over it.  The
   pipe penetrations were repaired; however, at one landfill where the
   problem was detected during construction, the repairs did not
   significantly decrease LDS flow rates; thus there must have existed a
   penetration defect that was not located.
o  Clogging and other problems with the leachate pumps or flow rate
   measuring system occurred at three landfills. These problems were
   attributed to design factors at one of the landfills and operation factors
   at the other two landfills. The problems, which were identified during
   routine operations, included: (i) clogging of the air lines and failure of
   the  compressor for the control system;  (ii) drift of the leachate level
   measurement system; (iii) drift of the "pump on" time setting; (iv) burn
   out  of pumps due to control system problems;  (v) clogging of pumps;
   (vi)  clogging of mechanical flowmeters; (vii) damage to electrical
   equipment by electrical storms; (viii) check valve failure; and (ix)
   inaccurate measurement of LCRS or LDS flow rates due to the above
   equipment problems. These problems  appear to have been primarily
   caused by: (i) inadequate overall mechanical system design; (ii) using
   equipment that was less reliable than was needed; (iii) using
   equipment that was not compatible with the landfill leachate; and (iv)
   not  performing equipment maintenance often enough. These
   problems were primarily remedied by equipment maintenance, repair,
   and replacement.
o  Liner system slope failure due to static  loading occurred at ten
   landfills. These problems were attributed to design factors at seven of
   the  landfills and operation factors at the remaining three landfills.
   Slope failure occurred during construction at two of the landfills and
   during operation at the remaining eight landfills.  The problem was
   detected by visual observation of mass movement of the liner system,
   cracking of soil layers near the slope crest, and tearing, tensioning, or
   wrinkling of geosynthetics.  The primary causes of failure were: (i)
   using unconservative presumed values for the critical interface shear
                         5-53
-------
   strength; (ii) not evaluating the critical condition for slope stability
   (e.g., liner system with waste at intermediate grades); (iii) not
   accounting for, or underestimating, seepage pressures; (iv) not
   accounting for moisture at the GM/CCL interface (which weakens the
   interface) due to spraying of the CCL and thermal effects; and (v) not
   maintaining the  drainage layer outlets free of snow and ice, which can
   lead to increased seepage pressures.  The slope failures were
   remedied by reconstructing the damaged liner systems, sometimes
   with different materials, and developing new construction procedures
   to reduce moisture at the  GM/CCL interface.
o  Cover system slope failure during construction occurred at four
   landfills.  These problems were attributed to design factors at two of
   the landfills and construction factors  at the remaining two landfills.
   Slope failure was detected by visual  observation of mass movement
   of the cover system, cracking  of soil  layers near the slope crest,  and
   wrinkling of geosynthetics at the toe  of the cover system slope. The
   primary causes  of failure were: (i) placing soil over the side slope
   geosynthetics from the top of the slope downward, rather from the toe
   of the slope upward; (ii) not considering the effects of variation in the
   tested geosynthetics, accuracy of test methods, and test conditions on
   the interface shear strength to use in design; and (iii) using
   unconservative  presumed values for the critical interface shear
   strength. The problems were remedied by reconstructing the cover
   systems using different cover system materials that result in higher
   interface shear strengths and  placing soil over side slope
   geosynthetics from the toe of the slope upward.
o  Cover system slope failure after rainfall or a thaw occurred at eleven
   landfills.  At all of these landfills, the failures were attributed to design
   factors.  Slope failure occurred during the post-closure period and was
   detected by visual observation of mass movement of the cover
   system, cracking of soil layers near the slope crest, and wrinkling of
   geosynthetics at the toe of the cover system slope. The primary
   causes of failure appeared to  be: (i) not accounting for, or
   underestimating, seepage pressures; (ii) clogging of the drainage
   system, which can lead to increased seepage pressures; and (iii) not
   accounting for moisture at the GM/CCL interface (which weakens the
   interface) due to rain falling on the CCL surface during construction
   and freeze-thaw effects. In general,  the problems were remedied by
   reconstructing the cover systems with new drainage systems or
   different materials.
o  So/7 cover damage due to earthquakes occurred at three landfills.
   These problems all occurred during operation and were attributed to
   design factors.  The damage,  which was detected by visual
   inspection, consisted of surficial cracking of soil intermediate cover
   occurring primarily near locations with contrast in seismic response
   characteristics (e.g., top of waste by  canyon walls). The damage was
                         5-54
-------
         expected and dealt with as an operation issue through post-
         earthquake inspection and repair (i.e., regrading and revegetating the
         cracked soil layers).
Almost all of the problems identified in this investigation were detected shortly
after they occurred by visual observation or evaluation of monitoring data.
Of the problems in this study for which the remedy was identified, six problems
were not completely repaired because their environmental impacts were not
expected to be significant and because:  (i) the source of the problem could not
be identified; (ii) the problem was not worsening; (iii) repair of liner systems or
LCRS pipes after waste placement would be extremely difficult and expensive;
and/or (iv) additional liner system damage could occur in any attempt to
excavate the waste and repair the liner system.
The problems only resulted in an  identified environmental impact to groundwater
or surface-water quality by leachate or landfill gas at one facility,  landfill L-8. At
this MSW landfill, groundwater impact by VOCs was attributed to gas migration
through a relatively permeable soil layer that secured the edge of the GM liner
and extended from the crest of the liner system side slope to beyond the liner
system. The problem was resolved by installing additional gas extraction wells
in the landfill. Without the measures taken to correct the problems at some of
the other facilities, however, adverse environmental impacts could have
eventually occurred  at these facilities.
The main impacts of the problems identified in this investigation are interruption
of waste containment system construction and operation,  increased
maintenance, and increased costs.
The identified problems that most often disrupted construction and were
required to be repaired before construction proceeded were related to:
      o holes in GM liners and  at pipe penetrations of liners;
      o large wrinkles in HOPE GM liners;
      o degradation of exposed geosynthetics;
      o uplift of constructed liners by groundwater or infiltrating surface water;
         and
      o erosion of unprotected  soil layers (CCLs, sand drainage layers, soil
         protection layers).
Problems that disrupt operation are generally more severe in terms of required
repairs than those that interfere with construction and may require waste
relocation.  Consequently, problems that disrupt operation generally require
more time to remedy than problems that are identified and repaired during
construction.  Problems that involve major breaches of liner systems or cover
systems (e.g., failure of landfill liner system slopes) may require months to
repair. The identified problems that most often disrupted operation and were
required to be repaired before operation proceeded were related to:
      o holes in GM liners and  at pipe penetrations of liners;
      o failure of one or more components of a liner system or cover system
         on landfill slopes; and
      o clogging of GTs in LCRSs.
                               5-55
-------
   •   Problems that require maintenance may be more severe in terms of required
       repairs than those that interfere with construction, but are generally less severe
       than those that interfere with operation.  In addition, problems that require
       maintenance are more likely to be reoccurring.  The identified problems that
       most often required maintenance were related to:
             o  erosion of soil layers (sand drainage layers, soil protection layers);
             o  repair of LCRS or LDS flow rate measuring and  removal systems; and
             o  cracking of soil intermediate cover after earthquakes.
   •   The costs of remedying the problems can be significant. For the identified
       problems, the costs at the times the remedies were implemented ranged from
       less than $10,000 for repairs of GM holes identified by leak location surveys
       during construction to more than several million dollars for repair of a liner
       system slope failure  that occurred during cell operation.  In general, problems
       that impacted operation were more expensive than those that impacted
       construction or maintenance.  However,  certain problems that impact
       maintenance, such as erosion of soil layers, may ultimately be more costly than
       other problems if these problems reoccur.
   •   Even though there was only evidence of environmental impact at one of the
       waste containment systems in this study, the landfill industry should do more to
       avoid future problems in order to: (i) reduce the potential risk of future
       environmental impact; (ii) reduce the potential health and safety risk to facility
       workers, visitors, and neighbors; (iii) increase public confidence  in the
       performance of waste containment systems; (iv) decrease potential impacts to
       construction, operation, and maintenance; and (v) reduce costs  associated with
       the investigation and repair of problems.
   •   Importantly, all of the design,  construction, and operation problems identified in
       this investigation can be prevented using available design approaches,
       construction materials and procedures, and operation practices. It is the
       responsibility of all professionals involved in the design,  construction, operation,
       and closure of waste containment systems to improve the  practice of waste
       containment system  engineering.  Owners must be prepared to  adequately fund
       the levels of design and CQA activity necessary to properly design and construct
       waste containment systems.  Design engineers must improve their practice to
       avoid the types of problems identified herein. Earthwork contractors,
       geosynthetics installers, and landfill operators all must be properly trained,
       supervised, and committed to the "quality goals" necessary to eliminate
       problems.

5.3.4 Recommendations
Based on an evaluation of the identified waste containment system problems, the
following general and specific design, construction, and  operation recommendations are
made to reduce the incidence of these problems. These measures are  not new; they
have been used extensively for other engineered structures, such as dams. The
measures include widely available design approaches, construction procedures,  and
operation practices.  Many recommendations for landfill  liner systems also apply to
                                      5-56
-------
cover systems, and vice versa. Because of this, the recommendations are grouped to
apply to the following broad categories:

   •   general;
   •   liners and barriers;
   •   drainage systems;
   •   surface layers and protection layers;
   •   liner system and cover system stability; and
   •   liner system and cover system displacements.

Recommendations for each of these categories are presented below.

General recommendations intended to reduce the occurrence of problems include:
   •   information dissemination (e.g., this report);
   •   training of design engineers to better understand waste containment system
       design fundamentals and to avoid the types of design problems described in this
       report;
   •   training of design engineers to be better prepared to develop waste containment
       system specifications and CQA plans that are complete and precise, that include
       the construction-related assumptions made during design, and that require
       construction and CQA procedures to identify and prevent the kinds of
       construction problems identified in this report;
   •   training of CQA personnel in standard CQA procedures to avoid the types of
       construction problems identified in this report; for engineering technicians, this
       training can be demonstrated through the National Institute for Training in
       Engineering Technologies (NICET)  certification program;
   •   training of contractors  to avoid the types of construction problems identified in
       this report;
   •   development of better construction materials, techniques, and quality
       control/quality assurance procedures to prevent the kinds of construction
       problems identified in this report;
   •   development of better operations manuals to describe and provide controls for
       procedures to be followed by landfill operations personnel;
   •   training of facility operators to better avoid the types of operation problems
       identified in this report;
   •   training of facility operators to better detect and quickly report problems
       occurring during operation; and
   •   performing  periodic independent audits to verify that the specified operation
       procedures are being practiced.

Specific recommendations are presented below in Table 5-12.
                                      5-57
-------
Table 5-12.  Specific Recommendations to Reduce Landfill Problems.
                      Recommendations
Category
Design
Construction
Operation
Liners and
Barriers
  Resin used to manufacture
  HOPE GM should be resistant
  to stress cracking.  This is
  currently evaluated using the
  notched constant tensile load
  test (ASTM D 5397). This test
  should be required in project
  specifications.
  Project specifications should
  require that both the inner and
  outer tracks of GM fusion seam
  samples taken for destructive
  testing meet the project seam
  requirements.  Failure of one
  track is generally indicative of
  overall seaming problems and
  can result in increased stress
  concentrations in the adjacent
  track.  In addition, testing both
  tracks may allow seaming
  problems to be identified and
  corrected quicker.
  The potential for GM damage
  during placement of a soil layer
  over a GM can be reduced by
  protecting the GM. Measures
  for GM protection should be
  incorporated into the design
  and specifications. Measures
  include placing a protection
  layer (e.g., thick GT cushion or
  GC drainage layer) over the
  GM, using a greater initial lift
  thickness of soil above the GM,
  and using construction
  equipment with low ground
  pressure to place soils over the
  GM. The protection measures
  should be selected based  on
  the characteristics of the soil to
  be placed (e.g., angularity,
  maximum particle  size), the
  thickness of the soil layer, the
  type of equipment placing  the
  soil, and whether CQA will be
  performed during soil
  placement. If the soil layer is
  placed during operation without
  CQA, extra GM protection is
  necessary.
  GMs located in areas
  subjected to high static and
  dynamic stresses from
  construction equipment, such
  as beneath temporary access
  roads, require an even higher
  level of protection  than GMs
  Construction equipment should
  be inspected for fuel and oil
  leaks, and those leaks should
  be repaired prior to using the
  equipment in liner construction
  to avoid liner and IDS
  contamination.
  Liners and barriers should be
  constructed in manageable
  increments that ensure
  protection of the liner and
  barrier materials under
  seasonal weather changes.
  CCLs should not be
  constructed with materials
  containing construction debris
  or large particles, even if prior
  to GM installation the CCL has
  a smooth surface and meets
  the hydraulic conductivity
  criterion. The debris may
  adversely impact the hydraulic
  conductivity of the CCL and/or
  damage an overlying GM.
  CCLs should not be left
  unprotected for an extended
  period of time.  They can
  desiccate and crack due to
  evaporation of water in the
  CCL, degrade when exposed
  to freezing and thawing
  actions, and be eroded by wind
  and water.
  Prior to deploying a GM, all
  extraneous objects (e.g., tools,
  sand bags) should be removed
  from the surface on which the
  GM is to be placed to avoid
  GM damage and, for
  composite liners, promote
  good contact between the GM
  and underlying GCL or GCL.
  HOPE GMs should be installed
  so that they are essentially
  stress-free at their lowest
  expected temperatures to
  avoid GM straining and,
  potentially, rupture.
  GMs should be covered with
  thermal insulation layers at
  very low temperatures since
  GM strain at break decreases
  with decreasing temperature.
  The  leading edge of an
  uncovered GM should be
  secured to prevent wind from
  flowing beneath the GM and
  uplifting it. This is typically
• Landfill operations manuals
  should include limitations on
  the types of equipment that
  may traffic over the liner
  system before the first lift of
  waste is  placed to prevent liner
  damage.
• Landfill operations personnel
  should be aware  of sensitive
  areas of  a liner system, such
  as at pipe penetrations or
  sumps, and should  protect
  these areas from damage.
  Sensitive areas can be
  identified with cones, flags, or
  other markers. They can also
  be isolated from traffic by
  berms, bollards, or  other
  means.
• Landfills  should be  operated to
  minimize the potential for
  waste fires. Measures to be
  taken could include not
  depositing loads  of hot waste
  in a landfill  and covering waste
  with a soil cover to decrease
  waste access to oxygen.
• Care should be taken to not
  damage  the liner system
  components when drilling into
  landfilled waste.  Settlement of
  the waste surface must be
  taken into account when
  selecting the depth  of drilling,
  and boreholes should not
  extend close (e.g., within 1 m)
  to the primary liner. Also, the
  limits of waste containment
  systems  should be  identified
  with markers or other means to
  reduce the  potential for liner
  system or cover system
  damage  by drilling or other
  invasive  activities.
                                                       5-58
-------
Table 5-12.  Specific Recommendations to Reduce Landfill Problems (Continued).
                      Recommendations
Category
Design
Construction
Operation
Liners and
Barriers
(continued)
  not subjected to high stresses.
  These protection measures
  should be incorporated into the
  design and specifications.
  GM should be protected during
  waste placement over the GM.
  Protection measures should be
  incorporated into the design
  and specifications. Measures
  include installing a protection
  layer  (e.g., thick GT cushion,
  GC drainage layer, or soil
  layer) over the GM, using
  spotters to direct equipment
  operators during waste
  placement over the GM, and
  placing only select waste over
  the GM.  Protection measures
  should be selected with
  consideration of waste
  characteristics and the
  equipment placing the waste.
  Sensitive areas  of a liner
  system (e.g., at  pipe
  penetrations) should be
  designed to be untraffickable
  by berms, bollards, or other
  means to decrease the
  potential for damage to these
  areas.
  It is difficult to construct pipe
  penetrations of liners to be
  defect free. Until new methods
  for constructing  better
  connections between GMs and
  ancillary structures have been
  developed and tested, designs
  without pipe penetrations (i.e.,
  designs with internal sumps)
  should be preferred.
  Internal sumps typically have
  sustained leachate heads at
  greater depths than other
  locations within the landfill and
  have  seamed corners, which
  may contain holes. To
  decrease the rate of leakage
  through GM holes at sumps,
  the sump design should
  include additional liner
  components, such as a GCL,
  beneath the GM liner in the
  sump area, even if the GM is
  already underlain by a CCL. A
  design with a prefabricated GM
  sump may also be considered.
  The potential for landfill gas to
  migrate over the geosynthetics
 accomplished by seaming
 adjacent panels of GM shortly
 after deployment and placing a
 row of sandbags along the
 edge of the GM.
 If sand bags are used to
 secure GM panels, the installer
 should ensure that the sand
 bags, and all other extraneous
 objects, are not trapped
 beneath the GM after seaming
 to avoid GM damage and, for
 composite liners, promote
 good contact between the GM
 and underlying CCL or GCL.
 For HOPE GMs, fusion seams
 are preferred over extrusion
 seams because fusion seams
 have higher seam integrity and
 lower stress concentrations at
 seams.  Extrusion seams
 should be minimized in the
 field by using prefabricated
 pipe boots, careful GM
 installation, etc.
 HOPE GMs must be cleaned
 along the seam path before the
 seam is constructed since dirt
 in the seam adversely impacts
 seam integrity. To minimize
 the potential for dirt to collect in
 the seam path, GM should be
 seamed shortly after
 deployment. A temporary
 protective plastic film may also
 be placed on the GM edges at
 the factory and removed from
 the GM just prior to seaming.
 In general, holes in HOPE GM
 seams should not be repaired
 by reseaming.  This reheating
 of seams can embrittle the
 HOPE at the repair and make it
 more susceptible to stress
 cracking.
 To the extent practicable, holes
 in GMs liners installed over
 GCLs should be repaired as
 soon as possible to avoid
 swelling of the GCL in case of
 hydration. GCL swelling
 results in a decrease in GCL
 shear strength and may impact
 landfill slope stability. Holes
 located in  areas where
 rainwater may pond should be
 patched first. The  patches
 should be sealed with a
                                                     5-59
-------
Table 5-12.  Specific Recommendations to Reduce Landfill Problems (Continued).
                     Recommendations
Category
Design
Construction
Operation
Liners and
Barriers
(continued)
  at the edge of the liner system
  must be considered in design.
  The potential for gas migration
  into the subsurface can be
  reduced by collecting gas
  generated in the landfill, using
  low-permeability soils over the
  edge of the liner system, and
  modifying the edge of the liner
  system so that the liner
  extends back up to the ground
  surface (like a reverse anchor
  trench).
 permanent seam and not only
 tack welded.
 When a GM is placed over a
 GCL, the GM should be
 covered with soils as soon as
 possible to minimize swelling
 of the GCL in case of
 hydration. GCL swelling
 results in a decrease in GCL
 shear strength and may impact
 landfill slope stability.
 Connections between GMs
 and ancillary structures should
 be carefully constructed and
 inspected to decrease the
 potential for construction-
 related GM defects.
 To decrease the potential for
 construction-related  GM
 defects in sumps, the GM
 panel layout should be
 configured to minimize seams
 in sumps or prefabricated
 sumps should be used.
 With respect to the potential for
 leakage, pipe penetrations are
 generally the most critical
 locations in landfills without
 internal sumps. If pipe
 penetrations are used, they
 should be carefully constructed
 and inspected.
 Sumps and pipe penetrations
 of liners should be leak tested
 by ponding tests, leak location
 surveys, gas tracer tests, or
 pressure  tests of double pipe
 boots as part of liner system
 CQA.  Leak testing of the liner
 on the landfill base (where
 leachate heads are the
 highest) may also be
 considered. Identified holes
 should be repaired.
 The entire installed GM should
 be inspected for damage and
 any damage should be
 repaired prior to placement of
 overlying materials.
 GM should be covered with a
 soil layer as soon as
 practicable after installation,
 but not during the hottest time
 of the day if the GM is
 significantly wrinkled, to reduce
 GM wrinkles, prevent GM
 uplift by wind, and protect the
 GM from  damage.	
                                                    5-60
-------
Table 5-12.  Specific Recommendations to Reduce Landfill  Problems (Continued).
                      Recommendations
Category
Design
Construction
Operation
Liners and
Barriers
(continued)
                                 Prior to placing soil over a GM,
                                 the GM should be inspected for
                                 wrinkles.  Excessive GM
                                 wrinkles and wrinkles that may
                                 fold over should be removed
                                 by waiting to backfill until the
                                 GM cools and contracts during
                                 the cooler nighttime and early
                                 morning hours, pulling the
                                 wrinkles out, or cutting the
                                 wrinkles out. The latter
                                 method is less desirable than
                                 the former methods because it
                                 requires intact GM to be cut,
                                 and it results in more GM
                                 seaming and testing.
                                 On long side slopes, it may be
                                 preferable to use textured GM
                                 rather than smooth GM to
                                 decrease the size of GM
                                 wrinkles that develop,
                                 especially near the slope toe.
                                 Composite liners and barriers
                                 constructed with a CCL should
                                 be covered with an insulation
                                 layer as soon as practicable to
                                 prevent CCL desiccation
                                 related to heating  or freeze-
                                 thaw action.
Drainage
Systems
  Adjacent materials conveying
  water should be designed to
  decrease the clogging potential
  of the downgradient material
  using filter criteria calculations
  and/or laboratory testing.
  If gap-graded soils are used as
  drainage materials, the effect
  of particle migration should  be
  evaluated during design using
  filter criteria calculations and/or
  laboratory testing. In fact, the
  effect of particle migration from
  all granular drainage materials
  should be evaluated since
  these materials have fines.
  Perforated pipes bedded in
  gravel should not be wrapped
  with a GT because the GT is
  useless, and, in some cases,
  even detrimental because the
  GT in this location is prone to
  clogging. Instead, the design
  should include a GT between
  the gravel and the surrounding
  soil or, possibly, no GT.
  Geosynthetic anchor trenches
  should be backfilled with low-
  permeability soil and the soil
  The drainage system should
  be kept free of debris that may
  impede the flow of liquid. In
  general, all sandbags should
  be removed from the drainage
  system. However, if the sand
  in the bags meets the project
  specifications for the overlying
  drainage layer material, the
  bags can be cut and removed
  and the sand left in place.
  GTs and GCs should be
  covered as soon as possible
  after installation to protect
  them from the environment
  (e.g., ultraviolet  light, water,
  high temperature, animals).
  The CQA consultant should
  verify that  all connections
  required for adjacent drainage
  system pipes have been made.
  When pipe is connected by
  butt fusion seaming, the seam
  should be  inspected for
  defects.
  Care should be taken to not
  damage drainage system pipes
  during  construction. The
  contractor should maintain
• Leachate may seep from
  landfill side slopes if the
  leachate can perch on layers of
  less permeable materials (e.g.,
  daily and intermediate cover
  materials) within the waste or
  drain from layers of more
  permeable materials (e.g.,
  tires) in the waste that are
  located relatively close to the
  side slope. The potential for
  seepage can be decreased by:
  (i) not placing layers of the
  more permeable materials near
  the side slopes; (ii) sloping
  layers of the less and more
  permeable materials away from
  the side slopes; (iii) distributing
  the more permeable materials
  throughout the waste; (iv)
  constructing leachate chimney
  drains to the LCRS around
  these layers; (v) removing
  perched leachate from wells
  installed to these layers; and
  (vi) using alternate daily covers
  (e.g., foams, tarps) that do not
  result in layers of less
                                                      5-61
-------
Table 5-12.  Specific  Recommendations to Reduce Landfill Problems (Continued).
                      Recommendations
Category
Design
Construction
Operation
Drainage
Systems
(continued)
  should be well compacted to
  reduce the potential for water
  to infiltrate into the trenches
  and flow into LCRSs or LDSs.
  If this is not practicable, the
  anchor trenches should be
  designed to drain freely and/or
  covered with a barrier, such as
  a GM. In addition, the ground
  surface should be graded away
  from the  trenches to  reduce
  runon from infiltrating into the
  trenches.
  Project specifications for
  needlepunched nonwoven GTs
  should require that the  GTs be
  needle-free and should require
  a certification from the
  manufacturer attesting to  this.
  Needles, if present, may
  damage  a nearby GM.
  The CQA Plan should require
  that deployed GTs near GMs
  be  inspected for needles
  before the GTs are covered
  with overlying materials.  If
  needles are found, the  GT
  should be rejected.
  If a GT is to be exposed to the
  environment for an extended
  time period after installation,
  the potential for degradation of
  the GT should be evaluated
  under all the anticipated
  environmental conditions.  EPA
  recommends that the effect of
  ultraviolet light on GT
  properties be evaluated using
  ASTM D 4355 (Daniel and
  Koerner, 1993).  This test is
  typically run for 500 hours;
  however, it can be run for
  longer time periods to meet
  project-specific conditions.  In
  any case, prior to covering the
  exposed GT, the  condition of
  the GT should be evaluated by
  laboratory testing to verify that
  the GT is still satisfactory.
  If test results indicate that the
  GT will not have the  required
  properties after exposure
  (typically a specified  strength
  retention), the GT should  be
  protected with a sacrificial
  opaque waterproof plastic tarp,
  soil layer, or other means.
  When the waste in a
 sufficient soil cover between
 construction equipment and
 the pipes during construction.
 Equipment operators should be
 aware of pipe locations, since
 pipes can be crushed by
 trafficking equipment.  Also,
 soil around pipes should be
 compacted using hand
 operated or walk-behind
 compaction equipment.
 After construction of a cell with
 an external sump, the pipe
 from the cell to the sump
 should be inspected to verify
 that the pipe is functioning as
 designed. The inspection may
 be performed by surveying the
 pipe with a video camera,
 pulling a mandrel through the
 pipe, flushing the pipe with
 water, or other means.
  permeable materials in the
  waste.
  Drainage system pipes should
  be maintained by cleaning the
  pipes at least annually and
  more frequently, if warranted.
  Landfills with external sumps
  may also include riser pipes at
  the low point of LCRSs as a
  precautionary measure to allow
  for leachate removal from the
  landfill, if necessary.
  Leachate flow measurement
  systems should be calibrated
  and adjusted as needed at
  least annually to ensure that
  the quantities measured are
  accurate.
  Due to the potential for
  problems in automated
  leachate metering and
  pumping equipment, landfill
  operations plans should
  include a verification program
  and contingency method for
  estimating the quantities of
  liquid removed from the LCRS
  and LDS.
  Leachate sump pumps should
  be self priming so the pumps
  will not become airlocked and
  shut down if air is pulled into
  the pumps.
  Leachate sump pumps should
  be selected to be compatible
  with sump geometries and
  anticipated leachate recharge
  rates so pump cycles are
  appropriate (e.g., not so  short
  that the  pumps turn on and off
  too frequently and burn out
  prematurely).
  The "pump on" levels in
  internal  sumps should be kept
  as low as practicable to reduce
  leakage if there are holes in
  the GM  liner in the sump,
  especially if the GM is not
  underlain by a GCL. It is
  recognized, however, that
  "pump on" liquid levels in
  internal  sumps may need to be
  larger than 0.3 m to achieve
  efficient sump pump operation.
  The potential for clogging of
  water-level indicators, pumps,
  and flowmeters must be
  considered when selecting the
                                                      5-62
-------
Table 5-12.  Specific Recommendations to Reduce Landfill Problems (Continued).
                     Recommendations
Category
Design
Construction
Operation
Drainage
Systems
(continued)
  containment system contains
  some fine particles that may
  migrate to the LCRS, the
  potential for LCRS clogging
  may be reduced by allowing
  those fine particles to pass
  though the drainage system to
  the LCRS pipes, which can
  subsequently be cleaned. The
  fine particles will pass more
  easily through the LCRS if no
  GTs are used in the LCRS or if
  the LCRS contains relatively
  thin open nonwoven GTs
  rather than thicker nonwoven
  GTs with a smaller apparent
  opening size.  Note that the
  above does not apply to an
  LCRS with only a GN drainage
  layer. Though a GN drainage
  layer has a high transmissivity,
  it is thin and is, therefore,
  generally more susceptible to
  clogging by sedimentation than
  a granular drainage layer.
                                types of equipment to use at a
                                facility.
                                Outlets of cover system
                                drainage layers should be kept
                                free of snow and ice so that
                                these layers can drain freely.
Surface
Layers
and
Protection
Layers
  Erosion of soil protection layers
  on liner system side slopes
  should be anticipated and dealt
  with in design.  The potential
  for erosion can be reduced by
  grading the liner system to
  avoid concentrated runoff and
  using a relatively permeable
  soil in the protection layer. In
  areas where the potential for
  erosion is relatively high,
  erosion control structures (e.g.,
  silt fence) can be used to
  reduce the need for intensive
  maintenance of soil protection
  layers. Protection layers can
  also be covered with a tarp or
  temporary erosion control mat.
  When a landfill is constructed
  on top of an existing landfill
  (vertical expansion), an
  exposed  GM liner can be
  uplifted by gases from the
  underlying landfill.  Therefore,
  in the case of a vertical
  expansion, unless gases from
  the underlying landfill are well
  controlled, GMs must be
  covered by a soil layer to
  prevent GM uplift and
  precautions must be taken to
  prevent erosion of this soil.
 Though it may be less costly
 for the owner to construct
 several landfill cells at once,
 this can leave new cells
 exposed to the environment for
 a significant time period.
 These cells will experience
 more erosion than cells filled
 sooner and will have more
 opportunity for liner damage.
 Additionally, every time an
 eroded soil layer is pushed
 back up the side slopes there
 is an opportunity for the
 underlying liner system
 materials to be damaged by
 construction equipment.
                                                     5-63
-------
Table 5-12.  Specific Recommendations to Reduce Landfill  Problems (Continued).
                     Recommendations
Category
Design
Construction
Operation
Surface
Layers
and
Protection
Layers
(continued)
  Better methods for protecting
  exposed soil layers on liner
  system side slopes from
  erosion or alternatives to soil
  layers (e.g., sand filled mats,
  Styrofoam sheets) are needed.
  Post-construction plans should
  be developed for portions of
  landfills that may sit idle for an
  extended period of time.  The
  plans should include
  procedures describing how the
  liner systems should be
  maintained prior to operation.
  For liner systems where soil
  protection layers are placed
  incrementally during landfill
  operation, a geosynthetic
  cushion (supercushion) better
  than the usual thick nonwoven
  GT needs to be developed to
  protect the liner system during
  soil placement.
  Erosion of surface layers on
  cover system side slopes
  should be anticipated and dealt
  with in design. In areas where
  the potential for erosion is
  relatively high, erosion control
  measures (e.g., silt fence, turf
  reinforcement and revegetation
  mat) can  be specified to
  reduce the need for intensive
  maintenance of soil layers.
  However, the erosion control
  measures themselves require
  maintenance.
  The length of cover system
  slopes between ditches or
  swales where runoff is
  collected should be selected to
  limit erosion to acceptable
  amounts (e.g., 5 tonnes/ha/yr).
  At a minimum, the potential for
  erosion should be evaluated
  using the universal soil loss
  equation. Cover system
  slopes may need to be 41-1:1 V
  or less and intercepted by
  swales at 6-m vertical intervals
  to meet acceptable erosion
  levels (EPA, 1994).
  Design flow velocities in
  drainage channels should be
  calculated so the appropriate
  channel lining can be selected.
                                                    5-64
-------
Table 5-12.  Specific Recommendations to Reduce  Landfill  Problems (Continued).
                      Recommendations
Category
Design
Construction
Operation
Liner System
and
Cover System
Stability
  The stability of liner system
  and cover system slopes
  should always be evaluated
  using rigorous slope stability
  analysis methods that consider
  actual shear strengths of
  materials, anticipated seepage
  pressures, and anticipated
  loadings.
  The majority of the slides
  described herein occurred
  along geosynthetic/
  geosynthetic interfaces. For a
  number of these cases, the
  interface shear strengths were
  estimated on the basis of
  published test data.  This
  approach should be avoided
  because there may be
  significant differences in
  interface shear strengths
  between similar materials from
  different manufacturers and
  even identical materials in
  different production lots from
  the same manufacturer.
  Because of this, geosynthetic
  interface shear strengths
  should be measured and not
  estimated.
  Interface shear strength test
  conditions (moisture, stresses,
  displacement rate, and
  displacement magnitude)
  should be representative of
  field conditions.
  The effects of variation in the
  tested geosynthetics, accuracy
  of test methods, and test
  conditions must be considered
  when selecting the design
  interface shear strength.
  Freeze-thaw of CCLs can have
  a significantly detrimental
  impact on GM/CCL interface
  shear strength and should be
  considered when selecting the
  interface shear strength to use
  in slope stability analyses.
  However, freeze-thaw effects
  on interface strength should
  not actually be a design
  consideration, since CCLs
  should be protected from
  freezing in the first place.
  The effect of construction on
  moisture conditions at the
  GM/CCL interface should be
  Soils should be placed over
  geosynthetics from the toe of
  slope upward to avoid
  tensioning the geosynthetics.
  Methods of soil placement that
  are not toe to top should be
  pre-approved by the engineer
  who analyzed the stability.
  Geosynthetic reinforcement
  should be  anchored prior to
  placing the soil layer to be
  reinforced.
  Outlets of drainage layers
  should be kept free of snow
  and ice so these layers can
  drain freely and prevent the
  buildup of seepage pressures.
  Soils or waste should be
  placed over geosynthetics from
  the toe of slope upward to
  avoid tensioning the
  geosynthetics.  Methods of
  waste placement that are not
  toe to top should be pre-
  approved by the engineer who
  analyzed the stability.
  Surficial cracking of soil cover
  layers during seismic loading,
  especially near locations with
  contrast in seismic response
  characteristics  (e.g.,  top of
  waste by rock canyon walls),
  should be anticipated and  dealt
  with as an operation  issue
  through post-earthquake
  inspection and  repair.
  Proposed changes to the
  landfill  filling sequence should
  be reviewed by the design
  engineer to ensure that these
  changes  will not adversely
  impact slope stability.
  Soil layers anchoring
  geosynthetics should be
  maintained during  landfill
  construction and operation.
                                                      5-65
-------
Table 5-12.  Specific Recommendations to Reduce Landfill  Problems (Continued).
                     Recommendations
Category
Design
Construction
Operation
Liner System
and
Cover System
Stability
(continued)
  considered when developing
  the specification for CCL
  construction and selecting the
  strength of liner system
  interfaces for slope stability
  analyses. The CCL
  construction specification
  should generally include
  limitations on maximum
  compacted moisture content,
  restrictions on applying
  supplemental moisture, and
  requirements for covering the
  CCL and overlying GM as soon
  as practical to minimize
  moisture migration to the
  GM/CCL interface.  If a CCL on
  a slope becomes desiccated, it
  should be reworked and not
  just moistened.
  Cover systems incorporating a
  low-permeability barrier layer
  should include a drainage layer
  above the barrier when the
  cover system side slopes are
  steeper than 5H:1V (EPA,
  1994). The purpose of this
  drainage layer is to prevent the
  buildup of seepage pressures
  in the cover system soil
  layer(s) overlying the barrier
  layer.
  When liner systems or cover
  systems are constructed over
  wastes, the potential for the
  wastes to generate gases that
  uplift the liners or barriers must
  be considered.  The  gas
  pressures decrease the shear
  strength along the bottom
  interface of the uplifted layer
  and may lead to slope
  instability.  Gas collection
  systems, therefore, may be
  required to prevent the buildup
  of gas pressures.
  Cover system drainage layers
  should be designed to handle
  the total anticipated flow to the
  drainage layer calculated  using
  a water balance or other
  appropriate analysis (e.g.,
  Giroud and Houlihan, 1995).
  Soong and Koerner (1997)
  recommend using a short-
  duration intensive storm in the
  water balance and do not
                                                    5-66
-------
Table 5-12.  Specific Recommendations to Reduce Landfill Problems (Continued).
                     Recommendations
Category
Design
Construction
Operation
Liner System
and
Cover System
Stability
(continued)
  recommend the EPA HELP
  computer model for this
  purpose.  The drainage layer
  flow rates output from the
  HELP model are an average
  for a 24-hour period and may
  be much less than the peak
  flow rates calculated using
  other methods if the
  precipitation data used in the
  HELP model are not carefully
  selected.
  Water collected in the drainage
  layer must be allowed to outlet
  to prevent the buildup of
  seepage pressures.
  Containment systems should
  be designed to limit seismic
  displacements to tolerable
  amounts.  To do this, designs
  may incorporate predetermined
  slip surfaces to confine
  movements to locations where
  they will cause the least
  damage (i.e., above the GM
  liner) and inverted liner system
  keyways to provide more
  resistance to movement.  For
  example,  a GM with a smooth
  top surface and a textured
  bottom surface could be  used
  in certain  liner systems to
  create a predetermined slip
  surface above the GM.
  Liner system anchor trenches
  should be designed to secure
  geosynthetics during
  construction, but release the
  geosynthetics before they are
  damaged by displacements
  during earthquakes.  An
  alternative is to unanchor the
  liner system after construction
  and secure it on a bench with
  an overlying soil layer.
  Stress concentrations at or
  near the liner system side
  slope crest should be avoided.
  Areas with stress
  concentrations are more
  problematic when subjected  to
  seismic displacements.  In
  particular, GM seams should
  generally  not be sampled near
  the slope  crest.	
                                                   5-67
-------
Table 5-12.  Specific Recommendations to Reduce Landfill Problems (Continued).
                  Recommendations
Category
Design
Construction
Operation
Liner System
and
Cover System
Displacements
 When liner systems or cover
 systems are constructed over
 existing wastes, the potential
 for the wastes to generate
 gases must be considered.
 The gases may uplift GMs,
 causing excessive stresses in
 the GMs and may impact slope
 stability. Some landfills may
 be generating little or no gas at
 the time of construction and
 may not need a gas collection
 system. Other landfills may be
 generating significant amounts
 of gas and may require a gas
 collection system beneath the
 entire liner system.
 Surface-water runoff should be
 managed to reduce foundation
 uplift problems during and after
 construction. Temporary and
 permanent surface-water
 diversion structures located
 near a cell may need to be
 lined to reduce infiltration,
 especially if the structures are
 located on relatively permeable
 soils and convey relatively
 large amounts of water.
 Runoff should not be allowed
 to pond near the cell, where it
 can infiltrate into the cell.
 Liner systems constructed over
 compressible, low shear
 strength waste materials
 should  be designed to
 accommodate the anticipated
 settlements. When GCL is
 used, seam overlaps should be
 wider than normal.
 Gas extraction well boots
 should  be designed to
 accommodate the anticipated
 landfill settlements.
 Cover systems with soil layers
 placed over compressible, low
 shear strength waste should
 use lightweight construction
 equipment and have good
 control of the thickness of soil
 placed over the waste so as
 not to cause bearing capacity
 failure of the waste and
 excessive displacement of the
 cover system.
5.4  Assessment of EPA HELP Model Using Leachate Generation Data

5.4.1  Introduction
The HELP model was developed by the U.S. Army Engineer Waterways Experiment
Station for the EPA Risk Reduction Engineering Laboratory, Cincinnati,  OH, to help
landfill designers compare the hydraulic performance of alternative waste containment
system designs.  However, the HELP model has increasingly been used to design
LCRS drainage layers and to estimate leachate generation  rates in order to size
leachate storage tanks.  There is little published information on the adequacy of the
HELP model for these purposes.  In this section of the  report, the HELP model is
assessed by comparing LCRS flow rate data from six landfill cells to leachate
                                            5-68
-------
generation rates predicted for these cells from HELP model simulations with typical
input parameters. The cells were selected to represent different waste types and
geographical regions of the U.S.  General data for the six landfill cells are included in
the landfill performance database described in Section 5.2.2.

The HELP model theoretical development and data requirements are described in
Section 5.4.2. A review of published literature on leachate generation rate predictions
by the HELP model is presented  in Section 5.4.3. The evaluation of the HELP model as
a design tool is described in Section 5.4.4, and results of this evaluation are presented
in Section 5.4.5.

5.4.2  Description of HELP Model
The HELP model simulates hydrologic processes for a landfill by performing daily,
sequential water budget analyses using a quasi-two-dimensional, deterministic
approach (Schroeder et al., 1994a, 1994b).  The hydrologic processes considered in the
HELP model include precipitation, surface-water storage (i.e., storage as snow),
interception of precipitation by foliage, surface-water evaporation, runoff, snow melt,
infiltration, plant transpiration, soil water evaporation, soil water storage, vertical drainage
(saturated and unsaturated) through non-barrier soil layers, vertical percolation (saturated)
through soil barriers, vertical percolation through GM and GM/soil composite barriers, and
lateral drainage (saturated).

Five main routines are used in  the HELP model to estimate runoff, evapotranspiration,
vertical drainage, vertical percolation, and lateral drainage.  Several other routines interact
with the main routines to generate daily precipitation, temperature, and solar radiation
values and simulate snow accumulation and melt, vegetative growth, interception, and
leakage through GM and GM/soil composite barriers. Runoff is computed using the runoff
curve-number method of the U.S.  Department of Agriculture Soil Conservation Service
(USDA-SCS) (USDA-SCS, 1985). Evapotranspiration is computed using a two-stage
modified Penman energy balance method developed by Ritchie (1972). Vertical drainage
is computed using Darcy's equation, modified to allow drainage under unsaturated
conditions using an unsaturated soil hydraulic conductivity calculated using an equation by
Campbell (1974). Percolation through a soil liner or barrier is also evaluated using Darcy's
equation, but under saturated conditions. Lateral drainage is modeled by an analytical
approximation to the steady-state solution of the Boussinesq equation.  Leakage through
GMs and GM/soil composite liners or barriers is evaluated based on the work of Giroud
and Bonaparte (1989a, 1989b) and Giroud et al. (1992).

Version 1 of the HELP model  was issued in 1984, and the model has been updated
several  times since then.  At the time this  report was prepared, Version 3 was the most
current. Data requirements for Version 3 of the HELP model are summarized in Table
5-13.  HELP requires daily and general climatic data, material properties data for the
landfill components being modeled, and landfill design data.  Required daily weather

                                       5-69
-------
data are precipitation, mean temperature, and total global solar radiation.  Daily
precipitation may be input manually, selected from a historical database (e.g., 1974-1977
data in HELP database, NOAATape, or Climatedata™), or generated stochastically
using a weather generation model developed by the U.S. Department of Agriculture-
Agricultural Research Service (USDA-ARS) (Richardson and Wright, 1984) with simulation
parameters available for 139 U.S. cities. Other daily climatologic data are generated
stochastically using the USDA-ARS routine. Required general weather data include
average annual wind speed  and latitude.  Default general weather data for 183 U.S.
cities are used by the model. The material properties of each layer being modeled are
either selected from the HELP model database of  default material properties or are
specified by the model user. Landfill design data,  including landfill general information
and layer configuration, are user specified.
Table 5-13.  Data Requirements for the EPA HELP Model, Version 3.	
WEATHER DATA REQUIREMENTS

     Evapotranspiration Data
          Default evapotranspiration option
               Location
               Evaporation zone depth
               Maximum leaf area index
          Manual evapotranspiration option
               Location
               Evaporation zone depth
               Maximum leaf area index
               Dates starting and ending the growing season
               Normal average annual wind speed
               Normal average quarterly relative humidity

     Precipitation Data
          Default precipitation option
               Location
          Synthetic precipitation option
               Location
               Number of years of data to be generated
               Normal mean monthly precipitation
          Create/Edit precipitation option
               Location
               One or more years of daily precipitation data
          NOAA Tape  precipitation option
               Location
               NOAA ASCII print file of Summary of Day daily precipitation data in as-on-tape format
          Climatedata™ precipitation option
              Location
              Climatedata™ prepared file containing daily precipitation data
          ASCII precipitation option
               Location
              Files containing ASCII data
              Years
                                         5-70
-------
Table 5-13.  Data Requirements for the EPA HELP Model, Version 3 (Continued).
          Precipitation Data (continued)
          HELP Version 2 data option
                Location
                File containing HELP Version 2 data

    Temperature Data
          Synthetic temperature option
                Location
                Number of years of data to be generated
                Years of daily temperature values
                Normal mean monthly temperature
          Create/Edit temperature option
                Location
                One or more years of daily temperature data
          NOAA Tape temperature option
                Location
                NOAA ASCII print file of Summary of Day data file containing years of daily maximum
                  temperature values or daily mean temperature values in as-on-tape format
                NOAA ASCII print file of Summary of Day data file containing years of daily minimum
                  temperature values or daily mean temperature values in as-on-tape format
          Climatedata M temperature option
                Location
                Climatedata™ prepared file containing daily maximum temperature data
                Climatedata™ prepared file containing daily minimum temperature data
          ASCII temperature option
                Location
                Files containing ASCII data
                Years
          HELP Version 2 data option
                Location
                File containing HELP Version 2 data

     Solar Radiation Data
          Synthetic solar radiation option
                Location
                Number of years of data to be generated
                Years of daily solar radiation values
                Latitude
          Create/Edit solar radiation option
                Location
          One or more years of daily solar radiation data
          NOAA Tape solar radiation option
                Location
                NOAA ASCII print file of Surface Airways Hourly solar radiation data in as-on-tape
                  format
          Climatedata™ solar radiation option
                Location
                Climatedata™ Surface Airways prepared file containing years of daily solar radiation
                  data
          ASCII Solar Radiation Option
               Location
                Files containing ASCII data
               Years
                                            5-71
-------
 Table 5-13.  Data Requirements for the EPA HELP Model, Version 3 (Continued).
      Solar Radiation Data (continued)
                HELP Version 2 Data Option
                     Location
                File containing HELP Version 2 data

 MATERIAL PROPERTY AND DESIGN DATA REQUIREMENTS

      Landfill General Information
           Project title
           Landfill area
           Percentage of landfill area where runoff is possible
           Method of initialization of moisture storage
           Initial snow water storage

      Layer Data
           Layer type
           Layer thickness
           Soil texture (Default, User-built, or manual options)
                Porosity
                Field capacity
                Wilting point
                Saturated hydraulic conductivity
           Initial volumetric soil water content
           Rate of subsurface inflow to layer

      Lateral Drainage Layer Design Data
           Maximum drainage length
           Drain slope
           Percentage of leachate collected from drainage layer that is recirculated
           Layer to receive recirculated leachate from drainage layer

      GM Liner Data
           Pinhole density in GM liner
           GM liner installation defects
           GM liner installation quality
           GM liner saturated hydraulic conductivity (vapor diffusivity)
           GT transmissivity

      Runoff Curve Number Information
           User-specified runoff curve number  used without modification
           User-specified runoff curve number  modified for surface slope and slope length
	Curve number calculated by HELP	
 5.4.3  Literature Review
 A number of researchers have performed field studies and analytical assessments to
 evaluate the HELP model (EPRI, 1984; Thompson and Tyler, 1984; Peters et al., 1986;
 Peyton and Schroeder, 1988; Barnes and Rodgers, 1988; Udoh, 1991; Lane et al., 1992;
 Benson et al., 1993;  Peyton and Schroeder, 1993; Field and Nangunoori, 1994;  Khire et al.,
 1994; Lange et al., 1997).  In particular, the studies evaluated the reliability of the HELP
 model as a tool to predict trends and magnitudes of the different landfill water balance


                                           5-72
-------
components (i.e., infiltration, runoff, etc.). The conclusions of these studies are not always
in agreement. For example, some of these studies found that HELP over-predicted
infiltration in humid climates and under-predicted infiltration in arid climates, but other
studies concluded just the opposite. In some cases, the HELP model was not able to
predict short-term trends. However, for a number of cases the HELP model analysis was
shown to give reasonable predictions of cumulative longer-term water balances. Despite
the wide use of the HELP model to predict landfill leachate generation rates, to the
author's knowledge, there are very few published studies comparing leachate generation
rates for modern landfills predicted with HELP model to actual leachate generation rates
measured at these landfills.  Two such studies were performed by Field and Nangunoori
(1994) and Lange et al. (1997). Results from these two studies are summarized below.

Field and Nangunoori (1994) used the HELP model, Version 2 to predict leachate
generation rates at an active, modern double-lined HW landfill in New York. They
compared the predicted rates to measured leachate generation rates at the landfill during
1992. Site-specific and default data were used in the model simulations. Actual site
rainfall data during 1992 were input manually.  The HELP model default values for a city
near the site were used for all other required weather data (i.e., evapotranspiration,
temperature, and solar radiation data).  The waste and liner system geometries used in
the model were selected to represent the actual landfill conditions in 1992. Data for the
intermediate and daily covers, waste, LCRS, and liner system layers (i.e., porosity, field
capacity, wilting point, and saturate hydraulic conductivity) were selected from the HELP
model database of default material characteristics. The permeabilities of the liner
system drainage layers (i.e., a 0.3-m thick sand layer and a GN) were modified to match
laboratory results or manufacturer's data.  The  authors did not provide modeling data
related to the runoff curve number or the percentage of the landfill area where runoff is
possible. The average annual leachate generation rate predicted by HELP was 36% of
the measured average annual LCRS flow rate.  Field and Nangunoori (1994) reported
that increasing the default hydraulic conductivity of the waste from 2x10"6 to 2x10"5 m/s
caused the predicted leachate generation rate to be within 17% of the measured rate.  It
is noted that the HELP model does not contain default properties for HWs, and Field and
Nangunoori used the default  properties for MSW in their simulations.

Lange et al. (1997) presented a case study of the use of the HELP model to predict
leachate generation rates at a modern MSW landfill in northeastern Ohio.  The liner
system for the landfill consisted of a GM/GCL/CCL composite liner and a GN drainage
layer. At the landfill permitting stage, the landfill designer used Version 2.05 of the HELP
model to estimate leachate generation rates. The designer used the model's default
weather data for a nearby city and default material property data.  The designer
assumed intermediate covers would be bare (i.e., unvegetated) and no surface water
runoff would leave the modeled landfill area.  The average annual leachate generation
rate estimated using the HELP model was 12,200 Iphd. The landfill began waste
placement operations in December 1992 on a 5-ha portion of the landfill.  Additional

                                      5-73
-------
areas were utilized with time, up to 25 ha by October 1995.  Actual leachate generation
rates were measured between December 1992 and February 1997.  Rates were highest,
up to 15,000 Iphd, during the first few months of operations and decreased with time. By
July 1996, LCRS flow rates had reached relatively steady levels of 800 to 900 Iphd.  In
this case, the leachate generation rate estimated during the design phase was in the
range of measured rates during the first few months of the landfill operation, but was not
representative of rates measured in the  following years.

Lange et al. (1997) evaluated the ability of the HELP model, Version 3.05 to predict
leachate generation rates that occurred at different times in the life of the landfill by using
input data that model the conditions of the landfill at these times.  In particular, actual
precipitation and temperature data recorded at a nearby city between 1992 and 1996 were
used in the HELP model. For comparison, one-year and five-year simulations were
performed.  Furthermore, the landfill was  modeled in terms of five areas with different cover,
slope, and vegetation conditions.  The extent of these areas varied during the operation of
the landfill. The leachate generation rate for the landfill at a certain time was estimated as
the area-weighted average of the average leachate generation rates for the five areas
during the considered simulation period (i.e.,  one year or five years). The landfill geometry
was modeled  based on aerial topographic maps obtained at different times in the life of the
landfill. The fraction of runoff that could exit the landfill areas was assumed to be zero for
the active area (i.e., working face) and areas with no waste, 75% for areas covered with
daily cover, and 100% for all other areas.  Good vegetation was assumed for intermediate
cover placed over areas that reached final grades. The material property data were, for the
most part, selected from the HELP default values.  The predicted leachate generation  rates
at the five different times in the life of the  landfill were between 90 and 230%  of the
measured rates.  It is noted that the length of the simulation period had limited effect on the
predicted average leachate generation rates.

5.4.4  Evaluation of HELP Model
The performance of the HELP model was evaluated as a "design  tool" to estimate
landfill leachate generation rates using a specific simulation methodology (i.e.,  specific
procedures for selecting simulation period and model input parameter values).  The
evaluation was conducted by comparing leachate generation rates estimated by HELP
Version 3.04a when used with the specific simulation methodology for six landfill cells to
measured LCRS flow rates at these cells. Table 5-14 presents operation information
and LCRS and liner system details for the six cells.  Four of the cells contain MSW, one
contains HW, and one contains MSW ash.  All of the MSW cells and the ash cell are
located in the NE; the HWcell is located in the W.  The cells varied in area between 2.2
and 6.4 ha and varied in maximum waste height between 21 and  46  m.  Average annual
rainfall at the  landfill sites was lowest (i.e., 280 mm) for Cell AC2 located in the Wand
highest (i.e., 1,190  mm) for Cells Y1  and Y2, located in the NE. As shown in Table 5-
14, all of the cells, except for Cell AC2, have a sand LCRS drainage  layer.  Cell
                                      5-74
-------
en
•Ij
en
                       Table 5-14. Operation Information and Liner System Details for Six Landfill Cells
                                        Modeled  Using EPA HELP Model.
Cell
No.



B1
B3
12
Y1
Y2
AC2
U.S.
Region11'



NE
NE
NE
NE
NE
W
Avg.
Annual
Rainfall

(mm)
1,070
1,070
990
1,190
1,190
280
Waste
Type«



MSW
MSW
MSW
ASH
MSW
HW
Cell
Area


(ha)
3.3
6.4
2.4
2.2
3.0
4.2
Max.
Waste
Height

(m)
21
25
46
10
15
30
Avg.
Liner
Base
Slope
(%)
2.0
2.0
2.5
5.5
5.5
2.0
LCRS Material
Type14'



S
S
S
S
S
G/GN
Thick.


(mm)
450
450
600
600
600
300/5
Hydraulic
Cond.(5)

(m/s)
1X10"4
1X10"4
1X10"4
5X1 0"5
5X1 0"5
5X10"3/0.1
LCRS Collector Pipe
Size (mm)
& Material'61


152PVC
152PVC
ND(3)
152PVC
152PVC
ND
Spacing


(m)
38
30
30
30
30
30
Primary Liner
Liner
Type'"


GM
GM/CCL
GM
GM/CCL
GM/CCL
GM/CCL
GM
Material16'


CSPE
CSPE
HOPE
HOPE
HOPE
HOPE
Thick.

(mm)
0.9
0.9
1.5
2.0
2.0
1.5
CCL
Thick.

(mm)
NA(3)
600
NA
450
450
450
Hydraulic
Cond.(5)
(m/s)
NA
1X10"9
NA
1X10"9
1X10"9
1X10"9
Notes:
(1) Regionsof the U.S. are:  NE = northeast, SE = southeast, W=west
(2) Waste types are: MSW = municipal solid waste, HW= hazardous waste, ASH = MSW ash
(3) ND = not determined; NA = not applicable
(4) LCRS material types are: S = sand, G = gravel, GN = geonet
(5) Hydraulic conductivity values shown represent minimum values for the LCRS drainage layer and maximum values for the CCL, as required in the project
   material specifications.
(6) Collector pipe and primary liner GM materials are:  PVC = polyvinyl chloride, HOPE = high-density polyethylene,  CSPE = chlorosulfonated polyethylene
(7) Liner types are: GM = geomembrane,  CCL = compacted clay liner
-------
AC2 has a gravel drainage layer underlain by a GN. LCRS collector pipes were used at
all of the cells and were spaced 30 to 38 m apart.

The simulation methodology used herein involved modeling four landfill scenarios
representing conditions that typically occur at different times and in different areas
within a cell during landfill operations. The first scenario assumes that essentially no
waste has been placed on the liner system and no measures have been implemented to
prevent direct infiltration of rainwater into the LCRS. This scenario may occur during
the first few months of cell operation.  The second scenario assumes waste has been
placed in the cell, and either no daily cover has been placed on the waste or no
measures have been implemented to divert clean rainwater from the cell. The third
scenario models an area of the  cell that has received waste and has been covered with
daily cover. It was assumed that the daily cover can shed away 50% of the storm-water
runoff in the cell. The fourth scenario assumes an intermediate cover has been placed
on waste that has almost reached final grades. The intermediate cover is assumed to
be vegetated with good grass and capable of diverting 100% of storm-water runoff.

The simulation methodology used herein utilized average annual (not peak) HELP
model results calculated over a  100-year simulation period.  Other methodologies could
also be used, although no other methodologies were used in  the preparation of this
report. Weather data for the simulation (i.e., daily precipitation, temperature,  and solar
radiation values) were  generated stochastically by the  HELP  program for the closest city
to the landfill site for which HELP has built-in weather parameters.  One hundred years
of data were generated and used in the HELP simulation to represent a wide range of
weather conditions that the landfill may experience.  The normal mean  monthly
precipitation values were modified to match the historical average annual precipitation
at the site. Table 5-15 summarizes HELP soil and design parameter values used for
the four different scenarios modeled at the six landfill cells. Daily and intermediate
covers and waste were modeled as vertical percolation layers. The LCRS drainage
material was modeled as a lateral drainage layer, and the CCL component of the
primary liner was modeled as a barrier soil liner. The layer material properties were
selected from the HELP model database of default material properties to represent
material properties described in the landfill  design plans and specifications.  LCRS
material hydraulic conductivity was modified in some cases to reflect values required by
the project specifications.  The cell geometry parameters (i.e., drainage length and
slope, waste height,  and surface slope and length) were selected based on the cell
design plans and on anticipated waste placement sequence and practices.

5.4.5 Study Findings
The results of the HELP model simulations are summarized in Table 5-16.  Reported for
each cell are actual average LCRS flow rates measured at the cell during the initial and
active periods of operation, as well as average flow rates obtained using HELP with the
four cell modeling scenarios.  Average annual LCRS flow rates over the 100-year

                                      5-76
-------
Table 5-15.  HELP Model Soil and Design Input Parameters for Select Cells.
Landfill General Information
- Project title
- Area of modeled portion of landfill (ha)
- Percentage of landfill area
where runoff is possible (percent) (1)
- Moisture storage initialization method
- Initial snow water storage (cm)
Layer Data

- Layer type
- Layer thickness (cm)
- Material texture default number
Porosity (vol. /vol.)
Field capacity (vol. /vol.)
Wilting point (vol./vol.)
Saturated hydraulic conductivity (cm/s)
- Rate of subsurface inflow to layer
Lateral Drainage Layer Design Data
- Maximum drainage length (m)
- Drain slope (%)
- Percentage of leachate collected from
drainage layer that is recirculated (%)
- Layer to receive recirculated leachate
from drainage layer
Geomembrane Liner Data
- Pinhole density per hectare
- Installation defects per hectare
- Geomembrane liner installation quality
- Geotextile transmissivity (m2/sec)
Runoff Curve Number Calculated by HELP

- Surface slope (%)
- Slope length (m)
- Default soil texture
- Quantity of vegetative cover
- Runoff curve number calculated by HELP

MSWCellBI located in the NE
1
0 for no waste and uncovered areas, 50 for areas with daily
cover, and 100 for areas with an intermediate cover
Program initialized to near steady state
0
Daily
Cover (1)
1
15
5
0.457
0.131
0.058
1x1 0'3
0
Intermed.
Cover (1)
1
30
10
0.398
0.244
0.136
1.2x10'4
0
Waste

1
varies (1)
18
0.671
0.292
0.077
1x1 0'3
0
Drainage
Material
2
45
1
0.417
0.045
0.018
1x10'2
0
GM CCL

4 NA
0.09 NA
40 NA
NA NA
NA NA
NA NA
3x1 0'12 NA
0 NA

20
2
0

NA


1
5
Good
NA
No
Waste (1)
2
20
1
Bare
75
No
Cover (1)
5
20
18
Bare
82
Daily
Cover (1)
5
20
5
Bare
85
Intermed.
Cover (1)
25
45
10
Good
82







Notes:
(1) Four scenarios are analyzed which model different conditions at different areas of the cell: (i) the no waste
   scenario models conditions at an area which has not received waste and where measures have not been
   implemented to prevent rainwater from entering the LCRS; (ii) the no cover scenario models an active waste
   disposal area which has not received daily or intermediate covers, and therefore, no runoff is allowed (waste
   thickness assumed at 3 m); (iii) the daily cover scenario models an area which received daily cover and allows
   rainwater runoff from 50% of the area (waste thickness assumed at 6 m); and (iv) the intermediate cover
   scenario models an area which received an intermediate cover and allows rainwater runoff from the entire area
   (waste thickness assumed at 12 m).
                                                5-77
-------
Table 5-15.  HELP Model Soil and Design Input Parameters for Select Cells (Continued).
Landfill General Information
- Project title
- Area of modeled portion of landfill (ha)
- Percentage of landfill area
where runoff is possible (percent) (1)
- Moisture storage initialization method
- Initial snow water storage (cm)
Layer Data

- Layer type
- Layer thickness (cm)
- Material texture default number
Porosity (vol. /vol.)
Field capacity (vol. /vol.)
Wilting point (vol./vol.)
Saturated hydraulic conductivity (cm/s)
- Rate of subsurface inflow to layer
Lateral Drainage Layer Design Data
- Maximum drainage length (m)
- Drain slope (%)
- Percentage of leachate collected from
drainage layer that is recirculated (%)
- Layer to receive recirculated leachate
from drainage layer
Geomembrane Liner Data
- Pinhole density per hectare
- Installation defects per hectare
- Geomembrane liner installation quality
- Geotextile transmissivity (m2/sec)
Runoff Curve Number Calculated by HELP

- Surface slope (%)
- Slope length (m)
- Default soil texture
- Quantity of vegetative cover
- Runoff curve number calculated by HELP

MSW Cell B3 located in the NE
1
0 for no waste and uncovered areas, 50 for areas with daily
cover, and 100 for areas with an intermediate cover
Program initialized to near steady state
0
Daily
Cover (1)
1
15
5
0.457
0.131
0.058
1x1 0'3
0
Intermed.
Cover (1)
1
30
10
0.398
0.244
0.136
1.2x10'4
0
Waste

1
varies (1)
18
0.671
0.292
0.077
1x1 0'3
0
Drainage
Material
2
45
1
0.417
0.045
0.018
1x10'2
0
GM CCL

4 3
0.09 60
40 16
NA 0.427
NA 0.418
NA 0.367
3x1 0'12 1x1 0'7
0 0

17
2
0

NA


1
5
Good
NA
No
Waste (1)
2
17
1
Bare
76
No
Cover (1)
5
17
18
Bare
82
Daily
Cover (1)
5
17
5
Bare
85
Intermed.
Cover (1)
25
60
10
Good
82







Notes:
(1) Four scenarios are analyzed which model different conditions at different areas of the cell: (i) the no waste
   scenario models conditions at an area which has not received waste and where measures have not been
   implemented to prevent rainwater from entering the LCRS; (ii) the no cover scenario models an active waste
   disposal area which has not received daily or intermediate covers, and therefore, no runoff is allowed (waste
   thickness assumed at 3 m); (iii) the daily cover scenario models an area which received daily cover and allows
   rainwater runoff from 50% of the area (waste thickness assumed at 6 m); and (iv) the intermediate cover
   scenario models an area which received an intermediate cover and allows rainwater runoff from the entire area
   (waste thickness assumed at 12 m).
                                               5-78
-------
Table 5-15.  HELP Model Soil and Design Input Parameters for Select Cells (Continued).
Landfill General Information
- Project title
- Area of modeled portion of landfill (ha)
- Percentage of landfill area
where runoff is possible (percent) (1)
- Moisture storage initialization method
- Initial snow water storage (cm)
Layer Data

- Layer type
- Layer thickness (cm)
- Material texture default number
Porosity (vol. /vol.)
Field capacity (vol. /vol.)
Wilting point (vol./vol.)
Saturated hydraulic conductivity (cm/s)
- Rate of subsurface inflow to layer
Lateral Drainage Layer Design Data
- Maximum drainage length (m)
- Drain slope (%)
- Percentage of leachate collected from
drainage layer that is recirculated (%)
- Layer to receive recirculated leachate
from drainage layer
Geomembrane Liner Data
- Pinhole density per hectare
- Installation defects per hectare
- Geomembrane liner installation quality
- Geotextile transmissivity (m2/sec)
Runoff Curve Number Calculated by HELP

- Surface slope (%)
- Slope length (m)
- Default soil texture
- Quantity of vegetative cover
- Runoff curve number calculated by HELP

MSW Cell 12 located in the NE
1
0 for no waste and uncovered areas, 50 for areas with daily
cover, and 100 for areas with an intermediate cover
Program initialized to near steady state
0
Daily
Cover (1)
1
15
5
0.457
0.131
0.058
1x1 0'3
0
Intermed.
Cover (1)
1
30
10
0.398
0.244
0.136
1.2x10'4
0
Waste

1
varies (1)
18
0.671
0.292
0.077
1x1 0'3
0
Drainage
Material
2
60
1
0.417
0.045
0.018
1x1 0'2
0
GM CCL

4 NA
0.15 NA
35 NA
NA NA
NA NA
NA NA
2x1 0"13 NA
0 NA

76
2.5
0

NA


1
3
Good
NA
No
Waste (1)
2.5
76
1
Bare
74
No
Cover (1)
5
20
18
Bare
82
Daily
Cover (1)
5
25
5
Bare
85
Intermed.
Cover (1)
20
35
10
Good
83







Notes:
(1) Four scenarios are analyzed which model different conditions at different areas of the cell: (i) the no waste
   scenario models conditions at an area which has not received waste and where measures have not been
   implemented to prevent rainwater from entering the LCRS; (ii) the no cover scenario models an active waste
   disposal area which has not received daily or intermediate covers, and therefore, no runoff is allowed (waste
   thickness assumed at 3 m); (iii) the daily cover scenario models an area which received daily cover and allows
   rainwater runoff from 50% of the area (waste thickness assumed at 6 m); and (iv) the intermediate cover
   scenario models an area which received an intermediate cover and allows rainwater runoff from the entire area
   (waste thickness assumed at 12 m).
                                               5-79
-------
Table 5-15.  HELP Model Soil and Design Input Parameters for Select Cells (Continued).
Landfill General Information
- Project title
- Area of modeled portion of landfill (ha)
- Percentage of landfill area
where runoff is possible (percent) (1)
- Moisture storage initialization method
- Initial snow water storage (cm)
Layer Data

- Layer type
- Layer thickness (cm)
- Material texture default number
Porosity (vol. /vol.)
Field capacity (vol. /vol.)
Wilting point (vol./vol.)
Saturated hydraulic conductivity (cm/s)
- Rate of subsurface inflow to layer
Lateral Drainage Layer Design Data
- Maximum drainage length (m)
- Drain slope (%)
- Percentage of leachate collected from
drainage layer that is recirculated (%)
- Layer to receive recirculated leachate
from drainage layer
Geomembrane Liner Data
- Pinhole density per hectare
- Installation defects per hectare
- Geomembrane liner installation quality
- Geotextile transmissivity (m2/sec)
Runoff Curve Number Calculated by HELP

- Surface slope (%)
- Slope length (m)
- Default soil texture
- Quantity of vegetative cover
- Runoff curve number calculated by HELP

MSWAsh Cell Y1 located in the NE
1
0 for no waste and uncovered areas and for areas with
daily cover; 1 00 for areas with an intermediate cover
Program initialized to near steady state
0
Daily
Cover (1)
1
15
5
0.457
0.131
0.058
1x1 0'3
0
Intermed.
Cover (1)

Not
Applicable





Waste

1
varies (1)
32
0.450
0.116
0.049
1x1 0'2
0
Drainage
Material
2
60
1
0.417
0.045
0.018
5x1 0'3
0
GM CCL

4 3
0.2 45
35 16
NA 0.427
NA 0.418
NA 0.367
2x1 0'13 1x1 0'7
0 0

17
5.5
0

NA


1
3
Good
NA
No
Waste (1)
5.5
17
1
Bare
76
No
Cover (1)
5
17
32
Bare
97
Daily
Cover (1)
5
17
5
Bare
85
Intermed.
Cover (1)

Not
Applicable









Notes:
(1) Four scenarios are analyzed which model different conditions at different areas of the cell: (i) the no waste
   scenario models conditions at an area which has not received waste and where measures have not been
   implemented to prevent rainwater from entering the LCRS; (ii) the no cover scenario models an active waste
   disposal area which has not received daily or intermediate covers, and therefore, no runoff is allowed (waste
   thickness assumed at 3 m); (iii) the daily cover scenario models an area which received daily cover and allows
   rainwater runoff from 50% of the area (waste thickness assumed at 6 m); and (iv) the intermediate cover
   scenario models an area which received an intermediate cover and allows rainwater runoff from the entire area
   (waste thickness assumed at 12 m).
                                               5-80
-------
Table 5-15.  HELP Model Soil and Design Input Parameters for Select Cells (Continued).
Landfill General Information
- Project title
- Area of modeled portion of landfill (ha)
- Percentage of landfill area
where runoff is possible (percent) (1)
- Moisture storage initialization method
- Initial snow water storage (cm)
Layer Data

- Layer type
- Layer thickness (cm)
- Material texture default number
Porosity (vol. /vol.)
Field capacity (vol. /vol.)
Wilting point (vol./vol.)
Saturated hydraulic conductivity (cm/s)
- Rate of subsurface inflow to layer
Lateral Drainage Layer Design Data
- Maximum drainage length (m)
- Drain slope (%)
- Percentage of leachate collected from
drainage layer that is recirculated (%)
- Layer to receive recirculated leachate
from drainage layer
Geomembrane Liner Data
- Pinhole density per hectare
- Installation defects per hectare
- Geomembrane liner installation quality
- Geotextile transmissivity (m2/sec)
Runoff Curve Number Calculated by HELP

- Surface slope (%)
- Slope length (m)
- Default soil texture
- Quantity of vegetative cover
- Runoff curve number calculated by HELP

MSW Cell Y2 located in the NE
1
0 for no waste and uncovered areas, 50 for areas with daily
cover, and 100 for areas with an intermediate cover
Program initialized to near steady state
0
Daily
Cover (1)
1
15
5
0.457
0.131
0.058
1x1 0'3
0
Intermed.
Cover (1)
1
30
10
0.398
0.244
0.136
1.2x10'4
0
Waste

1
varies (1)
18
0.671
0.292
0.077
1x1 0'3
0
Drainage
Material
2
60
1
0.417
0.045
0.018
5x1 0'3
0
GM CCL

4 3
0.2 45
35 16
NA 0.427
NA 0.418
NA 0.367
2x1 0'13 1x1 0'7
0 0

17
5.5
0

NA


1
3
Good
NA
No
Waste (1)
5.5
17
1
Bare
76
No
Cover (1)
5
17
18
Bare
82
Daily
Cover (1)
5
17
5
Bare
85
Intermed.
Cover (1)
30
30
10
Good
83







Notes:
(1) Four scenarios are analyzed which model different conditions at different areas of the cell: (i) the no waste
   scenario models conditions at an area which has not received waste and where measures have not been
   implemented to prevent rainwater from entering the LCRS; (ii) the no cover scenario models an active waste
   disposal area which has not received daily or intermediate covers, and therefore,  no runoff is allowed (waste
   thickness assumed at 3 m); (iii) the daily cover scenario models an area which received daily cover and allows
   rainwater runoff from 50% of the area (waste thickness assumed at 6 m); and (iv) the intermediate cover
   scenario models an area which received an intermediate cover and allows rainwater runoff from the entire area
   (waste thickness assumed at 12 m).
                                               5-81
-------
Table 5-15.  HELP Model Soil and Design Input Parameters for Select Cells (Continued).
Landfill General Information
- Project title
- Area of modeled portion of landfill (ha)
- Percentage of landfill area
where runoff is possible (percent) (1)
- Moisture storage initialization method
- Initial snow water storage (cm)
Layer Data

- Layer type
- Layer thickness (cm)
- Material texture default number
Porosity (vol. /vol.)
Field capacity (vol. /vol.)
Wilting point (vol./vol.)
Saturated hydraulic conductivity (cm/s)
- Rate of subsurface inflow to layer
Lateral Drainage Layer Design Data
- Maximum drainage length (m)
- Drain slope (%)
- Percentage of leachate collected from
drainage layer that is recirculated (%)
- Layer to receive recirculated leachate
from drainage layer
Geomembrane Liner Data
- Pinhole density per hectare
- Installation defects per hectare
- Geomembrane liner installation quality
- Geotextile transmissivity (m2/sec)
Runoff Curve Number Calculated by HELP

- Surface slope (%)
- Slope length (m)
- Default soil texture
- Quantity of vegetative cover
- Runoff curve number calculated by HELP

HW Cell AC2 located in the W
1
0 for no waste and uncovered areas, 50 for areas with daily
cover, and 100 for areas with an intermediate cover
Program initialized to near steady state
0
Daily
Cover (1)
1
15
5
0.457
0.131
0.058
1x1 0'3
0
Intermed.
Cover (1)
1
30
10
0.398
0.244
0.136
1.2x10'4
0
Waste

1
varies (1)
18
0.671
0.292
0.077
1x1 0'3
0
Protect.
Material
1
30
5
0.457
0.131
0.058
1x10'3
0
Drainage GM CCL
Material
1/2 4 3
30/0.5 0.15 45
1/20 35 16
0.417/0.85 NA 0.427
0.045/0.01 NA 0.418
0.018/0.005 NA 0.367
0.5/10 2x10"13 1x10"7
0 00

40
2.0
0

NA


1
3
Good
NA
No
Waste (1)
2
17
5
Bare
76
No
Cover (1)
5
17
18
Bare
82
Daily
Cover (1)
5
17
5
Bare
85
Interm.
Cover (1)
25
30
10
Good
83







Notes:
(1) Four scenarios are analyzed which model different conditions at different areas of the cell: (i) the no waste
   scenario models conditions at an area which has not received waste and where measures have not been
   implemented to prevent rainwater from entering the LCRS; (ii) the no cover scenario models an active waste
   disposal area which has not received daily or intermediate covers, and therefore, no runoff is allowed (waste
   thickness assumed at 3 m); (iii) the daily cover scenario models an area which received daily cover and allows
   rainwater runoff from 50% of the area (waste thickness assumed at 6 m); and (iv) the intermediate cover
   scenario models an area which received an intermediate cover and allows rainwater runoff from the entire area
   (waste thickness assumed at 12 m).
                                               5-82
-------
simulation were selected for comparison with measured flow rates. As shown in Table
5-16, estimated flow rates are highest for the "no waste" scenario and decrease
significantly with placement of waste and daily cover.  The lowest flow rates were
estimated for the "intermediate cover" scenario.  Therefore, when the HELP model is
used in the manner described above, it seems capable of modeling the trend  of
decreasing leachate generation rate with time observed at landfills. This trend is more
fully investigated in Appendix E.

Table 5-16.  Summary of Measured and Estimated LCRS Flow Rates for Six
            Landfill Cells.
Cell
No.


B1
B3
12
Y1
Y2
AC2
Initial Period of Operation
Time
Period
(months)
ND(1)
1-4
1-7
ND
1-10
1-6
Avg. LCRS
Flow Rate
(Iphd)
ND
15,304
6,627
ND
23,368
272
Active Period of Operation
Time
Period
(months)
20-54
5-93
8-76
13-78
11-54
7-88
Avg. LCRS
Flow Rate
(Iphd)
3,816
3,748
728
19,319
7,918
18
LCRS Flow Rates (Iphd) Estimated Using
HELP for Four Different Scenarios
No
Waste
14,335
16,075
11,698
17,685
17,685
1,885
No
Cover
9,622
10,820
7,290
17,066
12,597
1,250
Daily
Cover
9,124
10,142
7,462
15,504
12,225
1,025
Intermediate
Cover
4,392
5,035
3,533
NA(1)
6,708
130
Notes:  (1) ND = not determined, NA = not applicable.
       (2) Reported measurements are average values over the indicated time period. HELP
          simulation estimates represent average values for 100-year simulation period.

As shown in Table 5-16, the MSWash Cell Y1 exhibited much higher measured LCRS
flow rates than the MSW and HW cells.  During the active period of operation, the
average LCRS flow rate for Cell Y1 was about 19,300 Iphd and the average flow rates
for the other cells ranged from about 20 to 7,900 Iphd.  The higher flow rate in this range
is for MSW Cell Y2, located at the same site and having the same liner system details
and cell geometry as Cell Y1.  The higher LCRS flow rate for Cell Y1 than for Cell Y2
may be attributed to the slow placement rate of ash, the high hydraulic conductivity of
the ash, and lack of storm-water diversion from the ash cell.  The small thickness of
waste, high ash hydraulic conductivity, and lack of storm-water diversion allow for
relatively unimpeded infiltration of rainwater through the waste into the LCRS, and,
therefore, result in high leachate flow rates.  The lowest measured flow rate during the
active period of about 20 Iphd occurred for HW cell AC2,  located in the W at a site with
an average annual rainfall of only 280 mm.

LCRS flow rates  estimated using the HELP simulation methodology exhibited similar
trends as the measured flow rates.  The MSW ash cell Y1 had the highest estimated
flow rates; the HW cell in the W had the lowest estimated flow rates. Leachate
generation rates  of approximately 17,100 Iphd and 12,600 Iphd were estimated for the
MSW ash Cell Y1 and MSW Cell Y2, respectively,  for the "no cover" scenario,
demonstrating that, at least for the  considered cells, the HELP simulation methodology
is capable of predicting relative differences in LCRS flow rates for different waste types.
                                      5-83
-------
Figure 5-8 presents a comparison of average measured LCRS flow rates for the six
cells and the LCRS flow rates estimated for these cells using the HELP simulation
methodology. In this figure, the ranges of estimated LCRS flow rates given in Table
5-16 are plotted against the average measured LCRS flow rates given in the same
table. In particular, estimated LCRS flow rates for the "no waste" and "no cover"
scenarios are directly compared to average measured flow rates during the initial period
of operation and estimated LCRS flow rates for the "daily cover" and "intermediate
cover" scenarios are directly compared to average measured flow rates during the
active period of operation. Each complete data set is represented by a box when the
data points are connected as done in the figure. As shown in Figure 5-8, except for HW
Cell AC2 located in the W, the estimated leachate generation rates using the HELP
simulation methodology are generally of the same order of magnitude. The estimated
LCRS flow rates were somewhat higher than measured flow rates for MSW cells and
somewhat lower than the measured flow rate for the MSW ash cell.  For MSW Cells B1
and B3, the estimated flow rates were somewhat higher than measured flow rates. For
example, for Cell B3, estimated LCRS flow rates were in the range of 5,000 to 16,100
Iphd, while average measured flow rates were 15,300 Iphd during the initial period of
operation and 3,700 Iphd during the active period of operation.  For MSW ash Cell Y1,
the HELP methodology somewhat underpredicted leachate generation rates. Average
measured flow rates were 19,300 Iphd during the active period of operation;  the
estimated flow rates using the HELP simulation methodology were in the range of
15,500 to 17,700 Iphd. For MSW Cell 12 and especially for HW Cell AC2, the estimated
rates were significantly higher than measured rates. Average measured LCRS flow
rates for Cell AC2 were about 270 Iphd during the initial period of operation and 20 Iphd
during the active period of operation; the estimated flow rates using the HELP
simulation methodology were in the range of 130 to 1,000 Iphd.

For the evaluations performed in this section, the HELP model responded as expected
to changes in waste type and site climate. When used with default parameters, the
HELP model may generally overpredict LCRS flow rates for MSW landfills and landfills
in arid climates;  however, too few sites were evaluated to draw definitive conclusions.
Part  of the conservatism of the HELP model in predicting LCRS flow rates, especially as
waste is placed, may lie in the default moisture content value for the waste.  In the
HELP model, the waste is assumed to be at field capacity. However, MSW is typically
placed at moisture contents less than field capacity.  The moisture storage capacity of
waste is particularly important in arid climates since this storage capacity, if utilized,
may  hold essentially all the rainwater infiltrating the waste, and little leachate will be
generated. Also, some landfill cells,  and especially HW cells, have special methods of
handling rainwater that may not be taken into account in the  HELP model. For
example, in some HW cells, part of the waste is covered with a GM during cell
operation.  Rainwater collected on the GM is removed from the cell separate from
leachate, and, if clean, is discharged.

                                     5-84
-------
          25,000 -r
                                            Y1    XY2    oAC2
       LLI
                        5,000     10,000     15,000     20,000    25,000
                        MEASURED LCRS FLOW RATES (Iphd)
Figure 5-8. LCRS flow rates estimated using HELP versus measured LCRS flow
           rates for six landfill cells.
The authors believe that the HELP model can appropriately be employed as a tool to
estimate long-term average leachate generation rates to use with an appropriate level of
conservatism in the design of LCRS drainage layers and the sizing of leachate
management system components. The authors recommend that users develop a
consistent simulation methodology (analogous to the methodology used herein, with the
same or different underlying assumptions) for the HELP model and that they evaluate
the simulations, similar to the evaluations in Table 5-16 and Figure 5-8, using data from
existing local landfills. These simulations can be enhanced by performing parametric
analyses for key input parameters, such as initial waste moisture content.  With this
consistent, locally calibrated approach, the usefulness of the HELP model as a design
tool can be improved.
                                     5-85
-------
5.5 References
Barnes, F. J. and Rodgers, J. E. (1988), "Evaluation of Hydrologic Models in the Design
   of Stable Landfill Final Covers", U.S. Environmental Protection Agency, Office of
   Research and Development, Cincinnati, OH, EPA Project Summary, Report No.
   EPA/600/S-88/048.
Bass, J. M. (1986), "Avoiding Failure ofLeachate Collection and Cap Drainage Layers",
   EPA/600/2-86/058, U.S. Environmental Protection Agency,  Hazardous Waste
   Engineering Research Laboratory, Cincinnati, OH, 142 p.
Benson, C. H., Khire, M. V., and Bosscher, P. J. (1993),  "Final Cover Hydrologic
   Evaluation, Final Cover-Phase II", Environmental Geotechnics Report No. 93-4,
   University of Wisconsin, Madison, Wl, 151 p.
Bonaparte, R. and Gross B. A. (1990), "Field Behavior of Double-Liner Systems", Waste
   Containment Systems: Construction, Regulation, and Performance, ASCE
   Geotechnical Special Publication No. 26, R. Bonaparte ed., New York, pp. 52-83.
Bonaparte, R., Othman, M. A., Rad, N. S., Swan, R. H., and Vander Linde, D.  L. (1996),
   "Evaluation of Various Aspects of GCL Performance", Appendix F in Report of 1995
   Workshop on Geosynthetic Clay Liners, D.E. Daniel and H.E. Scranton, authors,
   EPA/600/R-96/149, EPA National Risk Management Research Laboratory,
   Cincinnati, OH, pp. F1-F34.
Bramlett,  J., Furman, C., Johnson, A., Ellis, W. D., Nelson, H.  and Vick, W. H. (1987),
   "Composition ofLeachates from Actual Hazardous Waste Sites",  U.S. Environmental
   Protection Agency, Office of Research and Development, Cincinnati, OH, EPA/600/2-
   87/043, 113 p.
Campbell, G. S. (1974), "A Simple Method for Determining Unsaturated Hydraulic
   Conductivity from Moisture Retention Data", Soil Science, 117(6), pp. 311-314.
Daniel, D. E. and Koerner, R. M. (1993), "Technical Guidance Document: Quality
   Assurance and Quality Control for Waste Containment Facilities", EPA/600/R-93/182,
   EPA Risk Reduction Research Laboratory, Cincinnati, OH, 305 p.
Dvirnoff, A. H. and Munion,  D. W. (1986), "Stability Failure of a Sanitary Landfill",
   International Symposium on Environmental Geotechnology, Lehigh, Pennsylvania,
   pp. 25-35.
EPA (1989), "Technical Guidance Document: Final Covers on Hazardous Waste Landfills
   and Surface Impoundments", EPA/530/SW-89/047, U.S. Environmental Protection
   Agency, Office of Solid Waste and Emergency Response, Washington, D.C., 39 p.
EPA (1994), "Seminar Publication Design, Operation, and Closure of Municipal Solid
   Waste Landfills",  EPA/625/R-94/008, U.S. Environmental Protection Agency, Center
   for Environmental Research Information, Cincinnati, Ohio, 86 p.
EPRI (1984), "Comparison of Two Groundwater Flow Models-UNSATID and HELP',
   Electric Power Research Institute, Topical Report, EPRI CS-3695, Project 1406-1.
Field, C. R. and Nangunoori, R. K. (1994), "Case Study-Efficacy of the HELP Model: A
   Myth or Reality?", Proceedings of the Waste Tech '94 Conference, National Solid
   Wastes Management Association, Charleston, SC., 9 p.
Ghassemi, M., Crawford, K., and Haro, M. (1986), "Leachate Collection and Gas
   Migration and Emission Problems at Landfills and Surface Impoundments",

                                     5-86
-------
  EPA/600/2-86/017, U.S. Environmental Protection Agency, Hazardous Waste
  Engineering Research Laboratory, Cincinnati, Ohio, 206 p.
Gibbons, R. D., Dolan, D., Keough, H., O'Leary, K., and O'Hara, R. (1992), "A
  Comparison of Chemical Constituents in Leachate From Industrial Hazardous Waste
  & Municipal Solid Waste Landfills", Proceeding of Fifteenth Annual Madison Waste
  Conference, University of Wisconsin - Madison, Sep, pp. 251-276.
Giroud, J. P. and Bonaparte, R. (1989a), "Leakage Through Liners Constructed with
  Geomembranes - Part I. Geomembrane Liners", Geotextile and Geomembranes, Vol.
  8, No. 1,  pp. 27-67.
Giroud, J. P. and Bonaparte, R. (1989b), "Leakage Through Liners Constructed with
  Geomembranes - Part II. Composite Liners", Geotextile and Geomembranes, Vol. 8,
  No. 2, pp. 77-111.
Giroud, J. P., Badu-Tweneboah, K., and Bonaparte, R. (1992),  "Rate of Leakage
  Through a Composite Liner Due to Geomembrane Defects",  Geotextile and
  Geomembranes, Vol.  11, No. 1, pp. 1-28.
Gross, B. A., Bonaparte, R., and Giroud, J.P. (1990), "Evaluation of Flow From Landfill
  Leakage Detection Layers", Proceedings of Fourth International Conference on
  Geotextiles, Vol. 2, The Hague, pp. 481-486.
Kenter, R. J., Schmucker, B. 0., and Miller, K. R. (1997), "The Day the Earth Didn't
  Stand Still: The Rumpke Landfill", Waste Age, Mar, pp. 66-81.
Khire, M. V., Benson,  C. H., and Bosscher, P. J. (1994), "Final Cover Hydrologic
  Evaluation-Phase III", Environmental Geotechnics Report No. 94-4, University of
  Wisconsin-Madison, Wl, 142 p.
Kmet, P., Mitchell, G., and Gordon, M. (1988), "Leachate Collection System Design and
  Performance - Wisconsin's Experience", Proceedings of ASTSWMO National Solid
  Waste Forum on Integrated Municipal Waste Management, Lake Buena Vista,
  Florida.
Lane, D. T., Benson,  C. H., and Bosscher, P. J. (1992), "Hydrologic Observations and
  Modeling Assessments of Landfill Covers", Final report No. 92-10, University of
  Wisconsin-Madison, 406 p.
Lange, D. A., Cellier, B. F., and Dunchak, T. (1997), "A Case Study of the HELP Model:
  Actual Versus Predicted Leachate Production Rates at a MSW Landfill in North-
  eastern Ohio", Proceedings of the SWAN A Conference, Sacramento, CA, 18 p.
NUS (1988), "Draft Background Document Summary of Data on Municipal Solid Waste
  Landfill Leachate Characteristics "Criteria for Municipal Solid Waste Landfills" (40
  CFR Part 258) Subtitle D of Resource Conservation and Recovery Act (RCRA)", U.S.
  Environmental Protection Agency, Office of Solid Waste, Washington, D.C.,
  EPA/530-SW-88-038, Jul.
Oweis, I. S.  (1985), "Stability of Sanitary Landfills", Geotechnical Aspects of Waste
  Management, seminar sponsored byASCE Metropolitan Section, New York.
Peters, N., Warner, R. S., Coates, A. L,  Logsdon, D. S., and Grube, W. E.  (1986),
  "Applicability of the HELP Model in Multilayer Cover Design:  A Field Verification and
  Modeling Assessment", Land Disposal of Hazardous Waste-Proceedings of the 1986
  Research Symposium, U.S. Environmental Protection Agency, Cincinnati, OH.

                                    5-87
-------
Peyton, R. L. and Schroeder, P. R. (1988), "Field Verification of HELP Model for
   Landfills", Journal of Environmental Engineering, ASCE, Vol. 114, No. 2, pp. 247-
   269.
Peyton, R. L. and Schroeder, P. R. (1993), "Water Balance for Landfills", Geotechnical
   Practice for Waste Disposal, D.E. Daniel, Ed., Chapman & Hall, London, pp. 214-
   243.
Richardson, G. and Reynolds, D. (1991), "Geosynthetic Considerations in a Landfill on
   Compressible Clays", Proceedings of Geosynthetics '91, Atlanta, Georgia, 1991, Vol.
   2, pp. 507-516.
Richardson, C. W. and Wright, D. A. (1984), "WGEN: A Model for Generating Daily
   Weather Variables", ARS-8, U.S. Department of Agriculture, Agricultural Research
   Service, 83 p.
Ritchie, J. T. (1972), "A Model for Predicting Evaporation from  a Row Crop with
   Incomplete Cover", Water Resources Research, 8(5), pp. 1204-1213.
Schmucker, B. 0. and Hendron, D. M. (1997), "Forensic Analysis of 9 March 1996
   Landslide at the Rumpke Sanitary Landfill, Hamilton County, Ohio", Slope Stability in
   Waste Systems, seminar sponsored by ASCE Cincinnati and Toledo Sections, Ohio.
Schroeder, P.  R., Lloyd, C. M., and Zappi, P. A. (1994a), "The Hydrologic Evaluation of
   Landfill Performance (HELP) Model, User's Guide for Version 3", U.S. Environmental
   Protection Agency, Office of Research and Development, Washington, D.C., Report No.
   EPA/600/R-94/168a.
Schroeder, P.  R., Dozier, T. S., Zappi, P. A., McEnroe, B. M., Sjostrom, J. W., and Peyton,
   R. L. (1994b), "The Hydrologic Evaluation of Landfill Performance (HELP) Model
   Engineering Documentation for Version 3", U.S. Environmental Protection Agency,
   Office of Research and Development,  Washington,  D.C., Report No. EPA/600/R-
   94/168b, 116 p.
Soong, T. Y. and Koerner, R. M. (1997), "The Design of Drainage Systems Over
   Geosynthetically Lined Slopes", GRI  Report #19,  Geosynthetic Research Institute,
   Philadelphia, Pennsylvania, 88 p.
Stark, T.  D. and Evans, W. D. (1997), "Balancing Act", Civil Engineering, Aug,  pp. 8A-
   11 A.
Thompson, F. and Tyler S. (1984),  "Comparison of Two Groundwater Flow Models
   (UNSAT1D and HELP) and their Application to Covered Fly Ash Disposal Sites",
   EPRI Document Series, Electric Power Research  Institute, Palo Alto, California, Aug.
Udoh, F.  D. (1991), "Minimization of Infiltration  Into Mining Stockpiles Using Low
   Permeability Covers", Dissertation Proposal,  Dept. of Materials Science and
   Engineering,  Mining Engineering Program, University of Wisconsin-Madison.
UDSA-SCS (1985), "Hydrology", Section 4 in National Engineering Handbook, U.S.
   Government  Printing Office, Washington, D.C.
                                     5-88
-------
                                 Chapter 6
                    Summary and Recommendations

6.1  Rationale and Scope of Chapter
The study discussed in this research report addressed three important areas of waste
containment system design and performance, namely:

   •   geosynthetic materials (puncture protection of GMs using GTs, wave behavior in
       HOPE GMs, plastic pipe behavior under high overburden stresses, and service
       life prediction of GTs  and GMs);
   •   natural soil materials  (slope stability of final cover systems with GCLs, kfieid of
       natural soil CCLs and soil-bentonite admixed CCLs, and hydraulic performance
       of CCLs in final cover systems); and
   •   field performance (LCRS and LDS flow quantities and chemical quality at
       landfills, assessment of EPA HELP computer code as a design tool using LCRS
       flow rate data, and lessons learned from waste containment problems at
       landfills).

All three areas were addressed through multiple tasks, each important in its' own right,
but also complementary to the other tasks because of the interrelationships between
waste containment system components. The ultimate goals of this study were to
assess the field performance of waste containment systems and to develop
recommendations for further improving the performance of these systems in
comparison to the current state-of-practice.

This chapter presents a summary of the tasks conducted for this study and provides
recommendations on practices to further improve the performance of waste
containment systems. These recommendations were developed, in part, using the
results of the various tasks.  Some, however, go beyond the scope of this study and are
offered by the authors with the understanding that the current level of "good" field
performance can be further improved within current material, design, testing, and
installation technology and practices.

6.1.1  Geosynthetics
As discussed  in Chapter 1, geosynthetics, including GMs, GTs, GNs,  GCs, plastic pipe,
and GCLs, are used in waste containment systems for a variety of functions.  Most
modern waste containment systems contain one or more geosynthetic components.
Notwithstanding their broad use, issues related to geosynthetic materials persist.
Indeed, the relative newness of these materials compared  to natural soil construction
materials requires that they continue to be studied and evaluated.  Chapter 2 of this
report described the results of the geosynthetic-related tasks of this research project.
These tasks addressed:
                                   6-1
-------
   •   protection of GMs from puncture using needlepunched nonwoven GTs;
   •   behavior of waves in HOPE GMs when subjected to overburden stress;
   •   plastic pipe stress-deformation behavior under high overburden stress;
   •   service life prediction of GTs; and
   •   service life prediction of GMs.

Key findings of the geosynthetic-related tasks are given below:

   •   Needlepunched nonwoven GTs can provide adequate protection of GMs against
       puncture by adjacent granular soils. A design methodology for GM  puncture
       protection was developed from the results of laboratory tests and was
       presented.
   •   Temperature-induced waves (wrinkles) in GMs do not disappear when the GM is
       subjected to overburden stress (i.e., when the GM is covered with soil), rather
       the wave height decreases somewhat, the width of the wave decreases  even
       more (i.e., the height-to-width ratio (H/W) of the wave increases), and the void
       space beneath the wave becomes smaller.  Residual stresses in HOPE  GMs
       installed in the field may be on the order of about 1 % to 22% of the  GM's short-
       term yield strength in the vicinity of GM waves, with higher residual  stresses
       associated with higher H/W values.  Significant residual stresses can reduce the
       GM service life.  The relationship between GM type, residual stress magnitude,
       and service life requires further investigation.
   •   If GM waves after backfilling are to be avoided,  light-colored (e.g., white) GMs
       can be used, GMs can be deployed and seamed without intentional slack,  GMs
       can be covered with an  overlying light colored temporary GT until backfilling
       occurs, and backfilling can be performed only in the coolest part of the day or
       even at night.
   •   Based on finite element modeling  results, use of the Iowa State formula  for
       predicting plastic pipe deflection under high overburden stress is reasonable.  In
       comparison to the FEM  predictions, the Iowa State formula overestimated  pipe
       vertical deflection under short-term conditions (which is conservative) and
       slightly underestimated pipe vertical deflection under long-term conditions
       (which is slightly unconservative, but typically accommodated by the
       incorporation of a factor of safety).
   •   PP GTs are slightly more susceptible to UV degradation than PET GTs,  and
       lighter weight GTs degrade faster than heavier GTs.
   •   GTs that are partially degraded by UV light do not continue to degrade when
       covered with soil, i.e., the degradation process is not auto-catalytic.
       Nonetheless, good practice dictates that GTs be covered with overlying
       protective materials in a timely manner to minimize exposure. Also, GTs should
       be protected  from exposure prior to installation (i.e., by keeping the  GT rolls in
       opaque bags).
                                   6-2
-------
   •   Buried HOPE GMs have an estimated service life that is measured in terms of at
       least hundreds of years. The three stages of degradation and approximate
       associated times for each as obtained from the laboratory testing program
       described in this report are: (i) antioxidant depletion (« 200 years), (ii) induction
       (« 20 years), and (iii) half-life (50% degradation) of an engineering property («
       750 years).  It is noted that these durations were obtained from the extrapolation
       of a number of laboratory tests performed under a limited range of conditions. It
       is recommended that additional testing be performed under a broader range of
       conditions to develop additional insight into the ultimate service life of HOPE
       GMs, and other types of GMs as well.

6.1.2 Natural Soils
CCLs, including those constructed from natural clay soils and those constructed from
soil-bentonite mixtures, have long been used in waste containment systems as
hydraulic barriers to inhibit liquid migration from the waste management unit. Either
used alone, or with a GM component in the form of a GM/CCL composite liner, CCLs
form an essential part of many liner systems and final cover systems. Other natural soil
materials used in liner and final cover systems include sands and gravels used for gas
conveyance systems or liquid drainage and collection systems, and soil layers used for
filtration, separation, or protection. Notwithstanding the widespread use of natural soil
materials in liner systems and final cover systems, questions and issues persist relative
to their use.  Several of these questions and issues were investigated, and the results
were reported  in Chapters 3 and 4 of this report. The subject areas that were
addressed are:

   •   slope stability of GCLs in final cover systems, as assessed from field test plots;
   •   kfieid of low-permeability natural soil CCLs;
   •   kfieid of admixed (soil-bentonite) CCLs; and
   •   CCL hydraulic performance in final cover systems;

These topics were selected on the basis of past research indicating areas where
additional insight was required,  or on the basis of concerns developed from relatively
recent field experience. Key findings of the natural soils related tasks are given below:

   •   Slope stability monitoring of final cover system test plots incorporating GCLs
       demonstrated acceptable performance for test plots constructed on 3H:1V
       slopes, but several of the test plots constructed on 2H:1V slopes failed.
       Importantly, for internally-reinforced GCLs, these failures were not due to
       inadequate internal strength, but inadequate interface strength.  Clearly, proper
       characterization of GCL interface shear strength is an important design step.
   •   The key to achieving low kfieid for natural soil CCLs is to ensure that 70 to 80%,
       or more, of the field-measured compaction (w vs. yd) points lie on or above the
       line of optimums for the particular CCL being placed.
                                    6-3
-------
   •   Practically no correlation was found between kfieid and frequently measured soil
       characterization parameters, such as plasticity index and percentage of clay,
       indicating that natural soil CCLs can be constructed with a relatively broad range
       of soil materials.
   •   Compaction density appears to be more significant than water content for
       achieving low kfieid in soil-bentonite liners.
   •   The long-term hydraulic performance of low-permeability (i.e., kfieid ^ 10~7 cm/s)
       CCLs in final cover systems may not be good in light of the effects of
       desiccation, freeze-thaw, root penetration, animal intrusion, and subsidence.

6.1.3  Field Performance
The premise of this portion of the study was that "modern"  waste containment systems
have been installed for up to a decade or more allowing for an assessment of their field
performance. Information on actual field performance can  be used to evaluate how
waste containment systems are performing now, and for extrapolation of their long-term
performance. Chapter 5 presented a discussion of the following specific topics related
to the performance of waste containment systems for modern landfills:

   •   evaluation of published information on field performance;
   •   collection and analysis of liquids management data;
   •   identification and assessment of problems;  and
   •   assessment of the EPA HELP model as a tool for LCRS design.

These topics were selected to develop an improved understanding of the actual field
performance of modern landfill liner systems, and, to the extent possible, to develop
answers to the questions identified in Section 5.1.1 of this report.  Key findings of the
field performance tasks are:

   •   LDSs from double-lined landfills will almost always  exhibit flow.  Much of this
       flow may be from sources other than primary liner leakage, particularly  in the
       time frame just after construction when construction water can be a significant
       source, and for GM/CCL composite liners, following waste placement when
       consolidation water from the CCL can be a major source.
   •   Average monthly active-period LDS flow rates for cells with HOPE GM primary
       liners  constructed with CQA (but without ponding tests or electrical leak location
       surveys) will often be less than 50 Iphd, but occasionally in excess of 200 Iphd.
       These flows are attributable primarily to liner leakage and, for cells with sand
       LDSs, possibly construction water.  Average monthly active-period LDS flow
       rates attributable to leakage through GM/GCL primary liners constructed with
       CQA will often be less than 2 Iphd, but occasionally in excess of 10 Iphd.
       Available data suggest that average monthly active-period LDS flow rates
       attributable to leakage through GM/CCL and GM/GCL/CCL primary liners
       constructed with CQA are probably similar to those for GM/GCL primary liners
       constructed with CQA.
                                   6-4
-------
•   Single liner systems with GM liners (installed on top of a relatively permeable
    subgrade) should not be used in applications where a true hydraulic efficiency
    above 90% must be reliably achieved, even if a thorough CQA program is
    employed. In these cases, single-composite liner systems or double-liner
    systems should be used. An exception to this may be made for certain facilities
    where electrical leak location surveys or ponding tests are used to identify GM
    defects and the defects are repaired. Higher true hydraulic efficiencies of 99%
    to more than 99.9% can be achieved by GM/GCL, GM/CCL, and GM/GCL/CCL
    composite liners constructed with good CQA.
•   Based on the existing data, GM/CCL and GM/GCL/CCL composite liners are
    capable of substantially preventing leachate migration over the entire period of
    significant leachate generation for typical landfill operation scenarios (i.e., for a
    landfill cell filled over a number of years, that does not undergo leachate
    recirculation or disposal of liquid wastes or sludges,  and that is capped with a
    final cover system designed to minimize percolation into the landfill; based on
    our existing understanding of their performance capabilities, these types of
    composite liners are capable of substantially preventing leachate  migration for a
    much longer  period, although field performance data of the type presented in
    this report do not yet exist for this longer period.
•   LCRS flow rates during operations (i.e., the initial  and active periods of
    operation) can vary significantly between landfills  located in the same
    geographic region and accepting similar wastes.  Large variations in flow rates
    (e.g., one order of magnitude difference) can even occur between cells at the
    same landfill.
•   LCRS flow rates were highest at the beginning of cell operations and decreased
    as waste thickness  increased and  daily and intermediate covers were  applied to
    the waste. Leachate generation rates decreased, on average, by a factor of four
    within one year after closure and by one order of magnitude two to four years
    after closure. Within nine years of closure, LCRS flow rates were negligible for
    the landfill cells evaluated in this study.
•   MSW cells produced, on average,  less leachate than HWand ISW cells.
•   For cells of a given waste type,  rainfall fraction (RF) values were highest in the
    northeast U.S. and  lowest in the west.
•   In general, HW landfills produced the strongest leachates and coal ash landfills
    produced the weakest  leachates.  MSW ash leachate was more mineralized
    than MSW leachate and the other  ISW leachates.
•   The solid waste regulations of the  1980s and 1990s have resulted in the
    improved quality of  MSW and HW landfill leachates.
•   The EPA HELP computer model, when applied  using an appropriate simulation
    methodology and an appropriate level of conservatism, provides a reasonable
    basis for designing LCRSs and  sizing leachate management system
    components. Use of the HELP  model for these purposes can  be enhanced
    through calibration to leachate generation rates at other landfills in the region
    and through parametric analyses that consider the potential range of values for
    key input parameters (e.g., initial moisture contents of waste).  Due to the
                                6-5
-------
       complexity and variability of landfill systems, however, the model will generally
       not be adequate for use in a predictive or simulation mode, unless calibration  is
       performed using site-specific measured (not default) material properties and
       actual leachate generation data.
   •   The frequency of occurrence of design, construction, and operational problems
       at landfills is significant.  The most common types of problems encountered
       involved liner system and final cover system slope stability.  Almost all of the
       problems were detected  shortly after they occurred, and  environmental impact
       due to the problems was only identified at one facility, which has since been
       remediated.  The main impacts of the  problems were interruption of waste
       containment system construction and  operation, increased maintenance, and
       increased costs.  Importantly, all of the problems identified in this investigation
       could have been prevented  using available design approaches, construction
       materials and procedures, and operation practices.

In light of the significant findings in each of the three areas of investigation,  it is obvious
that landfills are complex structures that require careful and  thorough design, testing,
construction and operation/maintenance. Procedures exist to avoid the types of issues
and problems identified in this report. Unfortunately, as most clearly demonstrated by
Appendix F of this report, landfill industry personnel  do not always utilize adequate
design, testing, construction, and operation/maintenance practices. The authors feel
strongly that current practices can and should be improved.  In the next four sections  of
this report, the authors highlight a number of areas related to landfill design,
construction, and operation where they believe practice improvement can be achieved
using readily available technology.

6.2  Liner Systems
Liner systems for the containment of solid waste consist of at least a low-permeability
barrier (liner) and an overlying LCRS. Depending on the nature  of the waste (and
obviously the pertinent regulations)  a single-liner system or a double-liner system with a
LDS between the two liners may be required.  In all cases, geosynthetics and/or natural
soils are typically utilized for the  liners, drainage layers, or both.  The design of these
multi-component and multilayered systems (see Figures 1-1 and 1-3) requires the
application of sophisticated engineering analysis methods. These systems also require
careful construction methods and CQA if they are to function as  intended.  This section
of the report is intended to  highlight several of the more important challenges faced by
engineers and contractors  in designing and constructing these systems. It is noted that
some of these challenges go beyond the tasks directly evaluated in this project;
however, these challenges are identified because they are important to waste
containment system performance.
                                    6-6
-------
6.2.1   Construction Quality Assurance
CQA has been shown to be of direct benefit in minimizing the potential leakage through
liner systems. This finding was originally put forth by Bonaparte and Gross (1990) on
the basis of sparse data and has been reinforced with the considerable additional data
generated since that time, including data presented in this study. Considerable
guidance exists for the development and implementation of liner system and cover
system CQA plans. Among the many requirements for such plans, the authors make
note of the following:

   •   soil and geosynthetic material conformance with the project specifications;
   •   proper pre-conditioning and placement of CCL lifts;
   •   proper compaction moisture content and density of CCLs;
   •   protection of CCLs from desiccation and freezing;
   •   placement of GMs without excessive waves and covering or backfilling the GMs
       in a manner that minimizes the trapping of waves; the goal of these measures is
       intimate contact between the GM and the underlying CCL or GCL;
   •   prevention of premature GCL hydration;
   •   inspection of GM seams, including nondestructive and destructive testing; and
   •   protection of GMs from puncture by adjacent materials or equipment.

6.2.2  Liner System Stability
This category of stability involves the liner system prior to waste placement.  The main
concern regarding liner system stability is for natural soils (particularly sand and gravel
drainage soils) or geosynthetics (particularly GTs and  GNs) to slide on underlying
geosynthetic surfaces. Sliding of drainage soils or sliding of drainage soils and GT
cushions on underlying GMs is unfortunately too common. The  instability is induced by
low shear  strength interfaces, steep and/or long slopes, equipment loads, seepage
forces, and/or seismic forces.  An area requiring particular attention is at access ramps
into below-grade landfills. These ramps are needed for operations, but are sometimes
overlooked in the assessment of landfill cell slope stability.  In some cases, ramps have
been installed by landfill operations personnel, without an evaluation of their effect on
liner system stability.  Another type of liner system  stability problem that requires careful
attention is sliding of GM layers on underlying CCLs or GCLs prior to waste placement.

Design of  liner systems for adequate slope stability is well within the design state-of-
practice. The available technical literature contains more than adequate  information to
design liner systems to be stable (see for example, Giroud and Beech, 1989; Koerner
and Hwu,  1991; Giroud et al., 1995; and Koerner and Soong, 1998). However, in the
authors' experience, the available methods are often not adequately utilized in design.
For example, it is not uncommon for seepage forces to be inadequately addressed
during the design process. Another significant design issue involves the  inadequate
characterization of interface shear strengths, apparently due to insufficient effort
                                   6-7
-------
expended in the laboratory evaluation of these strengths. Testing must be performed
under both project-specific and material-specific conditions, with considerable attention
given to the many variables that can influence the interface shear test results (e.g.,
boundary conditions, normal stresses,  hydration times, moisture conditions, and
displacement rates).  A number of papers, including those by Dove et al. (1997), Eid
and Stark (1997), Gilbert et al. (1997),  Sharma et al.  (1997), Sabatini et al. (1998),
Breitenbach and Swan (1999), and Sabatini et al. (2001), discuss variables that can
influence test results.

It is somewhat fortunate that many liner system slope stability problems can be repaired
at relatively small cost and with no environmental impact. This is particularly true of
slides that occur above the GM component of the liner system.  However, these facts
certainly do not justify a  less rigorous or careful design approach,  and overall
improvement in the rigor with which some owners and engineers address this design
issue is warranted.

6.2.3  Waste Stability
Of potentially greater significance than instability of the liner system before waste is
placed is a failure that occurs after waste has been deposited on top  of the liner system.
The Kettleman Hills landfill failure (Byrne et al.,  1992) is perhaps the  best known of this
type of occurrence. Design to resist this type of instability requires that the design
engineer specify acceptable waste configuration (e.g., intermediate slope angles) and
waste placement procedures, in addition to appropriately using slope stability analysis
methods and selecting liner system interface shear strengths.  For many facilities, waste
placement operations will need to be carefully sequenced.  Canyon-type landfills and
landfills built on soft foundation soils represent two classes  of facilities for which waste
mass stability deserves particular attention.

As with liner system stability, the technical analysis for waste mass stability is within the
state-of-practice, relying principally on  limit equilibrium slope stability  methods
developed in geotechnical engineering. The validity of the analysis is dependent  on the
choice of analysis methods, waste geometry and properties, interface shear strengths,
and moisture conditions in the landfill.  Particularly important with respect to waste
placement operations are the slope of the exposed surface of the waste, distance of this
exposed surface from the liner system  sideslopes, height of the waste, waste density,
and waste shear strength.  Discussion  of solid waste shear strengths to use in design
can be found in Kavazanjian et al. (1995). To help assure waste stability, the authors
recommend that the operations plans developed for landfills provide detailed criteria for
waste placement so that the landfill operator does not unknowingly fill the facility in a
potentially unstable manner.
                                    6-8
-------
Of particular importance in choosing waste and interface shear strengths is deformation
compatibility. It must be recognized that the amounts of deformation needed to
generate peak shear strengths in waste and along geosynthetic interfaces  are very
different. As discussed by Byrne (1994), Stark and Poeppel (1994), Gilbert et al.
(1997), and Sabatini et al. (2001), careful consideration must be given to the shear
strength deformation conditions used in design (i.e., peak, large displacement, or
residual).

It is interesting to note that several of the larger waste failures reported in the literature
occurred after periods of high rainfall, which had the effect of temporarily increasing the
density of the waste (Reynolds, 1991).  High rainfall can also impose seepage forces,
which will decrease stability accordingly.

Also important in some cases is seismic stability of the waste mass. While the
performance of several lined earthquakes in the 1994 California Northridge earthquake
was very good (Matasovic et al., 1995; Matasovic and Kavazanjian, 1996)  more needs
to be learned about this subject, particularly with respect to the seismic response of the
landfill and the determination of the acceptable magnitude of seismically-induced liner
system deformation. With respect to this latter criterion, it is the authors' experience
that design engineers often select a seismic deformation criterion of 150 to 300 mm
based on Seed and Bonaparte (1992).  However, these values may not be appropriate
in all applications. Careful consideration should be given to selection of an acceptable
level of deformation for design.  For example, all other factors being equal, a lower
allowable deformation should be used if the critical  interface is below the GM
component of the liner system (because excessive deformation would cause the GM to
rupture) than above  it.  Guidance on the seismic design of landfills can be found in
Richardson et al. (1995), Anderson and Kavazanjian (1995), and Kavazanjian (1998).

6.2.4  Performance of Composite Liner
For over a decade it has been known through theoretical analyses, laboratory tests, and
limited field data that composite liners are superior to either GMs alone or CCLs alone
for the containment of leachate or other liquids (Brown et al.,  1987; EPA, 1987; Giroud
and Bonaparte, 1989a,b; Bonaparte and Gross, 1990; Bonaparte and Othman, 1995).
This report has presented significant new field data that confirms the very good
performance characteristics of GM/GCL, GM/CCL,  and  GM/GCL/CCL composite liners
versus current types of single liner materials.

As discussed in Section 1.4.1.4, the basic premise of using a composite liner is that
leakage through a hole or defect in the GM upper component is impeded by the
presence of  a CCL or GCL lower component.  The GM  improves the performance of the
composite liner relative to that for a CCL or GCL alone by greatly limiting the portion of
the CCL or GCL exposed to leachate, and, for CCLs, lowering the potential for
                                   6-9
-------
desiccation cracking. Another benefit derived by using a composite liner is reduced
potential for diffusive transport through the liner. Diffusion is not an important transport
mechanism for inorganic ions through GMs.  However, as shown by Rowe (1998) and
others, diffusion rates of certain organic contaminants through GMs can be significant
when the concentrations of these contaminants are relatively high (i.e., diffusion through
GMs is generally not a concern at MSW landfills, but may be a concern at landfills or
impoundments where the liquid above the liner has relatively high concentrations of
volatile organic compounds). A CCL, and to a lesser extent GCL, component of a
composite liner will help retard organics that diffuse through the GM. Analysis methods
to design composite liners to account for diffusive transport are given by Rowe (1998).
If diffusion is the primary concern for a specific project, a CCL is preferred to a GCL as
the soil component of the composite liner.  To maximize retardation potential, adequate
thickness of CCLs is more important than low permeability.  One  approach to achieve a
composite liner with both low advective transport and low diffusive transport potential is
to specify a GM/GCL/CCL composite liner.  Giroud et al. (1997) present equations to
evaluate advective leakage rates through composite liners containing GCLs. While a
GM/GCL/CCL composite liner has advantages with respect to minimizing contaminant
transport potential, it may also create challenges with respect to slope stability factors of
safety. Shear strengths for both the GM/GCL interface and GCL/CCL interface require
careful evaluation.

6.2.5 Single vs. Double Liner System
As discussed in Section 1.2, federal regulations under Subtitle C  of RCRA require
permitted HWfacilities to be underlain by double-liner systems with leak detection
capability.  Also as discussed in Section 1.2, federal minimum design criteria for MSW
landfills include a single composite liner system. Several states have gone beyond
these minimum criteria for MSW landfills by requiring double-liner systems. A 1998
survey of 43 states has shown that for MSW landfills:

   •   31  (72%) states require single liner systems;
   •   6 (14%) states require double liner systems; and
   •   6 (14%) states provide options for the use of either a single liner or double liner
       system.

This survey highlights the differences in perspective (due to regional political
differences, population attitudes, hydrogeology, climate, drinking  water resources, and
other factors) between states as to the minimum requirements for liner systems at
RCRA Subtitle D landfills. These differences are even greater when it is realized that
the federal Subtitle D regulations contain both federal minimum design criteria and
performance-based criteria.  The performance-based critera require technical
demonstrations that are often made using the EPA HELP and MULTIMED computer
models, which do not address the potential for the migration of any landfill-generated
                                   6-10
-------
gas through the liner system. With respect to selection of the type of liner system for a
specific project, the authors offer the following thoughts:

   •   Caution should be exercised in using the EPA HELP  model to make a technical
       demonstration that the Subtitle D performance standard can be achieved with a
       liner system less (e.g., without a GM) than the federal minimum design criteria.
       Input parameters to the model can be selected to demonstrate a lesser potential
       for leachate generation than actually exists.  For example, the discussion in
       Chapter 5 of this report indicated that modeled leachate generation rates are
       sensitive to the assumed initial moisture content of the waste.  Because of the
       sensitivity of the HELP model results to the input parameters, when the model is
       used to make a technical performance demonstration, the model should be
       calibrated against  data (i.e., LCRS flow rates) from lined landfills in the same
       geographic area.  In addition,  the potential for  landfill  gas impacts to
       groundwater should also be considered as part of the technical demonstration.
   •   Based on the landfill operation data presented in this report, Subtitle D single-
       composite liner systems meeting federal  minimum design criteria can achieve a
       very high hydraulic efficiency and are capable  of preventing adverse impacts to
       groundwater. This conclusion is consistent with the previous conclusion
       reached by EPA regarding the performance capabilities of liner systems meeting
       federal minimum design criteria.
   •   Caution should be exercised in substituting a GCL alone for the CCL as the low-
       permeability soil component of a Subtitle D single-composite liner on the base of
       a landfill. While the hydraulic efficiency of a GM/GCL composite liner is as
       good, or better, than a GM/CCL composite liner, the GM/GCL  composite liner is
       more susceptible to diffusive transport  (Rowe,  1998)  and puncture than the
       GM/CCL composite liner. These concerns are less important for sideslope
       areas of the landfill where leachate heads are  lower;  thus, a GM/GCL composite
       liner is more likely  be appropriate for sideslopes than for base  areas from a
       hydraulic perspective. Also, a GM/GCL/CCL composite liner may be an
       effective low-permeability soil component for a single-composite liner.  In this
       case, it may be acceptable to specify a maximum hydraulic conductivity on the
       order of 1 x 1f>5 cm/s for the CCL of a  three-component composite liner used at
       MSW landfills.
   •   There may exist situations for MSW landfills where a  double-liner system would
       be preferred to a liner system meeting  the federal minimum design criteria.  In
       addition to the obvious situation where a  state regulation requires use  of a
       double-liner system, the project conditions favoring selection of a double-liner
       system include: (i) sites with especially vulnerable hydrogeology; (ii) sites where
       groundwater cannot be reliably monitored due to the  presence of complex
       hydrogeology, karst,  or other factors; and (iii) sites where, for whatever reason,
       a higher degree of reliability/redundancy  is required of the liner system than can
       be achieved by the Subtitle D federal design criteria.  In some  cases, it may be
       desirable to use a  double-liner system  beneath the base of the landfill, and,  for
       cost-effectiveness, a single-composite liner system beneath the sideslopes.
                                   6-11
-------
   •   The authors endorse a design strategy of providing additional protection at
       critical locations in the landfill. For a landfill underlain by a single-composite
       liner system, this strategy might take the form of installing a GM/GCL/CCL liner
       beneath each landfill sump. This design feature adds little cost to the overall
       project, but significant benefit in terms of design reliability and redundancy at the
       project's most critical location.
   •   Where double-liner systems are used, the authors prefer that the secondary
       liner be a GM/CCL or GM/GCL/CCL composite to a GM alone or a GM/GCL
       alone. Furthermore, the authors prefer a GM/GCL composite liner to either a
       GM alone or GM/CCL composite liner as the primary liner component of the
       double-liner system.  The preferred type of composite primary liner is much
       easier to construct on top of the LDS and secondary liner than is a GM/CCL
       composite primary liner and it minimizes LDS flow rates compared to the rates
       associated with the other types of primary liners considered in this report.
   •   Notwithstanding the specificity of the recommendations given above, the authors
       do endorse the use of creative thinking and good engineering to develop better-
       performing, more cost-effective liner systems.  The key to this approach,
       however, should be good engineering, and not, for example, manipulation of the
       HELP model or any other data or design tool to achieve a pre-conceived desired
       outcome.

6.2.6  Fate of Liner Systems
Of critical importance to the long-term performance of a liner system is the service life of
the GM component of the system.  Both CCLs and GCLs consist of geologic materials.
These materials can be expected to have service lives in excess of the design lives  that
have been defined for MSW, HW,  and LLRW disposal facilities constructed in the
United States.  This conclusion is only valid, however, if these materials stay hydrated
and stable, are adequately protected, and, for GCLs, are not subjected to unacceptable
chemical interactions (i.e., an increase in hydraulic conductivity of bentonite may result
if sodium in the bentonite is replaced by other cations present in the permeant). For the
CCL or GCL component of a composite liner beneath a waste mass these conditions
will often be met.

Perhaps the most important factor governing the service life of GMs is the polymer type,
and resin formulation. Of the variety of choices, HOPE is the most widely used polymer
and the resin formulation includes carbon black and an antioxidant package. As
described in this report and in Hsuan and Koerner (1998), lifetime predictions are
measured in terms of hundreds of years (but not "forever"). However, this report has
also pointed out that additional research in this area is warranted.

Perhaps the most significant issue  related to the use of HOPE GMs is the potential for
premature stress cracking before the end of its design life.  Currently, the notched
constant tensile load (NCTL) test is used with a minimum onset of brittle behavior of 100
                                   6-12
-------
hours.  (This is equivalent to a single point value, per ASTM D5397, of 200 hours). At
the designer's discretion, these values can be increased, and, depending on site-
specific conditions, this is encouraged.  Regarding HOPE formulations, the antioxidant
package included in the formulation is critically important, and specifications should
include a minimum OIT along with a minimum OIT retained value after oven aging and
laboratory simulated UV exposure.

6.3 Liquids Management
The liquids management strategy for a landfill generally refers to all liquids including:

   •   leachate collection and removal at the bottom of the waste mass, above the
       primary liner system;
   •   leakage collection and removal at the bottom of the waste mass between the
       primary and secondary liners;
   •   rainwater collection and removal via the final cover system drainage layer above
       the barrier material;
   •   gas condensate collection and removal via the gas collection piping system;  and
   •   groundwater collection and control via the pore pressure relief system in areas
       of high groundwater.

For the first three systems, drainage layers transmit liquid by gravity to a low point
where the liquid empties into a sump or gravity drain or is discharged from the waste
containment system, in the case of a final cover system drainage layer. In the case of a
sump, the liquid is withdrawn using submersible pumps or bailers.  For a gravity drain,
the liquid flows by gravity through a pipe that penetrates the liner system and
discharges to a storage or treatment system outside the limits of the landfill. From final
cover system  drains, the liquid flows by gravity either as sheet flow to the surrounding
land, or, more typically, into a perimeter stormwater collection and conveyance
structure.  For gas condensate collection and removal systems, liquids collected  in gas
collection piping systems typically drain to a low point in the piping system. From this
location, condensate is usually introduced back into the waste; however, sometimes
condensate is removed from the waste containment system and treated. With respect
to pore pressure relief system, these systems may consist of a series of wells or
perimeter trenches that are pumped to lower the groundwater table or may include a
drainage layer and sump installed beneath the liner system.

These liquid collection and removal systems were discussed in Section 1.4.2 of this
report.  That discussion is not repeated in this chapter.
                                   6-13
-------
6.3.1   Construction Quality Assurance
The authors provide the following commentary on CQA of liquids management systems:

   •   Placement of natural drainage soils on previously placed geosynthetics should
       be done with great care.  Minimum thicknesses of soil should be maintained
       between construction equipment and the underlying geosynthetics.  Only low
       ground pressure equipment should be allowed to spread the soil. Particular
       attention should be focused on preventing accumulation of excess slack and the
       propagation of waves in the underlying GM.
   •   When shallow trenching is conducted within a natural drainage soil in order to
       place plastic pipes above the previously installed GM, the trenching should be
       conducted by hand, with special lightweight backhoes equipped with rubber
       bucket tips, or by an alternate method that will not damage the GM.
   •   Pipe fittings and joints need careful observation during installation to confirm that
       the fittings are mated with one another and are fabricated by the same
       manufacturer.
   •   Sumps, sideslope or vertical riser pipes, and pipe penetrations through GMs
       should be carefully constructed since they are located in areas with the highest
       sustained hydraulic head on the liner. To this end, GM  seams in sumps should
       be limited to the extent possible.  The placement of an enhanced liner in the
       sump (e.g., GM/GCL/CCL instead of a GM/CCL) should also be considered. In
       addition, extra care should be taken to protect the liner around these features  by
       ensuring that required cushion layers are properly placed and that construction
       equipment maintains adequate separation from the GM.
   •   GN placement should be observed to ensure that no gaps exist between roll
       ends and edges and that sufficient plastic ties are used per the specifications.
   •   GNs must be carefully inspected for excess soil particles, fugitive bentonite from
       GCLs, vegetative growth, or other foreign matter, and these materials should be
       removed.  Flushing of water through the installed GN could be considered .
   •   Full coverage of required GT filters over drainage layers should be provided
       whether the drain is natural soil or a GN.
   •   If select waste is to be placed directly over the drainage stone (with no
       intervening filter layer), its placement should have full-time inspection by a CQA
       monitor to assure that the underlying materials are not disturbed.

6.3.2  Potential lor Clogging and Reduction of Flow Capacity
An important question regarding drainage systems is whether or not the system will clog
excessively. The phrase "excessively" is used because all drainage systems will,  over
time, undergo a decrease in their flow capability from the original installation or
manufacture; the issue is to what degree. The authors provide the following
commentary and recommendations on this topic:

   •   For LCRSs, the authors believe that the often cited regulatory value for drainage
       layer hydraulic conductivity of 1 x 10~2 cm/s for natural soil is too low in many
                                   6-14
-------
       applications. Furthermore, a value of 1 x 10~3 cm/s, which  is sometimes
       specified, will almost always be too low.  Hydraulic conductivities at these values
       result in drainage layers with substantial liquid storage (capillary) capacity and
       slow drainage rates. These conditions result in increased hydraulic head on the
       liner and, consequently, increased potential for clogging and leakage.  Design of
       LCRSs should be performed on a site  specific basis, using an adequate factor of
       safety. The soil should be free draining, with few fines, and little or no capillarity.
   •   For design of LCRSs, the HELP model can be an appropriate design tool for
       estimating leachate generation rates (see Chapter 5).  As previously indicated,
       however, HELP model results are sensitive to the input parameters provided.
       The authors believe design engineers  can do much more to calibrate their HELP
       model runs using data from already active landfills in the region.  In this regard,
       design engineers and landfill operators are encouraged to collect and
       disseminate this information.
   •   Landfill LCRS design should include not only an evaluation of leachate quantity,
       but also leachate quality.  This report presents considerable new data on landfill
       leachate characteristics.  From a design perspective, it is important to identify
       conditions (e.g., sludge co-disposal, special waste disposal) that would create a
       leachate with more than usual potential to clog a drainage layer.  For example,
       Koerner et al. (1994) identified leachate with high TSS and/or BOD5  values
       (e.g., above 10,000 to 15,000 mg/l) as a condition requiring special design
       consideration.  Interestingly, in the study of liquids management data described
       in Chapter 5, none of the landfill cells for which leachate chemistry data are
       available had average BOD5 values greater than 5,000 mg/l.
   •   For the internal drainage layer in a final cover system, water is the medium
       being transmitted and clogging of the drainage layer by water is generally not
       considered.  The primary issue for this layer is inadequate drainage capacity
       and the  buildup of seepage forces in the final cover system, leading to slope
       instability. A significant number of seepage-induced final cover system failures
       were identified in Chapter 5. The HELP  model must be used with caution to
       calculate liquid heads in the final cover system drainage layer, as experience
       has shown that these heads may be underpredicted if the peak daily rainfall
       used in the model is too low. Guidance on using the HELP model for this
       purpose is given in the upcoming EPA technical  guidance document titled,
       "Technical Guidance for RCRA/CERCLA Final Covers" (Bonaparte et al., 2002).
       Also, the manual procedure in Koerner and Daniel (1997) can be used to
       estimate liquid heads in the final cover system drainage  layer.

6.3.3 Perched Leachate
Perched leachate (which does not have full hydraulic connection to the underlying
LCRS) can occur as a result of a number of conditions in a landfill.  Excessively clogged
filters above the drainage layer, low-permeability buffer (or protection) soils placed
above the LCRS, low-permeability daily cover, and high moisture content sludges
(industrial or sewage) within the waste mass all can lead to the trapping of moisture in
                                   6-15
-------
pockets within the waste.  The perched leachate can increase the unit weight of the
waste and impact waste stability. Saturated conditions within the zone of perched
leachate will inhibit the generation of landfill gas and reduce the effectiveness of gas
extraction wells in the area. In addition, the "breakout" of perched leachate as seeps
has contaminated nearby surface waters, created odor problems, and killed vegetation.

6.3.4 Fate of Liquids Management Systems
Data in Chapter 5 and Appendix E indicate that in modern landfills, within a few  years to
a decade after a final cover system with a GM  is placed, LCRS flow rates become very
low to negligible. At a minimum, an LCRS should be designed for the anticipated peak
leachate flows during the facility's active life assuming  that some amount of clogging will
occur (the amount being specific to the anticipated leachate and the details of the
design). The authors recommend, however, that the system also be designed to retain
a significant flow capacity even after closure. The rationale for this recommendation is
that it provides the ability to continue to collect  and remove leachate at some future
date, should that need ever become necessary. The future need could arrive out of an
unforeseen, even if improbable, future event such as undetected damage to the final
cover system that allows significant new infiltration into the landfill.  The need could also
arrive out of a planned future event, such as the future use of a closed landfill cell as a
bioreactor or the placement of additional waste as in "piggybacking" operations. Finally,
the authors encourage the increasing tendency of design engineers to design LCRSs to
performance levels better than those required by regulatory  minimums.  Examples
include specifying higher permeability natural drainage materials (e.g., materials having
hydraulic conductivities of at least five to ten times larger than the regulatory minimum
of 1x10-2 cm/s specified by some states for MSW landfills), and designing to a  lower
maximum leachate head than the maximum allowed by regulation (e.g., designing to a
maximum leachate head of 0.03 or 0.1  m, rather than the maximum value  of 0.3 m
allowed by Subtitle C and Subtitle D regulations).

The internal drainage layer above the hydraulic barrier in a final cover system must
function for as long as the final cover system is required. Thus, the design must result
in a stable hydraulic condition within the final cover system over its' design life.   This will
typically require careful selection of the protection soil above a filter or drainage  layer so
that hydraulic equilibrium can be established (i.e., so that particles of the protection soil
are retained by the filter or drainage layer without clogging the layer).

6.4  Final Cover Systems
Conventional final cover systems placed over solid waste typically consist  of a barrier
material, internal drainage layer, cover soil, and surface material. Regulatory
requirements for these conventional systems were discussed in Section 1.2 of this
report. Increasingly, alternative design concepts are being applied to final cover
systems. These alternative concepts include evapotranspiration (ET) or capillary
                                   6-16
-------
barriers, rather than low-permeability hydraulic barriers such as GMs, CCLs, and GCLs.
ET and capillary barrier cover systems are finding increasing use at arid and semi-arid
sites.  These alternative cover systems are discussed in detail in the upcoming EPA
technical guidance document titled, "Technical Guidance for RCRA/CERCLA Final
Covers" (Bonaparte et al., 2002).

Both geosynthetics and natural soils are commonly used in final cover systems. Great
care is required during both design and construction in order to achieve adequate
performance.  While many of the authors' comments in Section 6.2 of this report on liner
systems also apply to the final cover systems, there are several differences between the
two.  For cover systems in comparison to liner systems:

   •   the barrier is meant to keep liquid  out of the waste mass, rather than containing
       liquid within;
   •   the liquid to be managed is infiltrating rainwater (and snow melt) which
       percolates through the cover soil rather than leachate;
   •   upward rising gases from the waste may need to be captured beneath the
       barrier and effectively transmitted  for proper management;
   •   the upward rising gases usually contain volatile constituents from the leachate,
       albeit at low concentrations for landfills (though potentially at higher
       concentrations at remediation sites), thus chemical mass transport and chemical
       compatibility of systems in contact with the gas should be considered;
   •   final cover systems slopes may be relatively steep and long, resulting in
       significant  slope stability design issues;
   •   final cover systems are subjected  to different environmental stresses than liner
       systems; these stresses include freeze-thaw and desiccation-wetting cycles; and
   •   the impact of waste settlements, both total and differential, on final cover system
       integrity should be considered for  proper design of all system components.

Several of the more important issues with respect to design, construction, and
maintenance of landfill final cover systems are discussed below.

6.4.1  Construction Quality Assurance
It seems  intuitive that if proper CQA produces improved performance for liner systems,
the same will be true for final cover systems.  The authors believe that in addition to the
CQA items for liner systems mentioned in Section 6.2.1 of this report, the following
items  require special attention when performing CQA of final cover systems:

   •   evaluation of the subgrade upon which the final cover system is to be placed to
       assure adequate bearing capacity and that buried waste will not damage
       overlying final cover system components;
   •   careful construction according to the design details for connections of GMs and
       GCLs to pipe vents;
                                   6-17
-------
   •   attention to construction details of the gas drainage layer (beneath the barrier)
       connection to the vent system to prevent GM blowouts;
   •   careful construction for connections of cover system internal drainage layers to
       their outlets;
   •   attention to proper location of haul roads and access roads according to lines
       and grades of the plans; and
   •   inspection of the erosion control features to verify that measures have been
       taken to obtain a healthy vegetative cover before the conclusion of construction
       activities.

6.4.2 Compacted Clay Barriers
There are serious concerns with respect to the use of CCLs in final cover systems,
particularly when used alone. These concerns are as follows:

   •   based on the  case histories presented in Section 4.3 of this report, desiccation
       of CCLs is a distinct long-term possibility in almost every final cover system
       application where the CCL  is not covered by a GM; it has been shown that upon
       rewetting, a desiccated CCL does not return to its original low hydraulic
       conductivity;
   •   the freeze-thaw sensitivity of CCLs has been demonstrated in laboratory studies
       whereby the CCL hydraulic conductivity is increased significantly and self-
       healing of the thawed CCL  is not likely (Othman et al., 1994);
   •   as discussed  in Section 4.3 of this report, there are documented cases of
       moisture migration through some CCLs used in final cover systems due to CCL
       degradation;
   •   depending upon the thickness and properties of the final cover system materials
       above the CCL, root penetration of CCLs may occur; these roots can cause the
       development of channels for water migration into the underlying waste;
   •   again depending upon the thickness and properties of the final cover system
       materials above the CCL, burrowing animal intrusion into CCLs is a possibility;
       animal intrusion could lead to relatively large pathways for water migration into
       the underlying waste mass; and
   •   distortion of CCLs due to total and (more importantly) differential settlement of
       the underlying waste may lead to CCL tensile strains that exceed the ultimate
       tensile strain by orders of magnitude; based on studies by Leonards and Narain
       (1963), Ajaz and Parry (1975a,b, 1976), and others, most CCLs tested under
       unconfined or low confinement conditions exhibit failure  at extensional strains of
       0.5% or less.

For these reasons, it  is felt that CCL barriers should typically not be used alone in the
final cover systems of landfills (particularly MSW landfills, which contain wastes that
undergo significant settlement) and that GMs or GCLs, by themselves, or as part  of a
composite cap, will typically be preferable.
                                   6-18
-------
6.4.3 Final Cover System Stability
Notwithstanding the availability of proven slope stability design methods (e.g., Koerner
and Hwu, 1991; Giroud et al., 1995), the sliding of cover soils on underlying
soil/geosynthetic and geosynthetic/geosynthetic interfaces has been a relatively
common problem for landfill final cover systems.  In evaluating final cover system
stability, consideration must be given to a variety of potential destabilizing forces (i.e.,
the gravitational mass of the cover soil,  equipment loadings, seepage forces, and
seismic forces).  As for liner systems, attention to detail by a qualified design engineer
has sometimes been lacking. This attention to detail should apply to the selection of the
input parameters to the slope stability analysis, to the evaluation of seepage, seismic,
and/or equipment forces to be applied to the cover system, the factor of safety used in
the analysis, and the analysis itself.  As for the evaluation of liner system stability, it is
recommended that the shear strengths of cover system materials and interfaces be
evaluated using the results of project-specific laboratory shear tests conducted in a
manner to simulate the anticipated field conditions.

In the experience of the authors, factors that contribute to the observed high frequency
of final cover system slope failures include:

   •   relatively steep slopes with long uninterrupted surfaces; these conditions can be
       mitigated by using flatter slopes, benches, intermediate berms, and/or tapered
       cover soil thicknesses;
   •   equipment loadings, which can be minimized by limiting the ground pressure of
       equipment and orienting the equipment in predetermined (and properly
       designed) paths; the  effect of even  low ground pressure equipment on cover
       system stability should be checked  by the design engineer;
   •   build-up of seepage forces within the drainage layer and/or cover soils due to
       inadequate drainage capacity, which is often the result of not performing a water
       balance for the internal drainage layer  and evaluating the potential for seepage
       forces; if the HELP model is used to estimate seepage forces, considerable care
       is needed in selecting a design storm event and other input parameters that do
       not lead to an underestimate of liquid head buildup in the drainage layer; as
       previously noted, the manual calculation method of Koerner and Daniel (1997)
       can also be used to estimate liquid  heads;
   •   inadequate design of drain transitions and outlets, such that water backs up in
       the drain and causes a buildup of pore pressure within the cover soil mass; and
   •   instability caused  by seismic forces, which is clearly a site-specific situation and
       one requiring careful  design and interpretation; paradoxically,  current regulations
       require seismic design of many MSW landfills but do not do so for HW landfills
       or abandoned landfills.

6.4.4 Cover Soil Erosion
The evaluation of cover soil erosion is also an  important step in the design of a landfill
cover system.  A possible design strategy to avoid seepage forces within a cover soil is
                                    6-19
-------
to use low permeability materials thereby preventing infiltration. Steep slopes, long
uninterrupted slope lengths, and/or poor vegetative cover all tend to increase runoff and
the potential for erosion of cover soils. The authors would like to highlight the following
design considerations with respect to cover soil erosion:

   •   temporary erosion control materials should be used more widely than current
       practice so as to minimize erosion until a healthy stand of vegetation is obtained;
   •   cover system construction projects often conclude in the late fall, with little time
       to establish vegetation before the end of the growing season; this condition
       should be avoided, if possible; it is critical to have a good stand of surface
       vegetation  prior to the end of the growing season if the potential for severe
       winter erosion is to be avoided;
   •   a careful choice of vegetation is critical in providing year-round protection of
       topsoil; a diverse  mixture of native vegetation that closely emulate a selected
       local "climax" community is preferred;
   •   channelization of  runoff is critical, and the design of soft or hard armor surface
       drainage swales and channels is necessary;  let-down chutes represent a
       particular design challenge due to the high water velocities that occur on steep
       slopes; and
   •   erosion control in  arid and semi-arid sites takes a completely different strategy
       than just described; the use of hard armor surfaces, particularly rock riprap, is
       common, with the selection of rock size being an important design output.

6.4.5 Fate of Final Cover Systems
Final cover systems play  a critical long-term role at landfills.  A properly functioning final
cover system will largely eliminate the long-term post-closure leachate generation
potential at solid waste landfills where there is no other input source for liquid (which is
usually the case).  If liquids are prevented from entering the waste mass, there is, in the
long term after the waste has biodegraded and/or stabilized, no significant potential for
continuing leachate and/or gas generation. Thus, the barrier component of the final
cover system should be as durable as the liner component of the liner system. Current
regulations call for a 30-year post-closure care period, and many design engineers
assume that this time frame represents the required design life for the final cover
system barrier.  The authors of this report recommend that the barrier in the final cover
system generally be designed for a longer design life, for example, a 100-year design
life.  The authors also offer the following  observations:

   •   The choice of GM resin in a final  cover system is influenced  by a number of
       factors.  For MSW landfills where settlement potential is significant, high out-of-
       plane deformation capability is a  desirable characteristic. This design criterion
       favors VFPE, fPP and PVC GMs. Long-term durability considerations favor
       HOPE (recall Sections 2.5 and 6.2.6), which does not perform well above the
       pressure rate of 7 kPa/min given in the out-of-place deformation test ASTM
                                    6-20
-------
D5617.  At lower pressure rates, where stress relaxation can occur, the situation
is different but the test is rarely conducted in a slow strain rate or creep mode.
In the current state-of-practice, chemical compatibility is rarely considered for
final cover system GMs since the upper surface of the GM is only exposed to
water infiltrating the cover soil. However, the lower surface of the GM may be
exposed to landfill gas, which invariably contains low concentrations of volatile
components present in the leachate. Thus, chemical  resistance is an issue that
should be considered based on site-specific conditions.
Both durability and chemical compatibility are issues with respect to the
reinforcing fibers or yarns of reinforced GCLs placed on sideslopes.  While the
GCL test plots described in Chapter 3 go far to show the validity of such GCL
reinforcement, GCLs have not been installed for a long enough time to
demonstrate the adequacy of this reinforcement over  a 30 or 100-year time
frame.
The design of internal drainage layers in final cover systems is too often
inadequate, i.e., the flow capacity is too low and outlets and transitions do not
have adequate flow capacity.  The potential for fines migration through the
drainage layer filter is not always considered.  The potential for freezing or other
blockage of the  drainage layer outlets is sometimes not assessed.
The design of final cover systems in seismic impact zones requires careful
consideration. The potential for amplification of free-field ground motions by the
waste mass combined with low shear strength geosynthetic interfaces makes
seismic performance an important consideration. EPA guidance (Richardson et
al. 1995) and Anderson and Kavazanjian (1995) provide procedures for
evaluating the potential for seismically-induced final cover systems
deformations. Considerations applicable to seismically-induced deformations of
liner systems (discussed in Section 6.2.3) are also applicable to final cover
systems.  An  additional consideration for final cover systems is that in high
seismic zones (e.g., near major active faults in  California), it may not be feasible
to design sloping final cover systems containing geosyntnetics to sustain non-
damaging deformations during major earthquakes.  As discussed by
Kavazanjian (1998), in these circumstances, it may be appropriate to design the
final cover system to an acceptable damage criterion. Acceptable damage
levels would be  based on preventing adverse environmental impact, cost of
repair, ease of repair, and any other impacts associated with the damage (e.g.,
loss of serviceability).  This approach would necessitate development of a
detailed post-earthquake response action plan  coupled with financial
assurances to provide the required funds to make the repairs at the time when
they are needed.
The fact that the waste mass is subsiding over time means that sideslope
angles are progressively decreasing. The amount is waste-dependent,  but the
mechanism is one that tends to progressively increase final cover system slope
stability factors of safety.
                            6-21
-------
6.5 Gas Management
The degradation of any putrescible organic fraction of the solid waste in a landfill
produces a number of gases. The condition is mostly applicable to MSW, but other
types of waste may also produce some type of gas by biological or chemical means.
The anaerobic decomposition of MSW produces two principal gases, methane and
carbon dioxide, in roughly equal quantities (i.e., 40 to 50% each of total gas volume)
and much smaller quantities of other gases. The gases produced in a MSW landfill is
generated over a relatively long period of time, especially for landfills at arid sites.
Using the EPA LandGEM computer program with Clean Air Act (CAA) default
parameters for gas generation at a temperate  site and considering an increment of
waste placed in one year, 10% of the gas from this waste is produced within two years
of waste placement, 40%  is produced  within 10 years of waste placement, and 80% is
produced within 30 years.  For an arid site, 20% of the gas from the waste is produced
within 10 years of waste placement, 45% is produced within 30 years, and 80% is
produced within 80 years.  The  gases  move within and from the landfill primarily by
convection, but also by diffusion. Gas emissions from MSW landfills are currently
governed by the RCRA Subtitle D regulations, which address the personal and
fire/explosion aspects of landfill gas, and the CAA regulations, which regulate emissions
of non-methane organic compounds as a surrogate to total landfill gas emissions.
Under the CAA, MSW landfills greater than a certain size must collect and combust their
landfill gas.

Some design engineers collect and vent or extract MSW landfill gases with vertical,
perforated collection wells (typically 5 wells per 2 hectares) without a continuous gas
transmission layer beneath the barrier system. This approach  can be justified if the
waste itself is sufficiently permeable to gas, if the gas wells are relatively closely
spaced, or, perhaps, at arid sites, where gas is generated relatively slowly. With gas
wells, the gas moves within the waste  to the perforations in the pipe and then flows or is
drawn out of the system. Another approach to venting or extracting gas from a  landfill
involves installing a continuous  gas transmission layer beneath the final cover system
barrier layer.  Shallow gas venting or extraction pipes will tie into the gas transmission
layer. Gas collection trenches with periodic vent or extraction pipes represents  a third
approach to gas collection beneath the final cover system.  Also, a combination of
these three gas venting/extraction systems can be used.

In any case (deep wells penetrating the waste, a continuous gas transmission layer
beneath the final cover  barrier layer, and/or collection trenches), the system outlets are
typically plastic pipes extending up through the final cover system.  Gas flow through
the pipes can be either  passive (vented to the atmosphere or flared) or active (collected
through a header using a blower system to create a small vacuum).  Without a gas
management system, gas pressure will build up in the landfill.  Note that with a GM in
the final cover system and relatively small cover soil thicknesses, gas pressures can
                                   6-22
-------
cause GM uplift.  Even if the GM is not physically lifted, positive gas pressure beneath
the GM can lower the effective stress at the interface between the GM and underlying
material (e.g., GCL), thereby reducing interface shear strength and potentially
contributing to a slope failure.

6.5.1   Construction Quality Assurance
As with all aspects of a waste containment system, CQA plays an important role in
achieving acceptable performance of a gas management system. For deep wells, the
number, location and extent of the pipe perforations are important.  Also, the wells must
be kept safely above the liner system beneath the waste.  Several examples exist
where gas well borings have extended into the liner system because of inadequate
survey control and not accounting for landfill settlement. For continuous gas
transmission layers beneath the barrier, continuity is important for either soil or
geosynthetic gas transmission layers.  If the latter, the material is often a GN with GTs
bonded to both sides. The overlapping of the GN along its edges and ends is important
as well as its joining with plastic ties per the specifications.  Both upper and lower GTs
need to be continuous with generous overlaps (often 300 mm) or sewn together to
prevent soil from entering and clogging the GN.

Lastly, the penetration of gas wells or vents through a GM barrier should have tightly
fitting prefabricated boots.  Unlike boots for liner penetrations at the bottom of the
landfill, boots for the final cover system GM must be designed to function while
accommodating the anticipated landfill settlement.  GCL tie-ins have similar
considerations.

6.5.2  Gas Uplift
As indicated above,  when using a GM in an MSW landfill final cover system, gas uplift
pressures will be exerted on the GM unless the gas is efficiently conveyed to the wells,
vents, or collection trenches.  If gas is not adequately managed,  uplift pressure will
either cause GM bubbles (or "wales") to occur displacing the cover soil and appearing at
the surface, or it will decrease the normal stress between the GM and  the underlying
material.  At several facilities, this latter effect has led to slippage of the GM and
overlying cover materials creating high tensile stresses as evidenced by compression
ridges in the cover soil and folding of the GM at the slope toe and tension cracks in the
cover soil near the slope crest.  Three situations need careful design consideration:

   •   if gas removal is by deep wells, the uppermost pipe perforations should be
       effective in capturing gas in the upper layers of waste;
   •   if gas removal is by a gas transmission layer beneath the GM and vents, the gas
       transmission layer should be designed with adequate long-term transmissivity;
       and
                                   6-23
-------
   •   if gas is removed by horizontal collection trenches, some of the trenches should
       be placed in close proximity to the bottom of the final cover system to prevent
       gas accumulation and uplift pressure on the cover system GM.

6.5.3   Landfill Settlement
The design of final cover system drainage systems and gas collection systems
(including the gas wells, vents, and/or collection trenches, and the network of piping for
gas and condensate transmission systems) is complicated by the magnitude of waste
settlement that typically occurs at solid waste landfills. Post-closure total settlement
may equal 10 to 20% of the landfill height for MSW landfills and up to 20 to 30% of the
waste height for some abandoned dumps. The design, construction, and  maintenance
of both the final cover system and gas management system must take these
settlements into account. Figure 6-1 illustrates the magnitude of post-closure
settlements that can occur at MSW landfills.  The settlement magnitudes given in this
figure should be considered to represent the upper range of values potentially
applicable to modern landfills because the database used to develop the figures
includes data for not only MSW landfills, but also abandoned dumps. Of equal concern
(but largely unquantified) is the differential settlement that may occur in isolated areas  of
the landfill.
      CO
      0)
      .
      o

      0)
      E
      0)
      "CD
                          10
       100

Time in Days (log)
1000
10000
Figure 6-1. Total post closure settlement data for MSW landfills and abandoned
           dumps [after Edgers et al. (1990); Konig et al. (1996); Spikula (1996)].
                                  6-24
-------
The time frames over which both total and differential settlement may occur are quite
long and depend on the many factors including the liquids management strategy
practiced at the site. Table 6-1 presents a framework for evaluating likely post-closure
total and differential settlements at MSW landfills and abandoned dumps.
Table 6-1. Impact of Liquids Management Practice on Final Cover System
           Settlement at MSW Landfills and Abandoned Dumps(1-2) (Koerner and
           Daniel, 1997).
Leachate
Management Practice
Standard leachate
withdrawal
Leachate recirculation
Total Settlement
Amount
10-20%
10-20%
Time
< 30 yrs.
< 1 5 yrs.
Differential Settlement(3>4)
Amount
Little to moderate
Moderate to major
Time
< 20 yrs.
< 10 yrs.
None, e.g., at abandoned   Up to 30%   > 30 yrs.   Unknown              > 20 yrs.
  landfills or dumps
1HW landfills, ISW landfills, and MSW ash monofills usually have much less settlement than the amounts
 listed in this table.
2The estimates in this table regarding the impact of the liquids management practice on settlement of
 landfill final cover systems are based on sparse data. They are meant to be a guide only, and site-
 specific estimates are required to develop more appropriate figures for any particular final cover system
 project.
3The estimates in this table regarding differential settlement amount and time are also based on
 very sparse data.  Clearly, field monitored data is needed in this regard.
4These qualitative assessment terms are also affected by the density of the waste; well-compacted
 waste produces less differential settlement than poorly-compacted waste.
6.5.4  Landfill Fires
While the incidence of landfill fires in MSW landfills has greatly diminished since the
days of the "open dump", they still sometimes occur. Air-to-methane mixture ratios of
20 to 50% have given rise to at least one fire, which damaged a geosynthetic final cover
system.  The vulnerable time frame of a facility with respect to landfill fires appears to
be after the GM is seamed and before cover soil is placed. Wind uplift of the GM  can
draw air  in through vents providing the oxygen necessary to create ignitable conditions.

Fires at depth within a waste mass may occasionally occur. The origin of such fires is
apparently spontaneous combustion and an air source is required for sustenance. The
key to preventing such a fire is to block air entry.  Identifying and blocking all potential
sources of air entry can sometimes be difficult.
                                    6-25
-------
6.5.5 Fate of Gas Management Systems
For large regionalized landfills where energy is utilized there is an incentive to maintain
the gas management system in good working order. When the energy conversion
becomes inefficient, however, the wells or vents may be decoupled from their external
piping systems and be allowed to vent to the atmosphere.  It is important, at that time, to
show the amount of gas being vented is below regulatory limits and does not present a
health or an environmental hazard.  It is also important to show that gas emissions
through the final cover system in the vicinity of the decoupled well or vent are below
regulatory limits.

6.6  Long-Term Landfill Management
The performance data for operating landfills presented in this report demonstrate that
landfills can be designed, constructed, and operated/maintained to achieve very high
levels of leachate and landfill gas containment and collection. The report has also
demonstrated that design, construction, and operation/maintenance issues and
problems persist at many landfills.  In the preceding  part of this chapter,  the authors
have attempted to provide guidance to design engineers on how to avoid the most
significant issues and problems that may typically arise.  Information on the anticipated
service lives of the various engineered components  of a landfill waste containment
system was also given.

The ultimate degradation of any individual waste containment system component of a
landfill after the completion of that component's useful service life may or may not lead
to a release of leachate or gas and contamination of groundwater.  Furthermore, a
release may, or may not, result in a significant environmental impact. In evaluating the
consequences of ultimate degradation,  the design engineer must consider a wide range
of factors including: the climatological and hydrogeologic setting; the composition, age,
and level of degradation of the waste; the potential for leachate and gas generation after
the component has completed its service life; the potential to maintain, rehabilitate, or
install other systems to achieve leachate and gas containment;  and collection, cost, and
social and institutional factors. These various factors should be considered within an
overall decision-making framework that addresses long-term landfill management.
Long-term landfill management strategies are discussed in Appendix G.

6.7  References
Anderson, D.G. and Kavazanjian, E., Jr. (1995), "Performance of Landfills Under
  Seismic Loading," Proceedings of the 3rd International Conference on Recent
  Advances in Geotechnical Earthquake Engineering and Soil Dynamics, St. Louis,
  Missouri, pp. 1557-1587.
Ajaz, A. and Parry,  R. H.  G. (1975a), "Stress-Strain  Behavior of Two Compacted Clays
  in Tension and Compression," Geotechnique, Vol. 25, No. 3, pp. 495-512.
                                   6-26
-------
Ajaz, A. and Parry, R. H. G. (1975b), "Analysis of Bending Stresses in Soil Beams,"
   Geotechnique, Vol. 25, No. 3, pp. 586-591.
Ajaz, A. and Parry, R. H. G. (1976). "Bending Test for Compacted Clays," Journal of
   the Geotechnical Engineering Division, Vol. 102, No. 9, pp. 929-943.
Bonaparte, R. and Gross, B.A. (1990), "Field Behavior of Double-Liner Systems,"
   Proceedings of the Symposium on Waste Containment Systems, ASCE Geotechnical
   Special Publication No. 26, pp. 52-83.
Bonaparte, R., Gross, B. A., Daniel, D. E., Koerner, R. M., and Dwyer, S. F. (2002),
   "Technical Guidance for RCRA/CERCLA Final Covers," U.S. EPA Office of Solid
   Waste and  Emergency Response, Washington, D.C. (in final review)
Bonaparte, R. and Othman, M. A. (1995), "Characteristics of Modern MSW Landfill
   Performance," Geotechnical News, Vol. 13, No. 1, pp. 25-30.
Breitenbach, A. J. and Swan, R. H.  (1999), "Influence of High Load Deformations on
   Geomembrane Liner Interface Strengths," Geosynthetics '99 Conference
   Proceedings, Industrial Fabrics Association International, St. Paul,  MN, Vol. 1, pp.
   517-529
Brown, K. W., Thomas,  J. C., Lytton, R. L, Jayawickarama, P. and Bahrt, S. C. (1987),
   "Quantification of Leak Rates Through Holes in Landfill Liners," report for cooperative
   agreement  CR-810940, EPA Risk Reduction Engineering Research Laboratory,
   Cincinnati, 147 p.
Byrne,  J. (1994),  "Design Issues with Strain-Softening Interfaces in Landfill  Liners,"
   Proceedings, Waste Tech '94, National Solid Waste Management Association,
   Charleston, 26 p.
Byrne,  R. J., Kendall, J. and Brown, S, (1992), "Cause and Mechanism of Failure,
   Kettleman Hills Landfill B-19, Unit 1A," Stability and Performance of Slopes and
   Embankments-ll, ASCE Geotechnical Special Publication No. 31,  R. B. Seed and R.
   W. Boulanger, eds.,  pp. 1188-1215.
Dove, J. E., Frost, D. J., Bachus, R. C. and Han, J. (1997), "The Influence of
   Geomembrane Surface Roughness on Interface Strength," Proceedings,
   Geosynthetics '97 Conference, Long  Beach, CA, Mar., Vol. 2, pp. 863-876.
Edgers, L., Noble, J. J. and Williams, E. (1990), "A Biologic Model for Long Term
   Settlement in Landfills, " Tufts University, Medford, MA.
Eid, H. T. and Stark, T. D. (1997), "Shear Behavior of an Unreinforced Geosynthetic
   Clay Liner," Geosynthetics International, Vol. 4, No. 6, pp. 645-659.
EPA (1987), Proposed Rulemaking, 40 CFR Parts 260, 264, 265, 270 and 271, "Liners
   and  Leak Detection for Hazardous Waste Land Disposal Units," Federal Register,
   Vol.  52, No. 103, pp.  20218-20311.
Gilbert, R. B.,  Scranton, H. D. and Daniel, D. E. (1997), "Shear Strength Testing for
   Geosynthetic Clay Liners," Testing and Acceptance Criteria for Geosynthetic Clay
   Liners, ASTM STP 1308, Larry W. Well, Ed., American Society for Testing and
   Materials.
                                  6-27
-------
Giroud, J. P., Bachus, R. C. and Bonaparte, R. (1995), "Influence of Water Flow on the
  Stability of Geosynthetic-Soil Layered Systems on Slopes," Geosynthetics
  International, Vol. 2, No. 6, pp. 1149-1180.
Giroud, J. P., Badu-Tweneboah, K. and Soderman, K. L. (1997), "Comparison of
  Leachate Flow Through Compacted Clay and Geosynthetic Clay Liners in Landfill
  Liner Systems," Geosynthetics International, Vol. 4, Nos. 3 and 4, September, pp.
  391-431.
Giroud, J. P. and Beech, J. F. (1989), "Stability of Soil Layers on Geosynthetic Lining
  Systems," Proceedings, Geosynthetics '89 Conference, Vol. 1, San Diego, pp. 35-46.
Giroud, J. P. and Bonaparte, R. (1989a), "Leakage Through Liners Constructed with
  Geomembranes, Part I:  Geomembrane Liners," Geotextiles and Geomembranes,
  Vol.8, No. 1, pp. 26-67.
Giroud, J. P. and Bonaparte, R. (1989b), "Leakage Through Liners Constructed with
  Geomembranes, Part II:  Composite Liners," Geotextiles and Geomembranes,  Vol.
  8, No. 2, pp. 77-111.
Hsuan, Y. G. and Koerner, R. M. (1998), "Antioxidant Depletion Lifetime in High Density
  Polyethylene Geomembranes," Journal of Geotechnical and Geoenvironmental
  Engineering, ASCE, Vol. 124, No. 6, pp. 532-541.
Kavazanjian, E., Jr. (1998), "Current Issues in Seismic Design of Geosynthetic Cover
  Systems," Proceedings of the 6th International Conference on Geosynthetics,
  Atlanta, pp. 219-226.
Kavazanjian, E., Jr., Matasovic, N., Bonaparte, R. and Schmertmann,  G. R. (1995),
  "Evaluation of MSW Properties for Seismic Analyses," Proceedings of the ASCE
  Specialty Conference Geoenvironment 2000, ASCE Geotechnical Special Publication
  No. 46, Y.B. Acarand D.E. Daniel, eds., Vol. 2, pp. 1126-1141.
Koerner, G. R., Koerner, R. M. and Martin, J. P. (1994), "Geotextile Filters Used for
  Leachate Collection Systems:  Testing, Design of Field Behavior", Journal of
  Geotechnical Engineering, ASCE, Vol. 120, No. 10, pp. 1792-1803.
Koerner, R. M. and Daniel, D. E. (1997), "Final Covers for Solid Waste Landfill and
  Abandoned Dumps," ASCE Press, 256 pgs.
Koerner, R. M. and Hwu, B. L. (1991), "Stability and Tension Considerations Regarding
  Cover Soils in Geomembrane Lined Slopes," Geotextiles and Geomembranes, Vol.
  10, No. 4, pp. 335-355.
Koerner, R. M. and Soong, T. Y. (1998), "Analysis and Design of Veneer Cover Soils,"
  Proceedings, Sixth International Conference on Geosynthetics," Industrial Fabrics
  Association International, St. Paul, MN, Vol.  1, pp. 1-26.
Konig, D., Kockel, R. and Jessberger, H. L. (1996), "Zur Beurteilung der Standsicherhert
  und zur Prognose  der Setzungen von Mischabfalldeponien," Proceedings 12th
  Nurnberg Deponieseminar, Vol. 75, Eigenverlag LGA, Nurnberg, Germany,  pp. 95-
  117.
Leonards, G. A. and Narain, J. (1963), "Flexibility of Clay and Cracking of Earth Dams,"
  Journal of the Soil Mechanics and Foundations Division, Vol. 89, No. 2, pp.  47-98.
                                  6-28
-------
Matasovic, N., Kavazanjian, E., Jr., Augello, A. J., Bray, J. D. and Seed, R. B. (1995),
  "Solid Waste Landfill Drainage Caused by 17 January 1994 Northridge Earthquake,"
  The Northridge, California Earthquake of 17 January 1994, California Department of
  Conservation, Division of Mines and Geology Special Publication No. 116, M.C.
  Woods and R.W. Sieple, eds., pp. 221-229.
Matasovic, N. and Kavazanjian, E., Jr. (1996), "Observations of the Performance of
  Solid Waste Landfills During Earthquakes," Proceedings of the 11th World
  Conference on Earthquake Engineering, Elsevier Science Ltd. Paper No. 341, 8 p.
  (on CD ROM).
Othman, M. A., Benson, C. H., Chamberlain, E. J., and Zimmie, T. F. (1994),
  "Laboratory Testing to Evaluate Changes in Hydraulic Conductivity of Compacted
  Clays Caused by Freeze-Thaw:  State-of-the-Art," Hydraulic Conductivity and Waste
  Containment Transport in Soils, STP 1142, D.E. Daniel and S.J. Trautwein (eds.),
  American Society for Testing and Materials, Philadelphia, PA, pp. 227-254.
Reynolds, R. T. (1991), "Geotechnical Field Techniques Used in Monitoring Slope
  Stability at a Landfill," Proceedings Field Measurements in Geotechnics, ed. G.
  Sorum, Rotterdam:  A. A. Balkema, pp. 883-891.
Richardson, G. N., Kavazanjian, E., Jr. and Matasovic, N. (1995), "RCRA Subtitle D
  (258) Seismic Design Guidance for Municipal Solid Waste Landfill Facilities," U.S.
  Environmental Protection Agency Report No. 600/R-95/051, 143 p.
Rowe, R. K. (1998), "Geosynthetics and the Minimization of Contaminant Migration
  through Barrier Systems Beneath Solid Waste," Proceedings, 6th International
  Conference on Geosynthetics, Atlanta, pp. 27-102.
Sabatini, P. J., Schmertmann,  G. R. and Swan,  R. H. (1998), "Issues in
  Clay/Geomembrane Interface Testing," Proceedings of the 6th International
  Conference on Geosynthetics, Atlanta, pp. 423-426.
Sabatini, P. J., Griffin, L. M., Bonaparte, R.,  Espinoza, R. D., Giroud, J. P. (2001),
  "Reliability of State-of-Practice for Selection of Shear Strength Parameters for Waste
  Containment System Stability Analysis,"  Proceedings, GRI-15 Conference on Hot
  Topics in Geosynthetics - II (Peak/Residual; RECMs; Installation Concerns),
  Geosynthetic Research Institute, pp 86-109.
Seed, R. B. and Bonaparte, R. (1992), "Seismic Analysis and Design of Lined Waste
  Fills: Current Practice," Stability and Performance of Slopes and Embankments - II,
  ASCE Geotechnical Special Publication No. 31, pp. 1521-1545.
Sharma, H. D., Hullings, D. E., and Greguras, F. D. (1997), "Interface Strength Tests
  and Application to Landfill Design, " Geosynthetics '97 Conference Proceedings,
  Industrial Fabrics Association International, St.  Paul, MN, Vol. 2, pp.  913-926.
Spikula, D. (1996), "Subsidence Performance of Landfills: A 7-Year Review,"
  Proceedings GRI-10 Conference on Field Performance of Geosynthetics and
  Geosynthetic Related Systems, Geosynthetic Research Institute, Philadelphia, pp.
  237-244.
                                   6-29
-------
Stark, T. D. and Poeppel, A. R. (1994), "Landfill Liner Interface Strengths from
  Torsional-Ring Shear Tests," Journal of Geotechnical Engineering, Vol. 120, No. 3,
  pp. 597-615.
                                   6-30
-------
               Appendix A
Behavior of Waves in High Density
   Polyethylene Geomembranes
                      by

             Robert M. Koerner, Ph.D., P.E.
                  Drexel University
               Philadelphia, PA 19104
                  performed under

           EPA Cooperative Agreement Number
                  CR-821448-01-0
                  Project Officer

                 Mr. David A. Carson
        United States Environmental Protection Agency
           Office of Research and Development
        National Risk Management Research Laboratory
                Cincinnati, OH 45268
-------
                               Appendix A
                           Behavior of Waves in
               High Density Polyethylene Geomembranes
A-1  Overview and Focus
Geomembranes (GMs) form the essential material component in many liner systems
which require a liquid or vapor barrier. Such applications are landfill liners, landfill
covers, liquid impoundment liners, and other waste pile liners. The usual assumption in
the placement of such liners is that they lay flat on the subgrade beneath them, e.g., on
the underlying compacted clay liner, geosynthetic clay liner, etc.  This is sometimes not
the case. Waves, or wrinkles, of different sizes can occur in the as-placed and seamed
GMs, see Figures A-1  and A-2.  These waves have given design engineers a certain
amount of concern as to the behavior of GMs after soil backfilling or covering. The
research study described in this appendix was developed to shed insight into the issue
of GM wrinkles.
Figure A-1. Relatively small waves, or wrinkles, in a field deployed GM.
                                    A-1
-------
Figure A-2.  Relatively large waves, or wrinkles, in a field deployed GM.
The approach to this study of the behavior of GM waves involved an extensive series of
laboratory tasks.  It is important to note that the purpose of the tests was to evaluate the
behavior of GM waves under field stresses. The tests were not designed to try to
quantify the effects of waves on hydraulic containment performance. The scope of the
laboratory testing program involved an assessment of the effects of the following four
variables on wave behavior:

      (a) normal stress;
      (b) original wave height;
      (c) thickness of GM; and
      (d) temperature.

Due to its common use in a variety of waste containment systems, high density
polyethylene (HOPE) GMs were used throughout the study.  In particular, one
manufacturer's commercially available GM was used. The only GM variable considered
was thickness. In all other cases, the thickness was maintained at 1.5 mm, which  is a
commonly used HOPE GM thickness in many applications.

GM waves, such as seen in Figures A-1  and A-2, can be classified as two different
types: thermally-induced GM waves and construction induced GM waves.
                                      A-2
-------
Thermally-induced waves in GMs are created due to the thermal expansion
characteristics of GMs after they are seamed together and before backfilling occurs.
These types of waves have been observed in GMs for many years (Schultz and Miklas,
1980). The height and/or width of the wave depends on the GM type (e.g.,  modulus,
thickness, surface texture, surface color), temperature difference after seaming and
before backfilling, and distance between points of fixity, e.g., previously backfilled
locations.

As an illustration of a thermally induced wave, a 30 m long section of 1.5 mm thick
HOPE GM (with a thermal expansion coefficient of 15 x 10~5/°C) undergoing a
temperature change from 15°C installation temperature to 50°C (sheet surface) under a
summer sun, will expand the following amount:

            AL  =AT(a)L = (50-15)(15x 10'5) (30) = 0.158 m = 158mm

Obviously, such thermally induced GM waves can be created in the field via the local
ambient conditions.

Alternatively, construction induced GM waves are sometimes created purposely.  In
North America, an adequate amount of slack is sometimes left in the GM liner to
compensate for the coldest temperatures envisioned (EPA, 1993). The philosophy is
that the majority of the slack will be removed when the GM is covered and the sheet
temperature is reduced.  Ultimately, when the envisioned coldest temperature is
reached, the rest of the built-in slack will be completely removed, therefore,  intimate
contact to the underlying soil will be achieved.

In order to estimate the wave dimensions that can be created by a given amount of
slack in GMs, Figure A-3 was developed. In the figure, the slack in the GM  which
results in the creation of waves with various height-to-width ratios is plotted  as a
function of wave height.

As seen in Figure A-3, a  slack of 158 mm, as calculated  in the earlier example, can
create the following different wave patterns:

      • a 120 mm high wave with height-to-width ratio of 1.0;
      • a 165 mm high wave with height-to-width ratio of 0.5;
      • a 215 mm high wave with height-to-width ratio of 0.33; or
      • a 265 mm high wave with height-to-width ratio of 0.2
                                      A-3
-------
            500.
            400.
   Slack in
Geomembrane
    (mm)
300.
            200.
            100.
              0
                                                   Height-to-Width
                                                        Ratio
                                                         1.0
                                                         0.5

                                                         0.33
                                                         0.2
                 0
             50      100      150     200
                     Height of Wave (mm)
                     250
          300
Figure A-3. Slack in GM resulting in the creation of waves with various height-to-
width ratios.
Alternatively, if the GM is relatively flexible or thinner than in the previous example, two
or even more smaller waves can be created within the same slack in the GM.  Table
A-1  summarizes the types of multiple waves that can be produced by a 158 mm
expansion of a GM.  In an actual facility, an expansion of this magnitude can certainly
create waves of the type seen in the photographs of Figures A-1 and A-2.  While this
example is based on the coefficient of thermal expansion/contraction of 1.5 mm thick
HOPE GMs,  it should be noted that all types of GMs currently used in the waste
containment  industry have similar values of coefficient of thermal expansion/contraction
(Koerner, 1998).

Table A-1. Types of Waves Produced by a Slack of 158 mm in a GM in a Distance
of 30 m with a Temperature Difference of 35°C
  Height-to-Width
   Ratio (H/W)
      Single Wave Height
             (mm)
Two Waves Height
      (mm)
Three Waves Height
       (mm)
1.00
0.50
0.33
0.20
120
165
215
265
60
80
105
130
40
55
70
90
                                     A-4
-------
However, such an ideal situation of a perfectly flat GM is very difficult to achieve.  The
reasons are as follows:

   • it is very difficult to quantify the actual difference between the two extreme
     temperatures, i.e., the installation temperature and the coldest temperature
     envisioned;
   • the focus of concern is the sheet temperature, not the ambient temperature;
   • the sheet temperature is a complicated function of ambient temperature,  surface
     color and texture, incidence of sun, weather, etc.;
   • accurate measurement of the coefficient of thermal expansion is difficult;
   • frictional forces mobilized between the interfaces can retard, or even constrain, the
     reduction of slack; and
   • it is extremely difficult to build into an installed and seamed GM, a prescribed
     amount of slack.

As a result, relatively large waves of the type seen in Figure A-2 are commonly seen in
the field.

It has been observed both in the field and in the laboratory that the installed wave
greatly distorts from its original shape under increasing normal stress. However, the
deformation pattern depends on the GM type  (e.g., thickness, modulus, flexural rigidity,
etc.), the original wave shape (e.g., height, height-to-width ratio, etc.), and the
surrounding environment (e.g., stress level, duration, temperature, etc.).

Figure A-4 illustrates some possible deformation scenarios.  Figure A-4a shows how the
wave distorts under relatively low normal stress. Figure A-4b shows that the profile of
the wave remains almost unchanged when  higher normal stress is applied.  This could
possibly be the case for waves in relatively thick and/or stiff GMs. When normal stress
is applied nonuniformly (e.g., with a horizontal component), the waves may roll over
towards one side as seen in Figure A-4c. When normal stress is applied to waves in
relatively thinner or more flexible GMs, they may become vertically flattened as seen in
Figure A-4d.  For extremely flexible GMs, they may even be flattened in a pancake
manner as seen in Figure A-4e.  Note that conditions as shown in Figure A-4d and e are
also possible when the service temperature is relatively high.

The concern as to the ultimate fate of GM waves should certainly receive attention as to
a rigorous understanding of the problem.  However, to date, all analyses and
investigations into GM waves have been semi-qualitative,  see Giroud and Morel (1992)
and Giroud (1995). Quantitative approaches which evaluate the  ultimate fate of GM
waves in a more rigorous manner are needed. For instance, there could be a  maximum
wave height, for a given set of conditions, where the GM wave will eventually
                                      A-5
-------
                            (a) under low normal stress
                       r\
                (b) after initial distortion            (C) one sided roll-over
                 (d) vertically flattened           (e) horizontally flattened

Figure A-4. Possible deformation scenarios in GM waves.
lay flat on the soil.  Given a maximum wave height, which could be specified in the
installation contract, the optimal fate of GM waves might be as follows:

   1.  The installer seams the GM with waves up to a maximum specified amount.
   2.  Typically a geotextile (GT) will cover the GM and the temperature of the GM will
      decrease. Thus, the wave(s) will decrease in size (Koerner and Koerner, 1995).
   3.  Upon backfilling over the GT covered  GM, the waves are fixed in position and
      contained by friction from further size  reduction stemming from future decreasing
      temperature.
   4.  Under increasing normal stress, due to soil, solid waste or liquids, the wave
      distorts from its original shape. As seen in Figure A-5, from results of this study,
      the wave becomes narrower in width at its base and only marginally shorter in its
      height. Thus, the wave's height-to-width ratio  is actually accentuated from  its
      initial condition (from approximately 0.33 to 0.44 for this particular GM).
   5.  Over time, creep and/or stress relaxation in the polymer structure occurs and the
      wave height decreases in size thus reducing the  H/W ratio.
   6.  Ultimately, it is hoped that the wave flattens to a H/W ratio of zero, so as to
      achieve contact with the underlying soil subgrade.

From the above description it is suggested that creep and stress relaxation play a key
(and essentially unknown) role in the ultimate elimination of GM waves.  Furthermore,
by knowing the characteristics of the "entombed" wave,  one can  possibly back-calculate
to the originally allowable maximum wave height.
                                       A-6
-------
Figure A-5.  Wave distortion under increasing normal stress from large-scale
laboratory experiments conducted in this study.
These issues then frame the essence of this study.  It is focused completely on GM
waves which exist in the GM at the point of backfilling and are caused by elevated
temperature above that which existed when the GM rolls were seamed.  In this study,
only GMs made from HOPE are evaluated. This is felt to be justified since HOPE
represents approximately 70% of the landfill liner market in North America. In other
countries, e.g., Germany, it is the only type of GM that is allowed.

A-2 Experimental Setup and Monitoring
A large-size experimental test box was constructed in the laboratory for the evaluation
of the behavior of HOPE GM waves.  Initially, the test box was utilized to conduct
preliminary tests to gain a better understanding of the problem to be investigated.  It
was then used for the justification of performing smaller scale experiments.  Finally, it
was designated for conducting a 10,000-hour control test. Details regarding each of  the
these items will  be presented after a description of the test box.

A photograph and schematic illustration of the test box is shown in Figure A-6.  The
basic components of the setup include a rigid box and a data acquisition system.  The
box has dimensions  of 1.8 m long by 1.0 m wide by 1.0 m high.  On the front panel,
there  is a 0.5 m-wide plexiglass window for the  purpose of visual observations. An air
bag which provides a uniform normal pressure up to 70 kPa is placed on top of the soil
and the reaction is transmitted through a 25 mm thick wooden board to five steel
reaction cross beams connected at the top of the box.  Also,  a  number of electrical
resistance strain gages are bonded on the test specimen at various locations with wires
extended out of the box and connected to a data acquisition system.
                                      A-7
-------
                                          1.8m
   1.0m
                                                                     0.6 m
        Void,



Sand above--
                                                                     0.4m
                                                                 — *"
                                                           To arata acqusition
Figure A-6. Photograph and schematic illustration of the large-scale experimental

test box used in this study.
                                     A-8
-------
The experimental monitoring of the behavior of HOPE GM waves includes two parts:
profile-tracing of the actual wave and strain gage monitoring. The profile-tracing
provides the opportunity of visual observation and recording the distortion of GM waves
under various experimental conditions. Important information such as the final
configuration, the final height-to-width ratio, and the locations of stress concentrations
can be obtained using this type of monitoring. Tracing the profile of GM waves is done
via the window on the front panel of the test box. An example of profile-tracing was
shown in Figure A-5. This type of monitoring was also performed routinely on trial runs
before the actual experiments began to determine the layout pattern of the other type of
experimental monitoring, i.e., the strain gage monitoring.

Strain gage monitoring quantifies the actual strain induced at different locations of the
GM wave under various experimental conditions.  When used in  conjunction with a data
acquisition system, this type of monitoring provides reliable information on the
experiment over the duration of the test. The strain gages used in this study are
electrical resistance (foil-type) strain gages having resistance of 120-ohms and gage
length of 12.7 mm. With proper configuration, this particular type of gage measures
strain within the range of ±5%.  The installation procedure recommended by the gage
manufacturer was precisely followed.  The surface cleaning and preparation was
considered most critical in this regard.  The photograph of an installed strain gage is
shown in Figure A-7. Note that a bondable terminal along with two curved "jumper
wires" are also used in the gage installation to prevent the gage from being  subjected to
any unexpected stresses.
                                     C onnection
                                      Terminal
Figure A-7. Strain gage with soldered connection installed on GM specimen.
                                       A-9
-------
Sets of preliminary tests were conducted using the large-scale experimental setup.
These tests were performed at an early stage of the task and were designed to gain
further understanding of the GM wave, as well as to evaluate the possibility of
transferring the large-scale tests to a small-scale experimental setup.  The material
used in these tests was a 1.5-mm thick smooth  HOPE GM.  Details of the preliminary
tests are presented as follows.

The first series of preliminary tests consisted of three separate experiments.  Namely, a
1.5-mm thick GM with relatively large, moderate, and relatively small waves.  The
waves were created by using specimens longer than the inner length of the test box.
After each specimen was placed  in the box, sand backfilling was started from the end to
the center of the  box in a symmetrical manner.  Consequently, the "slack" of specimen
was "pushed" toward the center and, as a result, a wave was formed.  The "original"
configuration of waves was defined as the wave profile under approximately 100 mm of
sand backfill. Using profile-tracing as previously described,  the shape of the original
wave was recorded. The same type of monitoring was repeated at various stages of
the backfilling and progressed until the maximum normal pressure provided by the
experimental setup, i.e., 70 kPa, was reached.

The results of the profile-tracing monitoring of these tests are shown in Figure A-8.  In
the figure, the outermost curves of all tests represent the original wave configuration
and the innermost curves correspond to the final wave profiles under 70 kPa.  As seen
in the figure, under increasing normal stress, the waves greatly distort from their original
shapes.  The waves become narrower in width at the base but only marginally shorter in
height. A quantitative parameter was devised by calculating a ratio of the wave  height
to its base width, i.e., a H/W ratio. For purposes of gaining perspective with field
installations, a somewhat accepted rule-of-thumb in  the field deployment of GMs is that
the height-to-width  (H/W) ratio should not be greater than 0.5. This being the case, the
"relatively large" and "moderate" waves in this study were already marginal from the
outset. The accentuated H/W ratios upon backfilling, 2.0, 0.9 and 0.4  as seen in Figure
A-8, were already considered as a valuable finding in the course of this study.

Even further, with respect to the empirical field guide, the drastic increase in the H/W
ratio for the "relatively large" and  "moderate" waves  indicated locations of high curvature
and therefore the possibility of high stress concentrations under even higher normal
stresses. Such waves should clearly be removed before the placing of backfill.  As a
result, the rest of this task focused on waves with an original height smaller than the
height of the relatively small wave shown in Figure A-8.

Also seen in Figure A-8 are the reference marks located at various portions of the
waves. These marks are very helpful in tracking the critical  locations of a wave  under
normal pressure with respect to the undeformed test specimen. Therefore, the
information will be used to establish the layout pattern of strain gage installation.
                                      A-10
-------
 (a) Relatively Large Wave
^\
Original
Final
Ht.
240 mm
125 mm
H/W
0.5
2.0
250 mm
                                                                       200 mm
                                                                       150 mm
                                                                       100 mm
                                                                       50 mm
                                                                       0 mm
(b) Moderate Wave
^\
Original
Final
Ht.
130 mm
70 mm
H/W
0.4
0.9
250 mm
                                                                       200 mm
                                                                       150 mm
                                                                       100 mm
                                                                       50 mm
                                                                       '0 mm
 (c) Relatively Small Wave
^\
Original
Final
Ht.
80 mm
35 mm
H/W
0.25
0.4
250 mm
                                                                       200 mm
                                                                       150 mm
                                                                       100 mm
                                                                       50 mm
                                                                       0 mm
Figure A-8.  Results of the profile-tracing monitoring of three preliminary tests.
                                     A-11
-------
As mentioned earlier, the pressurizing mechanism (i.e., the air bag and reaction beams)
in the large-scale test box can only provide a uniform normal pressure up to 70 kPa.  If
the average unit weight of typical solid waste is assumed as 12 kN/m3, such a normal
pressure is approximately equivalent to solid waste of 6 m in height.  This is relatively
low for a typical landfill. In order to evaluate the behavior of GM waves under high
normal pressures, e.g., greater than 1,000 kPa, transferring the experiments to smaller
setups which allow the application of higher normal pressures is necessary. Moreover,
smaller setups which can be housed in a environmental room will be especially
beneficial since the effect of temperature on the behavior of GM waves can then be
investigated.  However, such smaller tests must be justified on the basis  of this larger
test setup.

A small-scale setup justification test was designed and conducted to examine the
behavior of the wave itself. A wave, identical to the relatively small wave shown in
Figure A-8, was created in the large-scale test box. However,  instead of being
supported by the side walls of the test box, both ends of the specimen were held by
metal sticks 50 mm away from the walls of the box. In addition, both ends of the test
specimen were covered by 75 mm-wide smooth HOPE GM strips acting  as protective
slip-sheets. The experimental setup is shown in Figure A-9.
1 mm HOPE
strips on bot
                 50 mm
          c
                                          Air bag
                        GM
                         sides
                                             Supporting sticks
                                             removed after box
                                             is filled with sand
                                       reference marks
            u
Figure A-9.  Justification experiment for small-scale experimental setups.
                                      A-12
-------
Before the backfilling process was started, a 300 mm by 300 mm square region was
marked on the window of the test box. It was used as a virtual image of a smaller test
box in which the GM wave could be housed.  Two reference marks, immediately
adjacent to the square region, were made on the front edge of the wave, as seen in
Figure A-9. With the supporting sticks on both ends of the test specimen, backfilling
was carefully carried out until the test box was filled.  The supporting sticks were then
removed, leaving two horizontal spaces of 50 mm each on both ends of the specimen
(protected by the slip sheets) for possible lateral movement.

The GM wave was then pressurized using the air bag against the reaction beams. It
was observed that under a normal pressure of 70 kPa, the wave distorted in a manner
exactly like the relatively small wave shown in Figure A-8.  Moreover, the two reference
marks  remained completely stationary, i.e., there was no lateral movement of the GM.
This observation suggests that the frictional forces, mobilized  between the GM
specimen and the adjacent sand fill, were sufficient to restrict the horizontal portions of
the GM from any lateral movement and decrease in wave height.

In other words, the mobilized friction forces on the horizontal extensions of the wave
offered the same reaction as would a smaller test box simulated by the 300 mm by 300
mm square region. This important finding not only provided the justification of using a
smaller scale test  box, it also justified the use of both experimental setups, large and
small scale, to simulate situations in the field where the HOPE GMs waves are normally
much further apart.

Based on the above findings, four rigid boxes having dimensions of 300 mm long by
300 mm wide by 300 mm high were built.  Along with steel reaction frames and a
hydraulic pressurizing system, these boxes allow a application of normal pressure
higher than 1,500  kPa.  This is equivalent to a solid waste landfill of approximately 125
m in height, i.e., a so-called "megafill". In addition, all four boxes can be simultaneously
housed in a environmental room where constant environmental conditions can  be
maintained within  ranges of 0 to 55°C temperature and 0 to 98%  relatively humidity.
Photographs of one of four identical small scale test boxes and the environmental room
used in this task are shown in Figure A-10. As seen in the figure, data acquisition is
also available for strain gage measuring.

One of the objectives of the experimental  part of the task is to investigate the behavior
of HOPE GM waves under various conditions. As discussed earlier, the four small-
scale test boxes in conjunction with the environmental room are ideal in this regard.
The other objective of the experimental part of this study is to  obtain actual long-term
experimental data so that the validity of using rheologic models for the purpose of long-
term prediction can be evaluated.
                                      A-13
-------
Figure A-10.  Photographs of the small scale test box and the environmental room
used in this study.
Four sets of 1,000 hour experiments, utilizing the small scale test boxes just described
within an environmental room, were designed and conducted to evaluate the effect of
four experimental parameters on the behavior of HOPE GM waves.  These parameters
were the normal stress, original height of wave, thickness of GM, and testing
temperature. Table A-2 presents the experimental design of these tests. As seen, the
effects of different variables were evaluated by varying the particular one under
investigation while holding the others constant.  In all cases, smooth HOPE GMs were
used and strain gages were attached to the wave specimens at different locations with
continuous readout over the duration of the tests.  Note that all of the waves in the
experiments listed in Table A-2 were created with an original height-to-width ratio of
approximately 0.33.  Such a ratio was found typical for most of the naturally formed
HOPE GM waves in the laboratory covered by little-to-no backfill.

The large-scale test box was reserved and used for conducting a single long-term
(10,000 hours) control experiment.  A 1.5-mm thick smooth HOPE GM wave with
original height of 60 mm and a original height-to-width ratio of 0.33 was created and it
was subjected to a constant normal stress of 70 kPa at a temperature of 23±2°C. This
test is considered to be the control test for subsequent comparison of the results of the
small-scale tests.
                                     A-14
-------
Table A-2.  Experiments Conducted Using Small Scale Test Boxes	
 Experimental  	Experimental Conditions	
   Parameter     Normal Stress   Original Height       GM         Temperature
   Evaluated        (kPa)          of Wave     Thickness  (mm)       (°C)
                                    (mm)

Normal
Stress

180
360
700
1,100

60



1.5 23


 Original Height
    of Wave
      GM
   Thickness
    Testing
 Temperature
700
700
700
14
20
40
60
80
60
14
20
40
60
1.5
1.0
1.5
2.0
2.5
1.5
23
23
23
42
55
A-3  Experimental Results -1,000 hour Tests
The results of all twenty five of the 1,000 hour tests, as listed in Table A-2, will be
presented in this section. They will be given on a variable-by-variable basis.  Both
original and final (after 1,000 hours) shapes of the GM waves along with the
corresponding heights and height-to-width ratios will be shown.  Also, if applicable, a
comparison among results generated under different test conditions will be made to
evaluate the effect of that particular experimental variable.

As listed in Table A-2, four 1.5 mm thick HOPE GM wave specimens, having original
heights of 60 mm, were subjected to four different normal stresses, namely, 180, 360,
700,  and  1,100 kPa. The temperature was maintained at 23°C for all experiments over
the entire duration of the experiments, i.e., 1,000 hours.  The original (same for all
specimens) and the final shapes of all test specimens, obtained via profile-tracing
monitoring, are shown in Figure A-11.

Six strain gages,  numbered  from G1 to G6, were originally bonded at  the locations
shown in Figure A-11  for all specimens.  Note that gages G4 to G6 (shown as darker
circles in  Figure A-11) were bonded on the lower side of the GM since the gages which
                                     A-15
-------
Figure A-11. Original and final shapes of HOPE GM waves under various normal
stresses (grid lines have dimensions of 10 mm by 10 mm).

were used respond more accurately under tension than compression. As a result of
different normal stresses, these gages measured the strains corresponding to various
locations on the GM test specimens. A typical result of the test conducted under a
normal stress of 700 kPa is shown in Figure A-12 where the measured strains are
plotted against time. By viewing Figures A-11 and A-12 simultaneously, it is seen that
the upper portion of this particular wave specimen experienced measurable strain with a
maximum tensile strain of 3.4% recorded near the crest of wave.
   Strain (%)
               0
200
800
1000
                                    400        600
                                     Time (hours)

Figure A-12. Strain measurement results of experiment conducted at 700 kPa.
                                    A-16
-------
By investigating the results generated from both parts of the experimental monitoring,
i.e., the profile-tracing illustrated in Figure A-11  and the strain gage measuring
illustrated in Figure A-12, information such as final wave height, final height-to-width
ratio, maximum strain recorded, and the locations of high stress concentrations were
obtained.  Table A-3 summarizes such information obtained from the first series of
1,000 hour experiments.

As shown in Table A-3, the final wave height decreases with increasing normal stress.
However,  the height-to-width ratio increases with increasing normal stress even more
significantly.  It was seen that the effect on the height-to-width ratio is essentially
doubled in comparison with the effect on  the final wave height.  For example, a normal
stress of 700 kPa resulted in a 37% reduction in the wave height compared to its
original configuration.  However, the same normal stress caused a 76% increase in the
height-to-width ratio. Since high height-to-width ratios generally indicate large
curvatures and locations of high stress concentration, the overall effect of high normal
stress is obviously unfavorable.

Table A-3. Summarized Results of Test Series No.1 - Effect of Normal Stress
Normal
Stress
(kPa)
0
(original)
180
360
700
1,100
Final Wave
Ht.
(mm)
60
(original)
47
42
38
34
Final
H/W
Ratio
0.33
(original)
0.47
0.51
0.58
0.62
Max.
Strain
(%)
+ 1.7
(original)
+ 1.8
+ 2.0
+ 3.0
+ 3.2
Actual Location(s) of Highest
Stress Concentration
(Strain Gage Location)
Crest of wave (G1)
Crest of wave (G1)
Crest of wave (G1)
Crest of wave (G1)
Upper portion of wave
(G1, G2andG3)
Upper portion and base of wave
(G2 and G5)
The strain recorded in each experiment shows that tensile strain increases as normal
stress increases.  This is expected since the H/W values increase significantly with
greater curvature.  Nevertheless, the GM is tensioned significantly less than its yield
point. (Note that the tensile yield strain for this GM is in the range of 15 to 25%
depending on the temperature.) Therefore, tensile yield is not expected. However, the
general design objective is to place the GM with as little stress as possible.  This
concern will be re-examined later where the actual stresses induced will be quantified
using various rheologic models.

The second series of 1,000 hour experiments was designed to evaluate the  effect of the
original wave height on the behavior of HOPE GM waves. Five tests using 1.5 mm-thick
                                      A-17
-------
HOPE GM wave specimens were conducted. The original heights of the waves were
14, 20, 40, 60, and 80 mm, respectively. All specimens were subjected to a constant
normal stress of 700 kPa and maintained at a constant temperature of 23°C over the
entire duration of the experiment. The original and final  (after 1,000 hours) shapes of
the test specimens are shown in Figure A-13. Again, reference marks which identify the
locations and movement of the bonded strain gages are also shown in Figure A-13.

By summarizing the results generated from both parts of the monitoring, Table A-4 was
established.
                                     -G1--G2
                    (a) GM wave with original height of 14 mm
                    (b) GM wave with original height of 20 mm

Figure A-13.  Original and final shapes of HOPE GM waves with various original
wave heights (grid lines have dimensions of 10 mm by 10 mm).
                                     A-18
-------
                    (c) GM wave with original height of 40 mm
                                         G2
                                                 G5
                                                   G6
                                                  \
                    (d) GM wave with original height of 60 mm
                    (e) GM wave with original height of 80 mm

Figure A-13 (cont.).  Original and final shapes of HOPE GM waves with various
original wave heights (grid lines have dimensions of 10 mm by 10 mm).
                                    A-19
-------
Table A-4.  Summarized Results of Test Series No.2 - Effect of Original Wave Height
 Original   Original    Final     Final    Max.      Actual Location(s) of Highest
WaveHt.    H/W    Wave Ht.   H/W    Strain         Stress Concentration
  (mm)     Ratio      (mm)     Ratio     (%)	(Strain Gage Location)
14
20
40

60
0.
0.
0.

0.
.17
.15
.27

.33
8
12
25

38
0.
0.
0.

0.
14
.18
.38

.58
+ 0.
+ 1.
+ 2.

+ 3.
2
2
.4

.0
Negligible
Base of wave
Upper portion
(G2 and G4)
Upper portion

(G3)
and base

of wave


of wave


(G1, G2andG3)
80

0.

.33

47

0.

.65

+ 3.

.4

Upper portion
(G2 and G4)
and base

of wave

As seen in Table A-4, there was an approximate 40% reduction in height after 1,000
hours for all waves.  As to the final H/W ratio, it increases with increasing original wave
height. Note that for waves originally higher than 60 mm, the final H/W ratios exceeded
a value of 0.5.  With regard to the maximum strain recorded, an increasing trend is also
seen with increasing original height.  Moreover, there was no sign of achieving intimate
contact between the specimen and the underlying subgrade after 1,000 hours,  even for
the wave with the smallest original height, i.e., the 14 mm wave.

The third series of 1,000 hour experiments was designed to evaluate the effect of GM
thickness on the behavior of HOPE GM waves. Four tests using HOPE GM wave
specimens, with thicknesses of 1.0, 1.5, 2.0, and 2.5 mm, were conducted.  The original
heights of all wave specimens were approximately 60 mm. Owing to the various
stiffnesses of the GMs having different thicknesses, a constant value of original H/W
ratio could not be maintained, see Table A-5. All specimens were subjected to a
constant  normal stress of 700 kPa and maintained at a constant temperature of 23°C
over the entire duration of the experiments.  The original and final (after 1,000 hours)
shapes of the test specimens, along with reference marks which indicate the location
and movement of the strain gages, are shown in Figure A-14.

As shown in Table A-5, with the only exception being the 1.0-mm-thick GM wave, the
following observations are made.  First, the thickness of GM has very little effect on the
final height of GM waves.  There was an approximate 40% reduction in height after
1,000 hours for all waves. In other words, the original height essentially determined the
final height of GM waves.  Second, the GM thickness did show a significant effect on
the final H/W ratio of the waves.  That is to say, the final H/W ratio decreases with
increasing GM thickness.  The latter observation can be interpreted in an alternative
manner.  That is, for waves with the same original height, thicker GMs resulted in wider
voids  beneath the wave.  Third, the maximum strain recorded in each experiment shows
that tensile strain slightly increases as the thickness of GM increases.
                                      A-20
-------
Table A-5. Summarized Results of Test Series No.3 - Effect of GM Thickness
GM
Thickness
(mm)
1.0
1.5

2.0

2.5

Original
H/W
Ratio
0.24
0.34

0.18

0.21

Final
Wave Ht.
(mm)
27
38

33

38

Final
H/W
Ratio
0.52
0.56

0.34

0.32

Max.
Strain
(%)
+ 2.5
+ 3.0

+ 3.1

+ 3.3

Actual Location(s) of Highest
Stress Concentration
(Strain Gage Location)
Base of wave (G5)
Upper portion and base of
wave (G1, G2 and G3)
Upper portion and base of
wave(G1, G2, G4 and G5)
Upper portion and base of
wave (G2, G3, G4 and G5)
Note: Original heights of all wave specimens were approximately 60 mm
                                                 G5
                     (a) GM wave with thickness of 1.0 mm
                     (b) GM wave with thickness of 1.5 mm

Figure A-14. Original and final shapes of HOPE GM waves with various
thicknesses (grid lines have dimensions of 10 mm by 10 mm).
                                    A-21
-------
                      (c) GM wave with thickness of 2.0 mm
                      (d) GM wave with thickness of 2.5 mm

Figure A-14 (cont.). Original and final shapes of HOPE GM waves with various
thicknesses (grid lines have dimensions of 10 mm by 10 mm).
The fourth series of 1,000 hour experiments were designed to evaluate the effect of
temperature on the behavior of HOPE GM waves. Three sets of experiments, each
consisting of 1.5 mm thick HOPE GM waves with original heights of 14, 20, 40, and 60
mm, were conducted at temperatures of 23, 42 and 55°C.

The original shapes of all wave specimens were formed at 23°C with approximately 100
mm of sand backfill over them.  Temperature was then increased, as necessary, to the
desired value. This was meant to replicate field situations where the exposed GMs
experience an increase in temperature after placement and seaming.  The test boxes
were then filled with sand, followed by a decrease in temperature back to 23°C, to
simulate the decreasing in the sheet temperature of the field deployed  GMs after the
protection and drainage layers are placed.  After approximately 24 hours, a constant
normal stress of 700 kPa was applied. After another hour, temperature was  increased
                                     A-22
-------
from 23°C to the desired value and maintained for the remainder of the experiment.
The last step was intended to simulate a possible increase in the sheet temperature
over the entire lifetime of landfills.

The original and final shapes of the test specimens at the three temperatures, along
with reference marks which indicate the location and movement of the strain gages, are
shown in Figure A-15.

A typical strain measurement result of this series of experiments is shown in Figure
A-16. This particular test was conducted at a temperature of 42°C using a wave
specimen with an original height of 20 mm. As seen in the figure, temperature was
increased from 23 to 42°C one hour after the normal stress was applied. For this
particular experiment, a trend of increasing strain with increasing temperature was
observed in all measurements.  This is due to a combined effect of both thermal
expansion and material softening with increasing temperature.  Although such a trend is
seen in most of the other measurements, a decreasing trend was also observed in
some cases.  This suggests that the change of shapes due to the material softening
with increasing temperature can sometimes cause portions of the GM waves to  undergo
compressive stresses. When such an effect is more significant than the effect of
thermal expansion, a decreasing strain with increasing temperature is seen.

The summarized results generated from this test series of the monitoring is presented in
Table A-6. Note that the values of maximum strain listed in the table are corresponding
to the maximum final (after 1,000 hours) strain.

A-4 Experimental Results -10,000 hour Tests
The strain gage measurement results of the 10,000 hour test are presented graphically
in Figure A-17.  Experimental data up to 1,000 hours was used to establish the first set
of Kelvin-Chain models for predictions out to 10,000 hours.  The calculated curves
using these models are shown in dashed lines. As seen in Figure A-17, they agree with
the actual data measured between 1,000 and 10,000 hours very well.  This encouraging
finding is felt to justify the use of the Kelvin-Chain model for the  purpose of  long-term
prediction, Soong (1996). With  this in mind, the second set of Kelvin-Chain models was
developed using the entire  experimental strain gage measurements and another order
of extrapolation, i.e., predictions out to 100,000 hours, was made.  The resulting curves
are also shown in Figure A-17 as solid lines.

A-5 Analysis of Test Results
The experimental results, including the profile-tracing of the actual waves and the strain
gage monitoring, of the 1,000 hour tests were summarized and briefly discussed in the
previous section. Complete results of all of the strain gage monitoring,  along with the
                                      A-23
-------
                    (a) GM wave with original height of 14 mm
                    (b) GM wave with original height of 20 mm
                    (c) GM wave with original height of 40 mm
                                     G1
                                         G2
                                              G4
                                                 G5
                         23°
                         42'
3
                         >5°C
                    (d) GM wave with original height of 60 mm

Figure A-15. Original and final shapes of HOPE GM waves at various
temperatures (grid lines have dimensions of 10 mm by 10 mm).
                                    A-24
-------
       Strain (%)
               -1


•

+
0
t


•
• •

+ +
o o
23°C

•
» • ii ••
« i - » **
T
o
1 + + 1 *
, o i o°
23 to 42°
	 1 	
1
*f* • •<
t. « »<
*
I
K "
iii 1 1
| + +H

C 4

•%~.
«»«4*«
/rvpococr

-H-k-j... .

2°C

1

* G1
, + G2
o G3
• G4




                  -3
                10
-2
-1
                            10
                            10
10
                        10     10      10      10
                 I700 kPa stress applied     Tjme (hours)

Figure A-16.  Strain measurement results of test conducted on wave specimen
with an original height of 20 mm and at a temperature of 42°C.
Table A-6.  Summarized Results of Test Series No. 4 - Effect of Temperature
Original
Height (mm)/
H/W Ratio

14/0.17


20/0.15



40/0.27



60/0.33



Temp.
(°C)
23
42
55
23
42
55
23

42
55
23

42

55

Final Wave
Height
(mm)
8
10
5
12
14
12
25

25
25
38

30

28

Final
H/W
Ratio
0.14
0.19
0.20
0.18
0.21
0.30
0.38

0.42
0.40
0.58

0.52

0.55

Max.
Strain
(%)
+ 0.2*
+ 0.6
+ 1.3
+ 1.2
+ 1.6
+2.1
+ 2.4

+ 3.2
+ 2.1
+ 3.0

+ 4.9

+ 4.9

Actual Location(s) of Highest
Stress Concentration
(Strain Gage Locations)
Negligible
Negligible
Base of wave (G2)
Base of wave (G3)
Base of wave (G4)
Base of wave (G4)
Upper portion and base of
wave (G2 and G3)
Base of wave (G3)
Crest of wave (G1)
Upper portion and base of
wave (G1, G2, G3 and G5)
Upper portion and base of
wave (G1, G2, G3 and G5)
Upper portion and base of
wave(G1, G2 and G5)
Note:  "+" strain=tension
      "-" strain=compression
                                     A-25
-------
     Strain (%)
             1
                       10
10
10
10
10
10
10
                                     Time (hours)
Figure A-17. Experimental and modeled results of the 10,000 hour control test.

predicted behavior up to 10,000 hours, can be found in Soong (1996).  In this section,
the previous test results will be analyzed further.  Various aspects of the test results,
including final wave height, final height-to-width ratio, and the maximum strain at the
end of 1,000 hour experiments, will be utilized to quantify the effect of different
experimental variables on the behavior of HOPE GM waves.

In this section, the height of HOPE GM wave specimens at the end of the 1,000 hour
experiments as previously described are plotted against the relevant experimental
variables.  These variables include normal stress, original height of wave, thickness of
GM and testing temperature.  The results are shown in Figures A-18 through A-21.
Some observations are made and summarized in Table A-7.

Additionally, the height-to-width (H/W) ratio of the HOPE GM wave specimens at the
end of the experiments are plotted against various experimental variables, as shown in
Figures A-22 through A-25. Some observations are made and summarized in Table
A-8.

Lastly, the maximum tensile strain of the HOPE GM wave specimens at the end of the
experiments (irrespective of their locations) are plotted against various experimental
variables, as shown in  Figures A-26 through A-29. Some observations are made and
summarized in Table A-9.
                                     A-26
-------
             70
             60

       Final
       Wave
       Height 50
       (mm)
             40
             30
                0      200      400      600     800     1000     1200

                                 Normal Stress (kPa)

Figure A-18. Effect of normal stress on the final height of HOPE GM waves.
             50
             40

       Final
      Wave  30
      Height
       (mm)
             20
             10
                0
20        40        60        80

     Original Height of Wave (mm)
100
Figure A-19. Effect of original height of wave on the final height of HOPE GM
waves.
                                    A-27
-------
       Final
      Wave
      Height
       (mm)
             60
             50
40
             30
             20
                                                 o
                0.5
             1.0        1.5        2.0        2.5

               Thickness of Geomembrane (mm)
             3.0
Figure A-20.  Effect of GM thickness on the final height of HOPE GM waves.
              50
              40
       Final
      Wave
      Height
       (mm)
30.
              20
                                                •  Original
                                                O  Original
                                                   original
                                                   Original
                                            leight = 60 mm
                                            leight = 40 mm
                                            leignt = '2(1 mm
                                            leight = 14 mm
                 20
                30            40

                   Testing Temperature (°C)
50
60
Figure A-21.  Effect of testing temperature on the final height of HOPE GM waves
having various original heights.
                                     A-28
-------
Table A-7.  Effects of Different Experimental Variables on the Final Height of
HOPE GM Waves
    Experimental Variable
               Observations
Normal Stress
Original Height of Wave
Thickness of GM
Testing Temperature
Final wave height decreases with increasing
normal stress
% reduction in height = 27 log 
-------
07
06 -
0 5 -
Final
H/W
Poti-i 04-
Katio w-^
03 -
02 -
0 1 -
0.0 -





/<>
/



(
/
>


(
X
/
-S



./ c
) .X





)






               0
              20         40          60
                  Original Height of Wave (mm)
80
100
Figure A-23.  Effect of original height of wave on the final height-to-width ratio of
HOPE GM waves.

           0.7
      Final
      H/W
      Ratio
            0.6
0.5
            0.4
            0.3
            0.2
               0.5
              1.0        1.5        2.0        2.5

                Thickness of Geomembrane (mm)
          3.0
Figure A-24.  Effect of GM thickness on the final height-to-width ratio of HOPE GM
waves.
                                     A-30
-------
      Final
      H/W  o.5
      Ratio
            0.4
                                                •  Original Height = 60 mm
                                                O  Original Height = 40 mm
                                                °  Original Height = 20 mm
                                                   Original Height = 14 mm
            0.0
                20
30            40            50

   Testing Temperature (°C)
60
Figure A-25. Effect of testing temperature on the final height-to-width ratio of
HOPE GM waves having various original heights.
Table A-8. Effects of Different Experimental Variables on the Final Height-to-Width
Ratio of HOPE GM Waves
    Experimental Variable
                  Observations
Normal Stress
Original Height of Wave
Thickness of GM
Testing Temperature
  Final H/W ratio increases with increasing normal
  stress
  % reduction in height = 59 log 
-------
               4.0


               3.5



       Maximum
        Strain
         (%)   2.5


               2.0


               1.5


               1.0
                          200     400     600     800

                                   Normal Stress (kPa)
                                             1000
1200
Figure A-26.  Effect of normal stress on the maximum strain measured at the end
of experiments of HOPE GM waves.
      Maximum
       Strain
        to,
4.0

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0
                      A
                           20        40         60

                               Original Height of Wave (mm)
                                             80
  100
Figure A-27.  Effect of original height of wave on the maximum strain measured at
the end of experiments of HOPE GM waves.
                                    A-32
-------
             4.0
             3.5
      Maximum
       Strain
             3.0
             2.5
             2.0
                0.5
1.0         1.5        2.0        2.5
   Thickness of Geomembrane (mm)
 3.0
Figure A-28.  Effect of GM thickness on the maximum strain measured at the end
of experiments of HOPE GM waves.
                 7

                 6

                 5

       Maximum
        Tensile   4
         Strain
          (%)     3
                      • Original Height = 60 mm
                      O Original Height = 40 mm
                      D Original Height = 20 mm
                      • Original Height = 14 mm
                  20
    30           40          50
       Testing Temperature (°C)
60
Figure A-29.  Effect of testing temperature on the maximum strain measured at
the end of experiments of HOPE GM waves having various original heights.
                                     A-33
-------
Table A-9.  Effects of Different Experimental Variables on the Maximum Strain
Measured at the end of Experiments of HDPE GM Waves	
   Experimental Variable
                   Observations
Normal Stress
Original Height of Wave
Thickness of GM
Testing Temperature
• Maximum strain increases approximately
 linearly with increasing normal stress
• Max. % Strain = 0.0015 an + 1.6
 Where an = normal stress in kPa

• Maximum strain increases logarithmically with
 increasing original wave height
• Max. % Strain =4.1  log (OH) - 4.34
 Where OH = original height in mm

• Maximum strain increases approximately
 linearly with increasing GM thickness
• Max. % Strain =0.5t +2.1
 Where t = thickness of GM in mm

• Maximum strain increases with increasing
 temperature for waves originally shorter than 40 mm
• Maximum strain showed no clear trend with
 increasing temperature for waves originally higher
 than 40 mm
The Maxwell-Weichert model was seen to successfully predict the stress relaxation
behavior of HDPE GMs over the temperature range of -10 to 70°C (Soong et al., 1994;
Soong, 1995, 1996). This covers the range of interest in this study.  Moreover, the
effects of strain rate on the stress/strain relationships and the initial modulus of HDPE
GMs were also successfully described by the same model. As a result, the initial
modulus values of HDPE GMs at various temperatures, which are suitable for the use of
design and stress analysis, were quantified. Values of the initial modulus of HDPE
GMs, which will be used in the stress analysis to follow, have been assembled and
summarized in Table A-10.

Table A-10.  Modulus of HDPE GMs at Various Temperatures to be Used in the
Stress Analysis  to Follow
Temperature
(°C)
23
42
55
Initial Modulus
(MPa)
230
140
90
Using the modulus values presented in Table A-10, the stress induced in the GM can be
calculated with any known strain at a given temperature. However, such stresses will
relax over time.  The stress relaxation  behavior of the tested HDPE GM is dependent
                                     A-34
-------
upon temperature only.  In addition, master curves generated via time-temperature
superposition can be used for the prediction of long-term stress relaxation behavior.

As shown in Figure A-30, a normalized master curve for a 1.5 mm thick HOPE GM is
plotted against time at various temperatures. Via proper curve fitting, these normalized
master curves can be described using numerical expressions.  The resulting
expressions from the above procedure are given in Equations A-1, A-2 and A-3 for
temperatures of 10, 30 and 50°C, respectively.  Note that the "time" terms in these
equations are in the units of hours.

             120
             100

   Normalized
    Relaxation 80
     Modulus
              60
              40
                            V
                   -4    -3    -2   " -1  '0     1
                 10   10   10   10   10   10
                                                            50<
             10
  2   '3   ~ 4   '5     6    7
10   10   10   10   10   10
                                       Time (hours)

Figure A-30.  Normalized master curves of the long-term stress relaxation
behavior of HOPE GM at various temperatures.
Normalized stress relaxation behavior of HOPE GM at 10°C:

      (% Relaxation) = 51.4 + 8.9 log (time) -1.0 (log (time))2 + 0.05 (log (time))3  (A-1)
Normalized stress relaxation behavior of HOPE GM at 30°C:

      (% Relaxation) = 53.0 + 8.4 log (time) -1.2 (log (time))2 + 0.07 (log (time))3  (A-2)
Normalized stress relaxation behavior of HOPE GM at 50°C:

      (% Relaxation) = 48.0 + 5.3 log (time) -1.2 (log (time))2 + 0.19 (log (time))3  (A-3)
                                     A-35
-------
A procedure for analyzing the stress induced in the GM wave is proposed as follows.
Note that a worksheet, as shown in Table A-11, will be utilized to illustrate the
procedure conceptually. Also note that the numerical expression for the stress
relaxation behavior at 30°C, i.e., Equation A-2, will be used to analyze the results of
experiments conducted at 23°C.  As to the experiments conducted at 42 and 55°C, they
will be analyzed using the expression for the behavior at 50°C, i.e., Equation A-3.

Table A-11.  Elements of the Worksheet for the Stress Analysis of the
Experimental Results	
Time  Strain   Stress Induced     Relaxation       Relaxation
         s\         During       Behavior of ai0   Behavior of ai1
                                                                         Residual
                                                                        Stress, ar
  t
         so
                    = E"l"
                        x SQ
                                      Summation of
                                         stress
                                       (horizontally)
  t1      s1      ai2=Ex(e1-e0)    (1-Eqn*(ti-to)§)xai0

  t2      s2      ai2 = Ex(s2-s-|)     (1-Eqn(t2-to))xCTj0   (1-Eqn(t2-ti))xaj-|
 tn-1    sn-1
         sn     aj2 = Ex(en-en-i)    (1-Eqn(tn-to))xajQ   (1-Eqn(tn-ti))xan
 tf-1
 t.
 final
        sf-1
         sf
aif=
(1-Eqn(tfto))xai0    (1-Eqn(tf-ti))xaji
Notes:  t   Appropriate initial modulus value listed in Table A-10
          Equation A-3 for experiments conducted at 23°C
          Equation A-4 for experiments conducted at 42 and 55°C
       §   Replace the "time" terms in equations by the difference between the considered time and the
          corresponding stress induction time.

As seen  in Table A-11, the stress  induced between  any two adjacent instants of time is
determined via multiplying the differences in their corresponding strains by an
appropriate initial modulus value,  i.e., the values listed in Table A-10.  Immediately after
a stress  is induced, the GM will start to relax according to the appropriate modeled
behavior as expressed in Equations A-1, A-2 and A-3, depending on the temperature.
This concept is illustrated in the fourth and subsequent columns of Table A-11.   Finally,
as seen  in the last column of Table A-11, the instantaneous residual stress in the GM is
                                        A-36
-------
calculated by summing the remainder of all the discretized stresses corresponding to
that particular time instant.

Three example calculations which illustrate the above stress analysis procedure are as
follows. They are corresponding to the most critical strain measurements of three
different 1,000 hour experiments and their extrapolations.  Detailed information
regarding these three experiments is summarized in Table A-12.

Table A-12. Example Used to Illustrate the  Use of the Stress Analysis Procedure
Example   Thickness of    Original   Normal   Temperature    Location where
                GM          Wave     Stress                    Strain is the
                            Height                               Maximum
               (mm)	(mm)     (kPa)	(°C)
1 1.5
2 1.0
3 1.5


20
60
60


700
700
700


23
23
55


Near the base
of the wave where
the wave curvature
changes to
accommodate the
horizontal subgrade
Example A-1
As shown in Table A-12, this particular experiment was conducted at 23°C. Hence, an
initial modulus of 230 MPa and a relaxation behavior as expressed in Equation A-2 is
used in this particular stress analysis.  By inserting the strain data, along with the
appropriate constant and expression, into a preestablished spreadsheet, the strain data
is converted to stresses. The results are shown in Figure A-31, where strain and stress
are plotted against time. Note that the incorrect "modulus times strain" curve is also
shown in the figure to demonstrate the amount of stress relaxed over the entire duration
of time.

As seen in Figure A-31, a stress of 3,700 kPa was induced immediately after the full
load was applied to the wave specimen.  Subsequently, through the phenomenon of
stress relaxation along with the decreasing actual strain, the residual stress decreased
approximately 2,000 kPa to 750 kPa after 10,000 hours.

Example A-2
As shown in Table A-12, this particular experiment was also conducted at 23°C. Hence,
an initial modulus of 230 MPa and a relaxation behavior as expressed in Equation A-2
was used in this particular stress analysis.  A similar procedure to that used in Example
A-1 was carried out and the  results are shown in Figure A-32.
                                      A-37
-------
        Strain
                                                             5000
                                                             4000

                                                                 Stress
                                                                 (kPa)
                                                             3000
             0.5
             0.0
                  -3-2-101234
                 10    10   10    10    10   10   10    10
 5V
10
                                 Time (hours)


Figure A-31.  Results of the stress analysis of example 1.
           10    10     10    10    10    10    10
                                Time (hours)


Figure A-32.  Results of the stress analysis of example 2.
                                     A-38
-------
As seen in Figure A-32, a stress as high as 5,500 kPa was induced immediately after
the full load was applied to the wave specimen. Although there was only a slight
decrease in strain over the entire duration of time, a significant amount of stress was
still relaxed via the general stress relaxation phenomenon. As shown in the figure, the
residual stress decreased approximately 4,000 kPa to 1,500 kPa after 10,000 hours.

Example A-3
This experiment was started at 23°C and maintained at that temperature for one hour.
The temperature was then increased from 23°C to 55°C.  It took approximately nine
hours for the entire experimental setup to reach equilibrium at 55°C. Hence, for
analyzing strain data recorded during the initial one hour,  an initial  modulus of 230 MPa
and a relaxation behavior as expressed in Equation A-2 was used.  As for analyzing the
strain data recorded at twelve hours and beyond, an  initial modulus of 90 MPa and a
relaxation behavior as expressed in Equation A-3 was used.  Again, a similar procedure
as that used in the previous two examples was carried out and the results are shown in
Figure A-33.
      Strain
           3.0
           2.0
           1.0
                                                               10000
                                                               8000
                                                               6000
                                                                   Stress
                                                                    (kPa)
                                                              4000
                                                             Hint
                                                             ress
                                                          related
                                                              2000
                -3     -2     -1    0
               10    10    10    10   1!
                                      10    102    10   10    10
                                Time (hours)

Figure A-33.  Results of the stress analysis of example 3.
                                                               0
As seen in Figure A-33, a stress more than 9,000 kPa was induced immediately after
the full load was applied to the wave specimen. During the initial one hour of the test,
the stress relaxed to a residual value of approximately 4,500 kPa (i.e., 50% relaxation in
                                      A-39
-------
one hour). The effect of the subsequent increasing in temperature is clearly shown in
both curves between one and twelve hours. Finally, at a relatively high temperature of
55°C, the residual stress decreased approximately 2,900 kPa to 1,600 kPa after 10,000
hours.

The same procedure as illustrated in these three examples was carried out for all
twenty-five of the 1,000 hour experiments conducted in this study.  Again, only the most
critical strain measurement for each experiment was analyzed.  The complete results all
of analyses can be found in Soong (1996).

The residual stresses after 10,000 hours were also compared to the yield stress at the
particular temperature of the respective test.  The values of yield stress were obtained
via tensile tests conducted at the appropriate corresponding temperatures. The test
specimens were 1.5 mm thick HOPE GMs with a height of 50 mm and a width of 100
mm. The rate of extension used to conduct these tests was 12.7 mm/min (25%/min).
The short-term, but temperature corrected, yield stresses of HOPE GMs were evaluated
and are listed in Table A-13.

Table A-13.  Yield stresses of HOPE GMs at various temperatures to be used in
calculating the percent residual stresses to follow.
Temperature
(°C)
23
42
55
Yield Stress
(kPa)
15000
12000
9400
The entire procedure for obtaining the residual stress as a percentage of the yield stress
is summarized in a flow chart format as shown in Figure A-34.

The results of the stress analysis, in terms of the residual stress after 10,000 hours, are
summarized in Table A-14. Both the actual residual stress values and the percent of
the yield stress are presented. Some observations are made and summarized in Table
A-15.

A-6  Summary and  Conclusions
In this appendix,  the characteristics, fate, and behavior of waves of the type that are
seen in field deployed GMs were evaluated. The entire task was laboratory oriented.
However, full size waves were created,  thus it is believed that scale effects did not
significantly influence the test results. Due to their widespread use, the study focused
on HOPE GMs. The effects of four important experimental variables on the different
aspects of the  behavior of the waves were evaluated.  The variables are normal stress,
original wave height, GM thickness, and temperature.
                                     A-40
-------
                     Stress Analysis Procedure to Quantify
                        Residual Stresses in GM Waves
                           as % of the Yield Stress
   Experiment*
    Tensile Strains
    1,000 hou
 lly Measure
     out to
 • Duration
  Parameters [Evaluated:
  • normal str ;
  • original wa
  • thickness
  • temperaturs
        Extra
  1,000 hr. date
  using Kelvin
 !SS
   height
dfGM
        Perform
Stress Relaxalion
 Function of Time
 )olate
  to 10,000 hr.
 Chain Model
                                                            Tests as a
                                                            & Temp.
                   Use Maxwell-\fVeichert Model
                         To Del ermine
                    Instantaneois Modulus at
                       Creep Strain Rates
                                                                 Use Tim
                                                            Superposition
                                                             Stress Relax*
                           3-Temp.
                           for Generalized
                           ition Modulus
                         Convert all Me*
                                to Stre
                          (Instantaneou:
                        Normalized Residual
                             as a perc
                             oftheYiek
                          sured Strains
                          sses
                           & Relaxed)
                             Stresses
                          ;ntage
                           Stress
                                                           Experimentally Measure
                                                               Yield Stress at
                                                            Various Temperatures
Figure A-34.  Flow chart for the procedure of obtaining residual stresses in terms
of percent yield stress.
The experimental design for this task represented 25 separate tests each conducted for
1,000 hours.  In addition, a single control test was maintained for 10,000 hours (1.1
years).  Each of the tests utilized HOPE GMs with strain gages attached at a number of
critical locations.  This enabled extensional strain to be monitored for the duration of the
experiments.  The results of the strain gage measurements on the 1,000 hour tests
were then modeled and extrapolated one order of magnitude to 10,000  hours using the
Kelvin-chain model. The applicability of using the Kelvin-chain model was established
                                         A-41
-------
on the basis of the experimental results of the 10,000 hour control test. The other
important rheologic model presented in this study is the Maxwell-Weichert model.  It is
an analytic model that was calibrated using the results of large-scale stress relaxation
experiments. The Maxwell-Weichert model was  used to predict the stress relaxation
behavior of the modeled material at a range of temperatures.  In addition, the initial
portion of the stress/strain relationships of the modeled material at slow strain rates was
also predicted. As a result of combining both predictions, the design modulus of HOPE
GMs at various temperatures was determined. By incorporating the generalized stress
relaxation  behavior with such design modulus values, the measured strains were
converted  into tensile stresses.  These stresses were then expresses as a percent of
the tensile yield stress of the GM.

Table A-14.  Residual stress (after 10,000 hours) in the HOPE GM specimens of
experiments conducted in this study.
Experimental Parameter and Variables Residual Stress
(kPa)
Normal Stress 180kPa
360 kPa
700 kPa
1100kPa
Original Height of Wave 14mm
20 mm
40 mm
60 mm
80 mm
Thickness of GM 1.0mm
1.5 mm
2.0 mm
2.5 mm
Testing Temperature 23°C
14mm-42°C
55°C
23°C
20 mm - 42°C
55°C
23°C
40 mm - 42°C
55°C
23°C
60 mm - 42°C
55°C
1200
1300
2000
2100
130
740
1500
2000
2300
1600
2000
1600
1800
130
250
440
740
850
750
1500
1600
690
2000
2600
1600
Residual Stress
(% of Yield)
7.9
8.8
13.2
13.8
0.8
4.9
9.5
13.2
14.9
10.3
13.2
10.6
11.5
0.8
2.1
4.5
4.9
7.3
8.0
9.5
13.7
7.4
13.2
22.0
17.5
                                      A-42
-------
 Table A-15.  Effects of the Variables Evaluated in this Study on Residual Stress
 After 10,000 hours of HDPE GM Wave Experiments	
     Experimental Variable	Observations	
 Normal Stress                  •  Residual stress after 10,000 hours increases
                                 approximately linearly with increasing normal
                                 stress
                               •  Residual stress (% of yield) = 6.8 + 7 an
                                 where an = normal stress in MPa

 Original Height of Wave         •  Residual stress after 10,000 hours increases
                                 logarithmically with increasing original wave
                                 height
                               •  Residual stress (% of yield) = 18 log (OH) - 30
                                 where OH = original wave height in mm

 Thickness of GM               •  Thickness of GMs has no effect on
                                 variation of the residual stress

 Testing Temperature            •  Residual stress increases approximately linearly
                                 with increasing temperature - for waves originally
                                 shorter than 40  mm
                               •  Residual stress shows no clear trend with
                                 increasing temperature - for waves originally
	higher than 40 mm	
 The completed laboratory tests and the associated extrapolated results for 10,000 hours
 were evaluated and a number of observations were developed. These observations are
 subdivided according to the physical manifestation of the wave and its long-term stress
 condition.

 Regarding the original wave heights (which varied from 14 to 80 mm):
       • wave height decreased with increasing normal stress;
       • an average reduction in wave heights of 40% was observed after 1,000 hours;
       • GM thickness had a negligible effect on the decrease in wave height with
        normal stress;
       • there was a  slight decrease in wave height with increasing temperature;
       • final wave heights varied from 5 to 47 mm after 1,000 hours; and
       • intimate contact with the soil subgrade was not achieved after 1,000 hours,
        even for the smallest wave (14 mm) at the highest testing temperature.

 Regarding the original H/W values for the waves (which varied from 0.17 to 0.33):
       • H/W increased with increasing normal stress;
       • H/W increased approximately linearly with increasing original wave height;
       • H/W decreased approximately linearly with increasing GM thickness;
                                      A-43
-------
      • H/W decreased slightly with increasing temperature; and
      • final H/W values recorded from all experiments varied from 0.14 to 0.65 after
        1,000 hours.

Regarding the tensile strains measured at the end of the 1,000 hour experiments along
the top of the GM near the crest of the wave and the bottom of the GM near the
inflection points of the wave at its sides:
      • strains at the maximum point of curvature of the waves increased
        approximately linearly with increasing normal stress;
      • strains at the maximum point of curvature of the waves increased
        logarithmically with  increasing original wave height of the waves;
      • strains at the maximum point of curvature of the waves increased linearly with
        increasing GM thickness;
      • strains at the maximum point of curvature of the waves increased with
        increasing testing temperatures for waves originally shorter than 40 mm;
      • strains at the maximum point of curvature of the waves showed no clear trend
        with increasing testing temperatures for waves originally higher than 40  mm;
      • maximum recorded from all experiments varied from 3.2% to approximately
        4.9% after 1,000 hours.

Regarding the residual tensile stresses after the 1,000 hour experiments which were
then extrapolated to 10,000  hours:
      • residual tensile stress at the points of maximum curvature increased with
        increasing normal stress;
      • residual tensile stress at the points of maximum curvature increased with
        increasing original wave height;
      • thickness of the GM had essentially no effect on the residual tensile stresses;
      • residual tensile stresses increased with increasing testing temperature for
        waves originally shorter than 40 mm;
      • residual tensile stresses showed no clear trend with increasing testing
        temperature for waves originally higher than 40 mm; and
      • residual tensile stresses recorded from all experiments varied from 130  kPa
        (approximately 1 % of the yield stress) to 2,600 kPa (approximately 22% of the
        yield stress).

Based on the test results and the observations given above the following conclusions
are provided:

     • GM waves, which are induced in the field during placement and seaming  of GMs,
      distort upon the application of even a small normal stress. The distortion
      typically increases the height-to-width ratio of the wave.
     • The maximum tensile strain measured in this series of twenty-five 1,000-hour
      tests was approximately 5%.  Note that yield of HOPE GMs is in the range of 15
                                      A-44
-------
      to 25% strain (depending on the temperature), thus yielding of the GM was not
      observed in the tests.
     • The maximum  tensile stresses occur at locations of maximum tensile strain.
      These locations are on the side of the GM that undergoes extension, i.e., along
      the upper surface of the wave near its crest and along the lower surface where
      the wave curvature changes to accommodate the horizontal subgrade beneath
      the wave.
     • Based on an extrapolation to 10,000 hours to account for polymer stress
      relaxation, residual tensile stresses in the GM waves varied from 1% to 22% of
      the GM short-term tensile yield stress.
     • Over the 1,000-hour experimental time of stress application for the main series of
      tests,  the waves did not appear to significantly decrease, much less disappear.
     • It is important to note  that this study did not address the potential effects of the
      waves on liquid flow in lateral drainage layers above the GM, on liquid migration
      through the GM, or on the estimated GM service life.

A-7  Recommendations for the Field Placement of GMs
As illustrated in the Section A-1, the current practice of field placement of GMs in North
America is to install the GM with a certain amount of slack.  The concept is that the
majority of the slack will be removed when the GM is covered and the temperature of
the GM is reduced from its exposed temperature during installation and seaming.  The
goal  is that when the long-term steady-state temperature is reached during the GM's in-
situ service life, the slack will be completely removed as  a result of thermal contraction
and,  therefore, intimate contact by the GM with the subgrade will be achieved.  Many
construction  quality assurance (CQA) documents  in current practice include statements
referring to slack  in the GM.  For example, in EPA (1993), it states "The GM shall have
adequate slack such that it does not lift up off the  subgrade or substrate material at any
location within the facility, i.e., no "trampolining" of the GM shall be allowed to occur at
any time."

As a result of such statements, informal rules have been developed by some for the
deployment of HOPE  GMs.  One such informal rule is that the height of GM wave must
be such that it does not fold  over on itself during backfilling; another informal rule is that
the height-to-width ratio of the installed GM wave  should not be greater than 0.5. The
implicit assumption in allowing such waves is that the subsequent decrease in
temperature, along with the creep and stress relaxation inherent in the GM, will
eventually remove the waves and reduce residual stresses to negligible levels.

However, the experimental and analytic work presented  in this study brings into
question the acceptability of  these informal rules.  It was shown in this study that the
dissipation of waves that typically occur in GM liners under current installation
procedures is only nominal and much of the original wave remains over time.  The
implication is that contact with the subgrade material should not be expected to be
                                     A-45
-------
achieved, even with relatively small waves having an original height of 14 mm. This
was the smallest wave evaluated in this study.

The results of this study show that if waves are to be avoided, the GM must be
essentially flat on the underlying subgrade before backfilling. Waves having small
heights, e.g., less than 14 mm, might be acceptable for wet clay subgrades, providing
the underlying clay is soft enough so the normal stress can "deform" the adjacent wet
clay into the void that is created beneath the wave.  Further study in this regard is
needed. Based on this task, however, the size of such waves is likely to be very small,
e.g., 5 mm or less.

Even after accounting for the stress  relaxation that occurs over 10,000 hours, a
significant amount of tensile stress still remains in GM waves. Such tensile stress could
shorten the service life of a GM in comparison to GMs that are installed flat on the
subgrade.  As already noted, this issue was not evaluated as part of the current study.

One possible GM installation option to mitigate the potential negative consequences of
GM waves is to deploy and seam the GM without slack.  This installation procedure has
found increasing application in Germany. With this procedure, as the liner cools during
the night, it develops tensile stress due to restrained thermal contraction. The following
day, the temperature again rises and the GM is covered with soil at approximately the
same temperature that it was seamed.  In this way, contact with the subgrade is
achieved with only nominal tensile stress in the GM.  Unfortunately, subsequently
induced thermal stresses, if any, will not be dissipated through the phenomenon of
stress relaxation. This was shown by Lord et al. (1995).  Moreover, experiments
showed that going from high installation temperature, e.g., 40°C, to low final service
temperature, e.g., 25°C, can induce tensile stresses as high as 1,000 kPa, see Soong
(1996) for details.

It is suggested that a balance must be achieved so as to achieve contact with the
subgrade while only inducing a nominal amount of tensile stress in the GM. This
nominal amount of tensile stress is subjective at this time, Hsuan et al (1993).  Studies
are ongoing in this regard.  This balance may require some, or all, of the following
changes in the current practice of field deployment and seaming of GMs used in landfill
liner applications.

     1. GMs having light colored (e.g., white) surfaces can be used to advantage in
        decreasing the surface temperature of the GMs while exposed, hence the
        height of the waves will be smaller (Koerner and Koerner, 1995).
     2. GMs should be deployed and seamed without intentional slack.  However,
        installation should be carried out at a temperature as close to the coolest part
        of the day as possible. After the covering GT is placed, if one is required, the
        periphery of the seamed area can be ballasted with cover soil.
                                      A-46
-------
     3.  If a GT covering is not required, placement of an overlying light colored
        temporary GT may be necessary. This can prevent the GM from being
        exposed to direct sunlight before backfilling occurs.
     4.  Backfilling should be performed only in the coolest part of the day.  Quite
        possibly, it might have to be placed  at night.

The above procedures will help considerably  in gaining contact between the GM and the
underlying subgrade.  Since the GMs should  only experience small decreases in
temperature between installation, backfilling,  and in-situ service conditions, the induced
tensile stresses should be able to be accommodated with a properly selected stress-
crack resistant GM.

A-8  References

EPA (1993), Technical Guidance Document,  "Quality Assurance and  Quality Control for
   Waste Containment Facilities", EPA/600/R-93/182, September.
Giroud, J.P. and Morel, N. (1992), "Analysis of Geomembrane Wrinkles", Journal of
   Geotextiles and Geomembranes, Vol. 11,  No. 3, pp. 255-276 (Erratum: 1993, Vol.
   12, No. 4, p378).
Giroud, J.P. (1995), "Wrinkle Management for Polyethylene Geomembranes Requires
   Active Approach", Geotechnical Fabrics Report, Vol. 13, No. 3, pp. 14-17.
Hsuan, Y.G., Koerner, R.M. and Lord, A.E. Jr. (1993), "Notched Constant Tensile Load
   Test (NCTL) for High Density Polyethylene Geomembranes", Geotechnical Testing
   Journal, GTJODJ, Vol. 16, No. 4, December, pp. 450-457.
Koerner, R.M. (1998), Designing with Geosynthetics, 4th ed. New Jersey: Prentice Hall
   Inc.
Koerner G.R. and Koerner R.M. (1995), "Temperature Behavior of Field Deployed
   HOPE Geomembranes" Proceedings Conference on Geosynthetics, Nashville, TN,
   IFAI, pp. 921-937.
Lord, A.E., Jr., Soong T.-Y.  and Koerner, R.M. (1995), "Relaxation Behavior of
   Thermally-Induced Stress in HOPE Geomembranes", Geosynthetics International,
   Vol. 2, No. 3, pp. 626-634.
Schultz, D.W. and Miklas, M.P. Jr., (1980), Proceedings Disposal of Hazardous Waste,
   EPA-600/9-80-010, March, pp. 135-159.
Soong, T.-Y., Lord, A.E., Jr. and Koerner, R.M. (1994), " Stress Relaxation Behavior of
   HOPE Geomembranes", Proceedings 5th International Conference on Geotextiles,
   Geomembranes and Related Products, Singapore, pp. 1121-1124.
Soong T.-Y. (1995), "Effects of Four Experimental Variables on the Stress Relaxation
   Behavior of HOPE Geomembranes" Proceedings Conference on Geosynthetics,
   Nashville, TN, IFAI, pp. 1139-1147.
Soong, T.-Y. (1996), "Behavior of Waves in HOPE Geomembranes,"  Ph.D. Thesis,
   Drexel University, Philadelphia, PA.
                                     A-47
-------
                 Appendix B
  Antioxidant Depletion Time in High
Density Polyethylene Geomembranes
                         by

    Robert M. Koerner, Ph.D., P.E.       Grace Hsuan, Ph.D.
    Drexel University                Geosynthetic Research Institute
    Philadelphia, PA 19104           Philadelphia, PA 19104
                    performed under

              EPA Cooperative Agreement Number
                    CR-821448-01-0
                     Project Officer

                   Mr. David A. Carson
          United States Environmental Protection Agency
              Office of Research and Development
          National Risk Management Research Laboratory
                  Cincinnati, OH 45268
-------
                                 Appendix B
                       Antioxidant Depletion Time in
               High Density Polyethylene Geomembranes
B-1 Introduction
High density polyethylene (HOPE) geomembranes (GMs) have been used extensively
as barrier materials in waste containment applications, e.g., landfills, surface
impoundments, and waste piles.  The required service lifetime of such GMs varies
according to the type of waste, the sensitivity of the local environment, the stipulated
regulations (if any), and other factors. Service timeframes that have been considered
for landfills have typically fallen into the following  ranges:

•  regulatory minimum (post closure)         =  30 years
•  typical nonhazardous waste              =  50 - 200 years
•  hazardous/low level radioactive waste      =  200 -1000 years

Ideally, the service life of a GM should be at least equal to the service life of the landfill
structure. Thus, it is important to be able to quantify the anticipated service lifetime of
GMs used in waste containment applications.

The most direct way to assess service lifetime is to use information obtained from GMs
that have been installed at actual landfills.  However, the first generation of HOPE GM
lined waste facilities is only about 20 to 25 years old.  The available information
suggests that 20-year old HOPE GMs continue to perform in a manner consistent with
their as-installed properties.  An alternative approach  is needed to estimate GM service
life beyond the 20 to 25 year timeframe. In this appendix, the results of a set of
laboratory tests are presented and described.  The results are used to develop
estimates of the service lifetime of HOPE GMs.

The laboratory testing described herein involves aging the GM samples under an
environment that is designed to simulate actual field conditions. The reaction rate that
causes the degradation of the samples under such test conditions is accelerated by
incubating the samples at elevated test temperatures.  This results in an aging of the
samples in a relatively short  period of time, i.e., a few years under accelerated
conditions in comparison to perhaps hundreds of years under actual site conditions.
The degradation data from such elevated temperature testing can then be extrapolated
to predict the lifetime at a site specific ambient temperature by using the Arrhenius
method.
                                      B-1
-------
It should be emphasized that this appendix focuses on GMs that are covered or
backfilled in a "timely manner". Covering with another geosynthetic material or
backfilling with soil is necessary to protect the GMs from ultraviolet (UV) degradation
which is not considered in this task. Furthermore, the surface temperatures of GMs that
are exposed to sunlight are invariably much higher than the applications to which this
study on covered GMs is directed.

Note that this report focuses on lifetime of the antioxidants which are part of a HOPE
GM formulation. Subsequent stages of the total lifetime of the GM are induction time
and the onset of physical/mechanical property degradation. Due to the long term nature
of the incubation processes (up to 10 years), only the first stage of antioxidant depletion
time is reported in this appendix.  An example of the entire sequence of the three stages
of the long-term aging process of GMs was given in Section 2.5 of the main report.

B-2  Formulation, Compounding and Fabrication of HOPE GMs
Before going into a discussion on the long term aging mechanisms of HOPE GMs, the
various steps of producing HOPE GMs will be explained.  The components to  be
formulated, their compounding, and finally the manufacturing process are described in
this section.

The components of an HOPE GM consist of 96 to 97.5%  polyethylene (PE) resin, 2 to 3
% carbon black, and 0.5 to 1.0% antioxidants.  It should be recognized that HOPE GMs
are actually manufactured using PE resin with a density between 0.932 and 0.940 glee.
This resin density is classified as medium density PE according to ASTM D 883.  The
addition of carbon black and antioxidants, however, increases the formulated density of
the product to a range between 0.941  and 0.950 g/cc which is defined as HOPE in
ASTM  D 883. Therefore, the conventional term used in the industry of "HOPE" will be
used.

•   PE  - The resin used for HOPE GMs is a linear copolymer which  is produced by using
   ethylene and a-olefin as comonomer under low pressure and the appropriate type of
   catalyst.  The amount of a-olefin has a direct effect on the density of the resin; a
   greater amount of a-olefin added in the polymerization yields a  lower density PE
   polymer.

•   Carbon black - Carbon black is added into a HOPE GM formulation mainly for UV
   light stabilization. The loading range of carbon black in GMs is typically  2 to 3% by
   weight per ASTM D 1603.  Up to the level of opacity, the higher the loading of
   carbon black, the greater is the degree of UV light stability.  However, the addition of
   carbon black above the opacity level (which is around 3%) will not further improve
   UV resistance (Accorsi and Romero, 1995).
•   Antioxidants - Antioxidants are introduced to an  HOPE GM formulation for the
                                      B-2
-------
   purposes of oxidation prevention during high temperature extrusion and to improve
   the product long-term service life. There are a number of types of antioxidants used
   in GM manufacture and each of them has unique functional characteristics. Usually,
   synergistic mixtures of antioxidants of more than one type are used. Although the
   total amount of antioxidants in the GM is relatively small, less than 1%, their
   presence is vital to achieving the desired product service life. Note that this aspect
   of antioxidant depletion, and the corresponding time to depletion,  is the subject of
   this appendix.

The compounding methods that are used to mix the three components (PE resin,
carbon black and antioxidants) vary from manufacturer to manufacturer.  Three different
methods can be utilized. They are as follows:

•  GM Manufacturers Perform Their Own Mixing:
   GM manufacturers can purchase pure PE resin that contains no carbon black nor
   antioxidants from resin producers.  They then purchase carbon black powder and
   antioxidants from their respective suppliers. The appropriate amounts of these three
   ingredients are mixed in an extruder, forming pellets that consist of the proper
   proportion of each component.  These stabilized pellets are then transferred to
   another extruder for GM production.

•  Let-down From Concentrated Carbon Black Pellets:
   GM manufacturers can purchase PE resin that contains antioxidants only.
   Separately, they then purchase concentrated carbon black pellets consisting of
   approximately 25% carbon black in a PE resin carrier which is the same generic type
   as the parent PE resin.  During the production of the GMs, the exact proportion of
   PE  resin/antioxidant pellets and concentrated carbon black pellets are added to the
   extruder, resulting a product with the proper proportion of each component.

•  Completely Formulated Pellets:
   GM manufacturers can purchase pellets that consist of the proper proportion of PE
   resin, carbon black and antioxidants.  The completely formulated pellets go directly
   to the extruder for GM production.

Upon using a large extruder to mix, melt and filter the resin pellets into a flowing viscous
mass, there are two major processes used for manufacturing HOPE GMs. Their
differences are at the exit section of the  extruder which is some type  of die. One
process is flat sheet extrusion wherein a flat die (or "coathanger die") is utilized.  The
other process is blown film wherein a circular die is used.  Struve (1994) explains the
details of the two processes.
•  Flat die extrusion:
   Flat dies used  in the GM extrusion process are configured with adjustable lips from
   where the polymer sheet exits.  By adjusting the die lips (manually or automatically),
                                       B-3
-------
   the thickness of the GM can be accurately controlled.  Figure B-1 (a) is a schematic
   diagram of center fed flat die. The molten polymer from the extruder enters centrally
   into the die and spreads horizontally in both directions. On exiting, the somewhat
   cooled polymer sheet is deposited onto a series of chilled rolls.  For the production
   of a very wide sheet, two side by side coathanger dies can be joined together, see
   Figure B-1 (b).  The molten polymer in each side of the die is supplied from separate
   extruders. The two melt streams commingle together within the die. Again, the
   somewhat cooled polymer sheet is deposited onto a series of chilled rolls. After
   further cooling, the GM sheet is rolled onto a core for shipment and placement.
Figure B-1 (a). Flat die extrusion process to manufacture GM (Struve, 1994).
Figure B-1(b).
1994).
Dual flat die extruders used to manufacture wide GMs (Struve,
   Blow film extrusion:
   Circular dies are also utilized in the extrusion process of manufacturing PE GMs.
   They are oriented such that the polymer exits the die vertically. The molten polymer
   supplied from the extruder enters into an annular chamber through a number of
   symmetrically radial feed ports.  As the somewhat cooled polymer exits the die, a
   large cylinder of GM is formed, as can be seen in Figure B-2.  The cylinder is closed
   at the top where it passes between a set of nip rollers which draws the GM up and
   away from the die.  The dimensional stability of the cylinder is provided by internal
   and external air pressure. After the material passes through the nip rollers, the
   collapsed cylinder is cut longitudinally, opened to form a full width of GM sheet and
   rolled onto a core for shipment and placement.
                                      B-4
-------
          Nip rollers
          Extruder
                                 Cut here and
                                 unfolded
                                    Film
                                    bubble
Circular
Die
Figure B-2. Blow film extrusion process used to manufacture PE GMs.

B-3 Stages of Degradation in HOPE GMs
After proper placement of the GM sheets and seaming into a liner system, the GM will
hopefully serve as a barrier for many years. During the service period, aging takes
place in the GM.  The aging process of HOPE GMs can be considered to be a
combination of: physical aging and  chemical aging.  Both aging mechanisms take place
simultaneously. Physical aging implies a slow processes in which the material attempts
to establish equilibrium from its as-manufactured nonequilibrium state.  For semi-
crystalline polymers like HOPE, the process involves changes in the crystallinity of the
material (Petermann et al.,  1976).  Under this definition of physical aging there are no
primary (covalent) bonds broken.

On the other hand, chemical aging  indicates some type of degradation involving the
breaking of covalent bonds, e.g., thermal-oxidation, radioactive-degradation, etc.,
(Struik, 1978).  This process eventually leads to a reduction in engineering properties.
Therefore, from an applications point of view, chemical aging is the important
degradation mechanism and should be studied in great detail.  In the following sections
the different stages of chemical aging in HOPE GMs are described.

Conceptually, the chemical aging process of a HOPE GM can be considered to consist
of three distinct stages. They can be seen in Figure B-3.  These three  stages are
designated as  (a) depletion time of antioxidants, (b) induction time to the onset of
polymer degradation and (c) degradation of the polymer to decrease some engineering
property(s) to an arbitrary level, e.g., to 50% of its original value.
                                      B-5
-------
 T3
 CD
 C
 '.2
 CD
    100
     50
 CD
 Q.
 O
A 1 B 1 C

•* ^i^ ^i^ ^
1 1 \\
1 1 \
1 1 -^
1 1
i i
1 1
1 1 1



1
\
\


A - depletion time of
antioxidants
B = induction time to onset
of polymer degradation
C = time to reach 50%
degradation of a particular
property

                     Aging Time (log scale)

Figure B-3. The three conceptual stages in chemical aging of HOPE GMs.

B-3.1 Depletion of Antioxidants
The purpose of antioxidants in a HOPE GM formulation is to prevent degradation during
processing and to prevent oxidation reactions taking place during the first stage of
service life. However, there is only a limited amount of antioxidants in the formulation.
Hence, the lifetime for this stage is also limited. Once the antioxidants are completely
depleted, oxygen will begin to attack the polymer, leading to the induction time and
subsequently the deterioration of performance properties.  The duration of this
antioxidant depletion  stage depends strongly on the type of selected antioxidants.
Since many different  antioxidants can be selected, depletion time can vary from
formulation to formulation, subsequently affecting the lifetime of the GM.  Proper
selection, however, will be seen to contribute greatly to the overall lifetime of the GM.

The depletion of antioxidants may be consequence of two processes: chemical
reactions of the antioxidants, and physical loss of the antioxidants from the polymers.  In
addition, the rate of depletion is related to the type of antioxidants, to the service
temperature, and to the nature of the site specific environment. Regarding the chemical
reactions of antioxidants,  two main functions are involved: the scavenging of free
radicals, converting them  into stable molecules, and the reaction with unstable
hydroperoxide (ROOH) forming a more stable  substance.  Regarding their physical loss,
the process involves  the distribution of antioxidants in the GM and their volatility and
extractability.  Since antioxidants are the  main subject of this appendix, a detailed
investigation of these two processes will be presented in Section B-5.

B-3.2 Induction Time
In a pure PE resin, i.e., one with no carbon black and antioxidants, oxidation occurs
extremely slow at the beginning; often immeasurably slow. However, at the end of this
period acceleration occurs more rapidly.  Eventually, the reaction decelerates and once
                                       B-6
-------
again becomes very slow. This progression is illustrated by the curve in Figure B-4(a).
The initial portion of the curve (before measurable degradation takes place) is called the
induction period (or induction time) of the polymer.

In a stabilized polymer such as one with antioxidants, the acceleration stage takes a
considerably longer time to reach. The antioxidants create an additional depletion time
stage prior to the onset of the induction time, as shown in Figure B-4(b).
                                  Induction|  Acceleration   Deceleration
                                   period       period         period
                              c
                           c o
                           CD ~
                           o> 9-
                           >* o
                              X\J
                           s^ W
                           O .a
 Antioxidant  Induction  Acceleration  I  Deceleration
depletion time1   period'     period
                  c
               c  o
               CD  ~
                  w
               O .a
                                                               period
                                                               (b)
                                       Aging Time

Figure B-4. Curves illustrating the various stage of oxidation:  (a) unstabilized PE,
(b) stabilized PE.

Regarding the chemical process, the first step of oxidation in an unstabilized PE is the
formation of free radicals.  The free radicals subsequently react with oxygen and start
chain reactions.  The reactions are described in Eqs. B-1 to B-6 (Grassie and Scott,
1985).
Initiation stage:

     RH -» R» + H» (under energy or catalyst residues)

     R« + 02 -> ROO
                                                        (B-1)

                                                        (B-2)
                                       B-7
-------
Propagation stage:

     ROO + RH -» ROOM + R-                                              (B-3)

Acceleration stage:

     ROOM -» RO' + OH' (under energy)                                      (B-4)

     RO + RH -» ROM + R'                                                  (B-5)

     OH- + RH -» H20 + R-                                                  (B-6)

(where: RH represents the PE polymer chains, the symbol "•" represents free radicals
which are highly reactive.)

In the induction period, little hydroperoxide (ROOH) is present and when formed it does
not decompose.  Thus, the acceleration stage of the oxidation cannot be achieved. As
oxidation propagates slowly, additional ROOH molecules are formed. Once the
concentration of ROOH reaches a critical level, decomposition of ROOH begins and
accelerated chain reactions begin, signifying the end of the induction period (Rapoport
and Zaikov, 1986). This indicates that the concentration of ROOH has a major effect on
the duration of the induction period.

Viebke et al. (1994) have studied the induction time of an unstabilized medium-density
PE pipe. The pipes were internally pressure tested with stagnant water and externally
by circulating air at temperatures ranging from 70 to 105 °C.  They found the activation
energy of oxidation in the induction period to be 75 KJ/mol. Using their experimental
values, an induction time of 12 years can be extrapolated at a temperature of 25°C for
the material evaluated.

B-3.3 Material Property Degradation
The end of the induction period signifies  the onset of relatively rapid oxidation. This is
because the free radicals increase significantly via the decomposition of ROOH, as
indicated in Eqs.  B-4 to B-6.  One of the  free radicals is an alkyl radical (R») which
represents polymer chains that contain a free radical.  In the early stage of acceleration,
cross-linking occurs  in these alkyl radicals due to oxygen deficiency.  The reactions
involved are expressed by Eqs. B-7 and  B-8. The physical and mechanical properties
of the material subsequently respond to such molecular changes. The most noticeable
change is in the melt index, since it relates to the molecular weight of the polymer.  In
this stage, a lower melt index value is detected.  In contrast, the mechanical properties
do not seem to be very sensitive to cross-linking.  The tensile properties generally
remain unchanged or are unable to be detected.
                                      B-8
-------
                                — CH ~ - CR /| - CH _ —
CH9-CR1-CH9—
                                            _CH9-CRrCH9—
As oxidation proceeds further, and abundance of oxygen becomes available, the
reactions of alkyl radicals change to chain scission.  This causes a reduction in
molecular weight, as shown in Eqs. B-9 and B-10.  In this stage, the physical and
mechanical properties of the material change according to the extent of the chain
scission.  The melt index value reverses from the previous low value to a value higher
than the original starting value signifying a decrease in molecular weight. As for tensile
properties, break stress and break strain decrease.  Tensile modulus and yield stress
increase and yield strain decrease, although to a lesser extent. Eventually, the GM
becomes so brittle that all tensile properties change significantly and the engineering
performance is jeopardized. This signifies the end of the so-called "service life" of the
GM.

                       02&RH
— CH2-CRrCH2 — - ^ — CH2-CR1-CH2 —

                                              OOH
                                -CH2-CRi-CH2-  +    OH

                                         °*                               (B-9)

— CH2-CR, -Chi,  —	^   — CH2-CR,-0  +  -CH2 —

         °*                                                              (B-10)

Although quite arbitrary, the limit of service life of a GM is often selected as a 50%
reduction in a specific design  property.  This is commonly referred to as the half-lifetime,
or simply the halflife. The specific property could be tensile modulus, break stress,
                                      B-9
-------
break strain, impact strength, etc.  It should be noted that even at halflife the GM still
exists and can function albeit at a decreased performance level.

Hence, the lifetime of a GM will be equal to the depletion time of antioxidants, plus
induction time of the polymer, plus the time to reach a 50% reduction in a specific
engineering property. Graphically this was shown in Figure B-3 as the sum of "A", "B"
and "C".

B-4  Major Influences on Oxidation Behavior
There are many aspects of the polymer resin, its formulation, the ambient environment
and its service conditions that can effect the oxidation behavior of HOPE GMs.  This
section describes several of them placed in two categories: internal material effects and
external environmental/service effects.

B-4.7 Internal Material Effects
The chemical and physical structure of the polymer has a strong influence on the rate of
oxidation. This structure controls the formation of free radicals and the diffusion of
oxygen into the polymer.  Three major factors will be discussed:  branch density,
crystallinity and transition metals.

The medium density PE used to manufacture HOPE GMs is a copolymer. Apart from
the dominant ethylene monomer, a comonomer is added to the polymerization. The
comonomer is some type of a-olefin such as butene, hexene, methyl pentene, or octene
(Chu and Hsieh, 1992).  The comonomer forms short chain branches along the
backbone of the PE chain. Two examples are given in Figure B-5. The concentration
of the short chains varies from 5 to 8 per 1000 carbon atoms. The particular carbon
atom where the  branch attaches is surrounded by three other carbon atoms and is
defined as the tertiary carbon. The hydrogen atom attached to the tertiary carbon
possess a lower dissociation energy than other hydrogen atoms, thus free radicals are
most likely to occur at these locations. This is illustrated by Eq. B-11.  In other words,
PE with greater branch density concentration will generate more free radicals than
those with less branches under the same conditions.
                                      B-10
-------
          — CH2-CH2-CH - CH2 - CH2—         butene as the comonomer

                        CH2

                        CH3

          — CH 2- CH2 - CH - CH2 - CH2—        hexene as the comonomer

                        CH2
                        CH2

                        CH2
                        CH3
Figure B-5.  PE with butene and hexene as the comonomer.
   ,H   ,H    H   H   H              H   H   .    H   H
 -C-C-C-C-C	^-C-C-C-C-C   +   H.           (B-11)
   HH|HH              H   H   I     H   H
            CH2                            CH2
            CH3                            CH3

It is established (Michaels and Bixler, 1961) that the crystalline regions in PE are
sufficiently dense to severely limit oxygen penetration. The result of this impermeability
is that the diffusion of oxygen in the polymer is essentially controlled by the amorphous
region. Hence, the diffusion coefficient increases as crystallinity decreases.  During the
initial stage of the oxidation, alkyl radicals are probably produced in both the crystalline
and amorphous regions.  As oxygen gradually diffuses into the amorphous region, it
converts the radicals to alkylperoxy radicals, i.e., ROO, starting the chain oxidation
reactions.  On the other hand, those alkyl radicals that are trapped in the crystalline
matrix are unable to progress further (Billingham and Calvert, 1986).  In addition,
crystallinity  relates closely to the branch density of the polymer. This is because chain
branches interrupt the folding of the polymer chains, reducing the total amount of
crystallinity  in the polymer.  Therefore, as branch density increases, the crystallinity
decreases and the rate of oxidation increases.

The oxidation reaction of PE can be  increased in the presence of transition metals, e.g.,
Co, Mn, Cu, Al and Fe (Osawa and Ishizuka, 1973). The source of these elements
usually comes from residual catalyst used to polymerize the resin.  Although the
concentration of these elements is very low, they still can be a concern regarding the
long term durability of the polymer.  The transition metals break down hydroperoxides
                                      B-11
-------
via "redox" reactions, creating an additional amount of free radicals, as demonstrated in
Eqs. B-12 and B-13.

ROOM + Mn+ -> RO- + M(n+1) + OH-                                         (B-12)

ROOH + M(n+1) -> ROO' + Mn+  + H+                                         (B-13)
B-4.2 External Environmental Effects
The oxidation reaction in PE is rather sensitive to the surrounding ambient environment.
Any conditions that provide oxygen and accelerate the formation of free radicals,
particularly the decomposition of hydroperoxide, increase the rate of oxidation. Three
considerations are described: energy level, oxygen concentration and adjacent
materials.

Sunlight, heat and radiation are three types of energy which should be considered. For
an exposed GM, sunlight is the major concern. Coupled with heat there is a great
potential for free radical formation. Covering in a timely manner, however, avoids
photodegradation and greatly diminishes the heat from direct sunlight exposure.  As
mentioned previously, this study does not address sunlight exposed  GMs. Heat,
however, can come from other sources than direct sunlight. All  other things  being
equal, a GM will degrade faster at higher temperature as opposed to lower temperature.
In predicting lifetime,  it is essential to accurately estimate the service temperature of the
buried GM. For buried wastes that are radioactive there is a potential for the GM to be
exposed to high energy levels depending on the type of waste.  It is expected that low
level radioactive (LLR) and low level radioactive mixed (LLRM) wastes are orders of
magnitude too low to  produce energy levels that could cause degradation. Conversely,
high  level radioactive (HLR) and transuranic (TRU) wastes must be assessed
accordingly.  They are not within the scope of this study

The concentration of available oxygen is an obvious essential component to  any
oxidation reaction.  For exposed GMs, the availability of oxygen  is high and the oxygen
concentration is at its maximum.  Contrary, for the liner beneath a landfill, the available
oxygen will be extremely limited.  In the case of a liner for municipal solid waste landfill,
biodegradation of the waste will probably consume most of the available oxygen.
(Poland and Harper, 1986) showed that the biodegradation of solid waste changes from
aerobic and anaerobic after approximately 3 to 5 years).  Under this  situation, if
degradation occurs in PE, it may lead to crosslinking rather than chain scission, as
shown  in Eqs.  B-7 and B-8.  In surface impoundment applications, the portion of GM
that is covered by liquid is only exposed to approximately one eighth of the oxygen in
comparison to that exposed in air. Unfortunately, in unsaturated soil, the percentage of
oxygen present is very difficult to be defined since it is affected by the type of soil and
the moisture content.  Table B-1 lists an approximate ranking of GM  exposure to oxygen
                                      B-12
-------
on the bases of various applications. This leads directly to different experimental
incubation possibilities, e.g., immersed in liquid, liquid on top/air on bottom, and
completely in air.

Table B-1. Oxygen Availability to GMs in Several Common Applications.
Application
surface
impoundment
Liners

landfill liners

final covers

Location
top of slope
top of slope
base of slope
base of slope
beneath waste

above waste

GM
Surface
top
bottom
top
bottom
top
bottom
top
bottom
Oxygen
Availability
high
moderate
low
moderate
very low
low
high to moderate
very low
The type of material (soil or liquid) that makes direct contact with the GM has an
influence on the oxidation rate. If the adjacent soil contains a large amount of transition
metals (the amount is very subjective) and there is moisture or liquid present, the
transition metals can diffuse into the GM. This action catalyzes the oxidation by
accelerating the decomposition of ROOM, as explained in Eqs. B-12 and B-13.
Furthermore, if the liquid that is contained by the GM consists of a relatively large
amount of organic solvent, the amorphous phase of the GM can swell, increasing the
oxygen diffusion coefficient and accelerating the oxidation.

B-4.3  Commentary on Various Influences
This section described some of the  major influences (internal and external) on the rate
of oxidation of HOPE GMs.  In devising this particular long term experimental program
of approximately 10 years duration,  only one type of HOPE GM was selected. Thus the
internal effects from  the material itself such as  branch density, crystallinity and transition
metals are fixed to that particular GM type.
Additionally, in developing the incubation procedures, various external effects had to be
considered. Three separate incubation environments were evaluated. They are water
immersed, landfill simulated (water above and  air below) and air immersed.  It is
suggested that these different types of incubation procedures cover the range of typical
end uses illustrated in Table B-1.

B-5 Overview of Antioxidants
Since the subject of this appendix is the depletion of antioxidants in HOPE GMs, it is
essential to explain the performance of antioxidants during their depletion period.  Three
properties will be discussed in this section; the function of antioxidant, the types of
                                      B-13
-------
antioxidants along with their individual characteristics, and the antioxidant depletion
mechanisms.
B-5.1 Function of Antioxidants
The sequence of oxidation reactions in HOPE GMs indicated by Eqs. B-1 to B-6 can
also be interconnected by cycles "A" and "B", as illustrated in Figure B-6. There are four
important links in these two cycles, designated as (a) to (d).  If any of the links are
broken, the rate of oxidation of the polymer will be retarded.  If all four links are broken,
then oxidation will be stopped. The purpose of antioxidants in the polymer is to break
such links.
                               0.

                    RH
      ROH&H20
                                                        Numbers 1 to 6
                                                        represent the
                                                        Equations 1 to 6
                                                        in the text.
            (c)  X (4)
                          ROOM
Figure B-6. Oxidation cycles in PE (Grassie and Scott, 1985).

Since the involved molecular species in each of the four cycle links are not the same,
different types of antioxidants are designed to accommodate various requirements.
Antioxidants can be divided into two categories; primary and secondary.

Primary Antioxidants - They provide stabilization by trapping or deactivating free radical
species after they are formed, i.e.,  breaking links (a), (b) and (d). The antioxidants
which intercept the links (b) and (d) function in that they donate an electron. The
electrons react with free radicals ROO, RO and »OH converting them to ROOM, ROM
                                      B-14
-------
 and H20, respectively. The types of antioxidants that break the link (a) are electron
 acceptors. They convert the alkyl free radical (R») to form a stable polymer chain.

 Secondary Antioxidants - They are designed to intercept the link (c) in the "B" cycle.
 Their function is to decompose hydroperoxides (ROOM), preventing them from
 becoming free radicals.  The chemical reactions change the ROOM to a stable alcohol
 (ROM).

 B-5.2 Types and Characteristics of Antioxidants
 Apart from the two categories just described, antioxidants can be further classified into
 four large chemical types within which many different types are included. Table B-2 lists
 the chemical type and some of the commercial available antioxidants that can be used
 in PE GMs. To ensure long term durability, a manufacturer will use two or more types
 of antioxidants; typically one from each category.

 Table B-2. Types of Antioxidants (after Fay and King, 1994)	
      Category          Chemical Type            Examples of Commercially
	Available Antioxidants	
 Primary            Hindered Phenol         Irganox® 1076, lrganox®1010,
                                            Santowhite Crystals
                    Hindered Amines         Tinuvin® 622, Chimassorb® 922
                    (HALS*)

 Secondary          Phophites                lrgafos®168
                    Sulfur Compounds        distearyl thiodipropionate (DSTDP)
                    (Thiosynergists)          dilauryl thiodipropionate (DLTDP),
                    Hindered Amines         Irganox® 1076, Irganox® 1010,
	(HALS*)	Santowhite Crystals	
 * HALS = hindered amine light stabilizers

 There is another issue that needs to  be considered during the selection of antioxidants.
 That is the effective temperature range for each of the selected antioxidants. The
 antioxidant formulation or "package"  should protect the product at both the high
 temperature of the extrusion process and the significant lower temperature during its
 lifetime.  Thus the functioning temperature range for each type of antioxidant should be
 recognized. For the four chemical types listed above, the  effective temperature ranges
 are given in Figure B-7.  The graph shows that phosphites have an effective
 temperature range above 150°C.  They are considered to be process stabilizers.  Either
 thiosynergists or hindered amines will be added to the formulation to accommodate the
 low temperature service protection. On the other hand, for a formulation consisting of
 hindered phenols, a wide range of temperatures are covered; from ambient to process
 temperatures. However, hindered phenols are only primary stabilizers.  A secondary
 antioxidant is also required which can be either thiosynergists or hindered amines.
                                      B-15
-------
           Phosphites

     Hindered Phenols

       Thiosynergists

     Hindered Amines
                      0       50     100     150     200
                                         Temperature °C
250
300
Figure B-7. Effective temperature ranges of the four antioxidant types
(Fay and King, 1994).

In the mixing of the various types of antioxidants, one must beware of the possible
antagonistic effects between them. For products that require long term thermal stability
and light stability, a combination of phenolic and thiosynergist for thermal stability, and
hindered amine for light stability could be used. Unfortunately, the oxidation product of
the sulfur compound can be acidic which reacts with hindered amine, preventing its
interacting with free radicals (Kikkawa et al., 1987).  In HOPE GMs, the carbon black
can also influence the stability of the material, in particular, the thermal stability.
Materials containing carbon black absorb more heat than those without carbon black.
While this discussion is seemingly complicated, it should be recognized that the polymer
industry is quite advanced in the selection of antioxidants. There are many custom
designed packages for each product, including GMs, in order to accommodate a wide
range of processing and service requirements.

One last item to conclude this subsection is the issue of antioxidant cost.  Antioxidants
are comparatively much more expensive than the polymer resin. Thus cost of the final
product is weighted heavily by the amount and type of antioxidants used in the
formulation. A careful balance must be drawn between the required performance and
economy of the final product.

fi-5.3 Antioxidant Depletion Mechanisms
The amount of antioxidant in a HOPE GM decreases gradually as aging progresses.
The depletion can be caused by two mechanisms; chemical reactions of the
antioxidants and physical loss of antioxidants from the polymers by leaching. These
mechanisms can occur simultaneously.
                                     B-16
-------
Chemical reactions: As discussed previously, the antioxidants are consumed by free
radicals and alkylperoxides present in the material. The rate of consumption which
progresses from the surfaces of the GM inward depends on the concentration of these
two species.  Since phenolic and phosphite types of stabilizers are utilized in the
processing stage, the antioxidants that remain in GMs for longevity protection are
probably a combination of residue phenolic types along with thiosynergists or hindered
amines. Of these three types of antioxidants, hindered amines have a unique reactive
behavior. They can be cyclical and regenerative, both leading to a long functioning
time.  Only undesirable side reactions can terminate their efficiency (Fay and King,
1994).

Physical loss:  The two major concerns with respect to the physical stability of
antioxidants in the polymer are their volatility and extractability (Luston,  1986).
Research has indicated that the distribution of antioxidants in semicrystalline polymers
is not uniform, owing to the presence of crystalline and amorphous phases. It appears
that a greater concentration of antioxidants is found in the amorphous region which is
fortunate because the amorphous region is also the most sensitive to degradation.
Hence the mobility of antioxidants in the amorphous phase controls these two physical
processes.

The volatility of antioxidants is a thermally activated process and temperature changes
effect not only the evaporation of the stabilizers from the surface of the polymer but also
their diffusion from the interior to the surface layer. For HOPE GMs, the typical
operating temperature is well below 60°C. Hence volatility is probably not a major
concern. Because of this, one must avoid inducing such a mechanism in accelerated
laboratory aging tests. Very high testing  temperatures should not be utilized.  However,
elevated temperature is necessarily to  accelerate the laboratory aging study. Therefore
a careful balance is required  in the design of the experimental incubation setup.  As
noted previously this task is proposed to  be a 10-year effort.  The reason for such long
time is that the selected test temperatures are relatively low so as to minimize volatility
of antioxidants from occurring.

Extractability of antioxidants plays a part  wherever the GMs comes into contact with
liquids such as water or leachate.  The rate of extraction is controlled by the dissolution
of antioxidants from the surface and the diffusion from the interior structure to the
surface. However, dissolution is faster  than evaporation. Smith et al. (1992) performed
an aging study on a medium  density PE pipe material that was exposed to water
internally and air externally. They monitored the antioxidant depletion across the
thickness of the pipe via oxidative induction time (OIT). They found that the
consumption of antioxidants was three times faster in water than in air at temperatures
of 105, 95and80°C.
                                      B-17
-------
Thus in the physical loss of antioxidants, extraction takes a central role in lifetime
predictions.  Clearly, this is a concern if the GM contacts liquid during its service life.
Unfortunately, there is no data available regarding the effect of humidity on antioxidant
loss.

B-6 Experimental Design
As indicated in Table B-1, HOPE GMs can experience different levels of oxygen
concentration depending on the application, its site specific location, and the materials
in contact with the upper and lower surfaces.  It is important that the laboratory aging
tests simulate the site conditions as close as possible.  In this regard, four different
laboratory incubation protocols have been developed. They are described in Table B-3.
A detailed description of each incubation method is presented in this section.

Table B-3. Incubation Method of HOPE GMs in this Task
Incubation Incubation Applied Simulated GM Application
Series Method Stress
I
water none
(both sides)
surface impoundments below
liquid level
IV
air
(both sides)

water above/air
beneath

water
(both sides)
                               none
260 kPa
(compression)

30% yield stress
(tension)	
landfill covers and waste pile
covers

landfills liners beneath waste
surface impoundments along
side slopes below liquid level
The incubation method of Series I is designed to simulate GMs which are exposed to
liquids (water or leachate) on both sides and are essentially nonstressed, e.g., shallow
surface impoundments. HOPE GM samples are fully immersed in four water baths
maintained at constant temperatures of 55, 65, 75 and 85°C.  The dimensions of the
incubated samples are 150 mm by 150 mm. Samples are retrieved at various time
intervals and evaluated by a number of tests for their physical, chemical and mechanical
properties.

The incubation method of Series II is designed to simulate GMs which are exposed to
air on both of their surfaces, e.g., landfill covers and waste pile covers. GMs in these
applications will be exposed to air from above  and perhaps from beneath as well.  The
exact oxygen concentration that the GM will be subjected to is very difficult to define.
The incubation represents the extreme condition of full oxygen exposure. Hence the
lifetime predicted from this experiment will be a conservative value. In this series, GM
samples are exposed to a continuous flow of air.  HOPE GM samples are suspended in
                                      B-18
-------
five forced air ovens maintained at temperatures of 55, 65, 75, 95°C and 115°C.  The
dimensions of the samples are 100 mm by 150 mm.  Samples are retrieved at various
time intervals and are evaluated by a number of tests for their physical, chemical and
mechanical properties.

The incubation method of Series III is intended to simulate GMs situated beneath solid
waste landfills.  Circular shaped samples of 200 mm diameter are placed  in incubation
columns similar to those suggested by  Mitchell and Spanner (1985), as shown in Figure
B-8. Twenty (20) identical units of this type are used  in this incubation series. A static
compressive stress of 260 kPa is applied to each sample. This is approximately
equivalent to a 30 m high solid waste landfill. Above each sample is a layer of 100 mm
thick sand with 300 mm of water head.   Beneath each sample is dry soil with a limited
amount of air.  Four test temperatures of 55, 65, 75, and 85°C are being utilized.
Samples are retrieved at various time intervals and evaluated by a number of tests for
their physical, chemical and mechanical properties.
                  Insulation
                   heat tape
               thermocouple
                                 MINIM
                                  Sand
                                                   10

                                              Piezometer
                  ]
                                   Sand
B-
                                                         Load
. perforated steel
 loading plate

geomembrane
sample under
compression
Figure B-8.  A schematic diagram of the compression column for incubation
series III.

The incubation method of Series IV is intended to simulate GMs located on side slopes
of surface impoundments where tensile stress may be generated.  GM samples of
dimensions 38 mm by 150 mm are subjected to a constant tensile stress equal to 30%
of their room temperature yield stress.  The tensioned samples are completely
immersed in four water baths maintained at constant temperatures of 55, 65, 75 and
                                     B-19
-------
 85°C.  Samples are retrieved at various time intervals and evaluated by a number of
 tests for their physical, chemical and mechanical properties.

 The four incubation conditions used in this study are attempts at replicating the most
 commonly encountered site situations where GMs are being used.  How the resulting
 predicted lifetimes compare to one another is obviously a major focus of this task.  From
 a comparison of the results it is hoped that one can deduce effects on oxidation on GMs
 that are not directly studied, e.g., different types of stresses, tension effects, water
 extraction effects, etc.

 B-7 Evaluation Tests on Incubated Samples
 The HOPE GM samples in the four incubation series just described are retrieved after
 predetermined  lengths of time.  The times depend on the change in property behavior
 which cannot be estimated apriori. Hence the retrieval time is adjusted for each series.
 The progression of the aging process in each series is monitored by the results of a set
 of physical, chemical and mechanical tests. Table B-4 shows the tests that are used to
 track the behavior of the incubated GM samples. Most of tests are standard test
 methods commonly performed on HOPE  GMs test specimens. The only necessary
 commentary has to do with the different types of OIT tests, since these are the test
 results presented and analyzed in this appendix.

 Table B-4. Tests Used to Evaluate Incubated Samples	
	GM Property	ASTM Test Methods	Test Description	
 crystallinity                 D 1505                    density

 antioxidant amount (total)    D 3895 and D5885         Standard OIT and
                                                     high pressure OIT

 molecular weight (indirect)   D1238                     melt flow index

 mechanical properties       D 638                     tensile properties

 stress crack resistance      05397-appendix            single point notched
	constant tensile load	

 In this study, the total amount of antioxidant remaining in the incubated  GM samples is
 evaluated using two slightly different OIT tests.  They are the standard method and the
 high pressure method. Although OIT cannot identify the individual type or exact amount
 of each antioxidant present in the formulation, it does quantify the effectiveness of the
 antioxidants.  OIT is the time required for the GM test specimen to be oxidized under a
 specific pressure and temperature.  Since antioxidants protect the GM from oxidation,
 the length of OIT (in minutes) indicates the amount of antioxidants present in the test
 specimen. Howard (1973) showed that OIT is proportional to the antioxidant
                                      B-20
-------
concentration in the same formulation package. However, for different antioxidant
packages, direct comparison between two single OIT values can be very misleading
and caution must be expressed.  Since we are only investigating a single GM type and
its antioxidant package, this caution is not a concern in this task.

B-7.1 Standard Oxidative Induction Time (Std-OIT) Test
The Std-OIT test is performed according to ASTM D3895. The test uses a differential
scanning calorimeter (DSC) with a specimen testing cell that can sustain a 35 kPa
gauge pressure. A 5 mg GM specimen is heated from room temperature to 200°C at a
heating rate of 20°C/min under a nitrogen atmosphere. The gas flow rate is maintained
at 50 cc/min.  When  200°C is reached, the cell is maintained in an isothermal condition
for 5 minutes. The gas is then changed from nitrogen to oxygen. The pressure and
flow rate of oxygen are 35 kPa gauge pressure and 50 cc/min, respectively.  The test is
terminated after an exothermal peak is detected. The exothermal peak results from the
oxidation of the  GM specimen. An example thermal curve with its identified OIT value is
shown in Figure B-9.
                                   Oxidation Exotherm
                                       OIT
        ml
      elting peak
        I
nitrogen i
                                                oxygen
                                    Time (min.)
Figure B-9. Thermal curve of a standard OIT test.

B-7.2 High Pressure Oxidative Induction Time (HP-OIT)Test
The HP-OIT test is also performed using a DSC except now with a cell that can sustain
a pressure of 5500 kPa. This type of cell is called a high pressure cell and
consequently the test is called high pressure OIT (HP-OIT).  It is performed according to
ASTM D5885.

Tikuisis et al. (1993) have performed a detailed study on the effect of pressure and
temperature on  HP-OIT values.  A series of such tests was evaluated using 8 different
isothermal temperatures ranging from 150°C to 200°C under 8 different pressures, from
                                     B-21
-------
690 kPa to 5500 kPa.  They found an Arrhenius relationship between temperature and
HP-OIT values.  Pressure had very little influence on the HP-OIT values at temperatures
above 170 °C.  At 150 °C isothermal temperature, pressure greater than 3500 kPa
resulted in little change HP-OIT values.  As a result of their study, the generally agreed
upon pressure and isothermal temperature can be selected.  In the draft ASTM
standard, these values are 3500 kPa and 150°C, respectively.

The test protocol of the HP-OIT test used in this study is that a 5 mg GM specimen is
heated from room temperature to 150°C at a heating rate of 20°C/min under a nitrogen
atmosphere. The pressure of the cell in this nitrogen stage is maintained at 35 kPa
gauge pressure. The gas flow rate is not monitored. When 150°C is reached, the cell is
maintained in an isothermal condition for 5  minutes.  The gas is then changed from
nitrogen to oxygen.  The oxygen pressure in the cell is gradually increased to 3500 kPa
within 1 minute. The test is terminated after an exothermal peak is detected.  The
exothermal peak results from the oxidation  of the GM specimen. The resulting thermal
curve is similar to that shown in Figure B-9, except that the HP-OIT value is much
longer than the Std-OIT value for the same material. This is due to the lower testing
temperature.

B-7.3 Commentary on the Different OIT Tests
The major differences between the two OIT tests are oxygen pressure and isothermal
temperature. For the standard  OIT test, a low pressure and high temperature are used.
For the HP-OIT test, a high pressure and low temperature are utilized. Their differences
create somewhat of a dilemma insofar as the selection of a preferred test method for
OIT.  Table B-5 summaries the advantages and disadvantages.

The main reason behind developing the HP-OIT test is that the 200°C testing
temperature used in Std-OIT test is unable  to bring out the stabilization effect of
thiosynergists and hindered amine types of antioxidants.  As shown in Figure B-7, the
maximum effective temperature of both of these antioxidants is below 150°C. At 200°C,
both types of antioxidants rapidly volatilize from the GM thus losing their apparent
effect.  As a result, GMs with these types of antioxidants will exhibit a shorter OIT value
than those without.  Yet the long term performance of these GMs may be very similar,
or even better than those without these types of antioxidants.  In the HP-OIT test, the
test temperature is lowered to 150°C. Note that 150°C is the minimum temperature to
ensure complete melting of the HOPE GM specimen. The low testing temperature,
however, results in an extremely long test at the standard pressure of 35 kPa, making
the test somewhat unpractical. Hence a high pressure is applied.  At a higher oxygen
pressure, the concentration gradient of oxygen atoms across the specimen's  surface
becomes greater. This increases the number of oxygen atoms diffusing into the  molten
specimen, thereby accelerating the oxidation and reducing the testing time.
                                     B-22
-------
 Table B-5.  Differences Between the Standard and High Pressure PIT Tests	
	Test	Advantages	Disadvantages	
 Std-OIT            •  existing ASTM test protocol     •  high temperature may bias
 (200°C, 35 kPa)     •  short testing time (~ 100 min)     the test results for certain
                    •  standard test apparatus          types of antioxidants

 HP-OIT             •  existing ASTM test protocol     •  long testing time (>300
 (150°C, 3500 kPa)   •  able to distinguish the            min.)
                       stabilization effect of different   •  special testing cell and set
                       types of antioxidants in the       up are required
                       GM
                    •  lower temperature relates
	closer to service conditions	

 B-8 Data Extrapolation Method
 It is well established that chemical reactions of all types proceed more rapidly at higher
 temperatures than at lower temperatures. The relationship between chemical reaction
 rate (Rr) and temperature is usually expressed by the Arrhenius equation, Eq. B-14:

 Rr =Cexp(-Q/RT)                                                       (B-14)

 where:

 Rr =  reaction rate
 C  =  constant (independent of temperature)
 Q  =  activation energy of the reaction
 R  =  gas constant
 T  =  absolute temperature

 Taking the natural logarithm of both sides of Eq.  B-14, Eq. B-15 is obtained.

 ln(Rr) = lnC-Q/RT                                                        (B-15)

 If the log reaction rate is plotted against inverse temperature as shown in Figure B-10,
 the slope of the line will be  -Q/R and the intercept on the vertical axis will be the
 constant "C". The plot in Figure B-10 is called the "Arrhenius Plot" from which reaction
 rates at other temperatures (typically lower temperatures) can be extrapolated.  In order
 to produce a reliable extrapolation, there should  be a minimum of three data points,  i.e.,
 data from three different incubation temperatures, so that the experimental portion of
 the line can be reasonably  established. In addition, the test temperatures cannot  be so
 high that changes in material occur, thereby altering the nature of the reaction.
                                       B-23
-------
             CD
             -CD
c
o
t;
CD
CD
             O)
             O
                            experimentally obtained
                                portion of curve
                                                extrapolated portion
                                                     of curve
(e
high temperature
g. laboratory tests)
low temperature" x
(e.g. site specific condition)
                            Inverse Temperature (1/T)

Figure B-10.  Generalized Arrhenius plot for low-temperature reaction rate
predictions from high-temperature laboratory experimental data.

In this aging study, the four selected testing temperatures are 55, 65, 75 and 85°C.  The
reaction rate being evaluated in this particular report will be the antioxidant depletion
rate. Data obtained from the experimental portion will be extrapolated to a lower site
specific temperature.  Hence, the potential lifetime of the antioxidants in the HOPE GM
can be assessed. In Figure B-3, this  is Stage "A" of the overall predicted lifetime of the
HOPE GM being evaluated.

B-9  Results and Data Analysis on Antioxidant Depletion
This appendix presents the results to date on antioxidant depletion rates.  Incubation
Series I and III are presented.  Series II and IV are ongoing.  As shown in Table B-4,
GM test samples in Series I are incubated in water media under nonstressed conditions
whereas GM test samples in Series III are exposed to water above/air beneath and
under compressive stress.   The antioxidant depletion rate is measured using both the
Std-OIT and HP-OIT tests as explained previously.

B-9.7 Preparation of OIT Test Specimens
The incubated samples from each test series were retrieved after varying incubation
periods. The retrieved samples were equilibrated at room conditions for 24 hours.
They were then cleaned with tap water to remove surface contaminants. The cleaned
samples were placed in a plastic bag and stored inside a cabinet until testing.

OIT test specimens weighing 5 mg each were taken from the incubated samples. They
were cut from surface to surface across the thickness of the GM  near the center portion
                                      B-24
-------
of the sample.  Therefore, the resulting OIT values represent the average amount of
antioxidants across the thickness of the test specimens.

For the Std-OIT tests, three replicates were performed on each incubated sample and
the average value was used in the analysis.  For the HP-OIT tests, a single test was
performed on most of the incubated samples.  Some samples were tested twice to
verify the reproducibility of the test.

B-9.2 Results and Data Analysis of Incubation Series I
HOPE GM samples in incubation Series I were completely immersed in water at
temperatures of 55, 65, 75 and 85°C. The average OIT value for each incubated
sample was evaluated by Std-OIT and HP-OIT tests. In this subsection, the results of
both tests are presented together with the step-by-step data analysis which leads to the
prediction of antioxidant depletion.

The Std-OIT and HP-OIT test results are shown in Table B-6. The OIT values are
presented graphically by plotting OIT value against incubation time. Figures B-11 and
B-12 show Std-OIT and HP-OIT values, respectively.  The curves in both figures
indicate an exponential decrease in the amount of antioxidant present as incubation
time increases.  The curves also exhibit that the decrease in  OIT values  is greater for
the higher incubation temperatures than for the lower temperatures.

Since OIT values decrease exponentially as incubation time increases in both tests, a
linear relationship can be obtained between  In(OIT) and incubation time. Figures B-13
and B-14 show the In(OIT) versus incubation time plots for Std-OIT and HP-OIT,
respectively. The generalized equation for each of the straight lines is expressed by Eq.
B-16.

ln(OIT)=ln(P) + (S)*(t)                                                   (B-16)
                                      B-25
-------
Table B-6. OIT Test Results of Incubation Series I
Incubation
Time
(months)
0.1
1.0
3
9
12
18
30
55°C
Std-OIT
(min)
80.5
79.5
77.0
59.0
45.3
n/a
19.1

HP-OIT
(min)
210*
201
196
173
160
n/a
111
65°C
Std-OIT
(min)
80.5
78.2
74.0
40.2
24.2
17.0
10.8

HP-OIT
(min)
210*
204
157
135
120
109
87
Incubation
Time
(months)
0.1
1.0
3
9
12
18
30
75°C
Std-OIT
(min)
80.5
75.2
69.5
15.1
9.7
10.3
2.1

HP-OIT
(min.)
210*
172
154
82
87
76
38*
85°C
Std-OIT
(min)
80.5
70.5
63.4
12.9
6.2
3.4
0.5

HP-OIT
(min)
210*
181
127
72
50*
38
28*
Notes:
All Std-OIT values are the average of three replicate tests.
All HP-OIT values are from a single test with the exception of those marked with an
asterisk which are the average of two replicate tests.

where

OIT  = OIT time
S    = slope of the lines (i.e., OIT depletion rate)
t     = incubation time
P    = constant (the original value of OIT time in either the Std-OIT or HP-OIT tests)

Table  B-7  lists the depletion rates  that are obtained from  both Std-OIT  and HP-OIT
tests.
                                       B-26
-------
                           10      15     20      25     30      35
                             Incubation Time (month)
  Figure B-11.  Standard OIT versus incubation time plot for incubation
  Series I.
         250
             0      5      10      15     20      25

                             Incubation Time (month)

Figure B-12. HP-OIT versus incubation time plot for incubation Series I.
                                      B-27
-------
g
E,
H
o
                                                      30     35
        0       5      10      15     20      25
                        Incubation Time (month)
Figure B-13. Ln(OIT) versus incubation time plot for incubation
Series I  using Standard OIT tests.
     5.5
<§
H
O
       5  -
     4.5  -
       4  -
     3.5  -
         0      5      10      15     20      25      30     35

                         Incubation Time (month)
   Figure B-14. Ln(OIT) versus incubation time plot for incubation
   series I using HP-OIT test.
                                  B-28
-------
 Table B-7. Antioxidant Depletion Rates of Incubation Series I (i.e., the slopes of
 the straight lines in Figures B-13 and B-14).	
     Test Temperature	Std-OIT	HP-PIT	
           55°C                     -0.0467                    -0.0215
           65°C                     -0.0749                    -0.0342
           75°C                     -0.1280                    -0.0615
	85°C	-0.1765	-0.0822	

 The next step in the analysis is to extrapolate the OIT depletion rate to a lower
 temperature,  such as site specific temperature.  This is performed utilizing the
 Arrhenius equation, as described in Eqs. B-17 and B-18.

 S = A*Exp(-E/RT)                                                         (B-17)

 ln(S) = ln(A) + (-E/R)*(1/T)                                                  (B-18)

 where

 S = OIT depletion rate (slope of the lines listed in Table B-7)
 E = activation energy of the antioxidant depletion mechanism (KJ/mol)
 R = gas constant (8.31 J/mol°K)
 T = test temperature in absolute value (°K)
 A = constant

 Thus  a linear relationship can be established between ln(S) and inverse temperature,
 as indicated in Figure B-15. The activation energy deduced from the slope of the lines is
 43 KJ/mol for Std-OIT and 44 KJ/mol for HP-OIT. These two values are seen to be
 extremely close to one another. The corresponding Arrhenius Equation for Std-OIT and
 HP-OIT are expressed in Eqs. B-19 and B-20.

 ln(S) = 12.839 - 5210.2/T   for Std-OIT                                      (B-19)

 ln(S) = 12.372- 5311.8/T   for HP-OIT                                      (B-20)

 Using Eqs.  B-19 and B-20, the OIT depletion rates at a site specific temperature can be
 obtained. Koerner and Koerner (1995) and Yazadini et al.  (1995) found that the
 temperatures at the base of landfills in Pennsylvania and California, USA are around
 25°C. Thus 25°C is used to demonstrate the extrapolation calculation. OIT depletion
 rates of both tests are as follows:

 S = - 0.0096    for Std-OIT
 S = - 0.0043    for HP-OIT
                                       B-29
-------
        -1
    o:
    Q.
    Q
    H
    O
        -2-
        -3-
                                         y = 12.839-5210.2x RA2 = 0.995
                                              372-5311.8X RA2 = 0.990
        0.0027
0.0028
0.0029
0.0030
0.0031
Figure B-15. Arrhenius plot for incubation Series I (water immersion-
nonstressed).

In order to predict the aging time required to deplete the antioxidants in the HOPE GM
evaluated, Eq. B-16 is utilized. The calculation procedure is as follows:

•  For Std-OIT tests:
  The Std-OIT value for a pure unstabilized (i.e., no antioxidants) HOPE fluff was found
  to be  0.5 minutes. Thus 0.5 minutes is taken to be the OIT value when essentially all
  of antioxidants in the incubated HOPE GMs are consumed.  The calculation to find
  the time for this depletion at a service temperature of 25°C is as follows:

  In (OIT) = ln(P) + (S) *(t)                                                  (B-16)
  In (0.5) = In (80.5) + (-0.0096) (t)
  -0.69  = 4.39 - 0.0096 (t)
  t  = 529 months (44 years)

•  For HP-OIT tests:
  The OIT value for a  pure unstabilized HOPE fluff was found to be 25 minutes. (This
  relatively high value is due to the low isothermal temperature and switching nitrogen
  to oxygen in the test method). Thus 25 minutes is taken to be the OIT value when
  essentially all of the antioxidants in the incubated HOPE GM consumed.
                                      B-30
-------
  In (OIT) = ln(P) + (S)*(t)                                                   (B-16)
  In (25) = In (210) + (-0.0043)*(t)
  3.22 = 5.35 - 0.0043*(t)
  t = 495 months (41 years)

Thus it is seen that the predicted antioxidant lifetime at a service temperature of 25°C is
approximately 40 years for this particular HOPE GM formulation under this set of
immersion conditions.

B-9.3 Results and Data Analysis of Incubation Series III
HOPE GM samples in incubation Series III were exposed to water on top and air
beneath and a compressive stress of 260 kPa. The incubation temperatures are 55, 65,
75 and 85°C. The average amount of antioxidants in each aged sample was evaluated
by both Std-OIT and HP-OIT tests. In this subsection, the results of both tests are
presented together with the step-by-step data analysis which leads to the prediction of
antioxidant depletion.

The Std-OIT and HP-OIT values are shown in Table B-8. Also, the OIT values are
presented graphically by plotting OIT value against incubation time.  Figures B-16 and
B-17 show the Std-OIT and HP-OIT values, respectively.  Similar to Series I, the
depletion of OIT decreases exponentially as incubation time increases. The curves also
exhibit that the decrease in OIT values is greater for the higher incubation temperatures
than for the lower temperatures.

Since OIT values in this incubation series also decrease exponentially as incubation
time increases,  the data extrapolation steps will follow those used in Series I.  Based
on Eq. B-16, a straight line can be formed by plotting In(OIT) versus incubation time, as
indicated in Figures B-18 and B-19 for Std-OIT and HP-OIT, respectively.  The slope of
the lines represent the OIT depletion  rate at each particular temperature. Table B-9
lists the depletion rates that are obtained from both Std-OIT and HP-OIT tests.

Table B-8. OIT Test Results of Incubation Series III
Incubation
Time
(months)
0.1
3
9
12
18
24
55°C
Std-OIT
(min)
80.5
74.3
55.5
54.1
57.0
52.9

HP-OIT
(min)
210*
221
181*
175*
186
167
65°C
Std-OIT
(min)
80.5
77.9
50.5
36.8
19.0
25.9

HP-OIT
(min)
210*
189
164
135
105*
125
                                      B-31
-------
Table B-8 (cont.). PIT Test Results of Incubation Series
Incubation
Time
(months)
0.1
3
9
12
18
24
75°C
Std-OIT
(min)
80.5
66.2
45.3
27.9
17.5
12.6

HP-OIT
(min)
210*
192
143
113
103
92
85°C
Std-OIT
(min)
80.5
55.0
23.5
12.6
4.3
4.0

HP-OIT
(min)
210*
181
113
94
76
38
Notes:
All Std-OIT values are the average of three replicate tests.
All HP-OIT values are from a single test with the exception of those marked with an
asterisk which are the average of two replicate tests.
Table B-9. Antioxidant Depletion Rates of Incubation Series III (i.e., the slopes of
the straight lines in Figures B-18 and B-19).
Test Temperature
55°C
65°C
75°C
85°C
Std-OIT
-0.0217
-0.0589
-0.0798
-0.1404
HP-OIT
-0.0097
-0.0284
-0.0387
-0.0661
The next step in the analysis is to extrapolate the OIT depletion rate to 25°C using Eq.
B-18. Figure B-20 shows the Arrhenius plot for both OIT tests.  The activation energy
deduced from the slope of the lines is 56 KJ/mol for Std-OIT and 58 KJ/mol for HP-OIT.
Again these two values are seen to be extremely similar to one another.  However, they
are both slightly higher than those obtained in incubation Series I. This indicates that
the reaction mechanism for antioxidant depletion in Series III requires more energy than
that in Series I.  In other words, the OIT depletion rate in  Series III is slower compared
to Series I and the correspondingly lifetime prediction will be longer. The corresponding
Arrhenius Equations for Std-OIT and HP-OIT are expressed in Eqs. B-21 and B-22.

ln(S) = 16.885-6738.9/T      forStd-OIT                                   (B-21)

ln(S) = 16.856-6991.3/T      for HP-OIT                                   (B-22)
                                      B-32
-------
           100
                                 10        15
                             Incubation Time (month)
Figure B-16. Standard OIT versus incubation time plot for incubation Series
          250
          200  -
                                 10        15       20
                             Incubation Time (month)
Figure B-17. HP-OIT versus incubation time plot for incubation Series
                                     B-33
-------
       0
                                 10        15        20
                             Incubation Time (month)
Figure B-18.  Ln (OIT) versus incubation time plot for incubation Series III using
Standard OIT tests.
           5.5
             5  -
       E
       H
       0
4.5  -
             4  -
            5.5
                        5        10        15        20

                             Incubation Time (month)
                                                   25
Figure B-19.  Ln (OIT) versus incubation time plot for incubation Series III using
HP-OIT tests.
                                     B-34
-------
       -1
       -2-
CD
-t->
CO
Q:
c
o
    Q- _
    CD
    Q
    o
    =  -4H
       -5
       0.0027
                        Standard OIT
                        HP-OIT
                                   y= 17.045 -6798.2x RA2 = 0.953
                                   y= 16.856 -6991.3x RA2 = 0.943
                 0.0028
0.0029
0.0030
0.0031
Figure B-20. Arrhenius plot for incubation Series III (water top/air beneath-
compression stress).

Using Eqs. B-21 and B-22, the OIT depletion rates can be obtained.
S = - 0.0033
S =-0.0014
            for Std-OIT
            for HP-OIT
In order to predict the aging time that is required to deplete the antioxidants in the
HOPE GMs, Eq. B-16 is utilized. The calculation procedures to obtain the depletion
times at 25°C are  as follows:

•  For Std-OIT tests:
  The OIT value for a pure unstabilized HOPE fluff was found to be 0.5 minutes. Thus
  0.5 minutes is taken to be the OIT value when the antioxidants are consumed.
  In (OIT) = ln(P) + (S) *(t)
  In (0.5) = In (80.5) + (-0.0033)*(t)
  t = 1539 months (128 years)
                                                                      (B-16)
                                      B-35
-------
•  For HP-OIT tests:
  The OIT value for a pure unstabilized HOPE fluff was found to be 25 minutes.  Thus
  25 minutes is taken to be the OIT value when the antioxidants are consumed.

  In (OIT) = ln(P) + (S)*(t)                                                  (B-16)
  ln(25) = ln(210) + (-0.0014)*(t)
  t = 1521 months (126 years)

Thus it is seen that the predicted antioxidant lifetime at a service temperature of 25°C is
approximately 120 years for this particular HOPE GM formulation under this set of
simulated conditions.

B-9.4 Status  of Incubation Series II and IV
Both incubation Series II and  IV were started in June 1998.  Thus there is only a small
amount of data available and  it  is insufficient to perform an analysis and to generate
lifetime predictions.

However, incubation Series II does require a brief discussion.  Instead of using a single
HOPE GM in the evaluation, eight different HOPE GMs are being evaluated.  The GMs
were supplied by five different manufacturers.  Thus different antioxidants were most
likely to have  be used in the GM formulation packages.  In addition, higher
temperatures  are being applied to this particular incubation series.   The test
temperatures  are 55°, 65°C, 75°C, 95°C and 115°C.  The purpose of experimental
design is to investigate the highest possible incubation temperature for HOPE GMs
without changing the antioxidant depletion mechanism. The incubation samples from
the three highest test temperatures are  being retrieved every month in order to monitor
the rate of depletion of antioxidant content.

B-10  Summary
The three distinct stages of aging of HOPE GMs are described in this report. These
stages are (A) depletion time  of antioxidants, (B) induction time to the onset of polymer
degradation and (C) the time  to reach 50% degradation of a particular property.  The
lifetime of the GM is equal to the summation of these three stages.   The focus of this
task, however, is on the depletion time of antioxidants.

Four different incubation conditions were designed to simulate various field applications
of HOPE GMs.  The incubation  environments involve a combination of air, water,
compressive stress, and tensile stress.  In addition, the aging mechanisms in each
incubation condition were accelerated by elevated test temperatures which were set at
55, 65, 75 and 85°C.  In this appendix, only data from two of the incubation series,
Series I and Series III were presented and analyzed.  Samples in Series I were
completely immersed in water without any applied stress. Series III involved samples
                                      B-36
-------
that were exposed to water on top and air beneath, and a compressive stress of 260
kPa. Samples from  both series were retrieved after specified periods of time for
property evaluation.  The antioxidant depletion of the incubated samples was monitored
using both the Std-OIT test and HP-OIT test.

Data obtained from the elevated temperatures tests were then extrapolated to a site
specific (lower) temperature using the Arrhenius model.  For a site specific temperature
of 25°C, the time to consume the antioxidants in this particular HOPE GM formulation
will take 40 years under incubation Series I conditions.  On the other hand, it will take
120 years under incubation Series III conditions.  The shorter depletion time in Series I
is probably  due to the extraction rate of antioxidant which is higher in Series I than in
Series  III. The samples in  Series I were exposed to moving water on all of their
surfaces, whereas samples in Series III were exposed to static water on only one
surface. It  is known that moving water as in the Series I tests actually causes leaching
of antioxidants. Hence, the depletion time for incubation Series I is likely conservative
in comparison to most field situations since it is not common for both sides of the GM to
be exposed to moving liquids. In this regard, the results of the Series  III tests may
better represent HOPE GM in-service  conditions.  It is noted, however, that the Series
III tests were conducted with water rather than leachate. Certain strong leachates may
increase the antioxidant depletion rate. Additional research of this effect is needed.

Regarding the effect of compressive stress on the antioxidant depletion rate in the
Series  III incubation, a definitive result has not yet been obtained. On a preliminary
basis it appears that compressive stress may reduce the depletion rate, since the
depletion time for Series III samples is three times greater than that  of the Series I
samples. One possible hypothesis  is that the compressive stress may increase the
density of the amorphous phase of the HOPE material, consequently reducing the
diffusion rates of both antioxidants out of, and oxygen into,  the GM.

Finally, it should again be emphasized that the antioxidant depletion time represents
only the initial step in a three-step GM aging process, i.e., it is Stage A in Figure B-3.  At
the end of the antioxidant depletion time,  the physical and mechanical properties of the
HOPE  GM still remain essentially unchanged.  In order to establish the service life (i.e.,
the half lifetime) of an HOPE GM, the induction time plus the time to reach 50%
reduction in the relevant mechanical property must be obtained.  This will take longer
than the current incubation time of three years, thus the time frame of this study is
estimated to be ten years.  The second and third parts of the study will be presented in
due course.

B-11 Conclusion
Since this is only the first part in a series of three stages on the topic of HOPE lifetime
prediction, this conclusion will necessarily be preliminary.  Clearly though, this study
establishes that the depletion of antioxidants in the HOPE GM  under investigation is
                                      B-37
-------
quite long. Depending on the incubation method, the time for antioxidant depletion at
25°C is between 40 to 120 years.

These values, in and of themselves, are powerful indicators that HOPE GMs should last
well beyond the 30-year post closure period required in many environmental regulations
without any measurable degradation of mechanical properties.  Clearly, a service
lifetime measured in at least hundreds of years appears to be achievable.  It is hoped
that the results of the ongoing study will allow even better estimates of GM service
lifetime in the near future.

B-12 References
Accorsi, J. and Romero, E., (1995) "Special Carbon Blacks for Plastics" Plastics
   Engineering, April  1995, pp. 29-32.
ASTM D638 - Test Method for Tensile Properties of Plastics.
ASTM D883 - Terminology Relating to Plastics.
ASTM D1238 - Test Method for Flow Rates of Thermoplastics by Extrusion Plastomer.
ASTM D1505 - Test Method for Density of Plastics by the Density-Gradient Technique.
ASTM D1603 - Test Method for Carbon Black in Olefin Plastics.
ASTM D3895 - Test Method for Oxidative Induction Time of Polyolefins by Differential
   Scanning Calorimetry.
ASTM D5397 - Test Method for Evaluation of Stress Crack Resistance of Polyolefin
   Geomembranes Using Notched Constant Tensile Load Test
ASTM D5885 - Test Method for Oxidative Induction Time for Polyolefin Geosynthetics
   by High Pressure Differential Calorimetry.
Billingham, N.C. and Calvert, P.O., (1986), "The Physical Chemistry of Oxidation and
   Stabilization of Polyolefins", Developments in Polymer Stabilization - 3,  Chapter 5,
   edited by Scott, G., Published by Applied Science Publishers Ltd, London, pp. 139-
   190.
Chu, P.P. and Hsieh, E. T.,  (1992), "13C and 1H Nuclear Magnetic Resonance
   Methodologies in Industrial Polymer Research and Production", MQC/MQA and
   CQC/CQA of Geosynthetics, Edited by Koerner, R.M. and Hsuan, Y.G.,  GRI
   Conference Series, Published by IFAI, St Paul, MN., pp. 234-243.
Fay, J. J. and King R. E., (1994),  "Antioxidants for Geosynthetic Resins and
   Applications",  Geosynthetic Resins, Formulations and Manufacturing,  Edited by
   Hsuan, Y.G. and Koerner,  R.M., GRI Conference Series, Published by IFAI, St Paul,
   MN., U.S.A., pp. 77-96.
Grassie, N. and Scott, G., (1985), Polymer Degradation and Stabilization, Published by
   Cambridge  University Press, New York, U.S.A.
Howard, J.B., (1973),  "Data for Control of Stability in Polyolefin Wire and Cable
   Compounds",  Polymer Engineering Science, Vol. 13, No. 6, pp. 429-434.
Koerner, G.R. and Koerner, R.M., (1995), "Temperature Behavior of Field Deployed
   HOPE Geomembranes",  Geosynthetics '95 Conference Proceedings, Nashville,
   TN., pp. 921-938.
Kikkawa, K., Nakahara, Y., and Ohkatsu, Y., (1987), "Antagonism Between Hindered
   Amine Light Stabilizers and Sulfur-Containing Compounds", Polymer Degradation
   Stabilization, Vol. 18, pp. 237-245.
                                     B-38
-------
Luston, J., (1986), "Physical Loss of Stabilizers from Polymers", Developments in
   Polymer Stabilization - 2, Chapter 5, edited by Scott, G., Published by Applied
   Science Publishers Ltd, London, pp. 185-240.
Mitchell, D.H. and Spanner, G.E., (1985), Field Performance Assessment of Synthetic
   Liners for Uranium Tailings Ponds, Status Report, Battelle PNL, U.S. NRC,
   NUREG/CR-4023, PNL-5005.
Michaels, A.S. and Bixler,  H.J., (1961), "Solubility of Gases in Polyethylene", Journal of
   Polymer Science, 50, Vol. L, pp. 393-412.
Osawa, Z. and Ishizuka, T., (1973), "Catalytic Action of Metal Salts in Autoxidation and
   Polymerization. (X) - The Effect of Various Methyl Stearates on the Thermal
   Oxidation of 2,6,10,14-Tetramethylpentadecane", Journal of Applied Polymer
   Science,  Vol.  17, pp. 2897-2907.
Petermann, J., Miles, M. and Gleiter, H., (1976), "Growth of Polymer Crystals During
   Annealing", Journal of Macromolecular Science - Physics, B12(3), pp. 393-404.
Poland, F.G. and Harper, S.R., (1986), "Critical Review and Summary of Leachate and
   Gas Production from Landfills - Final Report", U.S. EPA Cooperative Agreement,
   CR 908 997.
Rapoport,  N.Ya. and Zaikov, G.E., (1986), "Kinetics and Mechanism of the Oxidation of
   Stressed Polymer", Developments in Polymer Stabilization - 4, Chapter 6, edited by
   Scott, G., Published by Applied Science Publishers Ltd, London, pp. 207-258.
Smith, G.D., Karlsoon, K. and Gedde, U.W., (1992), "Modeling of Antioxidant Loss From
   Polyolefins in  Hot-Water Applications. I; Model and Application to Medium Density
   Polyethylene Pipes", Polymer Engineering and Science, Vol. 32, No. 10, pp. 658-667.
Struik, L.C.E., (1978), Physical Aging in Amorphous Polymers and Other Materials,
   Elsevier Scientific Publishing Company, Amsterdam, The Netherlands.
Struve, F., (1994), "Extrusion of Geomembranes", Geosynthetic Resins, Formulations
   and Manufacturing, Edited by Hsuan, Y.G. and Koerner, R.M., GRI Conference
   Series, Published by IFAI, St  Paul, MN., U.S.A., pp. 77-96.
Tikuisis, T., Lam, P., and Cossar, M., (1993), "High Pressure Oxidative Induction Time
   Analysis by Differential Scanning Calorimetry", MQC/MQA and CQC/CQA of
   Geosynthetics, Edited by Koerner, R.M. and Hsuan, Y.G., GRI Conference Series,
   Published by IFAI, St Paul, MN., pp. 191-201.
Viebke, J., Elble,  E., Ifwarson, M. and Gedde, U.W. (1994), "Degradation of
   Unstabilized Medium-Density Polyethylene Pipes in Hot-Water Applications",
   Polymer Engineering and science, Vol. 34,  No. 17, pp. 1354-1361.
Yazadini, R., Campbell, J.L. and  Koerner, G.R., (1995), "Long-Term In-Situ Strain
   Measurements of a High Density Polyethylene Geomembrane in a Municipal Solid
   Waste  Landfill", Geosynthetics '95 Conference Proceedings, Nashville, TN., pp. 893-
   906.

B-13  Acknowledgments
Due to the nature of long-term research, which is typified by this 10-year study,
financing by a number of agencies and sources is necessary. This particular task has
commingled funds in the form of  partnering by different organizations.  The authors
express their sincere appreciation to all of the following:
                                     B-39
-------
1.  US Environmental Protection Agency, via its Risk Reduction Engineering Laboratory
   under Cooperative Agreement No. CR 821448.  Mr. Robert Landreth (retired) and
   Mr. David A. Carson are the Project Officers.

2.  National Science Foundation, via its Geomechanical, Geotechnical and Geo-
   Environmental Systems (G3S) program under Grant No. CMS-9312772.
   Dr. Priscilla P. Nelson is the Project Officer.

3.  The consortium of Geosynthetic Research Institute (GRI) member organization via a
   portion of their membership fees. A listing of the current members is as follows:

GSE Lining Systems, Inc.-  William W. Walling/Melody Adams
Earth Tech Consultants,  Inc. - Walt Studebaker/Charles P. Ballod
U.S. Environmental Protection Agency -David A. Carson
Polyfelt GmbH - Gernot Mannsbart/Philippe Delmas
Browning-Ferris Industries - Charles Rivette/Dan Spikula [BoD]
E.  I. duPont de Nemours & Co., Inc. - John L. Guglielmetti/Ronald J.  Winkler
Federal Highway Administration -Albert F. DiMillio/Jerry A. DiMaggio
Tensar Earth Technologies, Inc. - Peter J. Vanderzee/Donald G. Bright/Mark H. Wayne
National Seal Co. - Gary Kolbasuk [BoD]/George Zagorski
Poly-Flex, Inc. - James Nobert/George Yazdani
Akzo Nobel Geosynthetics  Co. - Wim Voskamp/Joseph Luna
Phillips Petroleum Co. - Rex L. Bobsein [BoD]
GeoSyntec Consultants Inc. - Jean-Pierre Giroud/James A. McKelvey Ill/Majdi Othman
NOVA Chemicals Ltd. - Nolan Edmunds
Tenax,  S.p.A. - Pietro Rimoldi [BoD]/Aigen Zhao
Amoco Fabrics and Fibers Co. - Gary Willibey
U.S. Bureau of Reclamation -Alice I. Comer/Jack Haynes
EMCON - Donald E. Hullings/Mark A. Swyka
Montell USA, Inc. - Robert G. Butala
TC Mirafi, Inc. - Michael M. Koutsourais/Dean Sandri
CETCO - James T. Olsta
Huesker,  Inc. - Thomas G.  Collins
Solvay Polymers - J. Michael Killough
Naue-Fasertechnik GmbH - Georg Heerten/Kent von Maubeuge
Synthetic Industries, Inc. - Marc S. Theisen/Deron N. Austin
Mobil Chemical Co. - PerK. Husby/Frank J. Velisek
BBA Nonwovens - John Matheny/Geoff Kempton
NTH Consultants, Ltd. - Jerome C. Neyer/Robert Sabanas
TRI/Environmental, Inc. - Sam R. Allen [BoD]/Richard Thomas
U.S. Army Corps of Engineers - David L.  Jams [BoD]
Chevron Chemical Co. - Pamela L. Maeger
                                    B-40
-------
Serrot Corp. - Robert A. Otto/Bill Torres [BoD]
Union Chemical Lab (ITRI) - Frank L. Chen
Haley and Aldrich, Inc. - Richard P. Stulgis [BoD]
Westinghouse-Savannah River - Michael Hasek
URS/Greiner/WCC - Pedro C. Repetto/John C.  Volk
S. D. Enterprise Co.,  Ltd. - David Eakin
Solmax Geosynthetiques - Robert Denis
EnviroSource Treatment & Disposal Services, Inc. - Patrick M. McNamara
Strata Systems, Inc. - John N. Paulson [BoD]
CARPI, Inc. -Alberto M. Scuero/JohnA. Wilkes
Rumpke Waste Service, Inc. - Bruce Schmucker
Civil & Environmental Consultants, Inc. - Richard J. Kenter
Firestone Building Products Co. - H. Joseph Kalbas
FITI (GSI-Korea) - Han-Yong Jeon
Waste Management Inc. - James R. (Ron) Jones
CETCO Europe, Ltd.  - Archie Filshill
                                    B-41
-------
                   Appendix C
Field  Performance  Data for Compacted
                   Clay  Liners
                          by

                  David E. Daniel, Ph. D., P.E.
                     University of Illinois
                     Urbana, IL 61801
                      performed under

               EPA Cooperative Agreement Number
                      CR-821448-01-0
                       Project Officer

                     Mr. David A. Carson
            United States Environmental Protection Agency
               Office of Research and Development
           National Risk Management Research Laboratory
                    Cincinnati, OH 45268
-------
                                 Appendix C
           Field Performance Data for Compacted Clay Liners

C-1 Introduction
The  performance of  compacted  clay  liners (CCLs)  constructed from  natural soil
materials and soil-bentonite blends was discussed  in Chapter 4.  A number of graphs
were presented correlating various parameters.

This appendix contains a summary of the data used in compiling the results presented
in  Chapter 4.   The  data are presented in  this appendix  in the form of tables  of
information.  The intent is to provide sufficient information so that future researchers can
add newly acquired data to the database and perform new analyses.  Also, data on
statistical variability  of certain parameters was  collected  and is  summarized  in this
appendix.

C-2 Data for Natural Soil Liner Materials
The data for natural soil liner materials are presented in four attached tables:

   •   Table C-1:  Material properties
   •   Table C-2:  Construction information
   •   Table C-3:  Quality assurance information
   •   Table C-4:  Hydraulic conductivity data

Each of the 89 sites is given a site number, which is shown  in column 1 of all the tables.
The symbols used are defined as follows:

    Clay Fraction = percent on a dry weight basis finer than 2 jim
    DF = maximum depth of penetration of wetting front into soil liner
    i = hydraulic gradient
    k = hydraulic conductivity
    L = thickness of soil liner
    LL = liquid limit of the soil
    MP = modified Proctor (ASTM D-1557)
    OWC = optimum water content
    Percent Fines = percent on a dry weight basis passing  the No. 200 sieve
    Percent Gravel = percent on a dry weight basis retained on the No. 4 sieve
    PI = plasticity index of the soil
    P0 = percent of (w,yd) points lying on or above the line of optimums
    RC = relative compaction (dry unit weight of compacted soil  divided by maximum
        dry unit weight from laboratory compaction test)
    RP =  reduced Proctor (less than the compactive effort from SP)
    ASj = degree of saturation of compacted soil minus degree of saturation on the line
        of optimum for the same dry unit weight
    SP = standard Proctor (ASTM D-698)
                                      c-1
-------
     TSB = two-stage borehole test
     w = water content as a percentage
     wopt = optimum water content
     Yd = dry unit weight
     Yd,max = maximum dry unit weight
     o' = effective stress in kPa
     \l/o = initial suction of soil liner

Some of the columns of data contain three data entries, one above the other, with the
following meaning:

   •  Upper number is the number of data points
   •  Middle number is the average  (geometric mean for hydraulic conductivity)
   •  Lower number is the standard  deviation

C-3  Data for Soil-Bentonite Admixed Liners
Data for soil-bentonite admixed liners are presented in tables as follows:

   •  Table C-4:  Material properties
   •  Table C-5:  Construction information
   •  Table C-6:  Quality assurance  data
   •  Table C-7:  Hydraulic conductivity data

The symbols are the same as those given in section C-2.
                                      c-2
-------
    Table C-1. Material Properties for Natural Clay Liner Materials in Database.
Site
No.
1


2

3




4
5

6

7


8

Location
and Date
SCA
Wilsonville, IL
Oct. 1992
Confidential

Keele Valley
Toronto, OT
1990
PADB
Livingston, LA
1987
Confidential

Confidential

Imperial, PA
Dec. 1990

Confidential

Source of Data
Benson et al.
1992

Benson & Boutwell
1992
Lahtietal. 1987
Readesetal. 1990

Johnson et al.
1990

Benson & Boutwell
1992
Benson & Boutwell
1992
GeoSyntec Report


Benson & Boutwell
1992
LL (%)
-
24
-
-
58
-
25
-
9
50
3
-
43
-
32
8
33
1
-
35
PI (%)
-
10
-
-
29
-
10
-
9
34
3
-
26
-
19
8
13
1
-
22
Percent
Gravel
-
4
-
-
;
-
2
-
-
0
-
-
3
3

8
5
2
-
1
Percent
Fines
-
65
-
-
85
-
85
-
-
95
-
-
87
-
88
8
77
6
-
75
Clay
Fraction
(%<2 urn)
-
37
-
-
50
-
22
-
-
47
-
-
32
-
35
8
27
2
-
45
wopt
(%)
10.2
9.0

26.8

12.3


17.9


14.3

13.5

14.1


14.5

Yd, max
(kN/m3)
20.1
21.3

14.6

19.0


16.8


18.6

19.5

18.6


18.8

Compactive
Effort
SP
MP

SP

SP


SP


MP

MP

SP


MP

o
-------
    Table C-1.  Material Properties for Natural Clay Liner Materials in Database (Continued).
Site
No.
9

10

11



12
13


14


15


16


17


Location
and Date
Sauk City, Wl
1988
Portage, Wl
1988
Marathon, Wl
1988
Marathon, Wl
1988

Imperial, PA
April 1991

Test Fill 2
July 1988

Confidential


Test Fill 1


Livingston, LA
Pad A
Oct. 1988
Source of Data
Gordon et al.
1989
Gordon et al.
1989
Gordon et al.
1989
Gordon et al.
1989

GeoSyntec Report


Mundell & Boos
1990

Benson & Boutwell
1992

Mundell & Boos
1990

Johnson et al.
1990

LL (%)
-
55
32
43
-
57
-
55
-
8
37
1
3
40
8
12
85
3
24
41
9
9
50
3
PI (%)
-
31
32
21
-
30
-
28
-
8
15
1
3
20
3
12
58
3
24
22
6
9
34
3
Percent
Gravel
-
4
-
1
-
-
-
-
-
8
2
1
6
0
-
1
0
-
20
0
-
-
0
-
Percent
Fines
-
_
-
~
-
-
-
-
-
8
78
5
6
70
8
1
99
-
20
77
-
-
95
-
Clay
(%<2 urn)
-
45
-
29
-
39
-
33
-
8
37
4
6
25
11
-
57
-
20
38
8
-
47
-
wopt
(%)
12.7

16.6

21.7

23.0


18.0


16.2


25.8


15.8


20.3


Yd, max
(kN/m3)
18.6

18.7

17.3

16.6


17.0


16.7


14.6


17.0


16.4


Compactive
Effort
MP

MP

MP

MP


SP


SP


SP


SP


SP


o
-------
    Table C-1. Material Properties for Natural Clay Liner Materials in Database (Continued).
Site
No.
18

19


20



21
22

23


24

25

Location
and Date
Tangipahoa Landfill
Amite, LA
March 1992
Confidential


Confidential

Confidential



1993
Green County, Wl
1987

Confidential

Confidential

Source of Data
Boutwell &
McManis
1995
Benson & Boutwell
1992

Personal Files

Trast
1993

Othman &
Luettich 1 994
Krantz & Bailey
1990

Trast
1993
Trast
1993
LL (%)
10
30
6
12
32
3
-
49
-
51
-
-
63
20
39
4.2
-
67
-
53
PI (%)
10
18
4
12
14
2
-
23
-
26
-
-
42
20
18
3.8
-
46
-
41
Percent
Gravel
5
0
-
12
1
1
-
1
-
1
-
-
;
-
-
-
-
0
-
0
Percent
Fines
5
52
3
12
85
3
-
94
-
90
-
-
96
-
73
-
-
94
-
88
Clay
Fraction
(%<2 urn)
5
16
1
12
44
4
-
43
-
36
-
-
;
-
30
-
53

-
36
wopt
(%)
13.0

10.5


18.5

11.8
18.0

20.5

20.0


16.0
21.5
11.5
16.1
Yd, max
(kN/m3)
18.7

20.1


17.2

18.5
17.0

16.3

16.5


18.4
16.3
19.8
18.0
Compactive
Effort
SP

MP


SP

MP
SP

SP

SP


MP
SP
MP
SP
o
CJ1
-------
   Table C-1.  Material Properties for Natural Clay Liner Materials in Database (Continued).
Site
No.
26


27


28




29
30

31


32

33

34

Location
and Date
Confidential


Confidential


Confidential


ERG Facility
Milan, Ml
1993
Confidential

Confidential


Confidential

Confidential

Confidential

Source of Data
Trast
1993

Trast
1993

Trast
1993

Bergstrom et al.
1995

Personal Files

Personal Files


Personal Files

Personal Files

Personal Files

LL (%)
-
33
-
-
31
-
-
35
-
-
27
-
-
32
7
40
1
-
45
-
29
-
44
PI (%)
-
19
-
-
18
-
-
19
-
-
10
-
-
19
7
24
1
-
27
-
15
-
16
Percent
Gravel
-
7
-
-
8
-
-
3
-
-
2
-
-
;
7
7
4
-
0
-
1
-
0
Percent
Fines
-
85
-
-
74
-
-
89
-
-
76
-
-
;
7
58
3
-
99
-
87
-
96
Clay
Fraction
(%<2 urn)
-
37
-
-
26
-
-
41
-
-
28
-
-
;
7
23
1
-
42
-
40
-
_
wopt
(%)
12.2
17.5
18.5
12.5
16.5
18.5
11.5
16.6
18.5
9.0
13.0
14.4
14.0

12.4


11.0

13.3


17.3
Yd, max
(kN/m3)
19.3
17.7
17.1
19.4
17.8
17.2
19.4
17.5
17.0
20.5
19.1
18.6
18.6

19.3


19.9

18.9


17.1
Compactive
Effort
MP
SP
RP
MP
SP
RP
MP
SP
RP
MP
SP
RP
MP

SP


MP

MP


SP
o
en
-------
   Table C-1. Material Properties for Natural Clay Liner Materials in Database (Continued).
Site
No.
35

36


37




38
39


40


41



42


Location
and Date
Confidential

Confidential


Indianapolis, IN
1994

ISGS Prototype
Urbana, IL
1986
ISGS Field-Scale
Urbana, IL
April 1988
Confidential


BP Chemicals
SDRI 1
Port Lavaca, TX
Nov. 1988
Celanese
Bishop, TX
July 1986
Source of Data
Personal Files

Personal Files


Personal Files


ISGS
Report

ISGS
Report

Personal Files


McBride-Ratcliff
Report


Personal Files


LL (%)
-
39
9
36
2.5
3
36
3
-
21
-
-
21
-
15
101
5
-
47
_

3
69
3.6
PI (%)
-
19
9
17
1.6
3
17
2
-
7
-
-
7
-
15
71
5
-
30
_

3
45
3.0
Percent
Gravel
-
0
9
2
1.6
3
10
5
-
9
-
-
9
-
-
0
-
-
.
_

2
0
0
Percent
Fines
-
97
9
74
2.6
3
48
3
-
60
-
-
60
-
-
98
-
-
66
_

3
79
3.0
Clay
Fraction
(%<2 urn)
-
-
9
30
3.4
3
16
1
-
26
- (4 urn)






-
.
_

2
49
4.2
wopt
(%)

22.2

13.2


12.4

10.3


10.3


31.6


19.5



23.4


Yd, max
(kN/m3)

17.7

18.3


19.0

20.4


20.4


13.4


16.3



15.1


Compactive
Effort

SP

SP


SP

SP


SP


SP


SP



SP


o
-------
   Table C-1. Material Properties for Natural Clay Liner Materials in Database (Continued).
Site
No.
43


44




45
46


47


48


49


Location
and Date
GCWDA Test Fill A
Texas City, TX
Nov. 1988
GCWDA Test Fill B
Texas City, TX
Nov. 1988
Texas Eastman
Longview, TX
1987
Puckett Plant
Ft. Stockton, TX
April 1988
Shell
Deer Park, TX
Dec. 1988
Confidential


Confidential


Source of Data
Personal Files


Personal Files


H.B. Zachry Co.
Report

Personal Files


Personal Files


Personal Files


Personal Files


LL (%)
119
62
4.1
119
62
4.1
8
44
4.0
31
35
1.8
41
39
4.7
60
41
2.7
60
42
1.7
PI (%)
119
42
4.3
119
42
4.3
8
28
3.2
31
16
2.3
41
24.3
5.2
60
23
-
60
22
-
Percent
Gravel
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Percent
Fines
119
86
6.2
119
86
6.2
8
70
2.4
31
98
1.1
41
69.5
-
-
86
-
-
86
-
Clay
Fraction
(%<2 urn)
-
-
-
-
-
-
-
-
-
2
22
8.4
-
-
-
-
-
-
-
-
-
wopt
(%)
22.4


22.4


19.5


23.3


14.6


20.0
18.0
13.3
20.0
18.0
13.3
Yd, max
(kN/m3)
15.4


15.4


16.4


15.4


17.7


16.2
16.7
18.7
16.2
16.7
18.7
Compactive
Effort
SP


SP


SP


SP


SP


RP
SP
MP
RP
SP
MP
o
oo
-------
   Table C-1. Material Properties for Natural Clay Liner Materials in Database (Continued).

Site
No.
50


51


52




53
54



55



56



57




Location
and Date
Confidential


Confidential


Emelle, AL
Oct. 1984

Confidential


Quarantine Rd
Landfill
Baltimore, MD
Jan. 1994
Savannah
River Plant Panel
A1
March 1988
Savannah
River Plant Panel
A2
March 1988
Savannah
River Plant Panel
B1
March 1988


Source of Data
Personal Files


Personal Files


Colder Assoc.
Report

Personal Files


Personal Files



Mueser-Rutledge
Report


Mueser-Rutledge
Report


Mueser-Rutledge
Report




LL (%)
88
43
3.4
62
40
1.9
2
37
1.4
8
54
2
-
_
.

-
66
.

-
66
_

-
69
_



PI (%)
88
24
-
62
22
-
2
18
2.8
8
31
3
-
_
.

-
35
.

-
35
_

-
38
_


Percent
Gravel
-
-
-
-
-
-
3
10
8.9
8
0
0
-
_
.

-
0
.

-
0
_

-
0
_


Percent
Fines
-
86
-
-
86
-
3
73
15.1
-
-
-
-
_
.

-
93
.

-
93
_

-
98
_

Clay
Fraction
(%<2 urn)
-
-
-
-
-
-
3
38
9.0
8
40
3
-
_
.

-
-
.

-
-
_

-
-
_


wopt
(%)
20.0
18.0
13.3
20.0
18.0
13.3
19.9


19.9


-
_
.

27.4



27.4



26.8




Yd, max
(kN/m3)
16.2
16.7
18.7
16.2
16.7
18.7
16.5


16.4


-
_
.

14.5



14.5



14.6




Compactive
Effort
RP
SP
MP
RP
SP
MP
SP


SP


-
_
.

SP



SP



SP



o
CD
-------
   Table C-1. Material Properties for Natural Clay Liner Materials in Database (Continued).

Site
No.
58



59





60

61



62



63



64




Location
and Date
Savannah River
Plant
Panel B2
March 1988
Savannah
River Plant Panel
B3
March 1988
Savannah
River Plant Panel
C1
March 1988
Savannah
River Plant Panel
C2
March 1988
Savannah
River Plant Panel
D1
March 1988
Savannah
River Plant Panel
D2
March 1988
BP Chemicals
Port Lavaca, TX
SDRI2
Dec. 1988


Source of Data
Mueser-Rutledge
Report


Mueser-Rutledge
Report


Mueser-Rutledge
Report


Mueser-Rutledge
Report


Mueser-Rutledge
Report


Mueser-Rutledge
Report


McBride-Ratcliff
Report




LL (%)
-
69
_

-
69
_

-
68
-

-
68
_

-
51
_

-
51
.

-
47
_



PI (%)
-
38
_

-
38
_

-
35
-

-
35
_

-
20
_

-
20
.

-
30
_


Percent
Gravel
-
0
_

-
0
_

-
0
-

-
0
_

-
0
_

-
0
.

-
-
_


Percent
Fines
-
98
_

-
98
_

-
95
-

-
95
_

-
73
_

-
73
.

-
66
_

Clay
Fraction
(%<2 urn)
-
-
_

-
-
_

-
-
-

-
-
_

-
-
_

-
-
.

-
-
_


wopt
(%)
26.8



26.8



26.6



26.6



20.2



20.2



19.5




Yd, max
(kN/m3)
14.6



14.6



14.6



14.6



15.9



15.9



16.3




Compactive
Effort
SP



SP



SP



SP



SP



SP



SP



o
o
-------
   Table C-1. Material Properties for Natural Clay Liner Materials in Database (Continued).
Site
No.
65


66



67
68


69


70

71

72


Location
and Date
BP Chemicals
Port Lavaca, TX
SDRI3
Dec. 1988
Confidential

Confidential


Confidential


Confidential


Confidential

Confidential

SDDS Longtree LF
Igloo, SD
Feb. 1990
Source of Data
McBride-Ratcliff
Report

Personal Files

Personal Files


Personal Files


Personal Files


Personal Files

Personal Files

S. Dakota Disposal
Systems Report

LL (%)
-
47

-
50
-
49
-
4
35
1
4
22
1
-
42
-
29
3
36
2.0
PI (%)
-
31

-
29
-
27
-
4
17
1
4
9
1
-
26
-
19
3
20
1.7
Percent
Gravel
-
-

-
;
-
-
-
4
2
2
4
6
3
-
0
-
4
-
0
-
Percent
Fines
-
66

-
75
-
62
-
4
67
10
4
50
2
-
88
-
83
-
85
-
Clay
Fraction
(%<2 urn)
-
-

-
;
-
-
-
4
22
4
4
16
1
-
45
-
34
-
35
-
wopt
(%)
13.5


19.0

19.3


14.8
11.5

10.0
8.5

14.9

12.2

18.0


Yd, max
(kN/m3)
19.2


16.1

16.1


17.7
19.0

19.9
21.4

18.7

19.6

16.5


Compactive
Effort
MP


SP

SP


SP
MP

SP
MP

MP

MP

SP


9
-------
   Table C-1.  Material Properties for Natural Clay Liner Materials in Database (Continued).
Site
No.
73


74


75



76
77


78


79


80


Location
and Date
Sea Drift, TX
Sept. 1988

Sea Drift, TX
Sept. 1988

McClellandtown, PA
Sept. 1990
Confidential


Arnoni LF Pad 1
Pittsburgh, PA
Feb. 1994
Arnoni LF Pad 2
Pittsburgh PA
Feb. 1994
DuPont Pad 1
Victoria, TX
Jan. 1989
DuPont Pad 2
Victoria, TX
Jan 1989
Source of Data
McClelland
Engineers Report

McClelland
Engineers Report

Cumberland Geot.,
Consultants Report
Personal Files


Personal Files


Personal Files


Engineering
Sciences Report

Engineering
Science Report

LL (%)
4
76
6.0
4
56
6.0
-
;
4
37
3.0
45
32
0.8
45
32
1.6
12
62
5.3
17
52
1.3
PI (%)
4
53
6.0
4
40
5.0
-
;
4
17
2.0
45
13
0.9
45
16
1.3
12
41
4.9
17
35
1.2
Percent
Gravel
-
-
-
-
-
-
-
;
-
1
-
-
-
-
-
-
-
-
-
-
-
-
-
Percent
Fines
-
-
-
-
64
-
-
;
-
92
-
-
-
-
-
-
-
9
82
3.2
15
84
6.0
Clay
Fraction
(%<2 urn)
-
-
-
-
-
-
-
;
-
-
-
-
=19
-
-
=25
-
-
-
-
-
-
-
wopt
(%)
21.0


18.0


21.0

19.2


9.9


11.5


25.0
17.8

19.6
14.4

Yd, max
(kN/m3)
15.5


16.9


15.6

16.6


19.7


19.6


14.9
16.5

15.9
18.0

Compactive
Effort
-


-


SP

SP


SP


SP


SP
MP

SP
MP

9
ro
-------
   Table C-1.  Material Properties for Natural Clay Liner Materials in Database (Continued).
Site
No.
81


82
83
84

85

86


87


Location
and Date
GE
Waterford, NY
July 1989
Findlay Township,
PA
Aug. 1988
Findlay Township,
PA
Aug. 1988
Montezuma Hills,
CA
Pad A (Dark)
Feb. 1991
Montezuma Hills,
CA
Pad B (Light)
Feb. 1991
Fernald, OH
Pad 1 (Ln. 1)
Nov. 1996
Fernald, OH
Pad 1 (Ln. 2)
Nov. 1996
Source of Data
Clough Harbor &
Assoc. Report

Paul Rizzo
& Assoc. Report
Paul Rizzo
& Assoc. Report
IT Corp.
Report

IT Corp.
Report

GeoSyntec Report


GeoSyntec Report


LL (%)
-
47
-
-
4
39
3.6
-

-

-
43
-
-
43
-
PI (%)
-
22
-
-
4
16
1.9
-

-

-
24
-
-
24
-
Percent
Gravel
-
-
-
-
4
10
3.3
-

-

-
-
-
-
-
-
Percent
Fines
-
-
-
4
84
5.4
4
81
2.7
-

-

-
84
-
-
84
-
Clay
Fraction
(%<2 urn)
-
-
-
4
54
6.7
4
48
8.6
-

-

-
37
-
-
37
-
wopt
25



18.2




17.7


17.7


Yd, max
(kN/m3)
15.3



17.6




17.1


17.1


Compactive
Effort
SP



SP




SP


SP


9
CO
-------
   Table C-1. Material Properties for Natural Clay Liner Materials in Database (Continued).
Site
No.
88


89


Location
and Date
Fernald, OH
Pad2(Ln. 1)
Nov. 1996
Fernald, OH
Pad 2 (Ln. 2)
Nov. 1996
Source of Data
GeoSyntec Report


GeoSyntec Report


LL (%)
-
25
-
-
25
-
PI (%)
-
14
-
-
14
-
Percent
Gravel
-
-
-
-
-
-
Percent
Fines
-
70
-
-
70
-
Clay
Fraction
(%<2 urn)
-
29
-
-
29
-
wopt
(%)
11.6


11.6


Yd, max
(kN/m3)
19.1


19.1


Compactive
Effort
SP


SP


9
-------
   Table C-2. Construction Information for Natural Clay Liner Materials in Database.
Site
No.
1

2
3

4

5



6
7

8

9

10

11

12

13

Compaction Criteria
w>OWC
RC > 90% MP
None
w>OWC
RC > 95% SP
w > OWC + 2, <+8
RC > 90% MP
w>OWC
RC > 90% MP
w>OWC
RC > 90% MP

w > OWC + 2
RC > 90% MP
w > OWC -2 to +4
RC > 90% MP
w>OWC
RC > 90% MP
w>OWC
RC > 90% MP
w>OWC
RC > 90% MP
w>OWC
RC > 90% MP
w > OWC + 2, <+5
RC > 90% MP
Compactor
CAT 825

Bomag 210PD
Rex 370

CAT 81 5

RexTrashmaster

CAT 825


CAT 825

Dynapac CT25

-

-

-

-

CAT 825

Compactor
Mass (kg)
32,400

-
30,000

19,800

36,000

32,400


32,400

12,600

-

-

-

-

32,400

Passes per
Lift
6

6
4

-

6

5


4

4

-

-

-

-

4

Lift Thickness
(mm)
150

150
150

150

150

150


150

150

150

150

150

150

150

Number of
Lifts
6

5
8

4

10

6


8

6

10

10

10

10

8

Pad Size
(m x m or m2)
36x15

32x14
30x30

15 x30

Liner

29x12


15x24

24x18

Liner

Liner

Liner

Liner

15x24

9
en
-------
   Table C-2. Construction Information for Natural Clay Liner Materials in Database (continued).
Site
No.
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Compaction Criteria
w > OWC + 2, <+5
RC > 90% MP
w>OWC
R0100%SP
w > OWC + 2, <+5
RC > 90% MP
w > OWC + 2, <+6
RC > 90% SP
Sj>78.5
RC > 90% MP
w>OWC
RC > 90% MP
Sj>82.0
w>OWC
RC > 95% SP
Sj>85.0
w>OWC
RC > 90% MP
w>OWC
RC > 90% MP
w>OWC
RC > 90% MP
w>OWC
RC > 90% MP
w>OWC
RC > 90% MP
Compactor
-
-
CAT 835
CAT 81 5
CAT D7G bulldozer
CAT 825
CAT 825
CAT 825
-
-
CAT815A
CAT815A
Dynapac CA25
Dynapac CA25
Compactor
Mass (kg)
-
-
39,000
19,800
25,000
32,400
32,400
32,400
-
-
18,900
18,900
18,900
18,900
Passes per
Lift
-
-
-
6
4
5
8
6
-
-
8-12
8- 12
4-6
4-6
Lift Thickness
(mm)
170
200
170
150
150
150
150
150
-
150
150
150
150
150
Number of
Lifts
6
7
5
4
5
10
6
6
-
10
5
5
6
6
Pad Size
(m x m or m2)
9x14
12x26
9x9
15x30
30x12
Liner
45x20
58x26
-
Liner
31 x15
31 x15
27x17
27x17
9
en
-------
   Table C-2. Construction Information for Natural Clay Liner Materials in Database (continued).
Site
No.
28
29
30
31
32
33
34
35
36
37
38
39
40
Compaction Criteria
w>OWC
RC > 90% MP
w > OWC-2, +5
RC > 90% MP
RC > 90% MP
w>OWC
RC > 95% SP
w > OWC, <+6
RC > 90% MP
w>OWC
RC > 90% MP
w>OWC
RC > 95% SP
w>OWC
RC > 95% SP
w > OWC, <+6
RC > 95% SP
w>OWC
RC > 95% SP
w>OWC
RC > 90% MP
w: 11 to 12%
RC > 90% SP
w > OWC, +5
RC > 92% SP
Compactor
Dynapac CA25
CAT 825
RexTrashmaster
CAT S563
CAT 825
CAT815B
IR SPF-56 & CAT
815
CAT 824B
CAT 81 5
FWD741
Hyster C852A
CAT815B
sheepsfoot
Compactor
Mass (kg)
18,900
32,400
27,000
-
32,400
19,800
19,800
32,400
17,100
-
-
19,800
59 kg/lin. cm
Passes per
Lift
4-6
6
-
-
-

4
(2 each)
6
8
4
12
12
-
Lift Thickness
(mm)
150
170
150
300
150
150
150
150
60
60
150
130
150
Number of
Lifts
6
9
6
2
8
8
3
6
6
5
6
6
6
Pad Size
(m x m or m2)
27x17
32x16
-
13X26
15x40
15x30
-
-
15x30
-
3x9
14.6x7.3
8x26
9
-------
   Table C-2. Construction Information for Natural Clay Liner Materials in Database (continued).
Site
No.
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
Compaction Criteria
RC > 95% SP
-
w > OWC+1
RC > 95% SP
w > OWC+1
RC > 90% SP
w > OWC, <+2
RC > 95% SP
w> OWC+1, <+3
RC > 95% SP
w > OWC+1 , <+5
RC > 90% SP
w > OWC, <+3
RC > 95% SP
w > OWC, <+3
RC > 95% SP
w > OWC, <+3
RC > 95% SP
w > OWC, <+3
RC > 95% SP
w > OWC, <+3
RC > 95% SP
-
-
-
Compactor
CAT 81 5
wedgefoot
IR SPF-48
IR SPF-48
CAT 81 5 &Bomag
BW213PD
Dynapac CA25
CAT815B
CAT815B
CAT815B
CAT815B
CAT815B
-
CAT 825C
-
CAT815B
Compactor
Mass (kg)
19,800
-
7,200
7,200
19,800
14,000
10,900
19,800
19,800
19,800
19,800
19,800
-
-
-
19,800
Passes per
Lifs
40

16
8
4
2
8
6- 10
5-8
3
3
4-5
-
4
-
12
Lift Thickness
(mm)
150
150
150
150
150
200 - 250
85
150
150
150
150
-
-
150
160
Number of
Lifts
4
4
5
5
4
5
10
4
4
4
4
-
4
4
4
Pad Size
(m x m or m2)
93
30x15
37x9
37x9
288
15x30
46x24
46x15
46x15
46x15
46x15
150
-
-
483
9
oo
-------
   Table C-2. Construction Information for Natural Clay Liner Materials in Database (continued).
Site
No.
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
Compaction Criteria
-
-
-
-
-
-
-
-
RC > 95% SP
RC>91% MP
w>OWC+1,<+5
RC > 95% SP
w>OWC+1,<+5
RC > 95% SP
w>OWC+1,<+5
RC > 95% SP
w>OWC+1,<+5
RC > 95% SP
w > OWC-2, <+4
RC > 90% MP
w > OWC-2, <+4
RC > 90% MP
w > OWC, <+6
RC > 95% SP
Compactor
Rex 3-50A & CAT
815B
CAT815B
CAT815B
CAT815B
CAT815B
CAT815B
CAT815B
CAT815B
CAT 81 5
CAT 81 5
CAT 81 5
CAT 81 5
CAT815B
CAT815B
-
-
CAT 825C
Compactor
Mass (kg)
19,800
19,800
19,800
19,800
19,800
19,800
19,800
19,800
19,800
19,800
19,800
19,800
19,800
19,800
-
-
32,400
Passes per
Lift
6 (Lifts 1-3)
21 (Lift 4)
6
12
12
12
12
12
12
40
80
2
2
6
6
7 to 10
7 to 10
8
Lift Thickness
(mm)
130
140
150
170
170
190
150
230
150
150
100
100
150
150
150
150
150
Number of
Lifts
4
4
4
4
4
4
4
4
4
4
10
11
4
4
6
6
4
Pad Size
(m x m or m2)
483
483
483
483
483
483
483
483
93
186
12x26
12x26
15x36
15x36
24x18
24x18
18x36
9
CD
-------
   Table C-2. Construction Information for Natural Clay Liner Materials in Database (continued).
Site
No.
73

74

75

76



77
78

79

80

81

82

83

85

Compaction Criteria
w > OWC,  95% SP
w > OWC,  95% SP
w > OWC+3, <+6
RC > 95% SP
w > OWC+3, <+5
RC > 95% SP
w > OWC+0.4
RC > 98% SP

w > OWC+1 .5
RC > 94% SP
w>OWC
RC > 95% SP
w>OWC
RC > 95% SP
w > OWC+4

RC > 96% SP

RC > 96% SP

RC > 90% MP

Compactor
CAT815B

CAT815B

CAT815B

CAT815B

IRSD-100D


IRSD-100D

CAT815B

CAT815B

Dresser VOS
PD84A
CAT 825 &
Vib. Smooth Drum
CAT 825 &
Vib. Smooth Drum
CAT815B

Compactor
Mass (kg)
19,800

19,800

19,800

19,800

10,200


10,200

19,800

19,800

16,200

32,400
-
32,400
-
19,800

Passes per
Lift
22

22

-

6

10


4

8

8

4

6
2
6
2
-

Lift Thickness
(mm)
1 @200
6@100
1 @200
6@100
1 @200
3@150
150

100


100

150

150

150

150

150

1 @300
3@200
Number of
Lifts
7

7

4

4

9


9

8

8

4

4

4

20x24

Pad Size
(m x m or m2)
12x23

12x23

15x30

18x30

15x30


15x30

465

465

465

223

223

-

s
o
-------
  Table C-2. Construction Information for Natural Clay Liner Materials in Database (continued).
Site
No.
85
86
87
88
89
Compaction Criteria
RC > 90% MP
W > OWC, <+4; RC > 95% SP
W > OWC, <+4; RC > 95% SP
W > OWC, <+4; RC > 95% SP
W > OWC, <+4; RC > 95% SP
Compactor
CAT815B
CAT 81 5
CAT 81 5
CAT 81 5
CAT 81 5
Compactor
Mass (kg)
19,800
19,800
19,800
19,800
19,800
Passes per
Lift
-
4
7
4
6
Lift Thickness
(mm)
1 @300
3@200
150
150
150
150
Number of
Lifts
20x24
6
6
6
6
Pad Size
(m x m or m2)
-
13x15
13x15
13x15
13x15
o
ro
-------
   Table C-3. Quality Control/Quality Assurance Data for Natural Clay Liners in Database.
Site
No.
1

2

3
4
5

6

7

8

9


w (%)
34
10.3
0.8
57
26.6
2.2
13.8
4
21.3
0.5
21
17.3
2.2
32
13.8
0.9
33
17.2
1.5
17
15.3
1.2
85
19.6
1.9

yd (kN/m3)
34
19.8
0.036
57
14.4
4.0
19.4
4
16.0
0.17
21
17.3
0.51
37
19.0
0.31
33
17.7
0.47
17
17.7
0.9
85
17.0
0.31

Po
44

28

98
80
95

32

88

8

90


ASj
-2.0

-4.0

+17.7
+3.0
-3.0

-8.2

+1.0

-12.6

+5.8


Distress
None

None

None
None
None

None

None

None

None


Purpose
Verify k< 1 x 10'7 cm/s

Verify k< 1 x 10~7 cm/s

Verify k< 1 x 10'8 cm/s
Verify k< 1 x 10'7 cm/s ;
show KF = KL using standard
construction methods
Verify k< 1 x 10~7 cm/s

Verify k< 1 x 10'7 cm/s

Verify k< 1 x 10~7 cm/s

Verify k< 1 x 10'7 cm/s

Monitor liner performance and
Verify k< 1 x 10'7 cm/s


Remarks





Compacted slightly wet of modified
Proctor optimum and wet of line of
optimums


Met CQA Spec, but dry of line of
optimums







£
ro
-------
   Table C-3. Quality Control/Quality Assurance Data for Natural Clay Liners in Database (continued).
Site
No.
10

11

12



13
14


15


16


17
18



w (%)
93
17.8
1.5
91
25.4
3.2
289
26.0
2.3
34
20.7
0.6
18
17.0
0.9
48
30.8
0.5
11
19.8
1.2
16
23.3
1.2
20
16.6
0.9

yd (kN/m3)
93
16.9
0.34
9100
16.0
0.66
289
16.1
0.50
34
16.7
0.24
18
16.8
0.16
48
14.1
0.27
11
16.1
0.22
16
15.7
0.44
20
17.35
0.25

Po
50

75

78


100


78

48
98
-

91

100

85


ASj
+3.5

-7.4

+4.6


+9.0


+4.0

48
+8.0
0.03

+6.0

+3.0

+4.0


Distress
None

None

None

None


None


None


None


None


None

Purpose
Monitor liner performance and
verify k< 1 x 10~7 cm/s

Monitor liner performance and
verify k< 1 x 10~7 cm/s

Monitor liner performance and
verify k< 1 x 10~7 cm/s

Verify k< 1 x 10~7 cm/s


Verify k< 1 x 10~7 cm/s


Verify k< 1 x 10'7 cm/s


Verify k< 1 x 10~7 cm/s


Verify k < 1 x 10~7 cm/s; show
KF = KL using standard
construction methods
Verify k< 1 x 10~7 cm/s



Remarks






















£
CO
-------
   Table C-3. Quality Control/Quality Assurance Data for Natural Clay Liners in Database (continued).
Site
No.
19

20

21

22

23
24

25

26
27

w (%)
584
13.6
0.72
37
17.6
0.52
18
19.5
0.3
-
60
22.0
2.8
19
23.6
1.1
18
18.9
0.31
53
15.5
36
13.5

yd (kN/m3)
584
19.0
0.19
37
16.9
0.33
18
16.9
0.15
-
60
16.4
0.57
19
15.8
0.32
18
16.9
0.42
53
17.6
36
18.0

Po
81

8

80

-
89

81

71

17
6

ASj
+3.8

-6.2

+4.3

-3.0
+7.4

+0.4

-0.5

-8.8
-10.4

Distress
None

None

None

None
None

None

None

None
None

Purpose
Monitor liner performance and
verify k< 1 x 10~7 cm/s

Verify k< 1 x 10'7 cm/s

Veryfy Suitability of Si Ity
Material for k < 1 x 10~7 cm/s

Verify k< 1 x 10~7 cm/s
Monitor liner performance and
perify k< 1 x 10~7 cm/s

Verify k< 1 x 10'7 cm/s

Verify k< 1 x 10~7 cm/s

Verify k< 1 x 10'7 cm/s
Verify k< 1 x 10~7 cm/s

Remarks


Compacted dry of line of optimums
using acceptable zone approach



Spec, required that So > 90%






In spec, but dry of line of optimums
In spec, but dry of line of optimums
s
-------
   Table C-3. Quality Control/Quality Assurance Data for Natural Clay Liners in Database (continued).
Site
No.
28

29


30



31
32

33

34

35

36


w (%)
54
16.2
92
13.9
0.71
-
16.2
-
13.1
-
-
13.9
-
13.4
-
17.8
-
20.7
-
15.5

yd (kN/m3)
54
17.7
92
18.8
0.28
-
18.6
-
19.1
-
-
19.2
-
18.7
-
17.1
-
16.8
-
17.6

Po

57

84


65

75


92

80

45

78

77

ASj

-0.3

+1.0


-2.3

-1.5


+7.5

+0.7

+2.5

+2.8

+3.4

Distress
None

None


None

None


None

None

None

None

None


Purpose
Verify K< 1 x 10~7 cm/s

Verify K< 1 x 1Q-7cm/s


Verify K< 1 x 1Q-7cm/s

Verify K< 1 x 1Q-7cm/s


Verify K< 1 x 1Q-7cm/s

Verify K< 1 x 10'7 cm/s

Verify K< 1 x 10~7 cm/s

Verify K< 1 x 1Q-7cm/s

Verify K< 1 x 1Q-7cm/s


Remarks
In spec, but straddles line of optimums

















Constructed with mine spoil

£
en
-------
   Table C-3. Quality Control/Quality Assurance Data for Natural Clay Liners in Database (continued).
Site
No.
37

38


39




40
41


42


43


44


45



w (%)
-
14.1
24
11.5
3.6
57
11.6
1.1
40
35.5
0.5
13
21.9
0.88
26
25.0
1.5
49
23.4
1.12
49
24.2
1.5
31
19.8
1.03

yd (kN/m3)
-
18.2
24
20.4
0.77
57
17.9
0.82
40
12.8
0.17
13
16.0
0.22
26
15.1
0.36
49
15.4
0.28
49
15.0
0.40
31
103.8
2.61

Po

45
Raw
Data
NA

10


100


92


81


63


47


71


ASj

-1.2

+ 10.2


-19


+0.03


+5.6


+5.5


+3.9


+ 1.0


+0.4


Distress
None

None


None


Desiccation
(Hot HOPE)

None


None


None


None


None



Purpose
Verify k< 1 x 10~7 cm/s

Verify k< 1 x 10'7 cm/s


Verify k< 1 x 10~7 cm/s


Verify k< 1 x 10~7 cm/s


Verify k< 1 x 10~7 cm/s


Verify k< 1 x 10'7 cm/s


Verify k< 1 x 10~7 cm/s


Verify k< 1 x 10'7 cm/s


Verify k< 1 x 10~7 cm/s



Remarks
Disturbance in tube by gravel
















East Side = A, West = B;
used light roller




Nothing unusual


£
en
-------
   Table C-3. Quality Control/Quality Assurance Data for Natural Clay Liners in Database (continued).
Site
No.
46


47




48

49



50



51



52


53



w (%)
32
27.3
3.4
51
16.5
1.25
160
17.8
0.40

152
18.9

1.05
216
18.6

1.16
152
17.8

1.10
9
21.2
1.08
32
21.6
2.2

yd (kN/m3)
32
15.4
0.22
51
17.7
0.50
160
17.0
1.1

152
16.7

0.25
216
16.9

0.37
152
17.0

0.40
9
16.1
0.32
32
15.5
0.36

Po

100


100


75



86



84



73



67

32
71


ASj

+8.0


+9.9


1.1



0.3



1.3



-0.7



+ 1.5

32
+0.02
0.04

Distress
None


None


potentially
desiccation or
freeze-thaw
damage

potentially
desiccation or
freeze-thaw
damage
potentially
desiccation or
freeze-thaw
damage
potentially
desiccation or
freeze-thaw
damage
None


None



Purpose
Verify k< 1 x 10~7 cm/s


Verify k< 1 x 10'7 cm/s


Verify k< 1 x 10~7 cm/s



Verify k< 1 x 10~7 cm/s



Verify k< 1 x 10'7 cm/s



Verify k< 1 x 10'7 cm/s



Verify k< 1 x 10'7 cm/s


Verify k< 1 x 10~7 cm/s



Remarks
Cell 1EhadSDRI,
Gs = 2.67 measured




Pads at Sites 48-51 were constructed
with same material by 4 different
contractors. Objective in each case to
obtain KF < 10~7 cm/s, with low bid/low
K contractor winning job.


















o
ro
-------
   Table C-3. Quality Control/Quality Assurance Data for Natural Clay Liners in Database (continued).
Site
No.
54
55

56



57
58


59


60


61



w (%)
-
27.0
0.7
30.6
0.7
-
29.6
1.1
-
30.7
1.3
-
29.4
1.3
-
26.8
0.8
-
29.8
0.4

yd (kN/m3)
-
15.0
0.22
14.2
0.17
-
14.0
0.30
-
14.3
0.22
-
14.4
0.31
-
15.1
0.20
-
14.4
0.14

Po
-
100*

100*


100*


100*


100*


100*


100*


ASj
-
+5.9

+6.6


+0.5


+8.6


+6.3


+8.4


+7.6


Distress
None
None

None

None


None


None


None


None



Purpose
Verify k< 1 x 10~7 cm/s
Verify suitability of soil for
k< 1 x 10~7 cm/s;
w « wopt
Verify suitability of soil for
k< 1 x 1Q-7cm/s;
w = wopt+3%
Verify suitability of soil for
k < 1 x 1 0'7 cm/s

Verify suitability of soil for
k < 1 x 1 0'7 cm/s

Verify suitability of soil for
k < 1 x 1 0'7 cm/s

Verify suitability of soil for
k< 1 x 10'7 cm/s;
vv~wopt
Verify suitability of soil for
k< 1 x 10'7 cm/s;
w = wopt+3%

Remarks
SDRI Test on Liner
k>1 x 10-7 because soil wasn't wet
enough

Wetting soil up to opt. +3% lowered k
(compared to site 55)
















£
oo
-------
   Table C-3. Quality Control/Quality Assurance Data for Natural Clay Liners in Database (continued).
Site
No.
62
63
64


65
66

67

68

69

70

w (%)
24.6
0.6
22.7
0.5
8
21.6
0.32
23
17.2
1.07
39
21.7
1.3
59
21.41.2

13
17.6
1.0
8
11.5
0.6
20.6

yd (kN/m3)
15.4
0.17
15.4
0.16
8
16.0
0.22
23
17.4
0.31
59
17.2
0.52
59
17.2
0.52
13
18.0
0.38
8
19.4
1.6
16.1

Po
100*
100*
88

0

95

98

100

75

60

ASj
+10.9
+3.2
+4.3

-6.1

+8.6

+8.1

+6.4

+8.0

-

Distress
None
None
None

None

None

None

None

None

None

Purpose
Verify suitability of soil for
k < 1 x 1 0'7 cm/s
Verify suitability of soil for
k < 1 x 1 0'7 cm/s
Verify k< 1 x 10 cm/s

Verify k< 1 x 10 cm/s

Verify k< 1 x 10 cm/s

Verify k< 1 x 10" cm/s

Verify k< 1 x 10" cm/s

Verify k< 1 x 10 cm/s

Verify k< 1 x 10" cm/s

Remarks















£
CD
-------
   Table C-3. Quality Control/Quality Assurance Data for Natural Clay Liners in Database (continued).
Site
No.
71
72
73


74
75

76

77

78


w (%)
14.3
6
23.7
1.0
36
25.2
1.2
30
19.6
1.6
7
25.4
2.0
76
21.8
0.4
111
11.0
0.6
109
12.4
1.5

yd (kN/m3)
18.0
6
15.5
0.2
36
14.8
0.86
30
16.1
0.02
7
15.2
0.7
76
15.9
0.3
111
19.2
0.6
109
18.8
0.5

Po
64
100
97

47

100

86

37

2


ASj
-
11.3
5

-3.7

11.1

1.1

1.1

-7.2


Distress
None
freeze-thaw,
but upper lift
re-worked
before SDRI
None

None

None

None

None

None


Purpose
Verify k< 1 x 10 cm/s
Verify k< 1 x 10 cm/s
Verify k< 1 x 10 cm/s

Verify k< 1 x 10 cm/s

Verify k< 1 x 10 cm/s

Verify k< 1 x 10 cm/s

Verify k< 1 x 10 cm/s

Verify k< 1 x 10 cm/s


Remarks


On site clay

Off site clay









9
CO
o
-------
   Table C-3. Quality Control/Quality Assurance Data for Natural Clay Liners in Database (continued).
Site
No.
79


80


81




82
83


84
85
86


87



w (%)
39
28.2
3.1
37
23.1
2.2
8
28.0
2.5
14
17.8
2.0
16
19.3
1.5
-
-
18
20.5
2.3
19
20.4
2.2

yd (kN/m3)
37
16.2
0.49
39
14.9
0.46
8
14.2
0.20
14
17.1
0.44
16
17.3
0.53
-
-
18
16.6
0.68
19
16.6
0.61

Po
-


-


-


-


-


-
-

100


95


ASj
12.6


11.7


2.7


2.4


12


-
-

5.5


4.1


Distress
None


None


None


None


None


None
None

None


None


Purpose
Verify k< 1 x 10 cm/s


Verify k< 1 x 10 cm/s


Verify k< 1 x 10 cm/s


Verify k< 1 x 10 cm/s


Verify k< 1 x 10 cm/s


Verify k< 1 x 10 cm/s
Verify k< 1 x 10 cm/s
Verify k< 1 x 10 cm/s


Verify k< 1 x 10 cm/s



Remarks















Dark clay
Light clay






9
CO
-------
   Table C-3.  Quality Control/Quality Assurance Data for Natural Clay Liners in Database (continued).
Site
No.
88


89



w (%)
24
13.2
0.8
29
13.2
1.1

yd (kN/m3)
24
19.2
0.35
29
19.1
0.45

Po

100


100


ASj

11.7


10.8


Distress

None


None


Purpose
Verify k< 1 x 10 cm/s


Verify k< 1 x 10 cm/s



Remarks






o
CO
ro
-------
   Table C-4.  Hydraulic Conductivity for Natural Clay Liner Materials in Database.
Site
No.
1


2

3



4
5


6


7


8


Thin-Wall Sampling Tube
k (cm/s)
4
3.2x10-8
0.32
2
3.6x10-9
109
8.0x10-9

4
5.0x10-9
0.34
3
8.7x10-9
0.21
4
2.4x10-8
0.46
8
8.4x10-8
0.35
5
9.0x10-9
0.58
Method, a', i
D5084
69
10
-
-
Flexible-Wall
165
20
9100
-
-
D5084
69
10
Flexible-Wall
-
-
D5084
-
-
D5084
-
-
SDRI
k (cm/s)
2.8x10-7



1.5x10-7
-

1.1 X10-7


9x10-9


2.7x10-7


5.8x10-8


1.2 X10-7


Size (m2)
1.44



1.82
-


2.33

1.49


2.33


2.33


2.33


Lysimeter
k (cm/s)

-


-
9x10-9

-


-


-


-


-


Size (m2)

-


-
15x15

-


-


-


-


-


TSB
k (cm/s)

-


-
-

-


-


-


5
4.3x10-8
0.12
-


30 cm Block
k (cm/s)

2.6x10-7


-
-

-


4x10-8


-


-


-



DF/L

1


1
-

0.3


0.6


0.7


-


-


Vo
(kPa)

60


70
-

-


80


70


-


-


9
CO
CO
-------
   Table C-4. Hydraulic Conductivity for Natural Clay Liner Materials in Database (continued).
Site
No.
9
10
11
12
13
14
15
16
Thin-Wall Sampling Tube
k (cm/s)
1.0x10-8
8.0x10-9
2.0x10-9
3.0x10-9
8
1.3x10-8
0.18
4
4.8x10-8
0.29
10
4.4x10-9
0.48
7
3.7x10-8
0.48
Method, a', i
-
-
-
-
D5084
69
-
9100
D5084
SDRI
k (cm/s)




1.3x10-8
2.0x10-8
3.3x10-9
3.0x10-8
Size (m2)




2.33
2.33
2.33
2.33
Lysimeter
k (cm/s)
7x10-9
3x10-8
3x10-9
2x10-9




Size (m2)








TSB
k (cm/s)




5
1.4x10-8
0.16

4
1.6x10-8
0.21

30 cm Block
k (cm/s)








DF/L




0.1
0.2
0.7
0.1
Vo
(kPa)








9
CO
-------
   Table C-4.  Hydraulic Conductivity for Natural Clay Liner Materials in Database (continued).
Site
No.
17


18


19



20
21


22

23

24


Thin-Wall Sampling Tube
k
7
3x10-9
0.19
5
1.5x10-8
0.12
8
1.9x10-8
0.46
9
3.0x10-9
0.63
2
3.1 X10-7
-
-
2.4x10-8
6
1.5x10-8
0.45
2
9x10-9
-
Method, a', i
9100
34
-
D5084
34
-
-
-
D5084
35
-
D5084
69
10
-
;
-
-
D5084
69
10
SDRI
k (cm/s)



9.8x10-9


-

8x10-7


2.5x10-7


2x10-8

-

1.5x10-8


Size (m2)



2.33


-

2.33


2.33


2.33

-

2.33


Lysimeter
k (cm/s)
6
6x10-9
0.25
-


4.4x10-
8

-


-


-

1.4x10-
8

-


Size (m2)
0.37


-


8x8

-


-


-

-

-


TSB
k (cm/s)
6
5x10-9
0.23
8
9.2x10-9
0.26
-

-


-


-

-

-


30 cm Block
k (cm/s)
-


4
1.4x10-8
0.34
-

-


2
2.2x10-7
-
-

-

1.1 X10-8


DF/L
1.0


-


-

1


1


-

-

1


Vo
(kPa)
-


-


-

30


-


-

-

45


9
CO
en
-------
   Table C-4.  Hydraulic Conductivity for Natural Clay Liner Materials in Database (continued).
Site
No.
25


26


27




28
29


30


31


32

Thin-Wall Sampling Tube
k (cm/s)
2
2.3x10-9
-
2
2.9x10-9
-
2
3.0x10-8
-
2
1.9x10-8
-
2
2.2x10-8
-
-
3.0x10-8
-
7
1 .6 x 1 0-8
0.26
2
3.0x10-8
Method, a', i
D5084
69
10
D5084
69
10
D5084
69
10
D5084
69
10
D5084
69
10
Flexible-Wall
-
22
D5084
-
-
Flexible-Wall
_
SDRI
k (cm/s)
8x10-9


2.0x10-7


1.8x10-7


9x10-8


3
1.7x10-8
-
1.1 X10-7


6.0x10-8


3.9x10-8

Size (m2)
2.33


2.33


2.33


2.33


1.85


2.33


2.33


2.33

Lysimeter
k (cm/s)
-


-


-


-


-


-


-


-

Size (m2)
-


-


-


-


-


-


-


-

TSB
k (cm/s)
-


-


-


-


-


-


6
4.7x10-8
0.034
-

30 cm Block
k (cm/s
6x10-9


1.8x10-7


1.5x10-7


1.7x10-7


2
1.7x10-8
-
-


~


-


DF/L
1


1


1


1


1


>0.7


1


1

Vo
(kPa)
35


-


-


-


-


-


32


0

9
CO
en
-------
   Table C-4. Hydraulic Conductivity for Natural Clay Liner Materials in Database (continued).
Site
No.
33

34

35


36
37

38
39


40


41
Thin-Wall Sampling Tube
k (cm/s)
2
1.3x10-8
6
1.5x10-8
0.62
2
3.0x10-8
6
9.1 X10-9
0.58
-
4.9x10-8
-
6
2.6x10-8
0.14
7
3.5x10-9
0.35
5.5x10-9
Method, a', i
-
;
D5084
69
10
-
-
D5084
21
20
Flexible-Wall
-
-
D5084
14
10
D5084
34
-
-
SDRI
k (cm/s)
3.9x10-8

4x10-7

3.7x10-8

3.0x10-8

1.3x10-8

<3. 6x1 0-8
2.6x10-9

4.3x10-9
2.2x10-8


1.0x10-7
Size (m2)
2.33

2.33

2.33

2.33

2.33

0.16
0.08 m2

1 .76 m2



2.33
Lysimeter
k (cm/s)
-

-

-

-

-


No
Flow

-


-
Size (m2)
-

-

-

-

-


-


-


-
TSB
K (cm/s)
-

-

-

-

-


-


7
1.6x10-8
0.33
-
30 cm Block
k (cm/s)
-

3
3.5x10-7
0.23
-

-

-


-


-


4.1x10-9

DF/L
1

1

>0.7

>0.7

0.5

0.1
0.5


0.8


0.24
Vo
(kPa)
0

-

-

25

~


-


-


-
9
CO
-------
   Table C-4.  Hydraulic Conductivity for Natural Clay Liner  Materials in Database (continued).
Site
No.
42
43


44


45




46
47
48


49


50


51


Thin-Wall Sampling Tube
k (cm/s)
-
3
2.4x10-9
0.12
3
2.4x10-9
0.13
12
5.8x10-9
0.63
9
1.5x10-8
0.12
-
3
1.1 X10-8
0.21
4
5.1 X10-8
0.67
3
7.4x10-8
0.31
3
4.1 X10-8
0.15
Method, a', i
-
Const. Head
-
-
Const. Head
-
-
-
-
-
-
-
-
-
D5084
69
10
D5084
69
10
D5084
69
10
D5084
69
10
SDRI
k (cm/s)
8x10-8
7x10-8


2x10-7



3.7x10-8


2x10-8

5x10-8
4x10-8


5.0x10-8


2.6x10-7


3.0x10-7


Size (m2)
2.33
2.33


2.33



2.33


2.33

2.33
2.33


2.33


2.33


2.33


Lysimeter
k (cm/s)
-
-


-



-

-


-
-


-


-


-


Size (m2)
-
-


-



-

-


-
-


-


-


-


TSB
k (cm/s)
-
-


-


-


-


-
5
2.1x10-8
0.57
5
3.2x10-7
1.07
5
7.5x10-8
1.20
5
1.1 X10-7
1.08
30 cm Block
k (cm/s)
-
-


-


-


-


-
4.8x10-8


7.7x10-8


3.1x10-6


5.3x10-7



DF/L
-
-


-


0.5


-


0.5
1


1


1


1


Vo
(kPa)
-
-


-


-


-


20
32


35


34


22


9
CO
oo
-------
   Table C-4.  Hydraulic Conductivity for Natural Clay Liner  Materials in Database (continued).
Site
No.
52


53


54
55


56
57

58
59
60
Thin-Wall Sampling Tube
k (cm/s)
-
-
-
4
1.7x10-8
0.21
-
-
8.1 X10-8
2.8x10-8
-
-
3.4x10-8
2.5x10-8
2.7x10-8
3.4x10-8
Method, a', i
-
-
-
D5084
-
-
-
-
-
:
-
-
-
-
-
-
SDRI
k (cm/s)
2
1.1 X10-7
0.10
2.2x10-8


7x10-8
1.3x10-7

2.4x10-8

5.6x10-8

5.0x10-8
9.4x10-8
1.2x10-7
Size (m2)
7.20


2.33


2.33
2.33

2.33

2.33

2.33
2.33
2.33
Lysimeter
k (cm/s)
-


-


-
-

-

-




Size (m2)
-


-


-
-

-

-




TSB
k (cm/s)
-


5
1.2x10-8
0.35
-
-

-

-




30 cm Block
k (cm/s)
-


-


-
-

-

-





DF/L
-


0.2


1
0.67

0.63

0.71

0.71
0.54
0.63
Vo
(kPa)
2


-


70
-

-

-




9
CO
CD
-------
Table C-4.  Hydraulic Conductivity for Natural Clay Liner Materials in Database (continued).
Site
No.
61

62
63

64
65
66


67


68


69


Thin-Wall Sampling Tube
k (cm/s)
-
4.3x10-8
1 .6 x 1 0-7
1.7x10-7
5.5x10-9
-
-
3
3.7 x10-8
0.31
3
3.0 x10-8
0.30
4
7.8 x10-9
0.14
4
2.1 x10-8
0.33
Method, a', i
-
;
-
-
;
-
-
D5084
-
-
D5084
-
-
D5084
22
34
D5084
22
34
SDRI
k (cm/s)
3.7x10-8

3.1 x10-7
3.9x10-7
2.3x10-7

1.8x10-7

1.2x10-8


8.3 x10-8

2
2.3 x10-8
0.017
2
1.3x10-8
0.002
Size (m2)
2.33

2.33
2.33
2.33

2.33

2.33


2.33


2.33


2.33

Lysimeter
k (cm/s)
-



-

-
-


-


-


-


Size (m2)
-



-

-
-


-


-


-


TSB
k (cm/s)
-



-

-
5
1.1 x10-8
0.26
5
8.5 x10-8
0.21
5
2.6 x10-8
0.11
5
5.6x10-8
0.12
30 cm Block
K (cm/s)
-



4.1 X10-9

-
-


-


-


-



DF/L
0.63

0.75
0.54
0.25

0.80

>0.5


>0.5


>0.7


>0.7

Vo
(kPa)
-



-

-

26


34


60


46

-------
Table C-4.  Hydraulic Conductivity for Natural Clay Liner Materials in Database (continued).
Site
No.
70
71
72
73
74
75
76
77
78
79
80
81
Thin-Wall Sampling Tube
k (cm/s)
2x10-8
2x10-8
2
1 .4 x 1 0-8
0.06

-
-
4
4.7x10-8
1.1
-
-
5
3.3x10-9
0.22
3
1.8x10-9
0.15
2
4.2x10-8
0.27
Method, a', i
Flexible-Wall
Flexible-Wall
52

-
-
52
-
-
-
-
-
SDRI
k (cm/s)
4x10-8
8.3x10-8
2.0x10-8
8x10-8
1 X10-9
5x10-8
3x10-8
2x10-8
2x10-8
2
4.5x10-8
2
4.0x10-8
1.5x10-7
Size (m2)
2.33
2.33
2.33
2.33
2.33
2.33
2.33
2.33
2.33
2.33
2.33
2.33
Lysimeter
k (cm/s)




-
-

-
-



Size (m2)




-
-

-
-



TSB
k (cm/s)




-
-

-
-



30 cm Block
k (cm/s)




-
-

-
-



DF/L


0.5
0.6
0.5
1
1
0.5
0.5

0.4
0.8
Vo
(kPa)




-
-

-
-



-------
   Table C-4.  Hydraulic Conductivity for Natural Clay Liner Materials in Data Base (continued).
Site
No.
82


83


84
85


86
87

88
89
Thin-Wall Sampling Tube
k (cm/s)
4
1.5x10-8
0.29
4
1.7x10-8
0.05
-
-
2
2.2x10-8
-
2
2.6x10-8
2
3.9x10-8
2
3.1 X10-8
Method, a', i
Flexible-Wall
34
-
Flexible-Wall
34
-
-
-
Flexible-wall
14
-
Flexible-wall
14
Flexible-wall
14
Flexible-wall
14
SDRI
k (cm/s)
3x10-8


4.5x10-8


1.3x10-7
2.8x10-8
1.5x10-8


1.4x10-8

2.3x10-8
2.1 X10-8
Size (m2)
2.33


2.33


2.33
2.33
2.25


2.25

2.25
2.25
Lysimeter
k (cm/s)
-


-


-
-
-


-



Size (m2)
-


-


-
-
-


-



TSB
k (cm/s)
-


-


-
-
-


-



30 cm Block
k (cm/s)
-


-


-
-
-


-




DF/L
0.4


0.4


0.8
0.8
-


-



Vo
(kPa)
-


-


-
-
-


-



o

4*.
IV)
-------
   Table C-5.  Material Properties for Soil-Bentonite Liners in Database.
Site
No.
1

2
3

4
5

6

7


8

Location
and Date
Oxford, NJ
1991
Southern
Nebraska
Southern
Nebraska
Southern
Nebraska

Southern
Nebraska
Kettleman
City, CA
1987
Kettleman
City, CA
1987
Borfer, TX
1988

Source of Data
Colder Assoc

D.L. Osadnick
D.L. Osadnick
D.L. Osadnick

D.L. Osadnick

Colder Assoc.

Colder Assoc.


McBride-Ratcliff


LL (%)
-
;
31
51
31
51
31
51
-
31
51
-
-
-
.
-
-
56

PI (%)
-
;
31
36
31
36
31
36
-
31
36
29
-
-
29
-
-
31
Percent
Gravel
-
;
32
13
32
13
32
13
-
32
13
-
-
-
.
-
-
-
Percent
Fines
-
;
32
32
32
32
32
32
-
32
32
81
-
-
81
-
-
55
Percent
Bentonite)
-
3.75
9.0
9.0
9.0
-
-
9.0



.


7.8
wopt
(%)

-
15.0
15.0
15.0


15.0
23.8


23.8


18.6
Yd, max
(kN/m3)

-
17.1
17.1
17.1


17.1
15.4


15.4


16.9
Compactive
Effort


SP
SP
SP


SP
SP


SP


SP
o
-------
   Table C-5.  Material Properties for Soil-Bentonite Liners in Database (continued).
Site
No.
9

10


11




12
Location
and Date
Borfer, TX
1988
San Mateo
County, Ca
1993
Lead, South
Dakota
1994
Mobile, AZ
1990

Source of Data
McBride-Ratcliff

BFI


Colder Assoc.


Colder Assoc.


LL (%)
-
65
-
51
-
-
.
-
-
60
-
PI (%)
-
39
-
36
-
-
.
-
-
38
-
Percent
Gravel
-
;
-
2
-
-
.
-
-
-
-
Percent
Fines
-
63
-
21
-
-
.
-
-
39
-
Percent
Bentonite)

10.5

10.0

-
14.7
-

4.0

wopt
(%)

20.1

9.0


17.7


13.5

Yd, max
(kN/m3)

16.5

19.9


16.8


18.5

Compactive
Effort

SP

MP


SP


SP

o
-------
   Table C-6. Construction Information for Soil Bentonite Liners in Database.
Site
No.
1
2
3
4
5
6
7
8
9
10
11
12
Compaction Criteria

w > OWC, <+4
RC > 95% SP
w > OWC, <+4
RC > 95% SP
w > OWC, <+4
RC > 95% SP
w > OWC, <+4
RC > 95% SP
w>OWC
RC > 90% MP
w > OWC +3
RC > 90% SP
w>OWC
RC > 95% SP
w > OWC +2
RC > 92% SP
w > OWC+2, +5
RC > 905 MP
w > OWC, <+3
RC > 98% SP
w > OWC +2
RC > 95% SP
Compactor
Ingersol
Rand S100
CAT 81 5
CAT 81 5
CAT 81 5
CAT 81 5
CAT 81 5
CAT 81 5
CAT 81 5
CAT 81 5
CAT 825
-
CAT 81 5 (1-4
lift), CAT
CP433B (5th
lift), Sakai SV
(6th lift)
Compactor
Mass (kg)
-
20,000
20,000
20,000
20,000
20,000
20,000
12,600
20,000
32,400
-

Passes per
Lift
10
6
6
4
4
2
2
6
6
4
-
4
Lift Thickness
(mm)
150
150
150
150
150
150
150
150
150
150
150-230
150
Number of
Lifts
4
6
6
6
6
7
7
6
6
6
3
6
Pad Size
(m x m or m2)
9x9
31 x11
31 x11
31 x11
31 x11
43x15
43x15
13x28
13x28
15x15
11 x11
36x18
o

-k
en
-------
   Table C-7. Quality Control/Quality Assurance Data for Soil-Bentonite Liners in Database.
Site
No.
1


2

3



4
5

6

7

8

9


w (%)
28
12.3
1.4
2
14.7
2
16.0
2
143
-
2
15.2
-
28.4
-
28.4
38
20.2
38
21.4

yd (kN/m3)
28
16.0
0.3
2
17.2
2
16.7
2
16.8
-
2
16.5
-
14.8
-
14.8
38
15.9
38
15.9

Po

-


~

~

-


~

54

54

~

"

ASj

-


~

~

-


~

~

~

~

"

Distress
None


None

None


None

None

None

None

None

None


Purpose
Verify k< 1 x 10'7 cm/s


Verify k< 1 x 10~7 cm/s

Verify k< 1 x 10'8 cm/s

Verify k< 1 x 10'7 cm/s


Verify k< 1 x 10'7 cm/s

Verify k< 1 x 10~7 cm/s

Verify k< 1 x 10'7 cm/s

Verify k< 1 x 10~7 cm/s

Verify k< 1 x 10'7 cm/s


Remarks




















o
-------
Table C-7. Quality Control/Qualit
Site
No.
10


11

12


w (%)
34
12.5
0.77
8
18.5
32
15.5

yd (kN/m3)
34
19.1
0.22
8
16.8
32
17.9

Po

100


75

"

ASj

-


~

"
\j Assurance Data for Soil-Bentonite Liners in Database (continued)

Distress
None


None

None


Purpose
Verify k< 1 x 10~7 cm/s


Verify k< 1 x 10'7 cm/s

Verify k< 1 x 10~7 cm/s


Remarks







o
-------
   Table C-8.  Hydraulic Conductivity for Soil-Bentonite Liners in Database.

Site
1

2
3

4
5

6
7

8

Thin-Wall Sampling Tube
k (cm/s)
4
5.5x10-8
2
3.0x10-8
2
1.9x10-8
2
6.0x10-8
-
2
7.5x10-8
7
6.9x10-9
7
6.9x10-9
0.35
-
_
Method, a', i
D5084
-
D5084
34
D5084
34
D5084
34
-
D5084
34
-
-
-
-
_
SDRI
k (cm/s)

?
3.0x10-8
1.0x10-8
3.0x10-8


2.0x10-8
1.6x10-8
6.2x10-8


2.2x10-8
Size (m2)

-
2.33
2.33
2.33


2.33
2.33
2.33


2.33
Lysimeter
k (cm/s)













Size (m2)













TSB
k (cm/s)













30 cm Block
k (cm/s)














DF/L













Vo
(kPa)













o

4*.
oo
-------
   Table C-8.  Hydraulic Conductivity for Soil-Bentonite Liners in Database (continued).

Site
9

10
11

12
Thin-Wall Sampling Tube
k (cm/s)
-
-
10
2.6x10-8
-
6
3.2x10-8
-
Method, a', i
-
-
-
-
;
-
SDRI
k (cm/s)

1.0x10-7
3.0x10-8
2.0x10-9
2.0x10-8

Size (m2)

2.33
2.33
2.33
2.33

Lysimeter
k (cm/s)






Size (m2)






TSB
k (cm/s)






30 cm Block
k (cm/s)







DF/L






Vo
(kPa)






o

4*.
CD
-------
                 Appendix D
Cincinnati  Geosynthetic Clay Liner
                   Test Site
                        by

                David E. Daniel, Ph.D., P.E.
                   University of Illinois
                   Urbana, IL 61081
                    performed under

             EPA Cooperative Agreement Number
                    CR-821448-01-0
                    Project Officer

                  Mr. David A. Carson
         United States Environmental Protection Agency
             Office of Research and Development
         National Risk Management Research Laboratory
                  Cincinnati, OH 45268
-------
                                 Appendix D
               Cincinnati Geosynthetic Clay Liner Test Site

D-1  Introduction
This appendix contains additional information that augments Chapter 3 in the main body
of the report regarding the research program related to field test plots constructed at a
site in Cincinnati, Ohio.  The test plots were constructed to evaluate the internal shear
resistance of geosynthetic clay  liners  (GCLs) that were constructed on 2H:1V and
3H:1V slopes in prototype landfill cover systems.

D-2  Test Plots
The  main objective in  constructing the field test plots was to  investigate the internal
(mid-plane) shear  strength of GCLs in carefully  controlled, field-scale tests.  Other
objectives were to verify that GCLs in landfill cover systems will remain stable on 3H:1V
slopes with a factor of safety of at least 1.5, to monitor the displacement and creep of
GCLs in the  field for as long as possible,  to  develop  information  on erosion  control
materials, and to better understand the field performance of GCLs as a component in
liner and cover systems.

Fourteen test plots have been constructed at the ELDA Landfill  in Cincinnati, Ohio.
Nine of the plots were constructed on 2H:1V slopes and five were constructed on 3H:1V
slopes.   Each  plot  is  about 9 m wide by 20  or  29 m long and is  covered  by
approximately 0.9 m  of cover soil.  Instrumentation was placed in each test plot (with a
few  exceptions)  in  order  to  monitor the  moisture  content of  the  subsoil  and
displacements of the  GCL.   An  additional  plot  consisting only of cover  soil  was
constructed on the 2H:1V slope.  This plot did not contain geosynthetic materials and
was  used as  a control plot to study the effect of erosion on the cover soil on a plot that
did not contain any synthetic erosion control material.

Slope angles of 2H:1V and 3H:1V were selected to test the shear strength limits of the
GCLs.  The rationale for selecting these slope inclinations was as follows. Many landfill
final  covers  have slopes of approximately 3H:1V.  If GCLs are to be widely used in
landfill covers, they will have to be stable at a slope angle of 3H:1V.  Thus, the 3H:1V
slope was selected to  be representative of  a typical landfill cover.  However, it is not
sufficient to demonstrate that GCLs are stable on 3H:1V slopes — it must be shown that
they are stable with an  adequate  factor  of  safety.    Many  regulators  and  design
engineers require that permanent  slopes have a  minimum factor of safety  for static
loading of 1.5.
For an infinite slope  in a cohesionless  material, with no seepage, the factor  of safety
(FS) is:

      FS = tan(4)) / tan(p)                                                   (D.1)

                                      D-1
-------
where (j> is the friction angle and p is the slope angle.  If a GCL remains stable on the
2H:1V slope,  the friction angle of the  GCL (assuming  zero cohesion) must be at least
26.6°, and for this friction angle, the factor of safety on a 3H:1V slope must be at least
1.5.  Thus, the logic was to try to demonstrate a minimum factor of safety of 1.5 on
3H:1V slopes, and in order to do this,  it was necessary to test the GCLs on 2H:1V
slopes.    It was recognized that constructing a 2H:1V slope was  pushing the test to
(and possibly beyond)  the limits of stability, not necessarily of the mid-plane of the
GCLs but certainly at various interfaces within the system.

D-2.1 Expectations at the Beginning of the Project
During the conception and design of the  field test plots, there were several expectations
concerning the performance of the GCLs. First, it was assumed that if the GCLs were
placed with the bentonite in contact with the subgrade soils that the bentonite would
hydrate by absorbing water from the adjacent soils. However, it was also assumed  that
if a  geomembrane  (GM) separated the bentonite component  of  the  GCL from  the
underlying subsoil,  and a GM was placed over the bentonite to encase the bentonite
between two  GMs, that the bentonite  would be isolated from adjacent soils (except at
edges) and would not hydrate.

A key expectation was that none of the GCLs would fail internally on any of the field  test
plots. This expectation was based on the results of mid-plane laboratory shearing  test
on fully-hydrated GCLs.    Interface  shear slides were viewed as possible,  but  the
greatest  concern  was  with   the  GCL/subsoil  interface.    It was   predicted   that
displacements of the GCLs would  be downslope with the largest displacements on the
2H:1V slopes.  Creep of the GCLs was considered possible.   Differential (shear)
displacements were expected to be nominal.

D-2.2 Layout of the Test Plots
Fourteen test plots containing  a GCL as a component  were  constructed.  The layout of
the plots is shown in Figure D-1. Each plot was assigned a letter. Five plots (plots A-E)
were constructed on a 3H:1V slope, and nine plots (plots F to L, N, and P) were built on
a 2H:1V slope.  An additional  plot, plot M, which  consisted of only cover soil and no
geosynthetics, was  an  erosion control  plot that  was installed  on a 2H:1V slope to
document the degree of erosion that would occur if no synthetic erosion control material
was  placed over the cover soil. In all other plots, a synthetic erosion control material
covered the surface of the test plot.   Plots on the 2H:1V slope  were about 20 m long
and 9 m wide; plots on the 3H:1V slope were about 29 m long and 9  m wide.

D-2.3 Plot Compositions
Four different types of GCLs were placed at the site: Gundseal, Bentomat ST, Claymax
500SP, and Bentofix. Two styles of Bentofix were employed:  Bentofix NW contained

                                      D-2
-------
                                   Geomembrane Placed over GCL
                    Crest of Slope

     3H:1V Plots:

                     Toe of Slope —»•


A
t
B

B

c

D

E
entomat f Bentofix f
ST I NS I
                                  Gundseal      Claymax      Gundseal
                                  (Bentonite       500SP       (Bentonite
                                     Up)                       Down)
                                                             Note:

                                                             All Test plots Were Nominally
                                                             9 m Wide and either 20 m Long
                                                             (21-1:1 V Slopes) or 29m Long
                                                             (31-1:1 V Slopes)
                               Geomembrane Placed
                                    over GCL
                                          No Geomembrane
                                                 Geomembrane
                                                  Placed over
                                                     GCL
D
CO
Crest of Slope
     2H:1V Plots:
             Toe of Slope —»•
G,
P

H

1

J

K

L

M

N
                          Gundseal  A   Claymax   A   Bentomat
(Bentonite
   Up)
                                        500SP
                                                       ST
                                              Bentofix
                                                NW
                                                    Bentofix
                                                      NW
Bentofix
  NS
                                                           No GCL
                                                           (Erosion
                             Bentomat ST (Plot G);
                             Gundseal (Plot P) with
                              Bentonite Up
                                              Claymax
                                               500SP
            Figure D-1.  Layout of field test plots.
-------
nonwoven geotextiles (GTs) on both surfaces.  Bentofix NS contained a woven GT on
the side that faced downward and a nonwoven GT on the side that faced upward.

Two  general  designs  were employed.   The  principal design involved  a subgrade
overlain by a  GCL, textured GM, geotextile/geonet/geotextile drainage composite, and
0.9 m of cover soil, as shown in Figure D-2.  This cross section is typical of many final
cover systems for landfills  being designed today.  The  GTs were heat-bonded to the
geonet (GN).  A nonwoven, needlepunched GT was used between the textured GM and
GN in an effort to develop  a high coefficient of friction between the  GM and drainage
layer.
                         Erosion
                         Control
                        Cover Soil
                        GT/GN/GT
                            GM
                           GCL

                         Subsoil •
Figure D-2.    Typical test  plot cross section employing a composite textured
              HOPE GM/GCL liner system.
The second design involves a GCL overlain by 0.3 m of drainage soil, a GT, and 0.6 m
of cover soil, as shown in Figure D-3.  This design is  also typical of current GCL designs
for final cover systems in which a GM  is not used.

The GM-supported GCL was employed  in two configurations, i.e., with the bentonite
encased between two GMs as shown in Figure  D-4A, or with the  bentonite in contact
with the subgrade, as shown in  Figure D-4B.  In the former case, the  bentonite was
designed to stay  dry.  In  the latter  case, it was expected that the bentonite would
hydrate by absorbing moisture from the subgrade.
                                     D-4
-------
                          Erosion
                          Control
                        Cover Soil
                        Geotextile
                     Drainage Sand
                             GCL

                          Subsoil
Figure D-3.    Alternative test plot cross section employing a GCL with no GM.
Geotextile-encased, needlepunched GCLs consisted of materials that either had woven
and nonwoven GTs on the surfaces, or two nonwoven GTs.  For the GCLs containing a
woven GT on one  surface, the woven  GT faced upward in some cases (Figure D-5A)
and downward in other cases (Figure D-5B)

Plot M is an erosion control section and consisted only of 0.9  m of cover soil.  There
were no geosynthetic materials or instrumentation at the erosion control  section.  The
plot was constructed to document the erosion that would occur without any geosynthetic
erosion control material on the surface.

General cross sections are shown in  Figure D-6 for plots constructed on the 2H:1V
slope and in Figure D-7 for plots constructed on the 3H:1V slope. A cross section in the
perpendicular direction is  shown in Figure D-8.  Each plot width was equal to two GCL
panels minus a 150-mm-wide overlap.  The spaces  between plots on the 2H:1V slope
ranged between 0 and 1.5 m,  and were typically 1.5  m on the 3H:1V slope. There were
graded drainage swales  only  on  the 3H:1V slopes.  Table D-1 lists the  slope angles,
plot, type of GCL, and a description of the plot cross-section from top to bottom. Table
D-2 lists the composition, dimensions, etc., of each plot.

D-2.4 Anchor Trenches
Anchor trenches were constructed at the crest of each test plot.   On the 3H:1V and
2H:1V  slopes all of the geosynthetic materials (GCL, GM, and GN, if present) were
brought into the anchor trench.  A GM cap strip was placed over the GCL in the anchor
trench with  the purpose of preventing moisture from  entering the GCL from the crest of
the plot. A typical anchor trench detail is shown in Figure D-9.

                                      D-5
-------
Soil in the anchor trench was nominally compacted. The anchor trench was used only
for  the  purpose of holding the geosynthetics  in position  during construction.  As
discussed later, the geosynthetics above the mid-plane of the GCLs were cut next to the
anchor trenches so that the shear force from the cover would be transmitted to the
internal  structure of the GCL (and  not simply carried by tension in the geosynthetics
overlying the mid-plane and anchored in the anchor trench).

D-2.5 Toe Detail
At the toe of the slope the GM and  GN were extended beyond the GCL in the plots on
the  3H:1V and 2H:1V slopes.  The  extension is shown in Figures D-6 and  D-7 for test
plots employing a GN for the drainage  layer.  Both the GM and the GN  were  extended
(daylighted) approximately 1.5 m past the end of the cover soil. For test plots in which a
sand drainage layer was used, a GN was extended beyond the sand drainage material
as shown in Figure D-10. The toe was designed to provide no buttressing effect for the
cover soil.

D-2.6 Instrumentation
The objectives of the instrumentation for the field test plots were to monitor the wetting
of the subsoil and the bentonite in the GCLs, and to monitor displacements of the GCLs.
Moisture sensors were  installed to verify that the  bentonite was hydrated, or in the case
of plots A, F,  and P, to  verify that the bentonite was dry. Extensometers were installed
to document  the internal shear and creep of the GCLs in each plot.  As there was  a
limited budget,  the  instrumentation was selected  based on simplicity, low cost, and
redundancy.

      D-2.6.1 Moisture Sensors.  Moisture sensors were installed in each test plot in
order to assess the  moisture conditions impacting the bentonite within the GCLs.  Two
types of sensors were used in  the project: a gypsum block sensor  and a fiberglass
mesh sensor (Figure D-16). The  gypsum block sensors were placed in  the subsoil
beneath the GCLs; the fiberglass sensors were placed within the  bentonite of the GCLs.
Both sensors operate on a resistance basis.  The fiberglass sensors contain a porous
fiberglass mesh embedded in two  wire screens .  The  resistance to flow of electric
current between the two screens is  dependent on the moisture present in the fiberglass
mesh. The resistance  is measured and converted to moisture content  by comparison
with a calibration chart.  The calibration is a function of soil type and the constituents of
the  soil moisture.   The gypsum block sensors have  two concentric  spirals of wire
between which resistance of gypsum is determined.  The electrical resistance of the
gypsum  is a function  of  the moisture content of the gypsum.   The resistance is
measured using a digital  meter manufactured  specifically  to measure resistance for
these sensors.

                                      D-6
-------
 (A) Plots with Bentonite Component Facing Upward
                   (B) Plots with Bentonite Component Facing Downward
      Cover Soil
   Geocomposite
   Drainage Layer

   Textured HOPE
   Geomembrane

GCL (Bentonite Up) {
        Subsoil
-Bentonite
                                    • Textured HOPE
                                    Geomembrane
                          Cover Soil
                       Geocomposite
                       Drainage Layer
                 GCL (Bentonite Down)
                                                                 Subsoil
Textured HOPE
Geomembrane
 Bentonite
   Figure D-4. Placement of Gundseal with bentonite facing upward or downward.
-------
(A)  Woven Geotextile Interfacing with Geomembrane
                    (B) Nonwoven Geotextile Interfacing with Geomembrane
       Cover Soil
    Geocomposite
    Drainage Layer

    Textured HOPE
    Geomembrane
GCL (BentonitSOp.) {
         Subsoil
Woven
Geotextile
  Bentonite
                                     ^ Nonwoven
                                       Geotextile
                            Cover Soil
Geocomposite
Drainage Layer


        GCL
                               Subsoil
Nonwoven
Geotextile
                                                                                                     • Bentonite
                                                              Woven
                                                              Geotextile
   Figure D-5. Orientation of GCL with either woven or nonwoven GT facing upward.
-------
Drainage Layer
(GT/GN/GT or Granular)
     Figure D-6. General cross section of plots on a 2H:1V slope.
      Geocomposite
         Drainage
         Material
     Figure D-7. General cross section of plots on a 31-1:1 V slope.
                                         D-9
-------
               0.9m
                               GCL
                                   GM
                                       Drainage Layer:
                                         GT/GN/GT
                                         or Granular
Erosion
Control
 Mat
                          Total VUdth of Test Plot: 10 to 13 m
Figure D-8. Cross section along width of test plots.
              Geomembrane
                Cap Strip
           Geocomposite-
           Drainage Layer
                                    All Geosynthetics above the
                                    Mid-Plane of the GCL Were Cut,
                                    Including the Upper Geotextile or
                                    Geomembrane Component of the
                                    GCL (If Present)
                      Anchor
                      Trench
                      Backfill
                                      Geomembrane
Figure D-9. Typical anchor trench detail.
                                       D-10
-------
Table D-1. Components of the GCL Field Test Plots.
Plot
A
B
C
D
E
F
G
H
I
J
K
L
M
N
P
GCL
Gundseal
Bentomat ST
Claymax 500SP
Bentofix NS
Gundseal
Gundseal
Bentomat ST
Claymax 500SP
Bentofix NW
Bentomat ST
Claymax 500SP
Bentofix NW
Erosion
Control
Bentofix NS
Gundseal
Target
Slope
(deg.)
18.4
18.4
18.4
18.4
18.4
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
26.6
Actual
Slope
(deg.)
16.9
17.8
17.6
17.5
17.7
23.6
23.5
24.7
24.8
24.8
25.5
24.9
23.5
22.9
24.7
Cross-section
(from top to bottom)
Soil/GN/GM/GCL (Bent, up)
Soil/GN/GM/GCL(Wup)
Soil/GN/GM/GCL (W-W)
Soil/GN/GM/GCL (NW up)
Soil/GN/GCL (Bent, down)
Soil/GN/GM/GCL (Bent, up)
Soil/GN/GM/GCL (W up)
Soil/GN/GM/GCL (W-W)
Soil/GN/GM/GCL (NW-NW)
Soil/GT/Sand/GCL (W up)
Soil/GT/Sand/GCL (W-W)
Soil/GT/Sand/GCL (NW-NW)
Soil
Soil/GN/GM/GCL (NW up)
Soil/GN/GM/GCL (Bent, up)
where:
      Soil = cover soil
      GN = geonet
      GM = textured GM
      GT = geotextile
      GCL = geosynthetic clay liner
      Bent, up = bentonite side of Gundseal facing upward (GM against subgrade)
      Bent, down =  bentonite side of Gundseal against subgrade
      W up = woven GT of GCL up, nonwoven side of GCL against subgrade
      NW up = nonwoven GT of GCL up, woven side of GCL against subgrade
      NW-NW = both sides of GCL nonwoven
      Bentofix I is Bentofix NW, with a nonwoven GT on both sides
      Bentofix II is Bentofix NS, with a woven GT facing upward.
                                   D-11
-------
Table D-2. Summary of Test Plots.
Plot
A
B
C
D
E
F
G
H
1
J
K
L
M
N
P
GCL
Type
Gundseal
Bentomat
Claymax
Bentofix NS
Gundseal
Gundseal
Bentomat
Claymax
Bentofix NW
Bentomat
Claymax
Bentofix NW
Erosion
Control
Bentofix NS
Gundseal
GM
(Y/N)
Y
Y
Y
Y
Y
Y
Y
Y
Y
N
N
N
N
N
Y
Drain
Type
GN
GN
GN
GN
GN
GN
GN
GN
GN
Sand
Sand
Sand
Sand
Sand
GN
Slope
3H:1V
3H:1V
3H:1V
3H:1V
3H:1V
2H:1V
2H:1V
2H:1V
2H:1V
2H:1V
2H:1V
2H:1V
2H:1V
2H:1V
2H:1V
Slope
Length
(m)
28.9
28.9
28.9
28.9
28.9
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
20.5
Crest
Elev.
(m)
179.2
179.2
179.2
179.2
179.2
157.9
157.9
157.9
157.9
157.9
157.9
157.9
157.9
157.9
157.9
Toe
Elev.
(m)
170.0
170.0
170.0
170.0
170.0
148.7
148.7
148.7
148.7
148.7
148.7
148.7
148.7
148.7
148.3
Test
Plot
Width
(m)
10.5
9.0
8.1
9.1
10.5
10.5
9.0
8.1
9.1
9.0
8.1
9.1
7.6
9.1
10.5
Notes:
     1.
     2.
     3.
     4.
Bentofix NW contained a nonwoven GT on both sides.
Bentofix NS was installed with the nonwoven GT facing upward.
Bentomat ST was installed with the woven GT facing upward.
Gundseal was installed with the bentonite facing upward in plots A, F, and P,
and with the  bentonite facing downward at plot E.
                                     D-12
-------
        GCL
Figure D-10.    Detail of drainage at toe for sections with GN drainage layer (not
              to scale).
           Gypsum Block
Fiberglass Moisture Sensor
                       40mm     40mm
                                         I!
              25 mm
   25mm
Figure D-11. Schematic diagrams of moisture sensors.
                                 D-13
-------
The  sensors were placed on the centerline of one of the  two GCL panels at  three
locations - top  , middle,  and bottom - of each plot as shown  in Figure D-12.   The
sensors were installed 5.2 m, 10.7 m, and 16.8 m from the crest on the 2H:1V slope and
6.1 m, 15.2 m, and 24.4 m from the crest on the 3H:1V slope. At each location two, and
in some cases three, moisture sensors were placed in the subsoil, at the subsoil-GCL
interface, and in a few instances, above the GCL.  The purpose of the sensors was to
monitor the moisture content of the  bentonite and soil adjacent to the  bentonite.
Because most plots contained a GM above the GCL, placing sensors in the cover soil
would  not provide information  on  moisture conditions within  or  near the GCLs.
Therefore, moisture sensors were generally placed  adjacent to or beneath the GCLs.  A
cross section of the typical moisture sensor installation  in all  plots except for plots A, F,
and  P, is  shown in Figure D-13.   Figure D-14 shows how the moisture sensors were
installed in plots A and F.

The  moisture sensors in  Plot P  were  installed differently than the other plots.  Only
fiberglass moisture sensors were installed in  Plot  P.   Sixteen moisture sensors were
placed in a 4 x 4 grid on the upper side of the bentonite of the GCL but underneath the
overlying GM.

The  gypsum blocks  and  digital  meter were obtained from Soil Moisture  Equipment
Corporation  of  Santa  Barbara,  CA.   The  fiberglass  sensors were  obtained  from
Techsas, Inc. of Houston,  TX.

1B
•
2m
3B

4m
•
5m
CREST
I •
I
I •
I
MIDDLE
I
I u
I
TOE •
                                           $ Cluster of
                                             Moisture Sensors
                                           | Extensometer
                     Left GCL Panel   Rig ht GCL Panel
Figure D-12. Locations of moisture sensors and extensiometers.

                                      D-14
-------
                                   Cover Soil
           GT/GN/GT -
                GM -

                GCL-
                Gypsum
                Block
                                            Subsoil
-Fiberglass Sensor

-Fiberglass Sensor
Figure D-13. Location of moisture sensors in all plots except A and F.
        GT/GN/GT
              GM

             GCL-
               Gypsum_
               Block
                                                Subsoil
  Fiberglass Sensor
Figure D-14. Location of moisture sensors in plots A, F, and P.
As mentioned above, the electrical resistance of a moisture sensor is measured and
converted to  moisture content by comparison with a calibration chart.  The moisture
sensor readout device used on this project reads from 0 to 100, with 0 corresponding to
no soil moisture and 100 corresponding to a very wet soil.  However, the calibration is a
                                      D-15
-------
function of soil type. There are generally four different soil types at the site.  Three soils
are distributed generally as shown in Figure D-15 for the 2H:1V test plots.  Soil A is a
gray fat clay, soil B is a clayey silt, and soil C is a silty clay (field classifications).  The
subsoil on the 3H:1V slope is primarily a clayey silt (soil D).

Calibration  tests were  performed for both the  gypsum block and fiberglass moisture
sensors for soils A, B,  C, and D.  A 1000 ml beaker was filled with soil, and a circular
piece of Gundseal was placed above the soil with the bentonite portion of the GCL in
contact with the soil.  A small layer of sand was placed over the GCL and a pressure of
18 kPa was applied to  the specimen. A gypsum  block was inserted within the subsoil,
and  a fiberglass moisture sensor was  placed at the interface of the GCL and the
subsoil.  The subsoil was incrementally  wetted, and after the moisture gauge reading
had  equilibrated,  the resistance  reading was recorded and a sample of the soil was
obtained for measurement of water content.  A typical calibration curve for the gypsum
block in the subsoil and the fiberglass moisture gauge at the soil/GCL interface is shown
for Soil A in Figure D-16.
                  Soil A: Gray Fat Clay
                  Soil B: Clayey Silt
                  Soil C: Silty Clay
           Crest
            Toe
B





B









B





B









B





B
x








B




X
X
/C









B

X
X
X

C









B"
X




C
X
'







C




'
."A









C

>
X
X


A









px
s




A

            PLOT:  FGHIJKLMN
Figure D-15. Soil types at 2H:1V test plots.

The calibration of the fiberglass moisture sensor with bentonite was performed as
follows.  A fiberglass sensor was sandwiched between two prewetted pieces of
Gundseal so that the  sensor was surrounded by bentonite. Sand was placed below and
above the GCLs, and a pressure of 18 kPa was applied.  After the moisture gauge
reading  had stabilized, the moisture gauge reading was recorded.
                                      D-16
-------
          O)
          CO
          (U
          b
          (U
          CO
          (U
100

 80-

 60+

 40

 20
OA
  •
                               '
  0
   0
—i	1	h-
 10      15      20

   Water Content (%
                                                          25
          30
Figure D-16.  Calibration of gypsum block moisture sensors (typical calibration).
The  calibration  curve  for the fiberglass  moisture sensor with bentonite is shown  in
Figure D-17. The scatter is due to the use of 15 different sensors in the development of
the calibration curve  (each  moisture sensor  should  ideally  have  its own individual
calibration curve).   This  calibration  curve  can  be used to qualitatively  distinguish
whether the  bentonite  is relatively dry or saturated.  Beyond that, however, statistical
scatter limits resolution. For example, a moisture gauge reading of 20 indicates that the
water content of the bentonite could range  between 40 and  150%, and for a  gauge
reading of 80 the water content of the bentonite could range between 190  and  290%.
However, a gauge reading of close to 0 clearly  indicates that the bentonite is dry, and a
reading close to 100 clearly indicates that  it is wet.

      D-2.6.2  Displacement Gauges.  Displacement gauges, or  extensometers, were
installed in each plot to measure displacements and to  assess shear strains  in the GCL
at multiple locations. Twenty displacement gauges were installed  in each plot (10 pairs
on each panel).  Five gauges in each panel were attached to the upper side of the GCL
Figure D-18.  With gauges on the upper  and lower side of the GCL, the difference in
total  displacement between the  upper  and  lower gauges provides  a measure  of
shearing displacement. Figure D-19 shows the  attachment of the hooks to the upper
                                     D-17
-------
              120
100    150    ZOO
  Water Content (%)
250
                                                            300
     Figure D-17. Calibration of fiberglass moisture sensors with bentonite.
and lower GTs of the GCLs.  Each extensometer consisted of a braided steel wire (for
flexibility) running from its point of attachment to above the crest of the slope.  The wire
was contained within a 6-mm OD (outside diameter) plastic tubing, and was connected
to a fishhook at the end of the wire (Figure D-19). The fishhook was attached by epoxy
to the surface of  the GT component of the GCL.   Gauges on the upper and  lower
surfaces  were used to measure  differential displacement, as  shown in Figure  D-20.
Each wire extended from the  fishhook to a monitoring station, or displacement table, at
the crest of the slope. A displacement table is shown  in Figure D-21.

D-2.7 Construction

Construction  of the plots  began  on November 15, 1994, and was  completed on
November 23, 1994. The construction sequence was as follows:
                                     D-18
-------
   1.   Subgrade preparation.
   2.   Installation of moisture sensors in  the  subgrade and  at  the  surface of the
       subgrade.
   3.   Placement of GCL.
   4.   Installation of the extensometers and displacement cables.
   5.   Installation of moisture gauges within the GCL (plots A, F, P).
   6.   Placement of GM (not applicable to plots J, K, L, and M).
   7.   Placement of GN composite or granular drainage layer (plots J, K, L, M).
   8.   Placement of GT (plots J, K, L, M only).
   9.   Placement of cover soil.
   10.  Construction of displacement tables.
      Crest
                                                          Toe
Figure D-18.  Locations of displacement sensors.
                                     D-19
-------

                               -mm OD Plastic Tubing
                                      3-mm Diameter Steel Cable
Figure D-19. Attachment of displacement monitoring hook to GCL.
                                 D-20
-------
           Geomembrane

         Bentonite
       Bentonite
        Extensometer
           Cable
                               O. 0)
                               ° E
      Upper Gauge

2 L = Differential Deformation
     Lower Gauge
                                                 Time
Figure D-20.  Location of displacement gauges to measure differential movement.
                                    D-21
-------
                                                   Displacement
                                                     Indicator
Figure D-21.  Displacement table at crest of slope.
D-2.8 Cutting of the Geosynthetics

With other geosynthetic materials besides the GCL leading into the anchor trench, part
of the down-slope component of force created by the cover soil is carried by tension in
these geosynthetic materials.  To concentrate all of the shear stress within the mid-
plane of the GCL, the geosynthetic materials above the mid-plane  of the GCL were
severed.  The geosynthetics above the mid-plane of the GCLs in plots A through D
(3H:1V slope) were cut on April 13, 1995, and the geosynthetics above the mid-plane of
the GCLs on the 2H:1V slopes and plot E (3H:1V slope) were cut on May 2, 1995.

In plots with GT-encased GCLs,  the  GN composite, GM, and the upper GT of the GCL
were cut at the crest of the slope down to the mid-plane of the GCL as shown in Figure
D-22. The geosynthetic materials in plots constructed with a granular drainage layer
were cut  down to the mid-plane of the GCL as shown  in Figure D-23.  The granular
drainage  material did not extend into the anchor trench, so the GT was cut as well as
the upper GT in the GCL.
                                     D-22
-------
                            Upper Geotextile
                               inGCLCut
         GT/GN/GT
Figure D-22.
Cross-section at crest of slope showing cutting of geosynthetics
down to mid-plane of GCL on test plots with a GM.
                        Upper Geotextile
                          in GCL Cut
       GCL-
                                                            Granular
                                                            Drainage
Figure D-23.
Cross-section at crest of slope showing cutting of geosynthetics
down to mid-plane of GCL on test plots without a GM.
                                   D-23
-------
The cutting of anchor trench materials in plots with Gundseal is shown in Figures D-24
and D-25.  In the case with the bentonite side of the GCL facing up (Figure D-24), the
GN and GM were cut leaving the entire GCL intact.  In the case with the bentonite side
of the GCL facing downward (Figure D-25), the GN and the GM of the GCL were cut.
                               Only GT/GN/GT
                                 & GM Cut
                                                            Topsoil
        Gundseal GCL
        (Bentonite Up)
Figure D-24. Cutting of slope with Gundseal, bentonite side facing upward.
                                GT/GN/GT & GM
                              Component of GCL
                                    Cut
             GT/GN/GT
          Gundseal GCL
         (Bentonite Down)
Figure D-25. Cutting of slope with Gundseal, bentonite side facing downward.
D-2.9 Supplemental Analyses of Subsoil Characteristics
In the summer of 1997, displacements developed in several test plots that appeared to
be  consistent with  the soil  patterns  depicted  in  Figure D-20.   The various soil
                                    D-24
-------
boundaries shown  in  Figure  D-15 were  determined in the field  at  the  time  of
construction, based on visual observations.  Cracks developed in several test  plots
parallel to the lines shown in Figure D-20.

In an attempt  to refine Figure D-15, additional samples  were obtained  between the
2H:1V test plots and analyzed for liquid and  plastic limits following American Society for
Testing Materials (ASTM) procedure D4318. Results are summarized in Figures  D-26
and D-27. There was some correlation between the originally mapped soil zones and
the results of liquid limit  (LL) and plasticity index (PI) determinations from the  1997
investigation,  particularly  in  terms  of the mapped  location  of the fat clay where  it
intersected with the other soils.

Several samples were also obtained from the 3H:1V test plots, and the results of LL and
PI determinations are shown in Figure D-28.   Little variability was noted. The soils  were
originally described as sandy, but it was clear that the sands were clayey because of
their plasticity.

D-2.10 Results of Water Absorption Tests
Tests were performed by GeoSyntec Consultants to evaluate  the probable hydration of
GCLs placed in contact with soils from the site.  Hydration  tests  were performed on
Claymax 500SP, Bentomat  ST, and Bentofix NW.   Dry GCLs were placed in sealed
containers and in contact with compacted soils from that site.  The soil used had a  LL of
41% and a PI of 19%.  The optimum  moisture content was 20% when  the soil was
compacted with standard Proctor (ASTM D698).

The soils were mixed to predetermined water contents equal to 16%, 20%, and 24%.
The soils were then compacted into  75-mm-diameter  molds to a depth of 150 mm.  The
GCL  specimens were  placed  in  contact with the soil and  subjected to a  nominal
compressive stress  of 10 kPa. The nonwoven GT component  of Bentomat  ST was
placed against the soil.  The apparatus was sealed, and GCL samples were removed at
various  times  for a  period  approaching three  months.  At the end of 75  days  of
hydration, the water contents were as shown in Table D-3.

As expected, the GCLs did  hydrate, and the data indicate that the bentonite was still
absorbing water from the wettest soils when the tests were discontinued.  The  soils at
the site  were  substantially  wetter than 24%  at the time  of  construction,   and  at
subsequent times.   For example, samples  of subsoils taken in June, 1997, from the
21-1:1 V test plots showed that the water content of the five soil samples averaged 40%.
                                     D-25
-------
                  Soil B (Clayey Silt)
Soil C (Silty Clay)
•^ ir-^ p^
/-\ s\ /V t r*s\ 	 .-«. .-«. 	 .X





4







9
ro
en

X

xx
/






3 5

1






X








24


2

36 x'
x

x
x

X




43
T^ v







X
X













^D

30
21XX
xx
xx

27




45

30



23



x





K












x^
xx

26
24

25

29




46

32 xx
X
X
X
x
/47
X








L




x
XX







^D

31
40

70

7QXX
X
x
X
X
X

47



26




X
xx

M











jj

32 x.
XX
s
69

24




37

21



38
x'
X
X




N











xx


















fc.
                                                  Soil A (Gray Fat Clay)
Note: Locations of soil types and descriptions of soils determined at the time of construction
      of the test plots in 1994 - - LL values shown were measured in 1997. The
      Information is shown together for purposes of comparison.
  Figure D-26. LL(%) measured in 1997 at 2H:1V test plots.
-------
                                  Soil B (Clayey Silt)
Soil C (Silty Clay)
g
•^ 	 ir^ 	 ^r
12 MX 10





1!










/'
x





5 2


1






X
X




8


?


14 x
x
X
x/
X
X
X

18








X
/J











10

3 xxX
/'
s

11


20



14
9



xx
X




K









x'

' 7

11
9


15


20



9 x-
X3/
/*








L




/
/





14

19
41


3VX
X
X
xx



21
10







X
X
x
X
M









19


15
x'
X
>y
38


9


16



13
21

x
X
xx





N









X

















X
xX^ k.
                                                                      Soil A (Gray Fat Clay)
                       Note:  Locations of soil types and descriptions of soils determined at the time of construction
                             of the test plots in 1994 - - LL values shown were measured in 1997. The
                             Information is shown together for purposes of comparison.
                   Figure D-27. PI (%) measured in 1997 at 2H:1V test plots.
-------


A


27(13)
24(9)
23(8)
30(10)
28(12)
V


B









C









D


22(8)
32(16)
22(7)
26(11)
24(9)
t


E


                 Liquid Limit(Plasticity Index) - both expressed as a percentage
Figure D-28.    Results of LL and PI tests on samples taken from subsoils in 1997
               at the 3H:1V test plots.
Thus, confirmation was provided that GCLs placed against the subgrade soils at the site
would be expected to hydrated to water contents in excess of 70 to 90%.
Table D-3.      Water Contents of GCLs after 75 Days of Hydration when Placed
               Against Soils with Various Initial Water Content (w).
GCL
Bentofix NW
Bentomat ST
Claymax 500SP
Soil at w= 16%
37
40
38
Soil at w=20%
58
67
55
Soil at w=24%
76
94
70
                                     D-28
-------
D-3  Laboratory Shear Tests
Due to scheduling  constraints, it was necessary  to construct the test plots  before
laboratory direct shear tests could be performed.  Experienced gained from the test
plots in the first few  weeks after construction was complete indicated that the interfaces
between the GCLs and adjacent materials (particularly between the GCL and overlying
textured High Density Polyethylene (HOPE) GM) were the potential failure surfaces of
greatest concern. To assist in evaluating if the interfaces between  materials  at the test
site  would  be  stable,  Drexel University's  Geosynthetic  Research  Institute  (GRI)
performed shear tests on critical interfaces.

At the time the shear testing was performed, thirteen plots had been constructed (all of
the test plots discussed earlier except for plot P, which was constructed later.  Five plots
were constructed on a 3H:1V slope, and eight plots were constructed on a 2H:1V slope.
The shear testing focused on the plots and interfaces  installed on the 2H:1V slope
because these were the more critical slopes. The interfaces of concern were:

     1.  The interface between the top of the GCL and the overlying textured GM.
     2.  The interface between the top of the GCL and the overlying sand.
     3.  The interface between the bottom of  the  GCL  and the underlying subgrade
        (particularly the clayey subgrade soils identified as Soil A  and Soil C in Figure
        D-20).

D-3.1  Testing Method
The shear tests were performed according to ASTM D5321 in a 300 mm square shear
box.  The specimens were hydrated for 10 days in the  shear device under a  normal
stress  of 18 kPa. This stress is the approximate normal stress acting on the GCLs in
the 2H:1V test plots.  The specimens were sheared at a strain  rate of 1 mm/min.  For
each  test,  the  peak and  large-displacement strengths were reported.  Single point
failure  envelopes were created by fitting a straight line through the origin and the failure
point.

The interface between the top of the GCL and the textured GM was the main focus of
the program of laboratory  shear testing because of the two  interface slides at the test
site. To simulate the field conditions as best as possible, site-specific products were
used in the tests.  In  order to obtain the large-displacement strengths, the specimens
were sheared to displacements of 35 mm.

D-3.2 Results
The results of the direct shear tests are presented in Attachment 1 of this appendix. The
testing results are summarized in Table D-4.
                                      D-29
-------
Table D-4. Summary of Results of Interface Direct Shear Tests.
Test Plot
A,E,F,
&P
B&G
C&H
1
K

Type of
GCL
Gundseal
Bentomat
ST
Claymax
500SP
Bentofix
NW
Claymax
500SP
Bentofix
NS
GCL Interface
Dry Bentonite
Surface of GCL
Woven Slit-Film
GT
Woven Slit-Film
GT
Nonwoven
Needlepunched
GT
Woven Slit-Film
GT
Nonwoven
Needlepunched
GT
Opposing
Interface
Textured HOPE
GM
Textured HOPE
GM
Textured HOPE
GM
Textured HOPE
GM
Drainage Sand
Textured HOPE
GM
Peak Secant
Friction
Angle
(cleg.)
37
23
20
37
31
29
Large-
Displacement
Secant Friction
Angle (deg.)
35
21
20
24
31
22
Note.  Plots J and L (plots with drainage sand and no GM) were not specifically evaluated because a
relatively high friction angle (31°) was measured for plot K, which like plots J and L also had drainage
sand and no GM.  It was assumed that the friction angle between the drainage sand and either Bentomat
ST (plot J) or Bentofix NW (plot L) was no less than the 31° value measured for Claymax 500SP.
The tests performed on Gundseal and summarized in Attachment 1 and Table D-4 were
performed with interface shear  on  dry  bentonite. The testing was performed  in this
manner on the assumption that the bentonite encased between two GMs would remain
dry.  The measured angle of internal friction was 37° for peak failure conditions and 35°
for  large displacement,  for the shearing  rate of 1  mm/min.   Daniel et al.  (1993)
performed internal shear tests on  Gundseal  on  smaller  (60-mm-diameter) samples.
Results are summarized in Figure D-29.  For shearing rates of 0.26 mm/min and 0.0003
mm/min, the respective angles of internal friction for the most comparable normal stress
used were 41° and 35°.  Considering the differences in materials and testing conditions,
the published results compare  favorably  with the results  of  tests  on 300x300mm
samples conducted for this research project.

Shearing tests were not  performed on hydrated Gundseal for this study because the
shearing  properties  of  hydrated   Gundseal  have  been  studied  extensively  and
documented  by  Daniel  et  al. (1993) and  Shan  (1993).  The tests described in the
                                      D-30
-------
          CO
          Q_
          _*:
          V)
          O)
          CO
          (D
          _c
          C/)
125


100


 75


 50


 25


  0
  o  Slow Shear (0.0003 mm/min)
-  D  Fast Shear (0.26 mm/min)
                                  50               100

                                  Normal Stress (kPa)
                                                     150
Figure D-29. Results of direct shear tests on dry Gundseal (from Daniel et al.,
                1993).
literature were  all fully drained tests performed  using either 60-mm-diameter  test
specimens in a direct shear apparatus or a 450x450mm test specimen in  a tilt table
submerged in a water bath. The direct shear tests were performed at an extremely slow
speed (0.0003 mm/min), which yields a lower shear strength compared to faster shear.
The tilt table tests were performed by slowly increasing the angle of tilt over a period of
several  weeks.   The secant friction angle from the  direct shear and tilt table tests is
plotted  versus  normal  stress  in Figure  D-30.   For a  normal  stress of 18kPa  (the
estimated value at the test plots), the angle of internal friction for the hydrated Gundseal
is approximately 20°.

The  influence of water content on the  shear strength  of unreinforced bentonite in
Gundseal has also been evaluated by Daniel et al. (1993). Results are summarized in
Figure D-31 for a normal stress of approximately 27kPa. The water contents  plotted are
the average final water content determined at the end of shear.  The test  at a water
content  of 145%  was for  fully hydrated bentonite.   The angle of  internal friction  was
found to be comparatively high for the as-manufactured water  content,  but for water
contents > 50%,  the friction angle was essentially independent of water content  and
equal to the value for fully hydrated bentonite.
                                      D-31
-------
Figure D-30.
                 co   40
                 O)
                 0
  g

  "G
  -i_
  u_

  "co
  s_
  0
  Jl


  •5


  "5)
                     30
                     20
                     10
                      0
                               Tilt Table

                               Direct Shear
                                  10         100


                                Normal Stress (kPa)
                                         1000
Influence of normal stress on the internal shear strength of

hydrated bentonite (from Shan, 1993).
            CD

            2
            O)
            CD
            T3
            O

            "O
                25
 20
            ^   15
            O)

            <
                10
                                50            100


                            Water Content of Bentonite (%)
                                             150
Figure D-31.    Influence of water content on the internal shear strength of

               Gundseal (after Daniel et al., 1993).
                                    D-32
-------
Interface shear tests were performed on the interface between the textured HOPE GM
and Bentomat ST, Bentofix NW, and Claymax 500SP.  As shown in Table D-4, the peak
friction angles for the GCL/HDPE interfaces varied from a low of 20° for Claymax 500SP
(with a woven, slit-film GT interfacing with the GM), to a high of 37° for Bentofix NW,
which had a nonwoven GT interfacing with the GM.  Bentomat ST was sheared with the
woven GT interfacing with the GM, and the resulting peak friction angle was 23°.  Large
displacement  shear strengths were  0 to 2° below the peak friction angle for GCLs
having a woven  GT interfacing with the GM, but for the GCL with a  nonwoven GT
interfacing with the GM, there was a much larger (13°) difference between peak and
large-displacement friction angles, probably as  a  result of "polishing" of the  GT with
large displacement. When Bentofix NS was sheared with the woven  side of the GCL in
contact with a GM, the  peak and  large-displacement friction angles were 29° and 22°,
respectively. This configuration was not  actually used for any of the  test plots because
the nonwoven side of Bentofix NS  was installed in contact with the GM at plots D and N.
The appropriate GM/GCL interface friction angles to assume for plots D and N are those
from the results  of tests on Bentofix NW, which  was tested with the  nonwoven GT
component of Bentofix interfacing with the GM.

Plots J, K, and L were installed with sand as a drainage layer directly overlying the GCL.
Placed above the sand  was a nonwoven GT overlain by cover soil that was protected
with an erosion control geosynthetic.  A GM was not installed. The concern in plots J,
K, and L was the shear strength  of the  interfaces between the GT component of the
GCL and the overlying sand. A direct shear test was performed  on  this interface. The
resulting  peak and large-displacement interface  friction angles were both 31° for tests
on Claymax 500SP.  The other  GCL-sand interfaces were not tested because the
lowest friction was expected for Claymax 500SP.  The two  other GT-encased GCLs
should have higher interface friction angles with the drainage sand that Claymax 500SP,
which contains woven slit-film GTs on both surfaces.

D-4 Performance of Test Plots
Displacement and moisture data from the test plots have been collected once every two
to three weeks since installation.  The post-construction data have been arranged into
three  different types of  graphs in  order to characterize how the test plots are moving
and hydrating with time:  1)  total down-slope  displacement vs. time;  2) differential
displacement between upper and  lower surfaces of the GCLs vs.  time; and 3) moisture
gage readings vs. time.  These graphs are presented in Attachments 2, 3, and 4 of this
Appendix.

D-4.1  Construction Displacement
Construction displacements  are the down-slope displacements of the GCLs observed
during construction of the test plots (during placement of the overlying geosynthetics or
                                     D-33
-------
drainage soil,  and cover soil).   Post-construction displacements are displacements
recorded after construction of the test plots and the associated displacement tables.

Construction displacements were measured by monitoring how far the displacement
cables moved in relation to reference stakes placed at the crest of the slope. Maximum
construction displacements measured from displacement gages attached to the top of
the GCL (at the crest and toe of the slope) and at the bottom of the GCL (at the crest
and toe of the  slope) are  listed in Table D-5.  However, the displacements varied with
time, and the maximum construction displacement did not always correspond to the last
construction displacement, probably because of  limitations  in  the resolution of the
extensiometers. Figure D-32 shows  the maximum construction displacements for the
left and  right panel at the toe of each plot.  Nearly all of the construction displacements
occurred when soil was placed above the GCLs.
Table D-5.  Maximum Construction Displacements.
Plot
A
B
C
D
E
F
G
H
1
J
K
L
N
Maximum Construction
Displacements(mm)
Above GCL
Crest
0
0
13
0
13
64
25
89
25
38
64
51
*
Toe
13
25
51
25
25
140
64
254
76
140
216
203
*
Below GCL
Crest
25
0
13
0
13
25
38
89
38
44
51
64
*
Toe
51
25
64
64
25
127
38
178
114
114
178
114
*
            note:  * = data not recorded
In general, the construction displacements were greater on the 2H:1V slopes than on
the 3H:1V slopes and greater at the toe of the slope than at the crest.  Table D-6 lists
ranges of maximum construction displacements from all of the extensiometer readings
on the 2H:1V and 3H:1V slopes.  Construction displacements for  plots on both slopes
ranged from  0 to 89 mm at the crest and 13 to 254 mm at the toe. In some instances,
                                     D-34
-------
such as plot I, the maximum construction displacement below the GCL was greater than
the maximum displacement  above the GCL.  This  indicates the bottom of the GCL
moved more than the top of the GCL at least one time during construction. However, it
was expected that the top of the GCL would move more than the bottom of the GCL.  It
should not be possible for the bottom of the GCL to deform downslope more than the
               (/>
               c
               o
              O
II
-------
D-4.2 Post-Construction Displacement of 3H: 1V Slopes
Post-construction displacements (both total and differential) are summarized in Table D-
7.  All 3H:1V slopes have remained stable during the 3-1/2 years of observation. Total
down-slope displacements have been less than 50 mm, and differential displacements
have been less than 10 to 40 mm. There has been no visual evidence of movement or
surface cracking.

      D-4.2.1  Test Plot A (Bentonite Between Two GMs).  The bentonite component of
Gundseal was expected to remain dry because the bentonite was encased  between two
GMs.   The  measured  peak and large-displacement interface  secant friction angles
between  dry bentonite and textured HOPE were 37° and 35°, respectively.  Because the
slope angle was 16.9°, the slope should be stable so long as the bentonite remains dry.

Fiberglass moisture sensors in plot A have provided variable results: two of the three
moisture  sensors have indicated that the  bentonite is dry, but one sensor near the crest
of the slope  had indicated some hydration. Two borings were drilled by hand  near the
crest and toe of the test plot in March, 1995, and 100-mm-diameter samples of  the GCL
were removed. The water  contents of the bentonite in the GCL at the crest and toe
were 27% and 24%, respectively. These values are essentially the same as the values
at the time of installation, confirming that the bentonite had not hydrated.

      D-4.2.2  Test Plots B, C, and D  (GT-Encased GCLs). Test plots  B, C, and  D
contain GT-encased  GCLs.  The bentonite in the GCL was expected to hydrate by
absorbing moisture from subgrade soils.  Most of the fiberglass moisture sensors have
indicated that  the  bentonite has hydrated, although less than expected.   One factor
inhibiting hydration may have been the relatively dry, sandy subsoils on the 3H:1V test
plots, compared to the 2H:1V test plots,  which  had more clayey, wetter subsoils. The
test plots have remained  stable because the slope angles are 17.5° to 17.8°,  i.e., less
than the interfacial friction angles.

      D-4.2.3  Test Plot  E  (Unreinforced GCL). Test plot E was constructed  with the
bentonite portion of Gundseal facing downward. The drained angle of internal friction
for fully hydrated bentonite is about 20°.  The slope angle at plot E was 17.7°; thus, the
test plot  is expected to  be  stable if the bentonite is hydrated,  but only with  F = tan
(20°)/tan  (17.7°) = 1.14 for an infinite slope (and possibly less, since the interface shear
strength with the underside of the GM may be slightly lower than the internal strength of
the bentonite).

As with most of the other test plots, the fiberglass moisture sensors for test plot E have
yielded variable results,  with some sensors indicating that the bentonite  has  become
hydrated  and others indicating that it has not  become hydrated. A boring was drilled

                                      D-36
-------
and a sample was taken from near the crest of the slope (the driest area) in March,
1995, and the water content of the bentonite was found to be 46%.   Eight more borings
were  drilled in April, 1996, at various locations along the full length of the slope.  The
water content varied between 54% and 79%, and averaged 60%.  The  average water
content of 60%  in test  plot  E  should be  sufficiently  large  to replicate the  strength
reduction associated with full hydration of the bentonite.

Table D-7. Summary of Post-Construction Performance of Field Test Plots.
Plot
A
B
C
D
E
F
G
H
I
J
K
L
N
P
Slope
3H:1V
3H:1V
3H:1V
3H:1V
3H:1V
2H:1V
2H:1V
2H:1V
2H:1V
2H:1V
2H:1V
2H:1V
2H:1V
2H:1V
Type of GCL
Gundseal
Bentomat ST
ClaymaxSOOSP
Bentofix NS
Gundseal
Gundseal
Bentomat ST
Claymax 500 SP
Bentofix NW
Bentomat ST
ClaymaxSOOSP
Bentofix NW
Bentofix NS
Gundseal
Stability of Test Plot As of
February, 1998
Stable
Stable
Stable
Stable
Stable
Internal Slide within the GCL
Occurred 495 Days after
Construction of Test Plot
Interface Slide between Lower
Side of GM and Upper Woven GT
Surface of GCL Occurred 20 days
after Construction of Test Plot
Interface Slide between Lower
Side of GM and Upper Woven GT
Surface of GCL Occurred 50 days
after Construction of Test Plot
Slumps and Surface Cracks
Developed about 900 Days after
Construction of Test Plot
Slumps and Surface Cracks
Developed about 900 Days after
Construction of Test Plot
Slumps and Surface Cracks
Developed about 900 Days after
Construction of Test Plot
Slumps and Surface Cracks
Developed about 900 Days after
Construction of Test Plot
Stable
Stable
Total
Displacement
(mm)
20
30
25
50
30



500
800
1200
500
30
NA
Differential
Displacement
(mm)
10
40
30
25
30
750
25
130
25
75
900
180
10
NA
Note: Total displacement is the total amount of down-slope movement measured after construction was
complete and displacement tables were installed; differential displacement is the difference between
down-slope movement of the upper and lower surfaces of the GCL that occurred after construction.
                                       D-37
-------
D-4.3 Post-Construction Performance of 2H:1V Plots
Slides have occurred at most of the 2H:1V test plots.  Two slides occurred at plots G
and  H  a few weeks after construction was complete.  Both involved  slippage at  the
interface between the  upper surface of the GCL (a woven GT in both  cases) and  the
lower surface of the textured HOPE GM. The next slide occurred in plot F about a year
and a half after construction.  In this case, the bentonite (which was encased between
two GMs) in this GCL unexpectedly became hydrated, and a slide resulted.  The test
plots remained stable  for the next two years, but then several slides  occurred in  the
subsoils  beneath  other test  plots.  The subsoils were plastic clays,  and the subsoil
slides (which occurred at the end of a wet spring season) were presumed to  be  the
result of hydration of the subsoil  clays and possibly the buildup of excess pore water
pressure in the subsoils, as well.

      D-4.3.1  Test Plots F and  P (Bentonite Encased  Between Two  GMs).  Plots F
and  P, like plot A, contained bentonite encased between two GMs.  The bentonite in
these test plots was expected to remain dry.  However, within three months after plot F
was  constructed,  two  of the three moisture sensors indicated that the bentonite had
become hydrated.

To evaluate the condition of the bentonite, 17 borings were drilled into Plot F in March,
1995, and 100-mm-diameter samples of the GCL were  recovered.  The water content of
the bentonite samples varied from 10% to 188%, and the data showed  that the right
panel was much more  hydrated than the left panel (Figure D-33).  In contrast to this field
data, Estornell and Daniel (1992)  reported laboratory test results for Gundseal in which
water migrated laterally through the GM-encased bentonite less than 100 mm  over a
test duration of 6 months.

Water may have  entered the bentonite at plot F through  cuts made in  the GM liner
overlying the GCL to allow insertion of the  extensiometer cables. Plot F was located at
a point where surface water at the crest of the slope was funneled directly to the anchor
trench area where the penetrations were made.  The mechanism for lateral movement
of water is probably waves in the overlying GM,  which would allow water to spread.
Alternatively, the source of water could have come from the V-shaped trough between
plots F and G, and spread through waves in the GM.  Unfortunately, the plot slid before
a complete forensic study could be performed.  Large-scale laboratory tests, described
later in this appendix, were performed to investigate the potential for lateral spreading of
water in the plane of the GCL.

Displacement sensors showed large movements in the right panel of plot F during  the
first year of observation, but not in the left panel until later.  Starting on about day 275
(August,  1995), the left panel began to move down slope, suggesting that the bentonite

                                     D-38
-------
in the left panel was finally becoming hydrated over a significant percentage of the total
area of the panel.
                                    Top of Slope

Water .,
Content —
(%)





Initial Water
Content of
Bentonite « 25%


^


421

• 1
27

50 •

70B


*- 	



34


«.


76 I
10
^
^^-— »
32
|
34

u
86
^9

Ii4
33
7
V
:::::^_
1
28



188
65-125

^7



	 Deformation
Sensor
Cables






                                     Toe of Slope

Figure D-33. Water content (%) measured in bentonite at plot F.

Plot F slid on March 24,  1996, 495 days after construction.  The cause of the slide is
hydration of the  bentonite; the peak angle of internal friction for hydrated bentonite at
the normal stress  existing  in the field was  20°, but the slope angle was 23.6°.  In
contrast, the peak  interface friction angle for dry bentonite was 37°.  Had the bentonite
not hydrated, the slope should have remained stable.

In response to the unexpected hydration, plot P was constructed on June 15,  1995.
The extensiometers were not installed  in plot P to eliminate all penetrations  in the
overlying GM.   The  number of  fiberglass  moisture sensors in the bentonite was
increased  from  3  in the  other  test  plots  to  16 in plot  P  to  provide additional
documentation of moisture conditions.   All but one of the 16 moisture sensors have
indicated that the bentonite has remained dry  in the 18 months of monitoring plot P.

      D-4.3.2  Test Plots G  and H.  Test plots G and H consisted of Bentomat ST and
Claymax 500SP, respectively.    Both plots slid at the interface between the upper GT
(a woven, slit-film  GT in  both cases) and the lower surface of the  overlying textured
HOPE  GM.   Plot  H slid 20 days after construction, and  plot G  slid 50 days after
construction.  Pre-slide displacements were  small (< 25 to 130 mm). There was no
                                      D-39
-------
warning of either slide.  Both  slides  occurred  at  night,  and the slides apparently
occurred quickly.

Test Plot H, which incorporated Claymax 500SP, was constructed on a 24.7° slope, but
the measured  peak and large-displacement interface  friction  angles  for the  relevant
materials  under hydrated  conditions  were only 20°.  Test plot H did not  slide
immediately  because the interfacial shear strength  of the  dry GCL was sufficient to
maintain a stable slope.  The slope slid when the bentonite hydrated.  Tests described
earlier showed that bentonite in the GCLs hydrated in a period of 10 to 20 days when
placed in contact with the subgrade soils from the test  plots. Tests reported by Daniel
et al.  (1993) showed similar results for other soils.   Thus, the sliding  time of  20  days
after construction is consistent with the  expected period to achieve nearly full hydration
of the bentonite.

When GCLs containing  a woven GT component become  hydrated, bentonite can swell
through the openings of the GT and lubricate the GCL-GM interface. After the slide, the
surface  of the GCL was very slick.  The tendency of bentonite to lubricate the GM/GCL
interface may be related to the thinness of the woven slit-film GT and  to differences in
apparent opening size between woven and nonwoven GTs.

Test plot G, which was constructed using Bentomat ST, was  slower  to slide, but the
slope angle  (23.5°) was 1.2° flatter than for plot H, and the interface shear  strength
between the GCL and overlying GM (23° peak, 21° large-displacement) was  1° to 3°
higher.  Also, a nonwoven GT faced downward in plot G, but a woven slit-film GT faced
downward in Plot  H.  GCLs are  expected to absorb  water more slowly from subgrade
soils when the GT separating the bentonite from the subsoil  is  a thicker nonwoven GT.
Thus, the reason  why  plot G slid 30 days  later than plot  H  appears to be  that the
bentonite in  the GCL at plot G was separated from wet subgrade soils by a thicker,
nonwoven GT, which slowed hydration.

      D-4.3.3  Plots I and N with Nonwoven GT Component Facing Upward.  Plots I
and N are similar to plot G, except that the GCL contained either one nonwoven GT with
the nonwoven  GT facing upward (plot  N) or two nonwoven GTs (plot I).  The slope
angles at plots I and N  were similar to the other 2H:1V plots.  However, the interface
friction angle between the nonwoven GT component of  Bentofix and the textured HOPE
(37° peak and 24° large-displacement) was much greater than for the woven slit-film GT
component of the  GCLs that slid. The geosynthetic components of plots I and N  have
remained stable because of the better interface shear resistance between a nonwoven
GT component of  a GCL compared to a woven GT component. The greater interface
shear resistance from the  nonwoven GT is attributed to: (1)  larger shear resistance
developed between nonwoven GTs and textured GMs in general, and (2) less hydrated
bentonite extrusion to the interface for the thicker nonwoven GT.

                                     D-40
-------
Large displacements began to develop in plot I and the adjacent test plots (J, K, and L)
about 3-1/2 years (900 days) after construction.  Several small slumps with downward
displacement of up  to about 100 mm along scarps,  and associated surface cracking,
were observed, with the slumps and cracks appearing in the lower half of the test plot.
The subsoils  in the  area of the slides are CL and CH clays, with the LL and PI of the
subsoil  next  to plot I  averaging  45% and  18%, respectively.  The  displacements
occurred at  the end of the wet spring season in 1997.   Examination of plot I and
adjacent test plots, coupled with excavation into the  subsoils, showed that sliding was
occurring 0.5 to  1  m  beneath  the  GCL,  in the clay subsoil.  It is  assumed that the
buildup  of pore water pressure behind the test  plots  helped to trigger the slides in the
subsoils. There was no indication of movement within the GCL or an interface with the
GCL.  Plot N  showed no signs of slumping or cracking,  but plot N was at the end of the
2H:1V test plots and likely was at a location where excess pore water pressures were
not as likely to develop.

      D-4.3.4 Plots J, K, and L with  No GM.   These test plots were  constructed  by
placing  drainage sand directly above the GCL.  All three test plots remained stable for
about 900 days after construction, and then all three  underwent significant down-slope
displacement (0.5 to 1.2 m, as shown in Table D-7). All three exhibited slumping in the
lower half to two-thirds of the test plots.  Scarps could be observed at several locations
within each  test plot.   Observation of the  depth of slumping clearly  showed that
displacement was occurring nominally 0.2 to 1  m beneath the GCLs.  Excavation into
the subsoils  showed that a layer of plastic clay was located at about this same depth.
The  sliding  mechanism was related to the  subsoils  and  not  to the  GCLs  or GCL
interfaces. Buildup of  pore water pressure in the clays following the wet spring season
was assumed to be the triggering mechanism.

The peak secant interface friction angle between the sand drainage material and GCL
was  31° for  a woven-slit  film  component (Table  D-7) and,  although  not  measured,
presumably  more for  a nonwoven  component.  An  interface friction angle of 31° is
significantly greater  than the slope angle (~ 25°), which explains the stability of the test
plots up until the point of sliding in the subsoil.

D-4.4 Moisture Gage Readings
Graphs  of moisture gage readings vs. time (attachment) provide a general indication of
how  and where the bentonite and subsoil is  hydrated  in each test plot.   Each graph
includes the  readings of the moisture sensors installed at one location along the slope
of the plot. The moisture sensors include the gypsum  block in the subsoil, the fiberglass
sensor at the interface of the GCL and the subsoil, and the fiberglass sensor  between
the GCL and the overlying GM (only in plots A, F, N, and P).
                                     D-41
-------
D-5 Tests to Study Lateral Spreading of Water in Bentonite
The Gundseal used  in plot F contained a lightweight GT backing called  "spidernet."
Because unexpected lateral spreading of water in the bentonite component occurred in
plot F, tests were performed to determine  if the  spidernet  contributed to the lateral
spreading of water.   Tests were  performed by  placing a 0.9-m-diameter piece of
Gundseal (with or without the spidernet) in  a large pipe with the bentonite facing up
(Figure D-34).  A sheet of textured HOPE GM with a 152 mm-diameter hole was placed
over the sheet of Gundseal. A 152 mm-diameter standpipe was sealed to the edges of
the hole in the overlying GM.  The volume of the large pipe surrounding the standpipe
was filled with aggregate.

Three series  of lateral rate of wetting tests were  performed.  In the first series,  two
specimens of Gundseal (with textured  HOPE GMs) were tested, one with  a spidernet
and the other without the spidernet. At the  beginning of each test, water was  poured
rapidly into the standpipe to a height  of 152 mm.  In both  tests (with spidernet  and
without spidernet) the lateral spread of water was greater than 460 mm after twenty-four
hours of the application of 152 mm of head.

In the first series of tests, it was believed that the application of water in the standpipe
was too rapid.  Therefore, a second series of tests was performed on two specimens of
Gundseal (with textured  HOPE GMs),  one with  a  spidernet and  one  without  the
spidernet. Water was introduced in 50 g increments per day for 7 days. This allowed
the bentonite to "pre-hydrate" and create a seal with the standpipe. After 7 days of pre-
hydration, water was added  to the standpipe  to  a  total  height of 150 mm.  The
specimens were left for an additional 7 days under 150 mm of head. After that time, the
water content of  the  bentonite  was determined at distances  from  the center of the
standpipe to  the  edge of the  Gundseal specimen.   Figure  D-35  shows the water
contents  from  the 7-day  inundation period.   The water contents were similar  for the
specimens with and without the spidernet.

Another series of tests was performed on two specimens of Gundseal, with  and without
the spidernet.   The Gundseal specimens had smooth HOPE  GMs.  This series was
performed exactly like the second series except that the specimens were allowed to sit
under 152 mm of head for 21 days. After 21 days, water contents were determined at
distances from the center of the  standpipe to the edge of the  Gundseal specimen. The
water contents were similar for the specimens with and without the spidernet (Figure D-
36).

The results of these tests  indicate that the spidernet has little effect on the lateral rate of
wetting of Gundseal for inundation periods of 21 days.  However, it appears that the rate
of inundation has an effect on the lateral spread of water.  The initial  series of tests
indicate that rapid inundation of water leads to greater lateral spreading of water.

                                     D-42
-------
                                   Textured HOPE Cover
T
             0.9m
 I
                           0.9-mlD   (PLAN VIEW)
                           Sonotube
                         Aggregate
                           Bentonite
                                                     150-mm ID Standpipe
                                      150mm
                                        XT'
                                       Aggregate
                 Gundseal - Bentonite Face Up
                          Textured  H DPE Geomembrane
                                    (SIDE VIEW)

Figure D-34. Apparatus to study lateral spreading of water in bentonite.
                                      D-43
-------
             600
             500
             400
             300
             200
             100
 I.D. of
standpipe
 76 mm
                                              • 7 days w/spidernet
                                              D 7 days w/out spidernet
                             50           100           150

                        Distance Measured from Center of Standpipe (mm)
                                                                    200
Figure D-35. Results from rate of lateral wetting tests after 7 days.
D-6  Erosion Control Materials
Erosion control materials were placed on the surfaces of all the test plots, except Plot
M, which was intentionally not covered with any erosion control material as a control
plot.  The purpose of the erosion control materials was to stabilize the slopes rapidly
and to maintain each slope's surface integrity. Erosion control materials provide for the
rapid growth of seeded grass by retaining heat from the sun and limiting erosion due to
overland runoff.   The erosion control materials give shelter to the seeds from flowing
water and winds.

Table D-8 summarizes the erosion control products that were placed on the various test
plots. Three plots (E, K, and N) had a sacrificial, biodegradable woody material applied
to the surface.  All erosion control materials were installed according to manufacturer's
specifications.  The erosion control materials were installed in an overlapping manner
and stapled together.   They were stapled to the soil at spacings of approximately 1 m
per the manufacturer's recommendations. Some plots were seeded prior to placement
of the erosion control  material,  and others were seeded after the erosion control
placement  (depending  on  the manufacturer's recommendation).   The  site  owner
provided for the seeding of the plots in December, 1994.
                                       D-44
-------

800 j
700 -

600 -
500 -
400 -
300 -
200 -
100 -

n
• 21 daysw/
1 D21 daysw/
I.D.of
standpipe
• B 76 mm
D



4 •
i 1 i D D.
spidernet
out spidernet









                             50           100          150

                        Distance Measured from Center of Standpipe (mm)
                                                                 200
Figure D-36. Results from rate of lateral wetting tests after 21 days.

All of the erosion control materials have worked well.  There were significant erosion
gullies and  some sloughing of the cover at the toe of the slope in the control Plot M that
did not contain any erosion control material.   All the erosion control materials appeared
to be functioning as designed and to have maintained the integrity of the surface of the
test plots.

D-7  Additional Laboratory Direct Shear Testing on an Unreinforced GCL
This section describes a laboratory-testing program designed to study the changes in
internal  shear strength of GCLs as a result of varying laboratory test preparations and
conditions.  Direct shear tests were  performed on the internal  portion  of  unreinforced
GCLs.  The test preparations and conditions evaluated include hydration time, shear
rate, and normal stress. An unreinforced GCL was selected to focus on the strength of
the bentonite itself.   The following section reports on results from  a one-dimensional
consolidation test of an unreinforced GCL

D-7.1 Materials Tested
A roll of Claymax  200R was  supplied  by the manufacturer and used  for testing.
Claymax 200R is an unreinforced GCL consisting of bentonite mixed with  an  adhesive
and sandwiched between two nonwoven GTs. Specimens of  Claymax 200R were used
in the direct shear and one-dimensional consolidation tests.
                                      D-45
-------
Table D-8 Geosynthetic Erosion Control Products.
Plot
A
B
C
D
E
F
G
H
1
J
K
L
M
N
P
Manufacturer
Tensar
Synthetic Industries
Synthetic Industries
Akzo
Akzo
Tensar
Tensar
Tensar
Synthetic Industries
Synthetic Industries
Akzo
Akzo
None
Akzo
Akzo
Product
TB 1000
Polyjute
Polyjute
Enkamat7010
Enkamat7010
TM 3000
TM 3000
TM 3000
Landlok 450
Landlok 450
Enkamat7010
Enkamat7010
Control Plot
Enkamat7010
Enkamat 7220
Color
Green
Beige
Beige
Black
Black
Black
Black
Black
Green
Green
Black
Black
-
Black
Black
Material
Polyolefin
Degradable
Polypropylene
Degradable
Polypropylene
Nylon
Nylon (with Excelsior)
Polyethylene
Polyethylene
Polyethylene
Polyolefin
Polyolefin
Nylon (with Excelsior)
Nylon
-
Nylon (with Excelsior)
Nylon
D-7.2 Direct Shear Tests
      D-7.2.1  Testing Equipment. Two different direct shear machines were utilized in
the direct shear tests.   A 64 mm diameter direct shear machine supplied  by ELE
(Engineering Laboratory Equipment,  Ltd.)  was used  in tests  involving  lower normal
stresses.  A 63.5 mm diameter Wykeham Farrance direct shear machine was used in
tests involving higher normal stresses.

      D-7.2.2   Testing Variables.  The testing variables evaluated include hydration
time, shear rate, and normal stress. The effects of one variable on the measured shear
strength was studied while the other two  variables were kept constant.  Table D-9
describes  each test  by listing the test number, hydration time before  shear, normal
stress, and shear rate.

      D-7.2.3  Hvdration Time.  Direct shear tests were performed after the GCLs had
been hydrated 24, 48, 72, and 152 hours.  Three direct shear tests were  performed for
each different length of hydration, except only one  shear test was performed after 152
hours of hydration.  All tests were performed with a normal stress of 17 kPa and a shear
rate of 1 mm/min.
                                     D-46
-------
Table D-9.  Outline of Shearing Program.
Test No.
1
2
3
4
5
6
7
8
9
23
14
15
19
20
21
22
24
25
26
27
29
37
38
32
36
28
30
Hydration
Period
(hrs)
24
24
24
48
48
48
72
72
72
153
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
Shear
Rate
(mm/min)
1
1
1
1
1
1
1
1
1
1
0.1
0.1
0.1
0.02
0.02
0.02
0.0025
0.0025
0.0025
0.0005
0.0005
1
1
0.024
0.024
0.0005
0.0005
Normal
Stress
(kPa)
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
172
172
172
172
172
172
      D-7.2.4.  Shear Rate and Normal Stress. The effect of shear rate on the shear
strength of GCLs was investigated at two normal stresses, 17 kPa and 170 kPa.  At a
normal stress of 17 kPa, the strengths were investigated at shear rates of 1, 0.1, 0.02,
0.0025,  and 0.0005  mm/min.  At a normal  stress of 170  kPa, the strengths were
investigated  at  shear rates of 1, 0.024, and 0.0005  mm/min.  All shear  tests were
performed after 24 hrs of hydration.
                                     D-47
-------
The effect of normal stress on the internal shear strength was  investigated at three
different shear rates, 1,  0.02, and 0.0005 mm/min.   At  each shear rate,  tests were
performed at normal stresses of 17 kPa and 172 kPa.  All shear tests were performed
after 24 hrs of hydration.

D-7.3  Specimen Description and Preparation
Direct shear tests were performed according to ASTM  D3080.  However, the procedure
was modified and is described in the following paragraphs.

Sections of 200 mm x 200 mm GCL were cut from a sheet of Claymax200R. The direct
shear specimen was trimmed to the diameter of the direct shear box from one of these
sections.   Trimming  was  carefully performed in  order  to  minimize  the  escape of
bentonite from the GCL.  Initial water contents were obtained.

A thin layer of vacuum grease was applied to the lower surface of the upper shear box,
to the upper surface of the lower shear box, and to the inside of the shear box. The
GCL specimen was  placed in the direct shear box  in the as-received (dry) condition.
Additional porous stones  and thin  plates of foil were used to adjust the vertical position
of the specimen in the direct shear box in order to force  the failure plane through the
mid-plane of the hydrated GCL.  The normal stresses used during shear were applied to
the  specimens  in one increment.  Water was added to the shear boxes,  and the
specimens were hydrated for a specific period of time.  The  height of the specimens
was monitored during hydration.

After hydration for the designated amount of time the specimen was prepared for shear.
The upper shear box was raised by turning the shear box  screws a one-quarter to one-
half turn.   However, the  shear  screws were not removed.  Instead, they were left in
contact with the lower shear box.  Shearing of the specimen commenced and the shear
stress and height were monitored. Shearing  was performed at a designated shear rate,
and the test was terminated after the maximum shear developed.  The specimen was
dismantled, and the water content after shear was obtained.

D-7.4  Correction for Shear Box Friction
The measured shearing resistance of the GCL was adjusted for friction that developed
between  the shear box screws and the lower shear  box. The friction of the screws
against the shear  box surface was measured by performing several shear tests on an
"empty" direct  shear box with  no normal  load  applied and  recording the maximum
resistance.  The maximum resistance in both shear machines was 1 kPa, and this value
was used as the friction.  After a GCL was sheared, the friction was deducted from the
maximum recorded shear resistance of the GCL.
                                     D-48
-------
D-7.5 Test Results
Individual curves of shear stress vs. displacement for all direct shear tests are shown in
Attachment 5. The shear stresses shown in Figures 1 to 11  in Attachment 5 have not
been adjusted for friction.

      D-7.5.1  Effect of Hvdration Time.   Table  D-10 lists  the adjusted peak shear
stress and the displacement at peak stress  from direct shear tests on GCL specimens
after 24, 48,  72 and 153  hours  of hydration.  The average, standard  deviation, and
coefficient of  variation for peak stress  and peak displacement are listed in Table D-10
for the different  lengths of hydration.   Overall, the averages of peak stress for each
hydration period are very similar.  However, within each hydration period,  the peak
stresses are variable.  The coefficient  of variation  for the peak shear stresses in each
hydration period range between 10 and 11%.

Table D-10. Shear Results for Various Hydration Periods.
Test
No.
1
2
3
Hydration
Period
(hr)
24
24
24
Shear
Rate
(mm/min)
1
1
1
^normal
(kPa)
17
17
17
Average
Standard Deviation
Coefficient of Variation
4
5
6
48
48
48
1
1
1
17
17
17
Average
Standard Deviation
Coefficient of Variation
7
8
9
72
72
72
1
1
1
17
17
17
Average
Standard Deviation
Coefficient of Variation
23
153
1
17
*
Tpeak
(kPa)
10.2
11.9
9.8
10.6
1.1
0.11
9.6
11.9
10.6
10.7
1.2
0.11
11
9.6
11.7
10.8
1.1
0.10
11.3
Speak
(mm)
3.2
4.3
4.3
3.9
0.7
0.17
2.3
4.3
3.6
3.4
1.0
0.30
3.6
2.3
2.4
2.8
0.7
0.25
2.9
            * peak shear stress is adjusted for machine friction
            ^normal = normal stress; Tpeak = peak shear stress; 5peak = displacement
            at Tpeak
                                      D-49
-------
Figure D-37 contains a Mohr-Coulomb diagram showing the results listed in Table D-10.
Friction angles for the tests in each hydration period were determined from regression
analysis forced through the origin (i.e. assuming no cohesion) and are listed in Table D-
11.  The results indicate that lengths of hydration between  24 and 153 hrs have little
effect on the measured internal shear strength of unreinforced GCLs at normal stresses
of 17 and 172  kPa.
             20
             15 -•
             10
H   24 hr hydration

O   48 hr hydration

X   72 hr hydration

A   153 hr hydration

    24 hr hydration regressic

    48 hr hydration regressic

    72 hr hydration regressic

    153 hr hydration regress:
                                         10
                                                      15
                                                                   20
                                  Normal Stress  (kPa

Figure D-37. Mohr-Coulomb diagram for GCLs hydrated for various durations.




Table D-11. Friction Angles of GCLs Hydrated for Various Durations.
Hydration
Time
(hr)
24
48
72
153

Number
of Tests
3
3
3
1

Friction Angle
(degrees)
29.5
30.0
30.5
31.7
      D-7.5.2.  Effect of Shear Rate and Normal Stress. Table D-12 lists the adjusted
peak shear stress and the displacement at peak stress from direct shear tests on GCL
                                      D-50
-------
specimens after being sheared at rates of 1, 0.1, 0.02, 0.0025, and 0.0005 while under
17 kPa normal stress.   Table  D-13 lists  the  adjusted peak  shear  stress and the
displacement at peak stress from direct shear tests  on GCL specimens after being
sheared at rates of 1, 0.024, and 0.0005  while under  170  kPa normal stress.   The
average,  standard  deviation,  and coefficient of variation for peak stress  and  peak
displacement are listed for each series of tests.  Figure D-38 shows the ratio of peak
strength to normal  stress versus  shear rate for tests conducted with 17 and 172 kPa
normal stress. Table D-14 summarizes the results.

Results from  the shear tests indicate that at both normal stresses,  the shear strength
decreases as the shear rate decreases.  It appears that only after  the shear rate has
decreased below 0.01 to 0.0001  mm/min does the shear strength become constant.
The friction angle for each group  of tests (at a particular shear rate  and normal stress)
was determined by forcing a failure envelope through  a  Mohr-Coulomb diagram of the
shear results where the envelope was forced through the origin. The results  show that
as the shear rate decreases, the friction angle decreases. At  17 kPa normal stress, the
friction angle is 29.5 for shear rates of 1 mm/min and decreases to less than 22 degrees
for shear rates below 0.0025 mm/min. At 172 kPa normal stress, the  friction angle is
15.7 degrees for shear rates of 1 mm/min and decreases to less than 13 degrees below
shear rates of 0.024 mm/min.

Results from  the shear tests also indicate that the  shear strength increases  as normal
stress increases. However,  both the  ratio of peak shear stress to normal stress and the
friction angle decrease as the normal  stress  increases.

D.7.6  One-Dimensional Consolidation Test
The purpose of performing a consolidation test was to  verify appropriate shearing rates
determined from the direct shear tests of  hydrated unreinforced GCLs specimens.  The
results from the direct shear tests (discussed in the  previous section) indicate  that when
the GCL is sheared slower than 0.01  to 0.001 mm/min at normal stresses of 17 to 170
kPa, the  measured shear strength will be  independent  of the shearing  rate.  A one-
dimensional  consolidation test was  performed  on  a  specimen of Claymax 200R  to
determine the estimated time to failure for normal stresses close to 17 and 170 kPa.

The consolidation test was performed in a Wykeham Farrance loading frame  in a fixed-
ring consolidation cell. The diameter  of the cell was 64  mm.

The specimen for the consolidation test was obtained from the same sheet of Claymax
200R  used  in the  direct shear  specimens testing. The specimen was trimmed  to
minimize escape of bentonite from the GCL.  The initial height of  the specimen was
measured,  and then the specimen  was placed in the consolidation  cell  in the as-
received condition.

                                     D-51
-------
Table D-12.    Results for GCLs Sheared at Various Rates under 17 kPa Normal
              Stress.
Test
No.
1
2
3
Shear
Rate
(mm/min)
1
1
1
Normal
Stress
(kPa)
17
17
17
Hydration
Period
(hr)
24
24
24
Average
Standard Deviation
Coefficient of Variation
14
15
19
0.1
0.1
0.1
17
17
17
24
24
24
Average
Standard Deviation
Coefficient of Variation
20
21
22
0.02
0.02
0.02
17
17
17
24
24
24
Average
Standard Deviation
Coefficient of Variation
24
25
26
0.0025
0.0025
0.0025
17
17
17
24
24
24
Average
Standard Deviation
Coefficient of Variation
27
29
0.0005
0.0005
17
17
24
24
Average
Standard Deviation
Coefficient of Variation
Peak
Shear
Strength
(kPa)
10.2
11.9
9.8
10.6
1.1
0.11
10.3
11.9
12.1
11.5
1.0
0.09
9.5
8.1
9.0
8.9
0.7
0.08
6.7
7.2
6.6
6.8
0.3
0.04
7.9
6.6
7.2
0.9
0.13
Displ. at
Peak
Strength
(mm)
3.2
4.3
4.3
3.9
0.7
0.17
2.5
3.4
3.3
3.1
0.5
0.16
2.8
3.5
3.1
3.1
0.4
0.11
3.4
3.4
2.8
3.2
0.4
0.11
2.4
2.4
2.4
0.05
0.02
                   peak shear stress is adjusted for machine friction
                                   D-52
-------
Table D-13.    Results for Various Rates and GCLs under 170 kPa Normal
              Stress.
Test
No.
37
38
Average
Standar
Coeffici
32
36
Average
Standar
Coeffici
28
30
Average
Standar
Coeffici
Shear
Rate
(mm/min)
1
1
*
d. Deviation
ent of Variat
0.024
0.024
i
d. Deviation
ent of Variat
0.0005
0.0005
*
d. Deviation
ent of Variat
Normal
Stress
(kPa)
172
172
ion
172
172
ion
172
172
ion
Hydration
Period
(hr)
24
24
24
24
24
24
Peak
Shear
Strength
(kPa)
50.7
59.1
54.9
6.0
0.11
43.8
38.7
41.2
3.6
0.09
46.2
38.9
42.5
5.2
0.12
Displ. at
Peak
Shear
Strength
(mm)
2.8
4.1
3.4
0.9
0.26
2.3
2.1
2.2
0.2
0.07
2.3
4.2
3.2
1.4
0.4
                  peak shear stress is adjusted for machine friction
Table D-14. Effect of Normal Stress and Shear Rate on Friction Angle.
Shear
Rate
(mm/min)
Number
of Tests
Friction Angle
(degrees)
Normal Stress = 17 kPa
1
0.1
0.02
0.0025
0.0005
3
3
3
3
2
29.5
31.9
25.6
20.3
21.6
Normal Stress = 170 kPa
1
0.024
0.0005
2
2
2
15.7
12.9
13.0
                                   D-53
-------
0.7 -







[
1










Illl



D 1 7 kPa Normal Stress
I 1 72 kPa Normal Stress
















i
i
i
i




i
i
i
i





-H







**•





















c
C
• • " i






jH"


1



X


-



x '
X


...


0.0001 0.001 0.01
Shear Rate (mm/min)
I
1


.... —






—






—



~"


— "


0.1
11
'

II
'ii


1
Figure D-38.  Effect of Normal Stress and Shear Rate on Shear Strength.
The consolidation test was performed according to ASTM D2435. A seating load of 11
kPa was applied, and water was introduced in to the consolidation cell.  The specimen
was allowed to hydrate under 11 kPa normal stress until the swelling ceased or until the
height of the specimen became constant.  After hydration, the first load was applied.
Subsequent loads were applied after the specimen had passed primary consolidation,
which was longer than 24 hours in most cases.  Loads were applied to a maximum of
307 kPa using a load-increment ratio of 2.  Then, the specimen was unloaded using a
load increment ratio of 4.

The hydration curve of the GCL in the consolidation test is shown in Figure D-39.  The
test lasted 48 days. Also shown  in Figure D-39 are hydration curves of several GCL
specimens tested in the direct shear tests.  The consolidation specimen swelled from an
initial thickness of 4.7 mm to a thickness of 14.6 mm under a  seating load of 11 kPa.
The hydrated  thickness of the consolidation specimen  was  greater than  all  of  the
hydrated thicknesses of the direct shear specimens.  However,  the GCL in direct shear
Test  No. 7 swelled close to  12 mm.   The difference in hydrated  thicknesses of  the
consolidation  and direct shear specimens may  have occurred  because of different
hydration conditions. First, the direct shear specimens were all hydrated under 17 kPa
normal stress, whereas the consolidation specimen was hydrated under 11 kPa normal
stress.  However, the  interpolated (between load  increments of 11 kPa and 20 kPa)
hydrated thickness of the GCL at 17 kPa  was 14.2 mm.  Therefore,  the difference in
normal stress between  11 kPa and 17 kPa might not be a factor in the differences in

                                     D-54
-------
hydration thicknesses of the consolidation specimen and  shear specimens.   Second,
most of the direct shear specimens were  hydrated for only 24 hours (only seven were
hydrated for durations of 2 to 7 days), whereas the consolidation specimen was allowed
to hydrate for 48 days.
                                      •  Test No. 7

                                      D  Test No. 8

                                      A  Test No. 23

                                      —I—Consolidation Test (Seating
                                          I

                                         150
 I

200
 I

250
                         50
                                 100
                                                                 300
                             Square Root of Time  (min
                                                   1/2
Figure D-39.  Hydration curves for Claymax200R.
The height of the specimen corresponding to the end of primary consolidation and end
of loading increment was determined using the square root of time method. The end of
primary consolidation  height and end  of loading  increment  heights are shown in the
consolidation curve in Figure D-40.

The minimum time to failure,  tf, required for a consolidated-drained direct shear test,
according to Gibson and Henkel (1954) and Terzaghi's theory of consolidation, is:
             f  2cv(l-Uf)

If the coefficient of consolidation, cv, which is given by:
                                                                         (D.2)
                 0.197Hc2
            c  =	
             V      f
                    L50
                        (D.3)
                                     D-55
-------
             16
             14
             12
             10
                                                  \
                   ""•  End of  Primary
                       Consolidation

                    D  End of  Increme:
                                10               100

                                    Stress (kPa)
                                                                 1000
Figure D-40. Consolidation curve of Claymax200R.
is  substituted  in equation  D.2, where Hc  is the average  drainage  distance during
consolidation (or  half the  specimen  height) and t50  is  the  time  required for  50%
consolidation, then equation D.3 becomes:
             f  (2)(.197)(Hc2)(l-Uf)
                                    tso
(D.4)
By assuming the thickness of the specimen is the same during consolidation and shear
(Hc is equal to Hd), and Uf is equal to 95%, Equation D.4 reduces to:
            tf=50t50
(D.5)
Equation D.6 is the method specified in ASTM D3080 for determining the rate of shear.

The  required times to failure were  calculated  based on Equation  D.5 for each load
increment of the consolidation test and are listed in Table  D-15.  Maximum  allowable
shear rates  in a consolidated drained  direct shear test  were calculated assuming a
                                     D-56
-------
displacement at failure of 2 mm. Based on times to failure calculated from consolidation
results, consolidated-drained shear tests with normal stresses of 17 kPa require a shear
rates less than .001 mm/min, and shear tests with normal stresses of 154 kPa require
shear rates less than .0002 mm/min.  The maximum shear rates that were determined
from the consolidation test  at normal stresses  of 17 kPa  and  154 kPa are consistent
with the maximum shear rates determined from the direct shear tests at similar normal
stresses.
Table D-15. Maximum Shear Rates.
Stress
(kPa)
19.7
38.3
77
154
307
tso
(min)
36
132
149
210
272
Estimated
Time to Failure, tf
(min)
1800
6600
7450
10500
13600
Maximum
Shear Rate
(mm/min)
0.001
0.0003
0.0003
0.0002
0.0001
D-8 References
Daniel, D.E., Shan,  H.Y., and Anderson, J.D. (1993),  "Effects of Partial Wetting on the
  Performance  of   the  Bentonite  Component  of  a  Geosynthetic  Clay  Liner,"
  Geosynthetics '93, Industrial Fabrics Association International, St. Paul,  Minnesota, 3:
  1483-1496.
Estornell, P. M. and  Daniel,  D. E. (1992), "Hydraulic Conductivity of Three Geosynthetic
     Clay Liners," Journal of Geotechnical Engineering, 118(10): 1592-1606.
Gibson, R.E., and Henkel, D.J. (1954), "Influence of Duration of Tests at Constant Rate
     of Strain on Measured 'Drained' Strength," Geotechnique, 4: 6-15.
Shan, H.Y.  (1993), "Stability of Final Covers Placed on Slopes Containing Geosynthetic
  Clay Liners," Ph.D. Dissertation, Univ. of Texas, Austin, TX, 296 p.
                                      D-57
-------
                            Appendix D
                           Attachment 1
       Results of Laboratory Direct Shear Tests on GCL Interfaces

(Tests Performed on 300 mm x 300 mm Samples in the Laboratories of the
                   Geosynthetic Research Institute,
                         Drexel University)
                               D-58
-------
  300

ft 25° *
3-200

| 150

§ 100

5  50
   GUNDSEAL (bentonite side) vs HOPE (textured) - DIRECT
                       SHEAR (12")
                                350 Ibs (2.6 psi) Normal
                                Stress
0.0      0.2     0.4     0.6      0.8     1.0
                        DISPLACEMENT (IN)
                                           1.2
                                                            1.4
1.6
V)
Q.
CO
to
£
CO

CO
0>
CO
                            234
                           Normal  Stress  (psi)
                        D-59
-------
           BENTOMAT (silt film) vs HOPE (textured) - DIRECT SHEAR
   160
'   140 •
S  120 •
ut  100
0   80
$  40
    20
     0
      0.0
0,2
                            350 (bs (2-6 pslj Normal
                            Stress
0.4      0.6       0.6
       DISPLACEMENT (IN)
                                                     1.0
                                              1.2
                                                                       1.4
       Cf)
        b
        OS
        0)
                                          *  ReaduaUei.4-displ.)
                                      2          3
                                   Normal  Stress  (psi)
-------
  zoo
  180
ff 160
3 140
8120
o 100
u.
    80
    60
    40
    20
     0
SHEAR
      0.0
BENTOFlX-type 11 (Slit Film) vs HOPE (textured) - DIRECT
                    SHEAR  (12")
                              350 lb£ (2.G psi) Normal
                              Stress
   0.2
                        0.4
  0.6       O.B
DISPLACEMENT (IN)
                                       1.0
                                                           1.2
                                                        1.4
                                   3          4

                       Normal Stress  (psi)
-------
                  SHEARPRO VS SAND - DIRECT SHEAR {12"}
  250
a,
g 150
DC
O
u_
IE 100
a
X
W  50
     0
                350 (bs (2.6 psi) Normal

                Stress
      O.O      0,2       O.4       O.G       O.Q

                                DEFLECTION (IN)
                                                   1.0
                                 1.2
   to
   O.

   (D
   tj
   O
   LU

   TO
   Q)

   CO
       4-
          O
y = 0.59696X  RA2 = 1.000
        T

        Z          3

       Normal Stress (psi)
                                 D-G2
-------
SAND - DIRECT SHEAR (12")
nnn
V?
s
III
K
£

2
CO


180
160
UO
120
100
BO
60
40
?o
r\
•

-



;
-
.

m •
H •
Hm"
•
m
1


0.0 0.2 0.4 0.6
m
-



• 350 tbs (2.6 psi) Normal
Stress


•

















0,8 1.0 1.2 1.4
DEFLECTION (IN)
co
CD
L>
OJ
-------
           BENTQFIX (NPNWGT) vs HOPE (textured) - DIRECT SHEAR


                                   (12")
  300
^ 250
at
to

=4 200 H
uJ
    50



     0
                      • •
                                         350 Ibs (2.6 psi} Normal

                                         Stress
      0.0       0.2       0.4      0.6      0,B


                              D1SPLACEMEHT (IN)
1.0
1.2
                                                                   1.4
      tn
      CO
                                  Normal Stress (psi)



                                 D-64
-------
   1.20


~ 1-00
*5

w  0.80
O

£  0,60

cc
S  0-40

10
   0.20
                SHEARPRO vs TEXTURED GEOMEMBRANE
0,00

   0.
                                               2.6 psi Normal Stress
        00       0.20       0.40       0.60       0.80

                              DISPLACEMENT (in)
1.00
1.20
 u

 o
LL
 i_
 (0

x:
                        y = 0.36502X  RA2 - 1.000
                                           3          4
                                     Normal  Stress  (pst)
                              !>65
-------
                      Appendix D
                     Attachment 2
Plots of Total Down-Slope Displacements of GCLs Versus Time
                         D-66
-------
 50 -L
  Downslope
50 -L
Downslope
                   Gauge 1 (Above GCL)
                                                           800
                                                 Time (days)
                   - Left Panel
                                -Right Panel
                  Gauge 5 (Above GCL)
                                                           800
                                                      Time (days)
                     - Left Panel
                                 -Right Panel
                   Gauge 1 (Below GCL)
                                                                                                                                  800
50 -L
Downslope
                  Gauge 5 (Below GCL)
                                                                                                                            700     800
                                                                                                                         Time (days)
                                                                                             - Left Panel
                                 -Right Panel
                             Figure A.1:  POST CONSTRUCTION DISPLACEMENT VS. TIME FOR PLOT A
                                               (Gundseal - Bentonite Side Up - 3:1 Slope)
-------
                             Gauge 1 (Above GCL)
D
oo
                         200    300    400    500    600     700
                                                                   800
  Gauge 1 (Below GCL)
                                                                                                                                       800
        50 -L
          Downslope
                           Gauge 5 (Above GCL)
                                                          Time (days)
                                  -Left Panel -o-Right Panel
Gauge 5 (Below GCL)
                                                                                                                                       800
                           Figure A.2: POST CONSTRUCTION DISPLACEMENT VS. TIME FOR PLOT B
                                                     (Bentomat- 3:1 Slope)
-------
                               Gauge 1 (Above GCL)
                                                                      800
                      Gauge 1 (Below GCL)
                                                                                                               400    500   _600

                                                                                                                    •B-B-
                                                                                                                                     700
                                                                                                                                            800
                                                                                    Downslope
                                                                                                       -Left Panel -n-Right Panel
(35
CD
                             Gauge 5 (Above GCL)
                                                900 A  AQCO A A 70t     800
                                                            Time (days)
          50^

           Downslope
                                   -Left Panel -D-Right Panel
                    Gauge 5 (Below GCL)
                                                                                                                                            800
                                                                                                                                  Time (days)
50^
 Downslope
                          - Left Panel -a- Right Panel
                             Figure A.3: POST CONSTRUCTION DISPLACEMENT VS. TIME FOR PLOT C

                                                        (Claymax-3:1 Slope)
-------
O

o
-50 -r


-40 --


-30 --


-20 --
    Downslope
Gauge 1 (Above GCL)
                                                   Time (days)
                                        500     600     700     800
                                     •-•—•—•-•—•-•
                        -Left Panel -o-Right Panel
                                                                                   -50 y

                                                                                   -40 --

                                                                                   -30 --


                                                                                   -20 --

                                                                                1-10-
                                                                                r    0
                                                                      B
                                                   20 --


                                                   30-

                                                   40 --

                                                   50 --
Gauge 1 (Below GCL)
                                                                                                      Time (days)
                                                             100  p  200    300     400    500     600     700     800
                                                                                                              •—•—•—•-•—•-•
                                                                                     Downslope
                                                                           -Left Panel -o-Right Panel
   Downslope
                      Gauge 5 (Above GCL)
                                                                         Gauge 5 (Below GCL)
                                                              800
                                                   Time (days)
                                                                                                                                       800
                                                                                                                            Time (days)
                                                      Downslope
                        - Left Panel •
                                     -Right Panel
                                                                          -Left Panel —n— Right Panel
                              Figure A.4: POST CONSTRUCTION DISPLACEMENT VS. TIME FOR PLOT D
                                                     (Bentofix II (NW up)  3:1 Slope)
-------
                  Gauge 1 (Above GCL)
                                            Time (days)
                                                      800
i 1  Downslope
               -Left Panel -n-Right Panel
                                                                                      Gauge 1 (Below GCL)
    Time (days)
600    700    800
        ID
                                                                                        -Left Panel —n—Right Panel
                                                                   Downslope
                                                      800
                                                                                        Gauge 5 (Below GCL)
                                                                                                                           800
                         Figure A.5:  POST CONSTRUCTION DISPLACEMENT VS. TIME FOR PLOT E
                                         (Gundseal - Bentonite Side Down - 3:1 Slope)
-------
D
ro
                           100
                                     Gauge 1 (Above GCL)


                                     200       300       400
Time (days)
500       600
               750 -1
                Downslope
                                        - Left Panel
                                                     - Right Panel
                                                                                        -so y

                                                                                         50 --

                                                                                      _ 150-

                                                                                      -§ 250 --
                                                                                      •E
                                                                                      E 350 --
                                                                                      8
                                                                                                     100
Gauge 1 (Below GCL)


200       300       400
Time (days)
500       600
                     550--

                     650 --

                     750 --
                       Downslope
                                                                                                                 -Left Panel
                                                           -Right Panel
                                     Gauge 5 (Above GCL)
                                                300
                                            Gauge 5 (Below GCL)
                                                         400
                                                                  Time (days)
                                                                   500      600
                  j -j-
                    Downslope
                                                                                          -50 y

                                                                                          50 -

                                                                                       ,-, 150 --

                                                                                       — 250 -
                                                                                       •E
                                                                                       E  350 -
                                                                                       8
                                                                                       f 450 -
                                                                                       V)
                                                                                       b  550 -

                                                                                          650-

                                                                                          750-
                                                                         Time (days)

                                                                          500       600
                                                                                             Downslope
                                    - Left Panel
                                                  Right Panel
                                                                                                             -Left Panel
                                                       -Right Panel
                                             Figure A.6:  POST CONSTRUCTION DISPLACEMENT VS. TIME FOR PLOT F
                                                                (Gundseal - Bentonite Side Up - 2:1 Slope)
-------
Gauge 1 (Above GCL)
-50 -,
-40-
-30-
§-20-
E,
^-10<
0
0
J 10 -
Q.
B 20 -
30-
40 -
50 -
Do











) 10 20 30 40
? 	 a 	 -^
	 1 	 1 	 r




tJ









Time (days)
50 60
	 r 	 A
^^^^^ I LJ 	 1




Failure: 788 in.


"nS °PS -A- Left Panel -n- Right Panel


Gauge 5 (Above GCL)
-50 j
-40 --
-30 -
? -20 -

iT -10 t) 10 20 3®~^2^t
0 „ A 	 ^~~-35
i— n i i T
0
n 10 --
Q.
•— 20
b
30 -
40 --
50 -L






0
^Vg














Tir^(days) 60
^— -r 	 *
^==I: 	 n '







Downslope Failure: 789 in.

-*-Left Panel -n- Right Panel



































Gauge 1 (Below GCL)
-50 -,
-40 -
-30
§-20-

0
0
J 10 -
Q.
B 20
30
40 -
50 J





) 10 20 5jl 	 n^40 50 60
A A ^^****n m i
	 1 	 m • I T-l I ttl 	 1





Downslope

-*- Left Panel -n- Right Panel

Gauge 5 (Below GCL)
-50 -r
-40 --
-30 -
f -20 -
-§

0 «
0
n 10 --
Q.
•— 20
b
30 -
40 --
50 J-
Downslope

-A- Left Panel -n- Right Panel

Figure A.7: POST CONSTRUCTION DISPLACEMENT VS. TIME FOR PLOT G
                      (Bentomat - 2:1 Slope)
-------

-50 T
-40 -
-30 -
~E -20 -
E
£-10,
E 0
0
8 10-
Q.
0 20 -
30 -
40 -
50 -




) 5
I






Downslope


Gauge 1 (Above GCL)









Time (days)

10 15 20 25 30
I I







—•—Left Panel —n— Right Panel

I I





Fa
















j|





-50 j
-40
-30
~B -20
E
£-10,
P 0
c *-*
0
co 10
Q.
5 20
30
40
50




) 5







Gauge 5 (Above GCL)









Time (days)

10 15 20 25 30
I I






Downslope

-•-Left Panel -o- Right Panel








Failure: 668 in.

















Gauge 1 (Below GCL)
-50 j
-40
-30
~E -20
E
£-10,
E 0
8 10
ro 1U -
Q.
0 20
30
40
50 J


Time (days)

) 5 10 15 20 25 n 30
III I A I






Downslope
—A— Left Panel —n— Right Panel


Gauge 5 (Below GCL)
-50 j
-40
-30
~B -20
E
£-10,
P 0
c *-*
0
co 10
Q.
5 20
30
40
50



Time (days)

) 5 10 15 20 25 30
III II






Downslope
-•-Left Panel -o- Right Panel

Figure A.8: POST CONSTRUCTION DISPLACEMENT VS. TIME FOR PLOT H
                      (Claymax - 2:1 Slope)
-------
D
en
              -50 y
                  0
                       Gauge 1 (Above GCL)

                   200    300    400    500
                                                           600    700    800
I 100
              150 +
              200--
              250 -1
                 Downslope
                                 -Left Panel -n-Right Panel
     -50 T
     Gauge 1 (Below GCL)
200    300    400    500
                                                                                                                                   600    700
                                                                                                                                                 800
                                                                                                                                                 —I
                                                                                    I  100
                                                                           150 --
                                                                                      200 --
                                                                          250 -1
                                                                             Downslope
                                                                                             -Left Panel -n-Right Panel
           -50 T
           50 --
         E 100 --
        0150 --
          200 --
          250 -L
           Downslope
                  Gauge 5 (Above GCL)

         100     200     300     400     500     600     700     800
                               -Left Panel -D-Right Panel
-50 T               Gauge 5 (Below GCL)
                                                                                             100     200     300     400     500     600     700     800
                                                                                  150 --
200 -L
 Downslope
           - Left Panel -a- Right Panel
                                           Figure A.9: POST CONSTRUCTION DISPLACEMENT VS. TIME FOR PLOT I
                                                                     (Bentofixl- 2:1 Slope)
-------
                          Gauge 1 (Above GCL)
                    200    300    400     500     600     700     800
        Downslope
                      -Left Panel
                                   -Right Panel
                                                                                Gauge 1 (Below GCL)
                                                                                                                            Time (days)
                                                                                                                                       800
                                                              Downslope
                                                                                              -Left Panel
                                                                                         -Right Panel
-50 x


  0
    Gauge 5 (Above GCL)
200     300     400     500
                              600
                                      700
                                              800
500 -L

  Downslope
                                - Left Panel
                                             - Right Panel
                                                                 100
    Gauge 5 (Below GCL)
200    300     400    500
                                                                                                       600
                                                                                                              700
                                                                                                                     800
                                                       500 -L

                                                          Downslope
                                                                                                    - Left Panel
                                                                                               - Right Panel
                           Figure A.10: POST CONSTRUCTION DISPLACEMENT VS. TIMEFOR PLOT J
                                            (Bentomat - Granular Drainage - 2:1 Slope)
-------
                    Gauge 1 (Above GCL)
-50 Y,
          100
                 200
                        300
                               400
                                      500
                                                Time (days)
                                             600    700    800
    Downslope
                   Left Panel
                                Right Panel
                                                                             Gauge 1 (Below GCL)
                                                                                  100
                                                                                         200
                                                                                                300
                                                                                                        400
                                                                                                               500
                                                                                                           Time (days)
                                                                                                      600    700     800
                                                                         50 --
                                                                        150 --
                                                                     | 250-
                                                                     r 350 --
                                                                      E 450 --
                                                                      0
                                                                      £ 550 --
                                                                        750 --
                                                                        850 --
                                                                        950 --
                                                                             Downslope
                                                                                            Left Panel
                                                                                         Right Panel
        100
    Gauge 5 (Above GCL)
                                  Time (days)
200    300     400    500    600     700    800
                                                                                             Gauge 5 (Below GCL)
  Time (days)
600    700    800
                                                                            Downslope
                                                                                            Left Panel
                                                                                                        -Right Panel
                       Figure A.11:  POST CONSTRUCTION DISPLACEMENT VS. TIME FOR PLOT K
                                         (Claymax - Granular Drainage - 2:1 Slope)
-------
                     Gauge 1 (Above GCL)
                                                 Time (days)

                                             600     700    800
    Downslope
                   - Left Panel
                               - Right Panel
                    Gauge 1 (Below GCL)
                                                                                                                        Time (days)
                                                                                                                                   800
    Downslope
                                                                                          - Left Panel
                              - Right Panel
D

oo
                     Gauge 5 (Above GCL)
-50 T
                        300
                               400
                                      500
                                             600
                                                    700
                                                           800
350 -L
    Downslope
                   - Left Panel
                                -Right Panel
                    Gauge 5 (Below GCL)
-50 T
                                                                                  100     200     300     400    500    600    700
    Downslope
                  - Left Panel -n- Right Panel
                       Figure A.12: POST CONSTRUCTION DISPLACEMENT VS. TIME FOR PLOT L
                                         (Bentofix - Granular Drainage 1-2:1 Slope)
-------
Plot M - Erosion control plot - no instrumentation installed



Plot P - Only moisture sensors were installed - no deformation sensors
                                    D-79
-------
                          Appendix D
                         Attachment 3
Plots of Differential Displacement between Upper and Lower Surfaces of
                       GCLs Versus Time
                             D-80
-------
                                        Left Panel
CD  25
o
.a
   50

E
                                                                                        800
                                                                          Time (days)
Q  75 1
                                      Extensiometer No.
                                      Right Panel
CD  25 --
o
.a
   50

E
                                                                                        800
                                                                            Time (days)
Q  75 -L
                                      Extensiometer No.
                                        •1
                Figure C.1:  RELATIVE DISPLACEMENT VS. TIME FOR PLOT A

                           (Gundseal - Bentonite Side Up - 3:1  Slope)
                                            D-81
-------
E
E
                                         Left Panel
                                                                                 800
                                   Extensiometer No.
                                 -1 -a-2 -»-3 -0-4 -*-5
E
E
                                         Right Panel
                                    Extensiometer No.
                                 -1
              Figure C.2: RELATIVE DISPLACEMENT VS. TIME FOR PLOT B

                                (Bentomat-3:1 Slope)
                                       D-82
-------
E
o
_a
                                          Left Panel
O  25

CD
   45 --
   65 --
o


"o_



O
                                    Extensiometer No.
                                    •1 -0-2 -»-3 -0-4 -A-5
E
                                          Right Panel
                                                                       Time (days)
                                                                                     800
                                      Extensiometer No.
-•-1
-n-
2
-^
3
-o-
4
-*-
5
               Figure C.3:  RELATIVE DISPLACEMENT VS. TIME FOR PLOT C

                                  (Claymax -3:1 Slope)
                                          D-83
-------
E
                                         Left Panel
                                     Extensiometer No.
                                   •1 -0-2 -»-3 -0-4 -A-5
                                     Right Panel
                                                                      Time (days)
                                                                           700       800
                                     Extensiometer No.
                                      •1
               Figure C.4: RELATIVE DISPLACEMENT VS. TIME FOR PLOT D
                                 (Bentofix NS-3:1 Slope)
                                         D-84
-------
E
E
                                       Left Panel
                                     Extensiometer No.
                                  -1 -0-2 -»-3 -0-4 -*-5
   E
   E
                                                  Right Panel
                                                                       Time (days)
                                                                                    800
                                    Extensiometer No.
   E
   
-------
E
E
                                       Left Panel

   E
   E
Right Panel
             Figure C.6:  RELATIVE DISPLACEMENT VS. TIME FOR PLOT F

                         (Gundseal - Bentonite Up - 2:1 Slope)
                                       D-86
-------
E
E
  -75 -r
                                            Left Panel
                                                                       Time (days)


                                      80     100     120     140     160     180     200
                               Failure
-•-1
-Q-
2
-»-
3
-o-
4
-*-
5
                               60
                                                 Right Panel
80
                        Time (days)

100     120    140     160     180    200
                                                                                     H
                                   Failure
                                         Extensiometer No.
-•-1
-n-
2
-•-
3
-o-
4
-*-
5
              Figure C.7: RELATIVE DISPLACEMENT VS. TIME FOR PLOT G

                                  (Bentomat-2:1 Slope)
                                         D-87
-------
|-75-
_i
O .50 -
5
splacement Bel'
ro
o en
b
' 25 -
E
,§
d 5°-
0
m
acement Abovi
O --J
o en
Q.
Q125 -
Left Panel


)
IB
1
|



t
Time (days)
M "xlO 15 20 25 30




-A Failure
^^



Extensiometer No.
-•-1 -a-2 -»-3 -0-4 -A- 5
                              Right Panel
                               15
20
Time (days)
25            30
                     Failure
                        Extensiometer No.
                    -1
Figure C.8:  RELATIVE DISPLACEMENT VS. TIME FOR PLOT H
                   (Claymax -2:1 Slope)
                           D-88
-------
E
E
                                          Left Panel
                                                                       Time (days)
                                      Extensiometer No.
-•-1
-o-
2
-•-
3
-o-
4
-*-
5
      -100
                                                   Right Panel
                                                                                     800
-•-1
-Q-
2
-•-
3
-o-
4
-*-
5
               Figure C.9: RELATIVE DISPLACEMENT VS. TIME FOR PLOT I

                                 (BentofixNW-2:1 Slope)
                                          D-89
-------
E
E
                                              Left Panel
                                                                  Time (days)
                                                                                 800
o
JO

Q.
                                     Extensiometer No.
                           •1  -0-2 -*-3 -0-4 -A-5
E
E
         Right Panel
                                                                                 800
E
-------
E
E
                                                                              800
                              -1 -0-2 -»-3 -0-4 -A-5
E
E
                                            Right Panel
                              -1
              Figure C.11: RELATIVE DISPLACEMENT VS. TIME FOR PLOT K

                        (Claymax - Granular Drainage - 2:1 Slope)
                                       D-91
-------
    E
    E
                                                     Left Panel
                                                                      Time (days)


                                                                 600      700      800
                                 1 -0-2 -*-3 -0-4 -*-5
O  -75
CD


1  -50
CO
Right Panel
                                                                      Time (days)

                                                                600       700
                                        800

                                        H
-•-1
-n-
2
-^
3
-o-
4
-*-
5
               Figure C.12:  RELATIVE DISPLACEMENT VS. TIME FOR PLOT L
                        (Bentofix NW - Granular Drainage - 2:1 Slope)
                                          D-92
-------
E
E
                                                 Left Panel
                             1  -0-2 -»-3 -0-4 -A-5
                                                                               800
1
   -50 +
                                                Right Panel
E
o  -25 --
"o_
Q
                                                                   Time (days)
               100      200      300      400      500      600       700
800
                                    Extensiometer No.
-•-1
-Q-
2
-^
3
-o-
4
-*-
5
           Figure C.13:  RELATIVE DISPLACEMENT VS. TIME FOR PLOT N
                             (BentofixNS-2:1 Slope)
                                      D-93
-------
                Appendix D
               Attachment 4
Plots of Moisture Sensor Readings Versus Time
                   D-94
-------
                  Plot Location: Crest
  100
 200
300     400     500
    Time (days)
600
700     800
  100     200     300     400     500     600     700     800
                      Time (days)
  100
 200     300     400
-Subsoil-GYPSUM
-w/inGCL- FIBERGLASS
               500     600     700     800
               -A-GCL/Subsoil - FIBERGLASS
Figure D1: MOISTURE READINGS VS. TIME FOR PLOT A
         (Gundseal - Bentonite up - 3:1 Slope)
                          D-95
-------
                     Plot Location: Crest
   100
200
300     400     500
    Time (days)
600
700
800
                    Plot Location: Middle
   100     200     300     400     500     600     700     800
                        Time (days)
                    Plot Location: Toe
   100
200
300     400     500
     Time (days)
600
700
800
       -Subsoil-GYPSUM
                 -GCL/Subsoil - FIBERGLASS
Figure D.2:  MOISTURE READINGS VS. TIME FOR PLOT B
                (Bentomat-3:1 Slope)
                            D-96
-------
                     Plot Location: Crest
   100     200     300     400     500     600     700     800
                       Time (days)
                     Plot Location: Middle
    100
200
300     400    500
    Time (days)
600
700
800
    100
200
300     400    500
    Time (days)
600
700
800
      -Subsoil-GYPSUM
                -GCL/Subsoil - FIBERGLASS
Figure D.3: MOISTURE READINGS VS. TIME FOR PLOT C
                (Claymax-3:1 Slope)
                           D-97
-------
                     Plot Location: Crest
   100     200
       300     400     500
            Time (days)
600
700
                                                 800
    100     200     300    400     500
                       Time (days)
                              600
        700
        800
    100
200
600
700
           300     400    500
               Time (days)
Subsoil - GYPSUM -A-GCL/Subsoil - FIBERGLASS
800
Figure D.4: MOISTURE READINGS VS. TIME FOR PLOT D
               (Bentofix NS-3:1 Slope)
                           D-98
-------
                    Plot Location: Crest
   100     200     300     400     500
                       Time (days)
600     700
800
   100     200     300     400     500     600     700     800
                       Time (days)
    100     200     300    400     500     600     700     800
                       Time (days)
  -•-Subsoil - GYPSUM -A-GCL/Subsoil - FIBERGLASS
Figure D.5: MOISTURE READINGS VS. TIME FOR PLOT E
      (Gundseal - Bentonite Side Down - 3:1 Slope)
                           D-99
-------
                    Plot Location: Crest
100
200       300
       Time (days)
                           400
                                               500
         600
100      200       300       400
               Time (days)
                                               500
                                        600
             100
            200
              300
           Time (days)
                                 400
500
600
       -Subsoil-GYPSUM
       -w/inGCL- FIBERGLASS
                          •GCL/Subsoil - FIBERGLASS
Figure D.6: MOISTURE READINGS VS. TIME FOR PLOT F
          (Gundseal - Bentonite up - 2:1 Slope)
                          D-100
-------
100.0 n
90.0 -
80.0 -
8 60.0 -
I 50.0-
1 40.0-
° 30.0 -
20.0 -
10.0 -
0.0 -
C
Plot Location: Crest
1

;
i
1

f
,/
-•^w-
7
) 100 200 300 400 500
Time (days)
100 n
o) 90 -
.<= 80 -
Hi 70 -
& 6°-
« 50-
§ 40-
to 30 -
o 20 -
5 10 -
n -
Plot Location: Middle
I




-I
^


Una
0



J/
/
f


^.^AAA I
100 200 300 400 500
Time (days)
                     Plot Location: Toe
         100
 Subsoil - GYPSUM
200         300
   Time (days)
                      400
500
GCL/Subsoil - FIBERGLASS
Figure D.7:  MOISTURE READINGS VS. TIME FOR PLOT G
               (Bentomat-2:1 Slope)
                          D-101
-------
Plot Location: Crest
100.0 -,
90.0 -
80.0 -
1 70'°-
8 60.0 -
I 50.0-
1 40.0 -
| 30.0 -
20.0 -
10.0 -
0.0 -
C

\
\
k



\
A I
) 100 200 300 400 500
Time (days)
   0)
   a:


   I
     100 -r
      90
      80 --
      70 -

      50 -
      40 -
      30 -
      20 --.
      10 ^
       0
         0
                                Plot Location: Middle
100        200         300

              Time (days)
                                                      400
500
   100 T
    90
a)  80 --
i   70 -
S   60-
«   50
3   40 -
    30 -
    20 -
    10 -
     0 --
•§
                              Plot Location: Toe
       0          100
          Subsoil - GYPSUM
          200         300

             Time (days)
                                                      400
500
                               GCL/Subsoil - FIBERGLASS
        Figure D.8:  MOISTURE READINGS VS. TIME FOR PLOT H
                         (Claymax-2:1 Slope)
                                   D-102
-------
                     Plot Location: Crest
   100
200
300     400     500     600
    Time (days)
        700
        800
                        Plot Location: Middle
   100      200     300     400     500
                       Time (days)
                                600
                                700
                800
     100
 200
 300    400     500
     Time (days)
600
700
800
 -Subsoil-GYPSUM
           -GCL/Subsoil - FIBERGLASS
Figure D.9: MOISTURE READINGS VS. TIME FOR PLOT I
               (Bentofix NW-2:1 Slope)
                           D-103
-------
                      Plot Location: Crest
    100
200
300     400     500
     Time (days)
 600
700
 800
                        lot Location: Middle
    100
200
300     400     500
    Time (days)
600
700
800
                       Plot Location: Toe
      100
  200
 300     400     500
      Time (days)
 600
700
800
   -Subsoil-GYPSUM
            -GCL/Subsoil - FIBERGLASS
Figure D.10:  MOISTURE READINGS VS. TIME FOR PLOT J
        (Bentomat - Granular Drainage - 2:1 Slope)
                            D-104
-------
                          Plot Location: Crest
       100
200
300    400     500
    Time (days)
600
700
800
                             Plot Location: Middle
       100     200     300     400     500    600     700
                          Time (days)
                                             800
                        Plot Location: Toe
      100     200     300     400     500    600     700     800
                         Time (days)
-•-Subsoil - GYPSUM -A-GCL/Subsoil - FIBERGLASS

 Figure D.11:  MOISTURE READINGS VS. TIME FOR PLOT K
         (Claymax - Granular Drainage - 2:1 Slope)
                            D-105
-------
                        Plot Location: Crest
     100     200     300    400     500
                         Time (days)
                                600
                               700
                800
                          Plot Location: Middle
     100
200
300     400     500
    Time (days)
600
700
800
                      Plot Location: Toe
    100     200     300     400     500     600     700     800
                        Time (days)
   -•-Subsoil - GYPSUM -A-GCL/Subsoil - FIBERGLASS
Figure D.12:  MOISTURE READINGS VS. TIME FOR PLOT L
       (Bentofix NW - Granular Drainage - 2:1 Slope)
                            D-106
-------
                         Plot Location: Crest
     100     200     300     400    500
                         Time (days)
                                600
        700
        800
                          Plot Location: Middle
       100    200     300    400     500
                          Time (days)
                                600
        700
       800
                      Plot Location: Toe
    100
200
600
700
            300     400     500
                Time (days)
Subsoil - GYPSUM -A-GCL/Subsoil - FIBERGLASS
800
Figure D.12:  MOISTURE READINGS VS. TIME FOR PLOT L
       (Bentofix NW - Granular Drainage - 2:1 Slope)
                            D-107
-------



D)
~
01
0
0

(ft
'o
5



Plot Location Crest
100 -,
90
80
70 -
60
50
40-
30
20
10-
Oj









~P— ••__
0 100 200 300 400 500 600
Time (days)
100 T
 90 -
 80 -
 70 -
 60 -
 50 -
 40 -
 30 -
 20 -
 10
  0
              100
                        200
                                  300

                               Time (days)
                                             400
                                                       500
                                                                  600
              100
                        200
                                  300

                               Time (days)
                                             400
                                                       500
                                                                  600
100 T
 90 --
 80 --
 70 --
 60 --
 50 --
 40 --
 30 --
 20 --
 10 ^
  0
             100        200        300        400

                               Time (days)
                                                       500
                                                                  600
      - inner left panel •
-outer left panel   • inner right panel —o— outer right panel
 Figure D.14: MOISTURE READINGS VS. TIME FOR PLOT P
          (Gundseal - Bentonite side up - 2:1 Slope)
                                     D-108
-------
                         Appendix D
                        Attachment 5
Results of Laboratory Direct Shear Tests Performed on 64-mm-Wide
          Specimens in University of Texas Laboratories

               (Results Described in Section D-7)
                            D-109
-------
      20
      15 --
                            Shear Rate = 1 mm/min

                      24 hour Hydration/17 kPa Normal Stress
 CO
 a.
 CD
 0)
 .c
 CO
                                                          Test No. 1

                                                          Test No. 2

                                                          Test No. 3
         02468

                              Displacement (mm)

    Figure 1:  Shear Stress vs.  Displacement for Test No. 1, 2, and 3
     20
CD
CL
 _
 CD
 0>
.c
CO
     15 --
     10 --
      5 --
                            Shear Rate = 1 mm/min

                     48 hr Hydration/17 kPa Normal Stress
Test No. 4

Test No. 5

Test No. 6
         02468

                              Displacement (mm)

    Figure 2: Shear Stress vs. Displacement for Test No. 4, 5, and 6
                                  D-110
-------
CO
0.
0)
-I—I
CO

CD
CD

CO
CD
CL
CD
-i— i
CO
L_
CD
CD

CO
     20
     15 --
£    10 --
      5 --
                             Shear Rate = 1 mm/min
                      72 hr Hydration/17 kPa Normal Stress
                                                         Test No. 7
                                                         Test No. 8
                                                         Test No. 9
         02468
                              Displacement (mm)

     Figure 3:  Shear Stress vs. Displacement for Test No. 7, 8, and 9
     20
                             Shear Rate = 1 mm/min
                     153 hr Hydration/17 kPa Normal Stress
     15 -•
                            468
                              Displacement (mm)
                                                         10
12
         Figure 4:  Shear Stress vs. Displacement for Test No. 23
                                   D-111
-------
 CO
 0.
 £
 -I— i
 CO
 CD
 0>
 .c
 CO
CD
CL
E
-i— i
CO
L_
CD
0>
.c
CO
       20
       15 --
10 --
        5 --
                              Shear Rate = 0.1 mm/min
                       24 hr Hydration/17 kPa Normal Stress
        0
                                                  Test No. 14
                                                  Test No. 15
                                                  Test No. 19
           0
                                                              8
                                Displacement (mm)
    Figure 5: Shear Stress vs. Displacement for Test No. 14, 15, and 19
      20
      15 --
      10 --
       5 --
                            Shear Rate = 0.02 mm/min
                       24 hr Hydration/17 kPa Normal Stress
                                                        Test No. 20
                                                        Test No. 21
                                                        Test No. 22
                                                                     8
                               Displacement (mm)
    Figure 6: Shear Stress vs. Displacement for Test No. 20, 21, and 22
                                    D-112
-------
CO
0.
£
-I— i
CO

CD
0>
.c
CO
      20
15 --
10 --
       5 --
       0
                          Shear Rate = 0.0025 mm/min

                      24 hr Hydration/17 kPa Normal Stress
                                                      Test No. 24
                                                      Test No. 25
                                                      Test No. 26
          0
                                                              8
                               Displacement (mm)

   Figure 7: Shear Stress vs. Displacement for Test No. 24, 25, and 26
CD
Q_
£
CO

CD
0)
.c
CO
      20 T
      15 -•
10 --
       5 --
                           Shear Rate = 0.0005 mm/min

                       24 hr Hydration/17 kPa Normal Stress
                                                    Test No. 27
                                                    Test No. 29
         0
                                                              8
                              Displacement (mm)

     Figure 8: Shear Stress vs. Displacement for Test No. 27 and 29
                                   D-113
-------
      100
CO
Q_
CO
L_
CD
0>

CO
 CD
 Q_
 £
 CO

 CD
 0)
 .c
 CO
                               Shear Rate = 1 mm/min

                        24 hr Hydration/172 kPa Normal Stress
                                                        Test No. 37
                                                        Test No. 38
                                       468
                               Displacement (mm)

             9: Shear Stress vs. Displacement for Test No. 37 and 38
                           Shear Rate = 0.024 mm/min

                      24 hr Hydration/172 kPa Normal Stress
                                                        a—^—n
                                                    Test No. 32
                                                    Test No. 36
           012345
                                Displacement (mm)

      Figure 10:  Shear Stress vs. Displacement for Test No. 32 and 36
                                   D-114
-------
CO
a.
 _
 CD
 0)
.c
CO
100


 90 -


 80 -


 70 -


 60 -


 50 -


 40 -


 30 -


 20 -


 10 -


   0
          0
                          Shear Rate = 0.0005 mm/min

                     24 hr Hydration/172 kPa Normal Stress
                                                    Test No. 28

                                                    Test No. 30
                        2         3
                         Displacement (mm)
    Figure 11: Shear Stress vs. Displacement for Test No. 28 and 30
                                  D-115
-------
 CO
 0.
 (to
 CD
 -I— I
 CO

 CD
 0>

 CO
CD
CL
£

CO
l_
CD
0>
.c
CO
         20
         15  --
  10 --
                                Shear Rate = 1 mm/min

                         24 hr Hydration/17 kPa Normal Stress
          •  Test No. 1: peak


          D  Test No. 2: peak


          O  Test No. 3: peak
           5  --
             •regression through
             origin
                                           10
                                                   15
                                                          20
                                  Normal Stress (kPa)

         Figure 12:  Mohr-Coulomb Diagram for Test No. 1, 2, and 3
       20
                                 Shear Rate = 1 mm/min
                          48 hr Hydration/17 kPa Normal Stress
       15  --
10 --
        5  --
        0
•  Test No. 4: peak


D  Test No. 5: peak


O  Test No. 6: peak
           'regression through
           origin
           0               5               10              15             20

                                  Normal Stress (psi)

         Figure 13:  Mohr-Coulomb Diagram for Test No. 4, 5, and 6
                                      D-116
-------
CO
0.
£

CO
l_
CD
0>
.c
CO
  CD
  Q_
  £
  -i— i
  CO

  CD
  0>
  .c
  CO
       20
       15  --
       10  --
 5 --
                               Shear Rate = 1 mm/min

                        72 hr Hydration/17 kPa Normal Stress
        •  Test No. 7: peak


        D  Test No. 8: peak


        O  Test No. 9: peak
           regression through
           origin
                                         10
                                                 15
20
                                 Normal Stress (kPa)

          Figure 14: Mohr-Coulomb Diagram for Test No. 7, 8, and 9
       20
        15  --
10 -•
         5 --
                                 Shear Rate = 1 mm/min

                         153 hr Hydration/17 kPa Normal Stress
            Test No. 23: peak


           -regression through
            origin
                                         10
                                                 15
20
                                 Normal Stress (kPa)

              Figure 15:  Mohr-Coulomb Diagram for Test No. 23
                                      D-117
-------
CO
a.
CD
0)
.c
CO
 CD
 CL
 £

 CO
 l_
 CD
 0>
 .c
 CO
      20
      15 --
      10 --
       5 --
                              Shear Rate = 0.1 mm/min

                        24 hr Hydration/17 kPa Normal Stress
       •  Test No. 14: peak


       D  Test No. 15: peak


       O  Test No. 19: peak
                 •regression through

                 origin
          0              5              10             15              20

                                Normal Stress (kPa)

      Figure 16: Mohr-Coulomb Diagram for Test No. 14, 15, and 19
       20
       15 --
10 --
        5 --
        0
                            Shear Rate = 0.02 mm/min

                       24 hr Hydration/17 kPa Normal Stress
•  Test No. 20: peak


D  Test No. 21: peak


O  Test No. 22: peak
          •regression through

           origin
                                        10
                                                15
                                                       20
                                Normal Stress (kPa)

      Figure 17: Mohr-Coulomb Diagram for Test No. 20, 21, and 22
                                     D-118
-------
CO
0.
(to
CD
-I— I
CO
L_
CD
CD

CO
CD
CL
(to
CD
-i— i
CO
L_
CD
CD

CO
       20
       15  --
       10  --
        5  --
                             Shear Rate = 0.0025 mm/min

                        24 hr Hydration/17 kPa Normal Stress
         •  Test No. 24: peak


        	regression through
            origin
         D  Test No. 25: peak


         O  Test No. 26: peak
           0
                                   10
15
20
                                Normal Stress (kPa)

      Figure 18: Mohr-Coulomb Diagram for Test No. 24, 25, and 26
      20
      15 --
10 --
       5 --
       0
                            Shear Rate = 0.0005 mm/min

                       24 hr Hydration/17 kPa Normal Stress
           regression through
           origin
           Test No. 27: peak
        D  Test No. 29: peak
                                        10
                                                 15
               20
                               Normal Stress (kPa)

        Figure 19: Mohr-Coulomb Diagram for Test No. 27 and 29
                                     D-119
-------
CO
a.
 _
 CD
 0)
.c
CO
                                 Shear Rate = 1 mm/min

                        24 hr Hydration/172 kPa Normal Stress
            regression through
            origin

            Test No. 37: peak
               D  Test No. 38: peak
                                             d> = 15.7 degrees
          0
                   50
                                   150
                               100

                       Normal Stress (kPa)

Figure 20:  Mohr-Coulomb Diagram for Test No. 37 and 38
200
CD
Q_
£

CO
l_
CD
0>
.c
CO
100  --


 90  --


 80  --


 70  --


 60  --


 50  --


 40  --


 30  --


 20  --


 10  --
        0
          0
                            Shear Rate = 0.024 mm/min

                       24 hr Hydration/172 kPa Normal Stress
regression through
origin

Test No. 32: peak
               D  Test No. 36: peak
                   50
                                   150
                               100

                        Normal Stress (kPa)

Figure 21:  Mohr-Coulomb Diagram for Test No. 32 and 36
200
                                   D-120
-------
     100
CO
a.
CD
0)
.c
CO
                          Shear Rate = 0.0005 mm/min

                    24 hr Hydration/172 kPa Normal Stress
                 •  Test No. 28: peak


                 D  Test No. 30: peak
regression through

origin
                                                    = 13.0 degrees
                                      100

                              Normal Stress (kPa)

       Figure 22:  Mohr-Coulomb Diagram for Test No. 28 and 30
                                   D-121
-------
                     Appendix E
Evaluation of Liquids Management Data
          for  Double-Lined Landfills
                             by

        Majdi A. Othman, Ph.D. P.E.           Rudolph Bonaparte, Ph.D. P.E.
        GeoSyntec Consultants             GeoSyntec Consultants
        Atlanta, Georgia 30342             Atlanta, Georgia 30342

        Beth A. Gross, P.E.                Dave Warren
        GeoSyntec Consultants             GeoSyntec Consultants
        Austin, Texas 70748               Atlanta, Georgia 30342
                        performed under

                 EPA Cooperative Agreement Number
                        CR-821448-01-0
                         Project Officer

                       Mr. David A. Carson
             United States Environmental Protection Agency
                 Office of Research and Development
             National Risk Management Research Laboratory
                      Cincinnati, OH 45268
-------
                                 Appendix E
  Evaluation of Liquids Management Data for Double-Lined Landfills


E-1    Introduction

E-1.1    Purpose and Scope of Appendix

The purpose of this appendix is to summarize and analyze liquids management data for
modern double-lined landfills located throughout the United States (U.S.).  Specifically,
leachate collection and removal system (LCRS) and leak detection system (LDS) flow
rate and flow chemistry data are presented for 189 cells at 54 municipal solid waste
(MSW), hazardous waste (HW), and industrial solid waste (ISW) landfills. These data
are used to evaluate:

   •   leakage rates and hydraulic efficiencies of landfill primary liners;
   •   landfill leachate generation rates (LCRS flow rates), including how leachate
       generation rates vary with waste type, geographic location (climate), and
       presence of final cover system; and
   •   landfill leachate chemistry  (LCRS flow chemistry), including how leachate
       chemistry varies with waste type, geographic location, and operation conditions,
       and whether federal solid waste regulations promulgated in the 1980's and early
       1990's have had an effect on the quantity of potentially-toxic trace chemicals
       found in leachate.

In addition to the field data presented herein, this appendix presents a literature review
summarizing significant previous work related to this study.

E-1.2    Organization of Appendix

The organization of this appendix  is as follows:

   •   a review of significant previously published work on primary liner performance
       and landfill leachate generation rates and chemistry is presented in Subsection
       E-2;
   •   a description of the data collected for this study and data reduction methods are
       presented in Subsection E-3;
   •   a summary and evaluation of data on leakage through landfill primary liners are
       presented in Subsection E-4;
   •   a summary and evaluation of landfill leachate generation rate data are presented
       in Subsection E-5;
   •   a summary and evaluation of landfill leachate chemistry data are presented in
       Subsection E-6;  and
   •   general conclusions are presented in Subsection E-7.
                                   E-1
-------
E-1.3    Definitions

E-1.3.1   Landfills

Landfills are land-based waste management cells that contain solid wastes. Waste
containment systems for landfills consist of liner systems that underlay the wastes
placed on them and final cover systems constructed over the wastes.  The goal of the
liner systems is to minimize, to the extent achievable, the migration of waste
constituents out of the landfills.  The goal of the final cover systems is to contain the
wastes, minimize, to the extent achievable, the percolation of water into the landfills,
and control the migration of gases, if any, from the wastes.

E-1.3.2  Liner, Liner System, and Double-Liner System

A liner is  a low-permeability barrier used to impede liquid or gas flow.  As discussed in
Giroud (1984) and U.S. Environmental Protection Agency (EPA) (1987a), no currently
available liner is totally impermeable. Since no liner is impermeable, liquid containment
within a landfill cell can only result from a combination of liners and drainage layers
performing complementary functions.  Liners impede leachate percolation and gas
migration out of the cell and improve the collection capability of overlying drainage
layers. Drainage layers collect and convey liquids on liners towards controlled collection
points (sumps) where the liquids can be removed from the cell. Drainage layers limit
the build  up of hydraulic head on underlying liners. Combinations of liners and drainage
layers in the cells are called liner systems.

A double-liner system consists of a primary liner and a secondary liner with an LDS
between the primary and secondary liners and an LCRS above the primary liner.
Essentially all landfill double-liner systems being constructed today have liners that
include geomembranes (GMs).  These liners can consist of a GM alone, GM on top of a
compacted clay liner (CCL), or GM on top of a geosynthetic clay liner (GCL). The latter
two liners are both referred to as "composite" liners. Only landfills with GM, GM/CCL
composite, or GM/GCL composite primary liners were considered  in this appendix.  In
addition, with the exception of six cells, all of the landfills considered herein have GM or
composite secondary liners. Older liner systems constructed with  CCL primary liners
were not  considered.

E-1.3.3  Double-Liner System  Components and Groups

Figure E-1.1 illustrates the double-liner system types considered in this appendix.  The
two main differences between the double-liner systems shown in Figure E-1.1 are the
primary liner type (GM, GM/CCL composite, or GM/GCL composite) and LDS drainage
layer type (granular material or geonet (GN)). The types of liners and drainage
materials used in double-liner systems significantly influence the frequencies of
occurrence, sources, and rates of flow from  LDSs. For purposes of this appendix, the
considered liner systems are grouped  into "Type I" through "Type VI" based on primary
liner and  LDS materials, as defined in Table E-1.1 and illustrated in Figure E-1.1.
                                   E-2
-------
Figure E-1.1. Liner System Types Considered in this Appendix.
                        E-3
-------
Table E-1.1.  Definitions of Liner System Types Considered in this Appendix.
Liner System
Type
I
II
III
IV
V
VI
Primary Liner
GM
Composite GM/CCL
or GM/GCL/CCL
Composite GM/GCL
LDS Material
Granular
GN
Granular
GN
Granular
GN
E-1.3.4   Cover, Daily Cover, Intermediate Cover, and Final Cover System

A cover is a barrier placed over material to isolate the material from the surrounding
environment.  Landfills require daily covers, intermediate covers, or final cover systems,
depending on the stage of landfill development.

At most landfills, a daily cover is applied to waste to control the spread of insects,
rodents, and other burrowing animals, which may use the waste as a food source and to
prevent the erosion of waste by wind and surface-water runoff. Daily cover usually
consists of a layer of soil, select waste, or other material such as foam or fabric.
Intermediate cover is often placed on open portions of landfill areas on which waste
placement has ceased, either permanently or for an extended period of time.
Intermediate cover serves the same purposes as daily cover, but at a higher level. It
usually consists of a thicker layer of soil or select waste than daily cover, and may
include a temporary GM.

As the active period of operation progresses, the landfill cell is filled with waste, and
waste placement ceases. Depending on the landfill, cells may be under intermediate
cover for up to several years before a final cover system is constructed over the waste.
Final cover systems are the only engineered covers on landfill  cells.   They serve the
same purposes as daily and intermediate covers, but are also designed to minimize
water infiltration into the waste (i.e., leachate generation), control the migration  of gases
produced by waste decomposition, and be aesthetically acceptable.   As shown in Figure
E-1.2, most final cover systems constructed today contain (from top  to bottom)  a
vegetated topsoil surface layer, soil protection layer, drainage layer,  barrier, gas
collection layer (at landfills with wastes that generate gases during decomposition), and
a foundation layer (if needed). Final cover systems considered in this appendix
primarily have a GM or composite barrier; however, ten  cells have a CCL barrier over
the entire landfill or the landfill side slopes.
                                   E-4
-------
COMPOSITE
BARRIER  •
     Figure E-1.2. Typical Final Cover System fora Modern Landfill.
                                E-5
-------
E-1.3.5  Waste Types in Landfills

The landfill cells in this study are grouped into three categories based on the
predominant waste type in a cell:

    •   MSW;
    •   HW;
    •   construction and demolition waste (C&DW);
    •   ash from MSW combustors (MSW ash); and
    •   ash from coal-burning power plants (coal ash).

C&DW and coal ash are ISWs.

E-1.3.6  Regions of the United States

The landfills in this study are grouped into three different geographic regions of the U.S.:
Northeast U.S. (NE), Southeast U.S. (SE), and West U.S. (W). These regions are
outlined in Figure E-1.3 and were chosen because of the climatic differences between
these regions (i.e., differences in average annual rainfall, potential evapotranspiration,
and days below freezing).

Generally, facilities in the SE receive relatively high rainfall, have relatively high
evapotranspiration, and experience few days below freezing annually.  Compared to the
SE facilities, those in the NE receive slightly lower rainfall, have lower
evapotranspiration, and experience a significant number of days below freezing
annually. Except for facilities near the northwest coast of the U.S., facilities in the W are
in relatively arid climates, with relatively low precipitation and relatively high
evapotranspiration, and may or may not experience a significant amount of days below
freezing annually (in arid climates this does not markedly affect leachate generation
rates).

The climatic differences  between regions will have a much larger impact on leachate
generation  rates at landfills than on LDS flow rates at landfills.  LDS flow rates can be
affected by climatic differences (higher LCRS flow rates means greater potential for
primary liner leakage), but LDS flow rates usually depend more on liner system
construction and performance than on climate.

E-1.3.7  LCRS Operational Stages

LCRS flow data for landfills are grouped into three different development stages  in the
life cycle of a landfill cell: (i) initial period of operation; (ii) active period of operation; and
(iii) post-closure period.  These stages are defined by characteristics in the LCRS flow
rates, as described below and shown for a MSW landfill in Pennsylvania in Figure E-1.4.
                                    E-6
-------
m
             Figure E-1.3. Geographic Regions and Locations of Landfills Evaluated in this Study.
-------
                 Initial Period

                 of Operation
                              Active Period of Operation
   160
   140 -
   120 -

T3~

•§. 100 J
O   80-
_i
LL
cn   60 H
Q

    40 -


    20 -


    0
        CD
        t
      oo
      oo
o>
oo
o>     o
oo     o
o
O)
                                        O)

                                        "5
CM
O)
CM
O)
CO
o
CO
O)
                                                        o>
                                                         c
                                                         (0
LO
O)
LO
O)
                                            DATE
Figure E-1.4.  LCRS and LDS flow rates over time at a MSW landfill in Pennsylvania.
                                     E-8
-------
The initial period of operation occurs during the first few months after the start of waste
disposal in  a cell.  During this stage, there is not sufficient waste in a cell to significantly
impede the flow of rainfall into the LCRS. To the extent rainfall occurs during this period,
it will rapidly find its way into the LCRS and, unless the LCRS drainage layer has a
relatively low permeability (i.e., less than 10~4 mis), to the LCRS sump. LCRS flow rates
during this stage are usually controlled by rainfall and can be directly correlated to local
climatic conditions.  LCRS flow rates are higher at landfills in wetter climates than at
those in arid climates. As such, LCRS flow rates during this stage are usually much
larger than  in later stages and vary widely with the amount of rainfall received.

During the active period of operation, the cell is progressively filled with waste and daily
and intermediate layers of cover soil. As waste placement continues, more of the
rainfall occurring during this stage falls onto the waste and cover soils rather than
directly onto the liner system. As a consequence, the LCRS flow rates decrease and
eventually stabilize. LCRS  flow rates during this stage are generally dependent on
rainfall quantity, waste thickness, waste properties (i.e., initial moisture content, field
capacity, and permeability), and storm-water management practices. Additional waste
reduces LCRS flow rates in two ways:

   •   it increases the total  storage capacity of water within waste in the landfill cell; and
   •   precipitation falling on cover soil is often directed out of the landfill as surface-
       water runoff.

During the post-closure period, the cell has been closed with a final cover system that
further reduces  infiltration of rainwater into the waste, resulting  in a further reduction in
LCRS flow rates.

E-1.3.8  LDS Operational  Stages

Similar to LCRS flow data, LDS flow data are also grouped into the same stages in the
life cycle of a landfill cell: (i) initial period of operation, (ii) active period of operation,
and (iii) post-closure period. During the initial period of operation, most of the LDS flow
is usually due to "construction water". After the active period of operation of a facility,
construction water has substantially drained (unless the LDS has a relatively  low
permeability, i.e.,  less than  1x10"4 m/s) and other sources of  LDS flow have a stronger
influence.  Sources of LDS  flow are defined in Section E-2.1.1. For simplicity in this
appendix, the LCRS and LDS operational stages for a landfill cell are assumed to occur
over the same time period.
                                    E-9
-------
E-2    Literature Review

E-2.1    Field Performance of Primary Liners

E-2 A A   Overview

The performance of primary liners at double-lined landfills and surface impoundments
has previously been assessed by comparing LDS and LCRS flow rate and chemical
constituent data. The first general study of this type, by Gross et al. (1990), identified
five potential sources of LDS flow (Figure E-2.1): (i) leakage through the primary liner;
(ii) water (mostly rainwater) that infiltrates the LDS during construction and continues to
drain to the LDS sump after the start of facility operation ("construction water"); (iii)
water that infiltrates the LDS during construction, is held in the LDS by capillary tension,
and is expelled from the LDS during waste placement as a result of LDS compression
under the weight of the waste ("compression water");  (iv) water expelled into the LDS
from the CCL and/or GCL components of a composite primary liner as a result of clay
consolidation under the weight  of the waste ("consolidation water"); and (v) water that
percolates through the secondary liner and infiltrates the LDS ("infiltration water").

Gross et al. (1990) presented the following five-step approach for evaluating the
sources of LDS liquid at a specific waste management cell:

    •  identify the potential  sources of flow for the cell based on double-liner system
       design, climatic and hydrogeologic setting, and cell operating history;
    •  calculate flow rates from each potential source;
    •  calculate the time frame for flow from each potential source;
    •  evaluate the potential sources of flow by comparing measured flow rates to
       calculated flow rates at  specific points in time; and
    •  compare LCRS and LDS flow chemistry data to further establish the likely
       source(s) of liquid.

Previously published studies which  evaluated the field performance of primary liners in
double-lined landfills using LCRS and LDS flow data of are reviewed in this section  of
the appendix, and the conclusions drawn from the studies are presented. The
remainder of this section is organized as follows:

    •  studies of landfills  with GM primary liners are reviewed in Section E-2.1.2; and
    •  studies of landfills  with GM/CCL and GM/GCL primary liners are presented in
       Section E-2.1.3.

It is noted that the field performance of liners can also be derived from leak location
surveys. Several authors (i.e.,  Darileket al., 1989;  Landreth, 1989; and Peggs, 1990,
1993) have described the use of the electrical leak location method for finding holes in
GM liners.  Darilek et al. (1989), Laine and Miklas (1989), Laine (1991), Laine and
Darilek (1993) and Darilek et al. (1995) reported on the  results of electrical leak location
                                   E-10
-------
surveys at GM-lined waste containment cells.  These results suggest that while GM
liners can be constructed with very low hole frequencies (e.g., less than 5 small holes
per 10,000 m2), a number of facilities have been constructed to lower standards
resulting in a significantly higher hole frequency, typically as a result of inadequate
seaming.

E-2.1.2   GM Primary Liners

E-2.1.2.1 Bonaparte and Gross (1990, 1993)

Bonaparte and Gross (1990, 1993) presented  data on LDS flows from several double-
lined landfill cells with GM primary liners; the later U.S. Environmental Protection
Agency (EPA) sponsored report is an expansion of the 1990 paper. Bonaparte and
Gross (1993) presented data for ten landfills containing 25 individually monitored  cells.
Eighteen of the cells had granular LDSs and seven had GN  LDSs. The facilities are
located  in the Eastern, North-Central, Gulf Coast, and Midwest U.S. Bonaparte and
Gross (1993) presented LDS flow rate data for two different time periods: (i) the initial
period of cell filling, just after the end of construction; and (ii) the active period of cell
filling. They concluded that all landfill cells in their study exhibited LDS flow attributable
to primary liner leakage. LDS flow rates averaged over the entire monitored portion of
the cell during the active period of cell filling (i.e., LDS flows excluded during the initial
period of cell filling) showed that about 40% of the cells with formal Construction Quality
Assurance (CQA) programs had LDS flow rates below 50 liters/hectare/day (Iphd), 80%
below 200 Iphd, and all had LDS flow rates below 1,000  Iphd.  In contrast, 70% of the
landfill cells constructed with no formal CQA programs had average active-life LDS flow
rates in  excess of 1,000 Iphd. The landfill cells with no formal CQA programs would not
be considered to be constructed to the current standard  of practice.

E-2.1.2.2 Mauleetal.  (1993)

Maule et al. (1993)  reported the results of LDS flow rate measurements for three  upper
midwest landfills, constructed during 1990 and 1991 with formal CQA programs.  Each
landfill has a GM primary liner and a GN LDS.  The three landfills had average LDS flow
rates of 10, 11, and 21 Iphd over the monitoring periods  of up  to 17 months.  Maximum
monthly LDS flow rates during the monitoring periods were 31, 24, and 66 Iphd.  These
LDS flow rates are all very low, at the lower end of the range reported by Bonaparte and
Gross (1993) for GN LDSs underlying GM liners (i.e., during the active  life period, 0 to
220 Iphd).

E-2.1.2.3 Tedder (1997)

Tedder (1997) presented LCRS and LDS flow  rate data for 22 cells at eight double-lined
landfills located in Florida. All of the cells  were constructed with GM primary liners and
granular or GN LDSs under formal CQA programs.  Data were reported for monitoring
periods  of 3 to 64 months. Tedder did not specify when  cell monitoring was performed
                                   E-11
-------
relative to the start of cell operations. Average LCRS flow rates ranged from 1,100 to
91,000 Iphd, while average LDS flow rates ranged from 14 to 4,300 Iphd.  Average LDS
flow rates were below 50 Iphd for 5 cells (23% of total number of cells), below 200 Iphd
for 13 cells  (59%), and above 1,000 Iphd for three cells (14%).  These average rates are
higher than those reported by Bonaparte and Gross (1990, 1993) for GM  liners with
CQA. The higher average LDS flow rates in the Tedder study in comparison to the
earlier studies may primarily result from the combining of data from the initial period of
cell operation (when LDS flow may be in large part attributed to sources other than
primary liner leakage) with data from the period of active cell filling. The higher LDS
flow rates may also be due to the relatively high LCRS flow rates that occurred in these
landfills.

E-2.12.4 Conclusions from Previous Studies

Conclusions drawn from the previous studies regarding the hydraulic performance of
GM liners are as follows:

    •  for landfills with GM primary liners, LDS flows during the active period of cell
       filling is primarily due to primary liner leakage;
    •  for landfills with GM primary liners constructed using formal CQA programs,
      average LDS flow rates are typically below 200 Iphd  and are rarely above 1,000
       Iphd; and
    •  average LDS flow rates from landfills with GM primary liners constructed without
      formal CQA are higher than for landfills with GM primary liners constructed with
      CQA.

E-2.1.3   Composite Primary Liners

E-2.1.3.1 Bonaparte and Gross (1990,  1993)

Bonaparte and Gross (1990, 1993) presented data on LDS flows from 51  cells at 18
double-lined landfills with composite primary liners; the later 1993 EPA-sponsored
report is an expansion of the 1990 paper.  The conclusions from their study for these
cells are as follows:

    •  double-lined landfills with GM/CCL composite primary liners almost always
      exhibited LDS flows due to consolidation water;  measured flow rates attributable
      to consolidation water were in the range of 0 to 1300 Iphd;
    •   LDS flows from landfills with GM/GCL composite primary liners were relatively
       low and ranged from 0 to 120 Iphd during active filling;
    •  the calculation methods presented by Gross et al.  (1990) for estimating
      consolidation water and construction water flow rates appear reasonable for the
      facilities  reported in this appendix.
                                  E-12
-------
Due to the masking effects of consolidation water and the limited available data,
Bonaparte and Gross were unable to quantify primary liner leakage rates. The authors
did conclude "the double-liner systems evaluated in this study have performed well.
Leakage rates through the primary liners have been low or negligible in most cases."

E-2.1.3.2 Feeney and Maxson (1993)

Feeney and Maxson (1993) used a methodology similar to that of Bonaparte and Gross
(1990) to evaluate LDS flows from 49 double-lined cells at eight HW landfills. The
landfills are located in humid, arid, and semi-arid regions of the U.S. All but two of the
cells have  a GM/CCL composite primary liner on the cell base and a GM primary liner
on the cell side slopes (i.e., Category 1 liner system). Two cells have a GM/GCL
composite primary liners on both the cell base and side slopes (i.e., Category 2 liner
system). All of the cells contain a granular soil/GN LDS on the cell base and a GN LDS
on the cell side slopes. All cells were constructed using third-party CQA programs.

For each landfill cell in their study, Feeney and Maxson (1993) reported minimum,
maximum, and average LCRS and LDS flow rates.  The reporting periods for the 49
cells ranged from 4 to 60 months.  At the time of the Feeney and Maxson report,  the
cells were  at different stages of operation,  from newly constructed to closed. For 41 of
the cells with Category 1  liner systems, average LDS flow rates during the monitoring
periods ranged from 0 to  310 Iphd.  Average LDS flow rates for 27 of the 41  Category 1
cells were  less than or equal to 100 Iphd. The authors attributed the observed LDS
flows primarily to consolidation water. The LDS flow rates reported by Feeney and
Maxson (1993) are in the same general range as those reported  earlier by Bonaparte
and Gross (1990, 1993).

LDS flow rates were temporarily higher for the remaining six Category 1 cells and the
two Category 2 cells.  Feeney and Maxson indicate that these eight cells initially
exhibited similar behavior to the other cells; however, during operations, the primary
liner in each cell was damaged, usually by heavy equipment operations in the cell.  The
primary liner in these cells was subsequently repaired.  Average  LDS flow rates for the
damaged cells were about an order of magnitude higher than the rates for the other
cells.  The authors do not provide any information on the high frequency of operational
damage to the liner systems in their study  and whether procedures were developed to
prevent similar damage in future cells.

E-2.1.3.3 Workman (1993)

Workman (1993) presented monitoring results for a MSW landfill having a primary liner
consisting  of a GM on the side slopes and a GM/CCL composite on the base. The LDS
consists of a GN drainage layer overlain by a geotextile (GT) filter. The portion of the
landfill described by Workman contains three cells constructed between 1989 and 1992.
The author does not indicate the level of CQA provided for the construction of each cell.
Average LDS flow rates from the three cells initially ranged from 50 to 700 Iphd, with
                                  E-13
-------
rates at the higher end of the range being associated with the fastest rates of waste
disposal. After cell filling ceased, LDS flow rates decreased to 20 to 30 Iphd.  Workman
attributed the observed LDS flows to consolidation of the CCL component of the
composite primary liner on the base of the landfill. This conclusion was supported by
the concentrations of major ions in the LDS liquids, which were different than the
concentrations of major ions in leachate.

Workman (1993) also reported that chemical analyses of the LDS liquids  from two of the
cells (Cells 1 and 4)  revealed the presence of several volatile organic compounds
(VOCs), including chloroethane, ethylbenzene, and trichloroethene, at low part-per-
billion concentrations, beginning about one year after the start of cell operation.
Workman noted that the detected compounds are common constituents of landfill gas
and that testing of the gas phase in the LDS indicated methane gas concentrations up
to 50% (i.e., up to -100% landfill gas). Workman also indicated that landfill  gas had not
yet been actively removed or passively vented from this landfill  at the time of the
measurements. He attributed the VOCs to the following source: "It is believed that
methane is impacting the LDS liquids of Cells 1 and 4. No organic constituents have
been detected in the Cell 2 LDS. The methane was first detected in Cells 1 and 4 about
one year after each cell was placed in operation.  This occurred about the same time
that the waste reached ground level and totally covered the liner system.  Since
methane is not actively vented at this time and can accumulate underpressure in the
leachate collection system, gradients can occur across the liner system.  The sideslopes
in this landfill are particularly vulnerable. As methane penetrated the liner and cooled,
the gas began to condensate and drain small quantities of liquid to the LDS sump."

E-2.1.3.4 Bergstrom et al. (1993)

Bergstrom et al. (1993) presented flow rate and chemical constituent data for the LDSs
of five cells at a HW landfill in Michigan.  The cells have a GM/CCL  composite primary
liner, with CCL consisting of a 1.5-m thick layer of compacted clayey till. The LDS
consists of a GN drainage layer overlain by a GT filter, with one layer of GN on the cell
side slopes and two  layers of GN on the cell bases. All cells were constructed using
third-party CQA programs.

Average LDS flow rates for the five cells ranged from approximately 200 to 700 Iphd
during active cell filling and 30 to 60 Iphd within one to two years after waste filling
ended. Bergstrom et al. (1993) attributed the observed LDS flows primarily to
consolidation water.  Bergstrom et al. also presented  inorganic  chemical constituent
data obtained from testing of LDS liquid, LCRS liquid, and groundwater. From these
data, they concluded that "each of these water sources has a unique chemical
composition and the leachate does not appear to be influencing LDS liquid
composition". The authors reported that VOCs had not been detected in the LDSs of
the five cells; however, details of the analyte list, analytical methods, and/or analytical
detection limits were not given.  They also estimated consolidation water volumes in the
LDS using the results of laboratory consolidation testing of the site-specific CCL
                                   E-14
-------
material along with records of waste placement in the landfill cells.  Estimated
consolidation water flow rates were 5 to 60% larger than the observed LDS flow rates.

E-2.13.5 Bonaparte et al. (1996)

Bonaparte et al. (1996) analyzed flow rate data for 26 MSW cells at six different landfills
containing GM/GCL composite primary liners. These data were collected as part of the
ongoing research investigation for the EPA mentioned earlier in this appendix.  The
authors used the data to calculate average and peak LCRS and LDS flow rates for three
distinct landfill development stages: (i) the "initial period of operation"; (ii) the "active
period of operation"; and (iii) the "post-closure period". These stages are the same as
those used in this appendix and  discussed previously in Section E-1.3.7. The mean
values calculated by Bonaparte et al. (1996) are presented in Table E-2.1.

Bonaparte et al. (1996) also calculated  "apparent" hydraulic efficiencies for the
composite primary liners of the 26 landfill cells. They defined liner apparent hydraulic
efficiency, Eaas:

             Ea (%) = (1 - LDS Flow Rate / LCRS Flow Rate) x 100                (E-1)

The higher the value of Ea, the smaller the flow rate from a LDS compared to the flow
rate from a LCRS. The value of Ea may range from zero to 100%, with a value of zero
corresponding to a LDS flow rate equal to the LCRS flow rate, and a value of 100%
indicating no flow from the LDS.  The parameter Ea is referred to as an "apparent"
hydraulic efficiency because, as  described earlier, flow into the LDS sump may be
attributed to sources other than primary liner leakage (Figure E-2.1).  The value of Ea is
calculated using total flow into the LDS, regardless of source. If the only source of flow
into the LDS sump is primary liner leakage, then Equation E-1 provides the "true" liner
hydraulic efficiency (Et). True liner efficiency provides a measure of the effectiveness of
a particular liner in limiting or preventing advective transport across the liner. For
example, if a primary liner is estimated to have an Et value of 99%, the rate of leakage
through the primary liner would be assumed to be 1% of the LCRS flow rate.

For the landfill cells with GM/GCL composite primary liners and sand  LDSs, Bonaparte
et al. (1996) found that  the Ea is  lowest during the initial period of operation (Eam =
98.60%; where Eam = mean apparent efficiency) and increases significantly thereafter
(Eam = 99.58% during the active  period  of operation and Eam = 99.89% during the post-
closure period). The lower Eam during the initial  period of operation was attributed to
LDS flow from construction water.  Bonaparte et al. (1996) stated that for cells with sand
LDSs, "calculated AE (Ea) values during the active period of operation and the post-
closure period may provide a reasonably accurate indication of true liner efficiency for
the conditions at these units during the  monitoring periods."

For six cells with GN LDSs,  the calculated value of Eam for the initial period of operation
was 99.96%. This value is higher than  the Eam for composite liners underlain by sand
LDSs for the same facility operational period (i.e., 98.60%). This higher apparent


                                  E-15
-------
Table E-2.1.  Mean LCRS and LDS Flow Rates in Iphd for 26 MSW Landfill Cells with
            GM/GCL Composite Primary Liners (from Bonaparte et al., 1996).
(a) LCRS
Initial Period of
Operation
Active Period of
Operation
Post-Closure Period
Number
of Units
25
18
4
Average Flow
Rate
m
5,350
276
124
o
3,968
165
-
Peak Flow Rate
m
14,964
752
266
o
11,342
590
-
(b) LDS
Initial Period of
Operation
Active Period of
Operation
Post-Closure Period
Number
of Units
26
19
4
Average Flow
Rate
m
36.6
0.7
0.2
0
68.5
1.1
-
Peak Flow Rate
m
141.8
7.7
2.3
0
259.9
13.7
-
       Notes: (1) m = mean value; o = standard deviation.
                                 E-16
-------
Q = TOTAL FLOW
Q = A+B+C+D
SOURCES:
A = PRIMARY LINER LEAKAGE
B = CONSTRUCTION  WATER  AND  COMPRESSION WATER
C = CONSOLIDATION WATER
D = INFILTRATION WATER
 Figure E-2.1. Sources of Flow from LDSs (from Bonaparte and Gross, 1990).
                                  E-17
-------
efficiency can be attributed to the differences in liquid storage capacity and hydraulic
transmissivity between sand and GN drainage materials.  A granular drainage layer can
store a much larger volume of construction water and releases this water more slowly
during the initial period of operation than does a GN drainage layer.  This suggests that,
during the initial period of operation, the main source of flow in a sand LDS underlying a
GM/GCL composite primary liner is construction water.

Bonaparte et al. (1996) concluded that "LDS flows attributable to [primary] liner leakage
vary from 0 to 50 Iphd, with most values being less than about 2 Iphd.  These flow rates
are very low.  The data shown in Table 4 [not included] suggest that the true hydraulic
efficiency of a composite liner incorporating a GCL may be greater than 99.90 percent.
A liner with this efficiency, when appropriately used as part of an overall liner system,
can provide a very high degree of liquid containment capability."

E-2.1.3.6 Conclusions from Previous Studies

The following conclusions are drawn from the previous studies regarding the hydraulic
performance of composite liners.

    •   LDSs underlying GM/CCL composite liners almost always exhibit flow due to
       consolidation water.  Measured  LDS flow rates attributable to consolidation
       water are in the range of 0 to 1,000 Iphd, with most values being less than 200
       Iphd. LDS flow rates attributable to consolidation water are a function of the
       characteristics of the CCL and the rate of waste placement in the overlying cell.
       Typically, the rate of flow decreases with time during the later portion of the
       active period of operation and the post-closure period.  LDS flow rates in the
       range of 0 to 100 Iphd have been reported within one to two years of the
       completion of active filling of cell.
    •   LDS flow attributable to leakage through GM/CCL primary liners was not
       quantified in the previous studies due to the masking effects of consolidation
       water, the very low anticipated flow rates from this source, the limited  available
       database on the chemical constituents in LCRS and LDS liquids, and the
       relatively long breakthrough times for advective transport through the CCL
       component of the liner.
    •   Flow rates from the LDSs of cells with GM/GCL composite primary liners are
       usually very low.  LDS flow  rates attributable to  leakage through this type of
       primary liner typically varied from 0 to 50 Iphd, with most values being less than
       about 2 Iphd.  The true hydraulic efficiency of GM/GCL composite liners may
       often exceed 99.9%.
    •   Average LDS flow rates may increase by an order of magnitude, or more, due to
       liner system damage induced by heavy equipment operations in the cell.
       Engineering and operational measures should be used to prevent this type of
       occurrence.
                                   E-18
-------
E-2.2    Leachate Generation Rates

E-2.2.1   Overview

Few published studies are available on leachate generation rates at modern landfills.
None of these studies included detailed comprehensive evaluations of the rates of
landfill leachate generation and how these rates vary with the landfill life cycle, landfill
waste type, geographic location, or cover condition for a large number of landfill cells.
However, the few published studies present valuable data on leachate flow rates at a
limited number of landfills. The findings of these studies are summarized in this section.

E-2.2.2   Feeney and Maxson (1993)

Feeney and Maxson (1993) presented LCRS flow rate data for 41 double-lined landfill
cells. Four of the 41 cells are located in arid regions, one is located in a semi-arid
region, and 36 are located in humid regions. The average monthly flow rates for the 41
cells ranged from 0 to 33,000 Iphd. Nine of the 41 cells (22% of total number of cells)
had LCRS flow rates less than 100 Iphd, six (15% ) had LCRS flow rates between 100
and 1,000 Iphd, 17 (41%) had LCRS flow rates between 1,000 and 10,000 Iphd, and
nine (22%) had LCRS flow rates greater than 10,000 Iphd. The four cells located in arid
regions had LCRS flow rates between 10 and 70 Iphd.  Feeney and Maxson noted that
15 of the 41 cells (37%) had received final cover systems during the monitoring periods
considered in their paper. The final cover systems consisted of, from bottom to top:
high-density polyethylene (HOPE) GM/CCL barrier overlain by a GN drainage layer; and
soil protection and surface layer.  For six of these closed cells, which were at one
landfill, average monthly LCRS flow rates decreased from approximately 1,400 Iphd to
470 Iphd within six months after closure. Over the next two years, flows decreased
further to 90 Iphd.  The mean  LCRS flow rate for all 15 closed cells decreased from
approximately 3,740 Iphd just before closure to 370 Iphd within several months after
closure.

E-2.2.3   Mauleetal. (1993)

Maule et al. (1993) presented LCRS flow rate and precipitation data from three active
upper midwest landfills for monitoring periods between 10 and 20 months.  Site A
received primary and secondary wastewater treatment sludge dewatered to
approximately 35% solids, along with smaller quantities of dry boiler ash, lime mud, and
wood waste. Site B received both dry boiler ash and a combined primary and
secondary sludge dewatered to an average of 30% solids.  Waste disposed in Site C
was baled MSW.  The LCRS flow rates for Sites A and B varied between 850 Iphd and
43,300 Iphd. During a 15-month monitoring period for Site A and an eight-month
monitoring period for Site B, precipitation at the site averaged approximately 25,000
Iphd and LCRS flow rates averaged approximately 22,000 Iphd for site A and 18,000
Iphd for Site B. Therefore, LCRS flow rates were on average 88 and 72% of
precipitation for sites A and B, respectively.  Site C had more uniform LCRS flow rates
                                  E-19
-------
that averaged approximately 7,000 Iphd over a 20-month monitoring period.  On
average LCRS flows accounted for 32% of precipitation.  Maule et al. (1993) attribute
the lower LCRS flow rates in Site C than in Sites A and B to the use of soil and synthetic
intermediate cover materials over a significant portion of Site C and to the absorptive
capacity of the baled MSW.  The differences in the wastes moisture contents at disposal
time may also have caused the observed differences in leachate generation  rates.

E-2.2.4   Haikolaetal. (1995)

Haikola et al. (1995) presented LCRS flow rate data for ten HW cells located at a landfill
in the NE. Three of these cells are double-lined.  Nine of the ten cells have received a
final cover system with a GM/CCL composite barrier.  Flow rates from the LCRSs of the
cells averaged from 7,800 to 15,100 Iphd while the cells were open, and from 220 to
1,870 Iphd after the cells were closed. The LCRS flow rates decreased from an
average of 1,150 Iphd during the year prior to closure to five times less after one year of
closure and to 40 to 60 times less after ten years of closure.  While the cells  were open,
LCRS flow rates were approximately 60% of annual precipitation at the site.  After ten
years of closure, LCRS flows accounted for less than 1% of annual precipitation.

E-2.2.5   Bonaparte etal. (1996)

Bonaparte et al. (1996) presented LCRS flow rate data for 25 MSW cells at six different
landfills.  Mean values of average and peak LCRS flow rates for the cells are presented
in Table E-2.1.  The mean average flow rates were 5,350 Iphd during the initial period of
operation, 276  Iphd during the active period of operation,  and 124 Iphd after closure.
These LCRS flow  rates are significantly lower than those reported  in prior studies. The
mean LCRS flow rates are controlled by very low LCRS flow rates at 16 cells at one
landfill (i.e., cells AX1-16 of this study). The authors  of this Appendix attribute the low-
LCRS flow rates at this landfill to surface-water management practices implemented at
the landfill.  Much  of the storm water falling onto a cell is diverted away from  the cell as
clean water.

E-2.2.6   Tedder (1997)

Tedder (1997)  presented LCRS flow rate data for 16  active MSW landfill cells located in
Florida for monitoring periods of 3 to 64 months.  Average LCRS flow rates for 15 of
these 16 cells ranged from 1,100 Iphd to 24,600 Iphd. The remaining cell had a very
high average LCRS flow rate of 90,800 Iphd. Approximately 70% of the cells had
average LCRS flow rates less than 9,000 Iphd.

E-2.2.7   Conclusions from Previous Studies

Based on the previous studies of landfill leachate generation rate, the following
conclusions are drawn.
                                   E-20
-------
    •  Open landfills (i.e., landfills without a final cover system) located in relatively
       humid regions have average leachate generation rates that are typically below
       20,000 Iphd.
    •  Average reported leachate generation rates for open landfills located in relatively
       humid regions can be up to 90% of precipitation that occurs at the landfill sites.
       This ratio is related to: (i) the type of waste and its initial  moisture content; and
       (ii) waste placement and covering practices. The ratio is lower for MSW landfills
       than for HW or ISW landfills and for wastes with low hydraulic conductivity daily
       and intermediate covers than for uncovered wastes.
    •  Open landfill cells located in arid regions have average leachate generation
       rates that are much lower (i.e., less than 100 Iphd) than cells in humid regions.
    •  Leachate generation rates decrease significantly after cell closure (i.e., after a
       final cover system is placed on the waste).  From the published studies, LCRS
       flow rates decrease by approximately one to three orders of magnitude within
       one year after closure, and by up to two orders of magnitude after ten years of
       closure.

E-2.3     Leachate Chemistry

E-2.3.1    Overview

This section summarizes published information on leachate chemistry for landfills in the
U.S. Leachate chemistry is primarily dependent on waste type.  Thus, this section is
organized by waste type as follows:

    •  MSW landfill leachate chemistry is discussed in Section  E-2.3.2;
    •  HW landfill leachate chemistry is discussed in  Section E-2.3.3; and
    •  ISW landfill (i.e., coal ash or C&DW landfill) leachate chemistry is discussed in
       Section E-2.3.4.

For the purposes of the discussions on leachate chemistry presented in this appendix,
MSW ash landfill leachate is grouped with leachate from ISW landfills.  This grouping is
considered appropriate because MSW ash landfill leachate is typically nonhazardous
and has chemical characteristics that are more similar to leachate from ISW landfills
that to leachates from MSW or HW landfills.

While the focus of this section is on the chemistry of leachate from newer landfills (i.e.,
landfills that began operation in the 1980's or 1990's), the chemistry of leachate from
older landfills is also presented for comparison purposes.  There are numerous
technical papers containing information on the chemistry of leachate at older MSW or
HW landfills in the U.S. (e.g., Sabel and Clark, 1984; Brown and Donnelly, 1988; Tharp,
1991). However, in many of these studies, poorly-defined co-disposal of wastes
occurred, the leachate had been diluted by ground water, or the samples were not taken
from controlled collection points since the older landfills did  not have LCRSs.  In
addition,  the leachate was almost always analyzed for indicator  parameters (e.g.,
                                   E-21
-------
specific conductance, total suspended solids (TSS), and chemical oxygen demand
(COD)) and major cations and anions (e.g., calcium and sulfate), but less frequently
analyzed for trace amounts of potentially toxic inorganic and synthetic organic
chemicals (i.e., arsenic and benzene) (hereafter referred to as "trace chemicals") that
may be present.  These studies, with potentially-significant co-disposal effects, leachate
diluted by ground water, leachate sampled from an unknown or uncontrolled collection
points, and/or few data on trace chemicals, were excluded from the data summary
presented herein. Consequently, most of the information on leachate from older MSW
and HW landfills  presented in this appendix is from EPA-sponsored studies (i.e.,
Bramlett et al., 1987; NUS Corporation (NUS),  1988; Gibbons et al., 1992).

With the federal solid waste regulations promulgated in the 1980's and early 1990's, it
is expected that the quality of landfill leachate would have improved over time (i.e., the
amount of trace chemicals would have  decreased). A brief summary of these
regulations and the anticipated effect of the regulations on landfill leachate chemistry
are presented below.

Prior to 1980, federal regulations for landfilling of solid waste were limited to the
guidelines for land disposal of solid waste in 40 CFR § 241 and the criteria for
classification of solid waste disposal facilities and practices in 40 CFR § 257.  These
regulations applied to MSW and ISW, but not to HW and other excluded wastes because
of "the lack of sufficient information on which to base recommended procedures". While
HWwas disposed of in designated HW landfills, the regulations allowed HWand other
excluded wastes to be disposed of in MSW and ISW landfills under certain circumstances.
Following the promulgation of the 1980 federal HW regulations (i.e., Resource
Conservation and Recovery Act (RCRA) Subtitle C regulations) that specifically defined
"HW (40 CFR § 261) and the 1982 federal regulations that governed disposal of HW in
landfills (40 CFR  § 264 Subpart N), landfilled HWwas essentially required to be disposed
of in specially designed and constructed HW landfills. Only small quantity generators could
dispose of HW in MSW and ISW landfills. With the passage of the 1980 and 1982
regulations, less HWwas disposed of in MSW and ISW landfills and the amount of trace
chemicals in leachate from these landfills would be expected to have decreased. In
addition, as the public became more aware of HWs after the passage of regulations and
communities instituted voluntary household HW programs, the amount of trace chemicals
entering the MSW waste stream would be expected to have decreased.

The Land Disposal  Restrictions (LDR) (i.e., 40 CFR § 268) required by the Hazardous
and Solid Waste Amendments (HSWA) to the RCRA were promulgated in 1984 and
prohibited the disposal of all listed and characteristic HWs and polychlorinated biphenyls
(PCBs) in landfills, except under limited circumstances or unless the HWs met the
treatment standards of 40 CFR § 268.40 and the PCB waste met the maximum
concentration requirement of 40 CFR § 268.32.  With the treatment standards, the
chemical concentrations in an extract of the waste or of the treatment residue of the
waste must not exceed certain values.  The chemicals that were covered under this
regulation were incrementally added to the list of restricted chemicals from 1986 to 1994
as EPA developed  treatment standards for the  wastes. The LDR would be expected to
                                   E-22
-------
result in a decrease in the amounts of trace chemicals in HW landfill leachate. Small
quantity generators of HWare not subject to the LDR; however, the cut-off weight for
small quantity generator HWwas reduced from that in the 1980 regulations. Thus, this
regulation may also have reduced the amount of trace chemicals in leachate from MSW
and ISW landfills.

To provide additional assurance that regulated HW and PCB waste were being
excluded from MSW landfills, the RCRA Subtitle D regulations (40 CFR § 258) for MSW
landfills, promulgated in 1991, included procedures in 40 CFR § 258.20 for detecting and
preventing disposal of these wastes.  The procedures may have reduced the amount of
trace chemicals in leachate from MSW landfills.

E-2.3.2   MSW

E-2.3.2.1 Introduction

Published information on leachate chemistry for MSW landfills in the U.S. is
summarized below. A study of leachate chemistry for five Canadian MSW landfills by
Rowe (1995) is also included because this study contains more data on the change in
leachate chemistry with time than the other studies.  It is expected that the same trends
would be observed at landfills in the U.S., though the specific constituents and
constituent concentrations may be significantly different due to differences in landfill
disposal regulations in the U.S. and Canada.  Select leachate chemistry data for the
landfills are presented in Table E-2.2. The listed parameters were selected based on
availability of parameters between studies, frequency of detection, and concentration.

It should be noted that  in some of the older MSW landfills, ISW and HW was co-
disposed with MSW. For example, in a study of 20 Wisconsin "MSW landfills by
McGinley and Kmet (1984), the principal  waste types were MSW for six landfills, MSW
and ISW for 11 landfills, and MSW, ISW,  and HW for three landfills. The McGinley and
Kmet (1984) data are included in the study by NUS (1988).  When the data for MSW
landfills are separated from the data for co-disposal landfills, such as in the Gibbons et
al. (1992) study, the co-disposal landfill data are not included in this appendix.

E-2.3.2.2 NUS (1988)

Under contract to EPA, NUS (1988) summarized leachate chemistry data for 83 MSW
landfills from six sources: (i) the Wisconsin Department of Natural Resources (McGinley
and Kmet, 1984); (ii) NUS (NUS,  1986,1987a); (iii) the Trade Association (EPA, 1988);
(iv) Sobotka & Co., Inc. (NUS, 1986);  (v)  K.W. Brown and K.C. Donnelly of Texas A&M
University (Brown  and Donnelly) (NUS, 1986); and (vi) Waste Management, Inc. (Baker,
1987).  The purpose of the study was to evaluate which chemicals were present in
MSW leachate, the chemical concentrations, and the effect of the 1980 Subtitle C
regulations on the chemistry of MSW leachate (i.e., did leachate from newer MSW
landfills contain less trace chemicals). The leachate chemistry data for the landfills are
not complete: organic chemical, inorganic chemical, and indicator parameter data are
                                  E-23
-------
Table E-2.2.  Summary of Select Leachate Chemistry Data from Literature.
Waste Type:
Start of Operation:
Location:
Number of Landfills:
Reference:
Parameter Units
pH pH units
Specific conductance |imhos/cm
TDS mg/l
TSS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Ammonia Nitrogen mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Barium |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Mercury |ig/l
Nickel |ig/l
Zinc |ig/l
Acetone |ig/l
Benzene |ig/l
Chlorobenzene |ig/l
Chloroform |ig/l
1 ,1 - Dichloroethane |ig/l
1 ,2 - Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methyl ethyl ketone |ig/l
4-Methyl-2-pentanone |ig/l
Methylene chloride |ig/l
Naphthalene |ig/l
Phenol |ig/l
Tetrachloroethylene |ig/l
Toluene |ig/l
1,1,1 - Trichloroethane |ig/l
1 ,1 ,2-Trichloroethane |ig/l
Trichloroethylene |ig/l
Vinyl chloride |ig/l
Xylenes |ig/l
MSW
pre-1 980
U.S.
37
NUS
(1988)
6.58
5,540
4,230
264
2,817
2,600
810
2,650
550
118
215
284
138
596
15
580
18
60
72
D(46)
164
880
320
D(56)
D(16)
D(13)
220
D(9)

168
D(59)
430
ND
1,100
D(48)
258
0(24)
420
D(26)
D(6)
D(38)
D(19)
D(63)
post-1 980
U.S.
9
NUS
(1988)
6.91
8,800
7,976
554
4,300
185
2,860
3,900
820
260
299
747
412
817
11
1,000
6.5
8
46
ND
185
335
4,000
ND
ND
ND
4(3)
ND

14
ND
9,900
D(40)
120
ND
1,700
ND
590
ND
ND
ND
ND
ND
pre-1 985
U.S.
24
Gibbons
etal.
(1992)














99

55

205
0.5


D(82)
65
736
D(3)
400
D(4)

492
198
D(65)
D(50)
898


D(5)
583
D(2)
ND
51
107
D(20)
post-1 985
U.S.
12
Gibbons
etal.
(1992)














ND(4)

ND(4)

ND(4)
ND(4)


D(100)(4)
7
D(4)
D(4)
116
D(6)

104
60
D(100)(4)
D(100)(4)
139


D(4)
406
178
D(1)
71
51
NA
1978-1989
Florida
6
Tedder
(1992)
7.07
7,490
4,645
122
1,731
288
103
172
928
174
481
278
30.4
324
36
169
22
56
91
1.2
114
150

9.9
3.9
1
5.9
2.0
ND

27


34.8
10
2,672
1
84
12

67
19
38
1 979-1 993
Ontario,
Canada
5
Rowe
(1995)
6.3-7.7



5,238-17,116
3,361 -12,367


851 - 2,247


322 - 1 ,643




15-50
1 1 0 - 390
1 00 - 570
0.26-5

1,320-11,000

<3-22



ND-10


71-170


ND-3,272


ND-25
156-1,226


ND-55
<13-65

1995
Texas
1
Hunt and
Dollins
(1996)
6.01 -7.18
1,640-2,880
1,110-2,500
NA
50(4)
2m
NA
642 - 2,000
73 - 1 37
84 - 1 ,470
4-14
203 - 460
27-73
155-314
ND-30
ND-700
ND
ND
ND
ND(4)
ND
ND
59-2,100
ND
ND
ND
5-20
ND
25-28
12-25
7-20
75(4)
71 -85
170(4)
ND(4)
ND(4)
ND
10-87
33(4)
ND
ND
10-12
33-38
                                          E-24
-------
Table E-2.2.  Summary of Select Leachate Chemistry Data from Literature (Continued).
Waste Type:
Start of Operation:
Location:

Number of Landfills:
Reference:

Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
TSS mg/l
COD mg/l
BOD5 mg/l
TOO mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Ammonia Nitrogen mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Barium |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Mercury |ig/l
Nickel |ig/l
Zinc |ig/l
Acetone |ig/l
Benzene |ig/l
Chlorobenzene |ig/l
Chloroform |ig/l
1 ,1 - Dichloroethane |ig/l
1 ,2 - Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methyl ethyl ketone |ig/l
4-Methyl-2-pentanone |ig/l
Methylene chloride |ig/l
Naphthalene |ig/l
Phenol |ig/l
Tetrachloroethylene |ig/l
Toluene |ig/l
1,1,1 - Trichloroethane |ig/l
1 ,1 ,2-Trichloroethane |ig/l
Trichloroethylene |ig/l
Vinyl chloride |ig/l
Xylenes |ig/l
MSW Ash
1 970 - 1 981
U.S.+

3+
NUS

(1987b)
7.44-8.58
4, 200 ->1 0,000
11,300-28,900

<5 - 1 ,200

59 - 636

1,803-18,500
94
1.2-36
21
NA
200 - 4,000
5-218
1,000
ND-44
6 - 1 ,530
12-2,920
1 -8
ND-412
ND- 3,300
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
1 975 - 1 988
U.S.


NUS

(1990)
5.2-7.4
9,400 - 46,000
8,030-41,000



1 7 - 420
44 - 744
7,700 - 22,000
14-5,080
4-35
386 - 8,390
1 5 - 367
1 ,240 - 3,800
ND-400
ND-3,080
ND-4
ND-32
ND-54
ND
NA
5-370
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
ND
ND-32
NA
NA
NA
NA
NA
NA
NA
Coal Ash
pre-1978
ash ponds
in U.S.

EPRI

(1978)


404 - 3,328
1 00 - 657





527 - 2,300

1 -563
20 - 1 56
294 - 982
15-84
300 - 40,000
10-52
23 - 1 70
24 - 200
0.2-15
15-130
160-2,700






















ash extract


Eisenberg
etal.
(1986)
4.15-11.1

304 - 1 ,743






250-916

108-310




ND-47
440 - 740
ND

ND-73
ND-680





















1974
U.S.

1
GeoSyntec

(1993)
10.9-11.2
1,440-2,200
736 - 990
NA
NA
NA
3-4
320 - 41 9
16-29
1 93 - 360
NA
8-24
ND
91-111
12-130
13-73
ND-1
ND-203
ND
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
                                          E-25
-------
Table E-2.2.  Summary of Select Leachate Chemistry Data from Literature (Continued).
Waste Type:
Start of Operation:
Location:
Number of Landfills:
Reference:
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
TSS mg/l
COD mg/l
BOD5 mg/l
TOO mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Ammonia Nitrogen mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Barium |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Mercury |ig/l
Nickel |ig/l
Zinc |ig/l
Acetone |ig/l
Benzene |ig/l
Chlorobenzene |ig/l
Chloroform |ig/l
1 ,1 - Dichloroethane |ig/l
1 ,2 - Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methyl ethyl ketone |ig/l
4-Methyl-2-pentanone |ig/l
Methylene chloride |ig/l
Naphthalene |ig/l
Phenol |ig/l
Tetrachloroethylene |ig/l
Toluene |ig/l
1,1,1 - Trichloroethane |ig/l
1 ,1 ,2-Trichloroethane |ig/l
Trichloroethylene |ig/l
Vinyl chloride |ig/l
Xylenes |ig/l
C&DW
< 1991
Texas
3
Norstrom
etal.
(1991)
6.5-7.3
2,920 - 6,850
2,412-4,270
1,000-43,000
3,080-11,200
1 00 - 320
76 - 1 ,080
1,710-6,520
1 25 - 240
<40
30-184
1 48 - 578
92-192
256-1,290
17-75
1,500-8,000
20-30
1 00 - 250
220-2,130
<2-9

1 ,700-8,630





















HW
1972-1983
U.S.
13
Bramlett
etal.
(1987)
8.2
14,690


10,217

3,097







13,100
NA
18.8
281
116
5
6,420
2,510
23,200
303
480
1,940
182
695
NA
166
334
13,600
775
14,100
468
26,100
430
8,340
1,830
21
2,040
ND
1,540
pre-1 983
U.S.
11
NUS
(1988)
6.9
20,000
10,562
1 ,470(3)
12,600
13,400
4,624
540(3)
2028
399
870(3)
72
25.4
377
2,780
840
600
110
480
D( )
272
536
60,000(3)
D( )
D( )
D( )
594
D( )

2,350
D( )


7,715
D( )
3,100
D( )
12,000
D( )
D( )
D( )
D( )

pre-1 987
U.S.
9
Gibbons
etal.
(1992)














18,600

52,500

405
82


D(92)
4,030
9,110
97,500
33,200
200,200

1,670
152,700
D(82)
D(80)
385,400


67,400
79,700
200,600
39,600
95,500
8,960
NA
post-1 987
U.S.
3
Gibbons
etal.
(1992)














NA

NA

NA
NA


NA
529
156
D(2)
132
4,250

81
185
NA
NA
1,970


112
1,670
D(4)
ND
38
40,100
NA
1 969 - 1 988
U.S.
10
Pavel ka
etal.
(1994)














13,800
860
430
240
100
0.8
1,800
1,150
46,800
425


135




59,300
41 ,900
39,500
200
32,000

3,310
1,900

4,210


                                          E-26
-------
Table E-2.2.  Summary of Select Leachate Chemistry Data from Literature (Continued).

Notes:
 (1) Parameter values are given as medians (NUS, 1988) or arithmetic averages (Bramlett et al., 1987; Gibbons et
    al., 1992; Tedder, 1992; Pavelka etal., 1994) of detected values, ranges of maximum concentrations (EPRI,
    1978), ranges of representative peak annual concentrations (Rowe, 1995), or ranges of all concentrations
    (Eisenbergetal., 1986; NUS, 1987b; NUS, 1990; GeoSyntec, 1993;  Hunt and Dollins, 1996).
 (2) ND = not detected; NA = not analyzed; D(x) = detected in x% of samples and concentration was not given;
    "  " = not reported.
 (3) Based on one detected concentration.
 (4) Based on one or two samples.
                                                 E-27
-------
available for 60 landfills; only inorganic chemical and indicator parameter data are
available for 16 landfills; and only organic chemical data are available for seven landfills.
In addition, the landfill leachates were not analyzed for the same inorganic and organic
chemicals.  For example, the leachates in the study by McGinley and Kmet (1984) were
analyzed for the 114 organic chemicals on the Priority Pollutant List (PPL). The PPL
does not include certain organic chemicals, such as ketones (e.g., acetone, methyl ethyl
ketone, 4-methyl-2-pentanone) and xylenes, which are commonly found in MSW landfill
leachate. The leachates in the studies by NUS (1986,1987a) were analyzed for the 210
organic chemicals in proposed Appendix IX of 40 CFR § 264, which contained ketones
and xylenes.  The McGinley and Kmet (1984), NUS (1986, 1987a), and Baker (1987)
studies were based on multiple data sets. Only one data set is available for each of the
landfills in the Trade Association (EPA, 1988), Sobotka & Co., Inc. (NUS, 1986), and
Brown and Donnelly (NUS, 1986) studies. It is unknown if the data set represented one
sampling event or was an average for several events.

NUS divided the leachate chemistry data into three categories: (i) data from 37 landfills
that started operation prior to 1980 (pre-1980 landfills); (ii) data from 9 landfills that
started operation after 1980 (post-1980 landfills); and (iii) data from 37 landfills whose
start of operation date was unknown. From comparison of these three categories of
data for MSW landfill leachate and data for HW landfill  leachate (discussed in Section
2.3.3.3), NUS found that 89 of the 275 organic chemicals analyzed for and 55 of the 58
inorganic chemicals and indicator parameters analyzed for were detected in MSW
leachate. The trace inorganic and organic chemicals detected in more than 60% of the
pre-1980 and post-1980 MSW leachate samples are arsenic, barium, boron, cadmium,
chromium, copper, lead, manganese, nickel, vanadium, zinc, acetone, methyl ethyl
ketone, methylene chloride, phenol,  and toluene.  Of these chemicals, barium, acetone,
methyl ethyl ketone, and phenol were detected at the highest average concentrations.
No significant difference in leachate  chemistry between the pre-1980 and post-1980
landfills was evident.  However, chemicals generally occurred at lower concentrations in
MSW landfill leachate than in HW landfill leachate. NUS also found that more that half
of the chemicals with federal health-based standards (e.g., maximum contaminant
levels (MCLs) for drinking water) were detected in MSW leachate at median
concentrations greater than the standards.

E-2.3.2.3 Gibbons et al. (1992)

In their study, Gibbons et al. evaluated the leachate chemistry for 48 MSW, HW,  and
MSW/HW co-disposal landfills and assessed whether the leachates from these facilities
were significantly different. The database included 347 leachate samples from 36 MSW
landfills owned or operated by Waste Management of North America (WMNA). Twenty-
four of these MSW landfills began operation before 1985 and were considered "old"
landfills. Twelve of the landfills began operation in 1985 or later and were considered
"new" landfills. The cutoff between the "old" and "new" landfills was based on a WMNA
policy implemented in 1985 that limited disposal of liquids in company landfills. Not all
samples were analyzed for the same parameters.  However, most were analyzed for the
56 PPL volatile organic compounds (VOCs) and some  were  analyzed for heavy metals.


                                  E-28
-------
Gibbons et al. primarily considered the PPL VOCs and four metals, arsenic, cadmium,
mercury, and lead, in their evaluation of leachate chemistry.

Gibbons et al. found that MSW leachates in the database were differentiated from HW
leachates based on the detection frequencies and concentrations of the considered
chemicals (i.e., PPL VOCs and four metals). While aromatic hydrocarbons (e.g.,
benzene, toluene), arsenic, cadmium, lead, and trans-1,2-dichloroethylene were
frequently detected in both MSW and HW leachate samples, certain chemicals, such as
chloroform, 1,2-dichloroethane, and tetrachloroethylene, were only found in more than
6% of samples for HW leachate.  Other chemicals, such as mercury, trichloroethylene,
and vinyl chloride, were primarily detected in HW leachates.  Gibbons et al. noted that
the difference in chemical detection frequencies between MSW and HW leachates is
probably underestimated because the detection limits for HW leachates were generally
significantly higher than those for MSW leachates due to sample matrix interference
(e.g., for benzene, the average detection limit was about 80 jig/l for MSW leachate and
20,000 jig/l for HW leachate).  Thus, a chemical could be present at higher
concentrations in HW leachate than in MSW leachate and be reported as "nondetected"
in the HW leachate.  Of the eight PPL VOCs detected in more than 10% of leachate
samples from both old  and new MSW landfills, six were found at lower average
concentrations in samples from new landfills than  in samples from old landfills.
However, only benzene was shown to be at a statistically lower concentration in
leachate from new MSW landfills.  In  contrast, almost all chemicals detected in more
than 10% of the MSW leachate samples  were at significantly lower concentrations in the
MSW leachate as compared to old HW landfill leachate.  Gibbons et al. also found that
the chemistry of leachate from MSW/HW co-disposal landfills was more similar to that of
MSW leachate than to  HW leachate.

E-2.3.2.4 Tedder (1992)

Tedder summarized leachate chemistry data for six active MSW landfills located in both
rural and heavily populated areas  in Florida and compared the chemical concentrations
to regulatory standards to assess  leachate quality. One of the landfills is  a co-disposal
facility that accepts both MSW and MSW ash.  Another landfill is operated as a
bioreactor with leachate recirculation.  Operation of the landfills began from 1978 to
1989; only one of the landfills was operated prior to 1980.  Tedder provided leachate
chemistry data for 146 samples collected between January 1987 and February 1992.
The suite of parameters that the leachate samples were analyzed for was not given,
and each sample was not analyzed for the same parameters.

The trace inorganic and organic chemicals detected at the highest average
concentrations are manganese, selenium, pentachlorophenol, and phenol. The
detection frequencies for these chemicals were not given.  Tedder compared the
detected concentrations to regulatory standards and found that the maximum detected
chemical concentrations in the leachates were below the toxicity characteristic
concentrations for solid waste given in 40 CFR § 261.24.  In addition, with the
exceptions often chemicals (i.e., beryllium, manganese, selenium, methylene bromide,


                                  E-29
-------
ethylbenzene, methylene chloride, pentachlorophenol, phenol, trichloroethylene, and
vinyl chloride), the average concentrations of detected chemicals were near or below
the Florida Department of Environmental Regulation standards or guidance
concentrations for drinking water.  With the exceptions of four inorganic chemicals (i.e.,
barium, beryllium, chromium, and selenium) and four phenolics (i.e., 2,4-dimethyl
phenol, p-nitrophenol, pentachlorophenol, and phenol), the average concentrations of
detected chemicals were near or below the levels reported for MSW leachate in the
study by NUS (1988). Tedder concluded that while the concentrations of inorganic
chemicals in the Florida landfill leachates were generally similar to those in the NUS
study, the concentrations of organic chemicals in the Florida landfill leachates were
generally significantly less.

E-2.3.2.5 Rowe (1995)

Rowe presented leachate chemistry data for five MSW landfills in  Ontario, Canada and
compared the leachate chemistry for these landfills to leachate chemistries for MSW
landfills in the U.S. and Europe. Operation of the  Canadian landfills began between
1972 and 1983.  One landfill was closed in 1988, the remaining four landfills were active
in 1993 at the time of the Rowe's study.

Rowe found the representative peak annual chemical concentrations  in leachate from
the Canadian landfills to be generally consistent with published U.S. and European
data.  For landfills with sufficient data, the concentration versus time trend for different
parameters was investigated. For one active landfill with the  most complete data set,
chloride, biological oxygen demand (BOD), and COD concentrations and the ratio of
BOD to COD increased with time over ten years of operation, though  these trends were
not monotonic from year to year.  The relatively low pH and the relatively high
BOD/COD values for the leachate were characteristic of leachate  during the acid  phase
of MSW decomposition.  Interestingly,  the concentrations of commonly detected VOCs,
such as ethylbenzene, methylene chloride, and trichloroethylene,  decreased over the
same time period. Rowe attributed this trend to degradation of the VOCs.  Rowe used
the VOC time trends at the landfills along with published information on VOC
degradation to estimate half-lives of selected VOCs.

E-2.3.2.6 Hunt and Dollins (1996)

Hunt and Dollins presented leachate chemistry data for a 1.8-ha MSW landfill cell in
northcentral Texas that became operational in June  1995.  After waste in the cell
reached interim grades in September 1995, waste placement in the cell was temporarily
ceased. Leachate collected from the cell from June to September was recirculated
back into the cell.  Leachate chemistry data for five sampling  events from June 1995 to
June 1996 are summarized in Table E-2.2. The leachate was analyzed for the 62
parameters in Appendix I of 40 CFR § 258 (i.e., detection ground-water monitoring
parameters for MSW landfills) for four of the sampling events and  the 213 parameters in
Appendix II of 40 CFR § 258 (i.e., assessment ground-water monitoring parameters for
                                  E-30
-------
MSW landfills) for one of the sampling events. The Appendix I parameters are included
in Appendix II. Only 17 of the Appendix II parameters were detected:

    •   arsenic;
    •   barium;
    •   acetone;
    •   methyl ethyl ketone;
    •   carbon disulfide;
    •   chloroethane;
    •   1,1-dichloroethane;
    •   1,1-dichloroethylene;
    •   cis-1,2-dichloroethylene;
    •   trans-1,2-dichloroethylene;
    •   ethylbenzene;
    •   4-methyl-2-pentanone;
    •   methylene chloride;
    •   1,1,1-trichloroethane;
    •   toluene;
    •   vinyl chloride; and
    •   xylenes.

The organic chemicals detected at the highest concentrations were acetone and
methylene chloride.  Interestingly, the data presented by Hunt and Dollins (1996) also
show that leachate pH has been increasing overtime from 6.01 in June 1995 to 7.18 in
June 1996 and the BOD to COD ratio was very low (i.e., 0.04) when  BOD and COD
were measured  in June 1996, suggesting that waste decomposition has moved from the
acid stage to the methane fermentation stage.

E-2.3.2.7 Conclusions from Previous Studies

The published leachate chemistry data show that leachate from MSW landfills is a
mineralized, biologically-active liquid containing trace concentrations of heavy metals
and synthetic organic chemicals. The limited data  on the change in MSW leachate
chemical concentrations over time from the Rowe (1995) and Hunt and Dollins (1996)
studies are consistent with theory (discussed in Section E-6.2.2).  During the active life
of a MSW landfill, waste decomposition is primarily in the acid stage.  In this stage, BOD
to COD ratios are relatively high and pH is relatively low.  As waste placement ceases,
BOD to COD  ratios decrease and pH increases. The trace chemicals were generally
found to occur at significantly lower frequencies and concentrations in MSW leachate
than in HW leachate. A significant difference between leachate from old and new MSW
landfills was not observed in the studies by NUS (1988) and Gibbons et al. (1992),
though most trace chemicals were detected at lower concentrations in leachate from
new MSW landfills than leachate from old MSW landfills.  However, the data  set for
newer landfills was relatively small and only included one landfill that became
                                   E-31
-------
operational after 1989.  It is anticipated that the quality of leachate from the post-1990
MSW landfills would be improved over that from pre-1980 MSW landfills.

Table E-2.3 presents a list  of the chemicals detected more than once in the U.S. MSW
leachate studies and that are also in Appendix II of 40 CFR § 258.  Two landfill
categories are considered  in this table: (i) old MSW landfills (i.e., pre-1980 landfills in
NUS (1988) and pre-1985  landfills in Gibbons etal. (1992)); and (ii) new MSW landfills
(i.e., post-1980 landfills in NUS (1988), post-1985 landfills in Gibbons et al. (1992),
landfills in Tedder (1992), and the landfill in Hunt and Dollins (1996)). In Table E-2.3,
chemicals detected more than once are indicated with one check, chemicals detected in
at least 30% of the samples are indicated with two checks, and chemicals detected in at
least 60% of the samples are indicated with three checks. The Tedder (1992) data
could only be used to determine if a chemical has been detected and could not be used
to calculate detection frequencies because the number of samples analyzed for a given
chemical was not indicated. Appendix II chemicals monitored for but detected once or
less are listed in Table E-2.4.

Sixty-one of the Appendix II parameters were detected in MSW leachate more than
once, and at least 116 of the  parameters were  not detected or were detected only once.
Nine of the  Appendix II inorganic chemicals and four of the Appendix II organic
chemicals were detected in more  than 60% of the MSW leachate samples analyzed for
the constituents. The Appendix II metals detected in the highest concentrations are
barium, nickel, and zinc. The four organic chemicals detected with the highest
frequency, acetone,  methyl ethyl ketone, methylene chloride,  and toluene, were also
present at higher concentrations than the other Appendix II organic chemicals.  These
organic chemicals generally have relatively high solubilities.  It appears, from Table E-
2.3, that certain Appendix II chemicals found in leachate from old MSW landfills were
not detected more than once in leachate from new MSW landfills (e.g., 2,4-D, dimethyl
phthalate, and 1,1,2-trichloroethane).  This effect likely results from the regulations
discussed in Section 2.3.1, but may also be partially an artifact of sampling size: there
were fewer samples from new MSW landfills analyzed for these chemicals, so it was
less likely that the chemicals would be detected.

E-2.3.3  HW

E-2.3.3.1 Introduction

Published information on leachate chemistry for HW landfills in the U.S. is summarized
below. Select leachate chemistry data are presented in Table E-2.2.  Most of the data
are from older landfills or include  leachate samples from older landfills, and, thus, do not
reflect the improvements in HW landfill leachate chemistry that are expected under the
previously-discussed solid  waste  regulations.
                                   E-32
-------
Table E-2.3.   Appendix II Chemicals Detected More than Once in Leachate from MSW
             Landfills.
Constituent
Antimony
Arsenic
Barium
Beryllium
Cadmium
Chromium
Cobalt
Copper
Cyanide
Lead
Mercury
Nickel
Selenium
Silver
Thallium
Tin
Vanadium
Zinc
Acetone
Acetonitrile
Benzene
Bis(2-ethylhexyl) phthalate
Carbon tetrachloride
Chlorobenzene
Chloroethane
Chloroform
p-Cresol
2,4-D
4,4-DDT
Di-n-butyl phthalate
o-Dichlorobenzene
p-Dichlorobenzene
Dichlorodifluoromethane
1,1-Dichloroethane
1,2-Dichloroethane
1,1-Dichloroethylene
cis-1 ,2-Dichloroethylene
trans-1 ,2-Dichloroethylene
1 ,2-Dichloropropane
Diethyl phthalate
Old Landfills
v'v'
v'v'v'
v'v'v'
S
sss
sss
ss
sss
sss
sss
s
sss
ss
ss
ss
s
sss
sss
sss
ss
ss
s

ss
s
s
sss
ss
s
s
s
s
s
s
s
s
No data available
>/>/
s
ss
New Landfills

v'v'v'
>/>/>/

>/>/>/
>/>/>/

•/ss
s
sss
s
sss
s
ss


sss
sss
sss
s
s
s
s
s
s
s
ss

ss
s

s
s
ss
s
s
sss
ss

s
                                  E-33
-------
Table E-2.3.    Appendix II Chemicals Detected More than Once in Leachate from MSW
              Landfills (Continued).
Constituent
2,4-Dimethylphenol
Dimethyl phthalate
Ethylbenzene
2-Hexanone
Isophorone
Methyl chloride
Methyl ethyl ketone
4-Methyl-2-pentanone
Methylene chloride
Naphthalene
Nitrobenzene
Pentachlorophenol
Phenol
Tetrachloroethylene
Toluene
1,1,1-Trichloroethane
1 , 1 ,2-Trichloroethane
Trichloroethylene
Trichlorofluoromethane
Vinyl chloride
Xylenes
Old Landfills
S
s
ss
s
ss
s
sss
ss
sss
ss
s
s
sss
s
sss
s
s
s
s
s
s
New Landfills
s

ss
ss
s

sss
sss
sss
s

s
ss
s
sss
s

s
s
s
s
Notes:  (1) S = detected more than once, SS = detected in at least 30 % of samples,
                        = detected in at least 60 % of samples.
                                   E-34
-------
Table E-2.4.  Appendix II Chemicals Monitored for but Detected Once or Less in
            Leachate from MSW Landfills .
Acenaphthene
Acenaphthylene
Acetophenone
2-Acetylanimofluorene
Acrolein
Acrylonitrile
Aldrin
Allyl chloride
4-Aminobiphenyl
Anthracene
Benzo[a]anthracene
Benzo[b]fluoranthene
Benzo[k]fluoranthene
Beno[g,h,i]perylene
Benzo[a]pyrene
Benzyl alcohol
alpha-BHC
beta-BHC
gamma-BHC
Bis(2-chloroethyl) ether
4-Bromophenyl phenyl ether
Butyl benzyl phthalate
Carbon disulfide
Chlordane
p-Chloroaniline
Chlorobenzilate
2-Chloronaphthalene
2-Chlorophenol
4-Chlorophenyl phenyl ether
Chrysene
m-Cresol
o-Cresol
4,4-DDD
4,4-DDE
Dibenz[a,h]anthracene
Dibenzofuran
1 ,2-Dibromo-3-chloropropane
1 ,2-Dibromoethane
m-Dichlorobenzene
3,3-Dichlorobenzidine
2,4-Dichlorophenol
2,6-Dichlorophenol
trans-1 ,3-Dichloropropene
Dieldrin
p-(Dimethylamino)azobenzene
7,12-Dimethylbenz[a]anthracene
3,3-Dimethylbenzidine
m-Dinitrobenzene
4,6-Dinitro-o-cresol
2,4-Dinitrophenol
2,4-Dinitrotoluene
2,6-Dinitrotoluene
Dinoseb
Di-n-octyl phthalate
Diphenylamine
Disulfoton
Endrin
Endrin aldehyde
Ethyl methacrylate
Famphur
Fluorene
Heptachlor
Heptachlor epoxide
Hexachlorobenzene
Hexachlorobutadiene
Hexachlorocyclopentadiene
Hexachloroethane
Hexachloropropene
lndeno(1 ,2,3-cd)pyrene
Isobutyl alcohol
Isodrin
Kepone
Methacrylonitrile
Methapyrilene
Methoxychlor
3-Methylcholanthrene
Methyl bromide
Methyl iodide
Methyl methanesulfonate
Methylene bromide
2-Methylnaphthalene
1,4-Naphthoquinone
1-Naphthylamine
2-Naphthylamine
m-Nitroaniline
o-Nitroaniline
p-Nitroaniline
o-Nitrophenol
p-Nitrophenol
N-Nitrosodi-n-butylamine
N-Nitrosodiethylamine
N-nitrosodimethylamine
N-nitrosodiphenylamine
N-nitrosodipropylamine
N-Nitrosomethylethalamine
N-Nitrosopiperidine
N-Nitrosopyrrolidine
Pentachlorobenzene
Pentachloronitrobenzene
Phenacetin
Phenanthrene
p-Phenylenediamine
Phorate
Polychlorinated biphenyls
Pronamide
Pyrene
Styrene
2,4,5-T
1 ,2,4,5-Tetrachlorobenzene
1 ,1 ,1 ,2-Tetrachloroethane
1 ,1 ,2,2-Tetrachloroethane
Toxaphene
1 , 2, 4-Trichloro benzene
2,4,6-Trichlorophenol
1 ,2,3-Trichloropropane
Vinyl acetate

                                  E-35
-------
E-2.3.3.2 Bramlett et al. (1987)

Bramlett et al. presented leachate chemistry data for 13 HW landfills that began
operating from 1972 to 1983. The landfills were located in all of the four geographic
regions of the U.S. defined by Bramlett et al. and had accepted a variety of HW.
Leachate samples were collected from the landfills in 1985 and analyzed for indicator
parameters and PPL metals, VOCs, and semivolatile organic compounds (SVOCs).

With the exception of beryllium, which was only detected in leachate from six landfills,
all of the PPL metals were detected in leachate from 10 or more landfills. The trace
metals detected at the highest average concentrations were arsenic, mercury, and
nickel. The most commonly detected organic chemicals were acetone, benzene, 2-
hexanone, methyl ethyl ketone, p-cresol, methylene chloride, phenol, and toluene.
These chemicals were also generally detected at higher concentrations than other
organic chemicals in the leachate.

E-2.3.3.3 NUS (1988)

As part of their study for EPA, NUS abstracted from TRW (1983) select leachate
chemistry data for 11 HW landfills.  The leachate chemistry data for the HW landfills are
not complete: organic and inorganic chemical data are available for nine landfills, only
inorganic chemical data are available for one landfill, and only organic chemical data are
available for one landfill.  The leachate samples were analyzed for 46 indicator
parameters and inorganic chemicals and 32 organic chemicals.

The trace metals detected at the most landfills were chromium, copper, and zinc. The
metals detected at more than one landfill and at the highest concentrations were
arsenic, barium, and zinc. Only the concentrations for nine of the 32 organic chemicals
were reported in NUS (1988).  Of these, acetone and methylene chloride were detected
at the highest concentrations.  As described in Section E-2.3.2.2, NUS also compared
the leachate chemistry data for the HW landfills and the MSW landfills in the database
and found that the inorganic and organic chemicals generally occurred at higher
concentrations in HW landfill leachate than in MSW landfill leachate.

E-2.3.3.4 Gibbons et al. (1992)

Gibbons et al. evaluated the chemistry of 945 leachate samples from 12 HW landfills
owned by Chemical Waste Management, Inc.  Nine of these HW landfills began
operation before 1987 and were considered "old" landfills. Three of the landfills began
operation in  1987 or later and were considered "new" landfills.  The cutoff between the
"old" and "new" landfills was based on the start of the LDR and more stringent landfill
design requirements. Not all samples were analyzed for the same parameters.
However, some were analyzed for trace metals and most were  analyzed for the PPL
VOCs. Gibbons et al. primarily considered four metals, arsenic, cadmium, mercury, and
lead, and the PPL VOCs in their evaluation of leachate chemistry.
                                  E-36
-------
Of the four metals, arsenic was detected at the highest frequency (i.e., in 93% of the
analyzed samples) and cadmium was detected at the highest concentration.  Methylene
chloride and toluene were the most frequently detected organic chemicals, found in over
75% of the analyzed samples. The organic chemicals detected at the highest average
concentrations were 1,2-dichloroethane, methylene chloride, and  1,1,1-trichloroethane.
Gibbons et al. found that most, but not all,  of the considered chemicals were detected
less frequently and at significantly lower concentrations in leachate from new HW
landfills  than in leachate from old HW landfills.  The two significant exceptions to this
are: (i) benzene, which was detected more than twice as  frequent and at statistically
higher average concentrations in leachate  from new landfills than  in leachate from  old
landfills; and (ii) vinyl chloride, which was detected at a statistically higher average
concentration in leachate from new landfills than in leachate from  old landfills.

E-2.3.3.5 Pavelkaetal. (1994)

Pavelka et al. evaluated the leachate chemistry data summarized  by EPA (1989) for 18
cells at ten HW landfills in the U.S. The landfills are located in all  four of the
geographical areas of the U.S. defined by  EPA (1989). Three of the 18 cells were
reportedly used for co-disposal of MSWand  HW, with MSW being the predominant
material. Waste disposal began at the landfills between about 1969 and 1988. The
EPA report contains leachate chemistry data for one sampling event conducted in  1989.
The leachate samples were analyzed  for 231 chemicals,  including VOCs, SVOCs,  and
trace metals.

All metals analyzed for were present.  The most frequently detected trace metals were
arsenic, barium, nickel, and zinc. These metals were also present at the highest
concentrations.  Twenty-nine of the 72 VOCs and 17 of the 107 SVOCs analyzed for
were detected. The three most frequently  detected VOCs, acetone, methyl ethyl
ketone,  and 4-methyl-2-pentanone, also were present at  the highest concentrations.
The dominant SVOCs  were phthalic acid and phenol. Pavelka et  al. evaluated the
relationship between leachate concentration  and chemicals properties and found that
that, excluding the alcohols, the  organic chemicals detected at the highest
concentrations generally had relatively high solubilities. The organic chemicals detected
at the lowest concentrations generally had relatively high octanol-water coefficients.

E-2.3.2.6 Conclusions from Previous Studies

Like MSW leachate, HW leachate is a mineralized liquid containing trace concentrations
of heavy metals and synthetic organic chemicals. HW leachate may also be biologically
active like MSW leachate, though generally to a much lesser degree, depending on the
characteristics of the HW. The trace inorganic and organic chemicals detected most
frequently and at the highest concentrations  in HW leachate were arsenic, barium,
nickel, zinc, acetone, methyl ethyl ketone,  methylene chloride, phenol, and toluene.
The organic  chemicals with the  highest concentrations have relatively high solubilities.
The limited leachate chemistry data from new HW landfills show that most, but not all, of
the PPL VOCs and selected metals were detected less frequently and at lower
                                   E-37
-------
concentrations in leachate from new HW landfills than in leachate from old HW landfills.
Thus, the Subtitle C regulations and the LDR appear to have resulted in improved HW
leachate quality.  From a comparison of the data for MSW and HW leachate, chemicals
are generally present at significantly higher concentrations in HW leachate than in MSW
leachate. Also, certain chemicals, such as 1,2-dichloroethane, are detected more often
in HW leachate than in MSW leachate.  As expected, all of the Appendix II chemicals
detected in MSW landfill leachate (Table E-2.3) and analyzed for in HW landfill leachate
were found in HW leachate. However,  several of the Appendix II chemicals detected
once or less in MSW leachate (Table E-2.4) were detected more than once in HW
leachate. Of these, acetophenone, o-cresol, 2,4-dichlorophenol, and isobutyl alcohol
were detected  most frequently in HW leachate.

E-2.3.4   ISW

E-2.3.4.1 Introduction

There are few published studies on the chemistry of leachate from ISW landfills.  The
primary reasons for this scarcity of information is that: (i) past studies focused on the
chemistry of MSW and HW landfills, which historically have resulted in more ground-
water contamination problems than ISW landfills; and (ii) ISW landfills have generally
been less regulated than MSW and HW landfills, and leachate chemistry data were not
typically required to be collected for the ISW landfills.  This section summarizes
published information on the leachate chemistry for three types of industrial wastes:  (i)
MSW ash (considered to be an ISW for the discussion of leachate chemistry in this
appendix); (ii) coal ash;  and (iii) C&DW. Select leachate chemistry data are presented
in Table E-2.2.

E-2.3.4.2 MSW Ash

NUS (1987b,1990) conducted two studies to assess the chemical properties of MSW
ash leachate. Though their studies included information on the chemistry of ash
extracts, leachate from MSW/MSWash co-disposal landfills, and leachate from MSW
ash landfills, only the chemistry of leachate from MSW ash landfills is considered
herein.

The first NUS study included a combined summary of leachate chemistry data from
publications and for three MSW ash landfills in the U.S.  Only limited data on the
chemistry of MSW ash landfill  leachate were found in publications from the U.S.,
Canada, Japan, and Europe.  These data were collected using a variety of sampling
procedures and analytical methods.  However, the published data generally fell within
the range of values for leachates from three MSW ash landfills in the U.S. These three
landfills became operational from 1970 to 1981.  Nine leachate samples were collected
from the three  landfills in 1986 and analyzed for metals, total organic carbon (TOC),
organic scan, base neutral extractables (BNAs), PCBs, polychlorinated dibenzo-p-
dioxins (PCDDs), and polychlorinated dibenzo-furans (PCDFs). The samples were not
analyzed for VOCs because VOCs are combusted during the incineration process and
                                   E-38
-------
are not detected in MSWash. The landfills were resampled for indicator parameters in
1987.

The second NUS study included leachate chemistry data for four MSWash landfills.
The landfills became operational between 1975 and 1988. Thirteen leachate samples
were collected in 1988 and 1989 and analyzed for indicator parameters, select metals,
Appendix IX SVOCs, and select PCDDs and PCDFs.

The results of the literature survey and chemical analyses indicate that MSWash
leachate can range from acidic to alkaline and has higher levels of specific
conductance,  total dissolved solids (TDS),  sulfate, and chloride and lower levels of COD
and TOC than MSW and HW leachates. Of the trace metals listed in Table E-2.2, all
were detected in MSW ash leachate at concentrations near or above those for MSW
leachate. Only barium, cadmium, and lead were detected at higher concentrations in
MSW ash leachate than in MSW or HW leachate. Eleven BNAs were detected in the
MSW ash leachate. The most frequently detected BNAs were  bis (2-ethylhexyl)
phthalate and dimethyl propanediol, both found in leachate from two of the seven
landfills described in the NUS reports.  Total PCDDs and PCDFs were detected in
leachate at very low concentrations, ranging from 0.06-543 jig/l and 0.04-823 jig/l,
respectively.   NUS hypothesized that suspended solids in the leachate were probably
the main contributor of PCDDs and PCDFs since these chemicals have a low water
solubility. The maximum detected concentrations of metals in the MSW ash leachate
were compared to the toxicity characteristic concentrations for solid waste given in 40
CFR § 261.24 and EPA MCLs and secondary MCLs (SMCLs).  While the detected
metals  concentrations were less than the toxicity characteristic concentrations, most
were greater than MCLs and SMCLs.

E-2.3.4.3 Coal Ash

There are few published data on the chemistry of coal ash landfill leachate. Most of the
data on coal ash leachate are from ash ponds or chemical extraction tests on ash
samples. Ash pond leachate may be more concentrated than landfill leachate due to
the relatively long contact time between the ash and water used to sluice the ash.
Leachate generated in a laboratory by extraction tests with ash samples may or may not
resemble landfill leachate depending on the extraction method.  Data on coal ash
leachate chemistry for one landfill in the eastern U.S. are presented in this subsection.
Chemistry data for ash ponds and coal ash leachate generated in the laboratory are
also presented for comparison purposes.

GeoSyntec Consultants  (1993) presented leachate chemistry data for a bituminous coal
ash landfill that began operating in 1974.  Leachate collected in the piping system
beneath the ash flows to one of two leachate ponds. Three leachate samples were
collected from the inlets to the ponds during 1992 and 1993 and analyzed for indicator
parameters and metals.  The leachate for this landfill is alkaline (i.e., pH of about 11),
which limits the concentrations of dissolved metals in the leachate, and  less mineralized
than leachates from MSW, HW, and MSWash landfills.  In addition, the concentrations
                                  E-39
-------
of trace metals in the coal ash leachate were within the range of values for MSW landfill
leachate.

Studies of the chemistry of coal ash leachate (not coal ash landfill leachate) were
conducted by the Electric Power Research Institute (EPRI) (1978) and Eisenberg et al.
(1986). The EPRI study presented maximum metals concentrations measured in fly
ash pond, bottom ash pond, and ash pond leachates.  Eisenberg et al. (1986) presented
data on the chemistry of three fly ash leachates prepared by extraction tests.  The
leachates were made by filtering mixtures of fly ashes and dissolved water.  All of the fly
ashes were produced in plants burning eastern bituminous coal. One of the leachates
was alkaline, one had a pH that was nearly neutral, and one was acidic.  With the
exception of chromium, the acidic leachate had the highest concentrations of metals.
Since most of the leachable metals on fly ash are in the form of acid-soluble metal
sulfates, a coal ash leachate with a high sulfate concentration generally has high metals
concentrations.  The coal ash leachates in the EPRI and Eisenberg et al. studies
generally contained higher levels of sulfate and metals than the coal ash landfill
leachate  in the GeoSyntec study.

E-2.3.4.4 C&DW

Norstrom et al. (1991) presented leachate chemistry data for three C&DW landfills
located near Houston, Texas. Wastes that are typically allowed at C&DW landfills
include brush, grass, lumber, concrete, plaster, asphalt, rock, soil, and metal.  Since the
landfills did not have LCRSs, leachate was collected from wells installed into the waste.
The leachate samples were analyzed for indicator parameters and metals.

Norstrom et al. found that the concentrations of chemicals in the C&DW leachate
generally fell within the lower half of the range  of concentrations reported for MSW
leachate.  Barium, lead, and zinc, however, were detected at higher concentrations in
C&DW leachate than in MSW leachate.

E-2.3.4.5 Conclusions From Previous Studies

ISW landfill leachate chemistry can vary significantly depending on the type of waste.
Leachate from MSW ash landfills can be acidic or alkaline and generally have higher
levels of inorganic chemicals than MSW and HW leachates.  Of the trace metals
analyzed for, all were detected in MSW ash leachate at concentrations near or above
those for MSW leachate and barium, cadmium, and lead were detected at
concentrations higher than those for HW leachates.  Coal ash landfill leachates can also
be acidic or alkaline and are generally less mineralized than MSW leachates.  Both
MSW ash and coal ash leachates have essentially no VOCs and few SVOCs.  MSW
ash leachates, however, can also contain trace amounts of PCDDs and PCDFs.

C&DW landfill leachate can have lower concentrations of inorganic chemicals than
leachate from MSW landfills. However, the trace metals barium, lead, and zinc can be
detected  at higher concentrations in C&DW leachate than in MSW leachate. C&DW
                                  E-40
-------
contains organic chemicals as evidenced by the relatively high TOC and COD levels.
Based on the relatively low BOD level for C&DW leachate, these organic chemicals are
not readily biodegradable.  The specific organic chemicals present in C&DW leachate
were not analyzed.

E-3    Data Collection and Reduction

E-3.1    Overview

The LCRS and LDS flow rate and flow chemistry data presented in this appendix were
obtained from engineering drawings, project specifications, as-built records, and
operation records, and interviews with facility owners, facility operators,  design
engineers, and regulatory agencies.  The data were collected in accordance with a
quality assurance project plan, which was reviewed and approved by the EPA.  Efforts
were made to obtain data from a wide variety of facilities,  and to obtain as complete a
record of data as possible, from construction of each facility to the time of data
collection.

A database was developed that includes information and data for 187 individual landfill
cells at 54 double-lined landfills.  The data collected includes: (i) general facility
information (including  location, average annual rainfall,  subsurface soil types, ground-
water separation distance from bottom of landfill), summarized in Table E-3.1; (ii)
general  cell information (including cell area, type of waste, height of waste, dates of
construction, operation, and closure), summarized in Table E-3.2; (iii) double-liner
system design details  (including type, thickness, and hydraulic conductivity of each
layer), summarized in  Table E-3.3; (iv) final cover system design details  (including type,
thickness,  and hydraulic conductivity of each layer) summarized in Table E-3.4; (v)
LCRS and LDS flow rate data, summarized in Tables E-3.5 and E-3.6, respectively; and
(vi) LCRS and LDS chemical constituent data, summarized in Table E-3.7.

E-3.2    General Description of Cells

The distribution of landfills and cells in the database by waste type and geographic
region is shown in Table E-3.8(a).  As shown in this table, most of the landfills in the
database are located in NE.  This is not surprising because: (i) the NE has a relatively
dense population; and (ii) double-liner systems are required for MSW landfills in several
states in the NE. In addition, the majority of the landfills in the database are used for
disposal of MSW. Based on the extent of the database and comparisons of these data
with the published data presented  in Section E-2, the database appears to adequately
characterize conditions for MSW landfills in the NE and SE, HW landfills in the NE and
SE, and MSW ash landfills in the NE. The database  is  quite sparse for landfills in the
W, coal  ash landfills, and C&DW landfills. Additional data from these facilities should be
collected and evaluated.
                                   E-41
-------
Table E-3.1.  Landfill Site Information.
Landfill
Designation
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
V
w
X
Y
z
AA
AB
AC
AD
AE
AF
AG
AH
Al
AJ
AK
AL
AM
Region
ofU.S.(1)
NE
NE
SE
SE
NE
NE
NE
NE
NE
NE
NE
SE
SE
SE
SE
SE
NE
NE
NE
NE
NE
SE
SE
SE
NE
NE
NE
SE
W
SE
NE
NE
NE
NE
NE
NE
NE
NE
W
Average
Annual
Rainfall
(mm)
1120
1070
1120
1630
1040
580
910
1020
990
760
1120
1140
1470
1520
1500
1500
890
1040
1040
1300
1070
1730
1500
1020
1190
1250
1140
1700
280
1830
1210
1120
840
840
1140
1250
760
790
430
Depth to
Ground
Water
(m)
ND(2)
3
2
2
5
2
2
0
1
ND
ND
6
ND
ND
ND
ND
<2
3
4
20
1
ND
ND
ND
ND
ND
<2
ND
91
0
3-7
2
<2
4
ND
ND
ND
3
14
Subsurface
Soil
Type
Silt & Sand
Coarse Sand
Sand & Clay
ND
ND
Sand
ND
ND
Silty Clay
Sand
Sand
Clay
ND
Sand
Sand
ND
Glacial Till
Fine Sand & Silt
Fine Sand
Silt & Silty Sand
ND
ND
Fine Sand / Clay Layers
ND
Glacial Till
Silt & Clay
Glacial Till
Chalk
Clay / Sandstone
Sandy Silt /Clay
Clay & Silt
ND
Clay
Silty Sand
Sand
Silt & Clay
Silt & Clay
Sand & Gravel
Silt
                               E-42
-------
Table E-3.1.  Landfill Site Information (Continued).
Landfill
Designation
AN
AO
AP
AQ
AR
AS
AT
AU
AV
AW
AX
AY
AZ
BA
BB
Region
ofU.S.(1)
NE
NE
W
NE
NE
SE
W
NE
NE
SE
NE
NE
NE
NE
SE
Average
Annual
Rainfall
(mm)
1120
830
380
970
1140
1420
740
1020
790
1120
1040
860
760
860
1090
Depth to
Ground
Water
(m)
5
3
> 1500
0
<2
1
3
3
3
2
1
ND
2
14
2
Subsurface
Soil
Type
Sand
Silty Sand
Shale
ND
Clay
ND
Clay
Clay & Sand
Silty Clay
Silty Sand
Sand & Gravel
ND
Sand
Silty & Sandy Clay
ND
Notes:  (1)  Region of the U.S. (See Figure E-1.3):  NE = northeast,
           SE = southeast, W = west.
       (2) ND = not determined.
                                E-43
-------
Table E-3.2. General Landfill Cell Construction and Operation Information.
Cell
No.
A1
A2
B1
B2
B3(4)
B4
B5
C1
C2
C3
C4
C5
C6
D1
D2
D3
D4
E1
E2
E3
E4
F1
G1
G2
G3
H1
H2
H3
H4
H5
H6
11
12
13
14
15
J1
J2
J3
J4
J5
J6
Waste
Type'1'
C&DW
C&DW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
HW
HW
HW
HW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
HW
HW
HW
HW
HW
HW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
Cell
Area
(ha)
5.3
6.4
3.3
3.5
6.4
2.9
3.9
3.2
3.7
3.6
3.7
2.6
3.6
0.4
0.3
0.3
0.4
2.4
2.4
1.2
1.2
1.8
3.0
1.6
1.7
0.5
1.1
0.9
1.2
0.8
1.3
3.2/2. 7(5)
4.2/2. 3(5)
3.4/1. 8(5)
4.7
4.7
1.0
1.0
1.0
1.0
1.0
1.0
Max.
Waste
Height
(m)
79
79
21
21
25
25
25
24
40
46
46
43
40
7.9
7.9
7.9
7.9
40
40
40
40
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
4.9
7.9
7.9
ND
ND
ND
ND
ND
ND
ND
ND
Avg.
Liner
Base
Slope
(%)
4.0
4.0
2.0
2.0
2.0
2.0
2.0
2.5
2.5
2.5
2.5
3.0
3.0
2.0
2.0
2.0
2.0
1.3
1.3
1.3
1.3
5.0
2.0
2.0
2.0
3.0
3.0
3.0
3.0
3.0
3.0
2.0
2.0
2.0
ND
ND
ND
ND
ND
ND
ND
ND
Final
Cover
Top/Side
Slopes
(%)
NA(3)
NA
5/25
5/25
5/25
NA
NA
NA
NA
NA
NA
NA
NA
ND(2)
ND
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
4/14
4/14
4/14
4/14
4/14
NA
5/33
5/33
5/33
5/33
5/33
ND
ND
ND
ND
NA
NA
End of
Construct.
Date
1989-90
Sep-92
May-84
May-84
Jul-87
Apr-91
May-92
Apr-90
Dec-90
Mar-91
Dec-91
Apr-92
May-93
Sep-85
Sep-85
Jun-87
Jun-87
Mar-88
Oct-87
May-90
Jul-90
Jul-92
May-89
May-89
Nov-92
1985
1986
1988
1990
1991
1992
Aug-87
Oct-87
Apr-88
May-92
Jul-92
Jul-90
Sep-90
Aug-91
Aug-91
Nov-92
Nov-92
Waste
Placement
Start
Date
1989-90
Oct-92
May-84
May-84
Jul-87
May-91
May-92
May-90
Apr-91
Aug-91
Feb-92
Nov-92
Aug-93
Oct-85
Jan-86
Jul-87
Jan-89
Mar-88
Oct-87
May-90
Jul-90
Jul-92
Jun-89
Jun-89
Dec-92
1985
1986
Jul-88
Dec-90
Sep-91
Nov-92
Aug-87
Oct-87
Apr-88
May-92
Jul-92
Oct-90
Apr-91
Nov-91
Jul-92
Jun-93
NA
Final
Closure
Date
NA
NA
Nov-88
Nov-88
NA
NA
NA
NA
NA
NA
NA
NA
NA
May-86
Mar-88
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
1990
NA
NA
NA
NA
NA
Oct-94
Oct-94
Oct-94
Jul-94
May-94
Fall 93
Fall 93
Fall 93
Fall 93
NA
NA
3rd Party CQA?
Liner
System
N
N
N
N
N
N
N
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
N
N
N
N
N
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Cover
System
NA
NA
Y
Y
NA
NA
NA
NA
NA
NA
NA
NA
NA
ND
ND
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
ND
NA
NA
NA
NA
NA
Y
Y
Y
Y
Y
Y
Y
Y
Y
NA
NA
Flow
Meas.
Methods
for LCRS
and LDS(2)
ND(3)
ND
5
5
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
ND
2,3
2,3
2
ND
ND
ND
ND
ND
ND
1
1
1
ND
ND
ND
ND
ND
ND
ND
ND
                                        E-44
-------
Table E-3.2.  General Landfill Cell Construction and Operation Information (Continued).
Cell
No.
K1
L1
L2
L3
L4
M1
N1
N2
O1
O2
O3
P1
P2
P3
Q1
Q2
R1
S1
S2
T1
T2
U1
U2
U3
U4
V1
V2
V3
V4
V5
W1-2
X1
Y1
Y2
Z1
AA1
AA2
AA3
AB1
AB2
AB3
AB4
Waste
Type'1'
MSW
HW
HW
HW
HW
ASH(M)
MSW
MSW
MSW
MSW
MSW
ASH(M)
ASH(M)
ASH(M)
MSW
MSW
MSW
MSW
ASH(M)
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
ASH(M)
MSW
ASH(C)
MSW
MSW
MSW
HW
HW
HW
HW
Cell
Area
(ha)
2.7
1.5
1.5
1.5
1.5
4.0
4.4
6.3
4.2
4.9
4.7
1.8
2.4
2.3
4.5
1.8
2.0
2.0
1.6
3.8
3.8
1.9
2.6
1.5
1.9
4.2
3.9
4.0
3.7
3.3
15.4
3.0
2.2
3.0
2.6
1.8
2.6
3.4
3.8
4.1
5.0
5.8
Max.
Waste
Height
(m)
15
7.6
7.6
7.6
7.6
ND
ND
ND
7.9
7.9
ND
ND
ND
ND
30
30
ND
ND
ND
21
21
26
26
24
23
ND
ND
ND
ND
ND
24
ND
ND
ND
23
37
37
37
>90
>90
>90
>90
Avg.
Liner
Base
Slope
(%)
4.5
2.0
2.0
2.0
2.0
2.0
2.0
2.0
3.2
3.2
3.2
2.0
2.0
2.0
2.0
2.0
3.5
0.5
0.5
2.0
2.0
2.9
3.0
ND
ND
4.0
4.0
4.0
4.0
4.0
1.0
2.0
5.5
5.5
3.2
2.0
2.0
2.0
3.0
3.0
3.0
3.0
Final
Cover
Top/Side
Slopes
(%)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
ND
ND
ND
ND
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
End of
Construct.
Date
Oct-89
Jun-90
Apr-93
May-90
Oct-91
Late 90
1988
1991
Aug-88
Feb-89
Jan-94
1991
1992
Late 93
Mar-90
Dec-93
1993
Sep-90
Aug-90
Jan-91
Jan-92
Jul-86
Jan-87
Nov-87
Mar-88
1989
1989
1990
1990
1993
May-92
1992
Sep-88
1990
Dec-91
Oct-90
Jul-90
Sep-91
1986
1986
Oct-87
Apr-89
Waste
Placement
Start
Date
Dec-89
Aug-90
Jan-94
Jun-90
Feb-92
Sep-91
1988
Jan-92
Mar-88
Mar-89
Feb-94
1991
Late 92-93
Late 93
Mar-90
Mar-94
May-93
Sep-90
Aug-90
May-91
Jan-92
Jul-86
Jan-87
Nov-87
Mar-88
Jan-90
Jan-90
Nov-90
Dec-90
May-93
May-92
Aug-92
Jan-89
Jan-91
Mar-92
Oct-90
Jul-90
Mar-92
1987
1987
Feb-88
May-89
Final
Closure
Date
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Sep-88
Sep-88
Sep-88
Oct-88
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
3rd Party CQA?
Liner
System
N
Y
Y
Y
Y
ND
ND
Y
Y
Y
Y
Y
Y
Y
Y
Y
N
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
N
Y
Y
Y
Y
Y
Y
Y
Cover
System
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
N
N
N
N
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Flow
Meas.
Methods
for LCRS
and LDS(2)
ND
ND
ND
ND
ND
ND
ND
ND
2
2
2
ND
ND
ND
ND
ND
ND
ND
ND
3
3
1,3
1,3
1,3
1,3
ND
ND
ND
ND
ND
ND
ND
1
1
ND
ND
ND
ND
1
1
1
1
                                       E-45
-------
Table E-3.2.  General Landfill Cell Construction and Operation Information (Continued).
Cell
No.



AC1
AC2
ACS
AC4
ACS
AC6
AC7
ACS
AD1
AD2
ADS
AD4
ADS
AD6
AD7
ADS
AD9
AD10
AD11
AD12
AD13
AD14
AD15
AD16
AD17
AD18
AE1
AE2
AE3
AF1
AG1
AH1
AH2
AH 3
AH4
AH 5
AH
AI2
AJ1
AK1
AK2
AL1
AM1
AM2
Waste
Type'1'



HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
MSW
MSW
MSW
ASH(M)
MSW
MSW/ASH(M)
MSW/ASH(M)
MSW/ASH(M)
MSW/ASH(M)
MSW/ASH(M)
C&DW
C&DW
MSW
MSW
MSW
MSW
MSW
MSW
Cell
Area
(ha)


4.6
4.2
4.6
3.8
6.9
6.9
4.0
6.1
0.6
0.6
0.6
0.6
0.5
0.9
1.5
1.5
1.3
1.3
1.1
1.1
1.1
1.1
1.1
1.1
1.7
1.7
4.5
5.3
4.0
2.2
8.1
1.2
1.1
1.1
1.1
1.0
3.5
2.8
3.6
1.4
1.6
14.9
3.2/2.4(5)
4.8/2.4(5)
Max.
Waste
Height
(m)

30
30
30
30
ND
ND
ND
ND
21
21
21
21
21
21
24
24
24
24
24
24
24
24
24
24
24
24
35
35
35
26
30
ND
ND
ND
ND
ND
49
49
ND
34
34
68
27
27
Avg.
Liner
Base
Slope
(%)
2.0
2.0
2.0
2.0
ND
ND
ND
ND
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.5
2.0
2.0
2.0
2.2
5.0
3.0
3.0
3.0
3.0
3.0
2.5
2.5
2.0
1.0
1.0
1.0
6.0
6.0
Final
Cover
Top/Side
Slopes
(%)
NA
NA
NA
NA
NA
NA
NA
NA
5/20
5/20
5/20
5/20
5/20
5/20
5/20
5/20
5/20
5/20
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
End of
Construct.
Date


Feb-87
Dec-87
Jun-89
Jul-89
Mar-92
Mar-92
Dec-93
Dec-93
May-85
Jul-85
Nov-85
Nov-85
Jun-86
Jun-86
Sep-87
Apr-88
Dec-88
Dec-88
Aug-89
Aug-89
Oct-90
Oct-90
Apr-92
Apr-92
Mar-94
Nov-93
Apr-88
Nov-90
Mar-94
Nov-89
Mar-92
Apr-91
Apr-91
Apr-91
Apr-91
Apr-91
Nov-90
Nov-90
Mar-94
May-93
Oct-94
1988-89
Sep-90
Sep-90
Waste
Placement
Start
Date

Mar-87
Mar-88
Jun-89
Jul-89
Mar-92
Mar-92
Jan-94
Jan-94
May-85
Jul-85
Nov-85
Nov-85
Jun-86
Jun-86
Sep-87
Apr-88
Dec-88
Dec-88
Aug-89
Aug-89
Oct-90
Oct-90
Apr-92
Apr-92
Mar-94
Nov-93
May-88
Dec-90
May-94
Jan-90
Apr-92
Nov-91
Nov-91
Nov-91
Nov-91
Nov-91
Nov-90
Nov-90
Jun-94
Oct-93
Oct-94
1990
Oct-90
Oct-90
Final
Closure
Date


NA
NA
NA
NA
NA
NA
NA
NA
Jul-88
Jul-88
Jul-88
Jul-88
Jul-88
Jul-88
Oct-93
Oct-93
Oct-93
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
3rd Party CQA?
Liner
System


Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
N
N
N
Y
Y
Y
Y
Y
Cover
System


NA
NA
NA
NA
NA
NA
NA
NA
Y
Y
Y
Y
Y
Y
Y
Y
Y
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Flow
Meas.
Methods
for LCRS
and LDS(2)
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
2
2
2
3
ND
ND
                                       E-46
-------
Table E-3.2.  General Landfill Cell Construction and Operation Information (Continued).
Cell
No.



AN1
AN2
AN 3
AN4
AO1
AO2
AO3
AO4
AP1
AQ1
AQ2
AQ3
AQ4
AQ5
AQ6
AQ7
AQ8
AQ9
AQ10
AR1
AR2
AS1
AT1
AU1
AV1
AV2
AV3
AV4
AV5
AW1
AW2
AW3
AW4
Waste
Type'1'



ASH(M)
ASH(M)
ASH(M)
ASH(M)
MSW
MSW
MSW
MSW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
MSW
MSW
HW
HW
MSW
HW
HW
HW
HW
HW
MSW
MSW
MSW
MSW
Cell
Area
(ha)


1.2
0.5
0.5
0.2
1.8
1.8
1.8
1.8
1.9
0.6
0.5
0.5
0.6
0.7
0.5
0.8
0.8
0.5
0.9
9.7
2.0
1.0
2.6
2.9
3.1
2.4
2.5
2.0
2.9
2.4
2.4
2.4
2.4
Max.
Waste
Height
(m)

52
52
52
52
23
23
23
23
13
21
21
21
21
21
21
21
21
21
21
11
11
9.8
11
23
20
20
20
20
23
21
21
21
21
Avg.
Liner
Base
Slope
(%)
2.5
ND
ND
0.80
1.0
1.0
1.0
1.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
3.0
1.5
1.5
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
Final
Cover
Top/Side
Slopes
(%)
NA
NA
NA
NA
NA
NA
NA
NA
NA
12/33
12/33
12/33
12/33
12/33
12/33
12/33
12/33
12/33
12/33
NA
NA
3/29
10/33
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
End of
Construct.
Date


Jun-91
Jun-92
Jun-92
Sep-93
Jan-92
Jul-92
Sep-93
Sep-93
Dec-89
Mar-86
Mar-86
Mar-86
Mar-86
Sep-86
Sep-86
Sep-86
Sep-86
Sep-86
Sep-86
1992
1995
Mar-89
Dec-87
Sep-90
Sep-86
Oct-87
Jun-88
Oct-89
Aug-91
Apr-93
Jul-93
May-94
Jun-94
Waste
Placement
Start
Date

Jun-91
Jun-92
Nov-92
Sep-93
Jan-92
Jul-92
Jun-94
Jun-94
Jul-91
Mar-86
Mar-86
Jul-86
Aug-86
Apr-87
Apr-87
Apr-87
Jan-89
Jan-89
Jan-89
Mar-92
Mar-95
Jul-89
Apr-88
Jan-91
Sep-86
Oct-87
Mar-89
Nov-89
Jan-92
May-93
Aug-93
May-94
Aug-94
Final
Closure
Date


NA
NA
NA
NA
NA
NA
NA
NA
NA
Early 90
Early 90
Early 90
Early 90
Mid 91
Mid 91
Mid 91
Mid 91
Mid 91
Mid 91
NA
NA
Oct-91
Nov-88
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
3rd Party CQA?
Liner
System


Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
N
N
N
Y
Y
Y
Y
Y
Y
Cover
System


NA
NA
NA
NA
NA
NA
NA
NA
NA
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
NA
NA
Y
Y
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Flow
Meas.
Methods
for LCRS
and LDS(2)
ND
ND
ND
ND
2
2
2
2
1,4
2
2
2
2
2
2
2
2
2
2
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
2
2
2
2
                                       E-47
-------
Table E-3.2.  General Landfill Cell Construction and Operation Information (Continued).
Cell
No.



AX1
AX2
AX3
AX4
AX5
AX6
AX7
AX8
AX9
AX10
AX11
AX12
AX13
AX14
AX15
AX16
AY1
AY2
AYS
AZ1
BA1
BA2
BB1
BB2
BBS
Waste
Type'1'



MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
HW
HW
HW
MSW
HW
HW
MSW
MSW
MSW
Cell
Area
(ha)


2.0
2.0
1.7
1.7
2.8
3.9
2.6
3.8
3.3
3.9
3.0
4.0
3.0
2.8
2.8
4.5
1.3
1.0
1.0
3.8
3.0
3.2
4.0
2.4
2.8
Max.
Waste
Height
(m)

24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
14
14
14
41
27
27
28
30
30
Avg.
Liner
Base
Slope
(%)
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
4.0
4.0
2.0
2.0
2.0
Final
Cover
Top/Side
Slopes
(%)
6/33
6/33
6/33
6/33
6/33
6/33
NA/33
NA/33
NA/33
NA/33
NA/33
NA/33
NA/33
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
End of
Construct.
Date


Jun-88
Jun-88
Aug-88
Aug-88
Sep-88
Dec-88
Jan-89
Jul-89
Dec-89
Feb-90
Feb-90
Oct-90
Jan-91
Apr-92
May-92
Jan-93
Jul-94
Jul-94
Jul-94
Sep-92
Jun-91
Nov-93
Dec-90
Jan-93
Jan-93
Waste
Placement
Start
Date

Jul-88
Jul-88
Sep-88
Sep-88
Oct-88
Dec-88
Feb-89
Jul-89
Dec-89
Jul-90
Feb-90
Oct-90
Jan-91
Apr-92
May-92
Jan-93
Oct-94
Aug-94
Aug-94
Dec-92
Oct-91
Apr-94
Feb-91
Jan-93
Jan-93
Final
Closure
Date


Feb-91
Feb-91
Apr-93
Apr-93
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
3rd Party CQA?
Liner
System


Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Cover
System


N
N
N
N
N
N
N
N
N
N
N
N
N
NA
NA
NA
NA
NA
NA
NA
Y
Y
NA
NA
NA
Flow
Meas.
Methods
for LCRS
and LDS(2)
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
ND
ND
ND
2
2
2
Notes:
(1) C&DW = construction and demolition waste, MSW = municipal solid waste, HW = hazardous waste.
   ASH(M) = MSW ash, ASH(C) = coal ash.
(2) Codes for method of flow measurement:
   1 = Automatic pumping system, liquid volume recorded from accumulating flow meter
   2 = Periodic pumping if liquid present in sump, volume recorded from accumulating flow meter
   3 = Periodically measure time to fill a known volume
   4 = Automatic pumping system from sump to a holding tank, volume transferred from holding tank measured
   5 = Periodic pumping if liquid present in sump, volume estimated from change in liquid level
   6 = Automatic pumping system, liquid volume estimated by multiplying pump capacity x time
(3) NA = not applicable, ND = not determined.
(4) Final cover system was  installed over 65% of Cell B3 at 65 months after start of waste placement.
(5) Values represent LCRS and LDS areas, respectively.
                                              E-48
-------
Table E-3.3. Landfill Double-Liner System Details.
Cell
No.
A1
A2
B1-2
B3
B4-5
C1-5
C6
D1-4
E1-4
F1
G1-2
G3
H1-6
11-3
I4-5, base
I4-5, sides
J1-4
J5-6
K1
L1-4, base
L1-4, sides
M1
N1
N2
O1 -3
P1-3, base
P1-3, sides
Q1-2
R1
S1-2
T1-2
LCRS
Avg. Pipe
or Swale
Spacing
(m)
17
17
38
30
ND(3)
91
91
37
46
37
46
23
46
30
ND
ND
ND
ND
ND
20
NA
ND
ND
ND
NA
NA
NA
46
ND
ND
30
Pipe or
Swale Size
(mm or m)
& Material
1 50 PVC
1 50 PVC
1 50 PVC
1 50 PVC
ND
200 PVC
200 HOPE
100 HOPE
ND
1 50 PVC
150 HOPE
1 50 PVC
1 50 PVC
ND
ND
ND
PVC
PVC
200 HOPE
ND
NA
ND
ND
ND
NA
NA
NA
1 00 PVC
150 HOPE
ND
1 50 PVC
Drainage Layer(s)
Material
S
S
S
S
S
S
S
S
S
S
S
S/G
S
S
TC/G
NA
S
S
S
S
S
S
S
S
G/S
S/GN
S/GN
S
G
S/GN
G/S
Thick.
(mm)
600
600
450
450
450
600
600
300
600
600
600
600/600
300
600
1 50/450
NA
600
600
600
300
300
600
600
600
300/300
600/5
600/5
450
600
600/5
1 50/450
Primary Liner
Type of
Liner(1)
GM
GM
GM
GM/CCL
GM/CCL
GM
GM/GCL
GM
GM
GM
GM
GM/CCL
GM
GM
GM/GCL
GM/GCL
GM
GM
GM
GM/CCL(4)
GM
GM
GM
GM
GM
GM
GM
GM
GM/CCL
GM
GM
GM
Type'2'
HOPE
HOPE
CSPE
CSPE
HOPE
HOPE
HOPE
HOPE
CSPE
HOPE
HOPE
HOPE
PVC
HOPE
HOPE
HOPE
PVC
PVC
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
Thick.
(mm)
1.5
2.0
0.9
0.9
1.5
2.0
2.0
2.0
0.9
1.5
1.5
1.5
0.8
1.5
1.5
1.5
1.5
0.8
2.0
1.5
1.5
1.5
2.5
1.5
2.0
1.5
1.5
1.5
1.5
2.0
1.5
GCL/
CCL
Thick.
(mm)
NA(3)
NA
NA
600
600
NA
6
NA
NA
NA
NA
450
NA
NA
6
6
NA
NA
NA
300
NA
NA
NA
NA
NA
NA
NA
NA
450
NA
NA
LDS
Avg. Pipe
or Swale
Spacing
(m)
17
17
38
30
ND
91
91
37
46
37
46
23
46
30
ND
NA
ND
ND
ND
20
NA
ND
NA
NA
ND
ND
ND
46
ND
ND
NA
Pipe or
Swale Size
(mm or m)
& Material
1 50 PVC
1 50 PVC
1 50 PVC
1 50 PVC
ND
1 00 PVC
1 50 PVC
100 HOPE
ND
1 50 PVC
100 HOPE
1 00 PVC
1 50 PVC
ND
ND
NA
PVC
PVC
150 HOPE
ND
NA
ND
NA
NA
ND
ND
ND
1 00 PVC
150 HOPE
ND
NA
Drainage Layer(s)
Material
S
S
S
S
GN
S
S
S
S
S
S
S
S
S
GN
GN
S
S
S/GN
S
S
S
GN
GN
S/GN
GN
GN
S
GN
GN
GN
Thick.
(mm)
600
600
450
450
5
450
450
300
600
300
300
300
300
450
10
5
ND
ND
300/5
300
300
300
5
5
300/5
5
5
300
10
5
13
Secondary Liner
Type of
Liner(1)
GM
GM
GM
GM
GM
GM/CCL
GM/CCL
CCL/GM
GM/CCL
GM
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM
GM
GM/CCL
GM/CCL
GM/CCL
GM
CCL
GM/CCL
GM/CCL
GM/GCL/CCL
GM/CCL
CCL
GM/CCL
GM/CCL
GM/CCL
GM
Type'2'
PVC
PVC
PVC
PVC
HOPE
HOPE
HOPE
HOPE
PVC
HOPE
HOPE
HOPE
PVC
HOPE
HOPE
HOPE
PVC
PVC
LLDPE
HOPE
HOPE
HOPE
NA
HOPE
HOPE
HOPE
HOPE
NA
HOPE
HOPE
HOPE
Thick.
(mm)
1.5
1.5
0.8
0.8
1.5
2.0
2.0
1.0
0.8
1.5
1.5
1.5
0.8
1.5
1.5
1.5
0.8
0.8
1.5
1.5
1.5
1.5
NA
1.5
1.0
1.5
1.5
NA
1.5
2.0
1.5
GCL/
CCL
Thick.
(mm)
NA
NA
NA
NA
NA
300
300
900
600
NA
600
600
900
300
150
150
NA
NA
600
900
900
NA
300
300
150
150
150
600
600
600
150
-------
Table E-3.3. Landfill Double-Liner System Details (Continued).
Cell
No.
U1
U2
U3-4
V1-5
W1-2
X1
Y1-2
Z1
AA1-3
AB1-4, base
AB1-4, sides
AC1-4, base
AC1-4, sides
AC5-8
AD1-6, base
AD1-6, sides
AD7-18, base
AD7-18, sides
AE1-3
AF1
AG1
AH 1-5
AM -2, base
AM -2, sides
AJ1
AK1-2
AL1
AM 1-2
AN1, base
AN1, sides
AN2-3, base
AN2-3, sides
AN4
AO1-4
LCRS
Avg. Pipe
or Swale
Spacing
(m)
30
24
ND
ND
91
NA
30
35
46
ND
NA
ND
NA
ND
100
NA
NA
NA
46
15
46
46
15
15
30
61
30
NA
NA
NA
NA
NA
NA
61
Pipe or
Swale Size
(mm or m)
& Material
1 00 PVC
1 00 PVC
ND
ND
200 HOPE
NA
1 50 PVC
1 50 PVC
1 50 PVC
ND
NA
ND
NA
ND
ND
NA
ND
NA
150 HOPE
150
1 50 PVC
1 50 PVC
1 50 PVC
1 50 PVC
1 50 PVC
200 HOPE
200 PVC
NA
NA
NA
NA
NA
NA
150 HOPE
Drainage Layer(s)
Material
S
S
S
S/GN
S/GN
S
S
S
S
G
GN
S
S
GN
S
GN
S
GN
S
S
G/S
S
S
S
S
S/GN
S
G/S
S/GN
S/GN
S/GN
S/GN
S/GN
S
Thick.
(mm)
600
600
600
600/5
600/5
600
600
600
600
300
5
300
300
5
300
5
300
5
300
5
1 50/300
450
600
600
600
600/5
600
300/1 50
600/5
450/5
600/5
450/5
600/5
600
Pr
Type of
Liner(1)
GM
GM
GM
GM
GM
GM
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM
GM/CCL
GM/GCL/CCL
GM/CCL
GM/CCL
GM/CCL
GM(5)
GM/CCL
GM
GM
GM/CCL
mary Liner
GM
Type'2'
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
LMDPE
LMDPE
HOPE
HOPE
HOPE
HOPE
Thick.
(mm)
1.5
1.5
1.5
1.5
1.5
1.5
2.0
1.5
2.0
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
2.0
1.5
2.0
2.0
2.0
2.0
1.5
1.5
1.5
2.0
2.0
2.0
2.0
2.0
2.0
1.5
GCL/
CCL
Thick.
(mm)
NA
NA
NA
NA
NA
NA
450
450
450
450
450
450
450
450
900
900
900
900
600
450
450
450
450
NA
450
6/600
900
450
450
NA
450
NA
NA
900
LDS
Avg. Pipe
or Swale
Spacing
(m)
30
24
ND
ND
NA
ND
30
ND
46
ND
NA
ND
NA
ND
30
NA
53
NA
46
61
46
46
15
NA
30
61
ND
NA
ND
ND
ND
ND
ND
NA
Pipe or
Swale Size
(mm or m)
& Material
1 00 PVC
1 00 PVC
ND
ND
NA
ND
1 50 PVC
1 50 PVC
1 50 PVC
ND
NA
ND
NA
ND
ND
NA
ND
NA
100 HOPE
150
1 50 PVC
1 50 PVC
1 50 PVC
NA
1 50 PVC
200 HOPE
1 50 PVC
NA
ND
ND
ND
ND
ND
NA
Drainage Layer(s)
Material
GN
GN
GN
S/GN
GN
GN
S
S
S
S
S
G/GN
GN
GN
S
GN
GN
GN
S
S
S
S
S
GN
S
GN
GN
GN
GN
GN
GN
GN
GN
GN
Thick.
(mm)
5
5
5
300/5
5
5
300
300
300
300
300
300/5
5
5
300
5
5
5
300
300
300
300
300
5
300
5
5
5
5
5
5
5
5
5
Secondary Liner
Type of
Liner(1)
GM
GM
GM
GM/CCL
GM/GCL/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
CCL
GM/CCL
GM
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/GCL
GM/CCL
GM
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM
Type'2'
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
NA
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
Thick.
(mm)
1.5
1.5
1.5
1.5
1.5
1.5
2.0
1.5
2.0
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
NA
2.0
2.0
1.5
1.5
1.5
1.5
1.5
1.5
1.5
2.0
2.0
2.0
2.0
2.0
1.5
GCL/
CCL
Thick.
(mm)
NA
NA
NA
150
6/150
150
600
600
600
900
900
900
900
900
900
900
900
900
600
900
NA
600
600
600
600
6
900
NA
600
900
600
900
600
600
-------
Table E-3.3.  Landfill Double-Liner System Details (Continued).
Cell
No.
AP1, base
AP1, sides
AQ1-4
AQ5-1 0
AR1-2
AS1, base
AS1, sides
AT1 , base
AT1 , sides
AU1-2
AV1-4
AV5, base
AV5, sides
AW1-2
AW3-4
AX1-16, base
AX1 -16, sides
AY1-3
AZ1
BA1-2
BB1-3
LCRS
Avg. Pipe
or Swale
Spacing
(m)
ND
ND
61
61
46
NA
NA
122
NA
30
30
30
NA
91
91
30
NA
ND
76
12
61
Pipe or
Swale Size
(mm or m)
& Material
HOPE
HOPE
ND
ND
250 PVC
NA
NA
2.4 x 0.3 G
NA
1 50 PVC
150 HOPE
150 HOPE
NA
150 HOPE
150 HOPE
150 HOPE
NA
ND
1 50 PVC
100 HOPE
150
Drainage Layer(s)
Material
S
GN
G/GN
G/GN
G/TC
GN
GN
S
GN
G
S
S
GN
S
S
G
G
GN
S
S
S
Thick.
(mm)
300
5
300/5
300/5
300/450
5
5
300
5
600
300
300
8
450
450
600
600
5
600
450
600
Primary Liner
Type of
Liner(1)
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/GCL/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/GCL
GM
GM/GCL
GM/GCL
GM/GCL
GM/GCL
GM/GCL/GM
GM/GCL
GM
Type'2'
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
Thick.
(mm)
2.0
2.0
2.0
2.0
1.5
2.0
2.0
1.5
1.5
1.5
2.0
2.0
2.0
1.5
1.5
1.5
1.5
1.5
1.5
2.0
1.5
GCL/
CCL
Thick.
(mm)
900
1400
450
450
6/300
900
600
450
450
900
1500
1500
1500
6
NA
6
6
6
250
6
6
LDS
Avg. Pipe
or Swale
Spacing
(m)
NA
NA
61
61
46
NA
NA
122
NA
NA
30
30
NA
91
91
30
NA
ND
NA
ND
ND
Pipe or
Swale Size
(mm or m)
& Material
NA
NA
ND
ND
1 00 PVC
NA
NA
1.2x0.3G
NA
NA
ND
25x61 OGN
NA
150 HOPE
150 HOPE
1 00 PVC
NA
ND
NA
25 GN
ND
Drainage Layer(s)
Material
GN
GN
G/GN
G/GN
GN
GN
GN
GN
GN
GN
GN
GN
GN
S
GN
G
GN
GN
GN
GN
GN
Thick.
(mm)
5
5
300/5
300/5
10
5
5
5
5
6
10
8
8
300
5
300
5
5
5
5
5
Secondary Liner
Type of
Liner(1)
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM
GM(6)
GM(6)
GM(6)
GM/GCL
GM/GCL/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM
GM
Type'2'
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
Thick.
(mm)
2.0
2.0
1.5
2.0
1.5
2.0
2.0
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
2.0
1.5
GCL/
CCL
Thick.
(mm)
900
900
900
900
600
900
900
900
900
NA
NA
NA
NA
NA
450
150
150
900
900
900
NA
     Notes: (1) GM = geomembrane, GCL = geosynthetic clay liner, CCL = compacted clay liner
              G = gravel, S = sand, GN = geonet,  TC = tire chips
           (2) HOPE = high-density polyethylene,  CSPE = chlorosulfonated polyethylene
              PVC = polyvinyl chloride, LLDPE = linear low-density polyethylene
              LMDPE = linear medium-density polyethylene
           (3) NA = not applicable, ND = not determined
           (4) GM and CCL are separated by a geotextile
           (5) HOPE GM on upper side slopes
           (6) Cells AV1 -5 sit on a clay formation
-------
Table E-3.4.  Landfill Final Cover System Details.
Cell
Number
B1-2, top
B1-2, sides
D1-2
H1,top
H1, sides
11-5
J1-4
U1-4
AD1-6
AD7-9
AQ1-10
AS1
AT1
AX1-4, top
AX1-4,sides
Protective
Soil
Thickness
(mm)
300
300
600
690
690
600
ND
600
600
600
600
450
300
600
600
Drainage Layer(s)
Type
S
S
GN/S
S
GT
GN
ND
NA
GN
GN
ND
NA
S
GN
S
Thickness
(mm)
450
150
5/150
200
ND(3)
5
ND
NA
5
5
ND
NA
200
5
150
Barrier Layer
Type<1>
GM
CCL
CCL/GM(4)
GM/CCL
GM/CCL
GM
GM
CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM/CCL
GM
CCL
GM
Material(2)
HOPE
NA(3)
HOPE
PVC
PVC
HOPE
HOPE
NA
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
NA
Thickness
(mm)
1.0
NA
2.0
1.0
1.0
1.0
ND
NA
1.5
1.5
1.5
1.5
1.0
1.0
NA
Soil
Thickness
(mm)
NA
300
600
600
600
NA
NA
300
600
600
900
900
600
NA
300
Notes:
 (1) GM = geomembrane, CCL = compacted clay liner, GN = geonet, GT = geotextile, S = sand
 (2) HOPE = high-density polyethylene,  PVC = polyvinyl chloride.
 (3) NA = not applicable, ND = not determined.
 (4) Barrier consists of a CCL underlain by a GM.
                                      E-52
-------
       Table E-3.5.  Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage.

Cell
No.

A1-2



B1





B2






B3








B4



B5



Waste
Type
(Region)
C&D(NE)



MSW(NE)





MSW(NE)






MSW(NE)








MSW(NE)



MSW(NE)


Primary
Liner/
LDS
Type
I



I





I






III








IV



IV


Initial Period of Operation
Time
Period01
(mos.)
1-24



1-19





1-19






1-4








1-12



1-12


Avg.
Flow
(Iphd)
ND



ND





ND






15,304








2,930



8,005


Peak
Flow
(Iphd)
ND



ND





ND






24,858








6,353



19,521


Active Period of Operation
Time
Period
(mos.)
25-38
39-42
43-48
Entire Period
20-31
32-43
44-54


Entire Period
20-31
32-43
44-54



Entire Period
5-16
17-28
29-40
41-52
53-64
65-76
77-88
89-93
Entire Period
13-24
25-36
37-47
Entire Period
13-24
25-35
Entire Period
Avg.
Flow
(Iphd)
3,108
6,124
2,938
3,568
2,245
5,223
3,975


3,816
2,732
3,740
2,337



2,954
5,700
9,272
7,575
2,859
1,189
403
560
578
3,748
5,315
4,157
2,462
4,022
5,543
2,943
4,300
Peak
Flow
(Iphd)
3,666
8,275
3,221
8,275
5,754
6,845
7,464


7,464
5,393
5,707
3,982



5,707
8,935
22,444
13,978
6,043
2,280
490
919
648
22,443
14,641
1 1 ,675
3,528
14,641
15,567
4,918
15,567
Post-Closure Period
Time
Period
(mos.)
NA



55-66
67-78
79-90
91-102
103-114
Entire Period
55-66
67-78
79-90
91-102
103-114
115-119
Entire Period
NA








NA



NA


Avg.
Flow
(Iphd)
NA



317
703
1,146
1,306
510
796
493
337
368
314
250
442
359
NA








NA



NA


Peak
Flow
(Iphd)
NA



670
1,877
1,956
1,943
718
1,956
1,040
654
1,796
855
537
848
1,796
NA








NA



NA





Notes
Flows are combined for Cells A1 & A2.
Cell A2 became operational in month 39.
This caused the elevated flow rates
for months 39-42.


















65% of cell was closed after 65 months
of start of waste placement in cell









m
-------
Table E-3.5.  Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

C1




C2



C3



C4



C5
C6
D1




D2



D3


D4

Waste
Type
(Region)
MSW(SE)




MSW(SE)



MSW(SE)



MSW(SE)



MSW(SE)
MSW(SE)
HW(SE)




HW(SE)



HW(SE)


HW(SE)
Primary
Liner/
LDS
Type
I




I



I



I



I
V
I




I



I


I
Initial Period of Operation
Time
Period01
(mos.)
1-9




1-12



1-8



1-4



1-12
1-10
1-7




1-9



1-12


1-11
Avg.
Flow
(Iphd)
ND




1,475



3,417



14,828



6,419
3,273
ND




6,803



20,292


31,281
Peak
Flow
(Iphd)
ND




2,585



9,558



41,331



12,528
12,155
ND




19,501



51 ,265


120,527
Active Period of Operation
Time
Period
(mos.)
10-21
22-33
34-45
46-56
Entire Period
13-24
25-36
37-45
Entire Period
9-20
21-32
33-41
Entire Period
5-16
17-28
29-35
Entire Period
13-26
11-17
NA




10-17
18-21
22-27
Entire Period
13-24
25-28
Entire Period
NA
Avg.
Flow
(Iphd)
789
259
159
103
332
435
300
161
311
311
314
268
301
937
438
407
624
2,513
393
NA




9,940
224,812
10,862
57,997
13,003
1,010
10,005
NA
Peak
Flow
(Iphd)
1,419
780
286
200
1,419
859
610
464
859
671
752
987
987
2,055
622
686
2,055
10,440
1,403
NA




30,405
407,523
32,341
407,523
44,895
2,413
44,895
NA
Post-Closure Period
Time
Period
(mos.)
NA




NA



NA



NA



NA
NA
8-19
20-26
27-38
39-50
Entire Period
28-33
34-47
Entire Period

NA


NA
Avg.
Flow
(Iphd)
NA




NA



NA



NA



NA
NA
ND
ND
376
715
456
321,768
40
96,559

NA


NA
Peak
Flow
(Iphd)
NA




NA



NA



NA



NA
NA
ND
ND
1,455
1,352
1,455
587,163
567
587,163

NA


NA



Notes
























Near-liquid waste was disposed of in
cell at several different times.






-------
       Table E-3.5. Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

E1



E2



E3
E4
F1


G1-2



G1-3


H1-6
11









Waste
Type
(Region)
MSW(NE)



MSW(NE)



MSW(NE)
MSW(NE)
MSW(NE)


MSW(NE)



MSW(NE)


HW(NE)
MSW(NE)








Primary
Liner/
LDS
Type
I



I



I
I
I


I






I
I








Initial Period of Operation
Time
Period01
(mos.)
1-7



1-12



1-12
1-12
1-12


1-12






ND
1-8








Avg.
Flow
(Iphd)
ND



ND



9,425
20,148
14,472


22,371






ND
ND








Peak
Flow
(Iphd)
ND



ND



25,394
55,785
45,010


46,120






ND
ND








Active Period of Operation
Time
Period
(mos.)
8-19
20-31
32-40
Entire Period
13-24
25-36
37-45
Entire Period
13-14
NA
13-24
25-30
Entire Period
13-24
25-36
37-42
Entire Period
43-54
55-67
Entire Period
ND
9-15
16-32
33-44
45-48
49-54
55-66*
67-78*
79-84*
Entire Period
Avg.
Flow
(Iphd)
8,432
11,521
6,525
9,035
5,821
4,547
4,434
4,979
6,062
NA
9,000
7,826
8,608
12,893
3,438
8,356
8,204
11,774
8,690
10,170
ND
16,224
ND
7,167
231
ND
624
541
904
4,149
Peak
Flow
(Iphd)
19,614
36,164
13,075
36,177
10,445
11,014
6,830
11,014
9,038
NA
25,450
10,932
25,450
23,485
1 1 ,652
10,303
23,485
48,159
20,923
48,159
ND
48,932
ND
22,020
332
ND
1,580
752
1,827
48,932
Post-Closure Period
Time
Period
(mos.)
NA



NA



NA
NA
NA


NA



NA


ND
85-93*








Avg.
Flow
(Iphd)
NA



NA



NA
NA
NA


NA



NA


ND
800








Peak
Flow
(Iphd)
NA



NA



NA
NA
NA


NA



NA


ND
1,794











Notes













Flows are combined for Cells G1 & G2.



Flows are combined for Cells G1 , G2,
& G3 for months 43-67.


Cell received intermediate cover
approximately 30 months after start
of waste placement in cell


"Flows are combined for cells 11 , 12, & 13
during these months.


Ol
Ol
-------
       Table E-3.5. Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

12







13







14
15
J1
J1-2
J1-3
J1-4
J1-5
J1-6
K1






Waste
Type
(Region)
MSW(NE)







MSW(NE)







MSW(NE)
MSW(NE)
MSW(NE)
MSW(NE)
MSW(NE)
MSW(NE)
MSW(NE)
MSW(NE)
MSW(NE)





Primary
Liner/
LDS
Type
I







I







VI
VI
I
I
I
I
I
I
I





Initial Period of Operation
Time
Period01
(mos.)
1-7







1-7







1-12
1-12
1-3





1-12





Avg.
Flow
(Iphd)
6,627







1 1 ,559







4,494
3,938
13,363





17,808





Peak
Flow
(Iphd)
13,959







21,081







17,251
7,985
16,182





24,832





Active Period of Operation
Time
Period
(mos.)
8-24
25-36
37-40
41-46
47-58*
59-70*
71-76*
Entire Period
8-24
25-36
37-40
41-46
47-58*
59-70*
71-76*
Entire Period
13-26
13-21
4-6
7-13*
14-21*
22-32*
33-38*
39-43*
13-24
25-36
37-48
49-60
61-66
Entire Period
Avg.
Flow
(Iphd)
ND
1,030
427
ND
624
541
904
728
ND
1 1 ,684
2,464
ND
624
541
904
3,684
2,041
3,108
5,235
4,813
3,904
3,858
2,824
ND
12,929
10,879
6,155
5,952
9,494
9,036
Peak
Flow
(Iphd)
ND
3,241
1,054
ND
1,580
752
1,827
3,241
ND
26,339
4,666
ND
1,580
752
1,827
26,339
4,282
1 1 ,669
5,824
7,697
6,445
9,880
5,672
ND
27,663
17,683
1 1 ,331
8,024
12,245
27,663
Post-Closure Period
Time
Period
(mos.)
77-85*







77-85*







27-36
22-34
ND





ND





Avg.
Flow
(Iphd)
800







800







567
189
ND





ND





Peak
Flow
(Iphd)
1,794







1,794







1,389
779
N D





ND








Notes
Cell received intermediate cover
approximately 28 months after start
of waste placement in cell

"Flows are combined for cells 11 , 12, & 13
during these months.


Cell received intermediate cover
approximately 22 months after start
of waste placement in cell

"Flows are combined for cells 11 , 12, & 13
during these months.





"Flows are combined for cells listed;
Cells J1-J4 received final closure by
month 32. Cells J5 & J6 were not closed.








Ol
O3
-------
      Table E-3.5. Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

L1
L1-2

L1-3



L1-4


M1
N1






N2




O1
O1 -2





O1 -3



Waste
Type
(Region)
HW(SE)
HW(SE)

HW(SE)



HW(SE)


ASH(M)(SE)
MSW(SE)






MSW(SE)




MSW(SE)
MSW(SE)





MSW(SE)


Primary
Liner/
LDS
Type
I
I

I



I


I
II






II




II
II





II


Initial Period of Operation
Time
Period01
(mos.)
1-2
3-7
8-12







ND
1-12






1-12




1-6









Avg.
Flow
(Iphd)
ND
ND
22,795







ND
ND






ND




ND









Peak
Flow
(Iphd)
ND
ND
51 ,266







ND
ND






ND




ND









Active Period of Operation
Time
Period
(mos.)
NA
13-18

19-25
26-37
38-43
Entire Period
44-55
56-61
Entire Period
ND
13-24
25-36
37-48
49-60
61-72
73-75
Entire Period
13-19
20-31
32-34
35-39
Entire Period
NA
7-18
19-30
31-42
43-54
55-64
Entire Period
65-76
77-80
Entire Period
Avg.
Flow
(Iphd)
NA
2,636

28,040
7,649
7,894
13,417
11,132
8,363
10,209
ND
ND
ND
ND
1,572
2,433
745
1,862
4,547
2,561
6,399
2,741
3,536
NA
4,407
4,023
7,089
6,201
8,661
5,987
10,691
1 1 ,605
10,920
Peak
Flow
(Iphd)
NA
4,293

68,107
17,349
15,145
68,107
35,338
10,954
35,338
ND
ND
ND
ND
5,601
17,597
888
17,597
5,741
3,460
7,274
3,170
7,274
NA
9,826
13,231
16,467
12,561
15,327
16,467
16,965
12,766
16,965
Post-Closure Period
Time
Period
(mos.)
NA









ND
NA






NA




NA









Avg.
Flow
(Iphd)
NA









ND
NA






NA




NA









Peak
Flow
(Iphd)
NA









ND
NA






NA




NA












Notes

































m
-------
       Table E-3.5. Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

P1

P2

P3
Q1




Q1-2
R1

S1




S2



T1
T1-2




U1-4

Waste
Type
(Region)
ASH(M)(SE)

ASH(M)(SE)

ASH(M)(SE)
MSW(NE)




MSW(NE)
MSW(NE)

MSW(NE)




ASH(M)(NE)



MSW(NE)
MSW(NE)




MSW(NE)
Primary
Liner/
LDS
Type
II

II

II
II




II
IV

II




II



II
II




II
Initial Period of Operation
Time
Period01
(mos.)
1-12

1-12

1-10
1-7





1
2-12
1-10




1-9



1-8
1-8




ND
Avg.
Flow
(Iphd)
ND

ND

ND
39,864





ND
1 1 ,592
2,226




2,185



2,137
ND




ND
Peak
Flow
(Iphd)
ND

ND

ND
111,129





ND
22,266
5,081




4,650



5,982
ND




ND
Active Period of Operation
Time
Period
(mos.)
13-34
35-39
13-22
23-27
11-15
8-19
20-31
32-43
44-48
Entire Period
49-58
13-23

11-22
23-28
29-40
41-45
Entire Period
10-17
18-33
34-46
Entire Period
ND
9-20
21-25
26-36
37-46
Entire Period
ND
Avg.
Flow
(Iphd)
ND
20,086
ND
8,935
24,490
9,598
9,290
4,610
3,166
7,263
7,287
9,323

653
ND
1,571
1,086
1,108
654
ND
1,255
1,026
ND
2,861
3,604
ND
661
1,552
ND
Peak
Flow
(Iphd)
ND
45,591
ND
12,277
60,420
16,006
21,862
6,761
5,210
21,862
16,042
17,889

1,220
ND
4,074
2,067
4,074
1,135
ND
3,638
3,638
ND
6,804
5,791
ND
2,174
6,804
ND
Post-Closure Period
Time
Period
(mos.)
NA

NA

NA
NA




NA
NA

NA




NA



NA
NA




ND
Avg.
Flow
(Iphd)
NA

NA

NA
NA




NA
NA

NA




NA



NA
NA




ND
Peak
Flow
(Iphd)
NA

NA

NA
NA




NA
NA

NA




NA



NA
NA




ND



Notes





























Ol
oo
-------
       Table E-3.5. Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

V1-2
V1-3
V1-4


V1-5


W1


W2



X1


Y1








Y2




Z1




Waste
Type
(Region)
MSW(SE)
MSW(SE)
MSW(SE)


MSW(SE)


MSW(SE)


MSW(SE)



MSW(SE)


ASH(M)(NE)








MSW(NE)




ASH(C)(NE)



Primary
Liner/
LDS
Type
II
II
II


II


II


II



II


III








III




III



Initial Period of Operation
Time
Period01
(mos.)
1-13







1-8
9-12

1-8



1
2-7
Period
1-12








1-10




1-12



Avg.
Flow
(Iphd)
13,622







ND
7,492

ND



1 1 1 ,031
32,469
43,693
ND








23,368




28,628



Peak
Flow
(Iphd)
49,828







ND
8,799

ND



111,031
104,645
111,031
ND








36,791




49,551



Active Period of Operation
Time
Period
(mos.)
ND
11
12-23
24-35
36-40
41-52
53-64
Entire Period
13-24
25-35
Entire Period
9-20
21-32
33-35
Entire Period
8-19
20-33
Entire Period
13-26
27
28-39
40-51
52-63
64-67
68
69-78
Entire Period
11-22
23-34
35-46
47-54
Entire Period
13-24
25-36
37-39
Entire Period
Avg.
Flow
(Iphd)
ND
5,149
14,112
6,967
18,045
8,443
1 1 ,683
10,923
2,693
943
1,856
4,288
4,813
719
4,125
5,926
2,188
3,913
19,645
ND
20,515
19,868
18,177
47,154
ND
8,937
19,319
10,353
1 1 ,344
4,404
4,397
7,918
34,520
36,866
32,265
35,312
Peak
Flow
(Iphd)
ND
5,149
33,926
18,142
41,601
31 ,668
23,420
41,601
6,365
1,572
6,365
9,389
10,524
2,141
10,524
14,315
5,376
14,315
58,673
ND
44,593
34,220
47,068
63,832
ND
21,355
63,832
19,204
25,308
6,308
5,199
25,308
68,294
92,207
35,763
92,207
Post-Closure Period
Time
Period
(mos.)
NA
NA
NA


NA


NA


NA



NA


NA








NA




NA



Avg.
Flow
(Iphd)
NA
NA
NA


NA


NA


NA



NA


NA








NA




NA



Peak
Flow
(Iphd)
NA
NA
NA


NA


NA


NA



NA


NA








NA




NA






Notes




































Ol
CD
-------
      Table E-3.5. Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AA1





AA2
AA2-3



AB1







AB2






AB3




AB4


Waste
Type
(Region)
MSW(NE)





MSW(NE)
MSW(NE)



HW(SE)







HW(SE)






HW(SE)




HW(SE)

Primary
Liner/
LDS
Type
III





III
III



III







III






III




III

Initial Period of Operation
Time
Period01
(mos.)
1
2-8




1-13




1-5







1-12






1-12




1-8
9-13
Avg.
Flow
(Iphd)
ND
4,084




14,533




479







878






4,050




ND
7,229
Peak
Flow
(Iphd)
ND
9,261




36,777




1,662







3,433






9,052




ND
9,558
Active Period of Operation
Time
Period
(mos.)
9-14
15-26
27-31
32-43
45-51
Entire Period
14-20
21-34
35-46
47-54
Entire Period
6-12
13-24
25-29
30-36
37-48
49-53
54-87
Entire Period
13-24
25-29
30-36
37-48
49-53
54-87
Entire Period
13-25
26-32
33-40
41-74
Entire Period
14-25
26-63
Avg.
Flow
(Iphd)
1,065
3,387
2,862
994
999
1,890
3,176
11,143
3,719
2,225
5,870
70
268
301
ND
430
141
ND
270
2,371
2,444
ND
1,410
1,162
ND
1,865
5,800
ND
3,944
ND
4,971
2,114
ND
Peak
Flow
(Iphd)
1,586
5,503
4,424
1,751
3,190
5,503
4,781
21,520
6,115
3,509
21,520
200
648
597
ND
724
272
ND
724
7,829
5,197
ND
2,004
1,326
ND
7,829
9,529
ND
4,569
ND
9,529
9,584
ND
Post-Closure Period
Time
Period
(mos.)
NA





NA
NA



NA







NA






NA




NA

Avg.
Flow
(Iphd)
NA





NA
NA



NA







NA






NA




NA

Peak
Flow
(Iphd)
NA





NA
NA



NA







NA






NA




NA




Notes

































m
-------
Table E-3.5.  Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AC1
AC2







ACS





AC4





ACS


AC6




AC7
ACS

Waste
Type
(Region)
HW(W)
HW(W)







HW(W)





HW(W)





HW(W)


HW(W)




HW(W)
HW(W)
Primary
Liner/
LDS
Type
IV
IV







IV





IV





IV


IV




IV
IV
Initial Period of Operation
Time
Period01
(mos.)
1-13
1
2-6





Period
1
2-12



Period
1-11





1-13


1-10




1-12
1-12
Avg.
Flow
(Iphd)
85
1,429
40





272
379
11



42
51





255


352




54
67
Peak
Flow
(Iphd)
169
1,429
77





1,429
379
37



379
255





977


2,990




120
168
Active Period of Operation
Time
Period
(mos.)
ND
7-18
19-30
31-42
43-54
55-66
67-78
79-88
Entire Period
13-24
25-36
37-48
49-60
61-73
Entire Period
12-23
24-35
36-47
48-59
60-73
Entire Period
14-25
26-34
Entire Period
11-16
17-28
29-34
35-40
Entire Period
13-18
13-18
Avg.
Flow
(Iphd)
ND
68
17
8
3
19
4
2
18
2
1
1
1
1
1
5
0
2
0
1
2
82
21
56
534
46
28
345
200
1,925
1,138
Peak
Flow
(Iphd)
ND
217
47
39
8
72
13
5
217
13
6
2
5
4
13
21
1
15
0
15
21
225
44
255
947
101
65
1,002
1,002
6,713
4,601
Post-Closure Period
Time
Period
(mos.)
NA
NA







NA





NA





NA


NA




NA
NA
Avg.
Flow
(Iphd)
NA
NA







NA





NA





NA


NA




NA
NA
Peak
Flow
(Iphd)
NA
NA







NA





NA





NA


NA




NA
NA



Notes

Cell received intermediate cover
approximately 56 months after start
of waste placement in cell





Cell received intermediate cover
approximately 41 months after start
of waste placement in cell



Cell received intermediate cover
approximately 40 months after start
of waste placement in cell













-------
       Table E-3.5.  Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AD1








AD2








ADS








AD4









Waste
Type
(Region)
HW(SE)








HW(SE)








HW(SE)








HW(SE)








Primary
Liner/
LDS
Type
III








III








III








III








Initial Period of Operation
Time
Period01
(mos.)
1-12








1-12








1-14








1-14








Avg.
Flow
(Iphd)
ND








ND








ND








ND








Peak
Flow
(Iphd)
ND








ND








ND








ND








Active Period of Operation
Time
Period
(mos.)
13-20
21-32







13-18
19-30







15-26








15-26








Avg.
Flow
(Iphd)
ND
373







ND
1,886







1,685








1,071








Peak
Flow
(Iphd)
ND
892







ND
3,783







4,197








4,523








Post-Closure Period
Time
Period
(mos.)
33-44
45-51
52-63
64-75
76-87
88-99
100-111
112-121
Entire Period
31-42
43-54
55-66
67-78
79-90
91-102
103-114
115-119
Entire Period
27-38
39-50
51-62
63-74
75-86
87-98
99-110
111-115
Entire Period
27-38
39-50
51-62
63-74
75-86
87-98
99-110
111-115
Entire Period
Avg.
Flow
(Iphd)
145
85
3
3
3
1
1
2
28
644
322
213
133
48
60
37
0
196
321
79
477
399
173
111
61
1,226
288
465
66
157
68
13
1
2
586
137
Peak
Flow
(Iphd)
652
130
22
42
21
4
2
9
652
1,152
609
342
410
110
119
106
0
1,152
478
209
888
807
555
905
698
1,325
1,325
2,870
374
478
111
110
9
4
1,413
2,870



Notes




































05
-------
        Table E-3.5. Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

ADS








AD6








AD7





ADS





AD9





Waste
Type
(Region)
HW(SE)








HW(SE)








HW(SE)





HW(SE)





HW(SE)




Primary
Liner/
LDS
Type
III








III








IV





IV





IV




Initial Period of Operation
Time
Period01
(mos.)
1-7








1-7








1-12





1-8





1-9




Avg.
Flow
(Iphd)
ND








ND








12,597





24,803





19,900




Peak
Flow
(Iphd)
ND








ND








26,492





39,997





42,854




Active Period of Operation
Time
Period
(mos.)
8-19








8-19








13-24
25-36
37-48
49-60
61-70
Entire Period
9-20
21-32
33-44
45-56
57-64
Entire Period
10-21
22-33
34-45
46-58
Entire Period
Avg.
Flow
(Iphd)
37,054








992








2,212
1,539
1,429
249
480
1,206
5,753
2,747
661
296
223
2,058
4,096
2,417
916
227
1,880
Peak
Flow
(Iphd)
115,663








1,853








2,857
2,755
2,813
629
614
2,857
10,545
5,352
2,393
1,070
355
10,545
8,051
4,343
1,680
699
8,051
Post-Closure Period
Time
Period
(mos.)
20-31
32-43
44-55
56-67
68-79
80-91
92-103
104-108
Entire Period
20-31
32-43
44-55
56-67
68-79
80-91
92-103
104-108
Entire Period
70-81
82-87



Entire Period
65-76
77-84



Entire Period
59-60
61-72
73-79

Entire Period
Avg.
Flow
(Iphd)
3,290
1,166
683
444
278
9
1
0
792
200
114
64
70
37
35
8
0
71
375
165



305
310
189



261
798
525
463

530
Peak
Flow
(Iphd)
5,541
1,801
1,385
767
1,045
58
2
0
5,541
776
190
86
156
100
125
42
0
776
533
334



533
415
315



415
825
797
611

825



Notes


















Cell received intermediate cover
approximately 12 months after start
of waste placement in cell



Cell received intermediate cover
approximately 12 months after start
of waste placement in cell



Cell received intermediate cover
approximately 12 months after start
of waste placement in cell


05
CO
-------
       Table E-3.5.  Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AD10






AD11






AD12





AD13




AD14




AD15


AD16


AD17
AD18

Waste
Type
(Region)
HW(SE)






HW(SE)






HW(SE)





HW(SE)




HW(SE)




HW(SE)


HW(SE)


HW(SE)
HW(SE)
Primary
Liner/
LDS
Type
IV






IV






IV





IV




IV




IV


IV


IV
IV
Initial Period of Operation
Time
Period01
(mos.)
1-12






1-5






1-13





1-12




1-12




1-13


1-12


1-12
1-12
Avg.
Flow
(Iphd)
20,960






1 1 ,875






25,609





18,604




20,104




24,664


17,442


5,761
5,035
Peak
Flow
(Iphd)
51 ,425






20,518






55,840





86,467




85,939




89,367


84,110


10,955
15,685
Active Period of Operation
Time
Period
(mos.)
13-24
25-36
37-50
51-62
63-74
75-79
Entire Period
6-17
18-29
30-41
42-53
54-65
66-69
Entire Period
14-25
26-37
38-49
50-61
62-69
Entire Period
13-18
19-30
31-42
43-55
Entire Period
13-24
25-36
37-48
49-55
Entire Period
14-25
26-39
Entire Period
13-24
25-39
Entire Period
13-16
13-19
Avg.
Flow
(Iphd)
7,382
1,370
789
470
308
154
1,883
3,168
3,552
4,305
4,975
2,442
1,255
3,536
5,612
1,362
1,260
1,091
1,064
2,150
19,758
3,287
2,064
3,815
5,403
10,197
3,014
2,023
1,232
4,452
571
234
390
31 ,485
7,080
17,927
4,046
1,186
Peak
Flow
(Iphd)
15,064
9,735
2,961
867
788
603
15,064
5,974
5,365
5,278
6,246
6,708
1,579
6,708
19,674
2,505
1,605
1,313
2,129
19,674
51 ,260
5,734
4,942
8,470
51 ,260
40,684
5,149
4,494
1,480
40,684
1,294
549
1,294
153,293
18,896
153,293
6,109
3,257
Post-Closure Period
Time
Period
(mos.)
NA






NA






NA





NA




NA




NA


NA


NA
NA
Avg.
Flow
(Iphd)
NA






NA






NA





NA




NA




NA


NA


NA
NA
Peak
Flow
(Iphd)
NA






NA






NA





NA




NA




NA


NA


NA
NA



Notes
Cell received intermediate cover
approximately 41 months after start
of waste placement in cell

1/2 of cell was closed approximately 59
months after start of waste placement
in cell
Cell received intermediate cover
approximately 41 months after start
of waste placement in cell




Cell received intermediate cover
approximately 41 months after start
of waste placement in cell



Cell received intermediate cover
approximately 41 months after start
of waste placement in cell


Cell received intermediate cover
approximately 41 months after start
of waste placement in cell


Cell received intermediate cover
approximately 41 months after start
of waste placement in cell
Cell received intermediate cover
approximately 41 months after start
of waste placement in cell


m
-------
       Table E-3.5. Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AE1

AE1-2



AE1-3

AF1





AG1


AH 1-5
AM



AI2




AJ1

AK1
AK1-2
AL1




Waste
Type
(Region)
MSW(NE)

MSW(NE)



MSW(NE)

ASH(M)(NE)





MSW(NE)


MSW/ASH(M)(NE)
C&DW(NE)



C&DW(NE)




MSW(NE)

MSW(NE)
MSW(NE)
MSW(NE)



Primary
Liner/
LDS
Type
III

III



III

III





III


III
III



III




III

IV
IV
IV



Initial Period of Operation
Time
Period01
(mos.)
1-12







1-12





1-12


ND
1-2
3-12


1-2
3-6



1
2-6
1-12

1-12



Avg.
Flow
(Iphd)
ND







25,383





1,780


ND
ND
15,552


ND
19,613



ND
17,133
9,867

ND



Peak
Flow
(Iphd)
ND







40,850





5,314


ND
ND
29,341


ND
22,964



ND
24,782
17,983

ND



Active Period of Operation
Time
Period
(mos.)
13-24
25-32
33-44
45-50
51-59
60-72
73-80
Entire Period
13-24
25-36
37-48
49-60
61-63
Entire Period
13-24
25-33
Entire Period
ND
13-24
25-36
37-50
Entire Period
7-18
19-30
31-42
43-50
Entire Period
7-13


13-15
13-29
30-41
42-54
Entire Period
Avg.
Flow
(Iphd)
ND
ND
ND
ND
ND
14,000
21,513
17,705
18,257
22,116
19,672
17,809
ND
19,463
3,963
4,022
3,988
ND
9,952
11,170
14,786
12,118
12,192
18,026
17,052
17,968
16,159
4,728


2,398
ND
934
1,349
1,150
Peak
Flow
(Iphd)
ND
ND
ND
ND
ND
48,977
33,542
48,977
26,768
29,973
41 ,880
40,718
ND
41 ,880
12,357
1 1 ,347
12,357
ND
22,419
23,200
29,477
29,477
20,328
32,402
34,159
28,114
34,159
9,936


3,130
ND
2,085
5,885
5,885
Post-Closure Period
Time
Period
(mos.)
NA

NA



NA

NA





NA


NA
NA



NA




NA

NA

NA



Avg.
Flow
(Iphd)
NA

NA



NA

NA





NA


NA
NA



NA




NA

NA

NA



Peak
Flow
(Iphd)
NA

NA



NA

NA





NA


NA
NA



NA




NA

NA

NA






Notes



































05
Ol
-------
       Table E-3.5. Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AM1






AM2






AN1
AN 1-2
AN 1-3
AN 1-4
AO1



AO2


AO3
AO4
AP1




Waste
Type
(Region)
MSW(W)






MSW(W)






ASH(M)(NE)
ASH(M)(NE)
ASH(M)(NE)
ASH(M)(NE)
MSW(NE)



MSW(NE)


MSW(NE)
MSW(NE)
HW(W)



Primary
Liner/
LDS
Type
IV






IV






IV
IV
IV
IV
IV



IV


IV
IV
IV



Initial Period of Operation
Time
Period01
(mos.)
1-9






1-9






1-7



1-5



1-5


1-8
1-8
1-12



Avg.
Flow
(Iphd)
ND






ND






ND



ND



15,881


16,746
20,017
3,093



Peak
Flow
(Iphd)
ND






ND






ND



ND



24,541


53,117
55,470
10,515



Active Period of Operation
Time
Period
(mos.)
10-21
22-33
34-45
46-57
58-69
70-81
Entire Period
10-21
22-33
34-45
46-57
58-69
70-81
Entire Period
8-12
13-17
18-31
32-34
6-17
18-29
26-36
Entire Period
6-17
18-30
Entire Period
NA
NA
13-24
25-36
37-48
Entire Period
Avg.
Flow
(Iphd)
270
336
111
20
18
11
111
32
35
17
67
64
112
55
10,844
8,455
14,087
21,598
1,984
1,299
1,144
1,485
3,027
1,688
2,331
NA
NA
4,885
3,353
4,579
4,272
Peak
Flow
(Iphd)
533
329
283
77
21
18
533
154
51
45
274
181
136
274
13,430
21,498
30,179
23,136
4,130
1,577
1,371
4,130
5,266
2,383
5,266
NA
NA
15,059
5,802
8,260
15,059
Post-Closure Period
Time
Period
(mos.)
NA






NA






NA



NA



NA


NA
NA
NA



Avg.
Flow
(Iphd)
NA






NA






NA



NA



NA


NA
NA
NA



Peak
Flow
(Iphd)
NA






NA






NA



NA



NA


NA
NA
NA






Notes
Cell received intermediate cover
approximately 5 months after start
of waste placement in cell




Cell received intermediate cover
approximately 5 months after start
of waste placement in cell





















05
O3
-------
       Table E-3.5.  Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).
Cell
No.
AQ1




AQ2




AQ3




AQ4




AQ5



AQ6



AQ7



AQ8



Waste
Type
(Region)
HW(NE)




HW(NE)




HW(NE)




HW(NE)




HW(NE)



HW(NE)



HW(NE)



HW(NE)



Primary
Liner/
LDS
Type
IV




IV




IV




IV




IV



IV



IV



IV



Initial Period of Operation
Time
Period01
(mos.)
1-6




1-6




1
2
3-12


1
2-12



1-12



1-12



1-12



1-12



Avg.
Flow
(Iphd)
10,203




13,050




ND
17,940
ND


18,970
ND



ND



ND



ND



ND



Peak
Flow
(Iphd)
18,944




20,721




ND
17,940
ND


18,970
ND



ND



ND



ND



ND



Active Period of Operation
Time
Period
(mos.)
7-48
47-58



7-48
49-58



13-44
45-54



13-43
44-53



13-35
36-47
48-50
Entire Period
13-31
32-43
44-52
Entire Period
13-30
31-42


13-14
15-26


Avg.
Flow
(Iphd)
ND
4,530



ND
2,181



ND
2,962



ND
1,049



ND
1 1 ,304
5,427
10,129
ND
2,226
3,405
2,732
ND
1,256


ND
21,329


Peak
Flow
(Iphd)
ND
10,531



ND
6,460



ND
13,430



ND
1,622



ND
34,353
6,692
34,353
ND
12,072
11,119
12,072
ND
4,821


ND
76,759


Post-Closure Period
Time
Period
(mos.)
59-65
66-77
78-89
90-97
Entire Period
59-70
71-82
83-94
95-97
Entire Period
55-66
67-78
79-90
91-93
Entire Period
54-57
58-70
71-82
83-92
Entire Period
51-62
63-74
75-84
Entire Period
53-64
65-76
77-80
Entire Period
43-54
55-66
67-79
Entire Period
27-38
39-50
51-63
Entire Period
Avg.
Flow
(Iphd)
5,835
644
1,367
1,615
1,997
218
206
747
217
377
626
420
721
1,851
686
577
4,292
495
534
1,779
1,040
603
561
745
472
504
551
497
35
78
261
129
586
557
578
573
Peak
Flow
(Iphd)
1 1 ,244
1,011
3,264
3,575
1 1 ,244
717
1,082
3,040
432
3,040
2,642
1,038
1,274
3,167
3,167
916
6,345
1,210
1,141
6,345
2,369
1,540
1,194
2,369
903
1,272
987
1,272
120
211
565
565
1,677
1,011
1,425
1,677
Notes
Cell final closure date is approximate














Cell final closure date is approximate




















05
-------
        Table E-3.5. Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AQ9



AQ10



AR1


AR1-2
AS1




AT1


AU1

AV1-4



AV1





Waste
Type
(Region)
HW(NE)



HW(NE)



MSW(NE)


MSW(NE)
HW(SE)




HW(W)


MSW(NE)

HW(NE)



HW(NE)




Primary
Liner/
LDS
Type
IV



IV



IV


IV
IV




IV


IV

IV



IV




Initial Period of Operation
Time
Period01
(mos.)
1-12



1-9



1-11



1-10




1-5


1-4

1-12








Avg.
Flow
(Iphd)
ND



ND



27,042



ND




ND


ND

ND








Peak
Flow
(Iphd)
ND



ND



65,871



ND




ND


ND

ND








Active Period of Operation
Time
Period
(mos.)
13-14
15-26


10-14
15-26


12-23
24-36
Entire Period
37-40
11-22
23-27


Entire Period
6-8


5-20
21-32
13-40
41-52
53-64
Entire Period
65-76
77-88
89-101
102-104
Entire Period
Avg.
Flow
(Iphd)
ND
4,579


ND
15,933


11,251
9,668
10,428
5,199
388
186


329
1,249


ND
4,991
ND
9,308
7,907
8,608
3,150
1,715
1,112
ND
1,821
Peak
Flow
(Iphd)
ND
24,946


ND
38,751


23,384
26,274
26,274
8,156
1,146
256


1,146
1,964


ND
1 1 ,597
ND
19,021
10,611
19,021
7,484
5,500
1,468
ND
7,484
Post-Closure Period
Time
Period
(mos.)
27-38
39-50
51-63
Entire Period
27-38
39-50
51-63
Entire Period
NA


NA
28-39
40-51
52-63
64-71
Entire Period
9-20
21-33
Entire Period
NA

NA








Avg.
Flow
(Iphd)
275
433
943
561
682
300
852
618
NA


NA
125
50
32
24
61
88
25
55
NA

NA








Peak
Flow
(Iphd)
497
1,425
2,667
2,667
2,251
1,709
1,588
2,251
NA


NA
190
103
52
61
190
434
107
434
NA

NA











Notes








Cell AR1 has four subareas. Waste
placement started in months 1, 5, 20, and
26 in subareas 1 through 4, respectively.




















05
oo
-------
       Table E-3.5. Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AV2




AV3




AV4




AV5



AW1
AW1.3
AW2
AW2.4
AX1




AX2





Waste
Type
(Region)
HW(NE)




HW(NE)




HW(NE)




HW(NE)



MSW(SE)
MSW(SE)
MSW(SE)
MSW(SE)
MSW(NE)




MSW(NE)




Primary
Liner/
LDS
Type
IV




IV




IV




IV



V
V,ll
V
V,ll
V




V




Initial Period of Operation
Time
Period01
(mos.)















1-12



1-12

1-12

1-2




1-5




Avg.
Flow
(Iphd)















18,789



6,358

3,553

16,718




15,521




Peak
Flow
(Iphd)















44,741



20,570

7,480

19,738




58,674




Active Period of Operation
Time
Period
(mos.)
52-63
64-75
76-88
89-91
Entire Period
35-46
47-58
59-71
72-74
Entire Period
27-38
39-50
51-63
64-66
Entire Period
13-24
25-37
38-40
Entire Period

13-24

13-21
3-14
15-26
27-33

Entire Period
6-14
15-26
27-33

Entire Period
Avg.
Flow
(Iphd)
5,520
1,829
1,469
ND
2,682
7,964
3,994
2,703
ND
4,466
3,748
2,086
2,787
ND
2,656
1 1 ,649
4,215
ND
6,949

1,409

2,330
1,128
202
108

540
320
210
352

281
Peak
Flow
(Iphd)
12,591
4,222
2,638
ND
12,591
15,363
9,294
5,964
ND
15,363
9,778
4,026
4,667
ND
9,778
27,720
1 1 ,933
ND
27,720

7,031

8,257
2,383
370
159

2,383
570
421
480

570
Post-Closure Period
Time
Period
(mos.)
NA




NA




NA




NA



NA
NA
NA
NA
34-45
46-57
58-69
70-83
Entire Period
34-45
46-57
58-69
70-83
Entire Period
Avg.
Flow
(Iphd)
NA




NA




NA




NA



NA
NA
NA
NA
52
47
48
58
66
248
142
109
208
178
Peak
Flow
(Iphd)
NA




NA




NA




NA



NA
NA
NA
NA
75
56
75
94
94
421
234
187
300
421



Notes

































m
-------
       Table E-3.5. Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AX3




AX4




AX5






AX6






AX7








Waste
Type
(Region)
MSW(NE)




MSW(NE)




MSW(NE)






MSW(NE)






MSW(NE)







Primary
Liner/
LDS
Type
V




V




V






V






V







Initial Period of Operation
Time
Period01
(mos.)
1-5




1-12




1-11






1-9






1-10







Avg.
Flow
(Iphd)
3,361




2,534




1,384






3,759






5,376







Peak
Flow
(Iphd)
7,985




12,688




3,394






7,171






12,155







Active Period of Operation
Time
Period
(mos.)
6-17
18-29
30-41
42-56
Entire Period
13-17
18-29
30-41
42-56
Entire Period
12-23
24-35
36-47
48-59
60-71
72-80
Entire Period
10-24
25-36
37-48
49-60
61-72
73-78
Entire Period
11-22
23-25
26
27-38
39-50
51-62
63-76
Entire Period
Avg.
Flow
(Iphd)
188
222
205
553
307
298
67
42
42
75
106
50
32
44
41
65
56
144
132
171
164
203
232
168
390
181
ND
126
206
175
289
234
Peak
Flow
(Iphd)
475
692
692
1,075
1,075
187
127
84
84
187
191
108
47
56
65
80
191
655
234
196
224
340
355
655
851
309
ND
580
393
281
1,412
851
Post-Closure Period
Time
Period
(mos.)
57-68
69-81


Entire Period
57-68
69-81


Entire Period
NA






NA






NA







Avg.
Flow
(Iphd)
227
187


206
23
70


47
NA






NA






NA







Peak
Flow
(Iphd)
458
320


458
37
84


84
NA






NA






NA










Notes
Cell received intermediate cover
approximately 34 months after start
of waste placement in cell


Cell received intermediate cover
approximately 34 months after start
of waste placement in cell


Cell received intermediate cover
approximately 33 months after start
of waste placement in cell

Cell was partially closed approximately 55
months after start of waste placement in
cell
Cell received intermediate cover
approximately 30 months after start
of waste placement in cell

Cell was partially closed approximately 53
months after start of waste placement in
cell
Cell was partially closed approximately 51
months after start of waste placement in
cell





m
-------
       Table E-3.5.  Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).
Cell
No.
AX8





AX9





AX10





AX11




AX12




AX13




Waste
Type
(Region)
MSW(NE)





MSW(NE)





MSW(NE)





MSW(NE)




MSW(NE)




MSW(NE)




Primary
Liner/
LDS
Type
V





V





V





V




V




V




Initial Period of Operation
Time
Period01
(mos.)
1
2-14



Period
1-9





1-7





1-16




1-12




1-7




Avg.
Flow
(Iphd)
21,038
578



4,881
1,047





2,786





4,675




3,494




6,683




Peak
Flow
(Iphd)
21,038
6,256



21,038
3,478





13,698





14,586




8,836




14,343




Active Period of Operation
Time
Period
(mos.)
15-26
27-38
39-50
51-62
63-71
Entire Period
10-21
22-33
34-45
46-57
58-65
Entire Period
8-19
20-31
32-43
44-55
56-59
Entire Period
17-28
29-40
41-52
53-62
Entire Period
13-24
25-36
37-48
49-56
Entire Period
8-19
20-31
32-43
44-53
Entire Period
Avg.
Flow
(Iphd)
545
489
352
325
499
439
92
30
12
30
40
41
330
285
342
502
486
374
219
112
121
148
150
1,376
711
493
354
768
1,734
3,058
250
424
1,408
Peak
Flow
(Iphd)
1,384
963
402
626
600
1,384
159
75
28
65
70
159
477
337
402
645
608
645
300
178
337
200
337
3,029
1,505
650
500
3,029
3,488
9,294
449
1,421
9,294
Post-Closure Period
Time
Period
(mos.)
NA





NA





NA





NA




NA




NA




Avg.
Flow
(Iphd)
NA





NA





NA





NA




NA




NA




Peak
Flow
(Iphd)
NA





NA





NA





NA




NA




NA




Notes
Cell was partially closed approximately 46
months after start of waste placement in
cell



Cell was partially closed approximately 41
months after start of waste placement in
cell



Cell was partially closed approximately 34
months after start of waste placement in
cell



Cell was partially closed approximately 39
months after start of waste placement in
cell


Cell may have been partially closed
approximately 31 months after start
of waste placement in cell


Cell may have been partially closed
approximately 28 months after start
of waste placement in cell


m
•vl
-------
        Table E-3.5.  Landfill LCRS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AX14




AX15


AX16


AY1
AY2
AYS
AZ1



BA1



BA2
BB1




BB2
BBS

Waste
Type
(Region)
MSW(NE)




MSW(NE)


MSW(NE)


HW(NE)
HW(NE)
HW(NE)
MSW(NE)



HW(NE)



HW(NE)
MSW(SE)




MSW(SE)
MSW(SE)
Primary
Liner/
LDS
Type
V




V


V


VI
VI
VI
VI



_(2)



_(2)
VI




VI
VI
Initial Period of Operation
Time
Period01
(mos.)
1-11




1-12


1-10


1-9
1-11
1-11
1
2-12


1-14



1-2
1-6




1-11
1-11
Avg.
Flow
(Iphd)
2,777




5,573


8,601


6,803
10,964
12,198
ND
4,093


ND



4,979
10,378




ND
ND
Peak
Flow
(Iphd)
6,582




1 1 ,809


17,756


12,439
23,914
32,326
ND
5,219


ND



5,860
22,130




ND
ND
Active Period of Operation
Time
Period
(mos.)
12-23
24-30
31-32
33-38
Entire Period
13-24
25-37
Entire Period
11-22
23-29
Entire Period
NA
NA
NA
13-22
23-25
26-31
Entire Period
15-26
27-38
39-42
Entire Period
3-12
7-18
19-24
25-35
36-47
Entire Period
12-23
12-23
Avg.
Flow
(Iphd)
216
324
ND
360
281
277
319
299
1,181
199
819
NA
NA
NA
4,006
ND
2,584
3,473
5,467
2,949
2,406
3,951
2,190
3,399
5,127
3,119
272
2,494
5,422
2,284
Peak
Flow
(Iphd)
300
430
ND
449
449
400
561
2,786
5,096
393
5,096
NA
NA
NA
5,054
ND
3,016
5,054
9,440
4,883
2,745
9,440
2,846
5,644
8,983
8,343
2,621
8,983
14,042
7,945
Post-Closure Period
Time
Period
(mos.)
NA




NA


NA


NA
NA
NA
NA



NA



NA
NA

NA
NA

NA
NA
Avg.
Flow
(Iphd)
NA




NA


NA


NA
NA
NA
NA



NA



NA
NA

NA
NA

NA
NA
Peak
Flow
(Iphd)
NA




NA


NA


NA
NA
NA
NA



NA



NA
NA

NA
NA

NA
NA



Notes


















Cell received intermediate cover
approximately 42 months after start
of waste placement in cell









m
         Notes:
            (1) NA = not applicable, ND = not determined.
            (2) Cells BA1 and BA2 include a GM/GCL/GM primary liner.
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage.

Cell
No.

A1-2



B1





B2



B3








B4



B5


Prim.
Liner/
LDS
Type
I



I





I



III








IV



IV



Waste
Type/
Region
C&DW(NE)



MSW(NE)





MSW(NE)



MSW(NE)








MSW(NE)



MSW(NE)


Initial Period of Operation
Time
Period^1
(mos.)
1-27



1-19





1-19



1-4








1-12



1-12


Avg.
Flow
(iphd)
ND



ND





ND



1,394








ND



11


Peak
Flow
(Iphd)
ND



ND





ND



4,250








ND



43


Active Period of Operation
Time
Period
(mos.)
28-38
39-50
51-58
Entire Period
20-31
32-43
44-54


Entire Period
20-31
32-43
44-54
Entire Period
5-16
17-28
29-40
41-52
53-64
65-76
77-88
89-93
Entire Period
13-24
25-36
37-47
Entire Period
13-24
25-35
Entire Period
Avg.
Flow
(Iphd)
45
38
30
39
266
424
892


517
404
996
665
689
124
101
262
231
45
92
102
98
135
32
1
73
34
3
16
9
Peak
Flow
(Iphd)
124
120
84
124
499
808
1,426


1,426
605
1,690
1,102
1,690
266
168
803
713
152
133
193
109
803
97
6
100
100
13
65
65
Post-Closure Period
Time
Period
(mos.)
NA



55-66
67-78
79-90
91-102
103-114
Entire Period
55-66
67-79

Entire Period
NA








NA



NA


Avg.
Flow
(Iphd)
NA



106
267
279
326
74
210
154
328

189
NA








NA



NA


Peak
Flow
(Iphd)
NA



222
1,134
451
612
97
1,134
393
514

514
NA








NA



NA


Notes



Flows are combined for Cells A1 & A2.
Cell A2 became operational in month 39.
Cell A2 became operational
in month 39.















65% of cell was closed after 65 months
of start of waste placement in cell









m

-!-j
CO
-------
       Table E-3.6.  Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

C1




C2



C3



C4



C5
C6
D1




D2



D3


Prim.
Liner/
LDS
Type
I




I



I



I



I
V
I




I



I



Waste
Type/
Region
MSW(SE)




MSW(SE)



MSW(SE)



MSW(SE)



MSW(SE)
MSW(SE)
HW(SE)




HW(SE)



HW(SE)


Initial Period of Operation
Time
Period^1
(mos.)
1-9




1-12



1-8



1-4



1-12
1-10
1-7




1-9



1-12


Avg.
Flow
(iphd)
ND




92



63



178



23
178
32




0



12


Peak
Flow
(Iphd)
ND




398



268



265



40
823
80




0



56


Active Period of Operation
Time
Period
(mos.)
10-21
22-33
34-45
46-56
Entire Period
13-24
25-36
37-45
Entire Period
9-20
21-32
33-41
Entire Period
5-16
17-28
29-35
Entire Period
13-26
11-17
NA




10-17
18-21
22-27
Entire Period
13-24
25-28
Entire Period
Avg.
Flow
(Iphd)
123
89
27
40
70
9
22
7
13
2
33
16
17
70
51
26
53
28
3
NA




776
101
63
388
7
283
76
Peak
Flow
(Iphd)
304
170
128
227
304
31
125
14
125
9
276
103
276
147
92
29
147
115
15
NA




2,426
122
133
2,426
73
341
341
Post-Closure Period
Time
Period
(mos.)
NA




NA



NA



NA



NA
NA
8-19
20-26
27-38
39-50
Entire Period
28-39
40-47

Entire Period
NA


Avg.
Flow
(Iphd)
NA




NA



NA



NA



NA
NA
102
1
5
64
41
9
1

6
NA


Peak
Flow
(Iphd)
NA




NA



NA



NA



NA
NA
886
10
70
156
886
51
2

51
NA


Notes



























Near-liquid waste was disposed of in
cell at several different times.





m
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

D4
E1

E2



E3
E4
F1


G1






G2



G3
Prim.
Liner/
LDS
Type
I
I

I



I
I
I


I






I



III

Waste
Type/
Region
HW(SE)
MSW(NE)

MSW(NE)



MSW(NE)
MSW(NE)
MSW(NE)


MSW(NE)






MSW(NE)



MSW(NE)
Initial Period of Operation
Time
Period^1
(mos.)
1-11
1-7

1
2-12
Period

1-12
1-12
1-12


1-12






1-12



1-12
Avg.
Flow
(Iphd)
233
2,144

3,518
179
483

1,595
996
124


ND






197



490
Peak
Flow
(Iphd)
801
5,026

3,518
536
3,518

1,951
2,362
479


ND






645



627
Active Period of Operation
Time
Period
(mos.)
NA
8-19
20-31
32-40
Entire Period
13-24
25-36
37-45
Entire Period
13-14
NA
13-24
25-30
Entire Period
13-24
25-36
37-42
43-51
52-63
64-67
Entire Period
13-24
25-36
37-42
Entire Period
13-25
Avg.
Flow
(Iphd)
NA
1,436
1,051
743
1,107
802
685
596
703
1,603
NA
66
67
66
ND
156
101
121
74
49
108
37
35
60
41
319
Peak
Flow
(Iphd)
NA
3,069
1,915
1,015
3,069
2,447
1,404
999
2,447
1,758
NA
83
77
83
ND
238
116
384
139
64
384
65
42
100
100
384
Post-Closure Period
Time
Period
(mos.)
NA
NA

NA



NA
NA
NA


NA






NA



NA
Avg.
Flow
(Iphd)
NA
NA

NA



NA
NA
NA


NA






NA



NA
Peak
Flow
(Iphd)
NA
NA

NA



NA
NA
NA


NA






NA



NA
Notes



























m
en
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).
Cell
No.
H1




H2








H3







H4


H5
H6
Prim.
Liner/
LDS
Type
I




I








I







I


I
I
Waste
Type/
Region
HW(NE)




HW(NE)








HW(NE)







HW(NE)


HW(NE)
HW(NE)
Initial Period of Operation
Time
Period^1
(mos.)
1-13




1-12








1-11







1-9


1-7
1-6
Avg.
Flow
(iphd)
ND




26








30







339


197
528
Peak
Flow
(Iphd)
ND




39








61







415


304
729
Active Period of Operation
Time
Period
(mos.)
14-25
26-37
38-49
50-55
Entire Period
13-24
25-29
30-32
33-42
43-45
46-57*
58-69*
70-76
Entire Period
12-15
16-25
26-28
29-30
31-42*
43-54*
55-58*
Entire Period
1 0-21 *
22-29*
Entire Period
8-20*
ND
Avg.
Flow
(Iphd)
19
8
4
2
9
6
8
ND
108
ND
35
16
9
31
ND
207
ND
81
37
32
27
79
91
58
78
97
ND
Peak
Flow
(Iphd)
65
21
7
5
65
9
10
ND
175
ND
49
25
12
175
ND
415
ND
83
62
39
29
415
128
66
128
149
ND
Post-Closure Period
Time
Period
(mos.)
56-67
68-79
80-88

Entire Period
ND








NA







NA


NA
NA
Avg.
Flow
(Iphd)
3
0
0

1
ND








NA







NA


NA
NA
Peak
Flow
(Iphd)
12
2
0

12
ND








NA







NA


NA
NA
Notes











"Cell partially closed after month 46



"Cell partially closed after month 31







"Cell partially closed


"Cell partially closed

m
CD
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

11









12







13







14
15
J1



Prim.
Liner/
LDS
Type
I









I







I







VI
VI
I




Waste
Type/
Region
MSW(NE)









MSW(NE)







MSW(NE)







MSW(NE)
MSW(NE)
MSW(NE)



Initial Period of Operation
Time
Period^1
(mos.)
1-5









1-7







1-7







1-12
1-12
1-3



Avg.
Flow
(iphd)
234









31







37







24
2
24



Peak
Flow
(Iphd)
508









77







87







70
11
34



Active Period of Operation
Time
Period
(mos.)
6-8
9-15
16-29
30-41
42-48
49-53
54-65
66-77
78-84
Entire Period
8-24
25-36
37-40
41-46
47-58
59-70
71-76
Entire Period
8-24
25-36
37-40
41-46
47-58
59-70
71-76
Entire Period
13-26
13-21
4-15
16-28
29-32
Entire Period
Avg.
Flow
(Iphd)
ND
5
ND
10
4
ND
2
13
79
16
ND
5
6
ND
8
8
5
7
ND
7
5
ND
4
13
17
9
26
11
9
9
ND
9
Peak
Flow
(Iphd)
ND
18
ND
44
10
ND
5
42
157
157
ND
35
11
ND
37
23
6
37
ND
23
8
ND
17
55
53
55
142
54
13
12
ND
13
Post-Closure Period
Time
Period
(mos.)
85-93









77-85







77-85







27-36
22-34
33-43



Avg.
Flow
(Iphd)
62









2







3







59
2
ND



Peak
Flow
(Iphd)
119









4







12







133
8
ND



Notes



Cell received intermediate cover
approximately 30 months after start
of waste placement in cell







Cell received intermediate cover
approximately 28 months after start
of waste placement in cell





Cell received intermediate cover
approximately 22 months after start
of waste placement in cell











m
-------
       Table E-3.6.  Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

J2

J3

J4

J5
J6
K1





L1





L2




L3



L4
Prim.
Liner/
LDS
Type
I

I

I

I
I
I





I





I




I



I

Waste
Type/
Region
MSW(NE)

MSW(NE)

MSW(NE)

MSW(NE)
MSW(NE)
MSW(NE)





HW(SE)





HW(SE)




HW(SE)



HW(SE)
Initial Period of Operation
Time
Period^1
(mos.)
1-12

1-7

1-7

1-7
1-5
1-12





1-3





1-7




1-5



1-5
Avg.
Flow
(iphd)
8

9

15

504
54
122





2,553





301




542



105
Peak
Flow
(Iphd)
13

14

43

786
236
163





3,464





458




974



461
Active Period of Operation
Time
Period
(mos.)
13-21
22-25
8-11
12-19
8-11

8-11
NA
13-24
25-36
37-48
49-60
61-66
Entire Period
4-15
16-27
28-39
40-51
52-61
Entire Period
8-20
21-35
36-47
48-59
Entire Period
6-18
19-30
31-41
Entire Period
6-18
Avg.
Flow
(Iphd)
8
ND
ND
19
32

139
NA
88
76
514
349
282
271
330
134
89
44
143
148
280
21
140
28
115
69
22
0
31
76
Peak
Flow
(Iphd)
11
ND
ND
28
44

149
NA
180
104
892
495
378
892
562
749
409
169
500
749
869
92
702
222
869
537
210
0
537
280
Post-Closure Period
Time
Period
(mos.)
26-36

20-21
22-30
12-16
17-22
NA
NA
NA





NA





NA




NA



NA
Avg.
Flow
(Iphd)
ND

29
ND
13
ND
NA
NA
NA





NA





NA




NA



NA
Peak
Flow
(Iphd)
ND

29
ND
43
ND
NA
NA
NA





NA





NA




NA



NA
Notes

































m
oo
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

M1



N1






N2




O1





O2




O3
P1

P2

P3
Prim.
Liner/
LDS
Type
I



II






II




II





II




II
II

II

II

Waste
Type/
Region
ASH(M)(SE)



MSW(SE)






MSW(SE)




MSW(SE)





MSW(SE)




MSW(SE)
ASH(M)(SE)

ASH(M)(SE)

ASH(M)(SE)
Initial Period of Operation
Time
Period^1
(mos.)
1-12



1-12






1-12




1-6





1-12




1-10
1-12

1-12

1-10
Avg.
Flow
(iphd)
76



ND






ND




293





6




359
ND

ND

ND
Peak
Flow
(Iphd)
114



ND






ND




620





24




768
ND

ND

ND
Active Period of Operation
Time
Period
(mos.)
13-24
25-36
37-43
Entire Period
13-24
25-36
37-48
49-60
61-72
73-75
Entire Period
13-19
20-31
32-34
35-39
Entire Period
7-18
19-30
31-42
43-54
55-64
Entire Period
13-24
25-36
37-48
49-59
Entire Period
11-16
13-34
35-39
13-22
23-27
11-15
Avg.
Flow
(Iphd)
63
45
49
53
ND
ND
ND
83
183
82
121
113
203
786
201
227
0
3
0
1
3
1
2
1
3
1
2
112
ND
716
ND
832
1,184
Peak
Flow
(Iphd)
93
66
69
93
ND
ND
ND
222
354
108
354
468
669
1,058
406
1,058
3
7
5
6
9
9
5
4
11
5
11
428
ND
2,585
ND
1,220
3,179
Post-Closure Period
Time
Period
(mos.)
NA



NA






NA




NA






NA



NA
NA
NA
NA

NA
Avg.
Flow
(Iphd)
NA



NA






NA




NA






NA



NA
NA
NA
NA

NA
Peak
Flow
(Iphd)
NA



NA






NA




NA






NA



NA
NA
NA
NA

NA
Notes







No GM in bottom liner




























m

-!-j
CD
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

Q1




Q1-2
R1




S1




S2



T1






T2





Prim.
Liner/
LDS
Type
II





II




II




II



II






II






Waste
Type/
Region
MSW(NE)





MSW(NE)




MSW(NE)




ASH(M)(NE)



MSW(NE)






MSW(NE)





Initial Period of Operation
Time
Period^1
(mos.)
1-3





1-2
3-5
6-8
9-12
Period
1-10




1-9



1-9






1
2-12




Avg.
Flow
(Iphd)
ND





ND
154
ND
296
235
12




5



ND






ND
0




Peak
Flow
(Iphd)
ND





ND
229
ND
475
475
39




24



ND






ND
0




Active Period of Operation
Time
Period
(mos.)
4-16
17-28
29-40
41-44
Entire Period
45-58
13-18
19-20
21-23

Entire Period
11-22
23-28
29-40
41-45
Entire Period
10-17
18-33
34-46
Entire Period
10-21
22-32
33-35
36
37-41
42-46
Entire Period
13-24
25-27
28-31
32
33-38
Entire Period
Avg.
Flow
(Iphd)
23
6
8
6
12
43
102
ND
110

105
38
ND
8
4
20
5
ND
5
5
2
5
ND
56
ND
96
17
0
ND
117
ND
80
36
Peak
Flow
(Iphd)
32
11
11
8
32
112
229
ND
253

253
68
ND
26
7
68
24
ND
8
24
19
29
ND
56
ND
375
375
0
ND
221
ND
414
414
Post-Closure Period
Time
Period
(mos.)
NA




NA





NA




NA



NA






NA





Avg.
Flow
(Iphd)
NA




NA





NA




NA



NA






NA





Peak
Flow
(Iphd)
NA




NA





NA




NA



NA






NA





Notes




































m

oo
o
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

U1



U2


U3
U4

V1





V2





V3




V4




Prim.
Liner/
LDS
Type
II



II


II
II

II





II





II




II





Waste
Type/
Region
MSW(NE)



MSW(NE)


MSW(NE)
MSW(NE)

MSW(SE)





MSW(SE)





MSW(SE)




MSW(SE)




Initial Period of Operation
Time
Period^1
(mos.)
1-7



1
2-14

1-12
1
2-8
1-13





1-13





1
2-7


Period
1-11




Avg.
Flow
(Iphd)
ND



ND
218

128
ND
14
117





135





669
220


284
247




Peak
Flow
(Iphd)
ND



ND
487

187
ND
40
153





256





669
258


669
573




Active Period of Operation
Time
Period
(mos.)
8-19
20-31
32-42
Entire Period
15-26
27-36
Entire Period
13-25
9-22

14-25
26-37
38-49
50-61
62-64
Entire Period
14-25
26-37
38-49
50-61
62-64
Entire Period
8-20
21-32
33-44
45-54
Entire Period
12-23
24-35
36-47
48-53
Entire Period
Avg.
Flow
(Iphd)
220
170
118
171
42
28
36
147
1

51
36
41
39
17
40
54
46
37
30
26
41
118
75
55
45
76
102
69
70
68
79
Peak
Flow
(Iphd)
373
285
181
373
74
44
74
528
14

68
58
227
77
18
227
70
86
68
33
26
86
139
97
73
58
139
392
197
100
92
392
Post-Closure Period
Time
Period
(mos.)
NA



NA


NA
NA

NA





NA





NA




NA




Avg.
Flow
(Iphd)
NA



NA


NA
NA

NA





NA





NA




NA




Peak
Flow
(Iphd)
NA



NA


NA
NA

NA





NA





NA




NA




Notes





Side slopes were capped after 24 months



Side slopes were capped after 19 months
Side slopes were capped after 10 months

Side slopes were capped after 7 months






















m
oo
-------
       Table E-3.6.  Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

V5



W1


W2



X1


Y1






Y2




Z1



Prim.
Liner/
LDS
Type
II



II


II



II


III






III




III




Waste
Type/
Region
MSW(SE)



MSW(SE)


MSW(SE)



MSW(SE)


ASH(M)(NE)






MSW(NE)




ASH(C)(NE)



Initial Period of Operation
Time
Period^1
(mos.)
1-4
5-8
9-14
Period
1-8
9-12

1-8



1
2-7
Period
1-6
7-12





1-10




1-12



Avg.
Flow
(Iphd)
300
2
249
193
ND
439

ND



364
4
55
ND
381





655




84



Peak
Flow
(Iphd)
618
3
432
618
ND
765

ND



364
25
364
ND
1,341





1,768




145



Active Period of Operation
Time
Period
(mos.)
15-24



13-24
25-35
Entire Period
9-20
21-32
33-35
Entire Period
8-19
20-33
Entire Period
13-23
24-30
31-42
43-54
55-66
67-78
Entire Period
11-22
23-34
35-46
47-54
Entire Period
13-24
25-36
37-39
Entire Period
Avg.
Flow
(Iphd)
64



34
19
27
594
204
32
358
5
0
2
242
ND
285
377
167
98
230
370
90
70
48
153
64
86
104
78
Peak
Flow
(Iphd)
125



109
44
109
1,826
1,217
52
1,826
45
2
45
1,031
ND
643
1,157
393
530
1,157
1,993
168
248
56
1,993
107
148
150
150
Post-Closure Period
Time
Period
(mos.)
NA



NA


NA



NA


NA






NA




NA



Avg.
Flow
(Iphd)
NA



NA


NA



NA


NA






NA




NA



Peak
Flow
(Iphd)
NA



NA


NA



NA


NA






NA




NA



Notes

































m
oo
-------
       Table E-3.6.  Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AA1





AA2


AA2-3



AB1







AB2







Prim.
Liner/
LDS
Type
III





III


III



III







III








Waste
Type/
Region
MSW(NE)





MSW(NE)


MSW(NE)



HW(SE)







HW(SE)







Initial Period of Operation
Time
Period^1
(mos.)
1
2-8




1-3
4-15
Period




1-5







1-12







Avg.
Flow
(Iphd)
ND
35




792
338
429




275







16







Peak
Flow
(Iphd)
ND
57




1,021
560
1,021




518







33







Active Period of Operation
Time
Period
(mos.)
9-14
15-26
27-31
32-43
44-51
Entire Period
16-20


21-32
33-44
45-54
Entire Period
6-17
18-29
30-41
42-53
54-65
66-77
78-87
Entire Period
13-24
25-36
37-48
49-57
58-63
64-75*
76-87*
Entire Period
Avg.
Flow
(Iphd)
1
93
251
37
39
73
92


128
63
63
87
55
91
90
66
98
37
17
66
20
24
29
30
27
64
82
41
Peak
Flow
(Iphd)
3
318
794
105
122
794
133


251
144
201
251
80
337
142
134
209
89
23
337
25
36
37
42
35
115
142
142
Post-Closure Period
Time
Period
(mos.)
NA





NA


NA



NA







NA







Avg.
Flow
(Iphd)
NA





NA


NA



NA







NA







Peak
Flow
(Iphd)
NA





NA


NA



NA







NA







Notes
































m

oo
CO
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AB3






AB4





AC1
AC2







ACS





Prim.
Liner/
LDS
Type
III






III





IV
IV







IV






Waste
Type/
Region
HW(SE)






HW(SE)





HW(W)
HW(W)







HW(W)





Initial Period of Operation
Time
Period^1
(mos.)
1-12






1-4
5-13




1-13
1-6







1-12





Avg.
Flow
(iphd)
39






ND
333




94
4







94





Peak
Flow
(Iphd)
224






ND
471




206
11







173





Active Period of Operation
Time
Period
(mos.)
13-25
26-32
33-44
45-56
57-68
69-74
Entire Period
14-25
26-37
38-49
50-59
60-63
Entire Period
NA
7-18
19-30
31-42
43-54
55-66
67-78
79-88
Entire Period
13-24
25-36
37-48
49-60
61-73
Entire Period
Avg.
Flow
(Iphd)
68
ND
40
21
13
15
34
77
50
50
35
ND
54
NA
65
66
96
30
19
12
7
43
123
71
47
19
12
51
Peak
Flow
(Iphd)
232
ND
108
45
21
18
232
107
98
64
47
ND
107
NA
107
128
164
40
34
17
13
164
159
98
82
26
25
159
Post-Closure Period
Time
Period
(mos.)
NA






NA





NA
NA







NA





Avg.
Flow
(Iphd)
NA






NA





NA
NA







NA





Peak
Flow
(Iphd)
NA






NA





NA
NA







NA





Notes

















Cell received intermediate cover
approximately 56 months after start
of waste placement in cell





Cell received intermediate cover
approximately 41 months after start
of waste placement in cell



m
oo
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AC4





ACS


AC6




AC7
ACS
AD1








Prim.
Liner/
LDS
Type
IV





IV


IV




IV
IV
III









Waste
Type/
Region
HW(W)





HW(W)


HW(W)




HW(W)
HW(W)
HW(SE)








Initial Period of Operation
Time
Period1'
(mos.)
1-11





1-13


1-10




1-12
1-12
1-12








Avg.
Flow
(iphd)
14





1


1




0
0
ND








Peak
Flow
(Iphd)
97





3


5




0
3
ND








Active Period of Operation
Time
Period
(mos.)
12-23
24-35
36-47
48-59
60-73
Entire Period
14-25
26-34
Entire Period
11-16
17-28
29-34
35-40
Entire Period
13-18
13-18
13-20
21-32







Avg.
Flow
(Iphd)
127
52
42
16
7
47
6
1
4
54
3
1
2
12
0
2
ND
107







Peak
Flow
(Iphd)
495
73
73
25
15
495
44
1
44
219
9
2
4
219
0
4
ND
603







Post-Closure Period
Time
Period
(mos.)
NA





NA


NA




NA
NA
33-44
45-56
57-68
69-80
81-92
93-104
105-116
117-121
Entire Period
Avg.
Flow
(Iphd)
NA





NA


NA




NA
NA
24
26
28
42
23
8
5
6
21
Peak
Flow
(Iphd)
NA





NA


NA




NA
NA
42
31
45
103
68
46
43
24
103
Notes



Cell received intermediate cover
approximately 40 months after start
of waste placement in cell






















m

oo
en
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).
Cell
No.
AD2








ADS








AD4








Prim.
Liner/
LDS
Type
III








III








III








Waste
Type/
Region
HW(SE)








HW(SE)








HW(SE)








Initial Period of Operation
Time
Period^1
(mos.)
1-12








1-16








1-16








Avg.
Flow
(iphd)
ND








ND








ND








Peak
Flow
(Iphd)
ND








ND








ND








Active Period of Operation
Time
Period
(mos.)
13-20
21-30







17-26








17-26








Avg.
Flow
(Iphd)
ND
258







231








303








Peak
Flow
(Iphd)
ND
1,517







665








1,517








Post-Closure Period
Time
Period
(mos.)
31-42
43-54
55-66
67-78
79-90
91-102
103-114
115-119
Entire Period
27-38
39-50
51-62
63-74
75-86
87-98
99-1 1 0
111-115
Entire Period
27-38
39-50
51-62
63-74
75-86
87-98
99-1 1 0
111-115
Entire Period
Avg.
Flow
(Iphd)
39
34
26
14
12
16
11
14
21
82
96
50
50
29
41
37
0
52
56
41
44
34
23
1
2
0
27
Peak
Flow
(Iphd)
62
45
44
43
28
27
47
25
62
118
138
148
73
57
72
108
1
148
90
51
79
43
44
2
9
1
90
Notes




























m

oo
CD
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).
Cell
No.
ADS








AD6








AD7





ADS






Prim.
Liner/
LDS
Type
III








III








IV





IV






Waste
Type/
Region
HW(SE)








HW(SE)








HW(SE)





HW(SE)






Initial Period of Operation
Time
Period^1
(mos.)
1-9








1-9








1-12





1-8






Avg.
Flow
(iphd)
ND








ND








135





201






Peak
Flow
(Iphd)
ND








ND








1,101





554






Active Period of Operation
Time
Period
(mos.)
10-19








10-19








13-24
25-36
37-48
49-60
61-69
Entire Period
9-19
20-24
25-36
37-48
49-60
61-64
Entire Period
Avg.
Flow
(Iphd)
547








763








71
96
17
33
64
56
108
785
82
46
45
79
134
Peak
Flow
(Iphd)
1,417








1,295








291
393
21
74
112
393
183
1,259
175
66
131
202
1,259
Post-Closure Period
Time
Period
(mos.)
20-31
32-43
44-55
56-67
68-79
80-91
92-103
1 04-1 08
Entire Period
20-31
32-43
44-55
56-67
68-79
80-91
92-103
1 04-1 08
Entire Period
70-81
82-87



Entire Period
65-76
77-84




Entire Period
Avg.
Flow
(Iphd)
56
43
66
94
119
142
155
270
106
117
61
41
30
26
12
0
0
39
73
105



83
248
313




274
Peak
Flow
(Iphd)
132
68
175
168
244
237
315
486
486
173
83
90
38
55
136
2
1
173
157
172



172
476
564




564
Notes



















Cell received intermediate cover
approximately 12 months after start
of waste placement in cell



Cell received intermediate cover
approximately 12 months after start
of waste placement in cell




m
oo
-------
        Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).
Cell
No.
AD9




AD10






AD11






AD12





AD13




Prim.
Liner/
LDS
Type
IV




IV






IV






IV





IV




Waste
Type/
Region
HW(SE)




HW(SE)






HW(SE)






HW(SE)





HW(SE)




Initial Period of Operation
Time
Period^1
(mos.)
1-9




1-12






1-5






1-13





1-12




Avg.
Flow
(iphd)
157




234






116






136





72




Peak
Flow
(Iphd)
421




436






155






369





365




Active Period of Operation
Time
Period
(mos.)
10-21
22-33
34-45
46-58
Entire Period
13-24
25-36
37-50
51-62*
63-74
75-79
Entire Period
6-17
18-29
30-41
42-53
54-65
66-69
Entire Period
14-25
26-37
38-49
50-61
62-69
Entire Period
13-18
19-30
31-42
43-55
Entire Period
Avg.
Flow
(Iphd)
129
27
23
18
49
181
67
41
150
124
118
111
ND
90
81
88
82
73
84
84
217
120
102
64
121
80
92
184
172
140
Peak
Flow
(Iphd)
297
60
36
27
297
404
159
64
471
350
237
471
ND
214
167
148
197
131
214
193
383
304
406
98
406
162
237
490
400
490
Post-Closure Period
Time
Period
(mos.)
59-60
61-72
73-79

Entire Period
NA






NA






NA





NA




Avg.
Flow
(Iphd)
92
17
10

22
NA






NA






NA





NA




Peak
Flow
(Iphd)
135
32
29

135
NA






NA






NA





NA




Notes

Cell received intermediate cover
approximately 12 months after start
of waste placement in cell


Cell received intermediate cover
approximately 41 months after start
of waste placement in cell

1/2 of cell was closed approximately 59
months after start of waste placement
in cell
Cell received intermediate cover
approximately 41 months after start
of waste placement in cell




Cell received intermediate cover
approximately 41 months after start
of waste placement in cell



Cell received intermediate cover
approximately 41 months after start
of waste placement in cell


m
oo
oo
-------
        Table E-3.6.  Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AD14




AD15


AD16


AD17
AD18
AE1






AE2




AE3
AF1





Prim.
Liner/
LDS
Type
IV




IV


IV


IV
IV
III






III




III
III






Waste
Type/
Region
HW(SE)




HW(SE)


HW(SE)


HW(SE)
HW(SE)
MSW(NE)






MSW(NE)




MSW(NE)
ASH(M)(NE)





Initial Period of Operation
Time
Period^1
(mos.)
1-12




1-13


1-12


1-12
1-12
1-12






1-8




1-11
1-12





Avg.
Flow
(iphd)
61




117


188


206
131
ND






ND




ND






Peak
Flow
(Iphd)
119




277


701


353
193
ND






ND




ND






Active Period of Operation
Time
Period
(mos.)
13-24
25-36
37-48
49-55
Entire Period
14-25
26-39
Entire Period
13-24
25-39
Entire Period
13-16
13-19
13-24
25-35
36-47
48-54
55-66
67-80
Entire Period
9-20
21-32
33-44
45-49
Entire Period
NA
13-24
25-36
37-48
49-60
61-63
Entire Period
Avg.
Flow
(Iphd)
132
137
95
192
133
134
150
143
163
222
196
104
83
ND
ND
108
83
256
130
150
205
197
149
101
174
NA
67
53
27
11
8
38
Peak
Flow
(Iphd)
364
329
144
281
364
246
270
270
568
459
568
119
113
ND
ND
192
119
1,111
166
1,111
267
241
179
129
267
NA
95
92
38
23
8
95
Post-Closure Period
Time
Period
(mos.)
NA




NA


NA


NA
NA
NA






NA




NA
NA





Avg.
Flow
(Iphd)
NA




NA


NA


NA
NA
NA






NA




NA
NA





Peak
Flow
(Iphd)
NA




NA


NA


NA
NA
NA






NA




NA
NA





Notes



Cell received intermediate cover
approximately 41 months after start
of waste placement in cell


Cell received intermediate cover
approximately 41 months after start
of waste placement in cell
Cell received intermediate cover
approximately 41 months after start
of waste placement in cell





















m

oo
CD
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AG1


AH1


AH2



AH3



AH4


AH5


AM



AI2




AJ1

AK1
AK1-2
Prim.
Liner/
LDS
Type
III


III


III



III



III


III


III



III




III

IV
IV

Waste
Type/
Region
MSW(NE)


MSW/ASH(M)(NE)


MSW/ASH(M)(NE)



MSW/ASH(M)(NE)



MSW/ASH(M)(NE)


MSW/ASH(M)(NE)


C&DW(NE)



C&DW(NE)




MSW(NE)

MSW(NE)
MSW(NE)
Initial Period of Operation
Time
Period^1
(mos.)
1-7


1-6


1-5



1-6



1
2-12

1-12


1-2
3-6
7-12
Period
1-2
3-6



1
2-6
1-12

Avg.
Flow
(Iphd)
85


8


10



8



ND
5

18


ND
1,177
359
687
ND
539



ND
279
206

Peak
Flow
(Iphd)
222


18


39



18



ND
17

48


ND
1,596
579
1,596
ND
893



ND
625
804

Active Period of Operation
Time
Period
(mos.)
8-19
20-33
Entire Period
7-18
19-28
Entire Period
6-13
14-25
26-28
Entire Period
7-12
13-18
19-28
Entire Period
13-18
19-28
Entire Period
13-24
25-28
Entire Period
13-24
25-36
37-50
Entire Period
7-18
19-30
31-42
43-50
Entire Period
7-13


11-15
Avg.
Flow
(Iphd)
1
0
1
64
63
64
ND
60
70
62
43
5
50
36
2
43
28
32
26
31
78
65
101
82
97
46
58
61
66
24


492
Peak
Flow
(Iphd)
2
1
2
92
103
103
ND
89
86
89
49
13
78
78
6
83
83
82
40
82
220
200
227
227
156
63
98
120
156
47


1,424
Post-Closure Period
Time
Period
(mos.)
NA


NA


NA



NA



NA


NA


NA



NA




NA

NA
NA
Avg.
Flow
(Iphd)
NA


NA


NA



NA



NA


NA


NA



NA




NA

NA
NA
Peak
Flow
(Iphd)
NA


NA


NA



NA



NA


NA


NA



NA




NA

NA
NA
Notes




































m
CD
O
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AL1



AM1






AM2






AN1
AN 1-2
AN 1-3


AN 1-4
AO1



AO2


AO3
AO4
Prim.
Liner/
LDS
Type
IV



IV






IV






IV





IV



IV


IV
IV

Waste
Type/
Region
MSW(NE)



MSW(W)






MSW(W)






ASH(M)(NE)





MSW(NE)



MSW(NE)


MSW(NE)
MSW(NE)
Initial Period of Operation
Time
Period^1
(mos.)
1-12



1-9






1-9






1-7





1-5



1-5


1-8
1-8
Avg.
Flow
(iphd)
ND



ND






ND






ND





ND



149


800
385
Peak
Flow
(Iphd)
ND



ND






ND






ND





ND



191


1,089
583
Active Period of Operation
Time
Period
(mos.)
13-29
30-41
42-54
Entire Period
10-21
22-33
34-45
46-57
58-69
70-81
Entire Period
10-21
22-33
34-45
46-57
58-69
70-81
Entire Period
8-12
13-17
18-21
22-31
Entire Period
32-34
6-17
16-28
29-36
Entire Period
6-17
18-30
Entire Period
NA
NA
Avg.
Flow
(Iphd)
ND
221
103
164
15
10
3
1
1
5
6
9
9
3
0
8
9
6
13
46
65
173
137
523
184
96
60
119
110
33
70
NA
NA
Peak
Flow
(Iphd)
ND
367
183
367
64
15
14
1
1
8
64
42
29
26
0
13
13
42
21
136
144
368
368
580
353
126
102
353
158
64
158
NA
NA
Post-Closure Period
Time
Period
(mos.)
NA



NA






NA






NA





NA



NA


NA
NA
Avg.
Flow
(Iphd)
NA



NA






NA






NA





NA



NA


NA
NA
Peak
Flow
(Iphd)
NA



NA






NA






NA





NA



NA


NA
NA
Notes







Cell received intermediate cover
approximately 5 months after start
of waste placement in cell




Cell received intermediate cover
approximately 5 months after start
of waste placement in cell



















m
CD
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AP1



AQ1




AQ2




AQ3




AQ4




AQ5




Prim.
Liner/
LDS
Type
IV



IV




IV




IV




IV




IV





Waste
Type/
Region
HW(W)



HW(NE)




HW(NE)




HW(NE)




HW(NE)




HW(NE)




Initial Period of Operation
Time
Period^1
(mos.)
1-12



1-6




1-12




1-12




1-3
4
5-13
Period

1-12




Avg.
Flow
(Iphd)
43



352




340




472




299
ND
356
342

127




Peak
Flow
(Iphd)
162



569




669




2,195




299
ND
642
642

688




Active Period of Operation
Time
Period
(mos.)
13-24
25-36
37-48
Entire Period
7-25
26-34
35-46
47-58
Entire Period
13-25
26-34
35-46
47-58
Entire Period
13-21
22-30
31-42
43-54
Entire Period
14-20
21-29
30-41
42-53
Entire Period
13-20
21-32
33-44
45-50
Entire Period
Avg.
Flow
(Iphd)
27
24
26
26
255
ND
197
116
196
437
ND
227
124
267
451
ND
287
161
286
121
ND
195
148
160
ND
81
131
119
109
Peak
Flow
(Iphd)
100
69
58
100
1,239
ND
435
143
1,239
1,550
ND
312
191
1,550
1,390
ND
427
265
1,390
220
ND
246
278
278
ND
140
709
179
709
Post-Closure Period
Time
Period
(mos.)
NA



59-65
66-77
78-89
90-97
Entire Period
59-70
71-82
83-94
95-97
Entire Period
55-66
67-78
79-90
91-93
Entire Period
54-65
66-77
78-89
90-92
Entire Period
51-62
63-74
75-84

Entire Period
Avg.
Flow
(Iphd)
NA



215
117
98
51
115
136
94
52
4
87
179
137
89
52
129
151
96
80
69
106
71
47
62

60
Peak
Flow
(Iphd)
NA



246
165
132
118
246
287
168
71
12
287
281
263
150
73
281
204
131
127
87
204
131
99
112

131
Notes







Cell final closure date is approximate














Cell final closure date is approximate









m
CD
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AQ6




AQ7



AQ8



AQ9



AQ10



AR1


AR1-2
AS1




AT1


Prim.
Liner/
LDS
Type
IV




IV



IV



IV



IV



IV



IV




IV



Waste
Type/
Region
HW(NE)




HW(NE)



HW(NE)



HW(NE)



HW(NE)



MSW(NE)



HW(SE)




HW(W)


Initial Period of Operation
Time
Period^1
(mos.)
1-12




1-6



1-12



1-12



1-9



1-11



1-3




1-5


Avg.
Flow
(iphd)
ND




38



54



24



14



292



252




ND


Peak
Flow
(Iphd)
ND




53



118



75



32



705



283




ND


Active Period of Operation
Time
Period
(mos.)
13-16
17-28
29-40
41-52
Entire Period
7-15
16-27
28-42
Entire Period
13-26



13-26



10-14
15-26

Entire Period
12-23
24-36
Entire Period
37-40
4-15
16-27


Entire Period
6-8


Avg.
Flow
(Iphd)
ND
289
202
173
222
ND
77
112
97
111



126



26
48

42
181
155
167
92
51
17


34
8


Peak
Flow
(Iphd)
ND
766
333
291
766
ND
160
488
488
349



429



32
250

250
470
442
470
217
110
52


110
15


Post-Closure Period
Time
Period
(mos.)
53-64
65-76
77-80

Entire Period
43-54
55-66
67-79
Entire Period
27-38
39-50
51-63
Entire Period
27-38
39-50
51-63
Entire Period
27-38
39-50
51-63
Entire Period
NA


NA
28-39
40-51
52-63
64-71
Entire Period
9-20
21-33
Entire Period
Avg.
Flow
(Iphd)
138
128
136

134
56
42
45
48
41
44
33
39
48
112
82
81
29
18
24
24
NA


NA
6
3
2
1
3
0
0
0
Peak
Flow
(Iphd)
227
197
166

227
108
79
64
108
77
68
75
77
95
206
179
206
48
63
75
75
NA


NA
35
17
7
4
35
1
0
1
Notes
























Cell AR1 has four subareas. Waste
placement started in months 1 , 5, 20, and
26 in subareas 1 through 4, respectively.









m

CD
CO
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AU1



AV1








AV2








AV3






Prim.
Liner/
LDS
Type
IV



IV








IV








IV







Waste
Type/
Region
MSW(NE)



HW(NE)








HW(NE)








HW(NE)






Initial Period of Operation
Time
Period^1
(mos.)
1-4



1-4
5-12







1-12








1-3
4-13





Avg.
Flow
(iphd)
ND



695
ND







ND








ND
409





Peak
Flow
(Iphd)
ND



1,735
ND







ND








ND
976





Active Period of Operation
Time
Period
(mos.)
5-16
17-29
30-32
Entire Period
13-24
25-33
34-45
46-57
58-69
70-81
82-93
94-104
Entire Period
13-20
21-32
33-44
45-56
57-68
69-80
81-90
91
Entire Period
14-25
26-34
35-46
47-58
59-71
72-74
Entire Period
Avg.
Flow
(Iphd)
56
30
ND
42
ND
ND
156
86
43
54
30
40
53
ND
88
59
16
19
13
17
ND
32
122
66
7
7
18
ND
41
Peak
Flow
(Iphd)
274
101
ND
274
ND
ND
660
114
55
86
55
71
660
ND
319
96
49
40
32
29
ND
319
188
137
20
12
48
ND
188
Post-Closure Period
Time
Period
(mos.)
NA



NA








NA








NA






Avg.
Flow
(Iphd)
NA



NA








NA








NA






Peak
Flow
(Iphd)
NA



NA








NA








NA






Notes
































m
CD
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AV4





AV5



AW1
AW1.3
AW2
AW2.4
AX1




AX2




AX3




Prim.
Liner/
LDS
Type
IV





IV



V
V,ll
V
V,ll
V




V




V





Waste
Type/
Region
HW(NE)





HW(NE)



MSW(SE)

MSW(SE)

MSW(NE)




MSW(NE)




MSW(NE)




Initial Period of Operation
Time
Period^1
(mos.)
1
2-5



Period
1-12



1-11

1-11

1-2




1-5




1-5




Avg.
Flow
(Iphd)
1,767
54



397
180



131

290

0




15




35




Peak
Flow
(Iphd)
1,767
68



1,767
325



524

514

0




45




151




Active Period of Operation
Time
Period
(mos.)
6-17
18-29
30-41
42-53
54-66
Entire Period
13-24
25-37
38-40
Entire Period

12-24

12-21
3-14
15-26
27-33

Entire Period
3-14
15-26
27-33

Entire Period
6-17
18-29
30-41
42-56
Entire Period
Avg.
Flow
(Iphd)
293
189
65
38
27
124
143
60
79
98

24

33
0
0
0

0
5
1
0

2
6
0
0
9
4
Peak
Flow
(Iphd)
323
372
121
65
52
372
205
127
136
205

127

83
0
0
0

0
21
2
0

21
44
1
0
47
47
Post-Closure Period
Time
Period
(mos.)
NA









NA
NA
NA
NA
34-45
46-57
58-69
70-83
Entire Period
34-45
46-57
58-69
70-83
Entire Period
57-68
69-81


Entire Period
Avg.
Flow
(Iphd)
NA









NA
NA
NA
NA
0
0
0
0
0
0
0
0
0
0
1
1


1
Peak
Flow
(Iphd)
NA









NA
NA
NA
NA
0
0
0
0
0
0
0
0
0
0
9
10


10
Notes



























Cell received intermediate cover
approximately 34 months after start
of waste placement in cell


m
CD
cn
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).
Cell
No.
AX4




AX5






AX6






AX7







Prim.
Liner/
LDS
Type
V




V






V






V







Waste
Type/
Region
MSW(NE)




MSW(NE)






MSW(NE)






MSW(NE)







Initial Period of Operation
Time
Period^1
(mos.)
1-12




1-11






1-9






1-10







Avg.
Flow
(iphd)
101




37






53






34







Peak
Flow
(Iphd)
860




92






93






47







Active Period of Operation
Time
Period
(mos.)
13-17
18-29
30-41
42-56
Entire Period
12-23
24-35
36-47
48-59
60-71
72-80
Entire Period
10-21
22-33
34-45
46-57
58-69
70-78
Entire Period
11-22
23-25
26
27-38
39-50
51-62
63-76
Entire Period
Avg.
Flow
(Iphd)
2
1
1
0
1
4
3
0
2
0
0
2
0
0
0
0
0
0
0
4
8
ND
1
3
0
0
2
Peak
Flow
(Iphd)
13
3
9
0
13
11
37
0
9
0
0
37
0
0
0
0
0
0
0
6
9
ND
9
9
0
0
9
Post-Closure Period
Time
Period
(mos.)
57-68
69-81


Entire Period
NA






NA






NA







Avg.
Flow
(Iphd)
0
0


0
NA






NA






NA







Peak
Flow
(Iphd)
0
0


0
NA






NA






NA







Notes

Cell received intermediate cover
approximately 34 months after start
of waste placement in cell


Cell received intermediate cover
approximately 33 months after start
of waste placement in cell

Cell was partially closed approximately 55
months after start of waste placement in
cell
Cell received intermediate cover
approximately 30 months after start
of waste placement in cell

Cell was partially closed approximately 53
months after start of waste placement in
cell
Cell was partially closed approximately 51
months after start of waste placement in
cell





m

CD
CD
-------
       Table E-3.6. Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).
Cell
No.
AX8





AX9





AX10





AX11




AX12




AX13




Prim.
Liner/
LDS
Type
V





V





V





V




V




V




Waste
Type/
Region
MSW(NE)





MSW(NE)





MSW(NE)





MSW(NE)




MSW(NE)




MSW(NE)




Initial Period of Operation
Time
Period^1
(mos.)
1-14





1-9





1-11





1-14




1-12




1-7




Avg.
Flow
(iphd)
48





1





0





0




0




0




Peak
Flow
(Iphd)
189





7





0





0




0




0




Active Period of Operation
Time
Period
(mos.)
15-26
27-38
39-50
51-62
63-71
Entire Period
10-21
22-33
34-45
46-57
58-66
Entire Period
12-23
24-35
36-47
48-59
60-63
Entire Period
15-26
27-38
39-50
51-62
Entire Period
13-24
25-36
37-48
49-56
Entire Period
8-19
20-31
32-43
44-53
Entire Period
Avg.
Flow
(Iphd)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Peak
Flow
(Iphd)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Post-Closure Period
Time
Period
(mos.)
NA





NA





NA





NA




NA




NA




Avg.
Flow
(Iphd)
NA





NA





NA





NA




NA




NA




Peak
Flow
(Iphd)
NA





NA





NA





NA




NA




NA




Notes

Cell was partially closed approximately 46
months after start of waste placement in
cell



Cell was partially closed approximately 41
months after start of waste placement in
cell



Cell was partially closed approximately 34
months after start of waste placement in
cell



Cell was partially closed approximately 39
months after start of waste placement in
cell


Cell may have been partially closed
approximately 31 months after start
of waste placement in cell


Cell may have been partially closed
approximately 28 months after start
of waste placement in cell


m
CD
-------
        Table E-3.6.  Landfill LDS Flow Rate Data, Summarized by Landfill Life Cycle Stage (Continued).

Cell
No.

AX14




AX15


AX16


AY1
AY2
AYS
AZ1


BA1



BA2
BB1





BB2

BBS

Prim.
Liner/
LDS
Type
V




V


V


VI
VI
VI
VI


J2>



_(2)
VI





VI

VI


Waste
Type/
Region
MSW(NE)




MSW(NE)


MSW(NE)


HW(NE)
HW(NE)
HW(NE)
MSW(NE)


HW(NE)



HW(NE)
MSW(SE)





MSW(SE)

MSW(SE)

Initial Period of Operation
Time
Period^
(mos.)
1-10




1-12


1-10


1-9
1-11
1-11
1-2
3-12

1-14



1-2
1-6





1-2
3-11
1-2
3-11
Avg.
Flow
(Iphd)
0




0


0


0
3
6
ND
0

ND



324
15





ND
1
ND
0
Peak
Flow
(Iphd)
0




0


0


0
12
28
ND
0

ND



425
65





ND
12
ND
0
Active Period of Operation
Time
Period
(mos.)
11-22
23-30
31-32*
33-38
Entire Period
13-24
25-37
Entire Period
11-22
23-29
Entire Period
NA
NA
NA
13-24
25-31
Entire Period
15-26
27-38
39-42
Entire Period
3-12
7-17
18-19
20-23
24-35
36-47
Entire Period
12-23

12-23

Avg.
Flow
(Iphd)
0
0
0
0
0
0
0
0
0
0
0
NA
NA
NA
3
0
2
136
73
21
92
12
13
2
14
1
1
6
0

0

Peak
Flow
(Iphd)
0
0
0
0
0
0
0
0
0
0
0
NA
NA
NA
22
3
22
316
209
30
316
39
25
5
18
2
4
25
0

1

Post-Closure Period
Time
Period
(mos.)
NA




NA


NA


NA
NA
NA
NA


NA



NA
NA





NA

NA

Avg.
Flow
(Iphd)
NA




NA


NA


NA
NA
NA
NA


NA



NA
NA





NA

NA

Peak
Flow
(Iphd)
NA




NA


NA


NA
NA
NA
NA


NA



NA
NA





NA

NA

Notes




















Cell received intermediate cover
approximately 42 months after start
of waste placement in cell












m

CD
oo
        Notes: (1)  NA = not applicable, ND = not determined
(2) Cells BA1 and BA2 include a GM/GCL/GM primary liner.
-------
Table E-3.7. Select LCRS and LDS Flow Chemistry Data for Landfills.

Cell No.
System
Primary Liner\LDS Type
Waste Placement Dates
Liquid Sampling Dates
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1,1-Dichloroethane |ig/l
1 ,2-Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methylene chloride |ig/l
1,1,1-Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl Chloride |ig/l
Xylenes l-ig/l
MSW LANDFILLS IN NORTHEAST U.S.
B1-3
LCRS1

LDS1
I
LCRS2

LDS2
I
5/84-11/88
3/85-1/95

6.56
10,972
6,910
10,073
4,077
3,092

911
162

298
727
88
< 17
40
110
51
< 6
< 27
< 3

< 45
14
299
< 67
< 104
376
< 5

5/85-1/95

6.54
2,697
1,810
2,037
484
491

195
43

43
149
61
< 9
28
< 58
36
< 5
< 56
< 3

< 43
< 5
< 29
< 112
< 119
< 178
< 5

3/85-1/95

6.42
9,381
5,991
10,938
3,415
3,051

894
135

261
757
68
< 15
53
117
38
< 6
< 44
< 3

< 34
< 12
< 186
< 105
< 123
337
< 5

4/85-1/95

6.42
6,821
2,143
2,357
1,083
803

344
36

69
292
110
18
108
< 101
< 38
< 9
< 25
< 7

< 19
< 8
< 24
< 140
< 123
140
< 6

LCRS3

LDS3
III
7/87-5/92
7/87-1 0/94

6.82
2,956
4,140
1,912
422
554

690
131


450

< 20
49
< 44
102
< 5
< 40
< 5


15
< 80
< 107

< 78
< 10

7/87-1 0/94

(1)
1,554
1,148
131
88
138

148
335


46
< 24
16
3
< 17
519
< 7
< 7
< 6


< 6
< 17
< 6

< 9
< 12

F1
LCRS

LDS
I
7/92-date
2/93-6/94

6.16
1,833
1,917
1,747
804
306
895
101
8
278
31
65
< 6
8
15
< 4
54
< 15
114
< 20

< 17
< 14
264
269
15
84
< 77
38
2/93-6/94

7.50
830
544
35
< 3.5(1)
10
408
108
40
64
8
147
< 1
< 7
< 28
< 4
< 30
< 1
< 1
< 1

< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
G1-3
LCRS/LDS1-3

LCRS/LDS1-2

LDS1-2
I
6/89-date
1/90-10/94

6.68
7,683
7,718
8,267
4,056
2,852
3,278
1,561
84
610
111
1,115
8
7
320
4
77
< 36
55
< 35
< 75
< 34
< 37
241
< 67
< 36
633
< 53
62
4/90

7.25
1,355

633
465
197
373
243
38
100
18
136
< 1
< 2
35
7
< 30












4/90

7.70
470

27
23
8
175
25
59
53

14
< 1
< 2
< 1
< 2
< 30












-------
        Table E-3.7.  Select LCRS and LDS Flow Chemistry Data for Landfills (Continued).

Cell No.
System
Primary Liner\LDS Type
Waste Placement Dates
Liquid Sampling Dates
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1 ,1-Dichloroethane |ig/l
1 ,2-Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methylene chloride |ig/l
1 ,1,1-Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl Chloride |ig/l
Xylenes ng/l
MSW LANDFILLS IN NORTHEAST U.S.
12
LCRS

LDS
I
10/87-8/94
3/88-2/92

6.30
5,332
3,663
3,350
4,510
1,104
1,602
356
29
497
175
225
< 10
10
< 50
< 40

21
294
20


85
1,303
84
46
959
< 60
277
5/88-1 0/93

6.65
2,345

< 50
< 6
96
497
131
98
89
60
46
< 10
< 5
< 50
< 30
< 40
1
110
< 1

4
1
480
220
24
5
< 1

J1-6
LCRS1-6

LDS1-6
I
LDS1
I
LDS2
I
LDS3
I
LDS4
I
LDS5
I
10/90-date
1 0/90-4/94

7.40
1,066

330


503
28
147
66
159
33
47
< 5
< 90

< 38
< 2
< 2
< 1
< 1
< 1
< 1
< 1
< 2
< 5
< 1
< 3
< 5
1 0/90-4/94

6.58
1,257

55

36
965
18
9
217
45
8
< 1
4
5
11

< 1
7
< 1
< 3
< 1
< 1
< 1
< 2
< 1
7
< 1
< 1
10/90-4/94

6.73
1,423

49

34
879
32
7
241
57
7

2
2
17

< 1
8
< 1
8
8
< 1
6
4
< 10

< 1
< 1
10/90-4/94

6.72
1,297

22

26
813
21
6
201
54
16

14
8
23

< 1
5
< 1
< 1
< 1
< 1
2
1
< 1
< 1
< 1
< 1
10/90-4/94

6.37
1,035

84

41
863
12
14
246
43
5

254
258
26

< 1
4
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
10/90-4/94

6.52
1,276

77

52
801
21
12
229
36
5

4
5
4

< 1
7
< 1
< 1
< 1
< 1
< 1
< 1
< 1
1
< 1
< 1
10/90-4/94

6.60
985

41

26
410
19
10
170
32
7

0.1

2

< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
27
< 1
< 1
K1
LCRS

LDS
I
12/89-date
12/91-2/93



3,820
1,619
481
957
2,401
341
118
362
84
413
17
< 8
42
88
58
20
23
< 100

< 100
87
107
< 100
< 100
652
< 100
267
12/91-2/93



1,127
252
459
203
898
208
72
168
30
69
31
< 4
28
< 71
< 20
14

< 100

< 100
57
85
< 100
< 100
437
< 100
179
8
-------
Table E-3.7. Select LCRS and LDS Flow Chemistry Data for Landfills (Continued).

Cell No.
System
Primary Liner\LDS Type
Waste Placement Dates
Liquid Sampling Dates
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1 ,1-Dichloroethane |ig/l
1 ,2-Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methylene chloride |ig/l
1 ,1,1-Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl Chloride |ig/l
Xylenes ng/l
MSW LANDFILLS IN NORTHEAST U.S.
Q1
LCRS/LDS

LDS
II
3/90-date
2/93-2/94
7.00
3,148
1,648
199
9

1,298
299
7
199
70
211
11
< 2
6
< 2
42












2/93-2/94

1,455
961
187
< 3

841
37
66
199
54
35
< 2
< 2
< 3
2
< 12












R1
LCRS/LDS

LCRS/LDS

LDS
II
5/93-date
5/93-5/94

2,665
2,041
3,742
1,133
593
973
181
114
231
55
147
10
< 2
20
6
220












5/94

3,240
2,450
2,910
1,500
880
1,240
240
62
367
70
189
16
< 2
35
9
366












5/94

2,830
2,180
2,420
3
720
1,270
182
< 5
347
68
141
24
< 2
38
6
579












S3
LCRS

LDS
II
2/87-1 0/92
7/91-4/94
6.30

1,533
1,499
572
520

118
27


99





14
239
7

16
21
90
18
7
264
9

10/91-10/93
6.59

835
77
38
61

22



92





< 5
< 16
< 5

< 5
6
8
< 6
< 8
< 5
< 10

-------
Table E-3.7. Select LCRS and LDS Flow Chemistry Data for Landfills (Continued).

Cell No.
System
Primary Liner\LDS Type
Waste Placement Dates
Liquid Sampling Dates
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1 ,1-Dichloroethane |ig/l
1 ,2-Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methylene chloride |ig/l
1 ,1,1-Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl Chloride |ig/l
Xylenes ng/l
MSW LANDFILLS IN NORTHEAST U.S.
T1-2
LCRS1

5/91 -date
6/94-3/95
6.85
13,035

5,160


4,985
1,163
< 20
583

1,058





< 14
68
< 13
< 80
< 13
108
171
< 13
< 84
1,050
< 27
351
LCRS2

LDS2
II
1/92-date
6/94-3/95
7.10
14,060

6,188


6,168
1,360
< 15
465

1,380





< 15
< 15
< 15
< 13
< 13
< 18
< 17
< 15
< 15
283
< 30
< 88
9/94-3/95
7.20
1 1 ,633



1,263

1,263
< 11
520







< 25
< 25
< 25
< 25
< 25
32
< 25
< 25

330
< 50
35
U2
LCRS

LDS
II
7/86-1 987/88
1/87-3/87
6.48
5,063
6,510
2,565
2,760

2,127
391
23



21

100

98

209




969


507


10/86-3/87
6.55
1,406
1,112
34
14

359
63
433



18





9




46
13
3



Y2
LCRS

LDS
III
1990-date
4/91-4/94
6.60
5,360
4,939
5,265
2,076
1,436
2,520
628
108
1,994

433
17
8
58
5
185
10
45



118
121

11
720

69
4/91-4/94
7.28
1,583
881
50
3
11
335
58
231
179
54
46
4
< 2
11
8
25









7


-------
        Table E-3.7.  Select LCRS and LDS Flow Chemistry Data for Landfills (Continued).

Cell No.
System
Primary Liner\LDS Type
Waste Placement Dates
Liquid Sampling Dates
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1 ,1-Dichloroethane |ig/l
1 ,2-Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methylene chloride |ig/l
1 ,1,1-Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl Chloride |ig/l
Xylenes ng/l
MSW LANDFILLS IN NORTHEAST U.S.
AA1-3
LCRS1

LDS1
III
10/90-1992
1994
6.79
2,364
8,511
6,603
1,778
2,978
4,789
1,571
222
548
218
415

1

10













1994
6.84
1,913
4,588
939
65
648
3,720
1,039
77
347
213
1,870

1

18













LCRS2-3

LDS2-3
III
7/90-date
1994
7.20
3,405
8,731
5,572
2,534
2,239
6,810
1,563
157
374
163
1,270

2

16













1994
6.52
575
1,810
114
60
20
800
70
651
438
73
34

< 0.2

< 1













AE1
LCRS/LDS

4/88-date
2/93-1 2/93




< 2
4

333
1,943



< 4
< 5
< 10
< 3
< 37
< 5
< 5
< 5


< 5
< 5
< 5
< 5
< 5
< 10
< 5
AG1
LCRS

LDS
III
4/92-date
1993
7.28
3,098
2,510
< 10
416
49
786
141
1,376

145
23
< 2
< 2
3
7
< 10












1993
7.35
1,559
1,273
32

31
181
18
720
261
72
12
12
< 2
4
14
< 10












8
-------
        Table E-3.7.  Select LCRS and LDS Flow Chemistry Data for Landfills (Continued).

Cell No.
System
Primary Liner\LDS Type
Waste Placement Dates
Liquid Sampling Dates
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1 ,1-Dichloroethane |ig/l
1 ,2-Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methylene chloride |ig/l
1 ,1,1-Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl Chloride |ig/l
Xylenes ng/l
MSW LANDFILLS IN NORTHEAST U.S.
AJ1
LCRS

LDS
III
6/94-date
6/94

6.50
597
480
< 10
2
407
384
5
20
144
10
3
13
< 0.5
< 50
2
< 100
< 5
< 5
< 5


< 5
< 10
< 5

< 5
< 10
< 5
6/94

7.30
468
352
< 10
< 1
2
164
14
92
50
10
7
10
1
< 50
6
< 100
< 5
< 5
< 5


< 5
< 10


< 5
< 10
< 5
AK1-2
LCRS

LDS
IV
10/93-date
1 2/93-3/95

6.65
1,592

1,062
< 2
245
711
94
45
387
51
64
< 3
< 19
< 27
< 48
< 50
< 5
46
< 2
< 1
< 1
< 7
603
134
< 1
95
< 11
< 30
1 2/93-3/95

7.20
679

13

4
331
4
25
116
29
5
< 1
< 3
< 2
< 5
< 50
< 1
< 1
< 1
< 1
< 1
< 1
< 5
3
< 1
< 1
< 5
< 3
AL1
LCRS

LDS
IV
1990-date
6/91-5/95

8.09
2,707
2,892
860
1,134
245
261
430
219
150
98
236
4
< 11
< 64
< 36
57
< 6
< 8
< 6
< 7
< 2
< 12
245
< 8
< 12
78
< 11
< 96
12/89-5/95

7.04
2,449
2,482
< 11
< 2
3
199
151
1,028
465
121
38
< 5
< 7
< 10
< 17
< 35
< 2
< 7
< 2
< 3
< 2
< 2
< 77
< 4
< 2
< 7
< 5
< 5
A01-2
LCRS

LDS1
IV
LDS2
IV
1/92-date
8/92-6/95

7.30
6,592
2,178
618

414
1,756
862
54
275
146
786
45
< 9
54
< 32
108
7
15
< 5
< 6
< 4
11
75
< 11
< 4
167
< 12
34
8/92-5/95

7.17
1,132
690
142

43
556
38
44
205
66
19
7
< 1
< 1
< 1
< 50
< 1
< 3
< 1
< 1
< 1
< 1
< 5
< 1
< 1
< 86
< 7
< 3
8/92-6/95

6.72
1,118
722
497

8
558
24
91
196
55
8
2
< 1
< 1
< 1
< 50
< 1
< 1
< 1
< 1
< 1
< 1
< 5
< 1
< 1
< 1
< 5
< 3
g
-------
Table E-3.7. Select LCRS and LDS Flow Chemistry Data for Landfills (Continued).

Cell No.
System
Primary Liner\LDS Type
Waste Placement Dates
Liquid Sampling Dates
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1 ,1-Dichloroethane |ig/l
1 ,2-Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methylene chloride |ig/l
1 ,1,1-Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl Chloride |ig/l
Xylenes ng/l
MSW LANDFILLS IN NORTHEAST U.S.
AR1
LCRS

LDS
IV
3/92-date
11/92-8/94

6.92
5,650
3,923
1,238
290
333
2,075
1,625
380
230
153
850
13
< 6
62
< 13
< 20
< 100
< 100
< 100
< 100
< 100
< 100
< 100
< 100
< 100
< 100
< 100
< 100
11/92-8/94

7.00
1,368
848
22
5
5
325
35
325
186
119
26
< 8
< 7
< 10
< 9
< 100
< 0.5
< 2
< 2
< 50
< 2
< 2
< 2
< 2
< 2
< 2
< 4
< 4
AX1-16
LCRS

LDS
V
7/88-date
9/88-1 2/93

6.47
6,046
3,751
4,149
2,825
2,016
2,828
493
104
602

366
< 18
< 6
35

27
< 32
73
< 34
< 53
< 32
< 46
651
< 60
< 29
419
< 65
72
7/88-3/95

6.79
2,364



67






185
















AZ1
LCRS

LDS
VI
12/92-date
9/93-3/95

6.14
4,810
5,200

2,890
1,009
1,445
617
< 7
660
106
328
< 12
< 15
< 37
< 3
< 132
< 80
170
< 80
< 120
< 110
< 90
4,150
270
< 90
275
< 300
< 170
6/93-3/95

6.60
5,235
3,570
154

55
1,183
57
1,065
133
39
1,102
235
2
< 8
8
< 40
< 2
11
< 3
< 4
< 4
< 3
29
< 9
< 3
< 3
< 8
< 4
Unspecified'-3'
LCRS/LDS

LCRS/LDS

LDS
III
5/92-date
5/92-4/94

7.80
2,946
1,466
666
18
191
1,106
319
94
197
57
250
13
< 2
10
3
40
8
< 5
< 5
< 5
< 5
5
< 5
< 5
< 5
9
< 5

4/94

7.52
2,500
1,328
164
35
40
805
310
280
101
43
262

< 2

2













4/94

7.90
3,160
1,627
145
> 39
41
1,640
433
35
172
56
348

7

4









7



-------
Table E-3.7. Select LCRS and LDS Flow Chemistry Data for Landfills (Continued).

Cell No.
System
Primary Liner\LDS Type
Waste Placement Dates
Liquid Sampling Dates
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1 ,1-Dichloroethane |ig/l
1 ,2-Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methylene chloride |ig/l
1 ,1,1-Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl Chloride |ig/l
Xylenes ng/l
MSW LANDFILLS IN NORTHEAST U.S.
Unspecified'1'
LCRS

LDS
I
3/92-date
11/92-8/94

6.62
4,290
1,200
6,800

160

342
< 34
520
100
300

1
6
36

19
32
< 1
< 3
< 1
17
< 118
< 48
2
375
5
52
11/92-8/94

7.21
1,090
800
56

22
505
19
35
187
35
22

< 0.2
< 1
< 1

< 1
< 1
< 1
< 1
< 1
< 1
< 2
< 1
< 1
< 1
< 1
< 3
Unspecified'1'
LCRS

LDS
IV
3/92-date
11/92-8/94



3,100




390
< 34
470







17
37
1
8
< 1
28
10
< 1
3
740
3
97
11/92-8/94

6.85
302
400
< 3

3
225
4
28
69
23
17

< 0.2
< 1
< 1

< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 3
Unspecified'1'
LCRS

LDS
VI
6/94-date
1/95



1,310
49
< 3
30
482
206
79
133
39
102
< 5
3
49
< 3
< 27
< 10
2
< 10
< 10
< 10
< 10
7
< 10
2
< 10
< 10
< 10
1/95



976
30
< 3
6
783
11
146
222
78
37
< 5
< 2
7
< 3
< 27
< 10
4
< 10
< 10
< 10
< 10
6
< 10
5
< 10
< 10
< 10
Unspecified'1'
LCRS

LDS
III
4/93-date
10/93

6.27
2,420
2,300
85
15
24
520
15
1,100
243
35
19
< 10
< 1
< 25
< 50
< 100
< 7
18
< 5


< 5
260
27
< 5
51
< 10
17
10/93


2,100

< 10
< 4
< 1
480
38
1,100
433
124
43
< 10
< 1
< 25
< 50
115
< 0.7
< 1
< 0.5

< 5
< 2
< 5
< 5
< 1
< 2
< 2

-------
Table E-3.7. Select LCRS and LDS Flow Chemistry Data for Landfills (Continued).

Cell No.
System
Primary Liner\LDS Type
Waste Placement Dates
Liquid Sampling Dates
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1 ,1-Dichloroethane |ig/l
1 ,2-Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methylene chloride |ig/l
1 ,1,1-Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl Chloride |ig/l
Xylenes ng/l
MSW LANDFILLS IN NORTHEAST U.S.
Unspecified11'
LCRS/LDS

LCRS

LDS
III
4/88-date
3/92-1 2/93
7.20
3,438
2,740
3,573
1,957
947
1,508
199
95
261
115
282
9
< 1
< 50
1
< 50












12/93
7.08

2,005
5,880
3,340
1,760
1,580
235
170
397
178
487
< 100
< 50
80
< 1
< 50












12/93
8.26
121
94
< 2

24
90
< 1
64
132
27
16

< 50

22













Unspecified1'
LCRS/LDS

11/92-date
11/93-2/94
6.70
2,910



45
203
19
750
508
62
115
< 2
11
32
1
202
< 10
54
< 10


29
200
96

290
< 10
144
Unspecified11
LCRS/LDS

1/85-date
1/85-11/93
6.73
8,388
6,858
2,935
2,985
1,962
2,324
1,237
133
331
154
875
6
< 7
< 5
27
68
< 3
< 11



< 7
< 12

< 62


< 20
Unspecified1'
LCRS/LDS

12/92-date
1/93-10/93


4,900
4,833
> 4,700
1,867
2,567
717
26
480
89
260
7


6
70

84



65
560


515

210
-------
Table E-3.7. Select LCRS and LDS Flow Chemistry Data for Landfills (Continued).

Cell No.
System
Primary Liner\LDS Type
Waste Placement Dates
Liquid Sampling Dates
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1 ,1-Dichloroethane |ig/l
1 ,2-Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methylene chloride |ig/l
1 ,1,1-Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl Chloride |ig/l
Xylenes ng/l
MSW LANDFILLS IN SOUTHEAST U.S.
C1-6
LCRS/LDS
I
5/90-date
4/91-12/94

6.81
5,339
3,302
1,119
986
401
2,426
1,298
61
263
49

7
< 5
< 74
37
< 64
11
< 10
< 5
< 1
< 5
56
< 123
< 9
< 94
< 123
< 7
81
V1-4
LCRS
II
12/89-date
12/89-11/92

6.51
8,983
8,640
804
595


2,263
69


333
17

21

27












W1-2
LCRS/LDS

LDS
III
5/92-date
8/92,2/93

5.90
4,100










< 10
< 5
< 24
< 5

< 25
260
< 25


49
260
35
< 25
230
< 25
160
8/92

5.90
340










< 10
< 5
< 10
9

< 5
150
< 5


< 5
102
58
< 5
20
< 5
< 5
AW1-4
LCRS/LDS

5/93-date
5/93-4/95

6.40
1,062
893
542
361
116
459
49


12
7
< 10
< 7
< 50
< 6
< 50
< 16
< 16
< 14

< 1
26
43
< 19
< 24
12
< 18
< 33
MSW LANDFILLS IN WEST U.S.
AM 1-2
LCRS1

LDS1
IV
LCRS2

LDS2
IV
10/90-2/91
4/91-2/95

6.62
2,451
1,709
396

108
1,204
183
18
376
95
92
236
< 20
< 30
< 2.0
< 40
17
172
6
324
< 5
57
< 6
< 21
3
267
< 21
122
2/92-1 2/93

6.90
17,250
14,330
184

116
407
2,725
1,500
1,700
283
1,348
< 2
< 20
< 30

< 40
< 1
< 3
< 1
< 1
< 1
< 1
3
< 1
< 1
< 1
< 1
< 1
4/91-1/95

6.60
2,795
1,413
65

22
867
332
9
319
74
88
235
< 20
< 57
< 7
< 91
13
136
< 11
547
< 8
29
< 20
< 16
< 12
146
< 11
71
2/92-2/95

7.26
17,063
14,363
76

28
163
2,659
1,425
1,819
299
1,929
< 10
< 10
< 30
3
< 40
< 1
< 2
< 1
< 1
< 1
< 1
< 2
< 1
< 1
< 1
< 2
< 2
-------
Table E-3.7. Select LCRS and LDS Flow Chemistry Data for Landfills (Continued).

Cell No.
System
Primary Liner\LDS Type
Waste Placement Dates
Liquid Sampling Dates
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1 ,1-Dichloroethane |ig/l
1 ,2-Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methylene chloride |ig/l
1 ,1,1-Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl Chloride |ig/l
Xylenes ng/l
MSW ASH LANDFILLS
M1
LCRS

LDS
I
9/91 -date
4/93-7/94
7.20
43,383
46,733









5
< 2
< 1
< 6













4/93-7/94
8.00
2,620
2,478









< 1
< 2
1
4













P1-2
LCRS
II
1991 -date
7/91-10/93
6.76
17,506
32,638









< 13
< 5
< 31
< 29













S2
LCRS

8/90-date
7/91-4/94
6.54

8,747
304
84


10,516
246


684





< 5
< 33
< 5

< 5
< 7
< 6
< 16
< 5
< 25
< 10

Y1
LCRS

LDS
III
1/89-date
1/89-4/94
7.34
16,709
10,773
355
61
39
160
5,848
146
1,332

1,994
5
< 7
< 12
3
< 40












1/89-4/94
7.34
1,806
2,628
32
8
4
8
1,580
148
262
54
244
< 8
< 6
< 8
7
18












AF1
LCRS

LDS
III
1/90-date
7/90-4/95
7.44
10,732
6,067
413
60
109
2,500
2,940
85
96
113
1,460
6
< 6
< 43
24
48
< 3
< 3
< 3
< 3
< 3
< 3
< 3
< 3
< 3
< 3
< 3
< 3
8/91-4/95
7.34
3,108
672
14
8
3
740
50
124
154
44
25
< 5
< 5
< 42
< 6
< 28
< 3
< 3
< 3
< 3
< 3
< 3
< 3
< 3
< 3
< 3
< 3
< 5
-------
Table E-3.7. Select LCRS and LDS Flow Chemistry Data for Landfills (Continued).

Cell No.
System
Primary Liner\LDS Type
Waste Placement Dates
Liquid Sampling Dates
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1 ,1-Dichloroethane |ig/l
1 ,2-Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methylene chloride |ig/l
1 ,1,1-Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl Chloride |ig/l
Xylenes ng/l
MSW ASH LANDFILLS
AN 1-4
LCRS/LDS

6/91-10/94
3/92-1 2/93



42,000
5,607
15
40
99
22,400
496
1,271
420
888
< 10
49
< 10
74
46
< 0.5
< 0.5
< 0.5
< 0.5
< 0.5
< 2
< 0.5
< 0.5
< 0.5
< 0.8
< 0.5
< 0.5
Unspecified11'
LCRS

LDS
III
10/90-date
7/91-10/92







5,010

3,430


880
17
< 5
84
3
< 24












7/91-7/93







2,130

1,380


105
< 20
< 5
14
< 9
< 20












COAL ASH LANDFILLS
Z1
LCRS

LDS
III
3/92-date
3/92-8/93

7.66
1,144
1,098
11
< 3
6
160
21
587
190
30
46
< 9
< 9
22
< 34
38
< 4
< 4
< 4

< 1
< 3
< 4
< 4
< 4
< 2
< 7
< 4
3/92-8/93

7.85
381
495
17
6
5
171
10
195
89
22
11
< 10
< 9
< 11
< 3
< 40












Unspecified11'
LCRS1

LCRS2

10/89-date
2/93-11/94

7.74
623
347



220

178

15

62
< 5
< 9
< 4













2/93-11/94

7.64
2,894
2,563



224

1,837

133

8
< 10
< 20
< 5













-------
Table E-3.7. Select LCRS and LDS Flow Chemistry Data for Landfills (Continued).

Cell No.
System
Primary Liner\LDS Type
Waste Placement Dates
Liquid Sampling Dates
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1 ,1-Dichloroethane |ig/l
1 ,2-Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methylene chloride |ig/l
1 ,1,1-Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl Chloride |ig/l
Xylenes ng/l
C&DW LANDFILLS
A1
LCRS

LDS
I
10/92-date
3/93-1 2/93


2,880
1,139

443

690
463
203
202
324

< 5

< 10













9/93-1 2/93


216
37
269
30

40
33
36
13
20

< 5
< 4
< 10
< 14












AM -2
LCRS1

LDS1
III
LCRS2

LDS2
III
9/89-date
9/91-5/94
6.58
5,275
4,125
3,293
845
1,055
2,300
655
35
335

333
13
< 1
37
3
< 100
18
92
2


75
382
44
10
561
11
251
9/91-5/94
6.40
800
803
< 65
4
8
238
42
113
32

104
< 2
< 1
< 11
3
< 100
< 1
42
< 1


< 17
36
24
< 1
9
< 1
231
9/91-5/94
6.28
4,355
4,325
4,083
1,407
1,415
2,600
687
61
428

235
16
< 1
40
3
< 12
17
108
3


57
452
58
< 11
665
6
168
9/91-5/94
6.20
830
515
123
20
21
536
36
64
34

78
< 2
< 1
< 17
5
< 100
4
66
< 1


32
252
39
< 1
26
3
480
-------
Table E-3.7. Select LCRS and LDS Flow Chemistry Data for Landfills (Continued).

Cell No.
System
Primary Liner\LDS Type
Waste Placement Dates
Liquid Sampling Dates
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1 ,1-Dichloroethane |ig/l
1 ,2-Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methylene chloride |ig/l
1 ,1,1-Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl Chloride |ig/l
Xylenes ng/l
HW LANDFILLS
AQ1&10
LCRS1

LDS1
IV
3/86-early 1990
2/91-11/93

7.90
23,200










189
< 5
22
24
1,190
< 8
14
< 5

18
< 5
5
9
60
< 5
13

2/91-11/93

7.36
2,582










< 10
< 5
12
31
< 40
< 4
< 5
< 3

< 2
< 5
6
< 8
19
7
< 10

LCRS10

LDS10
IV
Iate88-mid 1991
2/91-11/93

7.20
5,574















< 6
135
4

< 10
< 5
< 3
36
5
< 13
< 7

2/91-11/93

7.51
4,154










< 11
< 5
50
< 17
84
< 4
< 5
< 3

< 2
< 5
< 3
< 4
< 2
< 5
< 10

-------
Table E-3.7. Select LCRS and LDS Flow Chemistry Data for Landfills (Continued).

Cell No.
System
Primary Liner\LDS Type
Waste Placement Dates
Liquid Sampling Dates
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1 ,1-Dichloroethane |ig/l
1 ,2-Dichloroethane |ig/l
cis-1 ,2-Dichloroethylene |ig/l
trans-1 ,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methylene chloride |ig/l
1 ,1,1-Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl Chloride |ig/l
Xylenes ng/l
HW LANDFILLS
L1
LCRS
LDS
base III sides I
6/90-date
1/91-1/95

7.60
12,302



7

3,783
704


2,514
30


55
285

14



8
32
8

12

9
1/91

7.70
3,477



5

490
1,071


353











7





AD1&7
LCRS1

LDS1
III
5/85-late 1989
6/85-1 2/94

8.72
40,159



2,006

12,709
4,428


10,394
125,252
< 410
294
192

< 184
< 138
< 67

< 48
174
< 292
< 228
< 53
< 189
1,371

6/85-1 2/94

7.40
1,755



8

207
404


177
< 13
< 4
29
< 29

< 4
< 5
< 3

< 2
< 6
< 4
< 4
< 2
< 5
< 9

LCRS7

LDS7
IV
9/87-9/88, 5/93-6/93
9/87-1 2/94

9.99
39,036



4,471

10,759
6,105


5,549
34,572
< 56
158
< 306

555
< 604
< 2,181

< 238
< 850
< 602
< 466
< 240
1,042
< 7,438

9/87-1 2/94

7.45
2,512



7

301
6,105


250
< 14
< 3
< 12
< 16

< 4
< 6
< 3

< 3
< 6
< 3
< 3
< 25
< 5
< 9

AS1
LCRS

LDS
IV
7/89-10/91
12/89-2/95


















18
32
< 17


< 8
254
< 20
< 50
< 56
< 10
18
12/89-3/95




37













< 5
< 5
< 5


< 5
4
< 5
< 5
< 5
< 10
< 5
Notes: (1)" " = not analyzed; ND = all measurements reported as non-detect;  < = more than 50% of measurements reported as non-detec
     (2) Concentrations are arithmetic averages. Parameters reported as non-detect were taken as one-half the detection limit in calculating the avera
      (3) Unspecified landfills were not given a code name due to too few flow rate data to analyze.
-------
Table E-3.8(a).
Distribution  of LCRS/LDS Flow  Rate Data by Waste Type and
Geographic Region.

Waste Type
MSW
HW
MSW Ash
Coal Ash
C&DW
Geographic Region
NE
24 landfills
71 cells
5 landfills
26 cells
5 landfills
12 cells
1 landfill
1 cell
2 landfills
4 cells
SE
8 landfills
26 cells
5 landfills
31 cells
2 landfills
4 cells
-
-
W
1 landfill
2 cells
3 landfills
10 cells
-
-
-
Table E-3.8(b). Distribution of LCRS/LDS Flow Rate Data by Primary Liner and LDS
              Types.

Primary Liner Type
GM
GM/GCL Composite
GM/CCL or
GM/GCL/CCL
Composite
LDS Type
Sand or Gravel
13 landfills
41 cells
3 landfills
19 cells
13 landfills
31 cells
GN
11 landfills
28 cells
4 landfills
9 cells
16 landfills
57 cells
Table E-3.8(c). Distribution of LCRS Chemistry Data by Waste Type and Start of
              Operation Date.
Waste Type
MSW
HW
MSW Ash
Coal Ash
CADW
Pre-1990
Start of Operation
11 landfills
13 cells
3 landfills
5 cells
1 landfill
1 cell
1 landfill
1 cell
1 landfill
2 cells
Post-1990
Start of Operation
25 landfills
28 cells
1 landfill
1 cell
6 landfills
6 cells
1 landfill
1 cell
1 landfill
1 cell
                                   E-114
-------
From Table E-3.8(b), most of the cells at most of the landfills have either a GM primary
liner (37% of all cells) or a GM/CCL or GM/GCL/CCL primary liner (48%).  Fewer cells
(15%) have a GM/GCL primary liner.  About 48% of the cells have a sand or gravel LDS
and 52% have a GN LDS. Based on  the distribution of the data, the database appears
to be representative of typical double-liner system designs in landfills.

Most of the liquids management data are for open cells; only about 23% of the cells in
the database  had received a final cover system.

E-3.3    LCRS and LDS Flow Rate Data

All of the landfill cells in  the database were operated with a strategy of active liquid
removal from  both the LCRS and LDS. By federal regulation, MSW, MSWash, and HW
landfills must  generally limit the hydraulic head buildup in the LCRS to less than 0.3 m.
This is accomplished through design  of the LCRS drainage layer with adequate slope
and hydraulic conductivity, adequate collection pipe or swale spacing, and regular liquid
removal from  the LCRS sump.  LCRS and LDS flow rate data were reported on either a
daily, weekly,  or periodic basis, depending on facility. Using this source data, average
daily flow rates were calculated for both systems on a monthly basis by dividing the total
amount of liquid extracted from the systems during a month by the number of days in
the month and the landfill cell areas.  The volumes of flow used in the calculations were
obtained from landfill operations records, with flow measurements most often obtained
using accumulating mechanical flow meters. The reported flow rates should be
considered approximate.

Peak and average monthly LCRS and LDS flow rate data are summarized in Tables E-
3.5 and E-3.6, respectively. The data are separated into the three operational stages
described in Sections E-1.3.7 and E-1.3.8.  Both LCRS and LDS flow rate data are
available for 170 of the 187 cells.  The data set is not complete for the remaining 17
cells: 16 cells do not have available LCRS flow rate  data, and one cell does not have
available LDS flow rate data.

E-3.4    Landfill Chemistry Data

Leachate chemistry data are available for 59 cells at 50 landfills: 48 cells at 39  landfills
from the previously described database (including two cells from one landfill with
different waste types) and eleven cells from  eleven "unspecified" landfills with too little
flow  rate data to be given a landfill designation in the database (note that these
"undesignated" landfills  are in addition to the 54 "designated" landfills in  the database).
For the purposes of the  leachate chemistry evaluation conducted in Section 6 of this
appendix, the data are grouped as shown in Table E-3.8(c).

The MSW leachate chemistry data are available for 36 landfills located in all geographic
regions of the U.S.  This leachate chemistry data are believed to be representative of
modern MSW landfills in the U.S. operated without leachate recirculation or other
                                  E-115
-------
special activities (e.g., special waste disposal, induced aerobic degradation).  While the
leachate chemistry data from modern MSW landfills are extensive, they should not be
considered to reflect the full range of leachate chemistry associated with the anaerobic
decomposition process, from the acid stage to the methane fermentation stage.
Moreover, differences will exist from facility to facility based on a variety of climate, site,
waste, and operational factors. Additional data are needed from more facilities over a
longer time period to better identify the potential range of leachate chemistry
characteristics throughout the initial, active, and post-closure operational periods of a
facility.

Fewer leachate chemistry data are available for HW and ISW landfills.  In addition, the
types of wastes placed in HW and ISW landfills are generally more variable between
landfills than wastes placed in MSW landfills.  With the exception of the leachate
chemistry data for MSW ash landfills,  it is likely that the data presented in this appendix
do not fully characterize the variation in leachate chemistry for HWand ISW landfills.
The chemistry data for MSW ash landfill leachate may be representative of modern
MSW ash landfills in the U.S. because seven landfills are included in the database and
the chemistry of MSW ash is less variable than HW.

Select data from the leachate chemistry database are presented in Table E-3.7. The
table summarizes average values for 30 representative chemical parameters: water
quality indicator parameters (e.g., pH, specific conductance, TDS,  etc.); major inorganic
cations and anions (e.g., calcium, chloride, sulfate, etc.); trace metals (e.g., arsenic,
chromium, lead, etc.);  and VOCs (e.g., benzene, methylene chloride, trichloroethylene,
etc.).  The specific trace metals and VOCs were chosen for study because these metals
and VOCs are sometimes found in leachates from MSW, HW, and ISW landfills.  They
were also selected based on availability of parameters between landfills, frequency of
detection, and concentration.  It is recognized that the leachate chemistry database is
limited in terms of completeness and duration of monitoring.  In addition, key MSW and
HW leachate constituents,  such as alcohols and ketones, are poorly represented in the
database, and, thus, could not be included  in the list of select parameters.  It is
important that these additional data be collected so that our understanding of leachate
chemistry can continue to improve. For example, from the literature review in Section
2.3, ketones are found in high concentrations in MSW and  HW leachates.  However, the
majority of leachate samples in the database were not analyzed  for ketones.  Because
of this, there is little benefit in including ketones in the study.  The chemical data
presented herein are  intended to be representative, not comprehensive. The data
should not be considered complete for purposes of evaluating potential human health or
ecological impacts.

For all landfills in this study, the chemical data were reportedly obtained using sampling
and analysis procedures in accordance with EPA protocols (i.e., SW-846 (EPA, 1987b)).
EPA protocols contain quality control (QC) standards for both sampling and analysis,
including the use of method blanks, matrix spikes, and duplicates.  These protocols will
provide accurate analytical data for samples obtained from the LCRS and LDS sumps.
                                   E-116
-------
However, the potential for VOC volatilization from the sump liquids prior to sampling has
not been evaluated. Some of the chemical data reports obtained for the landfills only
contained a list of detected chemicals.  It was not known for which chemicals the
samples were analyzed.

The parameter values given in Table E-3.7 are arithmetic averages of the  data for a
given sampling point.  The arithmetic mean, rather than the geometric mean, was used
because more of the data in the literature are based on the arithmetic mean and
statistical tests indicated that neither mean was more appropriate. In calculating the
average value for each parameter, one-half of the given detection limit was
conservatively used for all results reported as non-detect.  If more than one-half of the
measurements for a parameter were reported as non-detect, the calculated  average
value given in Table E-3.7 is proceeded by a "<" symbol. If all of the measurements
were reported as non-detect, "ND" is given for the parameter value.  As with the flow
rate data, the chemical constituent data were obtained from landfill operations records.

E-4    Leakage Rates Through Primary Liners

E-4.1    Overview

The performance of primary liners at double-lined landfills is first assessed in this
Subsection in terms of primary liner leakage using the methodology of Gross et al.
(1990) described in Section E-2.1.1.  Briefly, this method requires the comparison of
LDS and LCRS flow rate data to quantify that portion of LDS flow attributable to primary
liner leakage as opposed to other sources.  The relative performances of the different
types of primary liners are then evaluated using the "apparent" liner hydraulic efficiency,
Ea, introduced by Bonaparte et al. (1996).  If the only source of flow into the  LDS is
primary liner leakage, then the "apparent" liner hydraulic efficiency is the "true" liner
hydraulic efficiency,  Et. The true efficiency of a liner is not constant but rather a function
of the hydraulic head in the LCRS and size of the area over which LCRS flow is
occurring (the area is larger at high flow rates compared to low flow rates). The true
efficiency of a liner is also a function of design: identical liners overlain by  different
LCRSs or placed on different slopes will exhibit different Et values. Also, the efficiency
of a liner for a given set of hydraulic conditions could change over time if the physical
condition of the liner changes. For example, long-term time dependent changes in GMs
could result from chemical degradation or brittle stress cracking under certain
conditions.  Time dependent changes in CCLs or GCLs can result from chemical
degradation, consolidation, or other factors. Notwithstanding all of these limitations, the
hydraulic efficiency concept is useful in characterizing liner hydraulic performance.

The methodology described above was used to  evaluate the hydraulic performance of
GM primary liners and GM/GCL composite primary  liners. Chemical constituent data
were not utilized  in the evaluation of these types of liners because the initial  hydraulic
assessment (i.e., comparing LCRS and LDS flow rates) yielded significant insight into
these liners' true hydraulic efficiencies.  However, the situation was found  to be more
                                  E-117
-------
complicated for GM/CCL and GM/GCL/CCL composite primary liners due the
generation of consolidation water by these liners not only during the initial period of
operation, but also during the active and post-closure periods. The performance
evaluation of these liners included the additional step of comparing the concentrations
of select chemical constituents in LDS liquids to concentrations of the same constituents
in LCRS liquids.  In particular, general water quality characteristics (i.e., major ion, COD,
BOD, and TOC concentrations) of the LCRS and LDS liquids were compared to assess
whether the liquids had different primary sources (e.g., leachate for LCRS liquids and
CCL pore water for LDS liquids). The concentrations of five key chemical constituents
(i.e., the inorganic anions sulfate and chloride and the aromatic hydrocarbons benzene,
toluene, and xylene) in the  LCRS and LDS flows were compared in more detail to
further assess whether primary liner leakage had contributed to LDS flows.

It is noted that the presence of chemical constituents in the LDS was evaluated
empirically.  Therefore, the  concentrations of chemicals collected in the LDS were
directly compared to concentrations of the same chemicals collected in the LCRS. No
fate and transport analysis  was performed that accounts for attenuation of the LCRS
chemicals migrating through the primary liner CCL. However, to overcome the need to
perform such an analysis, the five key chemical constituents evaluated were selected
based on their  high solubility in water, low octanol-water coefficient, high resistance to
hydrolization, and high resistance to anaerobic biodegradation in soil.

E-4.2    Leakage Rates Through GM Primary Liners

E-4.2.1   Description of Data

The performance of 31 of the 69 cells with GM primary liners are assessed in this
section. The remaining 38  cells with GM primary liners were excluded from the
assessment primarily because they do not have continuous LCRS and LDS flow rate
data available for an individual cell from the start of operation and for a significant
monitoring period. Cell D2  was excluded because near liquid wastes was disposed in
the cell at different times during active filling of the cell. Flow rate data are available for
the considered 31 cells at 14 landfills with monitoring periods of up to 114 months. Data
from 17 of the 31 cells in  this study were included in the previous EPA study by
Bonaparte and Gross (1993); however,  this study contains additional data for most of
these cells.  Descriptions of the liner systems installed at these landfills are presented in
Table E-4.1. Twenty-five cells have a 1.5- or 2-mm thick HOPE GM primary liner, and
the remaining six cells have a 0.9-mm thick chlorosulfonated polyethylene (CSPE) GM
primary liner. The LDS consists of a sand layer, a GN, or both. The secondary liner of
the double-liner system is a single GM or a GM/CCL composite liner.  Formal CQA
programs were used in the  construction of the liner system for 23 cells, while eight cells
were constructed without formal CQA.  It is noted that the six cells with CSPE GM
primary liners were all constructed without CQA, while only two of the 25 cells with
HOPE GM primary liners were constructed without CQA. After the end of their active
operation stages, six cells at three landfills received final cover systems with GM or
                                  E-118
-------
          Table E-4.1. Description of Liner System Components for Considered Landfill Cells with GM Primary Liners.
Cell
No.
B1-2
C1-5
D1,3,&4
E1-4
F1
G1-2
11-3
K1
N2
O1 -2
S1
S2
V1-2
W1-2
X1
Type(1)
of
Waste
MSW
MSW
HW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
ASH
MSW
MSW
MSW
LCRS
Material'2'
S
S
S
S
S
S
S
S
S
G/S
S/GN
S/GN
S/GN
S/GN
S
Thickness'3'
(mm)
450
600
300
600
600
600
600
600
600
300/300
600/5
600/5
600/5
600/5
600
GM(4)
Primary Liner
Type (and
Thickness (mm))
CSPE(0.9)
HDPE(2.0)
HDPE(2.0)
CSPE(0.9)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
HDPE(2.0)
HDPE(1.5)
HDPE(2.0)
HDPE(2.0)
HDPE(2.0)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
IDS
Material'2'
S
S
S
S
S
S
S
S/GN
GN
S/GN
GN
GN
S/GN
GN
GN
Thickness
(mm)
450
450
300
600
300
300
450
300/5
5
300/5
5
5
300/5
5
5
Secondary Liner
Geomembrane
Type (and
Thickness (mm))
PVC(0.8)
HDPE(2.0)
HDPE(1.0)
PVC(0.8)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
LLDPE(1.5)
HDPE(1.5)
HDPE(1.0)
HDPE(2.0)
HDPE(2.0)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
Lower Component
Material'5'
NA(6)
CCL
CCL
CCL
NA
CCL
CCL
CCL
CCL
CCL
CCL
CCL
CCL
CCL
CCL
Thickness
(mm)
NA
300
900
600
NA
600
300
600
300
150
600
600
150
150
150
3rd Party
CQA
Program
(Yes/No?)
No
Yes
Yes
No
No
Yes
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
m

CD
          Notes:  (1) Waste types:  MSW = municipal solid waste; HW= hazardous waste; ASH = MSW ash.
                 (2)  LCRS and LDS material types: GN = geonet or geocomposite; S =sand; G = gravel.
                 (3)  All material thicknesses are nominal values.
                 (4)  GM Types: HOPE = high density polyethylene; CSPE = chlorosulfonated polyethylene; PVC = polyvinyl chloride.
                    LLDPE = linear low density polyethylene.
                 (5)  liner lower component material types: CCL = compacted clay liner.
                 (6)  NA = not applicable
-------
GM/CCL barriers. Table E-4.2 contains a summary of the LCRS and LDS flow rate
database for the cells with GM primary liners.

E-4.2.2   Analysis of Data

E-4.2.2.1 Interpretation of Data

The interpretation of LDS flow rate data for cells with GM primary liners is relatively
straightforward because consolidation water is not a source of LDS flow and, for the
facilities included in this study, ground-water infiltration is not occurring. Thus, the only
potential sources of LDS flow are construction water, compression water, and primary
liner leakage. For cells with GN  LDSs, LDS flow should be primarily due to primary liner
leakage. For cells with sand LDSs, construction and compression water may be
significant sources of LDS flow for a time period of up to one year after cell construction
(Gross et al., 1990).  Because of this, cells with sand LDSs generally remain in the initial
period of operation longer than cells with GN LDSs.

E-4.2.2.2 Summary of Flow Rate Data

LCRS and LDS flow rate data for the 31 cells with GM primary liners are presented in
Table E-4.2. Average and peak monthly flow rates are reported for the three landfill
operational time periods defined  previously. For cells with long periods of active
operation and post-closure, data are summarized in approximately twelve-month
increments to facilitate the evaluation of temporal changes in flow rate.

From Table E-4.2, average monthly LCRS flow rates ranged from about 1,500 to 43,700
Iphd during the initial period of operation, from about 100 to 16,200 Iphd during the
active period of operation, and from about 320 to 1,300 Iphd after closure. Average
LDS flow rates for a cell ranged from about 5 to 2,100 Iphd during the initial period of
operation, 1 to 1,600 Iphd during the active period of operation, and 2 to 330 Iphd after
closure.  A review of the data in Table  E-4.2 indicates that peak monthly LCRS and LDS
flow rates are typically two to four times the average monthly values for individual
monitoring periods. This difference becomes larger in the LDS when the flow rates
become very low: this is mostly an  artifact of the LDS pumping schedule, which
becomes infrequent when the flow  rate becomes very low. Table E-4.2 shows that
between the initial and active periods of operation, LCRS flow rates typically decreased
up to about one order of magnitude and LDS flow rates decreased up to about one to
two orders of magnitude.  LCRS  and LDS flow rates continued to decrease after cell
closure: for the six cells that were closed, flow rates after closure decreased by up to
one order of magnitude compared to flow rates prior to closure.

E-4.2.2.3 Effects of LDS Material and CQA on LDS Flow Rates

Table E-4.3 and Figure E-4.1 present summaries of average LDS flow rates for the 31
cells with GM primary liners during each of the three landfill operational stages. Table
E-4.3 has been subdivided to separately report LDS flow rates based on two criteria:
                                  E-120
-------
       Table E-4.2. Summary of LCRS and LDS Flow Rate Data for Considered Landfill Cells with GM Primary Liners.
Cell
No.



B1




B2


C1



C2


C3


C4


C5
D1



D3

D4
E1


Cell
Area


(ha)
3.3




3.5


3.2



3.7


3.6


3.7


2.6
0.4



0.3

0.4
2.4


Start of
Waste
Placem.
(month-
year)
5-84




5-84


5-90



4-91


8-91


2-92


11-92
10-85



7-87

1-89
3-88



Final
Closure
(month-
year)
11-88




11-88


NA(5)



NA


NA


NA


NA
5-86



NA

NA
NA


Initial Period of Operation'1'
Time
Period

(months)
1-19




1-19


1-9



1-12


1-8


1-4


1-12
1-7



1-12

1-11
1-7


LCRS Flow(4)
Avg.

(Iphd)
ND(5)




ND


ND



1,475


3,417


14,828


6,419
ND



20,292

31,281
ND


Peak

(Iphd)
ND




ND


ND



2,585


9,558


41,331


12,528
ND



51,265

120,527
ND


LDS Flow
Avg.

(Iphd)
ND




ND


ND



92


63


178


23
32



12

233
2,144


Peak

(Iphd)
ND




ND


ND



398


268


265


40
80



56

801
5,026


Ea


(%)












93.74


98.16


98.80


99.64




99.94

99.25



Active Period of Operation'2'
Time
Period

(months)
20-31
32-43
44-54


20-31
32-43
44-54
10-21
22-33
34-45
46-56
13-24
25-36
37-45
9-20
21-32
33-41
5-16
17-28
29-35
13-26
NA



13-24
25-28
NA
8-19
20-31
32-40
LCRS Flow
Avg.

(Iphd)
2,245
5,223
3,975


2,732
3,740
2,337
789
259
159
103
435
300
161
311
314
268
937
438
407
2,513
NA



13,003
1,010
NA
8,432
11,521
6,525
Peak

(Iphd)
5,754
6,845
7,464


5,393
5,707
3,982
1,419
780
286
200
859
610
464
671
752
987
2,055
622
686
10,440
NA



44,895
2,413
NA
19,614
36,164
13,075
LDS Flow
Avg.

(Iphd)
266
424
892


404
996
665
123
89
27
40
9
22
7
2
33
16
70
51
26
28
NA



7
283
NA
1,436
1,051
743
Peak

(Iphd)
499
808
1,426


605
1,690
1,102
304
170
128
227
31
125
14
9
276
103
147
92
29
115
NA



73
341
NA
3,069
1,915
1,015
Ea


(%)
88.14
91.87
77.55


85.20
73.36
71.54
84.40
65.52
83.08
61.27
98.03
92.71
95.40
99.49
89.56
94.02
92.52
88.39
93.71
98.88




99.95
71.97

82.97
90.88
88.61
Post-Closure Period'3'
Time
Period

(months)
55-66
67-78
79-90
91-102
103-114
55-66
67-78

NA



NA


NA


NA


NA
8-19
20-26
27-38
39-50
NA

NA
NA


LCRS Flow
Avg.

(Iphd)
317
703
1,146
1,306
510
493
337

NA



NA


NA


NA


NA
ND
ND
376
715
NA

NA
NA


Peak

(Iphd)
670
1,877
1,956
1,943
718
1,040
654

NA



NA


NA


NA


NA
ND
ND
1,455
1,352
NA

NA
NA


LDS Flow
Avg.

(Iphd)
106
267
279
326
74
154
328

NA



NA


NA


NA


NA
102
1
5
64
NA

NA
NA


Peak

(Iphd)
222
1,134
451
612
97
393
514

NA



NA


NA


NA


NA
886
10
70
156
NA

NA
NA


Ea


(%)
66.48
62.02
75.64
75.01
85.41
68.83
2.80

















98.58
91.05






m
-------
       Table E-4.2. Summary of LCRS and LDS Flow Rate Data for Considered Landfill Cells with GM Primary Liners (Continued).
Cell
No.



E2


E3
E4
F1

G1





G2


I1(6)







|2<6)






Cell
Area


(ha)
2.4


1.2
1.2
1.8

3.0





1.6


3.2/2. 7(7)







4.2/2.3(7)






Start of
Waste
Placem.
(month-
year)
10-87


5-90
7-90
7-92

6-89





6-89


8-87







10-87







Final
Closure
(month-
year)
NA


NA
NA
NA

NA





NA


10-94







10-94






Initial Period of Operation'1'
Time
Period

(months)
1-12


1-12
1-12
1-12

1-12





1-12


1-5
6-8






1-7






LCRS Flow(4)
Avg.

(Iphd)
ND


9,425
20,148
14,472

22,371





22,371


ND
ND






6,627






Peak

(Iphd)
ND


25,394
55,785
45,010

46,120





46,120


ND
ND






13,959






LDS Flow
Avg.

(Iphd)
483


1,595
996
124

ND





197


234
ND






31






Peak

(Iphd)
3,518


1,951
2,362
479

ND





645


508
ND






77






Ea


(%)



83.08
95.06
99.14







99.12










99.53






Active Period of Operation'2'
Time
Period

(months)
13-24
25-36
37-45
13-14
NA
13-24
25-30
13-24
25-36
37-42
43-51
52-63
64-67
13-24
25-36
37-42
9-15
16-32
33-44
45-48
49-54
55-66
67-78
79-84
8-24
25-36
37-40
41-46
47-58
59-70
71-76
LCRS Flow
Avg.

(Iphd)
5,821
4,547
4,434
6,062
NA
9,000
7,826
12,893
3,438
8,356
ND
ND
ND
12,893
3,438
8,356
16,224
ND
7,167
231
ND
624
541
904
ND
1,030
427
ND
624
541
904
Peak

(Iphd)
10,445
11,014
6,830
9,038
NA
25,450
10,932
23,485
1 1 ,652
10,303
ND
ND
ND
23,485
1 1 ,652
10,303
48,932
ND
22,020
332
ND
1,580
752
1,827
ND
3,241
1,054
ND
1,580
752
1,827
LDS Flow
Avg.

(Iphd)
802
685
596
1,603
NA
66
67
ND
156
101
121
74
49
37
35
60
5
ND
10
4
ND
2
13
79
ND
5
6
ND
8
8
5
Peak

(Iphd)
2,447
1,404
999
1,758
NA
83
77
ND
238
116
384
139
64
65
42
100
18
ND
44
10
ND
5
42
157
ND
35
11
ND
37
23
6
Ea


(%)
86.22
84.93
86.56
73.56

99.27
99.15

95.47
98.79



99.71
98.98
99.28
99.97

99.86
98.49

99.68
97.60
91.26

99.52
98.67

98.67
98.54
99.49
Post-Closure Period'3'
Time
Period

(months)
NA


NA
NA
NA

NA





NA


85-93







77-85






LCRS Flow
Avg.

(Iphd)
NA


NA
NA
NA

NA





NA


800







800






Peak

(Iphd)
NA


NA
NA
NA

NA





NA


1,794







1,794






LDS Flow
Avg.

(Iphd)
NA


NA
NA
NA

NA





NA


62







2






Peak

(Iphd)
NA


NA
NA
NA

NA





NA


119







4






Ea


(%)
















92.25







99.71






m
-------
       Table E-4.2. Summary of LCRS and LDS Flow Rate Data for Considered Landfill Cells with GM Primary Liners (Continued).
Cell
No.



,3(6)






K1




N2



01(8)




02(8)



S1



Cell
Area


(ha)
3.4/1 .8(7)






2.7




6.3



4.2




4.9



2.0



Start of
Waste
Placem.
(month-
year)
4-88






12-89




1-92



9-88




3-89



9-90




Final
Closure
(month-
year)
10-94






NA




NA



NA




NA



NA



Initial Period of Operation'1'
Time
Period

(months)
1-7






1-12




1-12



1-6




1-12



1-10



LCRS Flow(4)
Avg.

(Iphd)
1 1 ,559






17,808




ND



ND




4,407



2,226



Peak

(Iphd)
21,081






24,832




ND



ND




9,826



5,081



LDS Flow
Avg.

(Iphd)
37






122




ND



293




6



12



Peak

(Iphd)
87






163




ND



620




24



39



Ea


(%)
99.68






99.31













99.86



99.45



Active Period of Operation'2'
Time
Period

(months)
8-24
25-36
37-40
41-46
47-58
59-70
71-76
13-24
25-36
37-48
49-60
61-66
13-19
20-31
32-34
35-39
7-18
19-30
31-42
43-54
55-64
13-24
25-36
37-48
49-59
11-22
23-28
29-40
41-45
LCRS Flow
Avg.

(Iphd)
ND
1 1 ,684
2,464
ND
624
541
904
12,929
10,879
6,155
5,952
9,494
4,547
2,561
6,399
2,741
4,407
4,023
7,089
6,201
8,661
4,023
7,089
6,201
8,661
653
ND
1,571
1,086
Peak

(Iphd)
ND
26,339
4,666
ND
1,580
752
1,827
27,663
1 7,683
11,331
8,024
12,245
5,741
3,460
7,274
3,170
9,826
13,231
16,467
12,561
15,327
13,231
16,467
12,561
15,327
1,220
ND
4,074
2,067
LDS Flow
Avg.

(Iphd)
ND
7
5
ND
4
13
17
88
76
514
349
282
113
203
786
201
0
3
0
1
3
2
1
3
1
38
ND
8
4
Peak

(Iphd)
ND
23
8
ND
17
55
53
180
104
892
495
378
468
669
1,058
406
3
7
5
6
9
5
4
11
5
68
ND
26
7
Ea


(%)

99.94
99.80

99.39
97.64
98.14
99.32
99.30
91.64
94.14
97.03
97.52
92.08
87.72
92.65
99.99
99.93
99.99
99.98
99.97
99.95
99.98
99.96
99.99
94.18

99.51
99.64
Post-Closure Period'3'
Time
Period

(months)
77-85






NA




NA



NA




NA



NA



LCRS Flow
Avg.

(Iphd)
800






NA




NA



NA




NA



NA



Peak

(Iphd)
1,794






NA




NA



NA




NA



NA



LDS Flow
Avg.

(Iphd)
3






NA




NA



NA




NA



NA



Peak

(Iphd)
12






NA




NA



NA




NA



NA



Ea


(%)
99.57




























m
-------
        Table E-4.2.  Summary of LCRS and LDS Flow Rate Data for Considered Landfill Cells with GM Primary Liners (Continued).
Cell
No.



S2


V1(8)
V2(8)
W1

W2


X1

Cell
Area


(ha)
1.6


4.2
3.9
15.4

15.4


3.0

Start of
Waste
Placem.
(month-
year)
8-90


1-90
1-90
5-92

5-92


8-92


Final
Closure
(month-
year)
NA


NA
NA
NA

NA


NA

Initial Period of Operation'1'
Time
Period

(months)
1-9


1-17
1-17
1-8
9-12
1-8


1
2-7
LCRS Flow(4)
Avg.

(Iphd)
2,185


13,622
13,622
ND
7,492
ND


111,031
32,469
Peak

(Iphd)
4,650


49,828
49,828
ND
8,799
ND


111,031
104,645
LDS Flow
Avg.

(Iphd)
5


117
135
ND
439
ND


364
4
Peak

(Iphd)
24


153
256
ND
765
ND


364
25
Ea


(%)
99.78


99.14
99.01

94.14



99.67
99.99
Active Period of Operation'2'
Time
Period

(months)
10-17
18-33
34-46
14-46
14-46
13-24
25-35
9-20
21-32
33-35
8-19
20-33
LCRS Flow
Avg.

(Iphd)
654
ND
1,255
ND
ND
2,693
943
4,288
4,813
719
5,926
2,188
Peak

(Iphd)
1,135
ND
3,638
ND
ND
6,365
1,572
9,389
10,524
2,141
14,315
5,376
LDS Flow
Avg.

(Iphd)
5
ND
5
40
41
34
19
594
204
32
5
0
Peak

(Iphd)
24
ND
8
227
86
109
44
1,826
1,217
52
45
2
Ea


(%)
99.20

99.63


98.72
97.98
86.15
95.76
95.50
99.92
99.99
Post-Closure Period'3'
Time
Period

(months)
NA


NA
NA
NA

NA


NA

LCRS Flow
Avg.

(Iphd)
NA


NA
NA
NA

NA


NA

Peak

(Iphd)
NA


NA
NA
NA

NA


NA

LDS Flow
Avg.

(Iphd)
NA


NA
NA
NA

NA


NA

Peak

(Iphd)
NA


NA
NA
NA

NA


NA

Ea


(%)












m
        Notes:
        (1) "Initial Period of Operation" represents period after waste placement has started and only a small amount of waste has been placed in the cell.
        (2) "Active Period of Operation" represents period when waste thickness in cell is significant and/or an effective intermediate cover is placed on the waste.
        (3) "Post-Closure Period" represents period after final cover system has been placed on the entire cell.
        (4) Flow rates are given in liter/hectare/day.
        (5) NA = not applicable; ND = not determined.
        (6) LCRS for Cells 11, 12, and 13 are combined after February 1992. Reported flow rates are the average for the three cells.
        (7) Values given represent LCRS and LDS areas, respectively.
        (8) LCRS flows are combined for Cells O1 and O2 and for Cells V1 and V2. Reported flow rates are the average for the two cells at each landfill.
-------
Table E-4.3. Average LDS Flow Rates in Iphd for Considered Cells with
             GM Primary Liners.
(a) Cells with Sand LDS
Constructed with Formal CQA
Cell
No.
C1
C2
C3
C4
C5
D1
D3
D4
G1
G2
11
12
13
O1
O2
V1
V2
Number
Range
Mean
Median
Initial Period
of Operation

92
63
178
23
32
12
233

197
234
31
37
293
6
117
135
15
6-293
112
92
Active Period
of Operation
70
13
17
53
28

76

138
41
16
7
9
1
2
40
41
15
1-138
37
28
Post-Closure
Period





35




62
2
3




4
2-62
25
19
Constructed without Formal CQA(1)
Cell
No.
B1
B2
E1
E2
E3
E4
F1
K1









Number
Range
Mean
Median
Initial Period
of Operation


2,144
483
1,595
996
124
122









6
122-2,144
911
740
Active Period
of Operation
517
689
1107
703
1603

66
271









7
66-1,603
708
689
Post-Closure
Period
210
241















2
210-241
226

(b) Cells with GN LDS Constructed with Formal CQA
Cell
No.
N2
S1
S2
W1
W2
X1
Number
Range
Mean
Median
Initial Period
of Operation

12
5
439

55
4
5-439
128
34
Active Period
of Operation
227
20
5
27
358
2
6
2-358
106
23
Notes: (1) All Landfill B and E cells have CSPE GM Primary liners while Cells F1 and K1 hav
        HOPE GM Primary liners.  See Section E-4.2 for discussion.
                                      E-125
-------
m
Fo
en
n
-±±- -i n nnn
•s 	 x 1 U,UUU
c
o
2
CD
Q.
O
o 1,000-
T3

CD
Q_
CD
< 100 -
O)
D
Q
CD
"CD
^ 10
1
LL
c/)
Q
CD
O) 1 -
CD
> LDS Typ
< GM Prin

D




D
i
D
D
R
i
D
D

D
D
>e: Sand
nary Liner Type: HOPE
CQA (Yes/No)? Yes
D
D
D
D
	
D
D
D


D
D

Sand Sand GN
HOPE CSPE HOPE
No No Yes
               Figure E-4.1. Comparisons of LDS Average Flow Rates During Active Period of Operation for
                           Cells with GM Primary Liners. (Note:  Horizontal Line Represents the Mean
                            LDS Flow Rate for Each Group).
-------
(i) type of LDS drainage material (i.e., sand or GN LDS); and (ii) whether or not a formal
CQA program was used to construct the cell liner system. These two criteria were
selected because of their anticipated influence on LDS flow rates. The type of LDS
drainage material is important due to the substantially large potential for cells with sand
LDSs to release construction and compression water over time, especially during the
initial and early active periods of cell operation.  This is in contrast to cells with GN
drainage layers, which drain rapidly. As described earlier, the use of CQA can have a
significant effect on LDS flow rates for cells with GM primary liners (Bonaparte and
Gross,  1990, 1993).  Figure E-4.1 presents a graphical demonstration of the effects of
LDS type, GM  type, and use of CQA on LDS flow rates for the double-lined cells with
GM primary liners considered in this appendix.  In this figure LDS flow rates are plotted
for four groups of data: (i) cells with sand LDSs and HOPE GMs constructed with CQA;
(ii) cells with sand LDSs and HOPE GMs constructed without CQA; (iii) cells with sand
LDSs and CSPE GMs constructed without CQA; and (iv) cells with GN LDSs
constructed with  CQA. The effects of LDS material, and CQA on LDS flow rate are
discussed below.

Of the 25 cells incorporating sand LDSs, 17 were constructed with a formal CQA
program and eight were constructed without CQA.  The cells constructed with CQA
exhibited average LDS flow rates of about 6 to 290 Iphd during the initial period of
operation, 1 to 140 Iphd during the active period of operation, and 2 to 60 Iphd during
the post-closure period. The cells constructed without CQA exhibited LDS flow rates of
about 120 to 2,140 Iphd during the initial period of operation, 70 to 1,600 Iphd during the
active period of operation, and 210 to 240 Iphd during the post-closure period. Mean
values  of LDS flow rates for all cells are about one order of magnitude lower for cells
constructed with formal CQA programs than for cells constructed without CQA.  This
large difference in mean LDS flow rate is in part attributed to the benefits of CQA. The
difference is also attributed to differences in the materials and construction methods
associated with the time of cell construction (CSPE GM liners are typically installed
using solvent seaming while HOPE GM liners are typically installed using thermal
seaming).  For example, two of the eight cells without formal CQA programs have
HOPE  GM primary liners, the same type of GM primary liner used at all of the facilities
that had formal CQA (i.e.,  HOPE). These two cells (F1, K1) had average LDS flow rates
during  the active period of operation of 66 and 271 Iphd respectively. While these flow
rates are about two to seven times greater than the mean LDS flow rate for all cells that
had formal CQA programs, the flow rates are not statistically different at the 90%
confidence level from those cells constructed with CQA.  In contrast,  the cells with
CSPE  GM primary liners and no formal  CQA exhibited average LDS flow rates in the
range of 520 to 1,600 Iphd. These flow rates are statistically different than flow rates
from cells constructed with CQA.

It is difficult to accurately separate the effects of CQA and GM type (i.e., HOPE vs.
CSPE) and construction methods on leakage rates through GM liners for the cells
constructed without CQA.  The causes of this difficulty are: (i) data are available to the
authors for only two cells with HOPE GM primary liners constructed without formal CQA;
                                  E-127
-------
and (ii) no data are available for cells with CSPE GM primary liners constructed with
formal CQA. Despite these limitations, the available data suggest that both CQA and
GM type and construction methods have significant effects on leakage through GMs.

All six cells with GN LDSs in Table E-4.3 were constructed using a formal CQA
program. As shown  in Table E-4.3, average LDS flow rates for these cells were in the
range of 5 to 440 Iphd during the initial period of operation and 2 to 360 Iphd during the
active period of operation. These rates are somewhat higher than LDS flow rates for
cells with sand LDSs constructed with CQA; however, the difference between the mean
flow rate from these groups is not significant at the 90% confidence level.

Table E-4.4 summarizes LDS flow rates for cells with HOPE GM primary liners
constructed with CQA. From this table, it can be seen that of the 18 cells for which
initial period of operation  data are available, eight exhibited average LDS flow rates of
less than 50 Iphd, six had average LDS flow rates in the range of 50 to 200 Iphd, and
four had average LDS flow rates from 200 to 500 Iphd. Of the 19 cells for which active
period of operation data are available, 13 exhibited average flow rates of less than 50
Iphd, four had average flow rates in the range of 50 to 200 Iphd, and two had average
LDS flow rates in the range of 200 to 500 Iphd.  Post-closure period data are available
for four cells which have undergone CQA: three exhibited average flow rates less than
50 Iphd and one had an average flow rate in the range of 50 to 200 Iphd.
Table E-4.4. Frequency of Average Measured LDS Flow Rates for Cells with HOPE
GM Primary Liners Constructed with CQA.

< 50 Iphd
50-200 Iphd
200-500 Iphd
Initial Period of
Operation
8
7
4
Active Period of
Operation
15
4
2
Post-Closure
Period
3
1
-
On the basis of the data presented above, it is concluded that LDS flow rates
attributable to leakage through properly constructed HOPE GM primary liners that have
undergone CQA monitoring will often be less than 50 Iphd,  but occasionally in excess of
200 Iphd.  These results are consistent with the findings of Bonaparte and Gross (1990,
1993) discussed earlier in Section E-2.1.2 of this appendix.

E-4.2.2.4  GM Primary Liner Efficiencies

Tables E-4.2 and E-4.5 present calculated Ea values for the GM primary liners.  Table
E-4.5 has  been subdivided to separately report Ea values for the same cell groups as in
Table E-4.3 (i.e., grouped based  on  LDS material type and use of CQA during
                                  E-128
-------
Table E-4.5. "Apparent" Hydraulic Efficiencies, Ea, (in %) of GM liners
                                                                            (1)
(a) Cells with Sand LDS
Constructed with Formal CQA
Cell
No.
C1
C2
C3
C4
C5
D1
D3
D4
G1
G2
11
12
13
O1
O2
V1
V2
Number
Range
Mean
Median
Initial Period
of Operation

93.74
98.16
98.80
99.64

99.94
99.26

99.12

99.53
99.68

99.86
99.14
99.01
12
98.16-99.94
98.82
99.20
Active Period
of Operation
78.92
95.82
94.35
91.51
98.89

99.24

97.29
99.50
99.61
99.04
99.76
99.98
99.96


13
78.92-99.98
97.91(3)
99.14(3)
Post-Closure
Period





94.82




92.25
99.71
99.57




4
92.25-99.75
96.59
97.20
Constructed without Formal CQA
Cell
No.
B1
B2
E1
E2
E3
E4
F1
K1









Number
Range
Mean
Median
Initial Period
of Operation




83.08
95.06
99.14
99.31









4
83.08-99.31
94.15
97.10
Active Period
of Operation
86.45
76.68
87.75
85.88
73.56

99.23
96.20









7
73.56-99.23
86.54
86.45
Post-Closure
Period
73.62
35.82















2
35.82-73.62
54.72

(b) Cells with GN LDS Constructed with Formal CQA
Cell
No.
N2
S1
S2
W1
W2
X1
Number
Range
Mean
Median
Initial Period
of Operation

99.45
99.77
94.14

99.87
4
94.14-99.87
98.31
99.61
Active Period
of Operation
93.58
97.33
99.51
98.55
91.32
99.95
6
91.32-99.95
96.71
97.94
Notes: (1) Ea = (1 - LDS Flow / LCRS Flow) x 100 %
      (2)  All Landfill B and E cells have CSPE GM Primary liners while Cells F1 and K1 have
         HOPE GM Primary liners.  See Section E-4.2 for discussion.
      (3)  Cell C1 was excluded from calculating mean and median values of Ea for GM liners
         constructed with formal CQA
                                        E-129
-------
construction). The data in Tables E-4.2 and E-4.5 suggest that while LDS flow rates
generally decrease with time, Ea values may increase or decrease with time depending
on the relative rates of decrease of LCRS flow versus LDS flow. For many of the cells
considered in this appendix, Ea values decreased with time due to the faster reduction in
LCRS flow rate compared to the reduction in LDS flow rate. For cells with sand LDSs
and CQA, Ea values are highest during the initial period of operation (Eam = 98.82%;
where Eam =  mean apparent efficiency for all cells) and they decrease thereafter (Eam =
97.91% during the active period of operation and Eam = 96.59% during the post-closure
period).  For  cells with GN LDSs and CQA,  Eam values were 98.31 and 96.71%,
respectively,  for the initial and active periods of operation.

For many of the cells with GM primary liners constructed with CQA, Ea values were very
high (i.e., greater than 99%). However, a significant number of the cells showed lower
Ea values, in  the range of 91 to 99%. Only one cell, C1, had an extremely low Ea value
(61.27 to 84.40%). Cell C1 had low LDS flow rates that decreased with time. This
suggests low leachate leakage through the  GM primary liner.  However, because the
LCRS flow rates were also very low,  calculated Ea values were lower than other cells
constructed with CQA and with similar LDS flow rates conditions. Therefore,  Cell C1
average LDS flow rate was not included in calculating the Eam values presented above for
the group of cells with sand LDSs and CQA.  The data suggest that GM primary  liners
constructed with a formal CQA program will have Et values in the range of about 90 to
100%, with most values being above 95%, and many values being above 99%.

Table E-4.5 also presents calculated Ea values for cells with sand LDSs constructed
without formal CQA.  From the eight  cells in this group, the two which had HOPE  GM
primary liners had Ea values from about 96% greater than 99%, and the six which had
CSPE GM primary liners had Ea values from about 36 to 95%.  The calculated Ea values
for the CSPE GM primary liners constructed without CQA are much lower than
respective values calculated for cells constructed with HOPE GM primary liners and
CQA.  Again, this data demonstrates the effects of GM material and construction
methods and of implementing a formal  CQA program to construct GMs.

E-4.2.3   Implications for Landfill Performance

The evaluation results reported above indicate that GM liners can achieve primary liner
leakage rates less than 50 Iphd and Et values of 99% or more.  This is a very good level
of performance. The results also indicate, however, that GM primary liners sometimes
will not achieve this performance level and that lower Et values, in the range of 90 to
99%, are not uncommon.  This relatively broad range of Et values is a consequence of
the potential  for even appropriately installed GMs to have an occasional small hole,
typically due  to an imperfect seam, but also potentially due to a manufacturing or
construction-induced defect not identified by the CQA program.  Leakage can occur,
relatively unimpeded, through a GM hole if a low-permeability material such as a CCL or
a GCL does not underlie the GM.  If a hole occurs  at a critical location where a
sustained hydraulic head exists, such as a landfill sump, the leakage rate through the
                                 E-130
-------
hole can be significant.  In contrast, the GCL or CCL component of a composite liner
can impede flow through a GM hole, even if it occurs at a critical location. The
conclusion to be drawn from the data evaluation is that single liner systems with GMs
liners (installed on top of a relatively permeable subgrade) should not be used in
applications where Et values as low as 90% would be unacceptable, even if a thorough
CQA program is employed. In these cases, single-composite liner systems or double-
liner systems should be utilized. An exception to this conclusion may be made for
certain facilities, such as surface impoundments or small, shallow landfill cells, with GM
primary liners that can be field tested over the GM sheet and seams using electrical
leak location surveys, ponding tests, or other methods. For these facilities, higher
efficiencies (i.e., greater than 99%) may be achieved with GM liners by identifying and
repairing the GM holes during construction and,  especially for surface impoundments,
during operation.

E-4.3     Leakage Rates Through Composite Primary Liners

E-4.3.1   Description of Data

The performances of all 28 cells with GM/GCL composite primary liners and 13 of the
88 cells with GM/CCL or GM/GCL/CCL composite primary liners are assessed in this
section. The remaining 75 cells with composite primary liners were generally excluded
from the assessment because: (i) they did not have continuous LCRS and LDS flow rate
data available for an individual cell from the start of operation; or (ii) there were
insufficient LCRS and LDS chemical constituent data to evaluate whether primary liner
leakage did or did not occur. Flow rate data are available for 41  cells at 16 landfills with
monitoring periods of up to 121 months. Data from several of the 41  cells in this study
were included in the previous  EPA study by Bonaparte and Gross (1993); however, this
study contains additional data for most of these cells.

The liner systems for all of the considered cells were constructed under a formal CQA
program.  Descriptions of the liner systems installed at these landfills are presented in
Table E-4.6.  The LDS consists of a sand layer, a gravel layer over a GN, or a GN. The
secondary liner of the double-liner system is a single GM, a GM/CCL composite, or a
GM/GCL composite. Tables E-4.7 and E-4.8 contain a summary of the LCRS and LDS
flow rate database for cells with GM/GCL and GM/CCL or GM/GCL/CCL primary liners,
respectively. Each of the cells for which post-closure data are available has a final
cover system with a GM or GM/GCL composite barrier. For cells with GM/CCL or
GM/GCL/CCL composite primary liners,  the database also includes LCRS and LDS
chemical constituent data. A summary of these data is presented in Table E-4.9.  For
reasons discussed subsequently, chemical constituent data were not used in the
evaluation of GM/GCL primary liners.
                                 E-131
-------
        Table E-4.6.  Description of Liner System Components for Considered Landfills with Composite Primary Liners.
Cell
No.
B3
Y2
AK1
AL1
AM 1&2
AO 1&2
AR1
AD1 (Base)
(Side slopes)
AD7 (Base)
(Side slopes)
AQ1
AQ10
AX 1-16
C6
AW1&2
BB1-3
AZ1
AY 1-3
I 4-5 (Base)
(Side slopes)
Type'1'
of
Waste
MSW
MSW
MSW
MSW
MSW
MSW
MSW
HW

HW

HW
HW
MSW
MSW
MSW
MSW
MSW
HW
MSW

LCRS
Material'2'
S
S
S/GN
S
G/S
S
G/TC
S
GN
S
GN
G/GN
G/GN
S
S
S
S
S
GN
TC/G
NA
Thickness'3'
(mm)
450
600
600/5
600
300/150
600
300/400
300
5
300
5
300/5
300/5
600
600
450
600
600
5
1 50/450
NA
Primary Liner
GM'4'
Type (and
Thickness (mm))
CSPE(0.9)
HDPE(2.0)
HDPE(1.5)
HDPE(1.5)
HDPE(2.0)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
HDPE(2.0)
HDPE(2.0)
HDPE(1.5)
HDPE(2.0)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
Lower Component
Material'5'
CCL
CCL
GCL/CCL
CCL
CCL
CCL
GCL/CCL
CCL
CCL
CCL
CCL
CCL
CCL
GCL
GCL
GCL
GCL
GCL
GCL
GCL
GCL
Thickness
(mm)
600
450
6/600
900
450
900
6/300
900
900
900
900
450
450
6
6
6
6
6
6
6
6
Max. Hydraulic
Conductivity (m/s)
1x1 0'9
1x1 0"9
5x10"11/1x10"9
1x1 0"9
1x1 0"8
5x10-10
5x10"11/1x10"7
1x1 0'9
1x1 0"9
1x1 0"9
1x1 0'9
1x1 0"9
1x1 0"9
5x1 0-11
5x1 0-11
5x1 0-11
5x1 0-11
5x1 0-11
5x1 0-11
5x1 0-11
5x10-11
LDS
Material'2'
S
S
GN
GN
GN
GN
GN
S
GN
GN
GN
G/GN
G/GN
S
S
S
GN
GN
GN
GN
GN
Thickness
(mm)
450
300
5
5
5
5
10
300
5
5
5
300/5
300/5
300
450
300
5
5
5
10
5
Secondary Liner
GM
Type (and
Thickness (mm))
PVC(O.S)
HDPE(2.0)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
HDPE(2.0)
HDPE(1.5)
HDPE(2.0)
HDPE(1.5)
HDPE(1.5)
HDPE(1.5)
HDPE(1.0)
HDPE(1.5)
HDPE(1.5)
Lower Component
Material
NA'6'
CCL
GCL
CCL
NA
CCL
CCL
CCL
CCL
CCL
CCL
CCL
CCL
CCL
CCL
GCL
NA
CCL
GCL
CCL
CCL
Thickness
(mm)
NA
600
6
900
NA
600
600
900
900
900
900
900
900
200
300
6
NA
900
6
150
150
m
w
ro
        Notes:  (1) Waste Types: MSW = Municipal Solid Waste; HW = Hazardous Solid Waste.
              (2) LCRS and LDS Material Types:  GN = GN or Geocomposite; TC = Tire Chips; S = Sand; G = Gravel
              (3) All material thicknesses are nominal values.
              (4) GM Types: HOPE = High Density Polyethylene;  CSPE = Chlorosulfonated Polyethylene; PVC = Polyvinyl Chloride.
              (5) Liner Lower Component Material Types:  CCL = Compacted Clay Liner;  GCL = Geosynthetic Clay Liner.
              (6) NA = Not Applicable.
-------
        Table E-4.7.  Summary of Flow Rate Data for Landfill Cells with GM/GCL Composite Primary Liners.
Cell
No.



AX1
AX2
AX3
AX4
AX5
AX6
AX7
AX8
AX9
AX10
AX11
AX12
AX13
AX14
AX15
AX16
C6
AW1
AW2
BB1
BB2
BBS
AZ1
Cell
Area


(hectare)
2.0
2.0
1.7
1.7
2.8
3.9
2.6
3.8
3.3
3.9
3.0
4.0
3.0
2.8
2.8
4.5
3.6
2.4
2.4
4.0
2.4
2.8
3.8
Start of
Waste
Placem.
(month-
year)
7-88
7-88
9-88
9-88
10-88
12-88
2-89
7-89
12-89
7-90
2-90
10-90
1-91
4-91
5-92
1-93
8-93
5-93
8-93
2-91
1-93
1-93
12-92
End of
Final
Closure
(month-
year)
2-91
2-91
4-93
4-93
NA(5)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Initial Period of Operation'1'
Time
Period'6'

(months)
1-2
1-5
1-5
1-12
1-11
1-9
1-10
1-14
1-9
1-7
1-16
1-12
1-7
1-11
1-12
1-10
1-10
1-12
1-10
1-6
1-11
1-11
2-12
LCRS Flow14'
Avg.

(Iphd)
16,718
15,521
3,366
2,534
1,384
3,759
5,376
4,881
1,047
2,786
4,675
3,494
6,683
2,777
5,573
8,601
3,273
6,358
3,553
10,378
ND(5)
ND
4,093
Peak

(Iphd)
19,738
58,671
7,985
12,688
3,394
7,171
12,155
21,038
3,478
13,698
14,586
8,836
14,343
6,582
11,809
17,756
12,155
20,570
7,480
22,130
ND
ND
5,219
LDS Flow
Avg.

(Iphd)
0
15
35
101
37
53
34
48
1
0
0
0
0
0
0
0
178
131
290
15
1
0
0
Peak

(Iphd)
0
45
151
860
92
93
47
189
7
0
0
0
0
0
0
0
823
524
514
65
12
0
0
Active Period of Operation'2'
Time
Period'6'

(months)
3-33
6-33
6-56
13-56
12-80
10-80
11-76
15-71
10-65
8-59
17-62
13-56
8-53
12-38
13-37
11-29
11-17


7-47
12-23
12-23
13-31
LCRS Flow
Avg.

(Iphd)
540
281
307
75
56
168
234
439
41
374
150
803
1,408
281
299
819
393


2,494
5,422
2,284
3,473
Peak

(Iphd)
2,383
570
1,075
187
191
655
851
1,384
159
645
337
3,029
9,294
449
561
5,096
1,403


8,983
14,042
7,945
5,054
LDS Flow
Avg.

(Iphd)
0
2
4
1
2
0
2
0
0
0
0
0
0
0
0
0
3


6
0
0
2
Peak

(Iphd)
0
21
47
13
37
0
9
0
0
0
0
0
0
0
0
0
15


25
0
1
22
Post-CIc
Time
Period'6'

(months)
34-83
34-83
57-81
57-81



















LCRS
Avg.

(Iphd)
66
178
206
47





















i























m
-------
           Table E-4.7. Summary of Flow Rate Data for Landfill Cells with GM/GCL Composite Primary Liners (Con
Cell
No.



AY1
AY2
AYS
14
15
Cell
Area


(hectare)
1.3
1.0
1.0
4.7
4.7
Start of
Waste
Placem.
(month-
year)
10-94
8-94
8-94
5-92
7-92
End of
Final
Closure
(month-
year)
NA
NA
NA
7-94
5-94
Initial Period of Operation'1'
Time
Period'6'

(months)
1-9
1-11
1-11
1-12
1-12
LCRS Flow14'
Avg.

(Iphd)
6,803
10,964
12,198
4,494
3,938
Peak

(Iphd)
12,439
23,914
32,326
17,251
7,985
LDS Flow
Avg.

(Iphd)
0
3
6
24
2
Peak

(Iphd)
0
12
28
70
11
Active Period of Operation'2'
Time
Period'6'

(months)



13-26
13-21
LCRS Flow
Avg.

(Iphd)



2,041
3,108
Peak

(Iphd)



4,282
11,669
LDS Flow
Avg.

(Iphd)



26
11
Peak

(Iphd)



142
54
Post-CIc
Time
Period'6'

(months)



27-36
22-34
LCRS
Avg.

(Iphd)



567
189


i





m
Notes:
(1) "Initial Period of Operation" represents period after waste placement has started and only a small amount of waste has been placed
(2) "Active Period of Operation" represents period when waste thickness in cell is significant and/or an effective intermediate cover is pi
(3) "Post-Closure Period" represents period after final cover system has been placed on the entire cell.
(4) Flow rates are given in liter/hectare/day.
(5) NA = Not Applicable;  ND = Not Determined.
(6) Breakthrough time for steady-state saturated flow through GCL component of composite liner is estimated to be 2 months based on
    using Darcy's equation and a saturated hydraulic conductivity of 5 x 10"11 m/s, hydraulic gradient of 5, and effective porosity of 0.2.
    calculation, it is assumed that flow through the GM component of the composite liner occurs through small holes and is instantaneo
-------
         Table E-4.8. Summary of Flow Rate Data for Landfill Cells with GM/CCL or GM/GCL/CCL Composite Primary Liners.
Cell
No.



B3(!))







Y2



AK1
AL1

AM1





AM2





A01


AO2

AR1

Cell
Area


(ha)
6.4







3.0



1.4
14.9

3.2/2 .4(7)





4.8/2 .4(7)





1.8


1.8

9.7

Start of
Waste
Placem.
(month-
year)
7-87







1-91



10-93
1990

10-90





10-90





1-92


7-92

3-92

End of
Final
Closure
(month-
year)
NA(6)







NA



NA
NA

NA





NA





NA


NA

NA

Initial Period of Operation'1'
Time
Period

(months)
1-4







1-10



1-12
1-29

1-9





1-9





1-5


1-5

1-11

LCRS Flo\A/4)
Avg.

(Iphd)
15,304







23,368



9,867
ND(6)

ND





ND





ND


15,881

27,042

Peak

(Iphd)
24,858







36,791



17,986
ND

ND





ND





ND


24,541

65,871

LDS Flow
Avg.

(Iphd)
1,394







655



206
ND

ND





ND





ND


149

292

Peak

(Iphd)
4,250







1,768



804
ND

ND





ND





ND


191

705

Active Period of Operation'2'
Time
Period

(months)
5-16
17-28
29-40
41-52
53-64
65-76
77-88
89-93
11-22
23-34
35-46
47-54

30-41
42-54
10-21
22-33
34-45
46-57
58-69
70-81
10-21
22-33
34-45
46-57
58-69
70-81
6-17
18-29
30-37
6-17
18-31
12-23
24-36
LCRS Flow
Avg.

(Iphd)
5,700
9,272
7,575
2,859
1,189
403
560
578
10,353
1 1 ,344
4,404
4,397

934
1,349
270
236
111
20
18
11
32
35
17
67
64
112
1,984
1,299
1,144
3,027
1,688
11,251
9,668
Peak

(Iphd)
8,935
22,444
13,978
6,043
2,280
490
919
648
19,204
25,309
6,380
5,199

2,085
5,885
533
329
283
77
21
18
154
51
45
274
181
136
4,130
1,577
1,371
5,266
2,383
23,384
26,274
LDS Flow
Avg.

(Iphd)
124
101
262
231
45
92
102
98
370
90
70
48

231
103
15
10
3
1
1
5
9
9
3
0
8
9
184
96
60
110
33
181
155
Peak

(Iphd)
266
168
803
713
152
133
193
109
1,993
168
248
56

367
183
64
15
14
1
1
8
42
29
26
0
13
13
353
126
102
158
64
470
442
Ea
(Average
for Active
Period)
(%)
97.8
98.9
96.5
91.9
96.2
77.3
81.8
83.0
96.4
99.2
98.4
98.9

75.3
92.4
94.4
95.8
97.3
95.0
94.4
54.4
71.9
74.3
82.4
100.0
87.5
92.0
90.7
92.6
94.8
96.4
98.1
98.4
98.4
Post-Closure Period'3'
Time
Period

(months)


































LCRS Flow
Avg.

(Iphd)


































Peak

(Iphd)


































LDS Flow
Avg.

(Iphd)


































Peak

(Iphd)


































Ea
(Average
for P-C
Period)
(%)


































m
CO
01
-------
            Table E-4.8. Summary of Flow Rate Data for Landfill Cells with GM/CCL or GM/GCL/CCL Composite Primary Liners (Continued).
Cell
No.



AD1







AD7




AQ1



AQ10


Cell
Area


(ha)
0.6







1.5




0.6



0.9


Start of
Waste
Placem.
(month-
year)
5-85







9-87




3-86



1-89


End of
Final
Closure
(month-
year)
7-88







10-93




early 90



mid 91


Initial Period of Operation'1'
Time
Period

(months)
1-12







1-12




1-6



1-9


LCRS Flo\A/4)
Avg.

(Iphd)
ND







12,597




10,203



ND


Peak

(Iphd)
ND







26,492




18,944



ND


LDS Flow
Avg.

(Iphd)
ND







135




352



14


Peak

(Iphd)
ND







1,101




569



32


Active Period of Operation'2'
Time
Period

(months)
13-20
21-32






13-24
25-36
37-48
49-60
61-69
7-25
26-34
35-46
47-58
10-14
15-26

LCRS Flow
Avg.

(Iphd)
ND
373






2,212
1,539
1,429
249
480
ND
ND
ND
4,530
ND
15,933

Peak

(Iphd)
ND
892






2,857
2,755
2,813
629
614
ND
ND
ND
10,531
ND
38,751

LDS Flow
Avg.

(Iphd)
ND
107






71
96
17
33
64
255
ND
197
116
26
48

Peak

(Iphd)
ND
603






291
393
21
74
112
1239
ND
435
143
32
250

Ea
(Average
for Active
Period)
(%)

71.4






96.8
93.8
98.8
87.0
86.6



97.4

99.7

Post-Closure Period'3'
Time
Period

(months)
33-44
45-51
52-63
64-75
76-87
88-99
100-111
112-121
70-81
82-87



59-65
66-77
78-89
90-97
27-38
39-50
51-63
LCRS Flow
Avg.

(Iphd)
145
85
3
3
3
1
1
2
375
165



5,835
644
1,367
1,615
682
300
852
Peak

(Iphd)
652
130
22
42
21
4
2
9
533
334



1 1 ,244
1,011
3,264
3,575
2,251
1,709
1,588
LDS Flow
Avg.

(Iphd)
24
26
28
42
23
8
5
6
73
105



215
117
98
51
29
18
24
Peak

(Iphd)
42
31
45
103
68
46
43
24
157
172



246
165
132
118
48
63
75
Ea
(Average
for P-C
Period)
(%)
83.4
69.5
-833
-1300
-667
-700
-400
-200
80.5
36.3



96.3
81.8
92.8
96.8
95.7
94.0
97.2
m
CO
CD
            Notes:
            (1) "Initial Period of Operation" represents period after waste placement has started and only a small amount of waste has been placed in the cell.
            (2) "Active Period of Operation" represents period when waste thickness in cell is significant and/or an effective intermediate cover is placed on the waste.
            (3) "Post Closure Period" represents period after final cover system has been placed on the entire cell.
            (4) Flow rates are given in liter/hectare/day.
            (5) 65 percent of Cell B3 received final cover after 60 months of start of waste placement.
            (6) NA = not applicable; ND = not determined.
            (7) Values given represent LCRS and LDS areas, respectively.
            (8) Estimated breakthrough time for steady-state saturated flow through CCL or GCL/CCL component of composite liner is given in Table E-4.12.
-------
          Table E-4.9.  Summary of Liquid Chemistry for the LCRS and LDS of Landfills with Composite Primary Liners.
Landfill ID
Cell No. -System
Waste Placement Period
Liquid Sampling Period
Parameter Units
pH
Specific Conductance ^mhos/cm
IDS mg/l
TSS mg/l
COD mg/l
BODS mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1,1 - Dichloroethane |ig/l
1 ,2 - Dichloroethane |ig/l
cis- 1 ,2-Dichloroethene |ig/l
Trans- 1 ,2-Dichloroethene |ig/l
Ethylbenzene |ig/l
Methylene Chloride |ig/l
1,1,1 - Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl Chloride |ig/l
Xylenes |ig/l
B(MSW)
B3-LCRS
B3-LDS
07/87-05/92
07/87-1 0/94
6.82
2,956
4,140
161
1,912
422
554

690
131


450

< 20
49
< 44
102
< 5
< 40
< 5


15
< 80
< 107

< 78
< 10

07/87-1 0/94

1,554
1,148
45
131
88
138

148
335


46
< 24
16
3
< 17
519
< 7
< 7
< 6


< 6
< 17
< 6

< 9
< 12

Y (MSW)
Y2-LCRS
Y2-LDS
1 990-date
04/91 -04/94
6.60
5,360
4,939

5,265
2,076
1,436
2,520
628
108
1,994

433
17
8
58
5
185
10
45



118
121

11
720

69
04/91-04/94
7.28
1,583
881

50
3
11
335
58
231
179
54
46
4
< 2
11
8
25









7


AK(MSW)
AK1-LCRS
AK1-LDS
1 0/93-date
1 2/93-03/95
6.65
1,592


1,062
< 2
245
711
94
45
387
51
64
< 3
< 19
< 27
< 48
< 50
< 5
46
< 2
< 1
< 1
< 7
603
134
< 1
95
< 11
< 30
12/93-03/95
7.20
679

60
13

4
331
4
25
116
29
5
< 1
< 3
< 2
< 5
< 50
< 1
< 1
< 1
< 1
< 1
< 1
< 5
3
< 1
< 1
< 5
< 3
AL (MSW)
AL1-LCRS
AL1-LDS
1 990 - date
06/91-05/95
8.09
2,707
2,892
110
860
1,134
245
261
430
219
150
98
236
4
< 11
< 64
< 36
57
< 6
< 8
< 6
< 7
< 2
< 12
245
< 8
< 12
78
< 11
< 96
1 2/89-05/95
7.04
2,449
2,482
24
< 11
< 2
3
199
151
1,028
465
121
38
< 5
< 7
< 10
< 17
< 35
< 2
< 7
< 2
< 3
< 2
< 2
< 77
< 4
< 2
< 7
< 5
< 5
m
CO
-------
           Table E-4.9. Summary of Liquid Chemistry for the LCRS and LDS of Landfills with Composite Primary Liners (Continued).
Landfill ID
Cell No. -System
Waste Placement Period
Liquid Sampling Period
Parameter Units
PH
Specific Conductance ^mhos/cm
IDS mg/l
TSS mg/l
COD mg/l
BODS mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium ng/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1,1 - Dichloroethane ng/l
1 ,2 - Dichloroethane |ig/l
cis- 1 ,2-Dichloroethene |ig/l
Trans- 1 ,2-Dichloroethene |ig/l
Ethylbenzene ng/l
Methylene Chloride |ig/l
1,1,1 - Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene ng/l
Vinyl Chloride |ig/l
Xylenes |ig/l
AM(MSW)
AM1-LCRS
AM1-LDS
10/90-02/91
04/91-02/95
6.62
2,451
1,709

396

108
1,204
183
18
376
95
92
236
< 20
< 30
< 2.0
< 40
17
172
6
324
< 5
57
< 6
< 21
3
267
< 21
122
02/92-1 2/93
6.90
17,250
14,330

184

116
407
2,725
1,500
1,700
283
1,348
< 2
< 20
< 30

< 40
< 1
< 3
< 1
< 1
< 1
< 1
3
< 1
< 1
< 1
< 1
< 1
AM2-LCRS
AM2-LDS
10/90-02/91
04/91 -02/96
6.60
2,795
1,413

65

22
867
332
9
319
74
88
235
< 20
< 57
< 7
< 91
13
136
< 11
547
< 8
29
< 20
< 16
< 12
146
< 11
71
02/92-01/96
7.26
17,063
14,363

76

28
163
2,659
1,425
1,819
299
1,929
< 10
< 10
< 30
3
< 40
< 1
< 2
< 1
< 1
< 1
< 1
< 2
< 1
< 1
< 1
< 2
< 2
AO (MSW)
AO-LCRS
AO1-LDS
AO2-LDS
01/92-date
08/92-06/95
7.30
6,592
2,178
1,310
618

414
1,756
862
54
275
146
786
45
< 9
54
< 32
108
7
15
< 5
< 6
< 4
11
75
< 11
< 4
167
< 12
34
08/92-05/95
7.17
1,132
690

142

43
556
38
44
205
66
19
7
< 1
< 1
< 1
< 50
< 1
< 3
< 1
< 1
< 1
< 1
< 5
< 1
< 1
< 86
< 7
< 3
08/92-06/95
6.72
1,118
722

497

8
558
24
91
196
55
8
2
< 1
< 1
< 1
< 50
< 1
< 1
< 1
< 1
< 1
< 1
< 5
< 1
< 1
< 1
< 5
< 3
AR (MSW)
AR1-LCRS
AR1-LDS
03/92-date
11/92-08/94
6.92
5,650
3,923

1,238
290
333
2,075
1,625
380
230
153
850
13
< 6
62
< 13
< 20
< 100
< 100
< 100
< 100
< 100
< 100
< 100
< 100
< 100
< 100
< 100
< 100
11/92-08/94
7.00
1,368
848

22
5
5
325
35
325
186
119
26
< 8
< 7
< 10
< 9
< 100
< 1
< 2
< 2
< 50
< 2
< 2
< 2
< 2
< 2
< 2
< 4
< 4
m
CO
CO
-------
           Table E-4.9. Summary of Liquid Chemistry for the LCRS and LDS of Landfills with Composite Primary Liners (Continued).
m
CO
CD
Landfill ID
Cell No. -System
Waste Placement Period
Liquid Sampling Period
Parameter Units
PH
Specific Conductance ^mhos/cm
IDS mg/l
TSS mg/l
COD mg/l
BODS mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic ng/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel ng/l
Benzene |ig/l
1,1 - Dichloroethane |ig/l
1 ,2 - Dichloroethane |ig/l
cis- 1 ,2-Dichloroethene ng/l
Trans- 1 ,2-Dichloroethene |ig/l
Ethylbenzene |ig/l
Methylene Chloride |ig/l
1,1,1 - Trichloroethane ng/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl Chloride |ig/l
Xylenes ng/l
AD(HW)
AD1-LCRS
AD1-LDS
05/85 - late '87
06/85-1 2/94
8.72
40,159




2,006

12,709
4,428


10,394
125,252
< 410
294
192

< 184
< 138
< 67

< 48
174
< 292
< 228
< 53
< 189
1,371

06/85-1 2/94
7.40
1,755




8

207
404


177
< 13
< 4
29
< 29

< 4
< 5
< 3

< 2
< 6
< 4
< 4
< 2
< 5
< 9

AD7-LCRS
AD7-LDS
09/87 - 09/88, 05-06/93
09/87-1 2/94
9.99
39,036




4,471

10,759
6,105


5,549
34,572
< 56
158
< 306

555
< 604
< 2,181

< 238
< 850
< 602
< 466
< 240
1,042
< 7,438

09/87-1 2/94
7.45
2,512




7

301
6,105


250
< 14
< 3
< 12
< 16

< 4
< 6
< 3

< 3
< 6
< 3
< 3
< 25
< 5
< 9

AQ (HW)
AQ1-LCRS
AQ1-LDS
03/86 - early 90
02/91 -1 0/93
7.90
23,200











189
< 5
22
24
1,190
< 8
14
< 5

18
< 5
5
9
60
< 5
13

02/91-10/93
7.36
2,582











< 10
< 5
12
31
< 40
< 4
< 5
< 3

< 2
< 5
6
< 8
19
7
< 10

AQ10-LCRS
AQ10-LDS
late 88 - mid 91
02/91-10/93
7.20
5,574
















< 6
135
4

< 10
< 5
< 3
36
5
< 13
< 7

02/91-10/93
7.51
4,154











< 11
< 5
50
< 17
84
< 4
< 5
< 3

< 2
< 5
< 3
< 4
< 2
< 5
< 10

-------
E-4.3.2   GM/GCL Composite Primary Liners

E-4.3.2.1 Interpretation of Data

The interpretation of LDS flow rate data for cells with GM/GCL primary liners is also
relatively straightforward because the potential sources of LDS flow for cells with
GM/GCL composite primary liners are construction water, compression water, and
primary liner leakage.  Ground-water infiltration is not a potential source  because all of
the facilities are reportedly located above the ground-water table. Consolidation water
is a potential source if the GCL hydrates prior to waste filling. While GCL installation
procedures are designed to keep the GCL as dry as possible, post-construction
changes  in moisture content can occur as a result of construction water  in the LDS.
The GCL components of composite  liners will produce little, if any, consolidation water,
depending on their  moisture content at the start of waste placement. The analysis of
these cells has been performed considering only LCRS and LDS flow rate data. LDS
chemical constituent data were not considered in the data interpretation  because the
chemistry data were unavailable for  many landfill cells, possibly due to the nonexistent
or very low LDS flow rates from these cells.

E-4.3.2.2 Summary of Flow Rate Data

Average and peak monthly flow rates for the 28 landfill cells with GM/GCL primary liners
are presented in Table E-4.7 for the  three landfill operational time periods described
earlier. Average monthly LCRS flow rates ranged from about 1,050 to 16,700 Iphd
during the initial period of operation,  about 40 to 5,420 Iphd during the active period,
and about 50 to 570 Iphd after closure. Average monthly LDS flow rates, excluding Cell
14, ranged from about 0 to 290 Iphd during the initial period of operation, 0 to 11 Iphd
during the active period, and 0 to 2 Iphd after closure. Cell 14 was excluded from the
data summary because of anomalous  LDS flow measurements during the active and
post-closure periods.  The LDS of this  cell exhibited much higher flow rates than any
other LDS underlying a GM/GCL composite liner, and also increasing LDS flow rates
under conditions of decreasing LCRS flow rates. The reason for this trend in data is
unknown. The landfill operator believes the high LDS flow rates are due to surface-
water runoff problems around the  landfill cell perimeter and direct infiltration of this
runoff into both the  LCRS and LDS via the liner system anchor trench at the cell
perimeter. This hypothesis has not been verified. A review of the data in Table E-4.7
indicates that peak  monthly LCRS and LDS flow rates are typically two to five times the
average monthly values calculated for the monitoring periods reported in Table E-4.7.
This  difference becomes larger in the LDS when the flow  rates become very low: this is
mostly an artifact of the LDS pumping  schedule, which becomes infrequent when the
flow rate  becomes very low.  Table E-4.7 shows that between the initial and active
periods of operation, LCRS flow rates typically decreased one to two orders of
magnitude and LDS flow rates decreased one to three orders of magnitude.
                                  E-140
-------
E-4.3.2.3 Liner Efficiencies

Table E-4.10 summarizes calculated Ea values for the GM/GCL composite primary
liners.  Table E-4.10 has been subdivided to separately report Ea values for cells with
sand LDSs and cells with GN LDSs due to the substantially large potential for cells with
sand LDSs to generate construction and compression water compared to cells with GN
LDSs.  For cells with sand LDSs, Ea values are lowest during the initial period of
operation (Eam = 98.60%) and increase significantly thereafter (Eam = 99.53% during the
active period of operation and Eam = 99.89% during the post-closure period). The lower
value of Eam during the initial period of operation is attributed to LDS flow from
construction water.  By conservatively ignoring long-term contributions to LDS flow from
construction water, compression water, and consolidation water, Ea values during the
active period of operation and the post-closure period can be interpreted to reflect true
liner efficiencies (Et) for the prevailing conditions.

For cells with  GN LDSs, Eam is 99.96% during the initial period of operation, 99.87%
during the active period of operation, and 98.78% during the post-closure period.
During the initial period of operation, the Eam value for cells with GN LDSs (i.e., 99.96%)
is higher than the Eam value for cells with sand LDSs (i.e., 98.60%). This higher
efficiency can be attributed to the differences in liquid storage capacity and hydraulic
transmissivity between sand and GN drainage materials.  By conservatively neglecting
the potential for consolidation water, Ea values for cells with GN LDSs can be
interpreted to reflect Et.

In summary, from Table E-4.7 (and excluding Cell 14), monthly average LDS flow rates
potentially attributable to primary liner leakage vary from 0 to 11 Iphd, with many values
reported as zero, and most values being less than about 2 Iphd. Primary liner leakage
rates of this magnitude are very low. The results in Table E-4.10 further indicate that Et
for a GM/GCL constructed to current standards with appropriate CQA will typically be in
the range of 99 to 100%, and will frequently be in excess of 99.9%.

E-4.3.3  GM/CCL and GM/GCL/CCL Composite Primary Liners

E-4.3.3.1 Interpretation of Data

The interpretation of flow rate data for cells with GM/CCL or GM/GCL/CCL composite
primary liners is more complicated than the interpretation for GM/GCL composite
primary liners because of the relatively large contribution of consolidation water to LDS
flow. Previous investigations, e.g., Bonaparte and Gross (1990), could not distinguish
the occurrence (or lack thereof) of primary liner leakage from the much larger volumes
of LDS consolidation water.  Also, breakthrough times (i.e., times of travel) for advective
transport through the CCL or GCL/CCL component of a composite liner can be quite
long, further complicating the evaluation.  Both LCRS and LDS flow rate and chemical
constituent data were used in this appendix to evaluate GM/CCL and GM/GCL/CCL
composite liner performance.
                                  E-141
-------
Table E-4.10.  "Apparent" Hydraulic Efficiencies, Ea, for GM/GCL Composite
              Liners (Excluding Cell 14).
Cells with Sand IDS
Cell
No.
AX1
AX2
AX3
AX4
AX5
AX6
AX7
AX8
AX9
AX10
AX11
AX12
AX13
AX14
AX15
AX16
C6
AW1
AW2
slumber
Range
Mean
Median
Initial Period
of Operation
100.00
99.90
98.97
96.01
97.30
98.58
99.37
99.02
99.91
100.00
100.00
100.00
100.00
100.00
100.00
100.00
94.57
97.94
91.84
19
91.84-100
98.60
99.90
Active Period
of Operation
100.00
99.33
98.70
98.75
96.67
100.00
99.20
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
99.29


17
96.67-100
99.53
100.00
Post-Closure
Period
100.00
100.00
99.55
100.00















4
99.55-100
99.89
100.00
Cells with GT/GN IDS
Cell
No.
BB1
BB2
BBS
AZ1
AY1
AY2
AYS
15











slumber
Range
Mean
Median
Initial Period
of Operation
99.86


100.00
100.00
99.97
99.95
99.95











7
99.86-100
99.95
99.95
Active Period
of Operation
99.78
100.00
100.00
99.94



99.65











6
98.73-100
99.87
99.94
Post-Closure
Period







98.78











1
98.78
98.78
98.78
Notes: (1) Apparent Efficiency = (1 - IDS Flow / LCRS Flow) x 100 %
                                    E-142
-------
£-4.3.3.2 Analysis of Flow Rate Data

Average and peak monthly flow rate data for the 13 landfill cells with GM/CCL or
GM/GCL/CCL composite primary liners are presented in Table E-4.8. Average monthly
LCRS flow rates ranged from about 9,900 to 27,000 Iphd during the initial period of
operation, 10 to 15,900 Iphd during the active period of operation, and 1 to 5,800 Iphd
during the post-closure period. Average monthly LDS flow rates ranged from about 10
to 1,400 Iphd during the initial period of operation, 0 to 370 Iphd during the active period
of operation, and 5 to 210 Iphd during the post-closure period. Peak monthly LCRS and
LDS flow rates are typically two to five times the reported average monthly value
calculated for the reported monitoring period. Between the initial period of operation
and later stages of the  active period of operation, flow rates for both the LCRS and LDS
typically decreased by  one to two orders of magnitude. For the active and post-closure
periods in particular, LDS flow rates for cells with GM/CCL and GM/GCL/CCL composite
primary liners are much higher than those for cells with GM/GCL composite primary
liners. These higher flow rates are attributable to consolidation water.

The flow rate data in Table E-4.8 were used to calculate Ea values for GM/CCL and
GM/GCL/CCL composite primary liners during the active period of operation and during
the post-closure period. Ea values were not calculated for the initial period of operation
because leachate breakthrough times for CCL or GCL/CCL components of the primary
liners were generally estimated to be greater than the initial period of operation.
Calculated Ea values are also presented in Table E-4.8. As can be seen in this table, Ea
values ranged from 54.5 to 100.0% during the active  period of operation, and from a
negative value to 97.2% during the post-closure period. These Ea values are lower than
corresponding values for GM/GCL composite primary liners, due to the presence of
consolidation water in the LDS. For comparison, Ea values for GM/GCL liners ranged
from 96.7 to 100% during the active period and  98.8 to 100% during the post-closure
period. Negative Ea values for Cell AD1 imply higher LDS than LCRS flow rates, a
consequence of continuing CCL consolidation and/or secondary compression  after cell
closure.  Additional observations with respect to the LDS flow rate data and calculated
Ea values for GM/CCL  and GM/GCL/CCL composite liners are given below.

    •  Consolidation water flow rates are dependent on the thickness and hydraulic
      properties of the CCL and the rate of overlying waste placement. LDS flows
      from this source may increase or decrease over time depending on the filling
      schedule for the landfill cell.  For most facilities, the filling schedule results in
      relatively large  LDS flow rates early in the active period. Average monthly flow
      rates during the active period may initially be as high as 200 to 400 Iphd, with
      flows attributed primarily to consolidation water. Average LDS flow rates in the
      range of 0 to 50 Iphd are not uncommon in the latter active life and after closure,
      although flow rates of more than 100 to 200 Iphd are occasionally observed.
      Continuing low  rates of LDS flow after closure are attributed primarily to
      continuing consolidation or secondary compression of the CCL component of the
      composite liner.
                                  E-143
-------
    •   While LDS flow rates tend to decrease with time, Ea values may increase or
       decrease with time depending on the relative rates of decrease of LCRS flow
       versus LDS flow.  For example, Ea values for Cell B3 were initially in the range of
       96 to 99%, but decreased to about 80% after 5 to 8 years due to steady LDS
       flow rates and decreasing LCRS flow rates.  For one cell (AD1), Ea values
       became negative.
    •   The highest value of Ea was achieved by Cell AM2  at a time period of about four
       years after the start of cell operations. For this cell, an Ea value of 100% was
       achieved under a condition  of very low rates of LCRS flow (i.e., <100 Iphd)
       during a period when overlying waste placement had  ceased.  Interestingly, the
       AM landfill is located in the W in a semi-arid environment. Leachate generation
       rates at this MSW landfill are, on average, an order of magnitude lower than the
       rates for the other facilities,  all of which are  located in the much wetter climate of
       the eastern U.S.
    •   The only GM/CCL or GM/GCL/CCL composite primary liner for which primary
       liner leakage rates and Et values can be estimated  from LDS flow rate data
       alone is for Cell AM2.  For this facility, zero leakage and an Et value of 100%
       was achieved for the low LCRS flow rate conditions noted above. Due to the
       interfering effects of consolidation water, LDS flow data for the other facilities do
       not alone allow conclusions on Et values to  be drawn. LCRS and LDS chemical
       constituent data are reviewed below to develop further insight into Et values for
       GM/CCL and GM/GCL/CCL composite liners.

E-4.3.3.3 Analysis of Chemical Data

Concentrations of chemical constituents in LDS liquids are compared to concentrations
of the same constituents in LCRS liquids for cells with GM/CCL or GM/GCL/CCL
composite liners in Table E-4.9. As indicated by the data in Table E-4.9,  the general
water quality characteristics of LDS liquids are different than the corresponding
characteristics for the LCRS liquids. This is due to the different origins of the primary
sources of the two liquids, leachate for the LCRS liquids and  CCL pore water for the
LDS liquids.  The different origins of the two liquids are reflected in different major  ion
chemistries, as well as differences in COD, BOD and TOC concentrations. Figures E-
4.2 through E-4.6 present Piper (1944) trilinear diagrams for six of the cells listed in
Table  E-4.9.  These diagrams demonstrate the different primary sources of LCRS and
LDS liquids from a given cell. Details for the construction and interpretation of Piper
trilinear diagrams can be found in most hydrogeology and geochemistry textbooks.

To further evaluate whether primary liner leakage had contributed to the observed  LDS
flows, the concentrations of five key chemical constituents  in  the LCRS and LDS flows
were investigated. Several factors  were considered in selecting the key constituents: (i)
common occurrence in leachate; (ii) high solubility in water and low octanol-water
coefficient; and (iii) high resistance  to hydrolization  and anaerobic biodegradation in soil.
Several chemicals that were considered as candidates due to their aqueous solubility
characteristics, notably several alcohols and ketones and methylene chloride, could not
                                  E-144
-------
     o  o  o   o  o
     88888
TOTAL DISSOLVED SOLIDS
    (Parts Per Million)
                          LEGEND

                        • 1 AK1-LCRS
                        o 2 AK1-LDS
 Ca
        80
             Calcium (Ca)

           CATIONS
                           20
No+K HC03+C03 20

  %meq /1
  40	^60

 Chloride (Cl)

A N I  O N  S
                                                                  80
          Figure E-4.2.  Piper trilinear diagram for cell AK1.
                                   E-145
-------
TOTAL DISSOLVED SOLIDS
    (Parts Per Million)
                          LEGEND

                        • 1  AL1-LCRS
                        o 2  AL1-LDS
 Ca
        80
             Calcium (Ca)

           CATIONS
                           20
No+K HCO3+CO3 20

  %meq /1
  40	^60

 Chloride (Cl)

A N I  O N  S
                                                                  80
          Figure E-4.3.  Piper trilinear diagram for cell ALL
                                   E-146
-------
TOTAL DISSOLVED SOLIDS
    (Parts Per Million)
                          LEGEND

                          1  AM1-LCRS
                        o 2  AM1-LDS
 Ca
        80
             Calcium (Ca)

          CATIONS
                           20
No+K HC03+C03 20

  %meq /1
  40	^60

 Chloride (Cl)

A N I O  N S
                                                                  80
          Figure E-4.4. Piper trilinear diagram for cell AM1.
                                  E-147
-------
TOTAL DISSOLVED SOLIDS
    (Parts Per Million)
                           LEGEND

                         • 1  AM2-LCRS
                         o 2  AM2-LDS
 Ca
        80
             Calcium (Ca)

          CATIONS
                           20
No+K HC03+C03 20

  %meq /1
  40	^60

 Chloride (Cl)

A N I O  N S
                                                                  80
          Figure E-4.5. Piper trilinear diagram for cell AM2.
                                  E-148
-------
     o  o  o  o  o
     88888
TOTAL DISSOLVED SOLIDS
    (Parts Per Million)
                         LEGEND

                        • 1 AO-LCRS
                        o 2 A01-LDS
                        x 3 A02-LDS
 Ca
        80
             Calcium (Ca)

          CATIONS
                           20
No+K HC03+C03 20

  %meq /1
  40	^60

 Chloride (Cl)

A N I O  N
                                                                 80
     Figure E-4.6.  Piper trilinear diagram for cell AO1 and AO2.
                                  E-149
-------
     o  o  o   o  o
     88888
TOTAL DISSOLVED SOLIDS
    (Parts Per Million)
                          LEGEND

                        • 1  AR1-LCRS
                        o 2  AR1-LDS
 Ca
        80
             Calcium (Ca)

           CATIONS
                           20
No+K HC03+C03 20

  %meq /1
  40	^60

 Chloride (Cl)

A N I  O N S
                                                                  80
          Figure E-4.7.  Piper trilinear diagram for cell AR1.
                                   E-150
-------
be used due to poor representation in the database or potential for presence in the
database as a laboratory contaminant. The five selected key constituents are the
inorganic anions sulfate and chloride and the aromatic hydrocarbons benzene, toluene,
and xylene.  The results of the comparison of key constituents are presented in Tables
E-4.11 and E-4.12. Table E-4.11 presents the concentrations of the five key constituents
as a function of time period after the start of landfill cell operation. These time periods
roughly correspond to the monitoring time periods given in Table E-4.8.  Table E-4.12
presents the results of the authors' assessment of the occurrence of key constituent
migration through  the composite primary liners. This assessment is based on a
qualitative comparison of the five key chemical constituents previously identified. Table
E-4.12 also presents an estimate of the advective breakthrough time for the CCL or
GCL/CCL component of each composite primary liner. The estimated breakthrough
times were calculated assuming that the GM component of the composite primary liner
has one or more holes through which leachate instantaneously migrates and that
leachate migration through the CCL or GCL/CCL component of the composite liner is
governed by Darcy's equation  assuming one-dimensional steady-state saturated flow.
Other assumptions used in the calculations are given in  the table. The effect of
chemical retardation was not considered in calculating the advective breakthrough
times. Retardation of chloride  and sulfate should  be negligible. Retardation
characteristics for benzene, toluene, and xylene will depend on the organic carbon
content of the CCL or GCL, redox conditions, and other factors.  It is expected, however,
that the effective retardation coefficient for these constituents would have been 2 or
more. These organic compounds were chosen for analysis notwithstanding their
retardation characteristics for a combination of reasons,  including relatively widespread
occurrence in leachate, and  relatively higher concentrations in leachate than other
organic compounds. In addition, these three constituents are not known as laboratory
contaminants, in contrast to methylene chloride, a constituent that is more mobile but is
also a common laboratory contaminant.

The current database is not sufficient to draw definitive conclusions on the performance
of GM/CCL and GM/GCL/CCL composite liners. However, using the data and
comparisons in Tables E-4.11 and E-4.12, the following observations can be offered with
respect to key chemical constituent migration through the composite primary liners:

    •  Three of the 13 considered landfill cells with GM/CCL or GM/GCL/CCL
      composite primary liners (i.e., Y2, AQ1, and AQ10) have insufficient chemical
      data to draw any conclusions on primary liner leakage rates based on the key
      chemical constituent data. The three cells exhibited average monthly LDS flow
      rates of 18 to 215 Iphd  for the active life and post-closure periods, with most
      values being less than  100 Iphd.  These flow rates are comparable to rates for
      the other cells included in the database for similar operational stages (except
      cells AM1 and AM2, located in a semi-arid climate).  Based on this observation,
      and on calculation results for long-term consolidation and secondary
      compression (see Gross et al. (1990)), the observed flows are attributed
      primarily to consolidation water and not primary liner leakage.  The potential
                                  E-151
-------
           Table E-4.11. Average Concentrations of Five Key Chemicals in LCRS and LDS Flows from Landfill

                        Cells with Composite Top Liners.
m
en
ro

Cell
No.
B3








Y2




AK1
AL1


AM1





AM2





Time
Period
(months)
1-4
5-16
17-28
29-40
41-52
53-64
65-76
77-88
89-93
1-10
11-22
23-34
35-46
47-54
1-12
1-29
30-41
42-54
1-9
10-21
22-33
34-45
46-57
58-69
1-9
10-21
22-33
34-45
46-57
58-69
Chemical'1'
Sulfate (mg/l)
LCRS
282
105
348
104
47
28
<127
<6

108
108
52


47
300
225
247
51
<2
<16
<27
<12
<3
96
<2
<13
<7
<2
<3
LDS
95
1286

500
14
123
90
301
48
231
299
326


16
1030
900
1375


1341
1200




1032
1300
1730
2405
Chloride (mg/l)
LCRS
25
207
352
580
355
899
203
998
1383
349
590
876


104
330
273
400
77
120
159
219
265
240
140
290
353
326
368
262
LDS
19
173
241
118

59
58


60
25
89


2
89
203
215


2260
2600




2175
2600
2700
2635
Benzene (ug/l)
LCRS
<11<2>
<1
<1
<5
8
7
<5
<5
<2

10



<5
<4
1
2
<21
18
19
18
10
13
11
17
19
<18
15
7
LDS
<25
<1

<5
6
<1
<5
<5
<5





<1
<2
<1
<1

<1
<1
<1



<1
<1
<1
<1
<1
Toluene (ug/l)
LCRS
150
<1
<1
354
233
101
14
<5
<1

720



88
133
5
2
219
160
336
290
199
56
22
89
266
286
148
65
LDS
<25
<1

<6
24
<1
<6
<4
<1

7



<1
<1 <3>
<1
<1

2
<1
<1



<1
<1
<1
<1
<1
Xylene (ug/l)
LCRS














30
540
<3
<3
150
90
121
122
90
82
34
57
95
94
<115
50
LDS














<3
<1

<1

<1
<1
<3



<2
<1
<3
<2
<2
-------
           Table E-4.11. Average Concentrations of Five Key Chemicals in LCRS and LDS Flows from Landfill

                        Cells with Composite Top Liners (Continued).
m


en
CO

Cell
No.
AO1



AO2


AR1


AD1










AD7







Time
Period
(months)
1-5
6-17
18-29
30-37
1-5
6-17
18-31
1-11
12-23
24-36
1-12
13-20
21-32
33-44
45-51
52-63
64-75
76-87
88-99
100-111
112-121
1-12
13-24
25-36
37-48
49-60
61-69
70-81
82-87
Chemical
Sulfate (mg/l)
LCRS

49
35
69

49
35
180
440
520
6353
5830
5470
4455
2223


1785
4488
3633
3870
2818
3620
7361
8213
6867
5740
6998
7480
LDS

88
41
2

89
93
600
170
265


480
498
308
443
338
456
339
296
369
340
683
586
954
1050
1148
1168
1132
Chloride (mg/l)
LCRS

930
988
570

930
988
1000
2200
1650
3930
24300
10763
11590
13960


13900
14550
14075
14800
3214
9550
10720
11535
14400
15775
12875
14267
LDS

58
27
46

16
24
49
8
41

289
210
185
217
240
131
377
137
114
138
109
216
219
469
418
387
387
357
Benzene (ug/l)
LCRS

6
10
<5

6
10


<100
492
429
33
5
35



26
18
<25
140
612
1168
644
778
687
540
<240
LDS

<1
<1
<5

<1
<1

<1
<50

<4
<4
<4
<4
<4
<4
<4
<4
<1
<1
<4
<4
<4
<4
<4

<4
<1
Toluene (ug/l)
LCRS

230
288
77

230
288


<100
305
292
133
60
108
<300


186
<30
<25
317
892
1859
2960
1660
1288
906
450
LDS

<1
44
<5

<1
<1

<2
<50

<6
<6
<6
<6
<6
<6
<6
<6
<1
<1
<6
<6
<6
<6
<6
<6
<1
<1
Xylene (ug/l)
LCRS

59
45
21

59
45


<100



















LDS

<3
<3
<10

<3
<4

<4
<50



















-------
m
en
Table E-4.11. Average Concentrations of Five Key Chemicals in LCRS and LDS Flows from Landfill
                Cells with Composite Top  Liners (Continued).

Cell
No.
AQ1




AQ10




Time
Period
(months)
1-58
59-65
66-77
78-89
90-97
1-15
15-26
27-38
39-50
51-63
Chemical
Sulfate (mg/l)
LCRS










IDS










Chloride (mg/l)
LCRS










IDS










Benzene (ug/l)
LCRS

<5
<8
<12
<5

<10
<10
<6
<5
LDS

<4
<4
<4
<4

<4
<4
<4
<4
Toluene (ug/l)
LCRS

<5
<5
<5
<5

<12
<30
<6
5
LDS

<10
<6
<14
<5

<5
<5
<5
<5
Xylene (ug/l)
LCRS


<5
<8
<5


190
<7
10
LDS










Notes:
(1) Reported concentrations represent average of 1 to 17 individual analysis results (typically on the order of 5) during incremental
    reporting period.
(2) Data preceded by "<" indicates more than half or more of the measurements for the parameter were reported as non-detects; in
    calculating average values, half of the test detection limit was conservatively used for all results reported as non-detects.
(3) For Cell AL1, toluene was not detected in nine often LDS flow samples obtained during the 1-41 months time period. Toluene was
   detected at a concentration of 91 |ig/l in month 30. This one detection is attributed to sampling or analysis error.
-------
         Table E-4.12.  Evaluation of Chemical Constituent Migration Through Landfill GM/CCL and

                     GM/GCL/CCL Composite Primary Liners.

Cell
No.



B3







Y2










Monitor.
Period
(months)


93







54 (no key
chemical
data after
34
months)







Estimated
Advective
Breakthr.
Time for
GCL/CCL
fmonthsf
46







35









Chemical
Sulfate




not diagnostic
due to
fluctuating Co(2)
in both LCRS
and LDS



not diagnostic
due to high 0,
in LDS consol.
water








Chloride




lower C0 in LDS
than in LCRS
and trend of
decreasing LDS
C0 with time not
indicative of
chloride
breakthr.
in LCRS, C0 =
170 to 1,160
mg/l with m(2) =
628 mg/l; in
LDS, C0 = 8 to
140 mg/l with m
C O nn « /I • n «
= 58 mg/l; no

indication of
chloride
breakthr.
Benzene




not diagnostic
due to very low
C0 in both
LCRS and LDS
(i.e., C0 almost
always below
DL(3) of 5 |ig/l)

no LDS data
available









Toluene




in LCRS, C0 up
to 700 |ig/l; in
LDS, C0 typically
below DL of 1 to
10 |ig/l; no
indication of
toluene breakthr.

only one C0
available from
each system (at
11 -22 months):
LCRS C0 = 720
|ig/l and LDS C0

~ / U-Q/I




Xylene




no data
available






no data
available










Summary of
Observations for Five
Key Constituents


no evidence of
significant leachate
migration into LDS
after almost 8 years of
cell operation, twice
the estimated CCL
breakthr. time

data are insufficient to
draw conclusions;
monitor, period is
about equal to
estimated CCL
breakthru. time; more
chemical data are

n^^H^H
1 ICCUCU


m
en
en
-------
         Table E-4.12.  Evaluation of Chemical Constituent Migration Through Landfill GM/CCL and

                     GM/GCL/CCL Composite Primary Liners (Continued).

Cell
No.



AK1








AL1








Monitor.
Period
(months)


12








54







Estimated
Advective
Breakthr.
Time for
GCL/CCL
fmonthsf
48








70







Chemical
Sulfate




in LCRS, C0 = 7
to 1 1 0 mg/l with
m = 47 mg/l; in
LDS, C0= 10 to
51 mg/l with m
= 16 mg/l; no
indication of
sulfate breakthr.

not diagnostic
due to high Q,
in LDS consol.
water





Chloride




in LCRS, C0 = 2
to 230 mg/l with
m = 104 mg/l;
in LDS, C0 = 2
to 6 mg/l with m
= 4 mg/l; no
indication of
chloride
breakthr.
increasing LDS
C0 with time
likely due to
decreasing
dilution of
consol. water by
construct, water


Benzene




not diagnostic
because C0 is
below DL in
both LCRS and
LDS




not diagnostic
because C0 is
below DL in
both LCRS and
LDS




Toluene




in LCRS, C0 = 5
to 300 |ig/l with
m = 88 |ig/l; in
LDS, C0 below
DL of 1 |ig/l; no
indication of
toluene breakthr.


in LCRS, C0 = up
to 600 |ig/l; in
LDS, toluene
below DL of 1
|ig/l; no
indication of
toluene breakthr.


Xylene




in LCRS, xylene
detected in half
of sampling
events at C0 up
to 79 |ig/l; in
LDS, C0 below
DLof3|ig/l; no
indication of
xylene breakthr.
not diagnostic
because C0 is
below DL in
LDS and in
LCRS too after
29 months




Summary of
Observations for Five
Key Constituents


no evidence of
significant leachate
migration into LDS;
however, monitor.
period is only about
1/4th of the estimated
GCL/CCL breakthr.
time; more chemical
data are needed
no evidence of
significant leachate
migration in to LDS;
monitor, period
somewhat less than
estimated CCL
breakthr. time; more
chemical data are
needed
m
en
en
-------
         Table E-4.12.  Evaluation of Chemical Constituent Migration Through Landfill GM/CCL and

                     GM/GCL/CCL Composite Primary Liners (Continued).

Cell
No.



AM1








AM2








Monitor.
Period
(months)


58








58







Estimated
Advective
Breakthr.
Time for
GCL/CCL
fmonthsf
4








4







Chemical
Sulfate




not diagnostic
due to high 0,
in LDS consol.
water





not diagnostic
due to high 0,
in LDS consol.
water





Chloride




not diagnostic
due to high 0,
in LDS consol.
water





not diagnostic
due to high 0,
in LDS consol.
water





Benzene




in LCRS, C0 =
12to20|ig/l; in
LDS, C0 below
DL of 1 |ig/l; no
indication of
benzene
breakthr.


in LCRS, C0 =
5 to 20 |ig/l; in
LDS, C0 below
DL of 1 |ig/l; no
indication of
benzene
breakthr.


Toluene




in LCRS, C0 =
40 to 420 |ig/l
with m = 267
|ig/l; in LDS, C0
below DL of 1
uq/l; no
r"\y' i **
indication of
toluene breakthr.

in LCRS, C0 =
1 0 to 400 |ig/l
with m = 1 46
|ig/l; in LDS, C0
below DL of 1
uq/l; no
r"\y' i **
indication of
toluene breakthr.

Xylene




in LCRS, C0 =
71 to150|ig/l
with m = 1 22
|ig/l; in LDS, C0
below DL of 1 to
3 |ig/l; no
indication of
xylene breakthr.

in LCRS, C0 = 2
to 1 30 |ig/l with
m = 71 mg/l; in
LDS, C0 below
DL of 1 to 3
mg/l; no
indication of
xylene breakthr.


Summary of
Observations for Five
Key Constituents


no evidence of
significant leachate
migration into LDS
after almost 5 years of
cell operation; monitor
period more than 12
times longer than
estimated CCL
breakthr. time
no evidence of
significant leachate
migration into LDS
after almost 5 years of
cell operation; monitor
period more than 12
times longer than
estimated CCL
breakthr. time
m
en
-•J
-------
         Table E-4.12.  Evaluation of Chemical Constituent Migration Through Landfill GM/CCL and

                     GM/GCL/CCL Composite Primary Liners (Continued).

Cell
No.



A01(4)









A02(4)










Monitor.
Period
(months)


37









31









Estimated
Advective
Breakthr.
Time for
GCL/CCL
fmonthsf
140









145









Chemical
Sulfate




not diagnostic
due to similar
LCRS and LDS
C0 ranges






not diagnostic
due to similar
LCRS and LDS
C0 ranges







Chloride




in LCRS, C0 =
320 to 1 300
mg/l with m =
860 mg/l; in
LDS, C0 = 7 to
1 00 mg/l with m
= 40 mg/l; no
indication of
chloride
breakthr.

in LCRS, C0 =
320 to 1 ,300
mg/l with m =
862 mg/l; in
LDS, C0 = 3 to
34 mg/l with m
= 24 mg/l; no
indication of
chloride
breakthr.

Benzene




in LCRS,
benzene
detected in half
of the sampling
events at C0 =
7 to 12|ig/l; in
LDS, C0 below
DLof 1 |ig/l; no
indication of
benzene
breakthr.
in LCRS,
benzene
detected in half
of the sampling
events at C0 =
7 to 12|ig/l; in
LDS, C0 below
DLof 1 |ig/l; no
indication of
benzene
breakthr.
Toluene




in LCRS, C0 =
1 0 to 550 |ig/l
with m = 167
|ig/l; in LDS, C0
below DL of 1
|ig/l in 2/3 of
sampling events;
no indication of
toluene breakthr.


in LCRS, C0 =
1 0 to 550 |ig/l
with m = 167
|ig/l; in LDS, C0
below DL of 1
UQ/I' no
f y , IV
indication of
toluene breakthr.



Xylene




in LCRS, C0 =
12to76|ig/l
with m = 34|ig/l;
in LDS, C0
below DLof 3
|ig/l; no
indication of
xylene breakthr.



in LCRS, C0 =
12to76|ig/l
with m = 34|ig/l;
in LDS, C0
below DLof 3
|ig/l; no
indication of
xylene breakthr.




Summary of
Observations for Five
Key Constituents


no evidence of
significant leachate
migration into LDS
after 3 years of cell
operation; however,
monitor, period is only
1/4th of estimated
CCL breakthr. time;
more chemical data
are needed

no evidence of
significant leachate
migration into LDS
after 3 years of cell
operation; however,
monitor, period is only
1/4th of estimated
CCL breakthr. time;
more chemical data
are needed

m
en
oo
-------
        Table E-4.12.  Evaluation of Chemical Constituent Migration Through Landfill GM/CCL and

                    GM/GCL/CCL Composite Primary Liners (Continued).

Cell
No.



AR1









AD1











Monitor.
Period
(months)


36









121










Estimated
Advective
Breakthr.
Time for
GCL/CCL
fmonthsf
2









70










Chemical
Sulfate




not diagnostic
due to similar
LCRS and LDS
C0 ranges






in LCRS, C0 =
1,785 to 6,353
mg/l with m =
4,234 mg/l; in
LDS, C0 = 296
to 498 mg/l with
m = 392 mg/l;
no indication of
sulfate breakthr.


Chloride




in LCRS, C0 =
600 to 2700
mg/l with m =
1625 mg/l; in
LDS, C0 = 8 to
74 mg/l with m
= 35 mg/l; no
indication of
chloride
breakthr.
in LCRS, C0 =
3,930 to 24,300
mg/l, with m =
13,450 mg/l; in
LDS, C0 = 114
to 337 mg/l,
with m = 204
mg/l; no
indication of
chloride
breakthr.
Benzene




not diagnostic
because C0 is
below DL in
both LCRS and
LDS





in LCRS, C0 =
<25to492|ig/l;
in LDS, C0
below DL of 1
to 4 |ig/l; no
indication of
benzene
breakthr.



Toluene




not diagnostic
because C0 is
below DL in both
LCRS and LDS






in LCRS, C0 =
<25to305|ig/l;
in LDS, C0 below
DLof 1 to6|ig/l;
no indication of
toluene breakthr.





Xylene




not diagnostic
because C0 is
below DL in
both LCRS and
LDS





no data
available










Summary of
Observations for Five
Key Constituents


no evidence of
significant leachate
migration into LDS;
monitor, period is
more than 10 times
the estimated
GCL/CCL breakthr.
time; data are not
diagnostic; more data
are needed
no evidence of
significant leachate
migration into LDS
after 10 years of cell
operation and closure,
1 .7 times more than
the estimated CCL
breakthr. time



m
en
tQ
-------
         Table E-4.12.  Evaluation of Chemical Constituent Migration Through Landfill GM/CCL and

                     GM/GCL/CCL Composite Primary Liners (Continued).

Cell
No.



AD7
















Monitor.
Period
(months)


87















Estimated
Advective
Breakthr.
Time for
GCL/CCL
fmonthsf
70















Chemical
Sulfate




in LCRS,
C0=2,818to
8,213 mg/l with
m=6,137 mg/l;
in LDS, C0=340
to 1,168 mg/l
with m=882
mg/l; increasing
LDS C0 after 36
months
attributed to
decreasing
dilution of
consol. water
by construct.
water
Chloride




in LCRS,
C0=3,214to
15, 775 mg/l
with m=1 1 ,547
mg/l; in LDS,
C0=109to469
mg/l with
m=320 mg/l;
increasing LDS
C0 after 36
months
attributed to
decreasing
dilution of
consol. water by
construct, water
Benzene




in LCRS, C0 =
<240to 1,168
|ig/l; in LDS,
C0 below DL of
1 to 4 |ig/l; no
indication of
benzene
breakthr.








Toluene




in LCRS, C0 =
31 7 to 2,960
|ig/l; in LDS, C0
below DL of 1 to
6 |ig/l; no
indication of
toluene breakthr.









Xylene




no data
available















Summary of
Observations for Five
Key Constituents


evidence of possible
breakthr. for sulfate &
chloride at 12-36
months; authors
attribute trend to
decreased dilution of
consol. water by
construct, water; no
evidence of organic
constituent breakthr.;
more chemical data
are needed




m
en
o
-------
           Table E-4.12.   Evaluation of Chemical Constituent Migration Through Landfill GM/CCL and
                          GM/GCL/CCL Composite Primary Liners (Continued).

Cell
No.



AQ1




AQ10




Monitor.
Period
(months)


97




63



Estimated
Advective
Breakthr.
Time for
GCL/CCL
fmonthsf
35




35



Chemical
Sulfate




no data
available



no data
available



Chloride




no data
available



no data
available



Benzene




not diagnostic
because C0 is
below DL in
both LCRS and
LDS
not diagnostic
because C0 is
below DL in
both LCRS and
LDS
Toluene




not diagnostic
because C0 is
below DL in both
LCRS and LDS

not diagnostic
because C0 is
below DL in both
LCRS and LDS

Xylene




no LDS data
available



no LDS data
available




Summary of
Observations for Five
Key Constituents


data are insufficient to
draw conclusions;
more data are needed


data are insufficient to
draw conclusions;
more data are needed


m
en
           Notes:

           (1) Advective breakthrough times for steady-state saturated flow through CCL or GCL/CCL component of composite liners were calculated
              using Darcy's equation and specified hydraulic conductivities, hydraulic gradient of 5 for GCLs and 1 for CCLs, and effective porosity of 0
              For this calculation, it is assumed that flow through the GM component of the composite liner occurs through small holes and is instantan
           (2) C0 = average concentration during incremental reporting period; m = mean concentration for the entire reporting period.
           (3) DL = detection limit.
           (4) Composite liquid quality samples were collected from the LCRSs of Cells AO1  and AO2; these samples are assumed to represent
              average conditions at the two cells.
-------
occurrence of primary liner leakage cannot be ruled out, however, because of
insufficient chemical data.
Of the ten remaining landfill cells considered in this study, none exhibited
obvious evidence of primary liner leakage.  One of these cells (AD7) exhibited a
potential indication of primary liner  leakage when sulfate and chloride
concentrations in LDS flows increased between 12 and 36 months after
construction. However, the concentrations of other chemicals did not increase
over time. The estimated breakthrough time for the composite primary liner is
70 months, several times greater than the time when sulfate and chloride
concentrations increased. The increase in concentration of these anions without
an increase in concentration of the  organic chemicals would suggest that the
source of the LDS flow is not leakage and that the increasing anion
concentrations could be attributed to decreasing dilution of consolidation water
over time by construction water.  However, Cell AD7 has a GN LDS, which
should not release much construction water. Also, if construction water was
significant, a similar trend would  be expected for Cell AD1 from the same facility.
Though Cell AD1 has a sand LDS,  which can release significant amounts of
water during cell operations, a similar trend was not observed.  Thus, the reason
for the  increase in anion  concentrations for Cell AD7 is unclear.
Five of the cells in this study (B3, AD1, AD7, AM1, and AM2) have key
constituent data of sufficient completeness and duration to conclude that
leachate migration into the  LDS at a rate of any engineering significance has not
occurred for a time  period exceeding the estimated breakthrough time for the
CCL component of the composite liner. Cell B3 has a monitoring period of
almost 8 years,  about twice the estimated CCL breakthrough time. The LDS
flow from Cell B3 exhibited  a general trend of decreasing chloride concentrations
with time while chloride concentrations in the LCRS tended to increase. Also,
while toluene concentrations as high as 700 jig/l were detected in LCRS flow,
there was no clear indication of toluene breakthrough into the LDS.  Cells AM1
and AM2 have a monitoring period  of almost 6 years, approximately 17 times
longer than the estimated CCL breakthrough time.  Concentrations of benzene,
toluene, and xylene in the leachate are generally in the 10 to 300 jig/l range
throughout the monitoring period; LDS concentrations for these three key
constituents are below the detection limit of 1 to 3 jig/l in all cases. Cell AD1  has
a monitoring period of 10 years, approximately 70% longer than the estimated
CCL breakthrough time.  Sulfate, chloride, benzene, and toluene data all exhibit
relatively high concentrations in the LCRS of this HWcell, with no indication of
breakthrough for any of these constituents. Cell AD7 has a monitoring  period of
about seven years with no indication of benzene or toluene breakthrough.
Et values can be estimated for Cells B3, AM1, AM2, AD1, and AD7 by
substituting the flow rate terms in Equation E-1 with constituent mass fluxes from
the LCRS and LDS. Mass fluxes were calculated using average flow rates and
chemical concentrations for benzene, toluene, and xylene during the active
operation and post-closure  stages.  The precision of this approach is limited by
the detection limits of the chemical  analytical methods used.  For the
                           E-162
-------
       calculations, key constituent concentrations were taken as one-half of the
       reported value where preceded by "<" in Table E-4.11.  The approach also
       assumes that all key constituents are conserved in solution during transport
       through the composite primary liner and that all of the constituent mass in the
       LDS is from the LCRS. Using this approach, the Cell B3 composite liner
       achieved an average Et of 99.8% through the active and post-closure period,
       Cells AM1 and AM2 achieved average Et values in excess of 99.9 and  99.5%,
       respectively, and Cells AD1  andAD7 achieved average Et values of 99.1 and
       greater than 99.9%, respectively.

E-4.3.4  Comparison to Liner Leakage Calculation Results

Giroud and Bonaparte (1989), and Giroud et al. (1992) presented equations for
calculating steady-state leakage rates through holes in the GM component of GM/CCL
and GM/GCL composite liners, respectively. Input parameters to the equations are the
hydraulic head acting on the liner, saturated hydraulic conductivity of the CCL  or GCL,
number and size (area) of holes in the GM, and average hydraulic gradient across the
GM.  Calculation results for a range of conditions are given in Table E-4.13.  Calculation
results are given for two different conditions of contact between the GM and CCL or
GCL components of the composite liner: (i) good contact; and (ii) poor contact. These
contact conditions are described in Table E-4.13.  The calculated leakage rates can be
compared to the observed LDS flow rates for cells with composite primary liners
summarized above:

   •  LDS flow rates reported in this appendix for the active life and post-closure
       period of landfill cells with GM/GCL composite primary liners (except Cell 14) are
       in the range of 0 to 10 Iphd, with most values less than 2 Iphd and many values
       reported as zero (Table E-4.7). From Table E-4.13, calculated steady-state
       leakage  rates for this type of composite liner are in the range  of 0.007 to 4 Iphd.
   •  True liner efficiencies (Et) may exceed 99.9% for GM/CCL and GM/GCL
       composite  liners, particularly at higher LCRS flow rates. This value of  Et would
       result in  composite liner leakage rates of 0.1, 1, and 10 Iphd, respectively, for
       LCRS flow rates of 100, 1,000, or 10,000 Iphd. These primary liner leakage
       rates can be compared to calculated values of 0.05 to 5 Iphd for GM/CCL
       composite  liners and 0.007 to 4 Iphd for GM/GCL composite liners.

The calculated leakage rates presented in Table E-4.13 are in the same range as the
primary liner leakage rates estimated from the data presented in this  appendix. These
results do not in themselves validate the liner leakage calculation models due  to the
assumptions and limitations in calculation model development. Also, there is no way to
know a priori when a cell liner will have a hole through which leakage can occur and
where the hole will be located, e.g., at a critical location such as the sump, or at a less
critical  location in upgradient portions of the cell.  However, the fact that the calculations
and monitoring data provide a consistent interpretation of composite liner hydraulic
performance is a useful finding.
                                  E-163
-------
Table E-4.13.    Calculated Steady-State Leakage Rates Through Composite Liners (in Iphd).
                                                                                       d)
Composite Liner
Type
GM/CCL
(Poor contact conditions)'2'
GM/CCL
(Good contact conditions)'3'
GM/GCL
(Poor contact conditions)
GM/GCL
(Good contact conditions)
CCLorGCL
Hydraulic
Conductivity (m/s)
1 x10'9
1x10-9
5x10-11
5x10-11
Hydraulic Head (m)
0.01
0.3
0.05
0.04
0.007
0.03
0.8
0.2
0.1
0.03
0.1
2
0.4
0.7
0.1
0.3
5
1
4
0.7
Notes:      (1)   Calculations are based on 2.5 GM holes per hectare. Area of each hole assumed to
                 be 10 mm2.  Calculations performed using equations from Giroud and Bonaparte
                 (1989) and Giroud et al. (1992).
           (2)   Good contact conditions correspond to a GM installed with relatively few wrinkles,
                 placed on top of a CCL that has been adequately compacted and has a smooth
                 surface, or a GCL with a smooth surface.
           (3)   Poor contact conditions correspond to a GM installed with a certain number of
                 wrinkles, placed on top of a CCL that has not been adequately compacted and does
                 not appear smooth, and/or placed on a GCL that is wrinkled.
                                       E-164
-------
E-4.3.5  Implications for Landfill Performance

The evaluation results presented in this section are encouraging with respect to the
types of composite liners used widely in landfills throughout the U.S. Leakage rates
through these types of composite liners appear to be very low, typically less than 2 Iphd
for GM/GCL composite liners. Conversely, Et values appear to be very high.  Et values
of 99.9% or more, particularly at higher rates of LCRS flow,  are achievable. Coupling
this conclusion with that from Section E-4.2 that the frequencies of holes in GMs
decrease when third-party CQA programs are employed in landfill  cell construction, and
also with available guidance for the installation of geosynthetic and natural soil liner
components (e.g., Daniel and Koerner (1995)),  provides a framework for reliably
achieving high levels of composite liner performance.

It is also noteworthy that composite liners with CCL components have advective
transport breakthrough times that are long relative to the timeframe for leachate
generation  in a modern landfill cell operated to:  (i) minimize leachate generation; (ii)
remove leachate as it is  generated; and (iii) "close as you go" using a final cover system
that limits the infiltration of precipitation into the  landfill. In the U.S., landfill cells are
typically operated for periods of one to five years, occasionally longer, and they are
promptly covered with a GM or other low-permeability barrier after filling. This
operations  sequence defines the timeframe for significant leachate generation in a
landfill cell that does not contain liquid wastes or sludges and that  does not undergo
leachate recirculation or moisture addition. For the cells in this study, estimated
advective breakthrough times through CCLs, assuming no chemical retardation, were
generally calculated to range from about 3 to 12 years. It thus appears that for modern
facilities that are  properly constructed and operated, GM/CCL and GM/GCL/CCL
composite liners  can prevent leachate leakage of any engineering significance through
the liner over the entire time period of leachate generation.  For cells with GM/GCL
liners, breakthrough times are anticipated to be faster than for cells with GM/CCL liners.

E-5    Leachate Generation Rates

E-5.1    Overview

The rate of leachate generation in a landfill is expected to be a function of several
factors including: (i) the operational stage of the landfill (i.e., initial  period of operation
vs. active period  of operation vs. post-closure period); (ii) the type  of waste placed in the
landfill (MSW vs. HW vs. ISW); (iii) the condition of the waste at the time of placement
(i.e., moisture content and density); (iv) the geographical region of the U.S. in which the
landfill is located (i.e., NE vs. SE vs. W);  and (v) waste placement  practices (i.e., size of
the active waste disposal area, characteristics of daily and intermediate covers, and
extent of storm water diversion away from the landfill active area).  These factors can
have significant impacts on  the volume of leachate generated from a landfill, and
therefore, on potential impacts to the surrounding environment and on the costs of
leachate collection and treatment. In  this section,  leachate generation rates are
                                   E-165
-------
evaluated for the landfills included in the current database and some of the factors that
influence leachate generation rates are identified.  It is noted that a few of the factors
mentioned above which may affect leachate generation rates can not be evaluated
using the current database (e.g., waste placement practices).  Thus, this section
concentrates on evaluating the effect of landfill operational stage, waste type, and
geographic region on leachate generation rates.

E-5.2    Description of Data

Average and peak monthly LCRS flow rates are provided in Table E-3.5 for the initial
and active periods of operation and for the post-closure period. The LCRS flow rates
are presented for approximately twelve-month increments. Tables E-5.1 through E-5.5
include summaries of the LCRS flow rates organized by type of waste and region of the
landfill.  Tables E-5.1 through E-5.4 include data for the initial and active periods of
operation for MSW,  HW, ash, and C&DW landfills, respectively.  Table E-5.5 includes
data for the post-closure period for MSW and HW landfills.

Data are available during operation conditions for 73 individually monitored MSW cells
from 32 landfills,  56 HW cells from twelve landfills, eight ash cells from six landfills, and
three C&DW cells from two landfills. Most of the 50 landfills considered are located  in
the NE (32 landfills), and only four are located in the W.  The available data covers
between approximately one and eight years of landfill operation. Almost half of the cells
had more than four years of LCRS flow rate data available.  Table E-5.5 includes LCRS
flow rate data for the post-closure period for eleven MSW cells at three landfills and  22
HW cells at five landfills. The data represents a maximum  of six and nine post-closure
years for MSW and  HW cells, respectively. Half of the eight landfills are located in the
NE and only one is located in the W (Landfill AT).

E-5.3    Analysis of Data

Tables E-5.1 through E-5.4 include summaries of data on LCRS flow rates measured
during operations of 140 individually monitored cells located at 50 landfills. Average
and peak monthly flow rates  are reported for the initial and active periods of operation.
The ratio (in %) of average LCRS flow rate to historical average annual rainfall is also
calculated in these tables. This ratio is referred to herein as the rainfall fraction (RF).

For the 44 MSW cells located in the NE for which initial period of operation data are
available, average LCRS flow rates varied between approximately 1,000 Iphd and
40,000 Iphd. Average LCRS flow rates were less than 10,000 Iphd for approximately
60% of these 44  cells and greater than 20,000 Iphd for only 14% of these cells.  During
the active period of operation average LCRS flow rates dropped significantly to a range
of approximately 40 Iphd to 18,000 Iphd.  For the 50 cells for which active period of
operation data are available,  50% had average LCRS flow rates less than 2,500 Iphd,
78% had average LCRS flow rates less than 5,000 Iphd, and only 4% had LCRS flow
rates greater than 10,000 Iphd. Peak monthly LCRS flow rates were typically two to
                                   E-166
-------
Table E-5.1. LCRS Flow Rates for MSW Landfills During Operations.
Cell
No.
Average
Annual
Rainfall
(mm)
LCRS Flow Rates (Iphd) During Operations
Initial Period
Months
Average
Peak
Fraction of
Rainfall (%)
Active Period
Months
Average
Peak
Fraction of
Rainfall (%)
(a)  Landfills located in the NE
B1
B2
B3
B4
B5
E1
E2
E3
E4
F1
G1-2
11
12
13
14
15
J1-6
K1
01
R1
S1
T1-2
Y2
AA1
AA2-3
AE1-2
AG1
AJ1
AK1-2
AL1
AO1
AO2
A03
A04
AR1
AU1
AX1
AX2
AX3
AX4
1,067
1,067
1,067
1,067
1,067
1,041
1,041
1,041
1,041
584
914
991
991
991
991
991
762
1,118
889
1,041
1,041
1,295
1,194
1,143
1,143
1,214
838
1,245
762
787
826
826
826
826
1,143
1,016
1,041
1,041
1,041
1,041
1-19
1-19
1-4
1-12
1-12
1-7
1-12
1-12
1-12
1-12
1-12
1-8
1-7
1-7
1-12
1-12
1-3
1-12
1-7
2-12
1-10
1-8
1-10
2-8
1-13
1-12
1-12
2-6
1-12
1-12
1-5
1-5
1-8
1-8
1-11
1-4
1-2
1-5
1-5
1-12
ND
ND
15,304
2,930
8,005
ND
ND
9,425
20,148
14,472
22,371
ND
6,627
11,559
4,494
3,938
13,363
17,808
39,864
11,592
2,226
2,137
23,368
4,084
14,533
ND
1,780
17,133
9,867
ND
ND
15,881
16,746
20,017
27,042
ND
16,718
15,521
3,361
2,534
ND
ND
24,858
6,353
19,521
ND
ND
25,394
55,785
45,010
46,120
ND
13,959
21,081
17,251
7,985
16,182
24,832
111,129
22,266
5,081
5,982
36,791
9,261
36,777
ND
5,314
24,782
17,983
ND
ND
24,541
53,117
55,470
65,871
ND
19,738
58,674
7,985
12,688


52
10
27


33
71
90
89

24
43
17
15
64
58
164
41
8
6
71
13
46

8
50
47


70
74
88
86

59
54
12
9
20-54
20-54
5-93
13-47
13-35
8-40
13-45
13-14
NA
13-30
13-42
9-84
8-76
8-76
13-26
13-21
4-38
13-66
8-48
13-23
11-45
9-46
11-54
9-51
14-54
60-80
13-33
7-13
13-15
30-54
6-36
6-30
NA
NA
12-36
21-32
3-33
6-33
6-56
13-56
3,816
2,954
3,748
4,022
4,300
9,035
4,979
6,062
NA
8,608
8,204
4,149
728
3,684
2,041
3,108
4,000
9,036
7,263
9,323
1,124
1,552
7,918
1,890
5,870
17,705
3,988
4,728
2,398
1,150
1,485
2,331
NA
NA
10,428
4,991
540
281
307
75
7,464
5,707
22,443
14,641
15,567
36,177
11,014
9,038
NA
25,450
23,485
48,932
3,241
26,339
4,282
1 1 ,669
9,880
27,663
21,862
17,889
4,074
6,804
25,308
5,503
21,520
48,977
12,357
9,936
3,130
5,885
4,130
5,266
NA
NA
26,274
11,597
2,383
570
1,075
187
13
10
13
14
15
32
17
21

54
33
15
3
14
8
11
19
29
30
33
4
4
24
6
19
53
17
14
11
5
7
10


33
18
2
1
1
0
                                       E-167
-------
Table E-5.1. LCRS Flow Rates for MSW Landfills During Operations (Continued).
Cell
No.
Average
Annual
Rainfall
(mm)
LCRS Flow Rates (Iphd) During Operations
Initial Period
Months
Average
Peak
Fraction of
Rainfall (%)
Active Period
Months
Average
Peak
Fraction of
Rainfall (%)
(a)  Landfills located in the NE (continued)
AX5
AX6
AX7
AX8
AX9
AX10
AX11
AX12
AX13
AX14
AX15
AX16
AZ1
1,041
1,041
1,041
1,041
1,041
1,041
1,041
1,041
1,041
1,041
1,041
1,041
762




1-11
1-9
1-10
1-14
1-9
1-7
1-16
1-12
1-7
1-11
1-12
1-10
2-12
Number
Min.
Max.
Mean
1,384
3,759
5,376
4,881
1,047
2,786
4,675
3,494
6,683
2,777
5,573
8,601
4,093
44
1,047
39,864
10,227
3,394
7,171
12,155
21,038
3,478
13,698
14,586
8,836
14,343
6,582
11,809
17,756
5,219
44
3,394
111,129
23,587
5
13
19
17
4
10
16
12
23
10
20
30
20
44
4
164
39
12-80
10-78
11-76
15-71
10-65
8-59
17-62
13-56
8-53
12-38
13-37
11-29
13-31
Number
Min.
Max.
Mean
56
168
234
439
41
374
150
768
1,408
281
299
819
3,473
50
41
17,705
3,527
191
655
851
1,384
159
645
337
3,029
9,294
449
561
5,096
5,054
50
159
48,977
11,308
0
1
1
2
0
1
1
3
5
1
1
3
17
50
0
54
13
(b)  Landfills located in the SE
C1
C2
C3
C4
C5
C6
N1
N2
O1 -2
V1-5
W1
W2
X1
AW1
AW2
BB1
BB2
BBS
1,118
1,118
1,118
1,118
1,118
1,118
1,524
1,524
1,499
1,727
1,499
1,499
1,016
1,118
1,118
1,092
1,092
1,092




1-9
1-12
1-8
1-4
1-12
1-10
1-12
1-12
1-6
1-10
9-12
1-8
1-7
1-12
1-12
1-6
1-11
1-11
Number
Min.
Max.
Mean
ND
1,475
3,417
14,828
6,419
3,273
ND
ND
ND
13,622
7,492
ND
43,693
6,358
3,555
10,378
ND
ND
11
1,475
43,693
10,410
ND
2,585
9,558
41,331
12,528
12,155
ND
ND
ND
49,828
8,799
ND
111,031
20,570
7,480
22,130
ND
ND
11
2,585
111,031
27,090

5
11
48
21
11



29
18

157
21
12
35


11
5
157
33
10-56
13-45
9-41
5-35
13-26
11-17
13-75
13-39
7-64
11-64
13-35
9-35
8-33
ND
ND
7-47
12-23
12-23
Number
Min.
Max.
Mean
332
311
301
624
2,513
393
1,862
3,536
5,987
10,923
1,856
4,125
3,913
ND
ND
2,494
5,422
2,284
16
301
10,923
2,930
1,419
859
987
2,055
10,440
1,403
17,597
7,274
16,467
41,601
6,365
10,524
14,315
ND
ND
8,983
14,042
7,945
16
859
41,601
10,142
1
1
1
2
8
1
4
8
15
23
5
10
14


8
18
8
16
1
23
8
                                       E-168
-------
Table E-5.1. LCRS Flow Rates for MSW Landfills During Operations (Continued).
Cell
No.
Average
Annual
Rainfall
(mm)
LCRS Flow Rates (Iphd) During Operations
Initial Period
Months
Average
Peak
Fraction of
Rainfall (%)
Active Period
Months
Average
Peak
Fraction of
Rainfall (%)
(c) Landfills located in the W
AM1
AM2
432
432
1-9
1-9
ND
ND
ND
ND

10-57
10-81
111
55
533
274
1
0
                                       E-169
-------
Table E-5.2. LCRS Flow Rates for HW Landfills During Operations.
Cell
No.
Average
Annual
Rainfall
(mm)
LCRS Flow Rates (Iphd) During Operations
Initial Period
Months
Average
Peak
Fraction of
Rainfall (%)
Active Period
Months
Average
Peak
Fraction of
Rainfall (%)
(a)  Landfills located in the NE
AQ1
AQ2
AQ3
AQ4
AQ5
AQ6
AQ7
AQ8
AQ9
AQ10
AV1
AV2
AV3
AV4
AV5
AY1
AY2
AYS
BA1
BA2
965
965
965
965
965
965
965
965
965
965
787
787
787
787
787
864
864
864
864
864




1-6
1-6
1-12
1-12
1-12
1-12
1-12
1-12
1-12
1-9
1-13
ND
ND
ND
1-12
1-9
1-11
1-11
1-14
1-2
Number
Min.
Max.
Mean
10,203
13,050
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
18,789
6,803
10,964
12,198
ND
4,979
7
4,979
18,789
10,998
18,944
20,721
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
44,741
12,439
23,914
32,326
ND
5,860
7
5,860
44,741
22,706
39
49












87
29
46
52

21
7
21
87
46
47-58
49-58
45-54
44-53
36-50
32-52
31-42
15-26
15-26
15-26
65-101
52-88
35-71
27-63
13-37
NA
NA
NA
15-42
3-12
Number
Min.
Max.
Mean
4,530
2,181
2,962
1,049
10,129
2,732
1,256
21,329
4,579
15,933
1,821
2,682
4,466
2,656
6,949
NA
NA
NA
3,951
2,190
17
1,049
21,329
5,376
10,531
6,460
13,430
1,622
34,353
12,072
4,821
76,759
24,946
38,751
7,484
12,591
15,363
9,778
27,720
NA
NA
NA
9,440
2,846
17
1,622
76,759
18,175
17
8
11
4
38
10
5
81
17
60
8
12
21
12
32



17
9
17
4
81
21
(b)  Landfills located in the SE
D3
D4
L1-3
AB1
AB2
AB3
AB4
AD1
AD2
ADS
AD4
ADS
1,626
1,626
1,143
1,702
1,702
1,702
1,702
1,829
1,829
1,829
1,829
1,829
1-12
1-11
8-12
1-5
1-12
1-12
9-13
1-12
1-12
1-14
1-14
1-7
20,292
31,281
22,795
479
878
4,050
7,229
ND
ND
ND
ND
ND
51,265
120,527
51,266
1,662
3,433
9,052
9,558
ND
ND
ND
ND
ND
46
70
73
1
2
9
16





13-28
NA
13-43
6-53
13-53
13-40
14-25
21-32
19-30
15-26
15-26
8-19
10,005
NA
13,417
270
1,865
4,971
2,114
373
1,886
1,685
1,071
37,054
44,895
NA
68,107
724
7,829
9,529
9,584
892
3,783
4,197
4,523
115,663
22

43
1
4
11
5
1
4
3
2
74
                                            E-170
-------
Table E-5.2. LCRS Flow Rates for HW Landfills During Operations (Continued).
Cell
No.
Average
Annual
Rainfall
(mm)
LCRS Flow Rates (Iphd) During Operations
Initial Period
Months
Average
Peak
Fraction of
Rainfall (%)
Active Period
Months
Average
Peak
Fraction of
Rainfall (%)
(b)  Landfills located in the SE (cont.)
AD6
AD7
ADS
AD9
AD10
AD11
AD12
AD13
AD14
AD15
AD16
AD17
AD18
AS1
1,829
1,829
1,829
1,829
1,829
1,829
1,829
1,829
1,829
1,829
1,829
1,829
1,829
1,422




1-7
1-12
1-8
1-9
1-12
1-5
1-13
1-12
1-12
1-13
1-12
1-12
1-12
1-10
Number
Min.
Max.
Mean
ND
12,597
24,803
19,900
20,960
11,875
25,609
18,604
20,104
24,664
17,442
5,761
5,035
ND
19
479
31,281
15,493
ND
26,492
39,997
42,854
51,425
20,518
55,840
86,467
85,939
89,367
84,110
10,955
15,685
ND
19
1,662
120,527
45,074

25
49
40
42
24
51
37
40
49
35
11
10

19
1
73
33
8-19
13-70
9-64
10-58
13-79
6-69
14-69
13-55
13-55
14-39
13-39
13-16
13-19
11-27
Number
Min.
Max.
Mean
992
1,206
2,058
1,880
1,883
3,536
2,150
5,403
4,452
390
17,927
4,046
1,186
329
25
270
37,054
4,886
1,853
2,857
10,545
8,051
15,064
6,708
19,674
51,260
40,684
1,294
153,293
6,109
3,257
1,146
25
724
153,293
23,661
2
2
4
4
4
7
4
11
9
1
36
8
2
1
25
1
74
11
(c)  Landfills located in the W
AC1
AC2
ACS
AC4
ACS
AC6
AC7
ACS
AP1
AT1
279
279
279
279
279
279
279
279
381
737




1-13
1-6
1-12
1-11
1-13
1-10
1-12
1-12
1-12
1-5
Number
Min.
Max.
Mean
85
272
42
51
255
352
54
67
3,093
ND
9
42
3,093
475
169
1,429
379
255
977
2,990
120
168
10,515
ND
9
120
10,515
1,889
1
4
1
1
3
5
1
1
30

9
1
30
5
ND
7-88
13-73
12-73
14-34
11-40
13-18
13-18
13-48
6-8
Number
Min.
Max.
Mean
ND
18
1
2
56
200
1,925
1,138
4,272
1,249
9
1
4,272
985
ND
217
13
21
255
1,002
6,713
4,601
15,059
1,964
9
13
15,059
3,316

0
0
0
1
3
25
15
41
6
9
0
41
10
                                            E-171
-------
Table E-5.3. LCRS Flow Rates for Ash Landfills.
Cell
No.
Average
Annual
Rainfall
(mm)
LCRS Flow Rates (Iphd) During Operations
Initial Period
Months
Average
Peak
Fraction of
Rainfall (%)
Active Period
Months
Average
Peak
Fraction of
Rainfall (%)
(a)  Landfills located in the NE
AF1
AN 1-4
S2
Y1
Z1
1,118
1,118
1,041
1,194
1,245




1-12
1-7
1-9
1-12
1-12
Number
Min.
Max.
Mean
25,383
ND
2,185
ND
28,628
3
2,185
28,628
18,732
40,850
ND
4,650
ND
49,551
3
4,650
49,551
31,684
83

8

84
3
8
84
58
13-60
8-34
10-46
13-78
13-39
Number
Min.
Max.
Mean
19,463
13,278
1,026
19,319
35,312
5
1,026
35,312
17,680
41,880
30,179
3,638
63,832
92,207
5
3,638
92,207
46,347
64
43
4
59
104
5
4
104
55
(b)  Landfills located in the SE
P1
P2
P3
1,499
1,499
1,499
1-12
1-12
1-10
ND
ND
ND
ND
ND
ND







13-39
13-27
11-15
Number
Min.
Max.
Mean
20,086
8,935
24,490
3
8,935
24,490
17,837
45,591
12,277
60,420
3
12,277
60,420
39,429
49
22
60
3
22
60
43
Table E-5.4. LCRS Flow Rates for C&D W Landfills
                                                d)
Cell
No.
Average
Annual
Rainfall
(mm)
LCRS Flow Rates (Iphd) During Operations
Initial Period
Months
Average
Peak
Fraction of
Rainfall (%)
Active Period
Months
Average
Peak
Fraction of
Rainfall (%)
A1-2
AM
AI2
1,118
1,143
1,143




1-24
3-12
3-6
Number
Min.
Max.
Mean
ND
15,552
19,613
2
15,552
19,613
17,583
ND
29,341
22,964
2
22,964
29,341
26,153

50
63
2
50
63
56
25-48
13-50
7-50
Number
Min.
Max.
Mean
3,568
12,118
16,159
3
3,568
16,159
10,615
8,275
29,477
34,159
3
8,275
34,159
23,970
12
39
52
3
12
52
34
Notes: (1) All of the landfills in this table are located in the NE of the U.S.
                                            E-172
-------
          Table E-5.5  LCRS Flow Rates (in Iphd) for MSW and HW Landfills During the Post-Closure Period.
Cell
No.
B1
B2
11
12
13
14
15
AX1
AX2
AX3
AX4
D1
AD1
AD2
ADS
AD4
ADS
AD6
AD7
ADS
AD9
AQ1
AQ2
AQ3
AQ4
AQ5
AQ6
AQ7
AQ8
AQ9
AQ10
AS1
AT1
Number
Avg.
S.D.
Type of
Waste
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
MSW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW
HW



Region

NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
SE
SE
SE
SE
SE
SE
SE
SE
SE
SE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
SE
W



Year
0
4061.2
2408.9
762.1
762.1
762.1
2114.0
2590.2
144.4
296.4
615.5
17.1

424.0
2035.3
1891.7
1219.4
42801.5
997.0
450.0
227.9
209.9
5067.7
1381.4
2081.1
1512.7
9062.3
2492.7
896.7
19817.8
3308.8
11314.5
329.7
1246.7
32
3853.2
8151.7
Year
1
568.5
500.7
675.9
675.9
675.9
566.6
204.7
64.7
292.2
254.0
61.6

119.6
1141.5
652.3
326.2
8312.2
969.5
374.6
309.6
561.9
477.1
189.0
351.5
4261.8
1406.4
3107.9
40.8
498.2
324.9
521.2
136.0
87.3
32
897.2
1606.2
Year
2
222.5
409.1




4.9
46.0
126.2
197.7
48.1
480.4
145.0
644.0
321.5
465.3
3288.9
200.3
164.9
188.7
549.5
766.7
205.9
419.9
1982.0
579.8
479.2
62.9
623.1
286.3
497.4
110.0
24.5
29
466.9
660.0
Year
3
747.7
500.3





49.9
145.7


493.3
52.5
321.8
79.4
66.0
1165.9
114.2



1766.4
746.8
721.4
639.0
644.7
532.8
212.7
507.1
670.9
900.6
38.6

22
505.3
426.4
Year
4
1331.0
242.2





120.8
240.8



0.0
213.0
476.8
156.6
682.8
63.9



1421.3
216.6
1850.6
466.5
490.9
405.7
360.7

2111.2
774.6
26.1

20
582.6
614.9
Year
5
1174.4
320.9





43.0
89.8



5.3
132.7
399.1
68.4
444.2
70.1













34.7

11
252.9
342.5
Year
6
651.1
257.7










1.4
48.4
172.7
13.3
278.1
36.6















8
182.4
218.8
Year
7












0.0
60.1
111.2
1.4
9.5
34.5















6
36.1
43.4
Year
8












0.2
37.1
61.2
1.8
0.6
8.2















6
18.2
25.3
Year
9












2.6
0.0


0.0
0.0















4
0.6
1.3
m




CO
-------
three times the average monthly flow rates.  For the initial period of operation, the peak
flow rates ranged between approximately 3,000 Iphd and 111,000 Iphd, and for the
active period of operation, between approximately 160 Iphd and 49,000 Iphd.  Peak
rates greater than 10,000 Iphd were exhibited by 70% of the cells during the initial
period of operation and by 38% of the cells during the active period of operation.

LCRS flow rates measured from the MSW cells located in the SE were in general
similar to those from MSW cells located in the NE, as shown in Table E-5.1.  However,
because rainfall amounts are in general about 20  to 50% greater in the SE than in the
NE, calculated RF values are smaller for the SE landfills than for the NE landfills. For
example, for the NE  MSW landfills during the active period of operation, RF values
ranged from 0 to 54%. More than half of the cells had RF values greater than 10%  and
25% of the cells had RF values greater than 20%. For the SE MSW Landfills during the
active period of operation, RF values ranged from 1 to 23%.  Only 25% of the cells had
an RF value greater than 10%.  It is interesting that the higher rainfall amounts in the SE
versus the NE did not yield higher LCRS flow rates for MSW landfills. This may be due
to the higher water evaporation rates in the SE (due to the higher temperatures) than in
the NE.  It may also  indicate that rainfall intensity  (i.e., amount of rainfall divided by
rainfall period) has an impact on leachate generation rates. Rainfall events in the SE
are more intense than in  the NE. More intense rainfalls have the potential to generate
larger surface-water runoffs and less infiltration, and thus less leachate.

Data are only available for two MSW cells located in the W. For these two cells,
average and peak LCRS flow rates were very low. Average LCRS flow rates were 111
Iphd and 55 Iphd for Cells AM1 and AM2, respectively.  Peak monthly rates were about
five times the  average rates.  It is noted that Landfill AM receives an  average annual
rainfall of 432 mm, which is relatively low. The calculated RF values are less than 1%.
These RF values are very low in comparison  to RF values  for landfills in the NE and SE.
This indicates that landfills in arid regions that receive average annual rainfalls less  than
500 mm may have very low leachate generation rates.

The evaluation of Tables E-5.1 through E-5.4 shows that the MSW landfills produced
less leachate than the other types of waste in all three regions of the U.S. The HW
landfills had LCRS flow rates during the active period of operation that were 50 to 60%
greater than those from the MSW landfills.  For the HW cells located in the NE and SE
for which initial period of  operation data are available, average LCRS flow rates varied
between approximately 500 Iphd and 31,000  Iphd. During  the active period of
operation, average LCRS flow rates were from approximately 300 Iphd to 37,000 Iphd.
Peak monthly LCRS flow rates were typically two  to three times the average monthly
flow rates.  For the initial  period of operation,  the peak flow rates ranged between
approximately 2,000 Iphd and 121,000 Iphd, and,  for the active period of operation,
between approximately 700 Iphd and 153,000 Iphd.  Peak  rates greater than 10,000
Iphd were exhibited by most of the cells during both the initial and active periods of
operation. For the NE HW landfills during the active period of operation, RF values
                                  E-174
-------
ranged from 4 to 87% (mean = 21 %).  For the SE HW landfills during the active period
of operation, RF values ranged from 1 to 74% (mean = 1%).

Data are available for ten HW cells located in the W.  For these cells, average LCRS
flow rates ranged from 40 Iphd to 3,000 Iphd during the initial period of operation and
from 0 Iphd to 4,000 Iphd during the active period of operation.  Peak flows were three
times greater than average flows.  RF values ranged from 1 to 30% and from 0 to 40%
during the initial and active  periods of operation, respectively.  For most of the cells, RF
values were less than 10%  during operations.

The limited number of ash and C&DW cells had significantly higher LCRS flow rates
than the MSW and HW cells. For the eight ash cells  located in the NE and SE, average
LCRS flow rates were in the range of approximately 1,000 Iphd to 35,000 Iphd during
operations and the peak flow rates were from approximately 4,000 Iphd to 92,000 Iphd.
The three C&DW landfills located in the NE had average rates during operations from
4,000 Iphd to  20,000 Iphd, and peak rates from  8,000 Iphd to 34,000 Iphd.  Flow rates
from these ash and C&DW cells during the active period of operation were 300 to 600%
higher than flow rates from  MSW cells.  Mean RF values during operations were 52%
for the ash cells and 43% for the C&DW cells. The waste characteristics (i.e., initial
moisture content, porosity, and permeability)  and the waste disposal and covering
practices may attribute to these differences in LCRS flows. For example, ash is often
sprayed with water to control dust and is not covered with soil as often as MSW.

Tables E-5.1 through E-5.4 show that landfill  geographic region (i.e., humid versus arid)
has a major impact on LCRS flow rates.  Landfills located in the humid NE and SE had
much higher LCRS flow rates and RF values  than landfills located in the arid W. In
particular, almost all of the cells that had historical average annual rainfall less than 500
mm had average LCRS flow rates that are less  than 2,000 Iphd and peak LCRS flow
rates that are less than 7,000 Iphd. The RF values for many of these cells were less
than 5%. Figures E-5.1 through E-5.4 show the effects of precipitation on LCRS flow
rates for the sites considered. These figures  show average and peak LCRS flow rates
for the initial and active periods of operation plotted as a function of historical average
annual rainfalls at the sites.  The ranges of average annual rainfalls covered by each of
the three regions are also shown on the figures. Though the LCRS flow rates are
considerably variable, a general trend can be observed from the four figures.  On each
figure an envelope was drawn that defines the approximate upper bound of the
measured LCRS flow rates.  As average annual rainfall increases, the average and
peak LCRS flow rates increase. Sites in the Wwith less than 500-mm average annual
rainfall exhibited low LCRS flow rates. In addition,  increasing annual rainfall beyond
1,100 to 1,200 mm (which only occurred for sites in the SE in this study) typically does
not seem to cause a corresponding increase  in  leachate generation rates.  As
previously stated, this may be attributed to differences in waste evaporation rates and
rainfall intensities between the SE and the NE.
                                  E-175
-------
m


en
                D,000 --
a!   20,000



LU

g   15,000
           =5-   35,000


           Q.
           [d   30,000
           <
           a:
           g   25,000

           _i
           Ll_
                            200     400     600     800    1,000   1,200   1,400



                                              AVERAGE ANNUAL RAINFALL (mm)
                                                                         1,600   1,800   2,000
             Figure E-5.1.  Average LCRS Flow Rate Versus Average Annual Rainfall During the Initial

                            Period of Operation.
-------
               135,000
m
          LU
          O
          CO
          o:
          o
          LU

          CL
                            200     400    600     800    1,000   1,200   1,400    1,600   1,800


                                             AVERAGE ANNUAL RAINFALL (mm)
2,000
               Figure E-5.2.  Peak LCRS Flow Rate Versus Average Annual Rainfall During the Initial Period

                             of Operation.
-------
              40,000
m

oo
                           200     400     600     800    1,000   1,200    1,400    1,600

                                            AVERAGE ANNUAL RAINFALL (mm)
1,800   2,000
              Figure E-5.3.  Average LCRS Flow Rate Versus Average Annual Rainfall During the Active
                            Period of Operation.
-------
              165,000
m


^i
CD
                            200    400     600     800    1,000   1,200   1,400   1,600    1,800



                                             AVERAGE ANNUAL RAINFALL (mm)
2,000
               Figure E-5.4.  Peak LCRS Flow Rate Versus Average Annual Rainfall During the Active

                            Period of Operation.
-------
Leachate generation rates during the post-closure period for eleven MSW cells and 22
HW cells are summarized in Table E-5.5. The data cover a maximum post-closure
period of nine years.  The table presents average LCRS flow rates for these cells for the
year prior to closure as well as for the years after closure. The mean LCRS flow rate for
all of the cells during the year before closure was approximately 3,900 Iphd. Within one
year after closure, the mean dropped by a factor of four and within two to four years
after closure the mean dropped approximately one order of magnitude. After six years
of closure, the eight cells for which data are available had LCRS flow rates of
approximately 5 to 1,200 Iphd (mean of 180 Iphd). As shown in Table E-5.5, four of the
six cells for which LCRS flow rate data  exists for nine years after closure had negligible
LCRS flow rates.  Figure E-5.5 shows LCRS flow rates for Cell ADS.  For this cell, the
average LCRS flow rate during the year before closure was 42,800 Iphd. After one year
of closure, the rate decreased by a factor of five and after two years of closure, the rate
decreased an order of magnitude from  the pre-closure value. After nine years of
closure, the LCRS rate became negligible. This significant decrease  in LCRS flow rates
demonstrates the ability  of well designed and constructed cover systems in minimizing
infiltration of rainwater into the waste and thus minimizing leachate generation.  Figure
E-5.6 shows the average LCRS flow rates measured for the 33 cells under closure.
This figure shows that after a cell is closed and a cover system is  installed, LCRS flow
rates will decrease with time and become negligible after approximately eight to ten
years of closure.

E-5.4    Implications for Landfill Performance

The data presented in this section show that for the considered landfills peak monthly
LCRS flow rates were typically two to three times the average monthly LCRS flow rates.
The data also shows that LCRS flow rates were typically two to three times smaller
during the active period of operation than during the initial period of operation.  During
the initial period of operation, LCRS flow rates will be greatly influenced by rainfall.
During the active period  of operation, the amount of waste in the cell  is greater and daily
and intermediate cover soils are placed on the waste, thus minimizing infiltration of
rainwater into the waste.

Evaluation of the LCRS flow rate data shows that average and peak LCRS flow rates
vary significantly even between similar  sites (i.e., sites in the same region and with the
same type of waste).  For example, for  44 MSW cells located in the NE, the average
LCRS flow rate  during the initial period  of operation varied between approximately  1,000
Iphd to 40,000 Iphd. Peak monthly LCRS flow rates were even more  variable for these
44 cells  ranging between approximately 3,000 Iphd and 111,000 Iphd. Similar results
are found for the other types of waste.  It is expected based  on these results that waste
placement practices have a very significant effect on leachate generation rates.
Although these effects are not directly quantified in this study, it is expected that a
landfill operator can minimize leachate  generation rates by using a small active disposal
area and implementing effective measures to minimize infiltration  of rainwater into  the
waste and to divert surface water away from the landfill.
                                  E-180
-------
m

oo
                  20,000
             a.

             LU
CO
a:
o

LU
O
<
a:
LLJ
                  15,000
                  10,000
                   5,000
                       flow rate during the year prior

                       to closure = 42,800 Ihd
                                                             4567

                                                           YEAR SINCE FINAL CLOSURE
                                                                                                       10
            Figure E-5.5. Average LCRS Flow Rates (Ihd) After Closure of Cell ADS, a HW Cell Located in the SE.
-------
m
oo
ro
i uu,uuu.u -
1
1
10,000.0 |

'S'
_c
_Q.
w 1,000.0
£
|
ut 100.0-
co
a:
o
_i
UJ |

< 10.0
a:
LLJ
<


1.0


n 1 -

I D
D
a n
^ n
g R
I S °
' e S 8
a n
D B
D




O








0123



H
O
i
§
g
8

D
D













4




O
o
^ e
n

^ D D
Q ^


n

n
D
D
n
a a


D
D

567891
                                                  YEAR SINCE FINAL CLOSURE
       Figure E-5.6. Average LCRS Flow Rates (Ihd) After Closure for Eleven MSW Cells (Shown as Circles) and 22 HW

                    Cells (Shown as Squares) (Note: Flow Rates of 0.0 Iphd are Shown as 0.1 Iphd).
-------
E-6    Leachate Chemistry Data

E-6.1    Introduction

This section presents an evaluation of the available leachate chemistry data presented
in Table E-3.7 for 59 cells at 50 landfills. The purpose of this section is to characterize
the chemistry of leachate from MSW, HW, and ISW landfills, compare  the leachate
chemistry data in this appendix to the published data summarized in Section E-2.3, and
evaluate the effect of the solid waste regulations discussed in Section  E-2.3.1  on
leachate quality (i.e.,  has the amount of trace toxic inorganic and synthetic organic
chemicals in leachate decreased). The remainder of this section is organized as
follows:

    •  landfill leachate chemistry is characterized in Section E-6.2;
    •  the leachate chemistry data are compared to published data in Section E-6.3;
       and
    •  the effect of the solid waste regulations on leachate quality is discussed in
       Section E-6.4.

E-6.2    Characterization of Landfill Leachate Chemistry

E-6.2.1   Introduction

The chemistries of leachates from MSW,  HW, and ISW landfills are characterized below
in terms of concentrations and relative detection frequencies (i.e., were the chemicals
detected in 50% or less of the samples or more than 50% of the samples) of 30
representative chemical parameters. A summary  of the average chemical
concentrations and the range of average  concentrations found in the landfill leachates is
given in Table E-6.1.  Federal MCLs for community drinking water systems (40 CFR §
141.11, 141.61, 141.62) are available for two of the heavy metals and  ten of the VOCs
considered in this study,  and are also listed in Table E-6.1.  The distributions of
chemistry data for MSW,  HW, and MSW ash landfill cells are shown in Figures E-6.1
through E-6.3, respectively, for the following select parameters and chemicals: BOD,
sulfate, chromium, benzene,  1,2-dichloroethane, and toluene.  For MSW landfills, the
chemical data for older and newer landfills were compared.  Older landfills are
considered to be those that started operating before 1990 (i.e., pre-1990 landfills).
Newer landfills started operating during 1990 or later (i.e., post-1990 landfills). The
distributions of chemistry  data for pre-1990  and post-1990 MSW landfill cells are shown
in Figure E-6.4 for the following select chemicals: chromium, benzene, 1,1-
dichloroethane, and toluene.  In Figures E-6.1 through E-6.4, non-detected chemicals
were graphed at  half  of their method detection  limits.

Leachate chemistry time trends are presented for  the one landfill cell (i.e., MSW Cell
B1) for which a relatively complete leachate chemistry data set is  available for a number
of years. Other parameters that may affect leachate chemistry, such as landfill
                                  E-183
-------
        Table E-6.1.  Summary of Landfill Leachate Chemistry Data.
Waste Type
Number of Landfills
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium \igl\
Lead \igl\
Nickel ng/l
Benzene |ig/l
1,1-Dichloroethane \igl\
1,2-Dichloroethane \igl\
cis-1,2-Dichloroethylene |ig/l
trans-1,2-Dichloroethylene |ig/l
Ethylbenzene \igl\
Methylene chloride \igl\
1,1,1-Trichloroethane |ig/l
Trichloroethylene \igl\
Toluene \igl\
Vinyl chloride |ig/l
Xylenes |ig/l
MCLs












50
5
100


5

5
70
100
700

200
5
1,000
2
10,000
MSW
10 Pre-1 990
Average
6.62
6,588
5,487
3,878
2,281
1,509
2,295
801
274
444
153
532
19
< 8
68
36
56
< 17
88
< 33
< 64
< 51
40
435
< 68
< 56
491
< 49
117
Minimum
6.30
3,438
2,740
804
<2
4
1,508
199
<23
261
84
225
<4
< 1
5
1
27
< 3
< 5
< 4
< 53
< 32
< 5
< 5
< 5
< 5
< 5
< 7
< 5
Maximum
7.20
8,983
8,640
8,267
4,510
2,852
3,278
2,263
1,943
610
279
1,115
78
< 17
320
90
98
< 36
294
< 100
< 75
< 100
87
1,303
100
114
959
< 100
277
No. of
Landfills
8
8
9
9
10
8
7
10
10
6
6
8
10
8
10
7
9
7
8
6
2
4
7
8
6
7
7
6
6
26 Post-1 990
Average
6.79
3,693
2,758
1,939
976
527
1,536
463
205
398
83
282
23
< 7
38
15
82
< 19
66
< 16
< 57
< 18
35
334
< 55
< 24
228
< 34
83
Minimum
5.90
597
480
< 10
<2
24
203
5
< 7
66
10
3
<2
< 1
3
1
10
<2
< 2
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 1
< 3
< 5
Maximum
8.09
13,548
8,621
6,800
4,700
2,609
5,800
1,625
1,376
1,994
191
1,219
236
< 20
90
50
220
< 100
260
< 100
436
< 110
118
4,150
270
100
740
< 300
220
No. of
Landfills
22
22
21
22
18
21
22
25
24
22
21
23
21
22
21
22
20
21
22
20
13
16
22
22
20
19
22
20
20
m
2
        Notes: (1)  " " = not analyzed; < = more than 50% of measurements reported as non-detect.
-------
       Table E-6.1. Summary of Landfill Leachate Chemistry Data (Continued).
Waste Type
Number of Landfills
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium |ig/l
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1,1-Dichloroethane |ig/l
1,2-Dichloroethane |ig/l
cis-1,2-Dichloroethylene |ig/l
trans-1,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methylene chloride |ig/l
1,1,1-Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl chloride |ig/l
Xylenes |ig/l
MCLs












50
5
100


5

5
70
100
700

200
5
1,000
2
10,000
HW
4
Average
8.17
22,096



1,623

7,758
2,985


5,243
26,710
< 119
124
109
738
< 131
123
< 382

< 79
< 133
161
< 99
< 76
< 173
< 1,475
14
Minimum
7.55
12,302



7

3,783
704


2,514
30
< 5
22
24
285
< 7
< 14
5

< 14
< 5
4
8
33
< 9
< 10
9
Maximum
9.36
39,598



3,239

11,734
5,267


7,972
79,912
<233
226
249
1,190
370
< 371
< 1,124

< 143
< 512
< 447
< 347
< 146
616
< 4,405
18
No. of
Landfills
3
3



2

2
2


2
3
2
2
3
2
3
4
3

2
4
4
4
3
4
3
2
MSW ASH
7
Average
7.06
22,083
24,493
1,670
55
62
1,942
10,426
881
900
267
1,181
9
< 12
< 30
23
<40
< 3
< 12
< 3
<2
< 3
< 4
< 3
< 7
< 3
< 10
< 5
<2
Minimum
6.54
10,732
6,067
304
15
39
99
2,940
85
96
113
684
5
<2
< 1
3
<24
< 1
< 1
< 1
< 1
< 1
< 2
< 1
< 1
< 1
< 1
< 1
< 1
Maximum
7.44
43,383
46,733
5,607
84
109
5,010
22,400
3,430
1,332
420
1,994
17
49
84
74
48
< 5
< 33
< 5
< 3
< 5
< 7
< 6
< 16
< 5
< 25
< 10
< 3
No. of
Landfills
5
4
6
4
4
3
4
4
5
3
2
5
6
6
6
6
4
3
3
3
2
3
3
3
3
3
3
3
2
m
00
Ol
       Notes: (1) " " = not analyzed; < = more than 50% of measurements reported as non-detect.
-------
       Table E-6.1. Summary of Landfill Leachate Chemistry Data (Continued).
Waste Type
Number of Landfills
Parameter Units
pH pH units
Specific conductance |imhos/cm
IDS mg/l
COD mg/l
BOD5 mg/l
TOC mg/l
Alkalinity mg/l
Chloride mg/l
Sulfate mg/l
Calcium mg/l
Magnesium mg/l
Sodium mg/l
Arsenic |ig/l
Cadmium |ig/l
Chromium \igl\
Lead |ig/l
Nickel |ig/l
Benzene |ig/l
1,1-Dichloroethane |ig/l
1,2-Dichloroethane |ig/l
cis-1,2-Dichloroethylene |ig/l
trans-1,2-Dichloroethylene |ig/l
Ethylbenzene |ig/l
Methylene chloride |ig/l
1,1,1-Trichloroethane |ig/l
Trichloroethylene |ig/l
Toluene |ig/l
Vinyl chloride |ig/l
Xylenes |ig/l
MCLs












50
5
100


5

5
70
100
700

200
5
1,000
2
10,000
COAL ASH
2
Average
7.70
884
723
11
< 3
6
190
21
383
190
22
46
36
< 7
< 16
< 19
38
<4
< 4
< 4

< 1
< 3
< 4
<4
< 4
< 2
< 7
<4
Minimum
7.66
623
347
11
< 3
6
160
21
178
190
15
46
< 9
< 5
< 9
< 4
38
<4
< 4
< 4

< 1
< 3
< 4
<4
< 4
< 2
< 7
<4
Maximum
7.74
1144
1098
11
< 3
6
220
21
587
190
30
46
62
< 9
22
< 34
38
< 4
< 4
< 4

< 1
< 3
< 4
< 4
< 4
< 2
< 7
<4
No. of
Landfills
2
2
2
1
1
1
2
1
2
1
2
1
2
2
2
2
1
1
1
1

1
1
1
1
1
1
1
1
C&DW
2
Average
6.43
4815
3553
2414
1126
839
2450
681
255
292
202
304
15
< 3
39
7
< 56
17
92
3


66
417
51
< 11
613
8
210
Minimum
6.43
4815
2880
1139
1126
443
2450
671
48
203
202
284
15
< 1
39
3
< 56
17
92
3


66
417
51
< 11
613
8
210
Maximum
6.43
4815
4225
3688
1126
1235
2450
690
463
382
202
324
15
< 5
39
10
< 56
17
92
3


66
417
51
< 11
613
8
210
No. of
Landfills
1
1
2
2
1
2
1
2
2
2
1
2
1
2
1
2
1
1
1
1


1
1
1
1
1
1
1
m
00
CD
       Notes: (1) " " = not analyzed; < = more than 50% of measurements reported as non-detect.
-------
m
oo
                  1E+04
                  1E+03
               c
               g
               '-i—*
               ro
               "c
               0
               o
               c
               ro
               g
               E
               0
               .c
               O
1E+02
                  1E+01
                  1E+00
• BOD (mg/l)
• Sulfate (mg/l)
A Chromium (ug/l)
x Benzene (ug/l)
o 1,1-Dichloroethane (ug/l)
• Toluene (ug/l)
                                                    Selected Chemicals
                    Figure E-6.1. Distribution of Select Chemical Data for MSW Leachate
-------
m
oo
oo
1 CTUt -
-\ F+m
1 ^TUO
c
O
1 1E+02-
0
o
o
O

o -iF+m
^ I ^TU I
E
0
f—
_L_
O
•i F+nn
1 ^TUU
•IF n-i

















I
I

I














1
1

1



>
i
,. >




i







i






>
>





( C

t
(


/
^
t

(
(





1
)

(
>

I
i
j

> f

4




1


I


I



1

»






* BOD (mg/l)

• Sulfate (mg/l)
A Chromium (ug/l)

x Benzene (ug/l)
O 1 ,1-Dichloroethane (ug/l)

• Toluene (ug/l)





                                                   Selected Chemicals
                     Figure E-6.2. Distribution of Select Chemical Data for HW Leachate
-------
m
oo
CD
1 CTUt
-ip+no _
C
g
i; -i F+D9 -
I
o
(J
"ro
•— 1F+D1 -

0
O
1 F+nn
IF m -



<
<
<



l
l
l
> i
>
>



i
l
l
1 L

l




i
i
i
fc

>
>
\
5



(

r
( (
( (



)
4

> 4
4
>



»

>
>


* BOD (mg/l)
• ^nlfpfp fmn/h
A Chromium (ug/l)
x Benzene (ug/l)
O 1 ,1-Dichloroethane (ug/l)
• Toluene (ug/l)


                                                 Selected Chemicals
                    Figure E-6.3. Distribution of Select Chemical Data for MSW Ash Leachate
-------
m
CD
O
i ^^ut -
-ip+no
1 ^^L/O
C
g
U— *
CO
"c
0
o
c 1E+02-
O
"CD
o
E
0
.c
O
1E+01 -





iF+nn




<


^

^




>


V

^
I 3
V 4

^



<







1
> <
4






i
£


A •





i


A

>
i *

/


t
• i
>


i

±



i t

















I |
6 !
L f\ \
t Vft ^
|
Q
H
V
C
C



c
i
c


/
>
\


>

t
«





!
i
r
t
, [
[
i
[


i


i

i

> E





1
j
4
]
3
]
3














i 1
n B
1 •
3 f
i
T

]



,





o Chromium (ug/l)

A Benzene (ug/l)
o 1,1-Dichloroethane (ug/l)
n Toluene (ug/l)










                                               Selected Chemicals
                Figure E-6.4. Distribution of Select Chemical Data for MSW Leachates from Pre-1990 and Post-1990
                       Landfills (Pre-1990 Data have open symbols, Post-1990 data have closed symbols).
-------
geographic location and date of waste placement relative to sample collection date,
were also considered.  However, there were too few data to draw meaningful
conclusions. For example, of the 36 MSW landfills, 31 are located in the NE, four are
located in the SE, and one is located in the W. In general,  the chemical concentrations
for leachate from the SE and W MSW landfills fall within the range of values detected
for leachate from the NE MSW landfills. It may be that geographic location has no
significant impact on leachate chemistry or it may be that there were too few leachate
chemistry data for MSW landfills in the SE and Wto observe an effect.

E-6.2.2   MSW

The MSW landfill leachates were found to be mineralized, biologically active liquids with
relatively low concentrations of heavy metals and VOCs. Of the trace metals and VOCs
in Table E-3.7, chromium, nickel, methylene chloride, and toluene were detected at the
highest concentrations.  Many of the chemicals exhibited significant concentration
variations (e.g., several orders of magnitude difference) between landfills and,
sometimes, for a given landfill as shown in Table E-3.7 and Figure E-6.1.  Based on the
pH values, the MSW landfill leachate was found to be, on average, slightly acidic (i.e.,
average pH of 6.7). This is expected  because carbon dioxide and organic acids are the
primary by-products of the first stage (i.e., the acid stage) of anaerobic degradation of
organic compounds in MSW landfills.  As the MSW in the landfill  ages and the
placement of fresh MSW ceases, methanogens begin to proliferate in a landfill and the
pH begins to approach neutrality as the acids are converted into  methane and a
bicarbonate buffering system is  established (the methane fermentation stage).  This
increase  in pH  with time is evident for landfill cell B1 after it was closed, as shown in
Table E-6.2. This cell became operational in 1984 and was closed in 1988.  In general,
a significant increase in pH with time was not observed for active cells receiving fresh
MSW. For cell B1, the BOD, COD, and BOD/COD ratios also decreased with time after
cell closure. Based on studies of older landfills (i.e., Chian and DeWalle, 1977), the
decrease in these parameters was expected. BOD and COD levels decrease as MSW
is degraded, but BOD decreases faster than COD.  Thus, the BOD/COD ratio also
decreases. The BOD/COD ratios for cell B1 are  characteristic of a landfill transitioning
from  the acid stage to the methane fermentation  stages.  As described by Ehrig and
Scheelhaase (1993), the acid stage is characterized by BOD/COD values greater than
0.4, and the methane fermentation stage is characterized by BOD/COD values less
than  0.1.

Average  chemical concentrations of cadmium, benzene,  1,2-dichloroethane,
trichloroethylene, and vinyl chloride in MSW landfill leachates exceeded MCLs. None of
the landfills had leachate with average chemical concentrations exceeding the MCLs for
ethylbenzene, toluene, or xylenes.
                                  E-191
-------
Table E-6.2 Average Concentrations of Select Parameters Over Time for Cell B1.
Date
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
PH
(pH units)
6.01
6.04
5.93
6.26
6.83
7.64
7.47
7.43
7.43
7.82
BOD
(mg/l)
1,180
2,440
16,700
3,120
2,065
730
124
68
130
112
COD
(mg/l)
2,900
9,650
38,600
5,780
4,250
1,670
1,850
572
1,440
1,050
BOD/COD
0.41
0.25
0.43
0.54
0.48
0.43
0.07
0.11
0.09
0.11
Chloride
(mg/l)

359
1,070
1,000
825
1,180
4,080
1,350
320
1,350
Methylene
Chloride
(ng/0

289
898
860
560
50
6
2
16
8
E-6.2.3    HW

The HW landfill leachates were more mineralized and had a higher organic content than
MSW leachates. No COD or BOD data were available for these landfills. All of the HW
leachates were alkaline, with pH values ranging from 7.55 to 9.36.  One possible
explanation for the alkaline pH values is the relatively common practice of solidifying
HWwith pozzolonic additives prior to disposal. These relatively high pHs decrease the
mobility of metals.  Even so, the metal concentrations in the HW leachates were
relatively high.  The average concentrations of the heavy metals were generally several
times to several orders of magnitude higher in HW leachate as compared to MSW
leachate. The biggest difference occurred for arsenic: the average arsenic
concentrations in leachates from HW and MSW landfills were 26,710 and 22 |ig/l,
respectively.  However, the high average arsenic concentration for HW leachate was
due to the high  arsenic levels in leachate from Landfill AD.  Excluding the data from this
HW landfill, the average arsenic concentration for HW leachate is 110 jig/l. The HW
leachates also had higher average concentrations than MSW leachates of all VOCs
except for methylene chloride, toluene, and xylenes. These differences can be seen
from comparing the  distributions of select chemical data for MSW and HW leachates
given in Figures E-6.1  and E-6.2, respectively.  Of the heavy metals and VOCs
considered in Table E-6.1, arsenic, nickel,  1,2-dichloroethane, and vinyl chloride were
detected at the  highest concentrations in HW leachates.

Average concentrations of arsenic, cadmium, chromium, benzene, 1,2-dichloroethane,
trichloroethylene, and vinyl chloride in HW landfill leachates exceeded MCLs.  None of
the landfills had leachate with average chemical concentrations exceeding the MCLs for
ethylbenzene, toluene,  or xylenes.
                                  E-192
-------
E-6.2.4   ISW

E-6.2.4.1 MSWAsh

For the purposes of the discussions on leachate chemistry, MSWash landfill leachate is
grouped with leachate from ISW landfills. This grouping is considered appropriate
because MSWash landfill leachate is typically nonhazardous and has chemical
characteristics more similar to leachate from ISW landfills than to leachate from MSWor
HWlandfills. The MSWash landfill leachates were more mineralized than the MSW
leachates, as evidenced by the high specific conductance, TDS, sulfate, and chloride
levels of the MSW ash leachates.  The concentrations of heavy metals were within the
range of those for MSW leachates. Unlike MSW leachates, the COD and BOD values
for the MSW ash leachates were very low because most of the organic portion of the
MSW had been combusted. As expected, no VOCs were detected.  The lower BOD
concentration, higher sulfate concentration, and  lower VOC concentrations in MSWash
leachate as compared to MSW leachate can be  seen by comparing Figures E-6.1 and
E-6.3.

Average concentrations of cadmium in the MSW ash landfill leachates exceeded MCLs.
None of the landfills had leachate with average chemical concentrations exceeding the
MCLs for any considered chemical except cadmium.

E-6.2.4.2 Coal Ash

Leachate chemistry data were obtained for the two coal ash landfills located in the NE.
The ash was produced in plants burning eastern bituminous coal. The coal ash
leachate was slightly alkaline and contained metals, but no VOCs.  This was expected
since any VOCs would have been combusted with the coal. In comparison to the MSW,
HW, and MSWash leachates, the coal ash leachate was less mineralized  and
contained metals at concentrations generally at the lower end of the concentration
range for MSW leachates. The relatively low concentrations of metals were expected
based on the relatively low levels of sulfate in the leachate.

Average concentrations of cadmium in the coal ash landfill leachates exceeded MCLs.
None of the landfills had leachate with average chemical concentrations exceeding the
MCLs for any considered chemical except arsenic and cadmium.

E-6.2.4.3 C&DW

The leachates from the two C&DW landfills were similar to MSW leachates in terms of
inorganic and  organic chemistry.  With the exception of benzene, trichloroethylene, and
vinyl chloride,  the considered heavy metals and  VOCs were detected at concentrations
below MCLs.
                                 E-193
-------
E-6.3    Comparison to Published Data

The leachate chemistry data collected for this study and summarized in Table E-6.1
were compared to the published data presented in Table E-2.2. In general, the leachate
chemistry data collected for this study fall within the range of published data.  The
exception to this are for the ISW landfills, which have not been fully characterized in the
literature. Also, the chemistry of ISW landfills is highly variable depending on waste
type.  Based on published data, the concentrations of metals in MSW ash leachates can
be quite high, higher than those for MSW leachates and, sometimes, higher than those
for HW leachates. However,  the database for MSW ash landfills presented herein does
not support this.  The concentrations of heavy metals for the MSW ash leachates were
within the range of those for MSW leachates. The literature also shows that the heavy
metals concentrations in C&DW leachates can  be quite  high.  However, the limited
database for C&DW landfills didn't support this.  It may be that the total  metal content
disposed of in landfills has been reduced since the previous studies  due to community
recycling programs.

E-6.4    Effect of Regulations on Leachate Chemistry

As discussed in Section E-2.3.1, with the solid waste regulations promulgated in the
1980's and early 1990's, it is expected that the quality of landfill leachate would have
improved over time.  To evaluate whether this has occurred, the MSW and HW leachate
chemistry data collected for modern facilities in this appendix were compared to data
from older facilities.

The concentrations of the 30  considered constituents in leachates from  the older
modern  MSW landfills constructed prior to 1990 (pre-1990 landfills) and the newer MSW
landfills  constructed after 1990  (post-1990 landfills) are shown in Table  E-6.1.  The
average concentrations of selected chemicals in the  landfill cells are shown in Figure E-
6.4.  No major differences between the leachate chemistry for the  pre-1990 landfills and
the post-1990 landfills were apparent.  The chemistry data sets had  no statistically
significant differences in the concentrations of trace metals or VOCs at the 90%
confidence level, though average VOC concentrations were generally lower in leachate
from the post-1990 landfills. The statistical analysis findings were limited by the data.
The limited number of landfills contributing to each data set and the wide range of
chemical concentrations led to large confidence intervals for each  parameter in the data
sets.  To further evaluate the  differences in leachate chemistry between older and
newer MSW landfills, the data for the post-1990 MSW landfills were  compared to
published leachate chemistry data for 61 older MSW landfills (i.e.,  pre-1980 landfills in
NUS (1988) and pre-1985 landfills in Gibbons et al. (1992)). The distribution of the
leachate chemistry data for the  older MSW landfills was not known, so the two data sets
could not be compared statistically. However, the average concentrations of trace
metals and VOCs in leachate from the newer landfills were almost always less than the
average concentrations in leachate from the older landfills.  Based on the above, it
appears that the solid waste regulations  have resulted in improved MSW landfill
                                  E-194
-------
leachate quality.  However, more data are needed to quantify this improvement.  From
the published information summarized in this report, the regulations may have also
reduced the occurrence of certain chemicals.  For example, acetonitrile, cyanide, and
naphthalene were detected more frequently in leachate from older landfills than in
leachate from newer landfills.

Published leachate chemistry data for 33  older HW landfills (i.e., pre-1984 landfills in
Bramlett et al. (1987), pre-1983 landfills in NUS (1988), and pre-1987 landfills in
Gibbons et al. (1992)) were compared to the data presented for HW landfills in this
report (i.e., newer HW landfills). The data set for newer HW landfills is small; only
leachate chemistry data for four landfills are available.  The concentrations of chemicals
in leachate from the newer landfills were found to be within the range of published
values for the older landfills. The distribution of the leachate chemistry data for the
older HW landfills was not known, so the two data sets could not  be compared
statistically.  However, on average, most heavy metal concentrations and almost all
VOC concentrations were lower in leachate from the newer landfills.  This reduction  in
leachate strength is likely a result of the Subtitle C regulations and the Land Disposal
Restrictions.

E-7    Conclusions

E-7.1    Primary Liner Leakage Rates and Efficiencies

E-7.1.1   GM Primary Liners

Performance of GM primary liners was evaluated using LCRS and LDS flow rate data
from 31  double-lined landfill cells at 14 landfills monitored for periods of up to 114
months.  Formal  CQA programs were used in the construction of 23 cells that had
HOPE GM primary liners. Six of the eight cells that were constructed without a CQA
program used CSPE GM primary liners and the remaining two cells used HOPE GM
primary liners.  The major findings of the evaluation are summarized below:

   •  LDS flows during the initial period  of operation are attributed primarily to
       construction water and primary liner leakage. LDS flows during the active and
       post-closure periods are attributed primarily to  primary liner leakage.
   •  Average monthly LDS flow rates for cells constructed with a formal CQA
       program ranged from about 5 to 440 Iphd  during the initial period  of operation, 1
       to 360 Iphd during the active period, and 2 to 60 Iphd during the post-closure
       period. Peak monthly flow rates for these cells  were typically below 500 Iphd
       and exceeded 1,000 Iphd in only two of the 23 cells.
   •  Based on an analysis of the available data, average monthly active-period LDS
       flow rates through HOPE GM primary liners constructed with CQA (but without
       ponding tests or electrical leak location surveys) will often be less than 50 Iphd,
       but occasionally in excess of 200 Iphd. These flows are attributable primarily to
       liner leakage, and, for cells with sand LDSs, possibly construction water.
                                  E-195
-------
    •   The eight cells constructed without a formal CQA program exhibited average
       monthly LDS flow rates that are about one to two orders of magnitude greater
       than LDS flow rates for cells constructed with CQA. The average flow rates from
       the eight cells ranged from 120 to 2,140 Iphd during the  initial period of
       operation, 70 to 1,600 Iphd during the active period, and, for the two cells for
       which post-closure data are available, 210 to 240 Iphd during the post-closure
       period. The large differences in LDS flow rates between cells constructed with
       CQA and cells constructed without CQA are partly attributed to the benefits of CQA
       and partly due to differences in the GM materials and construction (i.e., seaming)
       methods. The two cells that had HOPE GM primary liners and no formal CQA had
       average LDS flow rates that are about two to seven times  greater than the mean
       LDS flow rate for all cells constructed with a formal CQA program. In contrast, the
       cells with CSPE  GM primary liners and no formal CQA exhibited average LDS flow
       rates that are about one to two orders of magnitude greater than the  mean LDS
       flow rate for all cells that had CQA. There are not sufficient data, however, in this
       appendix to accurately separate the effects of CQA and GM type (i.e., HOPE vs.
       CSPE) and construction methods on leakage rates through GM liners.
    •   Based on an analysis of the available data, GM liners can be constructed to
       achieve very good hydraulic performance (i.e., Et values greater than 99%).
       However, even with a CQA program, GM liners sometimes will  not achieve this
       performance level and lower Et values, in the range of about 90 to 99%, will occur.
       This relatively broad range of Et values is a consequence of the potential for even
       appropriately installed GMs to have an occasional small hole, typically due an
       imperfect seam,  but also potentially due to a manufacturing or construction-induced
       defect not identified by the CQA program.  Leakage can occur, relatively
       unimpeded, through a GM hole if the GM is not underlain by a low-permeability
       material such as a CCL or a GCL.  If a hole occurs at a critical location where  a
       sustained hydraulic head exists, such as in a landfill sump, the rate of flow through
       the hole can be significant. In contrast, the GCL or CCL component  of a composite
       liner can impede flow through a GM hole, even if it occurs  at a critical location.

The conclusion to be drawn  from the data evaluation is that single-liner systems with
GM liners (installed on top of a relatively permeable subgrade) should  not be used  in
landfill applications where Et values as low as 90% would be unacceptable, even if a
thorough CQA program is employed.  In these cases, single-composite liner systems or
double-liner systems should be utilized.  An exception to this conclusion may be made
for certain facilities, such as surface impoundments or small, shallow landfill cells, with
GM primary liners that can be field tested over the GM sheet and seams using electrical
leak location surveys, ponding tests, or other methods.  For these facilities, higher
efficiencies (i.e., greater than 99%) may be achieved with GM liners by identifying and
repairing the GM holes  during construction and, especially for surface  impoundments,
during operation. In all  cases, GM liners should be manufactured and  installed using
formal quality assurance programs.
                                  E-196
-------
E-7.1.2   Composite Primary Liners

Performance of composite primary liners was evaluated using LCRS and LDS flow rate
and chemical constituent data from 41 double-lined landfill cells monitored for periods of
up to 121 months. All 41 of the cells were constructed with formal CQA programs.  The
major findings are summarized below:

    •   For cells with composite liners, LDS flows during the initial period of operation
       are attributed primarily to construction water. LDS flows during the active and
       post-closure periods are primarily attributed to primary liner leakage and
       compression water for cells with GM/GCL primary liners or consolidation water
       (including secondary compression) for cells with GM/CCL or GM/GCL/CCL
       primary liners.
    •   LDSs underlying GM/GCL  composite liners exhibited average monthly flow rates
       of 0 to 290 Iphd during the  initial period of operation, 0 to 11 Iphd during the
       active period, and 0 to 2 Iphd (with many values reported as zero) during the
       post-closure period.
    •   Average monthly active-period LDS flow rates from  cells with GM/GCL primary
       liners constructed with CQA will often be less than 2 Iphd, but occasionally in
       excess of 10 Iphd.
    •   LDSs underlying composite liners with a CCL or GCL/CCL lower component
       exhibited average monthly  flow rates of about 10 to 1,400 Iphd during the initial
       period of operation, 0 to 370 Iphd during the active period of operation, and 5 to
       210 Iphd during the post-closure period.
    •   Given the "masking" effects of consolidation water, key chemical constituent data
       must be used to assess the hydraulic performance of composite primary liners
       having a CCL or GCL/CCL lower component.  This approach was applied to 13
       landfill cells. There were insufficient data for three of the cells to draw any
       conclusions.  For the remaining ten cells, key LCRS and LDS chemical
       constituent data did not reveal obvious indications of primary liner leakage;
       however, for five of these latter cells, the data were  of insufficient completeness
       and duration to quantify primary liner performance.  Et values were estimated for
       the remaining five cells.  The calculated values range from 99.1 to greater than
       99.9%.
    •   The data in this appendix suggest that GM/GCL, GM/CCL, and GM/GCL/CCL
       composite liners of the type evaluated in this study can be constructed to
       achieve Et values of 99.9% or more. However, Et values in the range of 99 to
       99.9% will also occur.  These high  efficiencies demonstrate that the low-
       permeability soil component of a composite liner is effective in impeding leakage
       through holes in the GM component of the liner.
   •   Available leakage rate calculation methods for composite liners give leakage
       rates in the same range as the rates estimated from the data for composite
       primary liners presented in this appendix.  Notwithstanding  the uncertainties in
       both the assumptions used in the calculations and the estimated leakage rates,
       this is a useful finding.
                                  E-197
-------
    •   In the U.S., landfill cells are typically operated for periods of one to five years,
       occasionally longer, and they are promptly covered with a GM or other low-
       permeability barrier after filling.  This operations sequence defines the timeframe
       for significant leachate generation in a landfill cell that does not contain liquid
       wastes or sludges and that does not undergo leachate recirculation or moisture
       addition.  For the cells in this study, estimated advective breakthrough times
       through CCLs,  assuming no chemical retardation, were generally calculated to
       range from about 3 to 12 years.  It thus appears that GM/CCL and
       GM/GCL/CCL composite liners are capable of substantially preventing leachate
       migration over the entire period of significant leachate generation for typical
       modern landfills.

Finally, it is recognized that the current database for the evaluation of composite liner
performance is limited, in terms of both completeness and duration of monitoring.  Key
constituents, such as alcohols and ketones that could be better organic "tracers" than
the aromatic hydrocarbons used in this appendix, are poorly represented in the
database.  It is important that additional  data be collected so that our understanding of
the performance characteristics of these systems can continue to improve.

E-7.2    Leachate Generation Rates

Leachate generation rates at 140 individually monitored cells located at 50 modern
landfills were evaluated.  About 52%  of the cells are MSW cells, 40% are HW cells, and
only 8% are ash cells and C&DW cells.  Most of the landfills (64%) are located in the
NE, 28% are located in the SE, and only 8% are located  in the W.  Leachate generation
rates for 33 closed MSW and HW cells,  located primarily in the NE and SE, were also
evaluated.  The monitoring periods were up to 8 years for active operation conditions
and up to 9 years for post-closure conditions.  The major findings of these evaluations
are summarized below.

    •   LCRS flow rates during operations can vary significantly between landfills
       located in the same geographic region and accepting similar wastes.  Large
       variations in flow  rates (e.g., one order of magnitude difference) can even occur
       between cells at the same landfill. Differences in waste placement practices
       may be responsible for these  significant variations.  Limiting the size of the
       active disposal area and using effective measures to minimize rainfall infiltration
       into the waste and to divert surface-water runoff away from the waste will
       significantly decrease leachate generation rates compared to the rates observed
       under less controlled conditions.
    •   Average LCRS flow rates for MSW landfills located in the NE and SE varied
       between 1,000 Iphd and 44,000 Iphd during the initial period of operation and
       between 40 Iphd  and  18,000 Iphd during the active period of operation.  For this
       group of landfills during the initial period of operation, 60% of the cells exhibited
       average LCRS flow rates less than 10,000 Iphd and only 13% had rates greater
       than 20,000 Iphd.  For the same group during the active period of operation,
                                  E-198
-------
52% of the cells had average LCRS flow rates less than 2,500 Iphd, 79% of the
cells had average LCRS flow rates less than 5,000 Iphd, and only 5% had
average LCRS flow rates greater than 10,000 Iphd. Only two MSW cells are
located in the W. These two cells had very low average LCRS flow rates (i.e.,
55 and 110 Iphd).
RF values calculated for the MSW cells in the NE (means of 39% and 13% for
the initial and active periods of operation, respectively) were higher than RF
values for the SE cells (means of 33% and 8% for the initial and active periods
of operation, respectively).  It is possible that the higher water evaporation rates
and the higher runoff occurring with the shorter duration, more intense rainfall
events associated with the  SE offset any potential increases in leachate
generation rates caused by the  higher total amount of rainfall in the SE as
compared to the NE. RF values for the two MSW cells that are located at an
arid site in the W were less than 1%.
Average flow rates from HW landfills during the active period of operation were
50 to 60% higher than flow rates from  MSW landfills. The reason for these
higher leachate generation rates at the HW landfills in this study is unclear, but
may, in part, be due to differences in waste characteristics and operational
practices. Stabilized HW may be wetter and/or have a lower water storage
capacity than MSW. In addition, in comparison to landfills, cover materials  are
less frequently used in HW landfills to  divert clean storm water from the waste.
The ten HW cells located in the W had low average flow rates during operations
(i.e.,  less than 4,000 Iphd).
RF values calculated for the HW landfills in the NE (mean = 46% and 21 % for
the initial and active periods operation, respectively) were higher than RF values
for the landfills in the SE (mean = 33% and  11% for the initial and active periods
of operation, respectively).  Similar to the MSW landfills, the HW landfills in the
SE had lower RF values than landfills  in the NE.  RF values for the HW cells in
the W were typically below 10%.
The limited number of ash and C&DW landfills considered in this study exhibited
LCRS flow rates during the active period of operation that were 300 to 600%
higher than flow rates from MSW and  HW landfills. These average rates during
operations were between 1,000 Iphd and 35,000 Iphd for the ash cells and
between 4,000 and 20,000 Iphd for the C&DW cells.  It is possible that
differences in the waste characteristics and disposal practices are responsible
for the higher LCRS flow rates for ash and C&DW landfills.
Mean RF values during operations were about 53% for ash cells and 43% for
C&DW cells.
Peak monthly LCRS flow rates were typically two to three times the average
monthly flow rates for all types of waste and regions of the U.S.
Landfill geographic region has a major impact on LCRS flow rates.  For landfill
sites with historical average annual rainfall less than 500 mm, average LCRS
flow rates were low,  typically less than 2,000 Iphd. LCRS flow rates increased
with increasing rainfall up to a point. In general, for landfill sites with historical
                            E-199
-------
       average annual rainfall greater than 1,100 to 1,200 Iphd an increase in rainfall
       did not appear to cause a corresponding increase in leachate generation rate.
    •   LCRS flow rates were typically two to three times smaller during the active
       period of operation than during the initial period of operation.
    •   Leachate generation rates for the closed landfills in this study typically
       decreased by a factor of four within one year after closure and by one order of
       magnitude within two to four years after closure.  Six years after closure, LCRS
       flow rates were between 5 and 1,200 Iphd (mean of 180 Iphd).  Nine years after
       closure, LCRS flow rates were negligible.  These data show that well designed
       and constructed final  cover systems can be very effective in minimizing
       infiltration of rainfall into the waste, thus reducing leachate generation rates to
       near-zero values.

E-7.3    Leachate Chemistry

Select leachate chemistry data for 59 cells at 50 double-lined landfills are presented and
evaluated in this appendix. Most of the  data are for MSW landfills: there are 36 MSW
landfills, four HW landfills, and eleven ISW landfills in the  database. Fewer data are
available for HW and ISW landfills.  In addition, the types  of wastes placed in HW and
ISW landfills are generally more variable between landfills than wastes placed in MSW
landfills.  With the exception of the leachate chemistry data for MSW ash landfills, it is
likely that the data presented in this appendix do not fully  characterize the variation in
leachate chemistry for HW and ISW landfills. The chemistry data for MSW ash landfill
leachate may be representative of modern MSW ash landfills in the U.S. because seven
landfills are included in the database and the chemistry of MSW ash is less variable
than HW.
It is recognized that the database on leachate chemistry is limited in terms of
completeness and duration of monitoring. In addition, key MSW and HW leachate
constituents, such as alcohols and ketones,  are poorly represented in the database.  It
is important that these additional data be collected so that our understanding of the
performance characteristics of composite liners can continue to improve.
From the evaluation of landfill leachate chemistry data presented herein, the following
conclusions are drawn:

    •   For a given waste type, many of the leachate constituents exhibited significant
       concentration variations (e.g., several orders of magnitude difference) between
       landfill cells and, sometimes, for a given cell.
    •   For the leachate types for which data are available for more that two landfills, the
       average value of pH (pH units), specific conductance (jimhos),  COD (mg/l),
       BOD5 (mg/l), TOC (mg/l), and chloride (mg/l) were, respectively:
       .   MSW leachate: 6.7, 4,470, 2,500, 1,440, 380, and 560;
           HW leachate: 8.2, 22,100, not available, not available, 1,620, and 7,760;
           and
       .   MSW ash leachate: 7.1,  22,100, 1,670, 55, 62, and 10,400.
                                   E-200
-------
The MSW landfill leachates were mineralized, biologically-active liquids with
relatively low concentrations of heavy metals and VOCs.  On average, the
leachates were slightly acidic (i.e., average pH of 6.7), which is expected
because carbon dioxide and organic acids are the primary by-products of the
first stage (i.e., the acid stage) of anaerobic degradation of organic compounds
in MSW landfills. The chemistry of these leachates changed with time as the
organic compounds degraded (see, for example, Table E-6.2).  In general, the
leachate characteristics for cells receiving waste were more indicative of the acid
phase of degradation than the second stage (i.e., the methane fermentation
phase) of anaerobic degradation.  For closed cells, the leachate pH typically
increased with time and the BOD/COD ratio decreased with time, which  is
expected as the landfill is more fully in the methane fermentation phase  of
degradation. Of the heavy metals and VOCs considered in Table E-6.1,
chromium, nickel,  methylene chloride,  and toluene were detected at the  highest
concentrations in MSW leachates.  Average concentrations of cadmium,
benzene, 1,2-dichloroethane, trichloroethylene, and vinyl chloride in MSW
landfills leachates exceeded federal maximum contaminant levels (MCLs) for
drinking water.  None of the landfills had leachate with average chemical
concentrations exceeding the MCLs for ethylbenzene, toluene,  or xylenes.
The HW landfill  leachates were more mineralized and had a higher organic
content than MSW leachates.  All of the HW leachates were alkaline, with pH
values ranging from 7.5 to 9.4.  One possible explanation for the alkaline pH
values is the relatively common practice of solidifying HWwith pozzolonic
additives prior to disposal. These relatively high pHs decrease the mobility of
metals.  Even so, the average heavy metals concentrations were generally
several times to several orders of magnitude higher in HW leachates as
compared to MSW leachates.  The HW leachates also had higher average
concentrations of all VOCs, except methylene chloride, toluene, and xylenes. Of
the heavy metals and VOCs considered in Table E-6.1, arsenic, nickel, 1,2-
dichloroethane,  and vinyl chloride were detected at the highest  concentrations in
HW leachates. Average concentrations of arsenic, cadmium, chromium,
benzene, 1,2-dichloroethane, trichloroethylene, and vinyl chloride in HW landfill
leachates exceeded MCLs.  None of the landfills had leachate with average
chemical concentrations exceeding the MCLs for ethylbenzene, toluene, or
xylenes.
The chemistry of the ISW landfill leachates was highly variable  due to the wide
variety of wastes disposed in ISW landfills. The pH values for these leachates
ranged from 6.4 to 7.7. The MSW ash leachates, the most mineralized of the
ISW landfill leachates, were even more mineralized than the  MSW leachates in
this study, as evidenced by the high specific conductance, TDS, sulfate,  and
chloride levels of the MSW ash leachates.  Coal ash leachates  were the least
mineralized. Both the MSW ash and coal ash leachates had low BOD values
that were several orders of magnitude less than the BOD values for MSW
leachate because  most of the organic materials originally in the MSW and coal
had been combusted. The average BOD value for C&DW leachate, however,
                           E-201
-------
was within range of values reported for MSW leachate. Heavy metals
concentrations in MSW ash and C&DW leachates were similar to those for MSW
leachates. Metals concentrations in coal ash leachate were lower, generally at
the lower end of the concentration range for MSW leachates.  As expected, the
MSW ash and coal ash leachates did not contain VOCs.  However, published
data show that MSW ash leachates can contain trace amounts of base neutral
extractables (BNAs), polychlorinated dibenzo-p-dioxins (PCDDs), and
polychlorinated dibenzo-furans (PCDFs).  The one C&DW landfill for which
organic chemistry data are available produced leachate containing VOCs.
Average concentrations of cadmium in MSW ash and coal ash landfill leachates
and benzene, trichloroethylene, and vinyl chloride concentrations in C&DW
landfill leachates exceeded MCLs.
In general, the leachate chemistry data collected for the study fall within the
range of published data.
With the federal solid waste regulations promulgated in the 1980's and early
1990's, it is expected that the quality of MSW and HW landfill leachates would
have improved over time.  No statistically significant differences in
concentrations of the considered trace metals or VOCs in leachates from older
modern MSW landfills constructed prior to 1990 (pre-1990 landfills) and
leachates from the newer MSW landfills constructed after 1990 (post-1990
landfills) were observed at the 90% confidence level.  However, average VOC
concentrations were generally lower in leachate from the post-1990 landfills
(Table E-6.1). The statistical analysis findings were limited by the data. The
limited number of landfills contributing to each data set and the wide range of
chemical concentrations led to large confidence intervals for each parameter in
the data sets. To further evaluate the differences in leachate chemistry between
older and newer MSW landfills, the data for the post-1990 MSW landfills were
compared to published leachate chemistry data summarized in Table E-2.1 for
61 older MSW landfills (i.e., pre-1980 landfills in NUS (1988) and pre-1985
landfills in Gibbons et al.  (1992)).  The distributions of the leachate chemistry
data for the older MSW landfills were not known, so the two data sets could not
be compared statistically.  However, the average concentrations of trace metals
and VOCs in leachate from the newer landfills were almost always less than the
average concentrations in leachate from the older landfills. Based on the above,
it appears that the solid waste regulations have resulted in improved MSW
landfill leachate quality. However, more data are needed to quantify this
improvement. From the published information summarized in this report, the
regulations may have also reduced the occurrence of certain chemicals.  For
example, acetonitrile, cyanide, and naphthalene were detected more frequently
in leachate from older landfills than in leachate from newer landfills .
Published leachate chemistry data summarized in Table E-2.1 for 33 older HW
landfills (i.e., pre-1984 landfills in Bramlett et al. (1987), pre-1983 landfills in
NUS (1988), and pre-1987 landfills in Gibbons et al. (1992)) were compared to
the data presented for HW landfills in this report (i.e., newer HW landfills). The
data set for newer HW landfills is small; only leachate chemistry data for four
                           E-202
-------
       landfills are available.  The concentrations of chemicals in leachate from the
       newer landfills were found to be within the range of published values for the
       older landfills.  The distribution of the leachate chemistry data for the older HW
       landfills was not known, so the two data sets could not be compared statistically.
       However, on average, most heavy metal concentrations and almost all VOC
       concentrations were lower in leachate from the newer landfills.

E-8    References

Baker, J. (1987) personnel communication from J. Baker (WMI) to S. Monney (EPA)
    concerning MSW landfill leachate data, 1 Dec.
Bergstrom, W.R., Takacs, M.J., and Olson, M.A. (1993) "Behavior of Five Operating
    Landfill Leak Detection Systems", Proceedings,  Waste Tech '93, Marina Del Rey,
    CA, pp. 2.2-2.14.
Bonaparte, R. and Gross, B.A. (1990) "Field Behavior of Double-Liner Systems", Waste
    Containment Systems: Construction, Regulation, and Performance, ASCE
    Geotechnical Special Publication No. 26, R. Bonaparte, ed., New York,  pp. 52-83.
Bonaparte, R. and Gross, B.A. (1993) "LDCRS Flows from Double-Lined Landfills and
    Surface Impoundments", EPA/600/SR-93/070, EPA Risk Reduction Engineering
    Laboratory, Cincinnati, OH, 65 p.
Bonaparte, R., Othman, M.A., Rad, N.S., Swan, R.H., and Vander Linde, D.L. (1996)
    "Evaluation of Various Aspects of GCL Performance", Appendix F in Report of 1995
    Workshop on Geosynthetic Clay Liners,  D.E. Daniel and H.E. Scranton, authors,
    EPA/600/R-96/149, EPA National Risk Management Research  Laboratory,
    Cincinnati, OH, pp. F1-F34.
Bramlett, J., Furman,  C., Johnson, A., Ellis, W.D., Nelson, H. and Vick, W.H. (1987)
    "Composition ofLeachates from Actual Hazardous Waste Sites", EPA Office of
    Research and Development, Cincinnati, OH, EPA/600/2-87/043, 113 p.
Brown, K.W. and Donnelly K.C., (1988) "An Estimation of the Risk Associated with the
    Organic Constituents of Hazardous and  Municipal Waste Landfill Leachates",
    Hazardous Waste & Hazardous Materials, Vol. 5, No. 1,  pp. 1-30.
Chian, E.S.K. and DeWalle, F.B., (1997) "Evaluation of Leachate Treatment, Volume I,
    Characterization of Leachate", EPA Office of Research and Development,
    EPA/600/2-77/186a, Cincinnati, OH.
Daniel, D.E. and Koerner, R.M. (1995) 'Waste Containment Systems:  Construction
    Quality Assurance, and Quality Contror, ASCE Press, New York, 354 p.
Darilek, G.T., Laine, D.L., and Parra, J.O. (1989) "The Electrical Leak Location Method for
    Geomembrane Liners: Development and Applications", Proceedings, Geosynthetics
    '89 Conference, San Diego, CA, Vol. 2, pp. 456-466.
Darilek, G.T., Merizel,  R., and Johnson, A. (1995) "Minimizing Geomembrane Liner
    Damage While Emplacing Protective Soil", Proceedings, Geosynthetics '95
    Conference, Nashville, Tennessee, Vol. 2, pp. 669-676.
Ehrig, H.J. and Scheelhaase, T. (1993) "Pollution Potential and Long Term Behavior of
    Sanitary Landfills", Proceedings of the Fourth International Landfill Conference,
    Sardinia, Italy, pp. 1203-1225.
                                  E-203
-------
Eisenberg, S.H., Tittlebaum, M.E., Eaton, H.C., and Soroczak, M.M., (1986) "Chemical
    Characteristics of Selected Flyash Leachates", Journal of Environmental Health and
    Science, Part A, Vol. 21, No. 4, pp. 383-402.
EPA (1987a) "Background Document: Proposed Liner and Leak Detection Rule",
    EPA/530-SW-87-015, prepared by GeoSyntec Consultants, 526 p.
EPA, (1987b) "Test Methods for Evaluating Solid Waste, Physical/Chemical Methods",
    EPA Publication SW-846, Third Edition, 1986 and Revision I, 1987.
EPA (1988) "Report to Congress: Solid Waste Disposal in the United States",  Vols. I
    and II, revised final draft, EPA Office of Solid Waste, Washington, D.C.,
    unpublished.
EPA (1989) "Cooperative Industry Report,  Leachate  Treatability Study - Task 1 -
    Leachate Characterization Analytical Results", EPA Office of Solid Waste,
    Washington, D.C.
EPRI (1978) "The Impact of RCRA (PL 94-580) on Utility Solid Wastes", EPRI FP-878,
    EPRI,  Palo Alto, California, 133 pp.
Feeney, M.T. and Maxson, A.E. (1993) "Field Performance of Double-Liner Systems",
    Proceedings, Geosynthetics '93 Conference, Vol. 3, Vancouver, B.C., pp.  1373-
    1387.
GeoSyntec Consultants (1993), Confidential Report on Coal Ash Landfill.
Gibbons, R.D., Dolan, D., Keough, H., O'Leary, K., and O'Hara, R. (1992) "A Comparison
    of Chemical Constituents in Leachate From Industrial Hazardous Waste & Municipal
    Solid Waste Landfills", Proceedings, Fifteenth Annual Madison Waste Conference,
    Madison, Wl, pp. 251-276.
Giroud, J.P. (1984) "Impermeability: The Myth and a Rational Approach", Proceedings,
    International Conference on Geomembranes, Denver, CO, Vol. 1, pp. 157-162.
Giroud, J.P. and Bonaparte, R. (1989) "Leakage Through Liners Constructed with
    Geomembranes, Part II: Composite Liners", Geotextiles and Geomembranes, Vol.
    8, No.  2, pp. 77-111.
Giroud, J.P, Badu-Tweneboah, K., and Bonaparte, R. (1992) "Rate of Leakage Through
    a Composite Liner Due to Geomembrane Defects",  Geotextiles and
    Geomembranes, Vol. 11, No. 1, pp.  1-28.
Gross, B.A., Bonaparte, R., and Giroud, J.P. (1990) "Evaluation of Flow From  Landfill
    Leakage Detection Layers", Proceedings, Fourth International Conference on
    Geotextiles, Vol. 2, The Hague, pp. 481-486.
Haikola, B.M., Loehr, R.C., Daniel, D.E.  (1995) "Hazardous Waste Landfill Performance
    as Measured by Primary Leachate Quantity", Proceedings, Geoenvironment 2000,
    New Orleans, LA , Vol. 1, ASCE Special Topic Paper No. 46, pp. 554-567.
Hunt, W.C. and Dollins,  M. (1996) "Continuation of the Study of Leachate Quantity
    Generation Data Obtained From a Subtitle D Landfill Cell", Proceedings Texas Solid
    Waste Management Conference, Austin, TX, pp. 97-107.
Laine, D.L. (1991) "Analysis of Pinhole Seam Leaks Located in Geomembrane Liners
    Using the Electrical Leak Detection Method: Case Histories", Proceedings,
    Geosynthetics '91  Conference, Atlanta,  GA, Vol. 1, pp. 239-253.
                                  E-204
-------
Laine, D.L. and Miklas, M.P. (1989) "Detection and Location of Leaks in Geomembrane
    Liners Using Electrical Method: Case History", Proceedings, 10th National Superfund
    Conference, Washington, D.C., pp. 35-40.
Laine, D.L., and Darilek, G.T. (1993) "Locating Leaks in Geomembrane Liners of Landfills
    Covered with Protective Soil", Proceedings, Geosynthetics '93 Conference,
    Vancouver, B.C., Vol. 3, pp. 1403-1412.
Landreth, R.E. (1989) "Locating and Repairing Leaks in Landfill/Impoundment Flexible
    Membrane Liners", Proceedings, Geosynthetics '89 Conference, San Diego, CA, Vol.
    2, pp. 467-477.
Maule, J., McCulloch, J., and Lowe, R. (1993) "Performance Evaluation of Synthetically
    Lined Landfills", Proceedings, 1993 GRI Seminar, Philadelphia, PA, pp. 80-90.
McGinley, P.M. and Kmet,  P. (1984) "Formation, Characteristics, Treatment and Disposal
    ofLeachate from Municipal Solid Waste Landfills",  Bureau of Solid Waste
    Management, Wisconsin Department of Natural Resources, Madison, Wl, 288 p.
Norstrom, J. M., Williams,  C.E., and Pabor, P.A. (1991) "Properties of Leachate From
    Construction/ Demolition Waste Landfills", Proceedings, Fourteenth Annual Madison
    Waste Conference, Madison,  Wl, pp. 357-366.
NUS (1986) "Determination of Municipal Landfill Leachate Characteristics, Leachate
    Baseline Report", EPA Office of Solid Waste, Washington, D.C., unpublished.
NUS (1987a) "Characterization ofLeachates From Municipal Waste Disposal Sites and
    Codisposal Sites, Volume 6", EPA Office of Solid Waste and Emergency Response,
    Washington, D.C.
NUS (1987b) "Characterization of MWC Ashes and Leachates from MSWLandfills,
    Monofills, and Co-Disposal Sites", EPA Office of Solid Waste and Emergency
    Response, Washington, D.C., EPA/530/SW-87/028A.
NUS (1988) "Draft Background Document Summary of Data on Municipal Solid Waste
    Landfill Leachate Characteristics "Criteria for Municipal  Solid Waste Landfills" (40
    CFR Part 258) Subtitle D of Resource Conservation and Recovery Act (RCRA)",
    EPA Office of Solid Waste, Washington, D.C.,  EPA/530-SW-88-038.
NUS (1990) "Characterization of Municipal Waste Combustion Ash, Ash Extracts, and
    Leachates", EPA and Coalition on Resource Recovery and the Environment.
Pavelka, C.,  Loehr, R.C.,  and Haikola, B. (1994), "Hazardous Waste Landfill
    Characteristics", Waste Management, Vol. 13,  No. 8, pp. 573-580.
Peggs, I.D. (1990) "Detection and  Investigation of Leaks in Geomembrane Liners",
    Geosynthetics World, Vol.  1, No. 2, pp. 7-14.
Peggs, I.D. (1993) "Practical Geoelectric Leak Surveys with Hand-Held, Remote and
    Water Lance Probes",  Proceedings, Geosynthetics '93 Conference, Vol.  3, Vancouver,
    B.C., pp. 1523-1532.
Piper, A.M. (1944) "A Graphic Procedure in the Geochemical Interpretation of Water
    Analyses", Transactions of American Geophysical Union, Vol. 25, pp. 914-923.
Rowe, R.K. (1995) "Leachate Characteristics for MSW Landfills", Geotechnical Research
    Centre Report GEOT-8-95, Department of Civil Engineering, The University of
    Western Ontario, London,  Ont, 19 p.
                                  E-205
-------
Sabel, G.V. and Clark, T.P. (1984) "Volatile Organic Compounds as Indicators of
    Municipal Solid Waste Leachate Contamination", Waste Management & Research,
    No. 2, pp. 119-130.
Tedder, R.B. (1992) "A Comparison of Florida Landfill Leachate with Regulatory
    Standards and EPA Landfill Leachate", Florida Department of Environmental
    Regulation.
Tedder, R.B. (1997) "Evaluating the Performance of Florida Double-Lined Landfills",
    Proceedings, Geosynthetics 97 Conference, Long Beach, CA, Vol. 1, pp. 425-438.
Tharp, L.D. (1991) "Leachate Characteristics for Missouri Sanitary Landfills", Proceedings,
    Fourteenth Annual Madison Waste Conference, Madison, Wl, pp. 313-326.
TRW (1983) "Compilation of Hazardous Waste Leachate Data", EPA Office of Solid
    Waste, Washington,  D.C.
Workman, J.P (1993) "Interpretation of Leakage Rates in Double-Lined Systems",
    Proceedings,  7th Geosynthetic Research Institute Seminar, Drexel University,
    Philadelphia, PA, pp. 91-108.
                                  E-206
-------
                   Appendix F
      Waste Containment Systems:
     Problems and Lessons  Learned
                           by

Beth A. Gross, P.E.       Rudolph Bonaparte, Ph.D., P.E.    J. P. Giroud, Ph.D.
GeoSyntec Consultants     GeoSyntec Consultants        J. P. Giroud, Inc.
Austin, Texas 70746      Atlanta, Georgia 30342        Ocean Ridge, Florida 33435
                      performed under

               EPA Cooperative Agreement Number
                      CR-821448-01-0
                       Project Officer

                    Mr. David A. Carson
           United States Environmental Protection Agency
               Office of Research and Development
           National Risk Management Research Laboratory
                    Cincinnati, OH 45268
-------
                               Appendix F
   Waste Containment Systems: Problems and Lessons Learned

F-1       Introduction

F-1.1     Appendix Purpose and Scope
This appendix presents the results of an investigation into problems that have occurred
in waste containment systems (i.e., liner systems and final cover systems (hereafter
referred to as cover systems)) for 69 modern landfill and five modern surface
impoundment facilities located throughout the United States (U.S.). The term "modern
facility" refers to a facility designed with components substantially meeting current U.S.
Environmental Protection Agency (EPA) regulations (e.g., 40 CFR 258 for municipal
solid waste (MSW) disposal facilities or 40 CFR 264 for Resource Conservation and
Recovery Act (RCRA) hazardous waste (HW) disposal facilities) and constructed and
operated to the U.S. state of practice from the mid-1980's forward.  The purpose of the
study is twofold: (i) to better understand the nature, frequency, and significance of
identified problems; and (ii) to develop recommendations to reduce the future
occurrence of problems.

This appendix specifically excludes consideration of problems in older waste
containment systems not designed and constructed to current standards and
practices.  These problems include, for example,  the leachate collection and removal
system (LCRS) and cover system  internal drainage layer failures described by Bass
(1986), Ghassemi et al. (1986),  and Kmet et al. (1988). The appendix also does not
address foundation stability problems at older landfills, such as the problems
described by Oweis (1985), Dvirnoff and Munion  (1986), Richardson and Reynolds
(1991), Kenteretal. (1997), Stark and Evans (1997), and  Schmuckerand Hendron
(1997).  Problems at older facilities are often not  relevant to current standards and
practices.

F-1.2     Appendix Organization

This  appendix is organized as follows:

•  data on waste containment system problems  are presented in Subsection F-2;
•  the nature, frequency, detection, and remedy  of the identified problems are
   discussed in Subsection F-3;
•  the significance of the identified problems is discussed in Subsection F-4;
•  conclusions from this study are presented in Subsection F-5;
•  recommendations to reduce  the future occurrence of the identified problems are
   presented in Subsection F-6;
•  references are provided in Subsection  F-7; and
•  case histories of the identified problems are presented in Attachment F-A.

                                    F-1
-------
F-1.3     Terminology

Waste containment systems consist of liner systems and cover systems (Figure
F-1.1). Modern landfills and surface impoundments (hereafter referred to as
impoundments) both have liner systems that underlay the wastes placed in them
(Figure F-1.2). A liner system consists of a combination of one or more drainage layers
and low-permeability barriers (i.e., liners). A landfill single-liner system consists of a
liner overlain by an LCRS drainage layer. A landfill double-liner system consists of
primary and secondary liners, with a leak detection system (LDS) drainage layer
between the two liners and an LCRS drainage layer above the primary liner. Besides
drainage layers, the LCRS and LDS may also contain networks of perforated pipes,
sumps, pumps, flowmeters,  and other flow conveyance and monitoring components. A
liner system may  also include a protection layer over the LCRS drainage layer to further
isolate the liner from the environment (e.g., freezing temperature, stresses from
equipment).  Impoundment  liner systems are similar to those for landfills except that
they do not have an LCRS.
       Side Slope
      Side Slope
                     ^cr
                            Base

Figure F-1.1.  Waste containment systems.

Once an area of a landfill is filled to final grade, a cover system is constructed over
the area to contain the waste, minimize the infiltration of water into the waste, and
control the emissions of gases produced by waste decomposition or other
mechanisms.  In contrast to landfills, impoundments are typically clean closed (i.e.,
the impoundment is  removed and the site is reclaimed).  A cover system consists of
up to six basic components, from top to bottom: (i) surface layer; (ii)  protection layer;
(iii) drainage layer; (iv) barrier; (v) gas collection layer; and (vi) foundation layer.  In
some cases, the functions of several adjacent components can be provided by one
soil layer.  For example, a sand gas collection layer may also serve as a foundation
layer.

                                    F-2
-------
     Cover
    System
  Single-Liner
     System
     (a)
MMMM

 Surface and Protection Layer

     Drainage Layer
                Gas Collection Layer
                                 Barrier
                  Solid Waste
                LCRS Drainage Layer
                 Composite Liner
                                      Double-Liner
                                         System
                                                       Liquid Waste
                                                      Protection Layer
                                                                      Primary Liner
                                        LDS Drainage Layer
                                       Composite Secondary Liner
                              (b)
Figure F-1.2. Typical waste containment system components for landfills (a) and
    impoundments (b).

In general, the materials used to construct liners and barriers in modern waste
containment systems are geomembranes (GMs) alone and composites consisting of
GMs overlying compacted clay liners (CCLs) or geosynthetic clay liners (GCLs) (i.e.,
GM/CCL or GM/GCL composites). Drainage layers and gas collection layers are
typically constructed with sand, gravel, geonets (GNs), or geotextile (GT)/GN
composites (i.e., geocomposites (GCs)). Protection layers typically consist of soil or
thick GTs. For landfills, the protection layer over the LCRS drainage layer sometimes
consists of select waste. Surface layers for cover systems are typically constructed
with vegetated topsoil.

Liner systems for modern MSW landfills and nonhazardous MSW combustor ash
(MSWash) landfills must, based on state-specific implementation of federal RCRA
Subtitle D requirements, meet federal minimum design criteria or performance-based
design requirements (40 CFR 258.40).  The federal minimum design standard for new
MSW landfills and MSWash landfills requires a single-composite liner system  that
consists of the following, from top to bottom:

•   LCRS that limits the head of leachate on the composite liner to 0.3 m or less;
•   0.75-mm thick GM; and
•   0.6-m thick CCL with a maximum hydraulic conductivity of 1 x 10~7 cm/s.

If the GM is made of high-density polyethylene (HOPE), the GM must be at least 1.5
mm thick. While the federal minimum design criteria were adopted by many states, a
                                     F-3
-------
few states require that MSW landfills or MSWash landfills have a double-liner system.
The performance standard requires a liner system design that is demonstrated to
achieve certain groundwater compliance standards at a specified distance from the
landfill. This distance cannot exceed 150 m.

For RCRA HW landfills and impoundments, federal regulations (40 CFR 264) require a
double-liner system with at least the following components, from top to bottom:

•  for landfills, LCRS that limits the head of leachate on the primary liner to 0.3 m or
   less;
•  GM;
•  0.3-m thick granular LDS drainage layer with a minimum hydraulic conductivity of
   1 x 10~2 cm/s or a geosynthetic LDS drainage layer with a minimum hydraulic
   transmissivity of 3 x 10~5 m2/s;
•  GM; and
•  0.9-m thick CCL with a maximum hydraulic conductivity of 1 x 10~7 cm/s.

Cover systems for modern lined MSW  landfills and MSW ash landfills (40 CFR
258.40) must meet federal minimum design criteria or performance-based design
requirements (40 CFR 258.60).  The cover system meeting federal minimum design
criteria consists of the following, from top to bottom:
•  0.15-m thick soil surface layer;
•  0.5-mm thick GM; and
•  0.45-m thick CCL with a maximum hydraulic conductivity of 1x10  cm/s.

If the GM is made of HOPE, the GM must be at least 1.5 mm thick. The performance-
based cover system must perform equivalently to the cover system meeting federal
minimum design criteria with respect to reduction in infiltration and erosion protection.

For RCRA HW facilities, EPA gives cover system performance requirements in 40 CFR
264; there are no federal design criteria for cover systems for these facilities.  However,
EPA guidance (EPA, 1989) recommends that the cover systems for HW facilities
consist of the following, from top to bottom:

•  0.6-m thick soil surface and protection layer;
•  0.3-m thick granular drainage layer with a minimum hydraulic conductivity of
   1 x 10~2 cm/s or a geosynthetic drainage layer with a minimum  hydraulic
   transmissivity of 3 x 10"5 m2/s;
•  0.5-mm thick GM; and
•  0.6-m thick CCL with a maximum hydraulic conductivity of 1 x 10~7 cm/s.
                                     F-4
-------
For industrial solid waste (ISW) facilities, such as papermill landfills, coal ash landfills,
and construction and demolition waste (C&DW) landfills, EPA gives general
performance requirements (e.g., "provide adequate protection to ground and surface
waters") in 40 CFR 241.  Currently, there are no federal design criteria for liner
systems or cover systems for these facilities.
F-2       Data on Waste Containment System Problems

F-2.1     Data Collection Methodology

The data on waste containment system problems presented in this appendix were
obtained from the technical literature and from discussions with facility owners, facility
operators, design engineers, and federal and state regulators throughout the U.S.
The data were collected in accordance with a quality assurance plan, which was
reviewed and approved by the EPA.  Efforts were made to obtain information on
problems at ISW, MSW, and HW facilities and at facilities with different types of
waste containment system components and sites (e.g., facilities constructed on flat
terrain, in excavations, and in canyons,  and landfills constructed over existing
landfills). The investigation focused on  landfills, which resulted in more problems
being identified at landfills than at impoundments.  Based on the broad-based
method of data collection for this study,  it is believed that the problems in this
appendix are  representative of those for waste containment systems in landfills
nationwide. The study of impoundments was more limited and may not include
some of the more common impoundment problems. However, some impoundment
problems that have occurred but were not identified in this study may also have
occurred at landfills (e.g., operational problems related to the LDS). Thus, the
information on landfill problems in this appendix can also be used to identify
problems that may have occurred at impoundments.

F-2.2     Detection of Problems

Problems in waste containment systems are typically detected by visual observation
or an evaluation of monitoring  data.  Visual observation is the primary method of
detecting problems during construction of liner systems and cover systems.  For
example, visual observation is used to detect excessive GM wrinkling during
construction and uplift of geosynthetics  by groundwater. Leak location surveys and
other monitoring procedures can also be used to supplement visual observation
during construction.  For example, leak  location surveys are used to detect leaks
around pipe penetrations of liners.  Common liner system and cover system
nonconformities detected during construction quality assurance (CQA) monitoring
and repaired in accordance with the CQA plan, however, are generally not problems.
These nonconformities, such as CCL lifts  not compacted to specification and
                                    F-5
-------
defective GM seams, are part of normal construction.  During operation and after
closure of landfills, liner systems are covered with waste and cannot be visually
inspected. Then, evaluation of monitoring data is the primary method of detecting
problems. Visual observation is also important, however, for detecting problems that
are expressed at the surface of the waste containment system, such as erosion and
slope failure.

The waste containment system monitoring data typically collected and evaluated
during facility operation and closure are:

•  LCRS flow quantity and quality data;
•  LDS flow quantity and quality data;
•  groundwater quality data (at groundwater monitoring wells); and
•  landfill gas quality data (at gas monitoring wells).

LCRS data are used to evaluate: (i) effectiveness of runon controls and other
leachate minimization practices; and (ii) treatment requirements for leachate before
it may be discharged.  LDS data are used to evaluate whether primary liner leakage
is occurring.  Additional information on LDS flow rates relevant to the evaluation of
some of the case histories in this appendix is presented below. Groundwater and
gas data are used to evaluate whether migration of contaminants is occurring from
waste containment systems.

All liners in modern waste containment systems for HW landfills and for MSW
landfills meeting federal design requirements include a GM.  In a study of LDS flow
rates from double-lined waste containment systems, Bonaparte and Gross (1993)
found that all  landfill cells with GM primary liners appeared to have exhibited primary
liner leakage. Installed GMs typically have a small number of holes (e.g., 1 to 10 per
hectare) due to field seaming flaws and construction-related damage.  Leakage
through GMs primarily occurs by advection through these holes. The  leakage rate
through a hole increases as hole size and head on top of the hole increase.  If the
GM is underlain by a  low-permeability soil layer (i.e., CCL or GCL) to form a
composite liner, the rate of leakage decreases with decreasing hydraulic conductivity
of the soil layer. For a given head and hole size, the rate of leakage through a
composite liner is approximately 100 to 10,000 times less than that through a GM
alone. The head of leachate on liners is usually much smaller on side slopes than
on the base.   Therefore, all other things being equal, the leakage rate is greater on
the base than on the side slopes.  But if a composite liner is on the base and GM
liner is on the side slopes, the leakage rate is greater on the side slopes.
                                    F-6
-------
It is sometimes difficult to use LDS flow rates to evaluate if, and how much, primary
liner leakage is occurring because there are potential sources of LDS flow other than
leakage (Gross et al., 1990).  These sources are:

•  water (mostly rainwater) that infiltrates the LDS during construction and
   continues to drain to the LDS sump after the start of facility operation
   ("construction water");
•  water that infiltrates the LDS during construction, is held in the LDS by capillarity,
   and is expelled from the LDS as a result of LDS compression under the weight of
   the waste ("compression water");
•  water expelled into the LDS from the CCL component of a composite primary
   liner as a result of clay consolidation under the weight of the waste
   ("consolidation water");  and
•  water that percolates through the secondary liner and infiltrates the LDS
   ("infiltration water").

LDS flow quality data can be used and compared to LCRS flow  quality data to help
in the assessment of primary liner leakage.  LDS flow data are used to evaluate
whether primary liner leakage is occurring, but do not provide information on the leak
location.

F-2.3     Problem Classification

The types of problems identified during the investigation for this appendix are
categorized on the basis of two criteria. The first criterion addresses the component
or attribute of the landfill  liner system, landfill cover system, or impoundment liner
system affected by the problem. The specific components and attributes considered
in this study are: (i) landfill liner construction;  (ii) landfill liner degradation; (iii) landfill
LCRS or LDS construction; (iv) landfill LCRS or LDS degradation; (v) landfill LCRS
or LDS malfunction;  (vi) landfill LCRS or LDS operation; (vii) landfill liner system
stability; (viii) landfill liner system displacement; (ix) cover system construction; (x)
cover system degradation;  (xi) cover system stability; (xii) cover system
displacement; (xiii) impoundment liner construction; (xiv) impoundment liner
degradation; (xv) impoundment LDS; and (xvi) impoundment liner system stability;
and (xvii) impoundment liner system displacement.  Specific problems that may
affect these components and attributes and the significance of these problems are
discussed in Section 4.  Other components or attributes not specifically associated
with  landfill or impoundment integrity were not considered in the investigation.
These include landfill daily  and intermediate cover components (except for cracking
of soil intermediate cover from Northridge earthquake), leachate transmission and
treatment components beyond the leachate collection sumps or manholes, and
landfill gas extraction and management components.
                                    F-7
-------
The second criterion used to categorize the problem addresses the principal human
factor contributing to the problem. The principal human factors considered are: (i)
design; (ii) construction; and (iii) operation. While a principal human factor has been
assigned to each problem, it should be recognized that most problems have
complex causes and several contributing factors. Hereafter, the problem
classifications are shown as "component or attribute criterion'Tprincipal human
factor criterion"  (e.g., landfill liner system stability/design).

F-2.4     Problem Description

This investigation found 74 modern landfill and impoundment facilities that had
experienced waste containment system problems. This number of facilities is relatively
small in comparison to the total number of modern facilities nationwide. There are
approximately 3,900 active MSWand HW landfills and HW surface impoundments
nationwide (EPA, 1996; EPA, 1997); based on interviews with  regulators, it is
estimated that over half of these facilities (i.e., about 2,000) would be considered
"modern" using the criteria identified in Section F-1.1.  These numbers do not
include industrial landfills and surface impoundments. The search for problem
facilities for this investigation was not exhaustive, and it is certain that there are
other facilities that experienced problems similar to those described  in this  appendix.

Case histories of the problems are presented in Attachment F-A to this appendix.
Each case history includes a summary of the problem, information on how the problem
was resolved, and lessons learned for future projects. Landfills with liner system or
cover system  problems are designated with  an "L" or a "C", respectively (e.g.,  L-1).
Impoundments are designated with an "S".  The classification of the identified problems
at each facility, information source, and section of the attachment that describes each
problem are given in Table F-2.1.  The detailed references for the information sources
are listed in Attachment F-A  along with each case history. These references  are
repeated in Chapter F-7 of this appendix. Summaries of the problems are presented in
Table F-2.2. In Table F-2.3,  the problems are grouped by the classification criteria
presented in Section F-2.3.
Table F-2.1. Classification of Identified Problems at Landfill and Impoundment
    Facilities.
Facility
Designation
L-1
L-2
L-3
L-4
Information Source
Laineand Darilek (1 993)
Basnett and Bruner (1 993)
Darilek etal. (1995)
Adams etal. (1997)
Problem Classification
landfill liner construction/construction
landfill liner construction/construction
landfill liner construction/construction
landfill liner degradation/operation
Attachment
Section
F-A.2.1
F-A.3.1
F-A.2.2
F-A.3.2
                                     F-8
-------
Table F-2.1.  Classification of Identified Problems at Landfill and Impoundment
    Facilities (Continued).
Facility
Designation
L-5
L-6
L-7
L-8
L-9
L-10
L-11
L-12
L-13
L-14
L-15
L-16
L-17
L-18
L-19
L-20
L-21
L-22
L-23
L-24
L-25
Information Source
for F-A.2.3, Silva (1995) and
Tedder (1997)
forF-A.7.1, unpublished
Anderson (1993)
Loewenstein and Smrtic
(1994)
unpublished
unpublished
unpublished
unpublished
unpublished
Tisingeretal. (1993)
Tisingeretal. (1994)
unpublished
unpublished
Bonaparte and Gross (1 993)
Bonaparte and Gross (1993)
Paulson (1993)
unpublished
unpublished
unpublished
unpublished
Koerner et al. (1 993)
unpublished
unpublished
Mitchell etal. (1990);
Seed etal. (1990);
Byrne et al. (1 992)
Problem Classification
landfill liner construction/construction
landfill LCRS or LDS operation/operation
landfill liner construction/operation
landfill liner construction/operation
landfill liner construction/design
landfill liner construction/construction
landfill LCRS or IDS malfunction/design
landfill liner system displacement/design
landfill LCRS or LDS construction/
construction
landfill liner construction/construction
landfill liner construction/construction
landfill LCRS or LDS malfunction/design
landfill liner system displacement/design
landfill liner degradation/design
landfill LCRS or LDS malfunction/
operation
landfill LCRS or LDS malfunction/
construction
landfill liner degradation/operation
landfill liner construction/construction
landfill LCRS or LDS construction/
construction
landfill LCRS or LDS construction/
construction
landfill liner construction/construction
landfill LCRS or LDS malfunction/
construction
landfill liner construction/construction
landfill liner construction/construction
landfill liner degradation/construction
landfill liner system stability/design
landfill LCRS or LDS malfunction/
construction
landfill LCRS or LDS operation/operation
landfill liner system stability/operation
landfill liner system displacement/design
landfill liner system stability/design
Attachment
Section
F-A.2.3
F-A.7.1
F-A.2.4
F-A.2.5
F-A.2.6
F-A.2.7
F-A.5.1
F-A.9.1
F-A.4.1
F-A.2.8
F-A.2.9
F-A.5.2
F-A.9.2
F-A.3.3
F-A.6.1
F-A.5.3
F-A.3.4
F-A.2.10
F-A.4.2
F-A.4.3
F-A.2.11
F-A.5.4
F-A.2.12
F-A.2.13
F-A.3.5
F-A.8.1
F-A.6.2
F-A.7.2
F-A.8.2
F-A.9.3
F-A.8.3
                                   F-9
-------
Table F-2.1.  Classification of Identified Problems at Landfill and Impoundment
    Facilities (Continued).
Facility
Designation
L-26
L-27
L-28
L-29
L-30
L-31
L-32
L-33
L-34
L-35
L-36
L-37
L-38
L-39
L-40
L-41
L-42
L-43
L-44
L-45
L-46
C-1
C-2
C-3
C-4
C-5
C-6
Information Source
Anderson (1995);
Augelloetal. (1995);
Matasovic et al. (1995);
Matasovic and
Kavazanjian (1996);
Matasovic et al. (1998);
Stewart et al. (1 994)
Anderson (1995);
Augelloetal. (1995);
Chang etal. (1996);
Matasovic et al. (1 995)
unpublished
Koerneretal. (1998)
unpublished
unpublished
unpublished
unpublished
unpublished
unpublished
unpublished
unpublished
Boschuk(1991);
Giroud(1993)
Soong and Koerner (1997)
Soong and Koerner (1997)
Soong and Koerner (1997)
Soong and Koerner (1997)
unpublished
unpublished
Hullings and Sansone
(1996)
unpublished
Harris etal. (1992)
unpublished
Paulson (1993)
Bonaparte et al. (1996);
Vander Linde et al. (1 998)
Boschuk(1991)
Boschuk(1991)
Problem Classification
landfill liner system stability/design
landfill liner system stability/design
landfill LCRS or IDS construction/
construction
landfill liner construction/construction
landfill LCRS or IDS operation/
construction
landfill liner system displacement/design
landfill LCRS or LDS construction/
construction
landfill LCRS or LDS construction/
construction
landfill LCRS or LDS operation/operation
landfill LCRS or LDS operation/design
landfill LCRS or LDS malfunction/design
landfill LCRS or LDS malfunction/
operation
landfill liner system stability/design
landfill liner system stability/design
landfill liner system stability/design
landfill liner system stability/design
landfill liner system stability/operation
landfill liner degradation/construction
landfill liner degradation/design
landfill liner system stability/operation
landfill liner system stability/design
cover system degradation/design
cover system construction/construction
cover system stability/construction
cover system stability/construction
cover system stability/design
cover system stability/construction
Attachment
Section
F-A.8.4
F-A.8.5
F-A.4.4
F-A.2.14
F-A.5.5
F-A.9.4
F-A.4.5
F-A.4.6
F-A.7.3
F-A.7.4
F-A.6.3
F-A.6.4
F-A.8.6
F-A.8.7
F-A.8.8
F-A.8.9
F-A.8.10
F-A.3.6
F-A.3.7
F-A.8.11
F-A.8.12
F-A.11.1
F-A.10.1
F-A.12.1
F-A.12.2
F-A.12.3
F-A.12.4
                                   F-10
-------
Table F-2.1.  Classification of Identified Problems at Landfill and Impoundment
    Facilities (Continued).
Facility
Designation
C-7
C-8
C-9
C-10
C-11
C-12
C-13
C-14
C-15
C-16
C-17
C-18
C-19
C-20
C-21
C-22
C-23
S-1
S-2
S-3
S-4
S-5
Information Source
Boschuk(1991)
Boschuk(1991)
Boschuk(1991)
Boschuk(1991)
Boschuk(1991)
unpublished
unpublished
unpublished
Badu-Tweneboah et al.
(1994)
Calabria and Peggs (1996)
Soong and Koerner(1997)
Soong and Koerner(1997)
Soong and Koerner(1997)
Soong and Koerner(1997)
Anderson (1995);
Augelloetal. (1995);
Chang etal. (1996);
Matasovic et al. (1995);
Stewart etal. (1994)
Anderson (1995);
Augelloetal. (1995);
Matasovic et al. (1995);
Stewart etal. (1994)
Anderson (1995);
Augelloetal. (1995);
Matasovic et al. (1995);
Stewart et al. (1 994)
Peggs etal. (1991)
Paulson (1993)
unpublished
Bonaparte and Gross (1 993)
Bonaparte and Gross (1 993)
Problem Classification
cover system stability/construction
cover system stability/construction
cover system stability/construction
cover system stability/construction
cover system stability/construction
cover system degradation/design
cover system displacement/design
cover system stability/design
cover system stability/construction
cover system displacement/construction
cover system construction/construction
cover system stability/design
cover system stability/design
cover system stability/design
cover system stability/design
landfill liner system stability/design
landfill liner system stability/design
landfill liner system stability/design
impoundment liner degradation/
construction
impoundment stability/design
impoundment liner construction/
construction
impoundment liner construction/
construction
impoundment liner construction/
construction
Attachment
Section
F-A.12.5
F-A.12.6
F-A.12.7
F-A.12.8
F-A.12.9
F-A.11.2
F-A.13.1
F-A.12.10
F-A.12.11
F-A.13.2
F-A.10.2
F-A.12.12
F-A.12.13
F-A.12.14
F-A.12.15
F-A.12.16
F-A.12.17
F-A.12.18
F-A.15.1
F-A.16.1
F-A.14.1
F-A.14.2
F-A.14.3
                                   F-11
-------
Table F-2.2. Summary of Identified Problems.
Problem Classification
landfill liner construction/
construction
landfill liner construction/
construction
landfill liner construction/
construction
landfill liner construction/
operation
landfill liner construction/
operation
landfill liner construction/
design
landfill liner construction/
construction
landfill liner construction/
construction
landfill liner construction/
construction
landfill liner construction/
construction
landfill liner construction/
construction
landfill liner construction/
construction
landfill liner construction/
construction
landfill liner construction/
construction
landfill liner degradation/
design
landfill liner degradation/
operation
landfill liner degradation/
design
landfill liner degradation/
construction
landfill liner degradation/
construction
landfill liner degradation/
construction
Facility Designation/
Attachment Section
L-1/F-A.2.1
L-3/F-A.2.2
L-5/F-A.2.3
L-6/F-A.2.4
L-7/F-A.2.5
L-8/F-A.2.6
L-9/F-A.2.7
L-11/F-A.2.8
L-11/F-A.2.9
L-15/F-A.2.10
L-17/F-A.2.11
L-19/F-A.2.12
L-19/F-A.2.13
L-29/F-A.2.14
L-2/F-A.3.1
L-4/F-A.3.2
L-12/F-A.3.3
L-14/F-A.3.4
L-20/F-A.3.5
L-43/F-A.3.6
Problem Summary
leakage through holes in HOPE GM primary liner
leakage through holes in HOPE GM liners
leakage through holes in HOPE GM primary liner
leakage through holes in HOPE GM primary liner
leakage though HOPE GM/CCL composite
primary liner at pipe penetration
landfill gas migrated beyond liner system and into
vadose zone resulting in groundwater
contamination
leakage though HOPE GM primary liner at pipe
penetration
construction debris in CCL with initially smooth
surface protruded from CCL after CCL was left
exposed and subsequently eroded
leakage though HOPE GM primary liner at pipe
penetration
sand bag under installed GM liner approved by
CQA consultant
leakage through holes in HOPE GM primary liner
wind uplifted and tore HOPE GM liner during
construction
severe wrinkling of HOPE GM due to thermal
expansion during construction
large folded wrinkles in HOPE GM primary liner at
two exhumed leachate sumps
desiccation cracking of CCL in exposed HOPE
GM/CCL composite liner
HOPE GM/CCL composite liner damaged by waste
fire
leachate extraction well installed in landfill
appeared to puncture GM primary liner
HOPE GM liner damaged by fire believed to be
started by lightning strike
saturation of GCL beneath GM liner when
rainwater ponded on tack-seamed patch over GM
hole
water ponded between HOPE GM and CCL
components of composite secondary liner and
was contaminated from a source other than the
landfill
                                  F-12
-------
Table F-2.2. Summary of Identified Problems (Continued).
Problem Classification
landfill liner degradation/
design
landfill LCRS or LDS
construction/construction
landfill LCRS or LDS
construction/construction
landfill LCRS or LDS
construction/construction
landfill LCRS or LDS
construction/construction
landfill LCRS or LDS
construction/construction
landfill LCRS or LDS
construction/construction
landfill LCRS or LDS
degradation/design
landfill LCRS or LDS
degradation/design
landfill LCRS or LDS
degradation/construction
landfill LCRS or LDS
degradation/construction
landfill LCRS or LDS
degradation/construction
landfill LCRS or LDS
malfunction/operation
landfill LCRS or LDS
malfunction/design
landfill LCRS or LDS
malfunction/design
landfill LCRS or LDS
malfunction/operation
landfill LCRS or LDS
operation/operation
landfill LCRS or LDS
operation/operation
landfill LCRS or LDS
operation/operation
Facility Designation/
Attachment Section
L-44/F-A.3.7
L-10/F-A.4.1
L-15/F-A.4.2
L-16/F-A.4.3
L-28/F-A.4.4
L-32/F-A.4.5
L-33/F-A.4.6
L-9/F-A.5.1
L-11/F-A.5.2
L-13/F-A.5.3
L-18/F-A.5.4
L-30/F-A.5.5
L-12/F-A.6.1
L-22/F-A.6.2
L-36/F-A.6.3
L-37/F-A.6.4
L-5/F-A.7.1
L-23/F-A.7.2
L-34/F-A.7.3
Problem Summary
landfill gas well punctured GM component of
composite liner and extended into CCL
rainwater entered LDS through anchor trench
sand bags in LCRS drainage layer and debris in
LCRS pipe trench approved by CQA consultant
rainwater entered LDS through anchor trench
excessive needle fragments in manufactured
needlepunched nonwoven GT
HOPE LCRS pipe separated at joints
HOPE LCRS pipe separated at joints
erosion of sand LCRS drainage layer on liner
system side slopes
erosion of sand protection layer on liner system
side slopes
polypropylene continuous filament nonwoven GT
filter degraded due to outdoor exposure
polypropylene staple-fiber needlepunched
nonwoven GT filter degraded due to outdoor
exposure
HOPE LCRS pipe crushed during construction
LCRS pipes were not regularly cleaned and
became partially clogged, and LCRS drainage
layer may be partially clogged
waste fines clogged needlepunched nonwoven
GT filter wrapped around perforated LCRS pipes
waste fines clogged needlepunched nonwoven
GT filter around LCRS pipe bedding gravel
leachate seeped out landfill side slopes in the
vicinity of chipped tire layers
overestimation of LDS flow quantities due to
problems (e.g., clogging) with automated LDS
flow measuring and removal equipment
valves on LCRS pipes were not opened and
leachate could not drain, and waste and leachate
flowed over a berm into a new unapproved cell
LCRS leachate pump moved air and liquid
causing pump airlock and underestimation of
leachate quantities
                                  F-13
-------
Table F-2.2. Summary of Identified Problems (Continued).
Problem Classification
landfill LCRS or IDS
operation/design
landfill liner system
stability/design
landfill liner system
stability/operation
landfill liner system
stability/design
landfill liner system
stability/design
landfill liner system
stability/design
landfill liner system
stability/design
landfill liner system
stability/design
landfill liner system
stability/design
landfill liner system
stability/design
landfill liner system
stability/operation
landfill liner system
stability/operation
landfill liner system
stability/design
landfill liner system
displacement/design
landfill liner system
displacement/design
landfill liner system
displacement/design
landfill liner system
displacement/design
cover system construction/
construction
cover system construction/
construction
cover system degradation/
design
Facility Designation/
Attachment Section
L-35/F-A.7.4
L-21/F-A.8.1
L-24/F-A.8.2
L-25/F-A.8.3
L-26/F-A.8.4
L-27/F-A.8.5
L-38/F-A.8.6
L-39/F-A.8.7
L-40/F-A.8.8
L-41/F-A.8.9
L-42/F-A.8.10
L-45/F-A.8.11
L-46/F-A.8.12
L-9/F-A.9.1
L-11/F-A.9.2
L-25/F-A.9.3
L-31/F-A.9.4
C-2/F-A.10.1
C-16/F-A.10.2
C-1/F-A.11.1
Problem Summary
LCRS leachate pumps and flowmeters continually
clogged and IDS leachate pumps turned on too
frequently and burned out prematurely
sliding along PVC GM/CCL interface during
construction
sliding along GN/GCL (HOPE GM side) and
GCL(bentonite side)/CCL interfaces during
operation
sliding along HOPE GM/poly ester needlepunched
nonwoven GT and HOPE GM/CCL interfaces
during operation
two tears in HOPE GM liner and cracks in soil
intermediate cover from Northridge earthquake
extensive cracks in soil intermediate cover and
further tearing of GT cushion from Northridge
earthquake
sliding along needlepunched nonwoven GT/HDPE
GM primary liner interface after rainfall
sliding along needlepunched nonwoven GT/HDPE
GM liner interface after rainfall
sliding along gravel/HDPE GM liner interface after
rainfall
sliding along very flexible GM liner/needlepunched
nonwoven GT interface after rainfall
sliding along needlepunched nonwoven GT/PVC
GM liner interface after a thaw
sliding along needlepunched nonwoven GT/HDPE
GM liner interface after erosion of soil anchoring
geosynthetics
sliding along GN/HDPE GM primary liner interface
during construction
uplift of GM by landfill gas after erosion of
overlying sand LCRS drainage layer
uplift of geosynthetics by landfill gas after erosion
of overlying sand protection layer
uplift of composite liner by surface-water
infiltration during construction
uplift of composite liner by surface-water
infiltration during construction
portion of topsoil from off-site source was
contaminated with chemicals
high failure rate of HOPE GM seam samples
during destructive testing
failure of geosynthetic erosion mat-lined
downchute on 3H:1 V side slope
                                  F-14
-------
Table F-2.2. Summary of Identified Problems (Continued).
Problem Classification
cover system degradation/
design
cover system stability/
construction
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
construction
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system stability/
design
cover system
displacement/design
cover system
displacement/construction
Facility Designation/
Attachment Section
C-12/F-A.11.2
C-3/F-A.12.1
C-4/F-A.12.2
C-5/F-A.12.3
C-6/F-A.12.4
C-7/F-A.12.5
C-8/F-A.12.6
C-9/F-A.12.7
C-10/F-A.12.8
C-11/F-A.12.9
C-13/F-A.12.10
C-14/F-A.12.11
C-17/F-A.12.12
C-18/F-A.12.13
C-19/F-A.12.14
C-20/F-A.12.15
C-21/F-A.12.16
C-22/F-A.12.17
C-23/F-A.12.18
C-12/F-A.13.1
C-15/F-A.13.2
Problem Summary
erosion oftopsoil layer on 60 m long, 31-1:1 V side
slope
sliding along nonwoven GT/GM interface during
construction
sliding along topsoil/GCL interface after rainfall
sliding along sand/woven GT interface after
rainfall
sliding along sand/GM interface after rainfall
sliding along gap-graded sand/GM interface after
rainfall
sliding along gravel/GT interface during
construction
sliding along sand/calendered nonwoven GT
interface after rainfall
sliding along sand/GM interface after rainfall
sliding along topsoil/nonwoven GT interface
during construction
sliding along PVC GM/CCL interface after a thaw
sliding along geogrid/HDPE GM interface during
construction
sliding along sand/CCL interface during rainfall
sliding along sand/CCL interface immediately after
rainfall
sliding along sand/CCL interface after rainfall
sliding along sand/CCL interface after rainfall
minor cracks in soil intermediate cover from
Northridge earthquake
215-m long crack in soil intermediate cover from
Northridge earthquake
minor cracks in soil intermediate cover from
Northridge earthquake
cover system settlement caused tearing of HOPE
GM boots around gas well penetrations of GM
barrier
localized cover system settlement during
construction stretched, but did not damage, PVC
GM barrier and opened GCL joints
                                  F-15
-------
 Table F-2.2. Summary of Identified Problems (Continued).
Problem Classification
impoundment liner
construction/construction
impoundment liner
construction/construction
impoundment liner
construction/construction
impoundment liner
degradation/construction
impoundment liner system
stability/design
Facility Designation/
Attachment Section
S-3/F-A.14.1
S-4/F-A.14.2
S-5/F-A.14.3
S-1/F-A.15.1
S-2/F-A.16.1
Problem Summary
large wrinkles in HOPE GM primary liner at two
leachate ponds
leakage through holes in HOPE GM component of
composite primary liner
leakage through holes in HOPE GM primary liners
at two ponds
slow crack growth stress cracks and shattering
cracks in exposed HOPE GM liner at five ponds
sliding along polypropylene needle-punched
nonwoven GT/HDPE GM interface during waste
placement
 Table F-2.3.  Categorization of Identified Problems (Number of Identified
 Problems is Given)

Waste Containment System
Component or Attribute
Landfill Liner Construction
Landfill Liner Degradation
Landfill LCRS or LDS Construction
Landfill LCRS or LDS Degradation
Landfill LCRS or LDS Malfunction
Landfill LCRS or LDS Operation
Landfill Liner System Stability
Landfill Liner System Displacement
Cover System Construction
Cover System Degradation
Cover System Stability
Cover System Displacement
Impoundment Liner Construction
Impoundment Liner Degradation
Impoundment LDS
Impoundment Liner System Stability
Impoundment Liner System Displacement
Principal Human Factor
Contributing to the Problem
Design
1
3
0
2
2
1
9
4
0
2
16
1
0
1
0
0
0
Construction
11
3
6
3
0
0
0
0
2
0
2
1
3
0
0
1
0
Operation
2
1
0
0
2
3
3
0
0
0
0
0
0
0
0
0
0
From Table F-2.3, there were 85 identified problems at the 74 facilities.  The
investigation focused on landfill facilities: 94% of the identified problems described
                                    F-16
-------
occurred at landfills.  Based on the waste containment system component or attribute
criterion, the identified problems are distributed as follows:

•  landfill liner construction:  17%;
•  landfill liner degradation:  8%;
•  landfill LCRS or LDS construction: 7%;
•  landfill LCRS or LDS degradation: 6%;
•  landfill LCRS or LDS malfunction:  5%;
•  landfill LCRS or LDS operation: 5%;
•  landfill liner system stability:  14%;
•  landfill liner system displacement:  5%;
•  cover system construction:  2%;
•  cover system degradation: 2%;
•  cover system stability: 21%;
•  cover system displacement:  2%;
•  impoundment liner construction: 4%;
•  impoundment liner degradation:  1%;and
•  impoundment liner system stability: 1%.

No problems were identified under the impoundment LDS and impoundment liner
system displacement categories. Therefore, these categories are not discussed
further.

Based on the principal human factor contributing to the problem criterion, the
identified problems are distributed as follows:

•  design: 48%;
•  construction: 38%; and
•  operation: 14%.

F-3       Evaluation of Identified Problems

F-3.1     Introduction

In this section of the appendix, observations are made on the nature and frequency of
the identified problems for each of the 15 waste containment system component or
attribute criteria considered in this appendix and listed in Section F-2.3. The methods
by which these problems were detected, the time  it took to detect the problems after
they  developed, and the remedy of the problems are also discussed.  The most
common types of problems for each of the 15 categories and the reasons for these
problems are presented.
                                    F-17
-------
It should be noted that the problems and remedies described herein are facility
specific. It should not be inferred that these problems occur at most facilities or that
the specific remedies are appropriate for most facilities.

F-3.2     Landfill Liner Construction

F-3.2.1    Overview

Problems related to construction of landfill liners are one of the most common types of
problems identified in this study. This category represents 14 of the 85 problems (i.e.,
17%) described herein. The problems in this category were primarily attributed to
construction factors. Only one problem was attributed to design and two were
attributed to operation. The problems and the number of identified landfills with them
are as follows:

•  leakage through holes in an HOPE GM primary liner (5 landfills);
•  leakage through an HOPE GM primary liner or HOPE GM/CCL composite  primary
   liner at the LCRS pipe penetration of the liner (3 landfills);
•  severe wrinkling of  an HOPE GM during construction (2 landfills);
•  landfill gas migration beyond a liner system and into the vadose zone resulting in
   groundwater contamination (1 landfill);
•  protrusion of construction debris from a CCL with  an initially smooth surface after
   the CCL was left exposed and subsequently eroded (1  landfill);
•  presence of a sand bag under a GM liner approved by the CQA consultant (1
   landfill); and
•  uplift and tearing of a GM by wind during construction (1 landfill).

F-3.2.2    Leakage Through Holes in HOPE GM Primary Liner

The most common problem in this category, leakage through holes (construction- or
operation-related) in an HOPE GM  primary liner, occurred at landfills L-1, L-3, L-5, L-
6, and L-17. Leakage was detected during construction of landfills L-1 and L-3 by the
relatively high LDS flow rate (i.e., 2,900 liters/hectare/day (Iphd)) that occurred after
rainwater ponded in landfill L-1 and by electrical leak location surveys performed  as
part of CQA of landfill L-3. For both landfills,  GM holes were located by electrical leak
location surveys and repaired.  At landfill L-3, the leak location surveys performed as
part of CQA allowed extrusion seaming problems to be identified early and corrected,
decreasing the frequency of identified GM holes in subsequent installations.
Interestingly, a leak location survey showed that the HOPE GM component of the
secondary liner for landfill L-3 was damaged during placement of the overlying gravel
layer. Several of the holes were located  in the vicinity of the temporary ramps. The
HOPE GM primary liner, which was protected by a GT filter overlying a GN LCRS
drainage layer, was not damaged during placement of the gravel LCRS drainage layer
                                     F-18
-------
on the GT.  Thus, the potential for GM damage during placement of a soil layer over a
GM can be reduced by protecting the GM.

For the remaining three landfills (i.e., L-5, L-6, and L-17), primary liner leakage was
detected during landfill operation. Relatively high LDS flow rates were recorded from
landfill L-5.  However, due to problems with the LDS flow rate measuring system for
about the first three years of operation, the high readings appear to be partially due to
measurement error (e.g., flow rates calculated using the number of "pump on"
intervals were several times higher than rates calculated using flowmeter readings).
By the fourth year of operation, the LDS flow  rate measuring system appeared to be
functioning  adequately, and it was confirmed  that the average LDS flow rate was
relatively high (i.e., 4,660 Iphd). The exact locations of the HOPE GM primary liner
holes causing this leakage are unknown. However, it is likely that there is a hole in
the sump area because the LDS flow rates decreased significantly (i.e., to about
2,000 Iphd) when the "pump on" level in the internal LCRS sump was lowered from
0.6 m to 0.25 m. With respect to the potential for leakage, leachate sumps are
generally the most critical locations in landfills with internal sumps.  Leachate heads
are typically sustained and at higher levels in sumps than at other locations.  In
addition, GM liners  in sumps often have seamed corners to fit the sump geometry.
These seams may contain holes.  Even one GM  hole at a sump can cause relatively
high leakage rates due to the relatively high head of leachate in the sump. No other
remedial actions beyond lowering the "pump on"  level in the sump were implemented
for landfill L-5.

At landfills L-6 and L-17, primary liner leakage was detected within several months
after start of operation by relatively high LDS  flow rates (i.e., average flow rates of
1,200 Iphd and 1,030 Iphd,  respectively) and, for landfill L-17, by the color of and
chemical constituents in the LDS liquid. Landfill L-6 has an HOPE  GM primary liner
on the side slopes and a GM/CCL composite primary liner on the base. With this liner
configuration, leakage primarily occurred through the GM  primary liner on the side
slope. The GM primary liner is overlain only by a GN LCRS drainage layer and a GT
filter.  There is no soil protection layer on the  side slope; waste was placed directly on
the GT.  It is not known if the GM primary liner was damaged during construction or
operation.   However, without a thick protection layer, the potential for liner damage
during landfill operation increases. GM holes were located in landfill L-6 with a dye
test and visual inspection and were repaired.

For landfill L-17, the project specifications only required the inside track of dual track
fusion seams be destructively tested. When the  HOPE GM liner was inspected for
holes shortly after primary liner leakage was detected, liner holes and fusion seam
holes were  observed at several locations. At the seam holes, the outside track of the
seam had separated, allowing leachate to flow through the air channel between the
tracks and potentially through the liner if the inside track had holes.  Separation of the
outside track also increases stress concentrations at the inside track.
                                    F-19
-------
For the four landfills where holes were located in GMs, the holes of largest size were
found in panels and the majority of holes were found in seams.  Based on these five
landfills with GM primary liner leakage, and consistent with the study of LDS flow
rates from double-lined waste containment facilities conducted by Bonaparte and
Gross (1993), construction-related holes in GM liners should be anticipated, even  in
liners installed with CQA.  If there is a head of leachate over a liner hole,  leakage
occurs.  However, as shown for these landfills, GM primary liner holes resulting from
construction can be located by ponding tests, leak location surveys, or other methods
after liner system construction and be repaired before waste placement.

F-3.2.3    Leakage at Pipe Penetration of Primary Liner

Leakage at pipe penetrations of primary liners occurred at landfills L-7, L-9, and L-11.
Leakage was detected during operation of landfill L-7 when the average LDS flow  rate
increased from about 10 to 400 Iphd. This occurred after landfill operations personnel
regraded the sand LCRS drainage layer.  When the regraded area was inspected,
deep tire tracks, made by a rubber-tired loader, were found over the LCRS pipe
penetration of the composite primary liner. At the pipe penetration, the pipe was
broken and the CCL adjacent to the pipe was rutted. The damage was subsequently
repaired.

At landfills L-9 and  L-11, leakage at the LCRS pipe penetration of the HOPE GM
primary liner was detected  during construction after rainwater ponded over the
penetration.  For landfill L-9, this leakage occurred even though the penetration had
two special features to improve the connection integrity: (i) the GM was underlain by a
GCL at the penetration; and (ii) the penetration was sealed with two HOPE boots,
creating a space between them that could be pressure tested and later filled with
foam. Though the pipe penetration at landfill L-9 was inspected twice after
construction and a small extrusion seam hole was repaired, the rate of leakage did
not decrease  substantially. Since the pathway for this leakage could not  be  identified
during construction, this problem was not remedied.

At landfill L-11, leakage at the pipe penetrations in three cells was verified by dye
tests.  When the GM pipe boots were inspected, the boots in two cells were  loose
around the pipes and the boot in the third cell appeared adequate, but short. The
boots were repaired and the space between the boot and the primary liner was filled
with bentonite slurry.  Subsequently, leakage decreased.

These case histories demonstrate that it is difficult to construct pipe penetrations to be
hole free even when extra measures are taken to enhance the integrity of the
connections.
                                     F-20
-------
F-3.2.4    Severe Wrinkling of HOPE GM Liner

Severe wrinkling of the smooth HOPE GM liner was identified by visual observation
during construction of landfill L-19 and exhumation of two sumps at landfill L-29. At
landfill L-19, the large wrinkles were remedied before the liner was covered with an
overlying material. The HOPE GM liner for landfill L-19 was deployed and seamed in
the winter, when temperatures were near freezing, and not covered with sand until the
spring and summer, when temperatures were high and the GM was severely wrinkled.
Several thousand linear meters of wrinkles were required to be cut, leading to more
GM seams. In addition, the overlying sand layer was placed over the GM at night,
when temperatures were cooler and the GM was less wrinkled.

The HOPE GM liner in landfill L-29 also developed large wrinkles.  These wrinkles
were identified eight years after the liner system was constructed, when expansion of
the landfill was underway. Wrinkles were more numerous and larger near the slope
toe and near the sump than away from the slope toe and the sump. Some of the
wrinkles had folded over, and the GM at the folds had yielded.  It is unclear why these
large wrinkles were not noticed when the GM was installed. The GM had been
covered with a GT, which provided some thermal insulation of the underlying GM.
However, it is possible that the GM  developed wrinkles after the GT was placed and
the wrinkles were hidden.  The wrinkled GM was removed when the sumps were
reconstructed for the landfill expansion.  Both wide width tensile tests  (ASTM D 4885)
and single point notched constant tensile load tests (ASTM D 5397) were conducted
on samples of unwrinkled GM and wrinkled GM at folds to assess the effect of the GM
folding  on GM integrity. All of the samples had measured properties exceeding the
project specifications.  While the wrinkled and unwrinkled GM samples had wide width
tensile properties that were not significantly different from one another, the wrinkled
GM samples had a somewhat lower time to break than the unwrinkled samples.

F-3.2.5    Migration of Landfill Gas Beyond Liner System to Groundwater

Landfill gas migrated beyond the edge of a liner system and into the vadose zone
resulting in groundwater contamination at landfill L-8. Volatile organic compounds
(VOCs) were detected in shallow groundwater at a monitoring well located 60 m from
the downgradient edge of the landfill about four years after waste placement began.
This is the only case of groundwater or surface-water contamination by leachate or
landfill gas from a facility identified in this study. The migration occurred because
landfill gas was not well controlled and a pathway for gas migration was present.
Along the perimeter of the landfill, the composite liner was extended horizontally and
the GM was secured by covering it with a layer of relatively permeable soil. An
asphalt parking lot was constructed on top of a section  of the relatively permeable soil
layer and natural ground. As waste reached intermediate grades, it was covered with
a relatively low-permeability soil intermediate cover layer that graded into the
relatively permeable soil layer. The soil intermediate cover layer and the asphalt


                                    F-21
-------
served as confining layers, blocking landfill gas moving from the waste from venting to
the atmosphere.  Instead, some gas migrated through the relatively permeable soil
layer and then beyond the limit of the liner system.  The remedy consisted of
improving source control by installing additional gas extraction wells in the waste,
natural biodegradation of the VOCs in groundwater, and continued monitoring.

F-3.2.6    Other Problems

Construction debris protruding from the CCL, a sand bag under a GM liner approved
by the CQA consultant, and uplift and tearing of an  HOPE GM liner by wind were
identified by visual observation during construction of landfills L-11, L-15, and L-19,
respectively. These problems were remedied before the liner was covered with an
overlying material. At landfill L-11, the contractor constructed the CCL with a soil
containing a small amount of construction debris (i.e., about 0.02% by weight). The
completed CCL initially had a smooth surface; however, the surface of the CCL
eroded when it was left unprotected during the winter and debris protruded from the
CCL.  The CCL in this state was not suitable for placement  of the overlying GM. This
problem was remedied by covering the CCL on the  base of the landfill with a GCL and
covering the CCL on the side slope with a layer of debris-free clay.  Landfill L-15 was
remedied by removing the sand bag. Landfill L-19 was remedied by replacing the
damaged 1.5-mm thick HOPE GM.

F-3.3     Landfill Liner Degradation

F-3.3.1    Overview

Problems related to landfill liner degradation represent 7 of  the 85 problems (i.e., 8%)
described herein. Three problems in this category were attributed to design factors,
three were attributed to construction, and one was attributed to operation. The
problems and the number of identified landfills with  them are as follows:

•  liner damage by fire (2 landfills);
•  liner damage during well installation (2 landfills);
•  desiccation cracking of a CCL in an exposed HOPE GM/CCL composite liner (1
   landfill);
•  saturation of a GCL beneath a GM liner when rainwater  ponded on a tack-seamed
   patch over a GM hole (1 landfill); and
•  water from the CCL ponded between the HOPE  GM and CCL components of a
   composite secondary liner and was contaminated from a source other than the
   landfill (1 landfill).
                                    F-22
-------
F-3.3.2    Liner Damage by Fire

The liners at landfills L-4 and L-14 were damaged by fire during operation and
construction, respectively. The fires were detected by visual observation.  The GM
liner for landfill L-14 was damaged by fire that is believed to have been started  by a
lightning strike.  Rolls of GC drainage layer lined up on an HOPE  GM liner at the top
of the side slope caught fire  during a thunderstorm. The GM beneath the burnt rolls
was rippled and melted in some cases.  The damaged GM and GC were replaced.
The authors are not aware of other instances where a geosynthetics fire was started
by lightning.

At landfill L-4, a chemical reaction of one of the materials disposed of in the landfill
caused a waste fire that took about 11 months to contain and extinguish. Based on
temperature measurements made near the fire, the temperature in the vicinity of the
liner system may have approached 800°C. The liner system in the vicinity of the fire
was severely damaged: the  liner system geosynthetics were melted and disintegrated
and the CCL was desiccated.  The damaged materials were replaced.

F-3.3.3    Liner Damage During Well Installation

The liner at landfill L-12 was possibly damaged and the liner at landfill L-44 was
definitely damaged during installation of wells in the landfills. The possible liner
damage was detected in landfill L-12 by an increase in LDS flow rates; the liner
damage in well L-44 was detected when liner system components were observed in
auger cutting during well installation.  In both cases, the problems were attributed to
design factors.

At landfill L-12, a deep, 100-mm  diameter leachate extraction well was installed in the
double-lined landfill after the LCRS appeared to be clogged.  The well design called
for the well to extend into the sand LCRS drainage layer over the GM primary liner,
but the elevation of the top of the borehole was not surveyed immediately before well
installation. Considering waste settlement since the previous survey of the landfill, the
target borehole depth may have  been too deep. Following well installation, average
LDS flow rates increased from about 300 Iphd to 400 Iphd, and it was suspected that
the well had penetrated the GM primary liner. No remedial actions have been
implemented because it is not clear if the primary  liner was actually punctured and the
LDS flow rates have remained relatively low.

During installation of gas extraction wells in a active landfill, one of the 0.9-m diameter
boreholes for the wells was advanced into the composite liner.  The problem was
identified when portions of the liner system were observed in the cuttings from the
bucket auger. Upon observing these components in the auger cuttings, field
personnel poured bentonite  pellets into the bottom of the borehole to create an
approximately 0.9-m thick bentonite seal at the  borehole base.  It was later discovered


                                     F-23
-------
that a typographic error had been made on the design drawing for the gas extraction
system: the specified borehole depth at the location of the liner damage was greater
than the depth to the top of the liner. Though the potential environmental impact from
the damage was found to be negligible, the proposed remedy for this problem is
repair of the damaged liner system.

F-3.3.4    Other Problems

Saturation of a GCL beneath a GM liner was detected during construction of landfill L-
20; desiccation cracking of a CCL in an exposed composite liner and ponding  of
contaminated water between the GM and CCL components of a composite liner were
detected during operation of landfills L-2  and L-43, respectively. At landfill L-20, the
GCL component of a composite liner became saturated when rainwater ponded on a
tack-seamed patch over a GM hole during construction. This problem was identified
by visual inspection after the ponded water was removed.  The hydrated GCL had
uplifted the GM, and the composite liner was soft in the saturated area. The damaged
GCL was replaced.

At landfill L-2, the CCL component of a side slope composite liner for a cell
constructed three year earlier was observed to be severely desiccated when it was
partially exposed during construction of an adjacent cell. The composite liner, which
consisted of an HOPE GM over a CCL, had not been protected from the environment.
The design required that the sand LCRS drainage layer be placed incrementally up
the side slopes during landfill operation, with the sand advancing ahead of the waste.
During construction of the composite liner, water became trapped between the HOPE
GM and the CCL near the slope toe. The GM had to be cut so the water could drain.
The same phenomenon of trapped water occurred during the construction of the
adjacent cell. This occurrence of water was attributed to water vapor thermally driven
from the CCL into the space between the GM and CCL during the day as the CCL
heated.  The water then condensed on the  bottom of the GM at night as the GM
cooled and flowed downslope to the slope toe. The moisture contents of CCL
samples support this hypothesis.  The  CCL moisture content increased moving
downslope from crest to toe. At the crest, the CCL moisture content was significantly
less than the average construction moisture content; at the toe, the CCL moisture
content was greater than the average construction moisture content. No remedial
actions were implemented.

At landfill L-43, water ponded between the HOPE GM and CCL components of the
composite secondary liner on the side  slope.  This landfill has a GM primary liner on
the side slope and a GM/CCL composite primary liner on the base;  the secondary
liner is a GM/CCL composite. By about one year after construction, a large isolated
bubble of water developed between the GM and CCL components of the composite
secondary liner at the slope toe at a corner of the cell that  had not yet received waste.
The geosynthetics were cut to remove the water, the water was pumped out, and the


                                    F-24
-------
geosynthetics were repaired.  The water was chemically analyzed for organic
constituents and metals and found to be clean, with a chemistry similar to that of the
LDS liquid (i.e., water from the CCL component of the primary liner on the landfill
base).  A small bubble developed at the same location about one year later and the
water was removed, but not chemically analyzed.  When bubbles developed at the
same location about three years after construction, the bubble water was analyzed
and found to contain organic constituents. A testing program  to identify the source of
the water and contamination is underway. Preliminary results of the study indicate
that the source of the water was the CCL. It is believed that the CCL lost the water
due to desiccation and consolidation. While the composite primary liner on the base
was thermally insulated  by a 0.3-m thick sand LCRS drainage layer and the
composite secondary liner on the base was thermally insulated by the 0.3 m-thick
(each) sand LCRS or LDS drainage  layers and the 0.9-m CCL component of the
composite primary liner, the composite secondary liner on the side slopes was only
overlain by geosynthetics and was not sufficiently thermally insulated. Therefore, the
CCL component of the composite liner on the side slopes could potentially lose water
by thermal action.  Based on a chemical analysis,  landfill leachate and groundwater
were excluded as the source of the contamination.  Surface-water runoff from a
nearby former oil facility and fuel from equipment used to construct the liner system
are currently considered to be potential sources of the contamination.

F-3.4     Landfill LCRS or LDS  Construction

F-3.4.1    Overview

Problems related to landfill LCRS or LDS construction represent 6 of the 85 problems
(i.e.,  7%) described herein.  All of the problems in this category were attributed to
construction factors. The problems and the number of identified landfills with them
are as follows:

•  rainwater entered the LDS through the anchor trench (2 landfills);
•  HOPE LCRS pipe was separated at joints (2 landfills);
•  sand bags in the LCRS drainage layer and debris in the LCRS pipe trench
   approved by CQA consultant (1 landfill); and
•  excessive needles in a manufactured needlepunched nonwoven GT (1 landfill).

F-3.4.2    Rainwater Entering LDS Through Anchor Trench

Rainwater was found to be entering the LDS through the liner system anchor trench in
landfills L-10 and L-16 during construction and operation, respectively. This problem
was detected when LDS flow rates from the landfills were higher than expected.
When the landfill anchor trenches were  inspected, they were found to be full of water.
The GC or GN LDS drainage layers  in the trenches were conveying water from the
trenches into the LDSs.  This problem developed for landfill L-10 because its liner


                                    F-25
-------
system anchor trench had been backfilled with a sandy soil that allowed significant
water to infiltrate and pond on the geosynthetics. To remedy the problem, sections of
the back of the anchor trench were excavated to the outside slope of the perimeter
berm along the length of the trench, and the ends of the geosynthetics in the trench
sections were laid horizontal. The perimeter berm was reconstructed to grade with
gravel. This  allows water infiltrating the trench to drain to the outside slope of the
perimeter berm. To minimize infiltration of rainwater into the anchor trench, a GM was
placed over the top of the berm and covered with a 0.3-m thick layer of soil. At landfill
L-16, the anchor trench  soil was not initially well compacted. Over time, the anchor
trench soil settled, and a depression developed over the anchor trench. The
depression trapped runoff, which subsequently infiltrated into the trench.  The problem
for L-16 was remedied by removing, replacing, and regrading the trench soil and
grading the soil surface  to drain away from the trench.

F-3.4.3    HOPE Pipe Separated at Joints

During the initial video survey of the inside of the HOPE LCRS pipes in active landfills
L-32 and L-33, several pipe joints were found to be separated. The  separations were
typically less than 10 mm in width. The subsequent annual surveys  have revealed no
further separations in the pipe joints over time. The reason for the separations is
unclear. It may be that the pipes were never seamed together during construction or
that the quality of some  of the pipe seams was so poor that the seams failed during
construction.  No remedial actions have been taken.

F-3.4.4    Other Problems

Sand bags in the LCRS  drainage layer and debris in the LCRS pipe  trench approved
by the CQA consultant were found in landfill L-15, and excessive broken  needle
fragments in the manufactured needlepunched nonwoven GT were found in landfill L-
28.  These problems were detected during construction by visual observation and
remedied.  For landfill L-15, the sand bags and debris were removed.  For landfill  L-
28, the GT was placed on a GN LCRS drainage layer over a GM/CCL  composite liner.
By the time the needle problem was discovered, some of the GT had already been
covered by a 0.3-m thick soil protection layer. The contractor initially tried to fix the
GT that had been deployed and not covered with soil by manually searching for and
removing needles.  The  contractor also tried to locate needles in the GT beneath the
soil layer using a metal detector.  However, both methods proved to be too time
consuming to locate and remove the hundreds of needles.  Laboratory tests
conducted to evaluate the potential for GM puncture by needles of different lengths
and orientations showed that few needles should puncture the GM.  The  holes caused
by these needles would  be very small. Nonetheless, the defective GT was removed
and replaced with a "needle-free" GT. When the GT was removed, no GM damage
from the needles was observed.  The manufacturer of the defective GT installed
                                    F-26
-------
magnets in the manufacturing plant to remove broken needles from GTs produced in
the future.

F-3.5     Landfill LCRS or LDS Degradation

F-3.5.1    Overview

Problems related to landfill LCRS or LDS degradation represent 5 of the 85 problems
(i.e.,  6%) described herein. Two problems in this category were attributed to design
factors and three were attributed to construction. The problems and the number of
identified landfills with them are as follows:

•  erosion of the sand layer on the liner system side slopes (2 landfills);
•  degradation of polypropylene nonwoven GT filters due to outdoor exposure (2
   landfills);  and
•  crushing of HOPE LCRS pipe draining cell during construction (1 landfill).

F-3.5.2    Erosion of Sand Layer on Side Slopes

Progressive erosion of the sand layer on the liner system side slopes was detected at
landfills L-9 and L-11 during operation.  Landfill L-9 has 100-m long, 4H: 1V side
slopes, and landfill L-11 has 18-m high, 2.5H:1V side slope segments separated by
benches. The erosion caused gullies to develop in the sand layer and the deposition
of sand on the base of the landfills. In landfill L-9, the 0.6-m thick sand LCRS
drainage layer (specified minimum hydraulic conductivity of 1 x 10~4 m/s) also washed
into the exposed gravel around the LCRS pipes and in the  sump area.  In landfill L-11,
the 0.45-m thick sand protection layer (specified  minimum  hydraulic conductivity of 1 x
10~5 m/s) had eroded down to the liner system geosynthetics in two areas and these
areas were subsequently uplifted by landfill gas.  The erosion on this landfill was
exacerbated by runon from an adjacent MSW landfill. The erosion problem required
continual maintenance of both landfills:  sand was pushed back up the side slopes
numerous times and,  in landfill L-9, the  gravel in  the sump  areas was replaced twice.
Besides maintenance of the sand layer, the remedy for landfill L-11 also included
improvement to the runoff and runon control system. The erosion problem for the
landfills will be fully resolved when the sand on the side slopes is covered with waste.
However, for these landfills it may be several years before  this occurs.

F-3.5.3    Degradation of GT Filter Due to Outdoor Exposure

The polypropylene nonwoven GT filters at landfills L-13 and L-18 degraded due to
outdoor exposure. The degradation was detected during construction, and the
degraded GTs were replaced.  For landfill L-13, the  GT was designed assuming it
would be exposed to the environment for several months and then covered with a
sand protection layer.  This strategy was selected because the sand proposed for the


                                    F-27
-------
protection layer was very erodable and would require significant maintenance if left
exposed.  Though a 270 g/m2 GT met the project specifications, a heavier 540 g/m2
GT was selected, anticipating that this GT would retain enough strength after several
months of exposure to meet the specifications.  Due to construction delays, however,
the GT was exposed for more than six months.  By about 6.7 months of exposure, the
GT exhibited significantly reduced strength properties, did not meet the specification
for burst strength, and had developed holes in two areas near the side slope crest of
one of the perimeter berms.  The degraded polypropylene GT was replaced with a
270  g/m2 polypropylene GT and covered with a sand protection layer soon after
installation. Interestingly, a 540 g/m2 polyester  continuous filament nonwoven GT
filter was substituted for the polypropylene GT in part of the landfill. While the
mechanical properties of the polyester GT decreased with time, the rate of
degradation was slower than that for the polypropylene GT. After 14.5 months of
exposure, the polyester GT still met the project  specifications.

For landfill L-18, the 350 g/m2 polypropylene staple-fiber needlepunched nonwoven
GT component of a GC LCRS  drainage layer was exposed on the landfill side slope.
A soil protection layer was to be placed incrementally over the GC on the side slopes
during filling operations. By about one year after construction, waste had not been
placed and the GT component of the GC was falling apart, exposing the GN  and
underlying GM primary liner. The GT degradation was attributed to exposure to
ultraviolet light, sulfuric acid from industrial emissions, water, and high ambient
temperature.  The problem was remedied by replacing the GC LCRS drainage layer
on the side slopes and beginning waste placement in the cell soon afterwards.

F-3.5.4    Other Problems

The pipe draining one cell of landfill  L-30 was crushed during construction. However,
this problem was not detected  until landfill operation began. A valve on the HOPE
LCRS pipe that controlled water draining from the cell was kept closed until just
before the start of waste placement.  During this time, a significant amount of water
(i.e., more than meter deep) ponded in the cell.  When the valve on the pipe was
opened so water could drain, drainage occurred only very slowly. With no other on-
site location to dispose of waste, the baled waste was placed in the ponded water.
C&DWwas placed over the bales to keep the bales from floating.  The crushed
condition of the pipe was only identified when an attempt was made to flush the pipe
to  increase the water flow rate from  the cell. The C&DW contained relatively high
concentrations of sulfate.  As the waste decomposed, the sulfate was reduced  to
hydrogen sulfide gas, which caused gas problems at the landfill.  Due to the hydrogen
sulfide gas emissions,  the landfill was closed early, after only about 1.5 years of filling.
A gas extraction system with a flare was installed in the landfill, and gas emissions
from the facility are successfully being controlled.
                                     F-28
-------
F-3.6     Landfill LCRS or LDS Malfunction

F-3.6.1    Overview

    Problems related to landfill LCRS or LDS malfunction represent 4 of the 85
problems (i.e., 5%) described herein. Two problems in this category were attributed
to design factors and two were attributed to operation factors. The problems and the
number of identified landfills with them are as follows:

•  waste fines clogged the needlepunched nonwoven GT filter in the LCRS piping
   system (2 landfills);
•  LCRS pipes were not regularly cleaned and became partially clogged and LCRS
   drainage layer may be partially clogged (1 landfill);  and
•  leachate seeped out the landfill side slopes in the vicinity of chipped tire layers (1
   landfill).

Interestingly, no problems related to biological clogging of the LCRS or LDS at
modern waste facilities were identified.  If biological clogging  is a major problem at
landfills, it is expected that there would be evidence that it was  occurring. For
example, if the LCRS of a landfill was severely clogged, LCRS flow rates would be
relatively low during landfill operation, leachate head would build up in the landfill,
and, if the head was high enough, leachate would seep from  the landfill side  slopes.
Since most landfills don't exhibit the "symptoms" of biological clogging,  it is currently
not affecting landfill operation enough to be noticeable at  most  landfills.

F-3.6.2    Clogging of GT in LCRS Piping System

    Waste fines clogged the needlepunched nonwoven GT filters in the LCRS piping
systems of landfills L-22 and L-36.  The problems were detected during operation.
For landfill L-22, which contained industrial plant waste, lime-stabilized waste, and
slurried fines, the 540 g/m2 needlepunched nonwoven GT filter was wrapped around
LCRS perforated pipes bedded in the pea gravel LCRS drainage layer.  By about one
year after construction, it was apparent that the LCRS was not functioning adequately
because: (i) rainwater ponded on the waste surface and did not drain freely into the
waste; and (ii) the amount of leachate removed from the LCRS sump was less  than
expected. When the LCRS was excavated near the sump, the GT wrapping  the
LCRS pipes was found to be clogged by waste fines at the pipe perforations. As
described by Giroud (1996), the purpose of a GT is to  retain the material behind the
filter, not capture particles in motion. The GT around the  pipe serves no purpose. It
is not needed to prevent the gravel from falling through the pipe perforations. In fact,
this GT proved to be detrimental as it captured waste fines and biological particles at
the small flow areas at the pipe perforations.  It is not known how this problem was
remedied.
                                    F-29
-------
For landfill L-36, the needlepunched nonwoven GT around the LCRS pipe bedding
gravel is apparently clogged by fines from the incinerated MSWash placed  in the
landfill. LCRS flow rates are less than expected.  In addition, leachate ponded in the
landfill and seeped from the landfill side slopes. When a video camera was run
through the LCRS pipes, the pipes were found to be full of ash. The pipes were
flushed, but the sump still recharged very slowly even though the landfill was full of
leachate. From the gradation of the sand LCRS drainage layer and the apparent
opening size of the GT, it is expected that the clogging is most significant in the GT
around the pipe bedding gravel. The sand has larger openings than the GT and
passes fine ash particles. The owner's proposed remedy for this problem involves
installing a leachate collection manhole in the landfill  to facilitate leachate removal.

F-3.6.3    Other Problems

At landfill L-12, the LCRS appeared to be partially clogged when LCRS flow rates
decreased, but LDS flow rates increased, after soil intermediate cover was placed
over the landfilled  waste. In addition, the LDS flow rates were higher than those
typical of nearly filled landfills in that region of the country.  The LCRS drainage layer
is a sand with a specified minimum hydraulic conductivity of 1 x 10~5 m/s. The LCRS
and LDS pipes in the landfill were not flushed annually as is common practice in the
region. Rather than performing maintenance on the LCRS pipes, the landfill owner
decided to install a leachate extraction well in the landfill.  After the well was installed
through the waste, LDS flow rates increased, and it was suspected that the  well had
penetrated the GM primary liner.  Subsequently, the LCRS pipes were cleaned out
and are now scheduled to be flushed annually.  Insufficient time has past to determine
if cleaning the pipes solved the problem or if the LCRS drainage layer may be partially
clogged.

After about 1.2 million chipped tires were disposed of in MSW landfill L-37 as part of a
site cleanup, leachate was  observed to be seeping out the side slopes of the landfill in
the vicinity of chipped tire layers. The coarse tire chips have a higher hydraulic
conductivity than the MSW and, apparently, promote lateral drainage within the waste.
A bucket auger was advanced through the waste to the top of the sand LCRS
drainage layer at six locations near the seeps.  Perched leachate in the tire  chips was
found in some of the boreholes at depths of up to 3 m. The boreholes with  perched
leachate were completed as wells.  The wells allow some of the leachate collected in
the tire chip layers to readily drain to the LCRS. In addition, leachate levels in the
wells are inspected weekly, and the wells are pumped if leachate is present.
                                    F-30
-------
F-3.7     Landfill LCRS or LDS Operation

F-3.7.1    Overview

Problems related to landfill LCRS or LDS operation represent 4 of the 85 problems
(i.e.,  5%) described herein.  One problem in this category was attributed to design
factors and three were attributed to operation. The problems and the number of
identified landfills with them are as follows:

•  clogging and other problems with leachate pump or flow rate measuring system (3
   landfills); and
•  values on LCRS pipes were not opened and leachate could not drain  (1 landfill).

F-3.7.2    Malfunction of Leachate Pump or Flow Rate Measuring System

Clogging and other problems with leachate pumps or flow rate measuring systems
occurred at landfills L-5, L-34, and L-35. These problems were identified during
operation and were primarily remedied by equipment maintenance, repair, and
replacement. For landfills L-5 and L-34, the problems also led to overestimation of
LDS  flow quantities and underestimation of LCRS flow quantities, respectively.  At
landfill
L-5, there were numerous problems: (i) the control system that measured the liquid
levels in the sumps and operated of the pumps was prone to compressor failure and
clogging of air lines; (ii) the control system problems caused the LDS sump pump to
sometimes stay on even when there was no more liquid to be removed (i.e., it
pumped air) and caused pumps to run for too long of an interval, or even
continuously, until they burned out; (iii) the mechanical flowmeters frequently clogged
and became inoperable; (iv) the venturi flowmeters that replaced the mechanical
flowmeters were damaged by an electrical storm;  (v) a failed check valve allowed LDS
liquid that had been metered to flow back into the LDS of the cell and be  remetered;
and (vi) the leachate level measurement system in the LCRS sump experienced drift
due to the  buildup of landfill gas pressures in the sump. These problems were
remedied by a program of equipment maintenance, repair, and replacement.

For landfill L-34, the "pump on" time setting at the pump controller tended to drift
causing the pump to operate too long and LCRS flow rates to be underestimated.  As
a result of overpumping, air was pulled into the pump, and the pump tended to
become airlocked and shut down. An accumulating flowmeter was installed to
provide a better measurement of leachate flow quantities. However, when the air
pulled into the pump moved through the flowmeter, the flowmeter overestimated the
quantity of leachate removed.  Additionally, the pump did not reprime as the leachate
levels rose. When the landfill  operator noticed this, the pump was removed from the
sump and adjusted and the "pump on" time setting was reset.  This problem was
                                    F-31
-------
resolved by replacing the pumps with self-priming pumps from a different
manufacturer.

For landfill L-35, the submersible pumps and magnetic flowmeters in the LCRS
continually became clogged with a white precipitate. In addition, the LDS was
designed with large, shallow sumps to keep the liquid head in the LDS above the
pump intake, but no more than 0.3 m. To accomplish this, the pump cycle was very
short. The pump motor overheated from turning on and off so quickly and burned out.
The problems for landfill L-35 were resolved by disassembling the LCRS pumps and
flowmeters and cleaning them with citric acid about every month. Also, the LDS
pumps were replaced with smaller models to increase cycle times.

F-3.7.3   Other Problems

During operation of landfill L-23, it was discovered that valves on the LCRS pipes  in
two landfill cells were not opened prior to placement of waste in the cells.
Consequently, leachate could not drain from the cells.  Eventually, the waste became
buoyant due to rising leachate levels. After about 1.5 years of operation, a bulldozer
operating at the active face sunk in the waste and had to be removed with a crane. In
another cell, waste was placed too close to an intercell berm between it and a new
cell that had not yet been approved for waste.  Sufficient space between the waste
and the intercell berm should have been maintained to temporarily store runoff from
the waste. After a storm, leachate and waste washed over the berm and into the new
cell.  A temporary access road made out of waste was constructed over the intercell
berm to access the new cell and clean out the waste that had washed  into it.  At the
time, the sand LCRS drainage layer had not been placed over the berm liner system
geosynthetics. The waste placed directly on the HOPE GM primary liner damaged
the GM. The  corrective measures for this landfill have not yet been implemented.

F-3.8    Landfill Liner System Stability

F-3.8.1   Overview

Problems related to landfill  liner system stability are one of the most common types of
problems identified in this study. This category represents 12 of the 85 problems (i.e.,
14%) described herein. The problems in this category were primarily attributed to
design factors. Only three problems were attributed to operation.  The problems and
the number of identified landfills with them are as follows:

•  liner system slope failure due to static loading (10 landfills); and
•  liner system damage due to an earthquake (2 landfills).
                                    F-32
-------
F-3.8.2    Liner System Instability Due to Static Loading

Liner system slope failure due to static loading occurred at landfills L-21, L-24,
L-25, L-38, L-39, L-40, L-41, L-42, L-45, and L-46. The slope failures at landfills L-21
and L-46 occurred during construction; the slope failures at the other landfills occurred
during operation. All of the failures were detected by visual observation of mass
movement of one or more components of the liner system, including cracking of soil
layers near the slope crest, and/or tearing, tensioning, or wrinkling of geosynthetics.
The primary causes of failure were: (i) using unconservative presumed values for the
critical interface shear strength (landfills L-25, L-39, L-41, L-46); (ii) not evaluating the
critical condition for slope stability (e.g., liner system with waste at intermediate
grades, critical liner system interface) (landfills L-24 and L-38); (iii) not accounting for
or underestimating seepage pressures (landfills L-39 and L-40); (iv) not accounting for
moisture at the GM/CCL interface (which weakens the interface) due to spraying of
the CCL and thermal effects (landfill L-21); (v) not maintaining the drainage layer
outlets free of snow and ice, which can lead to increased seepage pressures (landfill
L-42); and (vi) not maintaining sufficient thickness of soil layer anchoring
geosynthetics (L-45).

During construction of landfill L-21, part of the single-composite liner system on the
upper 3.5H:1V side slopes slid downslope along the polyvinyl chloride (PVC) GM/CCL
interface.  Sliding occurred both after placement of the sand LCRS drainage layer
over the PVC GM liner and during placement of the lime-stabilized sludge protection
layer over the sand. The slide zone was identified by cracking of the sand layer or
stabilized sludge layer near the crest of the side slope and wrinkling of the GM liner
near the slope toe. When the GM in the slide zone was exposed, it was taut and, in
some cases, torn near the  slope crest.  The CCL beneath the liner was relatively wet:
while it had been constructed with an average measured moisture content of about 2
percentage points wet of standard Proctor optimum,  the moisture content measured
in the slide zone was about 7 percentage points wet of optimum. The increase in
moisture content at the  surface of the CCL between compaction and sliding is
believed to have resulted from condensation of water on the lower face of the GM due
to thermal effects and spraying of the CCL surface to prevent desiccation prior to
placement of the GM. A liner system slope stability analysis had not been conducted
as part of the landfill design.  However, the  liner system for another phase of the
landfill had been successfully constructed previously using the same liner system
components and geometry and similar site soils to construct the CCL.  After the
failure, the owner conducted direct shear interface tests and slope stability analyses.
The owner found that, on the steepest slopes (i.e., 3H:1V), the liner system was just
stable after construction. However, the liner system  became unstable as the CCL
surface became wetter  and the strength of the GM/CCL interface decreased. This
problem was remedied  by:  (i) placing a temporary protective cover over the GM liner
and CCL in the slide zone to protect the CCL from frost damage until the GM and
overlying soil layers could be reconstructed in the spring; (ii) installing a


                                     F-33
-------
polypropylene monofilament woven GT reinforcement layer between the GM liner and
overlying soils to carry the load of the soils for the liner system constructed on the
upper side slopes after the failure occurred; and (iii) developing new construction
procedures to reduce the potential for liner system sliding in the future.

At landfill L-24, a liner system slope failure occurred after the intermediate waste
slopes in part of the landfill were temporarily increased to about 2.5H:1V, significantly
steeper than the maximum slope of 4H:1V specified in the operations plans.  Sliding
occurred along the  GN/GCL (HOPE GM side) and GCL (bentonite side)/CCL
interfaces of the single-composite liner. (The GCL consisted of an HOPE GM with a
bentonite layer glued  to one side of the GM.)  The problem was detected when the
sand berm at the toe of the waste slope began to heave and the top of the waste
slope began to crack.  When part of the waste was excavated and the liner system
was exposed during the construction of a  landfill expansion, the GCL was observed to
be folded near the toe of the waste slope and, further back into the waste, the GCL
was taut and torn. The stability analysis conducted for design of landfill L-24 was
based on presumed interface shear strength values and did not evaluate the liner
system with the waste at intermediate grades; there was no regulatory requirement to
include this in the permit.  When the landfill owner later decided to overfill the cell with
waste, the stability of  the liner system with the relatively steep waste slopes was not
analyzed. This problem was remedied by excavating about 270,000 m3 of waste,
reconstructing the damaged liner system,  and regrading the waste slopes to 4H:1V.

At landfill L-25, which has  a double-composite liner system, the slope failure was
manifested by mass movement of the waste.  Cracks were observed on the landfill
surface in the early morning and, within about five hours, the waste had slid
horizontally up to 11 m and vertically up to 4 m. Around the side slopes, the soil cover
over the waste and the waste was cracked and, in some locations, the liner system
was torn. Sliding occurred primarily along the HOPE GM/CCL interface of the
composite secondary liner on the landfill base and the HOPE GM primary liner/GT
interface on the side slopes. Of note, there was no limit on the maximum CCL
moisture content in the specifications; the CCL material was compacted at an average
moisture content 5 percentage points wet of optimum. The landfill had been designed
using presumed interface shear strengths for the liner system. After the failure
occurred, interface  direct shear and pullout tests were conducted to evaluate the
shear strength of critical liner system interfaces, and stability analyses were
performed using the actual interface strengths.  The results of the laboratory tests
showed that only a  small amount of displacement (5 mm or less) is required to
mobilize the peak shear strength along an interface. At greater displacements, the
shear strength decreased  and approached the large-displacement value. Assuming
that peak shear strengths were mobilized  on the landfill base and 3H:1V side slopes
and large-displacement shear strengths were mobilized on the 2H:1V side slopes,  the
calculated factor of safety  for the three-dimensional failure surface was 1.08. Thus,
the measured interface shear strengths and the rapid decrease in shear strength with


                                    F-34
-------
displacement after peak strength has been reached can explain the landfill failure.
The problem was remedied by relocating waste into other phases of the facility,
repairing the liner system, and refilling the landfill with waste.

At landfill L-38, which has 3H:1V side slopes,  a stability analysis conducted during
design showed that the sand LCRS drainage layer would not be stable on the
underlying HOPE GM primary liner. To keep the sand from sliding downslope, the
design engineer added a needlepunched nonwoven GT between the sand and the
GM.  Apparently, the potential for sliding between the GT and GM was not evaluated.
After several rainfall events, the sand drainage layer became wet and the GT began
creeping downslope.  About one month after sliding started, a relatively heavy rainfall
occurred at the site and the liner system slope failed due to excessive creep and
tearing of the GT and cracking of the sand layer at the slope crest. At locations where
the GT tore and slid downslope, the underlying GM was abraded.  Since the failure
coincided with rainfall, seepage pressures in the sand probably contributed to the
failure. The method of repair was not given.

The slope failures  at landfills L-39 and L-40 were also partially attributed to seepage
pressures.  At landfill  L-39, a needlepunched nonwoven GT cushion was placed
between a LCRS gravel drainage material and an HOPE GM liner. About one to two
years after the liner system was constructed, a portion of the GT tore at the crest of
the 3H:1V side slope and slid to the slope toe after a heavy rainfall. A number of
successive  slides occurred during several subsequent rainfalls. Based on an infinite
slope analysis conducted by the authors of this appendix, the  GT cushion would have
been in tension even without seepage pressures. Information was not available on
whether the GT was designed to be in tension. The method of repair was not given.

At landfill L-40, a gravel LCRS drainage layer slid over an HOPE GM liner to the toe of
the 31-1:1 V side slope after a heavy rainfall.  After the failure, the gravel was inspected
and found to be contaminated with fines. The fines apparently inhibited drainage of
water from the gravel and allowed seepage pressures to develop.  The method of
repair was not given.

At landfill L-41, a very flexible polyethylene GM liner tore at the crest of the 2.5H: 1V
side slopes and slid downslope over an underlying needlepunched nonwoven GT.
The GM was overlain by a sand LCRS drainage  layer.  Failure occurred after  a heavy
rain and was attributed to seepage pressures in the sand drainage layer. However,
analyses conducted by the authors of this appendix using the  method of Giroud et al.
(1995) found that seepage pressures above the GM would not significantly affect the
stability of the GM/GT interface below the GM. Based on an infinite slope analysis
conducted by the authors of this appendix, the liner system was, at best, only
marginally stable after construction. Thus, the original design was only marginally
stable and the rainfall had the effect of "triggering" the slide. The method of repair
was not given.


                                    F-35
-------
At landfill L-42, a needlepunched nonwoven GT between a gravel LCRS drainage
layer and a PVC GM liner tore at the crest of the 41-1:1 V side slopes and slid
downslope over the GM. The failure occurred after frozen water in the gravel LCRS
drainage layer began to melt, but could not freely flow out of the gravel at the slope
toe because of ice at the toe. The method of repair was not given.

At landfill L-45, a needlepunched nonwoven GT protection layer slid downslope over
an HOPE GM liner after the soil layer anchoring the geosynthetics beyond the crest of
the side slope was eroded by landfill traffic.  The method of repair was not given.

At landfill L-46, a needlepunched nonwoven GT filter between a soil protection layer
and GN tore at the crest of a 3H:1V side slope and the GN separated at its panels
and slid downslope over an HOPE GM liner.  Slope stability analyses performed as
part of the liner system design used assumed interface shear strengths and relied on
the GT filter to carry the load of the overlying soil layer and construction equipment.
Laboratory interface shear strength testing conducted after the failure gave lower
shear strengths than those assumed for design. In addition, the design strength used
for the GT was too high and construction loads were  underestimated.  In areas where
the soil protection layer had been placed up the 11-m high slope, the soil was
removed and the damaged GT and GN were repaired. The placement of the soil
protection layer over the GT was subsequently limited to increase slope stability: the
soil layer was required to be placed in 6 m increments along the slope, advancing
upslope with waste placement.

Interestingly, the majority of the slides described above occurred along
geosynthetic/geosynthetic interfaces. For a number of case histories, the interface
friction angle between adjacent liner system components was estimated on the basis
of published tested data. This approach should be avoided because there may be
significant differences in interface shear strengths between similar materials from
different manufacturers and even identical materials from different production lots
from the same manufacturer. In fact, only a small error in the estimated interface
shear strength may cause slope instability. Because of this, geosynthetic interface
shear strengths should not be estimated, they should be measured.  Additionally, as
more geosynthetics are available on the market, the probability increases that there
will be significant differences in properties between geosynthetics that appear to be
similar.

F-3.8.3    Liner System Instability Due to an Earthquake

Liner system slope instability due to an  earthquake occurred at landfills L-26 and
L-27 during operation.  The instability was caused by the 17 January 1994 Northridge
earthquake (moment magnitude Mw 6.7), which generated estimated rock peak
horizontal accelerations at the landfill sites of 0.33g and 0.36g, respectively. The


                                     F-36
-------
damage, which was detected by visual inspection, consisted of: (i) tearing of the GM
liner at several locations near the side slope crest, parallel to the anchor trench in
landfill L-26; (ii) further tearing of the GT cushion above the GM liner on the side slope
in landfill L-27; and (iii) surficial cracking of soil intermediate cover, primarily near
locations with contrast in seismic response characteristics (e.g., top of waste by
canyon walls) at both landfills.

In two canyon fills at landfill L-26, the HOPE GM liner tore on benches above the
waste during the earthquake.  The tears were located near the side slope crest,
parallel to the anchor trench.  In one canyon, the GM tear was 4.3 m long and opened
up to  0.25 m wide; in  the other canyon,  there were three parallel tears with a total
length of about 23 m. Longitudinal cracks were present in the soil intermediate cover
at the top of the waste below the tear. The cracks were up to 0.3 m wide, with vertical
offset of 0.15 to 0.3 m.  At some locations, the cracks exposed the underlying GM
liner.  Forensic analyses indicated that the GM tears initiated from locations where
GM seam samples were cut for destructive testing.  Both  the stress concentrations
around the  hole (which  had been patched) and the high pullout capacity of the anchor
trench appear to have been factors in the initiation and propagation of the tears. As
the GM liner moved during the earthquake, it was constrained at the anchor trench
and subsequently tore at locations with  concentrated stresses.  Furthermore, in these
canyons, it  appears that the slope stability factor of safety of the waste at intermediate
grades was relatively low under the seismic loading of the Northridge earthquake.
The seismic-related damage at the landfill was remediated by repairing the damaged
GM, securing the liner system above the damaged GM using a soil berm rather than
an anchor trench,  and regrading and revegetating the cracked soil intermediate cover.

At landfill L-27, tears  in  the GT cushion  above the GM liner appeared to have
increased in size as a result of the earthquake. The tears were located on the side
slope above the waste.  No tears were observed in the GM liner.  In addition,
extensive cracks were observed in the soil  intermediate cover near its contact with the
side slope liner system. The cracks, which had up to 25 mm of vertical offset, may
have been the result of limited downslope movement of the GT. The damage was
remedied by repairing the GT and regrading and revegetating the cracked soil
intermediate cover.

F-3.9     Landfill Liner System Displacement

F-3.9.1    Overview

Problems related to landfill liner system displacement represent 4 of the 85 problems
(i.e., 5%) described herein. All of the problems in this category were attributed to
design factors. The problems and the number of identified landfills with them are as
follows:
                                     F-37
-------
•  uplift of liner system geosynthetics by landfill gas after erosion of the overlying
   sand layer (2 landfills); and
•  uplift of composite liner by surface-water infiltration during construction (2 landfills).

F-3.9.2    Uplift of Liner System Geosynthetics by Landfill Gas

Uplift of liner system geosynthetics by landfill gas occurred during operation of
landfills L-9 and L-11, after portions of the overlying sand layer on the liner system
side slope eroded.  The side slope liner system for both of these landfills was
constructed over existing MSW.  The designs called for gas beneath the liner system
of landfills L-9 and L-11 to be collected in a gravel trench at the crest of the side slope
and gas extraction wells, respectively. The designs were not based on site-specific,
estimated gas generation rates or field measurements of gas production.

During construction of  landfill L-9, an HOPE GM component of a single-composite
liner was uplifted after  the 0.6-m thick sand LCRS drainage layer began to erode,
decreasing the overburden pressure on the liner system.  Ten 6-m diameter bubbles
developed and uplifted the GM to about 1.5m.  In some areas, the GM yielded. The
uplift height was relatively large compared to the diameters of the uplifted areas;
consequently, the  estimated strain in the GM was relatively large (i.e., 16%). The
problem was remedied by cutting the GM at the bubbles, installing temporary gas
venting pipes through the liner system and into the underlying waste, and replacing
the sand  layer.

At landfill L-11, which has a double-liner system, the GC LCRS drainage layer, HOPE
GM primary liner, GN LDS drainage layer, and HOPE GM component of the
composite secondary liner were uplifted about 0.1 m by landfill gases in two area
where the 0.45-m thick sand protection layer had eroded. The diameters of the
uplifted areas were about 10 and 20 m. The uplift height was relatively small
compared to the diameters of the uplifted areas; consequently, the estimated strain in
the GM was relatively small (i.e., less than 0.002%). The GM  liner was cut at the two
gas bubbles to release the gas, and the liner system was repaired.  The sand
protection layer was replaced on the side slopes.

F-3.9.3    Uplift of Composite  Liner by Surface-Water Infiltration

The  GCL/CCL composite liners on the side slopes of landfills L-25 and L-31 were
uplifted by surface-water infiltration during construction.  This problem was detected
by visual  observation of ponded water beneath the GCL.  For both landfills, the GCL
component of the liner consisted of a 1.5-mm thick HOPE GM and a bentonite layer
glued to one side of the GM. The GCL was installed with the bentonite side down and
was  seamed by fusion seaming the GM component of adjacent panels.  Also for both
landfills, approximately the bottom  half of the side slope liner system was constructed
in an excavation against native soil (specified maximum hydraulic conductivity of 1 x


                                    F-38
-------
10~7 m/s) and the upper half was constructed against a more permeable rocky mine
spoil berm.

At landfill L-25, the ground surface outside of the berm was graded towards the berm,
allowing runoff to pond at the toe of the exterior berm slope. During liner construction,
runoff ponded at the toe of the exterior berm slope after rainwater was allowed to
evaporate and infiltrate  into the soil.  Subsequently, water pooled under the GM and
saturated the GCL and CCL near the toe of the side slope. When the GM was cut for
the water to drain  and the damaged  GCL was removed, the underlying  CCL was very
soft.  The water appeared to be originating from the interface of the berm mine spoil
and the native soil. Apparently, runoff was seeping through the more permeable berm
soils into cell. This problem was resolved by dewatering the berm using vertical wells,
routing runoff away from the berm toe, constructing a gravel underdrain beneath the
liner in the damaged area, and reconstructing the liner.

At landfill L-31, water conveyed in a  surface-water diversion ditch located on top of
the berm infiltrated into the  berm soils, pooled under the GCL, and saturated the GCL
and CCL.  When the GCL was cut for the water to drain and the damaged GCL was
removed, a 0.6-m  diameter cavity was found in the berm soils and CCL. The bottom
of the cavity was located near the interface of the mine spoil and native soil.
Apparently, the relatively high rate of water infiltration through the ditch and into the
mine spoil caused erosion of the mine spoil and CCL where the water exited the soil
and flowed beneath the GCL. The liner was repaired, and the ditch was lined with
clay to reduce infiltration.

F-3.10    Cover System Construction

Problems related to cover systems as they are built represent 2 of the 85 problems
(i.e.,  2%) described herein.  Both problems in this category were attributed to
construction factors.  The problems and the  number of identified landfills with them
are as follows:

•  portion of topsoil from an off-site  source was contaminated with chemicals (1
   landfill); and
•  high failure rate of HOPE GM seam samples during destructive testing
   (1  landfill).

The problems related to cover systems construction occurred at  landfills C-2 and C-16
and were detected and remedied during construction.  At landfill  C-2, a portion of the
topsoil from an off-site source was contaminated with chemicals.  This problem was
detected when several truckloads of topsoil brought to the  site had an aromatic odor.
Samples of the affected soil were analyzed and found to contain unacceptably high
concentrations of lead.  The problem was  resolved by removing the affected material
from the site and screening new material brought to the site for contamination.


                                    F-39
-------
At landfill C-16, a large proportion of the fusion and extrusion seam samples for a 1-
mm thick textured HOPE GM barrier failed destructive testing. The project
specifications required that destructive testing of the GM seams be performed by the
installer; the CQA consultant was only to monitor the installation. With respect to the
fusion seams, initially only the inside track of seam samples was destructively tested
in shear and peel by the installer.  The project specifications, however, required both
tracks of the fusion seam samples be destructively tested.  After about 50% of the GM
had been approved, based on passing destructive tests, and this GM had been
covered with a topsoil layer, the CQA consultant realized that the installer had not
tested both seam tracks. Archived fusion seam samples were subsequently obtained
and tested. About 60%  (i.e., 25 of 42) of the archived seam samples and 49% (i.e.,
44 of 90) of the seam samples for the entire GM failed the peel test, primarily due to
seam separation exceeding the  minimum specified value of 10%. With respect to
extrusion seams, 50% (i.e., 6 of 12) of the seam samples taken from GM not covered
with topsoil also failed. The installer attributed the high  seam sample failure
frequency to benzene, toluene, ethylbenzene, and xylenes (i.e., BTEX) in landfill gas
being absorbed  by the HOPE and inhibiting the formation of good seams.
Interestingly, the same installer had placed an HOPE GM barrier over an adjacent
section of the landfill about one year earlier and only had about 10% of seam samples
failing destructive testing.  After laboratory testing was conducted on the seam
samples, it was concluded that the primary cause of the poor seam quality was soil in
the seams (i.e.,  inadequate cleaning prior to seaming).  Other causes of failure were
overheating and, for extrusion seams, inadequate grinding.  The BTEX absorbed by
the GM had no apparent impact on seam quality. The failed seams were isolated and
repaired.

F-3.11     Cover System Degradation

Problems related to cover system degradation represent 2 of the 85 problems (i.e.,
2%) described herein. Both problems in this category were attributed to design
factors.  The problems and the number of identified landfills with them are as follows:

•  failure of a geosynthetic erosion mat-liner downchute on 31-1:1 V side slope (1
   landfill); and
•  erosion of a topsoil layer on a 60 m long, 3H:1V side slope (1 landfill).

The problems related to cover system degradation occurred at landfills  C-1 and C-12
and were detected and remedied during the post-closure period. At landfill C-1, a
polyethylene, three-dimensional, grass reinforcement type erosion mat was used
experimentally to line one downchute on a landfill cover system.  The other
downchutes were  lined with riprap. The erosion mat-lined downchute had a maximum
slope of 3H:1V and conveyed runoff from approximately 2 ha of cover system and 8
ha of adjacent property. The erosion mat was installed  and seeded in the fall, when


                                    F-40
-------
plant growth is relatively low, resulting in an extended period with poor to no grass
cover in the downchute.  Within one month after construction, following a series of
significant rainfall events, the channel was unserviceable. Soil had raveled along the
sides of the downchute, soil had eroded underneath the mat and along mat panel
overlaps, and the mat had moved downslope about 2 m. There was little grass in the
downchute.  The landfill owners concluded that the combination of large drainage
area, steep slope, and the inability of grass to sprout quickly in the channel lead to
failure of the downchute. The problem was resolved by relining the downchute with
riprap and placing topsoil in the eroded areas.

At landfill C-12, the cover system was constructed with 60 m long, 3H:1V unbenched
side slopes. Sand diversion berms were located at the top of the cover system and
about midway down the side slopes to divert runoff into six downchutes. Within three
years after construction,  deep gullies had developed on the landfill side slopes in the
vicinity of the riprap-lined downchutes and in areas where the sand berms  at the side
slope crest were breached.  Some of the gullies extended through the topsoil and
sand drainage layers down to the GM barrier. In several locations, the GM was
damaged by punctures and tears, and the subgrade beneath the GM was irregular.
The severe erosion was  attributed to the following: (i) the sand drainage layer
(specified  minimum hydraulic conductivity  of 1 x 10~5 m/s) in the cover system did not
have sufficient capacity;  (ii) sand diversion berms and downchutes did not  intercept
lateral flow in the sand drainage layer; (iii)  runoff collected by berms and downchutes
could  infiltrate through the topsoil layer and enter the drainage layer; and (iv) a lack of
access control resulted in unauthorized trafficking of four-wheel drive vehicles or dirt
bikes on the landfill. This problem was remedied by adding swales at the top of the
cover system to collect runoff and direct it  to the downchute, repairing the damaged
cover system, and installing a chain link fence around the perimeter of the  landfill to
limit vehicle access.

F-3.12    Cover System Stability

F-3.12.1   Overview

The most common type of problem identified in this study is related to cover system
stability. This category represents 18 of the 81 problems (i.e., 21%) described herein.
The problems in this category were primarily attributed to design factors. Only two
problem were attributed to construction and one was attributed to operation.  The
problems and the number of identified landfills with them are as follows:

•  cover system slope failure during construction (4 landfills);
•  cover system slope failure after rainfall or a thaw (11 landfills); and
•  soil cover damage due to an earthquake (3 landfills).
                                     F-41
-------
F-3.12.2 Cover System Failure During Construction

Cover system slope failure during construction occurred at landfills C-3, C-8, C-9, and
C-14. Slope failure was detected by visual observation of mass movement of the
cover system, cracking of soil layers near the slope crest, and wrinkling of
geosynthetics at the toe of the cover system slope. The primary causes of failure
were: (i) placing soil over the side slope geosynthetics from the top of the slope
downward, rather from the toe of the slope upward (landfills C-3 and C-14); using
unconservative presumed values for the critical interface shear strength (landfill C-8);
and (iii)  not considering the effects of variation in the tested geosynthetics, accuracy
of test methods, and test conditions on the interface shear strength to use in design
(landfill
C-9).

At landfill C-3, the design called for geosynthetic reinforcement to be installed over a
nonwoven GT cushion and covered with topsoil. The reinforcement was to be
secured on the top of the landfill by covering a length of geosynthetic with soil. Slope
stability  analyses were conducted assuming topsoil would be placed over the
reinforcement from the bottom of the slopes upward.  However, this condition was not
incorporated into the construction specifications. When construction began, access to
the bottom of the side slopes was not available.  So the contractor started placing
topsoil from the crest of the slope downwards. Shortly afterwards, a section of the soil
covered cover system slid along the  interface between the GT and an underlying GM
barrier.  The problem was remedied by repaired by placing new geosynthetic
reinforcement and GT layers over the GM barrier, and placing the topsoil over the  GT
from the bottom of the side slopes upward.

At landfill C-8, a gravel drainage layer placed on a 3H:1V side slope continually slid
down the slope, eventually damaging the underlying GM.  The contractor had tried to
place the gravel by pushing it up the slope with a bulldozer and by placing it on the
slope using a clamshell bucket, but neither method worked. The method of repair
was not given.

At landfill C-9, as topsoil was being placed over an already-installed sand drainage
layer on 3H:1V side slopes, the sand drainage layer slid downslope over a calendered
nonwoven GT.  Project-specific interface direct shear tests between the sand and GT
performed prior to the failure resulted in a secant interface friction angle of about 21°.
An infinite slope stability analysis performed with this  interface strength shows that the
sand should be stable on the 31-1:1 V slopes.  Tilt table tests performed after the failure
gave a secant friction angle for the sand/GT interface of about 18°.  The differences in
secant interface friction angles may be attributed to variation in the tested
geosynthetics, accuracy of the test methods, and differences in the test conditions.
The cover system was reconstructed with a needlepunched nonwoven  GT that had a
higher interface shear strength with sand than the calendered GT.


                                     F-42
-------
At landfill C-14, the design called for geogrid reinforcement to be installed between an
HOPE GM barrier and overlaying soil layers, with the first such layer being a sand
drainage layer. The design specified that the reinforcement be secured on the top of
the landfill by extending the reinforcement onto the top and covering it with the soil
layers.  Slope stability analyses were conducted assuming that the soil layers would
be placed over the reinforcement from the bottom of the slope upward.  However, this
condition was not incorporated into the construction specifications. When
construction began,  not all of the geogrid rolls were secured at the top of the slope
because landfill gas wells were in the way.  Access to the bottom of the side slopes
was limited at some locations due to wetlands near the slope toe.  As a consequence
of these conditions, the contractor placed a stockpile of sand over the geogrid on the
side slope near the crest and began placing the sand from the crest downward.
Shortly after sand placement began, the reinforcement snapped at the slope crest
beneath the sand stockpile and construction equipment placing the sand.  The GM
then tore near the slope crest and along outward diagonals down the length of the GM
on both sides of the stockpile. The cover system was redesigned without
reinforcement and reconstructed successfully.

F-3.12.3  Cover System  Failure After Rainfall or a Thaw

Cover system slope failure after rainfall or a thaw occurred at landfills C-4, C-5,
C-6, C-7, C-10, C-11, C-13, C-17, C-18, C-19, and C-20 during the post-closure
period.  Slope failure was detected by visual observation of mass movement of the
cover system, cracking of soil layers near the slope crest, and wrinkling of
geosynthetics at the toe of the cover system slope. The primary causes of failure
appeared to be: (i) not accounting for seepage  pressures (landfills C-4,  C-5, C-6, C-
17, and C-18); (ii) clogging of the drainage system, which can lead to increased
seepage pressures (landfills C-7, C-10, C-11, C-19, and C-20); and (iii)  not
accounting for moisture at the GM/CCL interface (which weakens the interface) due to
rain falling on the  CCL surface during construction and freeze-thaw effects (landfill C-
13).

The cover system slope failures at landfills C-4, C-5, C-6, C-17, and C-18 were
primarily attributed to rainfall-induced seepage  pressures in soil layers above the
failure surface. The cover systems for the landfills have 3H:1V or 2.51-1:1 V side slopes
and are up to about 60 m  in slope length. Failure occurred along a topsoil/GCL
interface at landfill L-4, a sand drainage layer/woven GT interface at landfill C-5, a
sand drainage layer/GM interface at landfill  C-6, and a sand layer/CCL interface at
landfills C-17 and C-18. The cover system for landfill C-4 has been redesigned with a
drainage layer; however, at the time of this appendix, the modified cover system had
not been constructed.  Landfill C-6 was repaired by reconstructing the cover system
with benches and collection pipes that drained water from the sand drainage layer into
the benches. The method of repair of landfills C-5, C-17, and C-18 was not given.


                                     F-43
-------
For landfills C-7, C-10, C-11, C-19, and C-20, clogging of components of the final
system drainage system prohibited the cover systems from draining freely.  When
significant pore pressures built up in the cover systems after rainfall, the cover
systems experienced slope failure. Failure occurred along the sand/GM interface at
landfills C-7 and C-10, topsoil/GT interface at landfill C-11, and sand/CCL interface at
landfills C-19 and C-20.  At landfill C-7, which was constructed with a  gap-graded
sand drainage layer in the cover system, clogging was caused by fines, presumably
washed into the sand from the topsoil and the sand  upslope of the failure zone.  This
problem was remedied by reconstructing the drainage layer with a uniformly graded
sand and a GN. At landfills C-10 and C-20, perforated pipes in the sand drainage
layer were wrapped with a GT filter. Eventually, fines clogged the GT at the pipe
perforations and water became trapped in the drainage layer. At landfill C-10, the
pipes were removed and replaced with perforated pipes bedded in gravel wrapped  in
a GT. The remedy for landfill C-20 was not given. At  landfill C-11, the GT beneath
the topsoil layer became clogged with  soil particles over time and did  not drain freely
into the underlying gravel. The remedy for landfill C-11 was not given. For landfill C-
19, the design called for water collected in the sand drainage layer to  drain  to the toe,
be collected in a gravel toe drain, and  exit the cover system through a pipe. An
adequate filter system was not established between the topsoil, sand  drainage layer,
and gravel toe drain.  The gravel became very contaminated with fines, which
presumably migrated into the gravel from the overlying sand and topsoil.  The remedy
of landfill C-19 was not given.

The cover system for landfill C-13 was constructed in the fall. During  the winter the
cover system was covered with snow and the ambient temperature was below
freezing until the spring.  A few days after the first thaw in spring, a PVC GM slid over
a CCL on the 41-1:1 V side slopes.  A forensic investigation showed that water could  not
exit from the sand drainage layer because the lower end of the drainage layer was
blocked by ice and  snow. As a result,  the cause of the slide was  initially assumed to
be the seepage pressures that developed when flow started after melting of the ice at
the lower end of the drainage layer. However, a subsequent slope stability  analysis
showed that seepage pressures above a GM have little effect on the factor  of safety
with respect to a slide that occurs at an interface located beneath the  GM.  With
seepage forces identified as only a minor contributor to the slope failure, an additional
investigation was conducted to evaluate the effect of temperature fluctuations on
GM/CCL interface shear strength.  Interface shear tests simulating the conditions
during the winter (-7°C) followed by thaw (+0.5°C) showed that the formation of ice
lenses at the GM/CCL interface at below-freezing temperature increased the water
content at the GM-CCL interface, resulting in a marked decrease of the interface
shear strength after a thaw, compared to the interface shear strength  before freezing.
With systematic measurements of the  water content of the CCL and a slope stability
analysis, the slope failure could be explained by the higher CCL water content and
lower GM/CCL interface shear strength in the area where the slide occurred than in


                                     F-44
-------
other areas.  The higher water content was attributed to the heavy rainfall that
preceded the installation of the GM in the area where the slide eventually occurred.
The fact that a PVC GM had ruptured with an apparently small strain, compared to the
typical 300% strain at break of PVC GMs, was also investigated. Tests on the PVC
GM showed that, at 0.5°C, the conditions after a thaw, the PVC GM had a yield strain
of 9%, which is much less than the strain at break of about 300% at 23°C. This 9%
yield strain explains the observed rupture of the GM and is consistent with the
observed displacements. The cover system was reconstructed in the slide area
without a GM.

F-3.12.3   Soil Cover Damage Due to an Earthquake

Soil intermediate cover damage due to an  earthquake occurred at landfills C-21,
C-22, and C-23 during operation. The damage was caused by the  17 January 1994
Northridge earthquake (moment magnitude Mw 6.7), which generated estimated rock
peak horizontal accelerations at the landfill sites ranging from 0.20g to 0.42g. The
damage, which was detected by visual inspection, consisted of surficial cracking of
soil intermediate cover occurring primarily  near locations with contrast in seismic
response characteristics (e.g., top of waste by canyon walls). At landfills C-21 and C-
23, the cracking was relatively minor. Cracks at landfill C-23 were up to 100 mm
wide. At landfill C-22, one crack near and parallel to the liner system anchor trench
was 215 m long, up to 150 mm wide, and vertically offset up to 100 mm. No waste
was exposed. At all  landfills, the damage was expected and was dealt with as an
operation issue through post-earthquake inspection and repair (i.e.,  regrading and
revegetating the cracked soil layers).

F-3.13     Cover System Displacement

Problems related to cover system displacement represent 2 of the 85 problems (i.e.,
2%) described herein.  One problem in this category was attributed to design factors
and one was attributed to construction.  The problems and the number of identified
landfills with them are as follows:

•  cover system  settlement caused tearing of HOPE  GM boots around gas well
   penetrations of GM barrier (1  landfill); and
•  localized cover system settlement during construction stretched the PVC GM
   barrier and opened GCL joints (1 landfill).

Problems related to cover system displacement occurred at landfills C-12 and C-15
and were detected during the post-closure period and construction, respectively. At
landfill C-12, a cover system, with vertical HOPE gas  collection wells that penetrated
the HOPE GM barrier, was  installed over MSW. At each gas well penetration, an
HOPE GM boot was  clamped to the well and extrusion seamed to the GM barrier to
seal the barrier around the well.  By about three years after the cover system was


                                    F-45
-------
installed, the MSW had settled up to 0.9 m. When several of the GM boots around
the wells were inspected, they were observed to be torn from the GM barrier.  The
boots were not designed to accommodate settlement of the waste, which would cause
downward displacement of the GM barrier relative to the wells. This problem was
remedied by replacing the old gas extraction well boots with new expandable boots
that can elongate up to 0.3 m.  These boots can also be periodically moved down the
well to accommodate landfill settlement.

At landfill C-15, a cover system was constructed over saturated, highly compressible
paper mill sludge.  To facilitate construction of the cover system, a stabilized sludge
working surface was spread over the in-place sludge. After the cover system
geosynthetics (GC drainage layer, 0.5-mm thick PVC GM barrier, GCL, and GC gas
collection layer) were installed, placement of the overlying soil layer began. The
repeated trafficking of low-ground pressure bulldozers over portions of the cover
system resulted in pumping of the underlying sludge into the stabilized sludge. This
pumping progressively reduced the shear strength of the stabilized sludge layer,
resulting in localized bulges and,  at times, placement of excessive thickness of soil.
Eventually, the weakened  stabilized sludge layer underwent a localized bearing
capacity failure in a 60-m long by 18-m wide area. The measured settlement of the
cover system in this area was up to 2.4 m, and the estimated average strain in the
geosynthetics where the settlement is 2.4 m  is 4.7%. Though none of the
geosynthetics appeared to have been damaged by the displacement, the PVC GM
was in tension and the GCL seams had separated at two locations along the length of
the panels. Adjacent GCL panels had been overlapped 0.15 m along the roll length;
however, calculations show that the seam would open if the average strain exceeded
3.2%. The affected area was repaired by removing the cover system materials in this
area, restabilizing and regrading the sludge,  and reinstalling the cover system with
new geosynthetic materials.  The bulldozers  used to spread the overlying soil  layer
had ground pressures less than that used previously, and additional grade control
measures were implemented to ensure that excess soil was not placed.

F-3.14     Impoundment Liner Construction

F-3.14.1   Overview

Problems related to impoundment liner construction represent 3 of the  85 problems
(i.e., 4%) described herein. All problem in this category were attributed to
construction factors.  The problems and the number of identified landfills with them
are as follows:

•  leakage through holes  in the HOPE GM primary liner or the HOPE GM component
   of the GM/CCL composite primary liner (2 impoundments); and
•  large wrinkles in an HOPE GM primary liner (1 impoundment).
                                    F-46
-------
F-3.14.2   Leakage Through Holes in HOPE GM Liner

Leakage through the primary liners of impoundments S-4 and S-5 was detected
during operation when LDS flow rates increased unexpectedly after the liquid levels in
the impoundments were raised.  For impoundment S-5, primary liner leakage was
confirmed  by the results of chemical  analyses of LDS  liquid.  For impoundment S-4,
which has  a composite primary liner  consisting of a 2.5-mm thick  HOPE GM over a
0.45-m thick CCL (specified maximum hydraulic conductivity of 1  x 10~9 mis), LDS
flow rates  increased significantly within one month after the liquid level in the pond
had reached its highest level of about 3 m.  When the pond liquid level was lowered,
the LDS flow rates returned to their normal levels.  GM holes were located and
repaired.  However, LDS flow rates increased again when the pond liquid level was
raised back to 3 m.  The leakage problem for impoundment S-4 was remedied when
the GM primary liner was inspected at an elevation corresponding to the maximum
liquid level, and holes were found and repaired. In both cases where LDS flow rates
increased when the pond liquid level was raised, the flow rate increase occurred over
a relatively short time period (less than  one month). Presumably  at least part of this
flow was due to primary liner leakage.  It is not clear how leakage entered the  LDS in
such a short time period given that the  primary liner is a composite. The primary liner
on the side slope is only protected by "sacrificial" GM; the primary liner is not
thermally insulated by a soil protection  layer.  It may be that the CCL underlying the
GM on the side slope has become desiccated due to thermal effects, and the
hydraulic conductivity of the CCL has increased by several orders of magnitude.

Impoundment S-5 has two ponds with HOPE GM primary liners.  Prior to operation,
leak location surveys were performed in both ponds, and identified primary liner holes
were repaired.  Even so, primary liner leakage was detected in both ponds shortly
after start of operation based on LDS flow rates, which increased with increasing pond
liquid level, and chemical analysis of the LDS  liquid. After about two years of
operation,  GM primary liner holes were located in the ponds and repaired.

F-3.14.3   Other Problems

The exposed HOPE GM primary liner in two ponds at  impoundment S-3 developed
large wrinkles after construction. The double-liner system for the ponds was
constructed in the winter when temperatures were cooler. At the  end of construction,
the GM primary liner was noticeably wrinkled,  but acceptable to the CQA consultant.
By early summer, the ponds had not yet been put into service, and the GM had
become more wrinkled under the increasing temperature. Wrinkles were more
numerous  and larger near the slope toe as they propagated downslope during several
months of  temperature cycling. (This is sometimes referred to as the "caterpillar
effect".) Wrinkles were,  on average,  about 100 mm high and several large wrinkles
near the slope toe were folded over.  This problem will be resolved by cutting out the
large wrinkles in the GM and seaming the cuts.


                                    F-47
-------
F-3.15    Impoundment Liner Degradation

Problems related to impoundment liner degradation represent 1 of the 85 problems
(i.e., 1%) described herein.  The problem in this category was attributed to design
factors.  Slow crack growth (SCG) stress cracks and rapid crack propagation (RCP)
shattering cracks developed in the exposed HOPE GM liner in five ponds at
impoundment S-1. The problem was detected during operation by visual inspection.
One of the pond liners had been installed with compensation panels to allow for liner
contraction at low temperatures; designed compensation panels had not been
installed in the other pond liners.  In general, at temperatures near freezing, the GM on
the side slope was taut. All of the GM seams had been constructed by lapping the
panels and applying a bead of HOPE extrudate (lap-type extrusion seam). The pond
liners were exposed and, therefore, subject to significant thermally induced tensile
stresses under the wide range of ambient temperatures at the site (i.e., -30 to 40°C).
By four years after installation, the five impoundment liners exhibited relatively short
SCG stress cracks in and adjacent to some seams.  There were no cracks below
water level. The stress cracks generally occurred in the lower GM.

During the winter, record low temperatures appeared to precipitate the cracking of
some seams from the side slope crest to toe. These long cracks were surrounded by
branching  RCP shattering cracks.  The liner in the pond containing the fewest visible
stress cracks had been installed with compensation panels. There was also some
indication that the most seriously damaged liners had been installed at high ambient
temperatures and would, therefore, require the largest amount of compensation in
order to be stress-free at the lowest operating temperature. The shattering cracks
that occurred in the pond liners can be explained by the conjunction of the following
(Giroud, 1994a):

•  the HOPE resin used in these GMs did not have a low stress-cracking
   susceptibility;
•  tensile stresses caused by thermal contraction;
•  strain concentrations caused by the seams;
•  decreased allowable yield strain at the low temperatures at which GM shattering
   occurred; and
•  increased crystallinity of the HOPE next to the seam.

Impoundment S-1 was remedied by replacing damaged GM with new GM containing
an adequate amount of slackness calculated using the temperature of the GM at the
time of the repair.  In addition, every third seam in the ponds was cut from the anchor
trench to the toe of slope and the GM was allowed to relax. Compensation panels
were installed at each cut seam.  Stress cracks in seams were repaired using a wide
bead extrusion technique developed specifically for this purpose.
                                    F-48
-------
F-3.16    Impoundment Liner System Stability

Problems related to impoundment liner system stability represent 1 of the 85
problems (i.e., 1%) described herein. The problem in this category was attributed to
design factors. The GT cushion in the double-liner system for sludge impoundment
S-2 failed during operation. The failure was detected by visual observation. The
failure was attributed to the method of sludge placement: sludge was dumped on the
GT cushion at the crest of the 10 m high, 2H:1V impoundment side slopes and
allowed to flow to the slope toe. On several occasions, the sludge adhered to a
polypropylene needlepunched nonwoven GT cushion layer, which overlies an HOPE
GM primary liner.  When this occurred, the sludge was pushed downslope by a low-
ground pressure bulldozer. Tension developed in the GT, and it eventually tore at the
slope crest and slid downslope over the GM. This problem was remedied by
replacing the damaged GT and  placing a thin GM slipsheet over the GT in the sludge
dumping area to facilitate the sliding of sludge downslope.
F-4       Significance of Identified Problems

F-4.1     Introduction

The main impacts of the problems identified in this investigation are interruption of
waste containment system construction and operation, increased maintenance,
increased costs, and negative public/regulator perception.  As discussed in Chapter
F-3, almost all of the identified problems were detected during construction or
operation,  shortly after they occurred.  In general, problems detected and repaired
during construction have no environmental impact; however, they can delay
construction, impact operation (e.g., delay waste placement in a landfill cell), increase
construction costs, and reduce public/regulator confidence.  Problems detected during
operation can potentially have an environmental impact.  However, if the problems are
detected and remedied soon after they occur, there is less likelihood of environmental
impact.  Problems that occur during operation are also more likely to interrupt facility
operation,  increase maintenance, and result in higher costs and greater impact to
public/regulator confidence than problems that occur during construction.  In addition,
problems that occur during operation may be difficult to repair.

Of the problems in this study for which the remedy was identified, six problems were
not completely repaired.  These problems are classified as landfill liner
construction/construction, landfill liner degradation/design, landfill LCRS or LDS
construction/construction, and landfill LCRS or LDS malfunction/operation. One
problem was detected during  construction, and five were detected during operation.
These problems are as follows:
                                    F-49
-------
Relatively high leakage rates through the GM primary liner on the base of landfill
L-5 were detected shortly after start of operation.  However, due to problems with
the flow measuring system, leakage rates were not accurately known until about
three years after operation began.  Leakage rates decreased when the head of
leachate in the LCRS sump was reduced; however, the leakage rates were still
relatively high. GM primary liner holes have not been located and repaired at this
landfill because: (i) there is no anticipated environmental impact of the primary
liner leakage given the expected performance capabilities of both the LDS and the
composite secondary liner; (ii) repair of liner systems after waste placement would
be extremely difficult and expensive; and (iii) additional liner system damage could
occur in any attempt to excavate the waste and repair the liner system.
Leakage at the LCRS pipe penetration of the HOPE GM primary liner for landfill L-
9 was detected during construction. The pathway for this leakage could not be
identified during construction, and the problem was not remedied. The
environmental impact from the leakage, however, is expected to be negligible
given that the landfill has a composite secondary liner.
Severe desiccation cracking of the CCL component of the composite liner for an
existing cell at landfill L-2 was detected when a new cell was constructed adjacent
to the existing cell which had been constructed three years earlier. The existing
cell with the cracked CCL was filled to about 70% of its waste capacity. No
actions were required for the older cell by the regulatory agency presumably
because: (i) the older cell was almost filled and would be closed shortly
afterwards;  (ii) the repair would  require that the waste be removed from the cell,
which is extremely difficult, and  costly; (iii) the CCL was only observed to be
desiccated on the side slope; all other thing  being equal, side slope liner holes are
less detrimental than base liner holes because the head of leachate on the side
slope is less than the head on the base slope; and (iv) environmental impacts from
the potential for increased liner  leakage are  expected to be negligible given that
the liner includes a GM. Two months after construction of the composite liner, the
CCL component of the liner on the side slope of the new cell was beginning to
crack too, and will likely become more desiccated until the exposed composite
liner for the cell  is covered with the sand drainage layer.
Several joints of the HOPE LCRS  pipes in active landfills L-34 and L-36 were
found to be separated when the inside of the pipes was videotaped. No action has
been required by the regulatory agency presumably because: (i) leachate flowing
out of the pipe at an open pipe joint can still  flow to the leachate sump (though the
localized head at the open joint  may be somewhat higher that those upgradient
and downgradient of the open joint); (ii) the pipe condition has remained
unchanged during subsequent annual videos; (iii) repair of LCRS pipes after
waste placement would be extremely difficult and expensive; and (iv)
environmental impacts from  having a localized higher head on the liner at the open
pipe joints are expected to be negligible.
A leachate well installed into landfill L-12 appeared to puncture the GM primary
liner in the landfill.  No action has been required by the regulatory agency because
                                 F-50
-------
   it is not clear if the primary liner was actually punctured and the flow rates from the
   LDS have remained relatively low.  Environmental impacts from this possible GM
   hole are expected to be negligible.

The impacts of the identified problems on the environment; construction, operation,
and maintenance, and cost are discussed below.

F-4.2     Environmental Impacts

F-4.2.1    Introduction

   The potential consequences of problems associated with landfill liner systems,
cover systems, and impoundment liner systems are presented in Tables F-4.1 to F-
4.3, respectively. As shown in these tables, the consequences range in severity and
potential for environmental impact. The potential environmental impacts are
described below. The tables also present methods for potentially preventing the
problems. Interestingly, the problems only resulted in an identified environmental
impact to groundwater or surface-water quality by leachate or landfill gas at one
facility, landfill L-8. At this MSW landfill, groundwater impact by VOCs was attributed
to gas migration through a relatively permeable soil layer that secured the edge of the
GM liner and extended from the crest of the liner system side slope to beyond the
liner system.  The problem was  resolved by installing additional gas extraction wells in
the landfill.  Without the measures taken to correct the problems at some of the other
facilities, however, adverse environmental impacts could have eventually occurred at
these facilities. Furthermore, the mere occurrence of problems, even in the absence
of an environmental impact,  undermines the confidence that the public holds in the
waste management professional community.

F-4.2.2    Landfills Liner Systems

Potential environmental impacts of the problems that can affect landfill liner systems
can be ranked as follows, from the most to the least serious:

•  major liner breach;
•  hole in GM or GCL;
•  increased risk of leakage through existing holes;
•  increased risk of hole in liner; and
•  incorrect monitoring.
                                    F-51
-------
        Table 4-1.  Consequence and Prevention of Landfill Liner System Problems.
        PROBLEM TYPE
                                         POSSIBLE CONSEQUENCE
                                      PROBLEM PREVENTION
(In
1. LINER CONSTRUCTION
• GM holes
  .  Seams, Panels

  .  Connections

• GM condition
  .  Wrinkles
  .  Scratches
• GM uplift by wind
• CCL defects
  .  Stones, Debris
  .  Improper compaction
• Material in contact with GM
  .  Overlying (LCRS or LDS material)
  .  Overlying (Waste, no protection layer)
  .  Overlying (Placement of soil protection
     layer during operations)
  .  Underlying (Debris, Sandbags)
• Liner system anchor trench/edge covered
  with a permeable material overlain by a
  less permeable material
• Contamination of CCL in top liner	
                                                 Defect

                                                 Defect
Yield (Potential defect)
Yield (Potential defect)
Rupture or yield

High permeability, Damage to GM
High permeability

Puncture (Defect)
Damage to GM
Damage to GM

Damage to GM
Gas migration over and beyond liner
system

Incorrect interpretation of LDS chemistry
Specifications, CQA, Ponding test, Leak
survey, Conductive GM
Connection design, CQA, Ponding test, Leak
survey, Gas tracer test, Conductive GM

Installation temperature and method, CQA
CQA
Design, Installation method, CQA

Specifications, CQA
Specifications, CQA

Cushion, CQA, Ponding test, Leak survey
Cushion, Soil protection layer, Operation QC
Cushion, CQA, Operation QC

CQA
Design, CQA
                                                                                       Runon control, Equipment maintenance, CQA
        2. LINER DEGRADATION
        • GCL outdoor exposure


        • GM stress cracking
          GM outdoor exposure
          CCL outdoor exposure
          GM/CCL composite outdoor exposure
          Fire
          Advance borehole through liner	
                                         Hydrated and swollen (Low shear
                                         strength, May trigger instability, See
                                         "Stability" in 7)
                                         Major liner breach
                                         Weakening, Embrittlement
                                         Cracking, Erosion
                                         CCL:  Cracking
                                         GM:  Major breach; CCL: Cracking
                                         Major liner breach
                                      Installation method, CQA
                                      Resin selection, Protection layer, Minimize
                                      tension (Design, Installation temperature and
                                      method)
                                      GM selection, Protection layer
                                      Protection layer
                                      Protection layer
                                      Operation
                                      Design, Do not drill in vicinity of liner
-------
       Table 4-1. Consequence and Prevention of Landfill Liner System Problems (Continued).
       PROBLEM TYPE
                                         POSSIBLE CONSEQUENCE
                                     PROBLEM PREVENTION
       2. LINER DEGRADATION (Con.)
       •  GM exposure to chemicals
       •  CCL/GCL exposure to chemicals
                                         Weakening, Permeation, Holes (Defect)
                                         Increased permeability
                                     GM selection, Operation (Chemical control)
                                     Material selection, Operation (Chemical
                                     control)
       3. LCRS AND LDS CONSTRUCTION
       •  GT holes due to burning

       •  Drainage layer not sealed in anchor trench

       •  Pipe not connected at joints
       •  Debris in pipe trenches
       •  Sandbags in LCRS or LDS
       •  GT needles
                                         Clogging of drainage layers, pipes, or
                                         sumps (See "Clogging" in 5)
                                         Water intrusion in LCRS and LDS,
                                         Incorrect monitoring
                                         Excessive leachate head
                                         Potential clogging of pipes or sumps
                                         Excessive leachate head
                                         Damage to GM
                                     Installation method, CQA

                                     Design, CQA

                                     CQA
                                     CQA
                                     CQA
                                     ManufacturerQC, GT plant inspection, CQA
(In
CO
4. LCRS AND LDS DEGRADATION
• Soil protection layer outdoor exposure
• GT outdoor exposure

• GN or GC compressive creep
• Pipe failure
  .  Pipe crushing
           Pipe weld separation
Erosion
Holes in GT (Clogging of LCRS material,
See "Clogging" in 5)
Excessive leachate head

Excessive leachate head, Potential GM
damage
Excessive leachate head, Potential GM
damage
Design
Protection layer

Design, Testing

Design, Pipe selection

Construction method, CQA
       5. LCRS AND LDS MALFUNCTION
       •  GT clogging
         .  GT over drainage layer

         .  GT around pipe
       •  Drainage layer clogging

       •  Pipe clogging

       •  Water expelled from CCL into LDS
       •  Leachate lateral seepage
                                         Delay in leachate collection, Excessive
                                         leachate head
                                         Excessive leachate head
                                         Excessive leachate head, Incorrect
                                         monitoring
                                         Excessive leachate head, Incorrect
                                         monitoring
                                         Incorrect monitoring
                                         Leachate migration through cover, Cover
                                         uplift	
                                     GT selection, No GT if used to remove
                                     moving particles
                                     No GT around pipe
                                     Drainage material selection

                                     Pipe selection, Maintenance

                                     Design
                                     Design, Operation
-------
       Table 4.1. Consequence and Prevention of Landfill Liner System Problems (Continued).
       PROBLEM TYPE
                                         POSSIBLE CONSEQUENCE
                                      PROBLEM PREVENTION
       6. LCRS AND LDS OPERATION
       • Leachate generation greater than leachate
         removal/storage/treatment rate
       • Equipment failure
         .  Pumps,  Cleanout

         .  Flowmeter, Check valve
       • Forget to open valves	
                                         Excessive leachate head
                                         Excessive leachate head

                                         Incorrect monitoring
                                         Excessive leachate head
                                      Design (e.g., temporary storm-water isolation
                                      berms)

                                      Design (e.g., sump capacity) , Equipment
                                      selection
                                      Equipment selection
                                      Operation QC, Simpler design
       7. STABILITY
         (STATIC, HYDRODYNAMIC, SEISMIC)
       • Global (Foundation failure)
       • Waste/I i n e r syste m s I id e
       • Liner system slide	
                                         Major liner breach
                                         Major liner breach
                                         Major liner breach
                                      Investigation, Design
                                      Design, Testing, Waste placement sequence
                                      Design, Testing, Construction method
dn
8. DISPLACEMENT
• Foundation settlement or subsidence

• Differential settlement at connections
• Liner system uplift
  .  Water

  .  Gas

  .  Wind
• Waste settlement
  .  Downdrag force on liner

  .  Downdrag force on manhole
GM: rupture or yield; GCL: open joints

GM: rupture or yield

GM: rupture or yield; GCL: open joints
(May trigger instability, See "Stability" in
7)
GM: rupture or yield; GCL: open joints
(May trigger instability, See "Stability" in
7)
GM: rupture or yield

GM: rupture or yield; GCL: open joints

Differential settlement (GM: rupture or
yield; GCL: open joints)	
Investigation, Design, Foundation
improvement, GM selection
Design, GM selection

Underdrain, Protection layer

Gas collection system, Protection layer

Design, Installation method, CQA

Liner system design, Waste placement
(compaction)
Manhole foundation design, Manhole-liner
connection design
-------
       Table 4-2. Consequence and Prevention of Landfill Cover System Problems.
       PROBLEM TYPE
                                         POSSIBLE CONSEQUENCE
                                     PROBLEM PREVENTION
(In
en
1. COVER SYSTEM CONSTRUCTION
• Periphery sealing
• Contaminated soil used in cover system
• GM holes
  .  Seams, Panels

  .  Connections

• GM condition
  .  Wrinkles
  .  Scratches
• GM uplift by wind
• CCL defects
  .  Stones, Debris
  .  Improper compaction
• Material in contact with GM
  .  Overlying (Drainage layer material)
  .  Underlying (Debris, Sandbags)
• GT holes due to burning

• Sandbags  in drainage layer

• GT needles
                                                Gas migration
                                                Contamination of runoff

                                                Defect

                                                Defect
Yield (Potential defect)
Yield (Potential defect)
Rupture or yield

High permeability, Damage to GM
High permeability

Puncture (Defect)
Damage to GM
Clogging of drainage layers (See
"Clogging" in 2)
Excessive water head (May trigger
instability, See "Stability" in 3)
Damage to GM
Design
Specifications, CQA

Specifications, CQA, Leak survey, Conductive
GM
Connection design, CQA, Leak survey, Gas
tracer survey, Conductive GM

Installation temperature and method, CQA
CQA
Design, Installation method, CQA

Specifications, CQA
Specifications, CQA

Cushion, CQA
CQA
Installation method, CQA

CQA

ManufacturerQC, GT plant inspection, CQA
       COVER SYSTEM DEGRADATION
       •  GCL outdoor exposure
         GM stress cracking
       •  GM outdoor exposure
       •  GM/CCL composite outdoor exposure
       •  Surface layer exposure
                                         Hydrated and swollen (Low shear
                                         strength, May trigger instability, See
                                         "Stability" in 3)
                                         Major breach
                                         Weakening, Embrittlement
                                         CCL: Cracking
                                         Erosion (May trigger uplift by gas, See
                                         "Displacement" in 4 )	
                                      Installation method, CQA,
                                     Resin selection, Protection layer, Minimize
                                     tension (Design, Installation temperature and
                                     method)
                                     GM selection, Protection layer
                                     Protection layer
                                     Design
-------
       Table 4.2.  Consequence and Prevention of Landfill Cover System Problems (Continued).
       PROBLEM TYPE
                                         POSSIBLE CONSEQUENCE
                                      PROBLEM PREVENTION
       2.  COVER SYSTEM DEGRADATION (Con.)
       • GT clogging
         .  GT over drainage layer

         .  GT around pipe

       • Drainage layer clogging
                                         Excessive water head (May trigger
                                         instability, See "Stability" in 3)
                                         Excessive water head (May trigger
                                         instability, See "Stability" in 3)
                                         Excessive water head (May trigger
                                         instability, See "Stability" in 3)
                                      GT selection

                                      No GT around pipe

                                      Drainage material selection
       3. STABILITY
       (STATIC, HYDRODYNAMIC, SEISMIC)
       •  Surface layer or protection layer slide or
          cracking of intermediate soil cover during
          earthquakes
       •  Cover system slide	
                                         Erosion (May trigger uplift by gas, See
                                         "Displacement" in 4); Outdoor exposure of
                                         underlying materials
                                          Major barrier layer breach
                                      Design, Testing, Construction method
                                      Design, Testing, Construction method
dn
4.  DISPLACEMENT
• Waste settlement or subsidence
• Differential settlement at connections
• Cover system uplift
  .  Leachate

  .  Gas

  .  Wind
GM: rupture or yield; GCL: open joints
GM: rupture or yield

GM: yield
(May trigger instability, See "Stability" in
3)
GM: rupture or yield
(May trigger instability, See "Stability" in
3)
GM: rupture or yield	
Design, GM selection, Construction,
Operation
Design, GM selection

Underdrain, Protection layer

Gas collection system, Protection layer

Design, Installation method, CQA
-------
       Table 4-3. Consequence and Prevention of Impoundments Liner System Problems.
       PROBLEM TYPE
POSSIBLE CONSEQUENCE
PROBLEM PREVENTION
(In
       1. LINER CONSTRUCTION
       •  GM holes
         .  Seams, Panels

         .  Connections

       •  GM condition
         .  Wrinkles
         .  Scratches
       •  GM uplift by wind
       •  CCL defects
         .  Stones, Debris
         .  Improper compaction
       •  Material in contact with GM
         .  Overlying (Soil protection layer)
         .  Underlying (Debris, Sandbags)
       •  Contamination of CCL in top liner
Defect

Defect
Yield (Potential defect)
Yield (Potential defect)
Rupture or yield

High permeability, Damage to GM
High permeability

Damage to GM
Damage to GM
Incorrect  interpretation of LDS chemistry
Specifications, CQA, Ponding test, Leak
survey, Conductive GM
Connection design, CQA, Ponding test, Leak
survey, Gas tracer test, Conductive GM

Installation temperature and method, CQA
CQA
Design, Installation method, CQA

Specifications, CQA
Specifications, CQA

Cushion, CQA
CQA
Runon control, Equipment maintenance, CQA
       2. LINER DEGRADATION
       •  GCL outdoor exposure


       •  GM stress cracking
       •  GM outdoor exposure
       •  CCL outdoor exposure
       •  GM/CCL composite outdoor exposure
       •  Fire
       •  GM exposure to chemicals
       •  CCL/GCL exposure to chemicals

       •  Wave action
       •  Drifting objects or ice
       •  Dropped objects	
Hydrated and swollen (Low shear
strength, May trigger instability, See
"Stability" in 4)
Major liner breach
Weakening, Embrittlement
Cracking, Erosion
CCL: Cracking
GM:  Major breach; CCL: Cracking
Weakening, Permeation, Holes (Defect)
Increased permeability

GM: rupture or yield
GM: puncture
GM: puncture
Installation method, CQA
Resin selection, Protection layer, Minimize
tension (Design, Installation temperature and
method)
GM selection, Protection layer
Protection layer
Protection layer
Operation
GM selection, Operation (Chemical control)
Material selection, Operation (Chemical
control)
Protection layer
Protection layer, Operation QC
Protection layer, Operation QC
-------
       Table 4-3.  Consequence and Prevention of Impoundments Liner System Problems (Continued).
       PROBLEM TYPE
                                         POSSIBLE CONSEQUENCE
                                      PROBLEM PREVENTION
01
oo
       3. LDS
       • GT holes due to burning
         Drainage layer not sealed in anchor trench

         Pipe not connected at joints
         Debris in pipe trenches
         Sandbags in LDS
         GT needles
         Soil protection layer outdoor exposure
         Clogging of GT around pipe
         Drainage layer clogging
         Pipe failure
         .  Pipe crushing

         .  Pipe weld separation
• Water expelled from CCL into LDS
• Equipment failure
  .  Pumps, Cleanout

  »  Flowmeter, Check valve
Clogging of drainage layer, pipes, or
sumps (See "Clogging" below)
Water intrusion in LDS, Incorrect
monitoring
Excessive leachate head
Potential clogging of pipes or sump
Excessive leachate head
Damage to GM
Erosion
Excessive leachate head
Excessive leachate head

Excessive leachate head, Potential GM
damage
Excessive leachate head, Potential GM
damage
Incorrect monitoring

Excessive leachate head

Incorrect monitoring
Installation method, CQA

Design, CQA

CQA
CQA
CQA
ManufacturerQC, GT plant inspection, CQA
Design
No GT around pipe
Drainage material selection

Pipe selection/Design

Construction method, CQA

Design

Design (e.g., sump capacity), Equipment
selection
Equipment selection
       4. STABILITY AND DISPLACEMENTS
       • Global stability (Foundation failure)
       • Liner system slide
       • Foundation settlement or subsidence

       • Differential settlement at connections
       • Liner system uplift
         .  Water
            Wind
                                         Major liner breach
                                         Major liner breach
                                         GM: rupture or yield; GCL: open joints

                                         GM: rupture or yield

                                         GM: rupture or yield; GCL: open joints
                                         (May trigger instability, See "Stability"
                                         above)
                                         GM: rupture or yield
                                      Investigation, Design
                                      Design, Testing, Construction method
                                      Investigation, Design, Foundation
                                      improvement, GM selection
                                      Design, GM selection

                                      Underdrain, Protection layer
                                                                                      Design, Installation method, CQA
-------
Each of these five categories of problems is discussed below. The categories are
based on the assumption that the liners have a GM component. Gas migration from an
active landfill without a cover system may have a major environmental impact, such as
the groundwater contamination by VOCs that occurred at landfill L-8.  This problem
does not directly affect liner integrity and, therefore, does not fit within the above
categories. The gas migration beyond the liner system is believed to be a rare
occurrence related to placement of an asphalt layer over the edge of the  liner system
and inadequate landfill gas control.  The problem is not discussed further in this section.

Major breaches in a landfill liner are likely to cause leakage into the ground that may
have a significant  impact on the environment.  Operation of the landfill cell where such a
breach occurs must generally be stopped.  As indicated  in Table F-4.1, major breaches
in landfill liners can occur mostly as a result of liner system instability (i.e. slide).  Major
breaches in landfill liners can also occur as a result of relatively rare events, such as
stress cracking, fire, and major mistakes (e.g., advancing a borehole through  a liner).

GM or GCL holes  can provide a pathway for leachate migration and, therefore, can
have a detrimental impact on the environment.  As shown  by many studies, even
properly installed GMs with adequate CQA programs will exhibit a small number of
construction holes (i.e., 1 to 10 per hectare). These few holes are mostly due to
imperfect seaming and are virtually impossible to eliminate. There have been no
identified environmental impacts results from these holes.  Nonetheless, every effort
should be  made to reduce the frequency of seam holes through rigorous  control of the
seaming and CQA processes. In addition,  every effort should be made to eliminate
other less  frequent causes of GM holes. Many of the holes result from construction
activities, and they should be detected during construction. As indicated  in Table F-4.1,
GM liner holes can occur in landfills due to a variety of causes or as a result of a variety
of activities:

•  manufacturing  holes (a rare occurrence);
•  installation holes;
•  damage by materials placed in contact  (or in the vicinity) of GMs;
•  damage during operations (e.g.,  GM rupture by blade of bulldozer placing  soil
   protection layer); and
•  GM tear due to excessive stretching resulting from various types of displacements.

Holes in GCLs can occur due from displacements that result in open GCL joints and
from wrinkles or protrusions that cause GCL thinning.

A number  of other problems can increase the potential for leakage through the landfill
liner.  These include problems that result in: (i) an increase of the leachate head on top
of the liner; (ii) a CCL constructed with a high hydraulic conductivity due to improper
materials or compaction; or (iii) degradation of the GCL or CCL component of a
composite liner due to outdoor exposure or exposure to chemicals. As shown in Table


                                      F-59
-------
F-4.1, leachate heads in landfills can increase due to problems related to LCRS or LDS
installation, degradation, malfunction, and operation, such as:

•  unconnected LCRS or LDS pipes;
•  clogging of the LCRS or LDS components related to holes in GT filters, debris in the
   LCRS or LDS, and buildup of sediments, chemical precipitates, or biological mats in
   the LCRS  or LDS;
•  pipe failure;
•  GN or GC compressive creep;
•  failure of leachate pumps or sump liquid level indicator; and
•  operation errors.

All problems that result in a weakening of the GM increase the risk of GM holes. As
shown in Table F-4.1, these problems include:

•  GM installation problems, such as wrinkles and scratches;
•  action of materials in contact with the GM that cause scratches on the GM surface or
   cause the  GM to yield;
•  exposure of the GM  to a variety of agents that weaken or embrittle it (e.g., ultraviolet
   light, chemicals); and
•  displacements that cause yield of the GM.

Problems that result in incorrect monitoring of the leachate quantity removed from a
landfill generally do not  have a direct impact on the environment. However, it is
important to have a properly functioning leachate measuring system, especially for
LDSs.  As discussed above, LDS flow rates have been used to evaluate whether a liner
system is functioning adequately and if liner repairs  are necessary.  LDS flow chemistry
has been used to evaluate whether primary liner leakage occurred.

The problems identified in this study fall in all five of the above categories. With the
exception of the problems at landfills L-2, L-5,  L-8, L-9, L-12, L-29, and L-36, the
problems were detected and remedied shortly after they occurred and did not result in
any identifiable environmental impact.  Groundwater contamination was found at landfill
L-8 and is being remedied. Leakage through the GM primary liner of landfills L-5 and L-
12 and leakage at the LCRS pipe penetration of the GM primary liner of landfill L-9 were
detected but not remedied. The desiccated CCL component of the single composite
liner on the side slopes of landfill L-2 was also not remedied. As previously discussed,
the potential for significant environmental impacts from the conditions at landfills L-2, L-
5, L-9, and L-12 is expected to be negligible, especially given that landfills L-5, L-9, and
L-12 have composite secondary liners.  The separated LCRS  pipes in landfills L-29 and
L-36 were not repaired,  but they only result in a small increase in the potential for liner
leakage as a result of a localized increase in leachate head on the liners.
                                      F-60
-------
F-4.2.3    Cover Systems

Problems that affect cover systems are generally less likely to increase the potential for
environmental impacts related to waste containment system integrity than problems that
affect liner systems. From Table F-4.2, it appears that, with the exception the problem
of using contaminated topsoil as the surface layer, the identified problems that affected
landfill cover systems can generally be grouped into two categories based on their
potential environmental impact: (i) problems that result in leakage, or increased risk of
leakage, through the cover system; and (ii) problems that result in the release of landfill
gas to the atmosphere or the ground.  These two categories are discussed below.
Using contaminated topsoil as the cover system surface layer may contaminate runoff.
However, this is a rare occurrence. Though it is not listed in Table F-4.2, sediment and
runoff loading from cover systems to offsite properties may also have a potential
environmental impact. However, since the impacts from transported sediments and
runoff do not affect waste containment system integrity, the focus of this appendix,
these impacts are not discussed further.

Problems that result in leakage, or increased risk of leakage, through a landfill cover
system are far less critical than problems that result in leakage, or increased risk of
leakage, through a liner system for the following reasons: (i) most leakage through the
cover system is absorbed by the underlying waste and controlled by the liner system;
and (ii) the cover system is exposed and, thus, can be visually inspected for  problems
and maintained.  Therefore,  major cover system  breaches, which occur mostly as a
result of instability, would likely be detected and repaired soon after they occurred.
Deformations of the cover system, which could result in holes in the cover system
barrier layer, would also be apparent by visual inspection and could be repaired.

From the foregoing discussion it appears that the only impact to the environment of a
landfill cover that needs to be considered is the release of gas. The release  of gas to
the environment is a serious problem,  but it is relatively easy to solve. Typical
measures are:

•  use of a low-permeability barrier layer that includes a GM;
•  proper sealing of the cover at its periphery; and
•  use of an adequate gas extraction system.

The problems identified in this study fall in both of the above categories.  In all cases,
the problems were identified and remedied shortly after they occurred and did not result
in any detectable environmental impact.

F-4.2.4    Impoundment Liner Systems

Potential environmental impacts of the problems that can affect impoundment liner
systems can be ranked as follows, from the most to the least serious:


                                      F-61
-------
•  major liner breach;
•  hole in GM or GCL;
•  increased risk of leakage through existing holes;
•  increased risk of hole in liner; and
•  incorrect monitoring.

Each of these five categories of problems is discussed below.

Major breaches in an impoundment liner system are likely to cause leakage into the
ground that may have a significant impact on the environment. Operation of the
impoundment where such a breach occurs  must generally be stopped, and the
impoundment must be emptied for repair.  However, emptying an impoundment is
relatively easy, as compared to removing waste from a landfill, and the leakage stops as
soon as the impoundment is empty. As indicated in Table F-4.3, major breaches in
impoundment liners can occur as a result of three principal causes: (i) stress cracking,
which has occurred in the past in a number of impoundments, but occurs far less often
since progress in GM resin selection has been made; (ii) liner system instability; and (iii)
major operational mistakes (e.g., the dropping of a large heavy object on the liner).
Major breaches in impoundment liners can  also occur as a result of relatively rare
events, such as fire.

GM or GCL holes can provide a pathway for leachate migration and, therefore, can
have a detrimental impact on the environment. As shown by many studies, even
properly installed GMs with adequate CQA programs will exhibit a small number of
construction holes (i.e., 1 to 10 per hectare). These few holes are mostly due to
imperfect seaming and are virtually impossible to eliminate.  There  have been no
identified environmental impacts results from these holes. Nonetheless, every effort
should be made to reduce the frequency of seam holes through rigorous control of the
seaming and CQA processes.  In addition, every effort should be made to eliminate
other less frequent causes of GM holes.  Many of the holes result from construction
activities, and they should be detected during construction. For example, small leaks
can be detected by performing a ponding test, which is relatively easy in the case of an
impoundment. As indicated in Table F-4.3, GM liner holes can occur in impoundments
due to a variety of causes or as  a result of a variety of activities:

•  manufacturing holes (a rare occurrence);
•  installation holes;
•  damage by materials placed in contact (or in the vicinity) of GMs; and
•  GM tear due to excessive stretching resulting from various types of displacements.

Holes in GCLs can  occur due from displacements that result in open GCL joints and
from wrinkles or protrusions that cause GCL thinning.
                                     F-62
-------
A number of other problems can increase the potential for leakage through the
impoundment liner. These are problems that result in: (i) a CCL constructed with a high
hydraulic conductivity due to improper materials or compaction; (ii) degradation of the
GCL or CCL component of a composite liner due to outdoor exposure or exposure to
chemicals; and (iii) an increase of the leachate head on top of the secondary liner (i.e.,
problems related to the  LDS).

All problems that result  in a weakening of the GM increase the risk of GM holes.  As
shown in Table F-4.3, these problems include:

•  GM installation problems, such as wrinkles and scratches;
•  action of materials in contact with the GM that cause scratches on the GM surface or
   cause the GM to yield;
•  exposure of the GM  to a variety of agents that weaken or embrittle it; (e.g., ultraviolet
   light, chemicals); and
•  displacements that cause yield of the GM.

Problems that result in incorrect monitoring of the liquids in the LDS generally do not
have a direct impact on the environment.  However, it is important to have a properly
functioning  liquids measuring system, especially for LDSs. LDS flow rates have been
used to evaluate whether a liner system is functioning adequately and if liner repairs are
necessary.

The problems identified in this study fall in all five of the above categories. In all cases,
the problems were detected and remedied shortly after they occurred and did not result
in any identifiable environmental impact.

F-4.3     Construction,  Operation, and Maintenance Impacts

The main impacts of the problems in this investigation were interruption of waste
containment system construction and operation and increased maintenance. The
identified problems that most often disrupted construction and were required to be
repaired before construction proceeded were related to:

•  holes in GM liners and at pipe penetrations of liners;
•  large wrinkles in HOPE GM liners;
•  degradation of exposed geosynthetics;
•  uplift of constructed  liners by groundwater or infiltrating surface water; and
•  erosion of unprotected soil layers (CCLs, sand drainage layers, soil protection
   layers).

Relative to problems that disrupt operation and may require waste relocation, problems
that interfere with construction are relatively easy to remedy. In  addition, the frequency
of these problems can be reduced with good design,  construction, and  CQA. Holes in


                                      F-63
-------
GM liners and at connections can generally be located with leak location surveys,
ponding tests, dye, or other nondestructive methods and then repaired.  Large wrinkles
in HOPE GM liners can be pulled out or cut out or the GM can be allowed to cool during
the evening hours and contract. GM wrinkling can also be reduced through
development of an adequate strategy for placing and covering the GM for each
particular site.  Exposed GTs, GCLs, or CCLs that degrade are replaced with new
materials or, for CCLs, reconstructed.  Eroded soil layers are replaced and/or regraded.
Uplift of constructed liners is probably the most complex and time consuming problem
because it requires that the water or gas uplifting the liner be controlled before a new
liner is constructed.

The identified problems that most often disrupted facility operation and were required to
be repaired before operation proceeded were related to:

•  holes in GM liners and at pipe penetrations of liners;
•  failure of liner system and cover system slopes; and
•  clogging of GTs in LCRSs.

Problems that disrupt operation are generally more severe in terms of required repairs
than those that interfere with construction and may require waste relocation.
Consequently, problems that disrupt operation generally require more time to remedy
than problems that are identified and repaired during construction. Problems that
involve major breaches of liner systems or cover systems (e.g., failure of landfill liner
system slopes) may require months to repair. However, the frequency of these
problems can be reduced with good design, construction, CQA, and maintenance.

The identified problems that most often required maintenance were related to:

•  erosion of soil layers (sand drainage layers, soil protection layers);
•  repair of LCRS or LDS flow rate measuring and removal systems; and
•  cracking of soil intermediate cover layers  after earthquakes.

Problems that require maintenance may be more severe in terms of required repairs
than those that interfere with construction, but are generally less severe than those that
interfere with operation. In addition, problems that require maintenance are more likely
to be reoccurring.  The frequency of these problems, however, can be reduced with
good design, construction, CQA, and maintenance.

F-4.4     Cost Impacts

The costs of remedying the problems can be  significant. For the identified problems,
the costs at the time the remedies were implemented ranged from less than $10,000 for
repairs of GM holes identified by leak location surveys during construction to more than
several million dollars for repair of a liner system slope failure that occurred during cell


                                      F-64
-------
operation. In general, problems that impacted operation were more expensive than
those that impacted construction or maintenance.  Problems that impacted operation
were likely to cause the most damage, require relocation of waste, and result in
suspension of waste placement in the containment system.  However, certain problems
that impact maintenance, such as erosion of soil layers, may ultimately be more costly
than other problems if the problems that impact maintenance reoccur.
F-5       Conclusions

Based on the results of the investigation into waste containment system problems
presented in Chapters F-2 to F-4 of this appendix, the following conclusions are drawn:

•  This investigation identified 74 modern landfill and impoundment facilities (i.e.,
   facilities designed with components substantially meeting current EPA regulations and
   constructed and operated to the U.S. state-of-practice) that had experienced a total
   of 85 waste containment system problems. This number of facilities is relatively
   small in comparison to the over 2,000 modern landfills and surface impoundments
   nationwide. The search for problem facilities for this study was not exhaustive, and
   it is certain that there are other facilities with problems similar to those described in
   this appendix.
•  The investigation focused on landfill facilities:  94% of the identified problems
   occurred at landfills.  Among the landfill problems, 70% were liner system  related
   and 30% were cover system related. The ratio of liner system problems to cover
   system problems is probably exaggerated by the fact that a number of the facilities
   surveyed were active and did not have a cover system.
•  Based on the waste containment system component or attribute criterion, the
   identified problems were classified as follows, in order of decreasing frequency:
   .   cover system stability:  21%;
   .   landfill liner construction: 17%;
   .   landfill liner system stability:  14%;
   .   landfill liner degradation: 8%;
   .   landfill LCRS or LDS construction:  7%;
   .   landfill LCRS or LDS degradation:  6%;
   .   landfill LCRS or LDS malfunction:  5%;
   .   landfill LCRS or LDS operation:  5%;
   .   landfill liner system displacement:  5%;
   .   impoundment liner construction: 4%;
   .   cover system construction: 2%;
   .   cover system degradation: 2%;
   .   cover system displacement:  2%;
   .   impoundment liner degradation: 1%;and
   .   impoundment liner system stability: 1%.
                                     F-65
-------
•  Using this criterion, these problems can also be grouped into the following general
   categories (Figure F-5.1):
   .   liner system or cover system slope stability or displacement: 45%;
   .   liner, LCRS or LDS, or cover system construction: 28%;
   .   liner, LCRS or LDS, or cover system degradation: 17%; and
   .   LCRS or LDS malfunction or operation: 10%.
                                Degradation
                                    17%
  Construction
      28%
LCRS or LDS
 Malfunction/
  Operation
    10%
                                                 Stability or
                                                Displacement
                                                    45%
Figure F-5.1. General distribution of problems by waste containment system
component or attribute criterion.
   Based on the principal human factor contributing to the problem criterion, the
   identified problems are classified as follows (Figure F-5.2):
   .   design: 51%;
   .   construction: 35%; and
   .   operation: 14%.
                                     F-66
-------
      Construction
         35%
Operation
   14%
                                           Design
                                            51%
Figure F-5.2.  Distribution of problems by principal human factor contributing to
   the problem criterion.
   Problems that occurred at two or more facilities and the number of facilities at which
   they occurred are as follows, listed in the order of presentation in Chapter F-3:
   .   leakage through holes (construction- or operation-related) in an HOPE GM
      primary liner (5 landfills);
   .   leakage through an HOPE GM primary liner or HOPE GM/CCL composite
      primary liner at the LCRS pipe penetration of the liner (3 landfills);
   .   severe wrinkling of an HOPE GM liner during construction (2 landfills);
   .   liner damage by fire (2 landfills);
   .   liner damage during well installation (2 landfills);
   .   rainwater entered  the LDS through the anchor trench (2 landfills);
   .   HOPE LCRS pipe was separated at joints (2 landfills);
   .   erosion of the sand layer on the liner system side slopes (2 landfills);
   .   degradation  of polypropylene nonwoven GT filters due to outdoor exposure (2
      landfills);
   .   waste fines clogged the needlepunched nonwoven GT filter in the LCRS piping
      system (2 landfills);
   .   clogging and other problems with the leachate pump or flow rate measuring
      system (3 landfills);
   .   liner system  slope failure due to static loading (10 landfills);
   .   liner system  damage due to earthquakes (2 landfills);
   .   uplift of liner system geosynthetics by landfill gas after erosion of the overlying
      sand layer (2 landfills);
   .   uplift of composite liner by surface-water infiltration during construction (2
      landfills);
                                      F-67
-------
.   cover system slope failure during construction (4 landfills);
.   cover system slope failure after rainfall or a thaw (11 landfills);
.   soil cover damage due to earthquakes (3 landfills); and
.   leakage through holes in the HOPE GM primary liner or the HOPE GM
   component of the GM/CCL composite primary liner (2 impoundments).
For problems that occurred at three or more facilities, the principal human factor
contributing to the problem criterion, detection of the problem, causes of the
problem, and remedy of the problem are described below:
.   Leakage through holes in an HOPE GM primary liner occurred at five landfills. In
   each case, the holes were attributed to construction or, at one landfill, possibly
   operation factors.  At two of the landfills, leakage was first detected during
   electrical leak location surveys performed as part of CQA and by the relatively
   high LDS flow rates that occurred after rainwater ponded in a landfill. At the
   remaining three landfills, leakage was first detected during operation by the
   relatively high LDS flow rates and the  color of and chemical constituents in the
   LDS liquid.  The cause of the leakage was attributed to construction-related holes
   in the GM.  However,  at one landfill, where waste was placed directly on liner
   system geosynthetics (i.e.,  there is no soil protection layer), the GM may have
   been damaged during waste placement. The leakage problem was resolved at
   four landfills by repairing the GM holes; at the remaining  landfill, the problem,
   clearly identified only after the cell had been covered with waste, was partially
   remedied by lowering the "pump on" liquid level in the LCRS sump.
.   Leakage through an HOPE GM primary liner or HOPE GM/CCL composite
   primary liner at the LCRS pipe penetration of the liner occurred at three landfills.
   This leakage was attributed to construction factors at two of the landfills and
   operation factors at the third landfill. At two of the landfills, leakage at the pipe
   penetration was detected during construction after rainwater ponded over the
   penetration and LDS flow rates increased. The cause of the leakage was
   construction defects in the pipe penetration; it is difficult to construct a defect-free
   pipe penetration, even when extra measures are taken to enhance the integrity of
   the connection. At the remaining landfill, leakage was detected during operation
   when the average LDS flow rate increased significantly.  For this landfill, the pipe
   penetration was damaged during operation when a rubber-tired loader trafficked
   over it. The pipe penetrations were repaired; however, at one landfill where the
   problem was detected during construction, the repairs did not significantly
   decrease LDS flow rates; thus there must have existed a penetration defect that
   was not located.
.   Clogging and other problems with the  leachate pumps or flow rate measuring
   system occurred at three landfills. These problems were attributed to design
   factors at one of the landfills and operation factors at the  other two  landfills. The
   problems, which were identified during routine operations, included: (i) clogging
   of the air lines and failure of the compressor for the control system; (ii) drift of the
   leachate level measurement system;  (iii) drift of the "pump on" time setting; (iv)
   burn out of pumps due to control system problems; (v) clogging of pumps; (vi)
                                   F-68
-------
clogging of mechanical flowmeters; (vii) damage to electrical equipment by
electrical storms; (viii) check valve failure; and (ix) inaccurate measurement of
LCRS or LDS flow rates due to the above equipment problems. These problems
appear to have been primarily caused by: (i) inadequate overall mechanical
system design; (ii) using equipment that was less reliable than was needed; (iii)
using equipment that was not compatible with the landfill leachate; and (iv) not
performing equipment maintenance often enough.  These problems were
primarily remedied by equipment maintenance, repair,  and replacement.
Liner system slope failure due to static loading occurred at ten landfills. These
problems were attributed to design factors at seven of  the landfills and operation
factors at the remaining three landfills.  Slope failure occurred during construction
at two of the landfills and during operation at the remaining eight landfills. The
problem was detected by visual observation of mass movement of the liner
system, cracking of soil layers near the slope crest, and tearing, tensioning, or
wrinkling of geosynthetics. The primary causes of failure were: (i) using
unconservative presumed values for the critical  interface shear strength; (ii) not
evaluating the  critical condition for slope stability (e.g.,  liner system with waste at
intermediate grades); (iii) not accounting for or underestimating seepage
pressures;  (vi) not accounting for moisture at the GM/CCL  interface (which
weakens the interface) due to spraying of the CCL with water and thermal
effects; and (v) not maintaining the drainage layer outlets free of snow and ice,
which can lead to increased seepage pressures. The slope failures were
remedied by reconstructing the damaged liner systems, sometimes with different
materials, and  developing new  construction procedures to reduce moisture at the
GM/CCL interface.
Cover system slope failure during construction occurred at four landfills. These
problems were attributed to design factors at two of the landfills and construction
factors at the remaining two landfills. Slope failure was detected by visual
observation of mass movement of the cover system, cracking of soil layers near
the slope crest, and  wrinkling of geosynthetics at the toe of the cover system
slope.  The primary causes of failure were:  (i) placing soil over the side slope
geosynthetics from the top of the slope downward,  rather from the toe of the
slope upward;  (ii) not considering the effects of variation in the tested
geosynthetics, accuracy of test methods, and test conditions on the interface
shear strength to use in design; and (iii) using unconservative presumed values
for the critical interface shear strength. The problems were remedied by
reconstructing  the cover systems using different cover system materials that
result in higher interface shear  strengths and placing soil over side slope
geosynthetics from the toe of the slope upward.
Cover system slope failure after rainfall or a thaw occurred at eleven landfills.  At
all of these landfills,  the failures were attributed to design factors.  Slope failure
occurred during the post-closure period and was detected by visual observation
of mass movement of the cover system, cracking of soil layers near the slope
crest, and wrinkling of geosynthetics at the toe of the cover system slope. The
                                F-69
-------
   primary causes of failure appeared to be: (i) not accounting for, or
   underestimating, seepage pressures; (ii) clogging of the drainage system, which
   can lead to increased seepage pressures; and (iii) not accounting for moisture at
   the GM/CCL interface (which weakens the interface) due to rain falling on the
   CCL surface during construction and freeze-thaw effects.  In general, the
   problems were remedied by reconstructing the cover systems with  new drainage
   systems or different materials.
.  So/7 cover damage due to earthquakes occurred at three landfills. These
   problems all occurred during operation and were attributed to design factors.
   The damage, which was detected by visual inspection, consisted of surficial
   cracking of soil intermediate cover occurring primarily near locations with contrast
   in seismic response characteristics (e.g., top of waste by canyon walls).  The
   damage was expected and dealt with as an operation issue through post-
   earthquake inspection and repair (i.e., regrading and revegetating the cracked
   soil layers).
Almost all of the problems identified in this investigation were detected shortly after
they occurred by visual observation or evaluation of monitoring data.
Of the problems in this study for which the remedy was  identified, six problems were
not completely repaired because their environmental impacts were not expected to
be significant and because: (i) the source of the problem could not be identified; (ii)
the problem was not worsening; (iii) repair of liner systems or LCRS pipes after
waste placement would be extremely difficult and expensive;  and/or (iv) additional
liner system damage could occur in any attempt to excavate the waste and repair
the liner system.
The problems only resulted in an  identified environmental impact to groundwater or
surface-water quality by leachate or landfill gas at one facility, landfill L-8. At this
MSW landfill, groundwater impact by VOCs was attributed to gas migration through
a relatively permeable soil layer that secured the  edge of the  GM liner and extended
from the crest of the liner system  side slope to beyond the liner system.  The
problem was resolved by installing additional gas extraction wells in the landfill.
Without the measures taken to correct the problems at some of the other facilities,
however, adverse environmental  impacts could have eventually occurred at these
facilities.
The main impacts of the problems identified in this investigation are interruption of
waste containment system construction and operation, increased maintenance, and
increased costs.
The identified problems that most often disrupted construction and were required to
be repaired before construction proceeded were related to:
.  holes in GM liners and at pipe penetrations of liners;
.  large wrinkles in HOPE  GM  liners;
.  degradation of exposed geosynthetics;
.  uplift of constructed liners by groundwater or infiltrating surface water; and
.  erosion of unprotected soil layers (CCLs,  sand drainage layers, soil protection
   layers).
                                   F-70
-------
Problems that disrupt operation are generally more severe in terms of required
repairs than those that interfere with construction and may require waste relocation.
Consequently, problems that disrupt operation generally require more time to
remedy than problems that are identified and repaired during construction.
Problems that involve major breaches of liner systems or cover systems (e.g., failure
of landfill liner system slopes) may require months to repair.  The identified problems
that most often disrupted operation and were required to be repaired  before
operation proceeded were related to:
.   holes in GM liners and at pipe penetrations of liners;
.   failure of one or more components of a liner system or cover system on landfill
   slopes; and
.   clogging of GTs in LCRSs.
Problems that require maintenance may be more severe in terms of required repairs
than those that interfere with construction, but are generally less severe than those
that interfere with operation.  In addition, problems that require maintenance are
more likely  to be reoccurring. The identified problems that most often required
maintenance were related to:
.   erosion  of soil layers  (sand drainage layers, soil protection layers);
.   repair of LCRS or LDS flow rate measuring and removal systems; and
.   cracking of soil intermediate cover after earthquakes.
The costs of remedying the problems can be significant. For the identified problems,
the costs at the times the remedies were implemented ranged from less than
$10,000 for repairs of GM holes identified by leak location surveys during
construction to more than several million dollars for repair of a liner system slope
failure that occurred during cell operation.  In general, problems that impacted
operation were more expensive than those that impacted construction or
maintenance.  However, certain problems that impact maintenance, such  as erosion
of soil layers,  may ultimately be more costly than other problems if the problems that
impact maintenance reoccur.
Even though there was only evidence of environmental impact at one of the waste
containment systems in this study, the landfill industry should do more to avoid
future problems in order to: (i) reduce the potential risk of future environmental
impact; (ii) reduce the potential health and safety risk to facility workers, visitors, and
neighbors; (iii) increase public confidence in the performance of waste containment
systems; (iv) decrease potential impacts to construction, operation, and
maintenance; and (v) reduce costs associated with the investigation and repair of
problems.
Importantly, all of the design, construction, and operation problems identified in this
investigation can be prevented using available design approaches,  construction
materials and procedures, and operation practices. It is the responsibility of all
professionals involved in the design, construction, operation, and closure of waste
containment systems to improve the practice of waste containment system
engineering.  Owners must be prepared to adequately fund the levels of design and
CQA activity necessary to properly design and construct waste containment
                                   F-71
-------
   systems. Design engineers must improve their practice to avoid the types of
   problems identified herein.  Earthwork contractors, geosynthetics installers, and
   landfill operators all must be properly trained, supervised, and committed to the
   "quality goals" necessary to eliminate problems.
F-6       Recommendations to Reduce Incidence of Identified Problems

F-6.1     Introduction

This chapter of the appendix presents general and specific design, construction, and
operation measures to reduce the incidence of the waste containment system problems
described  in this appendix. These measures are not new; they have been used
extensively for other engineered structures, such as dams. The measures include
widely available design approaches, construction procedures, and operation practices.
Many recommendations for landfill liner systems also apply to cover systems and
impoundment liner systems, and vice versa.  Because of this, the recommendations are
grouped to apply to following broad categories:

•  general;
•  liners and barriers;
•  drainage systems;
•  surface layers and protection layers;
•  liner system and cover system stability; and
•  liner system and cover system displacements.

Recommendations for  each of these categories are presented below.  The designations
of the facilities with problems that lead to the recommendations are also listed for each
recommendation except those in the general category.  It should be noted that the
problems and remedies are site specific.

F-6.2     General

General recommendations intended to reduce the occurrence of problems include:

•  information dissemination (e.g., this appendix);
•  training of design engineers to better understand waste containment system design
   fundamentals and to avoid  the types of design problems described in this appendix;
•  training of design engineers to be  better prepared to develop waste containment
   system specifications and CQA plans that are complete and precise, that include the
   construction-related assumptions made during design, and that require construction
   and CQA procedures to identify and prevent the kinds of construction problems
   identified in this appendix;
                                      F-72
-------
•  training of CQA personnel in standard CQA procedures to avoid the types of
   construction problems identified in this appendix; for engineering technicians, this
   training can be demonstrated through the National Institute for Training in
   Engineering Technologies (NICET) certification program;
•  training of contractors to avoid the types of construction problems identified in this
   appendix;
•  development of better construction materials, techniques, and quality control/quality
   assurance procedures to prevent the kinds of construction problems  identified in this
   appendix;
•  development of better operations manuals to describe and provide controls for
   procedures to be followed by landfill operations personnel;
•  training of facility operators to better avoid the types of operation problems identified
   in this appendix;
•  training of facility operators to better detect and quickly report problems occurring
   during  operation; and
•  performing periodic independent audits to verify that the specified operation
   procedures are being practiced.

F-6.3     Liners and Barriers

The following recommendations are made to reduce the frequency of liner and barrier-
related problems.

Design

•  Resin used to manufacture HOPE GM should be resistant to stress cracking.  This is
   currently evaluated using the notched constant tensile load test (ASTM D 5397).
   This test should be required in project specifications. (S-1)
•  Project specifications should require that both the inner and outer tracks of GM
   fusion seam samples taken for destructive testing meet the project seam
   requirements.  Failure of one track is generally indicative of overall seaming
   problems and can results in increased stress concentrations  in the adjacent track. In
   addition,  testing both tracks may allow seaming problems to be identified and
   corrected quicker.  (L-17, C-16)
•  The  potential for GM damage during placement of a soil layer over a GM can be
   reduced by protecting the GM.  Measures for GM  protection should be incorporated
   into the design and specifications. Measures include placing a protection layer (e.g.,
   thick GT  cushion or GC drainage layer) over the GM, using a greater initial lift
   thickness of soil above the GM,  and using construction equipment with low ground
   pressure to place soils over the  GM.  The protection measures should be selected
   based on the characteristics of the soil to be placed  (e.g., angularity,  maximum
   particle size), the thickness of the soil layer, the type of equipment placing the soil,
   and whether CQA will be performed during soil placement. If the soil layer is placed
   during  operation without CQA, extra GM protection is necessary. (L-3)


                                      F-73
-------
•  GMs located in areas subjected to high static and dynamic stresses from
   construction equipment, such as beneath temporary access roads, require an even
   higher level of protection than GMs not subjected to high stresses.  These protection
   measures should be incorporated into the design and specifications. (L-3)
•  GM should also be protected during waste placement over the GM. Protection
   measures should be incorporated into the design and specifications. Measures
   include installing a protection layer (e.g., thick GT cushion, GC drainage layer, or soil
   layer) over the GM, using spotters to direct equipment operators during placement of
   waste over the GM, and placing only select waste over the GM. Protection
   measures should be selected with consideration of waste characteristics and the
   equipment placing the waste. (L-6)
•  Sensitive areas of a liner system (e.g., at pipe penetrations) should be designed to
   be untraffickable by berms, bollards, or other means to decrease the potential for
   damage to these areas.  (L-7)
•  It is difficult to construct pipe penetrations of liners to be defect free. A method that
   was successful for one landfill (i.e., L-11) was to fill the space between the pipe and
   pipe boot with bentonite slurry.  Until new methods for constructing better
   connections between GMs and ancillary structures have been developed and tested,
   designs without pipe penetrations (i.e., designs with internal sumps) should be
   preferred.  (L-9, L-11)
•  Internal sumps typically have sustained leachate heads at greater depths than other
   locations within the landfill and have seamed corners, which may contain holes. To
   decrease the rate of leakage through GM holes at sumps, the sump design should
   include additional liner components, such as a GCL, beneath the GM liner in the
   sump area, even if the GM is already underlain by a CCL.  A design with a
   prefabricated GM sump may also be considered. Bonaparte (1995) provides
   additional  discussion of this  design approach.  (L-5)
•  The potential for landfill gas  to migrate over the geosynthetics at the edge of the liner
   system must be considered  in design. The potential for gas migration into the
   subsurface can  be reduced by collecting gas generated in the landfill, using low-
   permeability soils over the edge of the liner system, and modifying the edge of the
   liner system so that the liner extends back up to the ground surface (like a reverse
   anchor trench).  (L-8)

Construction

•  Construction equipment should be inspected for fuel and oil leaks, and those leaks
   should be repaired prior to using the equipment in liner system construction to avoid
   liner and LDS contamination. Runon should be controlled so that it does not
   contact, and potentially contaminate, the liner  and LDS during construction.  (L-43)
•  Liners and barriers should be constructed in manageable increments that ensure
   protection of the liner and barrier materials under seasonal weather changes. (L-19)
•  CCLs should not be constructed with materials containing construction debris or
   large particles, even if prior to GM installation the CCL  has a smooth surface and


                                      F-74
-------
meets the hydraulic conductivity criterion. The debris may adversely impact the
hydraulic conductivity of the CCL and/or damage an overlying GM. (L-11)
CCLs should not be left unprotected for an extended period of time. They can
desiccate and crack due to evaporation of water in the CCL, degrade when exposed
to freezing and thawing actions, and be eroded by wind and water. (L-11)
Prior to deploying a GM, all extraneous objects (e.g., tools, sand bags) should be
removed from the surface on which the GM is to be  placed to avoid GM damage
and, for composite liners, promote  good contact between the GM and underlying
CCLorGCL.  (L-15)
HOPE GMs should be installed so that they are essentially stress-free at their lowest
expected temperatures.  GMs contract with decreasing temperature. If there is not
sufficient slack in the GMs at higher temperatures they may yield and rupture at
lower temperatures.  (S-1,S-3)
GMs should be covered with thermal insulation layers at very low temperatures (e.g.,
-20°C for HOPE GMs or 0°C for PVC GMs) since  GM strain at break decreases with
decreasing temperature. The effect of temperature on GM strain at break depends
on the type of GM polymer.  (C-13, S-1)
The leading edge of an uncovered  GM should be secured to prevent wind from
flowing beneath the GM and uplifting it.  This  is typically accomplished by seaming
adjacent panels of GM shortly after deployment and placing a row of adjacent
sandbags along the edge of the GM. (L-19)
If sand bags are used to secure GM panels until the panels are seamed, the installer
should ensure that the sandbags, and all other extraneous objects, are not trapped
beneath the GM after seaming to avoid GM damage and, for composite liners,
promote good contact between the GM and underlying CCL or GCL. (L-15)
For HOPE GMs, fusion seams are  preferred over extrusion seams because fusion
seams have higher seam integrity and lower stress concentrations at seams.
Extrusion fillet seams are preferred over extrusion flat seams because fillet seams
have lower stress concentrations at seams.  Extrusion seams should be minimized
in the field by using prefabricated pipe boots,  careful GM installation, etc.  (S-1)
HOPE GM must be thoroughly cleaned  along a seam path before  the seam is
constructed since dirt in the seam adversely impacts seam integrity.  To minimize
the potential for dirt to accumulate along a seam path,  GM should be seamed shortly
after deployment. A temporary plastic film may also be placed on the GM edges at
the factory and removed from the GM just prior to seaming.  (C-16)
In general, holes in HOPE GM seams should  not be repaired by reseaming.  This
reheating of seams can embrittle the HOPE at the repair and make it more
susceptible to stress cracking. (S-1)
To the extent practicable, holes in GM liners installed over GCLs should be repaired
as soon as possible to avoid swelling of the GCL in case of hydration. GCL swelling
results in a decrease in GCL shear strength and may impact slope stability. Holes
located in areas where rainwater may pond should be  patched first. The patches
should be sealed with a permanent seam and not only tack welded. (L-20)
                                  F-75
-------
•  When a GM is placed over a GCL, the GM should be covered with soils as soon as
   possible to minimize swelling of the GCL in case of hydration. GCL swelling results
   in a decrease in GCL shear strength and may impact slope stability.  (L-20)
•  Connections between GMs and ancillary structures should be carefully constructed
   and inspected to decrease the potential for construction-related GM holes.  (L-9,
   L-11)
•  To decrease the potential for construction-related GM holes in sumps, the GM panel
   layout should be configured to minimize seams in sumps or prefabricated sumps
   should be used.  (L-5)
•  With respect to the potential for leakage, pipe penetrations are generally the most
   critical  locations in landfills without internal sumps. If pipe penetrations are used,
   they should be carefully constructed and inspected.
•  Sumps and pipe penetrations of liners should be leak tested by ponding tests, leak
   location surveys, gas tracer tests, or pressure tests of double pipe boots as part of
   liner system CQA.  Leak testing of the liner system on the facility base (where
   leachate heads are the highest) or of the entire liner system may also be considered.
   Identified holes should be repaired. (L-1,  L-3, L-6, L-9, L-11)
•  The entire installed GM should be inspected for damage and any damage should be
   repaired prior to placement of overlying materials.  (L-3)
•  GM should be covered with  a soil layer as soon as practicable after installation, but
   not during the hottest part of the day if the GM is significantly wrinkled, to reduce GM
   wrinkles, prevent GM uplift by wind, and protect the GM from damage. (L-19,
   L-29, L-43, S-3)
•  Prior to placing soil over  a GM, the GM should be inspected for wrinkles. Excessive
   GM wrinkles and wrinkles that may fold over  should be removed by waiting to
   backfill until the GM cools and contracts during the cooler nighttime and early
   morning hours, pulling the wrinkles out, or cutting the wrinkles out. The latter
   method is less desirable  than the former methods because  it requires intact GM to
   be cut, and it results in more GM seaming and testing.  (L-19, L-29, S-3)
•  On long side slopes, it may be preferable  to use textured GM rather than smooth
   GM to decrease the size of GM wrinkles that  develop, especially near the slope toe.
   Interestingly, Giroud (1994b) has shown analytically that GM wrinkles are shorter
   and spaced closer together when the shear strength between the GM and the
   underlying material is increased. Therefore,  based on analysis, the use of textured,
   rather than smooth, GM decreases the risk that large wrinkles will form.   (L-29)
•  Composite liners and barriers constructed with a CCL should be covered with an
   insulation layer as soon as practicable to prevent CCL desiccation related to heating
   or freeze-thaw action.  (L-2, L-43)

Operation

•  Landfill operations manuals should include limitations on the types of equipment that
   may traffic over the liner  system before the first lift of waste is placed to  prevent liner
                                      F-76
-------
   system damage. These limitations are enforced during construction when CQA is
   implemented; they should also be applied during operation.  (L-7)
•  Landfill operations personnel should be aware of sensitive areas of a liner system,
   such as at pipe penetrations or sumps, and should protect these areas from
   damage. Sensitive areas can be identified with cones, flags, or other markers. They
   can also be isolated from traffic by berms, bollards, or other means. (L-7)
•  Landfills should be operated to minimize the potential for waste fires. Measures to
   be taken could include not depositing loads of hot  waste in a landfill and covering
   waste with a soil cover to decrease waste access  to oxygen. (L-4)
•  Care should be taken to not damage the liner system components when drilling into
   landfilled waste. Settlement of the waste surface must be taken into account when
   selecting the depth of drilling, and boreholes should not extend close (e.g., within 1
   m) to the liner.  Also, the limits of waste containment systems should be identified
   with markers or other means to reduce the potential for liner system or cover system
   damage by drilling or other invasive activities. (L-12,  L-44)

F-6.4     Drainage Systems

The following recommendations are made to reduce the frequency of drainage system-
related problems.

Design

•  Adjacent materials conveying water should be designed to decrease the clogging
   potential of the downgradient material using filter criteria calculations and/or
   laboratory testing. (C-7, C-10, C-11, C-19, C-21)
•  If gap-graded soils are used as drainage materials, the effect of particle migration
   should be evaluated during design using filter criteria  calculations and/or laboratory
   testing.  In fact, the effect of particle migration from all granular drainage materials
   should be evaluated during design, at least qualitatively, since all coarse granular
   drainage materials have some fines.  (C-7)
•  Perforated pipes bedded in gravel should not be wrapped with a GT because the GT
   is useless, and, in some cases, even detrimental because the GT in this location is
   prone to clogging. This has been discussed by Koerner et al. (1993) and
   demonstrated by several case histories presented in their paper.  Instead, the design
   should include a GT between the gravel and the surrounding soil or, possibly,  no
   GT. (L-22,  C-20)
•  Geosynthetic anchor trenches should be backfilled with low-permeability soil and the
   soil should be well compacted to reduce the potential for water for infiltrate into the
   trenches and flow into LCRSs or LDSs. If this is not practicable, the anchor trenches
   should be designed  to drain freely and/or covered  with a barrier, such as a GM. In
   addition, the ground surface should be graded away from the trenches to reduce
   runon from  infiltrating into the trenches. (L-10, L-16)
                                      F-77
-------
•  Project specifications for needlepunched nonwoven GTs should require that the GTs
   be needle-free and should require a certification from the manufacturer attesting to
   this.  Needles, if present, may damage a nearby GM.  (L-28)
•  The CQA Plan should require that deployed GTs near GMs be inspected for needles
   before the GTs are covered with overlying materials. If needles are found, the GT
   should be rejected. (L-28)
•  If a GT is to be exposed to the environment for an extended time period after
   installation, the potential for degradation of the GT should be evaluated under all the
   anticipated environmental conditions. EPA recommends that the effect of ultraviolet
   light on GT properties be evaluated using ASTM D 4355 (Daniel and Koerner, 1993).
   The test is typically run for 500 hours; however, it can be run for longer time periods
   to meet project-specific conditions. In any case, prior to covering the GT, the
   condition of samples of the exposed GT taken from the field should be evaluated by
   laboratory testing to verify that the exposed GT is  still satisfactory.  (L-13, L-18)
•  If test results indicate that the GT will not have the required properties (typically a
   specified strength retention) after exposure, the GT should be protected with a
   sacrificial opaque waterproof plastic tarp, soil layer, or other means. Tisinger et al.
   (1993) suggest that this may be the best strategy since a heavy degraded GT that
   meets the  specifications is more sensitive to stress concentrations than a new lighter
   GT that meets the same specifications.  (L-13, L-18)
•  When the waste in a containment system contains some fine particles that may
   migrate to the LCRS, the potential for LCRS clogging may  be reduced by allowing
   those fine particles to pass though the LCRS to the leachate collection pipes, which
   can subsequently be cleaned.  The fine particles will pass more easily through the
   LCRS if no GTs are used in the LCRS or if the LCRS contains relatively thin open
   nonwoven GTs rather than thicker nonwoven GTs with a smaller apparent opening
   size.  Note that the above does not apply to an LCRS with  only a GN drainage layer.
   Though a GN drainage layer has a high transmissivity, it is thin and is, therefore,
   generally more susceptible to clogging by sedimentation than a granular drainage
   layer. (L-36)

Construction

•  The drainage system should be kept free of debris that may potentially impede the
   flow of liquid.  In general, all sandbags should be removed  from the drainage
   system. However, if the sand in the bags meets the project specifications for the
   overlying drainage layer material, the bags can be cut and  removed and the sand
   left in place.  (L-15)
•  GTs and GCs should be covered as soon as possible after installation to protect
   them from the environment (e.g., ultraviolet light, water, high temperature, animals).
   (L-13, L-18)
•  The CQA consultant should verify that all connections required for  adjacent drainage
   system pipes have been  made. When pipe is connected by butt fusion seaming, the
   seam should be inspected for defects. (L-32, L-33)


                                     F-78
-------
•  Care should be taken to not damage drainage system pipes during construction.
   The contractor should maintain sufficient soil cover between construction equipment
   and the pipes during construction. Equipment operators should be aware of pipe
   locations, since pipes can be crushed by trafficking equipment. Also, soil around
   pipes should be compacted using hand operated or walk-behind compaction
   equipment. (L-30)
•  After construction of a cell with an external sump, the pipe from the cell to the sump
   should be inspected to verify that the pipe is functioning as designed.  The
   inspection may be performed by  surveying the pipe with a video camera, pulling a
   mandrel through the pipe, flushing the  pipe with water, or other means.  (L-30)

Operation

•  Leachate may seep from landfill side slopes if the leachate can perch on layers of
   less permeable materials (e.g., daily and intermediate cover materials) within the
   waste or drain from layers of more permeable materials (e.g., tires) within the waste
   that are located relatively close to the side slopes. The potential for seepage can be
   decreased by: (i) not placing layers of the more permeable materials near the side
   slopes; (ii) sloping layers of the less and more permeable materials away from the
   side slopes; (iii) distributing the more permeable materials throughout the waste; (iv)
   constructing leachate chimney drains to the LCRS around these layers;  (v) removing
   perched leachate from wells installed to these layers; and (vi) using alternate daily
   covers (e.g., foams, tarps) that do not results in layers of less permeable materials in
   the waste. (L-37)
•  Drainage system pipes should be maintained by cleaning the pipes at least annually
   and more frequently, if warranted. (L-12, L-36)
•  Landfills with external sumps may also include riser pipes at the low point of LCRSs
   as a precautionary measure to allow for leachate removal from the landfill,  if
   necessary. (L-12)
•  Leachate flow measurement systems should be calibrated and adjusted as needed
   at least annually to ensure that the quantities measured are accurate.
   (L-5, L-34)
•  Due to the potential for problems in automated leachate metering and pumping
   equipment, landfill operations plans should include a verification and contingency
   method for estimating the quantities of liquid removed from the LCRS and LDS.
   (L-5, L-34)
•  Leachate sump pumps should be self priming so the pumps will not become
   airlocked and shut down if air is pulled into the pumps. (L-34)
•  Leachate sump pumps should be selected to be compatible with sump geometries
   and anticipated leachate recharge rates so pump cycles are appropriate (e.g., not so
   short that the  pumps turn on and off too frequently and burn out prematurely). (L-35)
•  The "pump on" levels in internal sumps should be kept as low as practicable to
   reduce leakage if there are holes in the GM liner in the sump, especially if the GM is
   not underlain  by a GCL.  It is recognized, however, that "pump on" liquid levels in


                                      F-79
-------
   internal sumps may need to be larger than 0.3 m to achieve efficient sump pump
   operation. (L-5)
•  The potential for clogging of water-level indicators, pumps, and flowmeters must be
   considered when selecting the types of equipment to use at a facility. For example,
   the impeller and filter screen in the mechanical flowmeters initially used at landfill
   L-5 frequently became clogged. After the mechanical flowmeters were replaced with
   venturi flowmeters, clogging was not a problem.  (L-5, L-35)
•  Outlets of cover system drainage layers should be kept free of snow and ice so that
   these layers can drain freely.  (L-42)

F-6.5     Surface and Protection Layers

The following recommendations are made to reduce the frequency of surface layer and
protection layer-related problems.
Design

•  Erosion of soil protection layers on liner system side slopes should be anticipated
   and dealt with in design. The potential for erosion can be reduced by grading the
   liner system to avoid concentrated runoff and using a relatively permeable soil in the
   protection layer. In areas where the potential for erosion is relatively high, erosion
   control structures (e.g., runoff diversion berms, silt fence) can be used to reduce the
   need for intensive maintenance of soil protection layers. Protection layers can also
   be covered with a tarp or temporary erosion control mat.  (L-9, L-11)
•  When a liner system is constructed on top of an existing landfill (vertical expansion),
   an exposed GM liner can be uplifted by gases from the underlying landfill.
   Therefore, in the case of a vertical expansion, unless gases from the underlying
   landfill are well controlled, GMs must be covered by a soil layer to prevent GM uplift
   and precautions must be taken to prevent erosion of this soil.  (L-9, L-11)
•  Better methods for protecting exposed  soil layers on liner system side slopes from
   erosion or alternatives to these soil layers (e.g., sand filled mats,  Styrofoam sheets)
   are needed. A new geosynthetic that will insulate the underlying  liner materials
   could be developed for this purpose. (L-9)
•  Post-construction plans should be developed for portions of landfills that may sit idle
   for an extended period  of time. The plans should include procedures describing how
   the liner system should be maintained prior to operation.  (L-9)
•  For liner systems where soil protection layers are placed incrementally during landfill
   operation, a geosynthetic cushion (supercushion) better than the usual thick
   nonwoven GT needs to be developed to protect the liner system during soil
   placement.
•  Erosion of surface layers on cover system side slopes should be  anticipated and
   dealt with in design. In areas where the potential for erosion is relatively high,
   erosion control measures (e.g., runoff diversion berms, silt fence, turf reinforcement


                                      F-80
-------
   and revegetation mat) can be specified to reduce the need for intensive
   maintenance of soil layers.  However, the erosion control measures require
   maintenance (e.g., periodic  removal of soil retained by silt fence). (C-1, C-12)
•  The length of cover system  slopes between ditches or swales where runoff is
   collected should be selected to limit erosion to acceptable amounts  (e.g.,
   5 tonnes/ha/yr). At a minimum, the potential for erosion should be evaluated using
   the universal soil loss equation.  As described by EPA (1994), cover system slopes
   may need to be 4H: 1V or less and intercepted by swales at 6-m vertical intervals to
   meet acceptable erosion levels.  (C-12)
•  Design flow velocities in drainage channels should be calculated so the appropriate
   channel lining can be selected. (C-1)

Construction

•  Though it may be less costly for the owner to construct several landfill cells at once,
   this can leave new cells exposed to the environment for a significant time period.
   These cells will experience more erosion than cells filled sooner and will have more
   opportunity for liner damage. Additionally, every time an eroded soil layer is pushed
   back up the side slopes there is an opportunity for the underlying liner system
   materials to be damaged by construction equipment.  (L-9)

F-6.6     Liner System and Cover System Stability

Design

•  The stability of liner system  and cover system slopes should always be evaluated
   using  rigorous  slope stability analysis methods that consider actual shear strengths
   of materials, anticipated seepage pressures, and anticipated  loadings.  (L-21, L-24,
   L-25,  L-38 to L-42, L-46, C-3 to C-11, C-17 to C-20, S-2)
•  The majority of the slides described herein occurred along geosynthetic/
   geosynthetic interfaces. For a number of these cases, the interface shear strengths
   were estimated on the basis of published tested data. This approach should be
   avoided because there may be significant differences in interface shear strengths
   between similar materials from different manufacturers and even identical  materials
   in different production lots from the same manufacturer.  Only a small error in the
   estimated interface shear strength may cause slope instability.  Because of this,
   geosynthetic interface shear strengths should not be  estimated, they should be
   measured. Additionally, as  more geosynthetics are available on the market,  the
   probability increases that there will be significant differences in  properties between
   geosynthetics that appear to be similar.  (L-21, L-24,  L-25, L-38 to L-42, L-46, C-4 to
   C-11,  C-17)
•  Interface shear strength test conditions (moisture, stresses, displacement  rate, and
   displacement magnitude) should be representative of field conditions. (L-25)
                                      F-81
-------
The effects of variation in the tested geosynthetics, accuracy of test methods, and
test conditions must be considered when selecting the interface shear strength to
use in design.  (L-25,  C-9)
Freeze-thaw of CCLs can have a significantly detrimental impact on GM/CCL
interface shear strength and should be considered when selecting the interface
shear strength to use in slope stability analyses. However, freeze-thaw effects on
interface strength should not actually be a design consideration, since CCLs should
be protected from freezing in the first place. (C-13)
The effect of construction on moisture conditions at the GM/CCL interface should be
considered when developing the specification for CCL construction and selecting the
strength of liner system  interfaces for slope stability analyses. The CCL construction
specification should generally include limitations on maximum compacted moisture
content, restrictions on applying supplemental moisture, and requirements for
covering the CCL and overlying GM as soon as practical to minimize moisture
migration to the GM/CCL interface.  If a CCL on a slope becomes desiccated, it
should be reworked and not just moistened. (L-21, L-25,
C-13)
Cover systems incorporating a low-permeability barrier layer should include a
drainage layer above the barrier when the cover system side slopes are steeper
than 5H:1V (EPA, 1994). The purpose of this drainage layer is to prevent the
buildup of seepage pressures in the cover system soil layer(s) overlying the barrier
layer. (C-4)
When liner systems or cover systems are constructed over wastes, the potential for
the wastes to generate gases that uplifts the liners or barriers must be considered.
The gas pressures decrease the shear strength along the bottom  interface of the
uplifted layer and may lead to slope instability.  Gas collection systems, therefore,
may be required to prevent the buildup of gas pressures.  (L-9, L-11)
Cover system drainage layers should be designed to handle the total anticipated
flow to the drainage layer calculated using a water balance or other appropriate
analysis (e.g., Giroud and Houlihan, 1995). Soong and Koerner (1997) recommend
using a short-duration intensive storm in the water balance and do not recommend
the EPA Hydrologic Evaluation of Landfill Performance (HELP) computer model for
this purpose.  The drainage layer flow rates output from the HELP model are an
average for a 24-hour period and may be much less than the peak flow rates
calculated using other methods if the precipitation data used in the HELP model are
not carefully selected. (C-4, C-5, C-12,  C-17 to C-19)
Water collected in the drainage layer must  be allowed to outlet to  prevent the buildup
of seepage pressures. (L-42)
Containment systems should be designed to limit seismic displacements to tolerable
amounts.  To do this,  designs may incorporate predetermined slip surfaces to
confine movements to locations where they will cause the least damage (i.e., above
the GM liner) and inverted liner system keyways to provide more resistance to
movement. For example, a GM with a smooth top surface and a textured bottom
                                   F-82
-------
   surface could be used in certain liner systems to create a predetermined slip surface
   above the GM.  (L-26, L-27)
   Liner system anchor trenches should be designed to secure geosynthetics during
   construction, but release the geosynthetics before they are damaged during
   earthquakes.  An alternative is to unanchor the liner system after construction and
   secure it on a bench with an overlying soil layer.  (L-26)
   Stress  concentrations at or near the liner system  side slope crest should be avoided.
   Areas with stress concentrations are more problematic when subjected to seismic
   loading. In particular, GM seams should generally not be sampled near the slope
   crest. (L-26)
Construction

•  Soils should be placed over geosynthetics from the toe of slope upward to avoid
   tensioning the geosynthetics.  Methods of soil placement that are not toe to top
   should be pre-approved by the engineer who analyzed the stability.  (C-3, C-14)
•  Geosynthetic reinforcement should be anchored prior to placing the soil layer to be
   reinforced.  (C-14)

Operation

•  Outlets of drainage layers should be kept free of snow and ice so these layers can
   drain freely and prevent the buildup of seepage pressures.  (L-42)
•  Soils or waste should be placed over geosynthetics from the toe of slope upward to
   avoid tensioning the geosynthetics.  Methods of placement that are not toe to top
   should be pre-approved by the engineer who analyzed the stability.  (S-2)
•  Surficial cracking of soil cover layers during seismic loading, especially near
   locations with contrast in seismic response characteristics (e.g., top of waste by rock
   canyon walls), should be anticipated and dealt with as an operation issue through
   post-earthquake inspection and repair.  (L-26, L-27, C-21 to C-23)
•  Proposed changes to the landfill filling sequence should be reviewed by the design
   engineer to ensure that these changes will not adversely impact slope stability.
   (L-24)
•  Soil layers anchoring geosynthetics should be maintained during landfill construction
   and operation.  (L-45)

F-6.7     Liner System and Cover System Displacements

The following recommendations are made to reduce the frequency of liner system and
cover system displacement-related problems.
                                      F-83
-------
Design

•  When liner systems or cover systems are constructed over existing wastes, the
   potential for the wastes to generate gases must be considered. The gases may
   uplift GMs, causing excessive stresses in the GMs and may impact slope stability.
   Some landfills may be generating little or no gas at the time of construction and may
   not need a gas collection system. Other landfills may be generating significant
   quantities  of gas and may require a gas collection system beneath the entire liner
   system. (L-9, L-11)
•  Surface-water runoff should be managed to reduce foundation uplift problems  during
   and after construction. Temporary and permanent surface-water diversion
   structures  located near a cell may need to be lined to reduce infiltration, especially if
   the structures are located on relatively permeable soils and convey relatively large
   amounts of water. Runoff should not be allowed to pond near the cell, where it can
   infiltrate into the cell.  (L-25, L-31)
•  Liner systems and cover systems constructed over compressible,  low shear strength
   waste materials should be designed to accommodate the anticipated settlements.
   When GCL is used, seam overlaps should be wider than normal.  (C-15)
•  Gas extraction well boots should be designed to accommodate the anticipated
   landfill settlements. (C-12)

Construction

•  Cover systems with soil layers placed over compressible, low shear strength waste
   should use lightweight construction equipment and have good control of the
   thickness of soil placed over the waste so as not to cause bearing capacity failure of
   the waste  and excessive displacement of the cover system. (C-15)
F-7       References

Adams, FT., Overmann, L.K., and Cotton, R.L. (1997), "Evaluation and Remediation of
    a Fire Damaged Geosynthetic Liner System", Proceedings of Geosynthetics '97,
    Long Beach, California, Vol. 1, pp. 379-392.
Anderson, K.A., "Leachate Collection Management at a Hazardous Waste Landfill",
    Hydrological Science and Technology, Vol. 9, No. 1-4, 1993, pp. 30-53.
Anderson, R.L., "Earthquake Related Damage and  Landfill Performance", Earthquake
    Design and Performance of Solid Waste Landfills, Yegian, M.K. and Liam Finn,
    W.D., eds., ASCE Geotechnical  Special Publication No.  54, 1995, pp. 1-16.
Augello, A.J., Matasovic, N., Bray, J.D., Kavazanjian, E., Seed, R.B.,  "Evaluation of
    Solid Waste Landfill Performance During the Northridge  Earthquake", Earthquake
    Design and Performance of Solid Waste Landfills, Yegian, M.K. and Liam Finn,
    W.D., eds., ASCE Geotechnical  Special Publication No.  54, 1995, pp. 17-50.
                                     F-84
-------
Badu-Tweneboah, K., Williams, N.D., and Haubeil, D.W., "Assessment of a PVC
    Geomembrane Used in a Landfill Cover System", Proceeding of Fifth International
    Conference on Geotextiles, Geomembranes and Related Products, Singapore,
    1994, pp. 1029-1032.
Basnett, C.R. and  Bruner, R.J., "Clay Desiccation of a Single-Composite Liner System",
    Proceedings of Geosynthetics '93, Vancouver, British Columbia, 1993, Vol. 3, pp.
    1329-1340.
Bass, J.M., "Avoiding Failure ofLeachate Collection and Cap Drainage Layers",
    EPA/600/2-86/058, EPA Hazardous Waste Engineering Research Laboratory,
    Cincinnati, Ohio, 1986, 142 p.
Bonaparte, R., "Long-Term Performance of Landfills'" Proceedings of the ASCE
    Specialty Conference Geoenvironment 2000, ASCE Geotechnical Special
    Publication No. 46, 1995, Vol. 1, pp. 415-553.
Bonaparte, R. and  Gross, B.A., "LDCRS Flow Rates from Double-Lined Landfills and
    Surface Impoundments", EPA/600/SR-93/070, EPA Risk Reduction Research
    Laboratory, Cincinnati, Ohio, 1993, 65 p.
Bonaparte, R., Othman, M.A., Rad, N.R., Swan, R.H., and Vander Linde, D.L.,  "Evaluation
    of Various Aspects of GCL Performance", Appendix F in Report of 1995 Workshop on
    Geosynthetic Clay Liners, Daniel, D.E. and Scranton, H.B., Editors,  EPA National
    Risk Management Resource Laboratory, Cincinnati, Ohio, EPA/600/R-96/149, 1996,
    pp. F-1-F-34.
Boschuk, J. Jr., "Landfill Covers An Engineering Perspective", Geotechnical Fabrics
    Report, IFAI, Mar 1991, pp. 23-34.
Byrne, R.J., Kendall, J., and Brown, S., "Cause and Mechanism of Failure Kettleman
    Hills Landfill B-19, Phase 1A", Stability and Performance of Slopes and
    Embankments-II, ASCE Geotechnical Special Publication No. 31, 1992, pp. 1188-
    1215.
Calabria, C.R. and Peggs, I.D., "Investigation of Geomembrane Seam Failures: Landfill
    Cover", Proceedings of the 10th GRI Conference Field Performance of
    Geosynthetics and Geosynthetic Related Systems, Geosynthetic Research
    Institute, Philadelphia, Pennsylvania, 1996, pp. 234-257.
Chang, S., Bray, J.D., and Seed, R.B., "Engineering Implications of Ground Motions
    from Northridge Earthquake", Bulletin of the Seismological Society of America, Vol.
    86, No.  1, Part B Supplement,  1996, pp. S270-S288.
Daniel, D.E. and Koerner, R.M., "Technical Guidance Document: Quality Assurance and
    Quality Control for Waste Containment Facilities",  EPA/600/R-93/182, EPA Risk
    Reduction Research Laboratory, Cincinnati, OH, 1993, 305 p.
Darilek, G.T., Menzel, R., and Johnson, A., "Minimizing Geomembrane Liner  Damage
    While Emplacing Protective Soil", Proceedings of Geosynthetics '95, Nashville,
    Tennessee, 1995, Vol. 2, pp. 669-676.
Dvirnoff, A.H. and  Munion, D.W., "Stability Failure of a Sanitary Landfill", International
    Symposium on Environmental Geotechnology, Lehigh, Pennsylvania, 1986, pp. 25-
    35.
                                     F-85
-------
Environmental Protection Agency, "Technical Guidance Document: Final Covers on
    Hazardous Waste Landfills and Surface Impoundments", EPA/530/SW-89/047, EPA
    Office of Solid Waste and Emergency Response, Washington, D.C., 1989, 39 p.
Environmental Protection Agency, "Seminar Publication: Design, Operation, and Closure
    of Municipal Solid Waste Landfills", EPA/625/R-94/008, EPA Center for
    Environmental Research Information, Cincinnati, Ohio, 1994, 86 p.
Environmental Protection Agency, "List of Municipal Solid Waste Landfills", EPA/530/R-
    96/006, EPA Office of Solid Waste and Emergency Response, Washington, D.C.,
    1996, pp. 1-2.
Environmental Protection Agency, Resource Conservation and Recovery Act Information
    System (RCRAIS) database,  EPA National Oversite Database,  Jun 1997.
Ghassemi, M.,  Crawford, K., and  Haro, M., "Leachate Collection and Gas Migration and
    Emission Problems at Landfills and Surface Impoundments", EPA/600/2-86/017,
    EPA Hazardous Waste Engineering Research Laboratory, Cincinnati, Ohio, 1986,
    206 p.
Giroud, J.P., "Lessons Learned From Studying the Performance of Geosynthetics",
    Proceeding of Geotextiles-Geomembranes Rencontres 93, Vol. 1, Joue-les-Tours,
    France, Sep 1993, pp. 15-31.
Giroud, J.P., "The Mechanics of Geomembrane Stress Cracking", Computer Methods
    and Advances in Geomechanics, Siriwardane, H.J. and Zaman, M.M. (editors),
    Balkema, Proceedings of the  Eighth International Conference on Computer
    Methods and Advances in Geomechanics, Morgantown, West Virginia, 1994a, Vol.
    1, pp. 177-188.
Giroud, J.P., "Quantification of Geosynthetics  Behavior", Proceeding of the Fifth
    International Conference on Geotextiles, Geomembranes, and Related Products,
    Singapore, 1994b, Vol. 4,  pp. 1249-1273.
Giroud, J.P., "Granular Filters and Geotextile Filters", Proceedings ofGeofilters '96,
    Lafleur, J.,  and  Rollin, A.L., Editors, Montreal, 1996, pp. 565-690.
Giroud, J.P. and Houlihan, M.F., "Design of Leachate Collection Layers", Proceedings
    of Fifth International Landfill Symposium, Sardinia, Italy, Vol. 2,  1995, pp. 613-640.
Giroud, J.P., Bachus,  R.C.,  and Bonaparte, R., "Influence of Water Flow on the Stability
    of Geosynthetic-Soil Layered  Systems on Slopes", Geosynthetics International,
    IFAI, Vol. 2, No. 6, 1995,  pp.  1149-1180.
Gross,  B.A.,  Bonaparte,  R.B.,  and Giroud, J.P., "Evaluation of Flow from Landfill
    Leakage Detection Layers", Proceedings of Fourth International Conference on
    Geotextiles, Vol. 2, The Hague, The Netherlands,  1990, pp. 481-486.
Harris,  J.M.,  Rivette, C.A., and Spradley, G.V., "Case Histories of Landfill Erosion
    Protection Using Geosynthetics", Geotextiles and Geomembranes, No. 11, 1992,
    pp. 573-585.
Hullings, D.E. and Sansone, L.J.,  "Design Concerns and Performance of Geomembrane
    Anchor Trenches", Proceedings of the 10th GRI Conference Field Performance of
    Geosynthetics and Geosynthetic Related Systems, Geosynthetic Research
    Institute, Philadelphia, Pennsylvania, 1996, pp. 219-233.
Kenter, R.J., Schmucker, B.O., and Miller, K.R., "The Day the Earth Didn't Stand Still:
    The Rumpke Landfill", Waste Age, Mar 1997, pp. 66-81.
                                     F-86
-------
Kmet, P., Mitchell, G., and Gordon, M., "Leachate Collection System Design and
    Performance - Wisconsin's Experience", Proceedings,  of ASTSWMO National Solid
    Waste Forum on Integrated Municipal Waste Management, Lake Buena Vista,
    Florida, 1988.
Koerner, G.R., Koerner, R.M., and Martin, J.P., "Field Performance of Leachate
    Collections Systems and Design Implications", Proceedings of 31st Annual SWANA
    Conference, San Jose, California, 1993, pp. 365-380.
Koerner, G.R., Eith, A.W., and Tanese, M., "Properties of Exhumed HOPE Field Waves
    and Selected Aspects of Wave Management", Proceedings of the 11th GRI
    Conference on Field Installation of Geosynthetics, Geosynthetic Research Institute,
    Philadelphia, Pennsylvania, 1998, pp. 152-162e.
Laine, D.L. and Darilek, G.T., "Locating Leaks in Geomembrane Liners of Landfills
    Covered with a Protective Soil", Proceedings of Geosynthetics '93, Vancouver,
    British Columbia, 1993, pp. 1403-1412.
Loewenstein, D.L., and Smrtic, M.J., "Primary Liner System Repair  Made During
    Operation of the City of Albany Landfill: A Case History", Modern Double Lined
    Landfill Management Seminar, An Operation and Maintenance  Perspective,
    sponsored by New York State Association for Solid Waste Management in
    cooperation with New York State Department of Environmental  Conservation Division
    of Solid Waste, Saratoga Springs, New York, Jan 1994, 6 p.
Matasovic, N. and Kavazanjian, E., Jr., "Observations of the Performance of Solid
    Waste Landfills During Earthquakes", Eleventh World Conference on Earthquake
    Engineering, Acapulco, Mexico, Elsevier Science Ltd., 1996, CD-ROM Paper No.
    341.
Matasovic, N., Kavazanjian, E., Jr., Augello, A.J., Bray, J.D., and Seed, R.B., "Solid
    Waste Landfill Damage Caused by 17 January 1994 Northridge Earthquake", The
    Northridge, California, Earthquake of 17 January 1994, Woods, M.C. and Seiple,
    W.R., eds., California Department of Conservation, Division of Mines and Geology
    Special Publication 116, 1995, pp. 221-229.
Matasovic, N., Kavazanjian, E., Jr., and Anderson, R.L., "Performance of Solid Waste
    Landfills  in Earthquakes", Earthquake Spectra, Vol.  13, No. 5,  May 1998, pp. 319-
    334.
Mitchell, J.K., Seed, R.B., and Seed, H.B., "Kettleman Hills Waste Landfill Slope Failure.
    I: Liner-System Properties", Journal of Geotechnical Engineering, Vol. 116, No. 4,
    Apr 1990, pp. 647-668.
Oweis, I.S., "Stability of Sanitary Landfills", Geotechnical Aspects of Waste Management,
    seminar sponsored by Metropolitan Section, ASCE, New York,  1985.
Paulson, J.N., "Veneer Stability Case Histories: Design Interactions Between
    Manufacturer/Consultant/Owner", Proceedings of the 7th GRI  Seminar
    Geosynthetics Liner Systems: Innovations, Concerns, and Designs, Geosynthetic
    Research Institute, Philadelphia, Pennsylvania, 1993, pp. 235-241.
Peggs, I.D., Winfree, J.P., Giroud, J.P., "A Shattered  Geomembrane Liner Case History:
    Investigation and Remediation", Proceedings of Geosynthetics '91, Atlanta,
    Georgia, 1991, Vol. 2, pp. 495-505.
                                     F-87
-------
Richardson, G. and Reynolds, D., "Geosynthetic Considerations in a Landfill on
    Compressible Clays", Proceedings of Geosynthetics '91, Atlanta, Georgia, 1991,
    Vol.2, pp. 507-516.
Schmucker, B.O. and Hendron, D.M., "Forensic Analysis of 9 March 1996 Landslide at
    the Rumpke Sanitary Landfill, Hamilton County, Ohio", Slope Stability in Waste
    Systems, seminar sponsored by ASCE Cincinnati and Toledo Sections, Ohio, 1997.
Seed, R.B., Mitchell, J.K., and Seed, H.B., "Kettleman Hills Waste Landfill Slope Failure.
    II: Stability Analyses", Journal of Geotechnical Engineering, Vol. 116, No. 4, Apr
    1990, pp. 669-690.
Silva, M., "Some  Landfills Starting to Leak, State Study Says", Miami Herald, 18 Jun
    1995, p. 6B-7B.
Soong, T.Y. and Koerner, R.M., "The Design of Drainage Systems Over
    Geosynthetically Lined Slopes", GRI Report #19, Geosynthetic Research Institute,
    Philadelphia, Pennsylvania, 1997,  88 p.
Stark, T.D.  and Evans, W.D., "Balancing Act",  Civil Engineering, Aug 1997,  pp. 8A-11A.
Stewart, J.P., Bray, J.D., Seed, R.B., and Sitar, N., "Preliminary Report on the Principal
    Geotechnical Aspects of the January 17, 1994 Northridge Earthquake",  Report No.
    UCB/EERC-94/08, College of Engineering, University of California at Berkeley,
    Berkeley, California, 1994, 238 p.
Tedder, R.B., "Evaluating the Performance of Florida Double-Lined  Landfills",
    Proceedings of Geosynthetics '97, Long Beach, California,  1997, Vol. 1, pp. 425-
    438.
Tisinger, L.G., Clark, B.S., Giroud, J.P., and Christopher,  B.R.,  "Analysis of  an Exposed
    Polypropylene Geotextile", Proceedings of Geosynthetics '93, Vancouver, British
    Columbia, 1993, pp. 757-771.
Tisinger, L.G., Clark, B.S., Giroud, J.P., and Schauer, D.A., "Performance of Nonwoven
    Geotextiles Exposed to a Semi-Tropical  Environment", Proceedings of Fifth
    International Conference on Geotextiles, Geomembranes, and Related  Products,
    Singapore, 1994, pp. 1223-1226.
Vander Linde, D.L., Luettich, S.M., and Bonaparte, R., "Lessons Learned From the
    Failure of a Landfill  Cover System", Geosynthetics: Lessons Learned From Failures,
    IFAI, to be published in 1998.
                                      F-88
-------
                                Attachment F-A
             Case Histories of Waste Containment System Problems

F-A.1     Introduction

The purpose of this attachment to Appendix F is to present case histories of the waste
containment system problems described in Chapter F-2 of the appendix. The case
histories are divided into 15 categories based on the waste containment system
component or attribute criterion affected by the problems: (i) landfill liner construction;
(ii) landfill liner degradation; (iii) landfill LCRS or LDS construction; (iv) landfill LCRS or
LDS degradation; (v) landfill LCRS or  LDS malfunction; (vi) landfill LCRS or LDS
operation; (vii) landfill liner system stability; (viii) landfill liner system displacement; (ix)
cover system construction; (x) cover system degradation; (xi) cover system stability; (xii)
cover system displacement; (xiii) impoundment liner construction; (xiv) impoundment
liner degradation; and (xv) impoundment liner system stability. Each of these 15
categories of case histories is presented in a separate section of this attachment (F-A.2
to F-A. 16).  The case histories are also categorized by the principal human factor
contributing to the problem: (i) design; (ii) construction; and (iii) operation. The
classification of each case history is shown as "component or attribute criterion"/
"principal human factor criterion" (e.g., landfill liner system stability/design). The nature
of the problem in each case history is  described.  When information is available, the
method by which the problem was detected and the remedies, if any,  that have been
implemented are also presented.  Lessons learned for future projects are given.

F-A.2     Landfill Liner Construction

F-A.2.1   LJ_

Problem Classification:  landfill liner construction/construction
Region of U. S.: unknown
Waste Type:  unknown
Reference: Laine, D.L. and Darilek,  G.T., "Locating Leaks in Geomembrane Liners of
    Landfills Covered with a Protective Soil", Proceedings of Geosynthetics '93,
    Vancouver, British Columbia,  1993, Vol. 3,  pp. 1403-1412.

Problem Summary:  leakage  through holes in HOPE GM primary liner

Problem Description: Laine and Darilek (1993) described the results of an electrical
leak location survey conducted over the 0.4 ha base of a double-lined landfill cell
exhibiting primary liner leakage. The 1.5-mm thick HOPE GM primary liner for the cell
was overlain by the following components, from top to bottom:

•  0.3-m thick sand protection layer;
•  GT; and
                                      F-89
-------
•  0.3-m thick gravel LCRS drainage layer.

The purpose of the survey was to locate GM primary liner holes so they could be
repaired prior to placing waste in the cell.  The cell had been constructed with third-party
CQA and air pressure and vacuum box testing of GM seams.  However, primary liner
holes were suspected because the flow rate of water from the GN LDS drainage layer
into the LDS sump increased to about 2,900 Iphd after rainwater ponded over the base
of the cell. Laine and Darilek (1993) did not give the height of ponded rainwater. Three
small holes were found in the GM primary liner during the survey of the 0.4 ha landfill
base. Two holes were located in the GM panels, and one was located in a seam.  One
of the panel holes was 2.5-mm in diameter. The sizes of the other two holes and the
type of seam containing one of the holes (i.e., fusion or extrusion) were not given.

The GM primary liner holes found by Laine and Darilek are not unexpected.  In a study
of LDS flow rates from double-lined waste containment systems, Bonaparte and Gross
(1993) found that all landfill cells with  GM primary liners appeared to have  exhibited
primary liner leakage. The LDS flow rate from the cell surveyed by  Laine and Darilek
(1993) was higher than the typical LDS flow rates of less than  200 Iphd reported by
Bonaparte and Gross (1993) for operating landfill cells with GM primary liners
constructed with CQA. The relatively high flow rate from this cell as compared to the
landfill cells in the study by Bonaparte and  Gross (1993) is due to: (i) the relatively high
head of rainwater on the holes as compared to the typical head of leachate over holes
in a landfill; and/or (ii) the greater size and frequency of the holes as compared to that
typical of landfills constructed with CQA. As proposed by Giroud (1984), leakage
through a GM hole can be evaluated using Bernoulli's equation for free flow through  an
orifice.  Considering only the 2.5-mm diameter hole, the authors of this appendix found
that Bernoulli's equation gives a flow rate of 2,900 Iphd through the  hole (considering
the 0.4 ha cell base) when the head of liquid above the hole is 1.0 m.

Resolution: The identified holes were repaired. The LDS flow rate after the repairs was
not given by  Laine and Darilek (1993).

Lessons Learned for Future Projects'.  Based on the available  information,  the following
lessons can be learned from this case history:

•  Construction-related holes in GM liners should be anticipated, even in liners installed
   with CQA. If there is a head of liquid over a liner hole, leakage occurs.
•  GM primary liner holes can be located by leak location surveys  or other methods
   even after the LCRS drainage layer and protection layer have been placed. In the
   case history presented above, the primary liner leakage rate was perceived to be
   relatively high, and the landfill owner decided to decrease the leakage rate by
   identifying and repairing GM holes prior to waste placement.
•  When water is ponded over a GM  primary liner in a landfill, relatively high flow rates
   can occur through small holes in the GM, depending on the head of water over the
                                      F-90
-------
   holes. Therefore, these small holes can be easily detected and repaired prior to
   waste placement if water can be ponded over them.
F-A.2.2   L-3

Problem Classification: landfill liner construction/construction
Region of U. S.:  southcentral
Waste Type: HW

Reference: Darilek, G.T.,  Menzel, R., and Johnson, A., "Minimizing Geomembrane Liner
    Damage While Emplacing Protective Soil", Proceedings of Geosynthetics '95,
    Nashville, Tennessee, 1995, Vol. 2,  pp. 669-676.

Problem Summary: leakage through holes in HOPE GM liners

Problem Description:  Darilek et al. (1995) described the results of electrical leak
location surveys conducted over the base of three double-lined landfill cells located in
southeast Texas. The surveys were performed in addition to normal CQA procedures.
For each  cell, the following four liner system configurations were surveyed: (i) the 2-mm
thick HOPE GM component of the composite secondary liner;  (ii) the GM component of
the composite secondary liner after placement of the overlying 0.3-m thick pea gravel
(maximum particle size about 9 mm) LDS drainage layer; (iii) the 2-mm thick HOPE GM
primary liner; and (iv) the  GM primary liner after placement of the overlying GN LCRS
drainage  layer,  440 g/m2 needlepunched nonwoven GT filter, and 0.3-m thick pea gravel
LCRS drainage layer.  These configurations allow an assessment of GM damage
associated with placing gravel directly on the GM and the degree of protection provided
to the GM by the overlying GN and GT.

With respect to the surveys of the six exposed GMs (three GM primary liners and three
GM components of secondary liners), the frequency of construction-related holes found
during surveys  of the bottom GMs in two cells was 33 and 14/ha.  An extrusion seaming
problem was corrected, and the number of holes  located in the bottom GM of the third
cell and the top GMs for the three cells dropped to 5/ha or less. Darilek et al. (1995)
suggested that the hole frequency might also have decreased due to the GM seaming
equipment operators taking more care to avoid seam holes. Most of the located GM
holes were in extrusion seams; none were in fusion seams.  The four largest holes were
punctures or slits in the GM panels. The sizes and possible causes of these holes were
not discussed by Darilek et al. One hole was found in the prefabricated HOPE sump
structure.

The gravel LCRS and LDS drainage layer material was dumped from the top of the side
slopes of the cells onto a  sacrificial piece of GM placed over the GM liners. A grader
with its blade on a extending boom was used to push the gravel down the slopes until a
temporary access ramp could  be developed in each cell.  Then low ground-pressure
                                     F-91
-------
bulldozers pulled the gravel down the ramps and spread it across the cells. The
thickness of the gravel between the bulldozers and the GMs was at least 0.3 m. After
the gravel was placed, the GMs were resurveyed. The frequency of additional holes
detected in the bottom GMs ranged from about 1 .3 to 4.8/ha. The holes ranged in size
from small punctures to a 60 mm in diameter.  Several of the holes were located in the
vicinity of the temporary ramps. No additional holes were detected in the GM primary
liners, which were separated from the gravel by the GN and GT.

Resolution: The identified GM holes were repaired.

Lessons Learned for Future Projects: Based on the available information, the following
lessons can be learned from this case history:

•  Construction-related holes in GM liners should be anticipated, even in liners installed
   with CQA. If there is a head of leachate over a liner hole, leakage occurs.  In the
   case history described above,  the performance of electrical  leak location surveys
   during sequential installation of multiple GM liners allowed extrusion seaming
   problems to be identified early and corrected, decreasing the frequency of identified
   GM holes in subsequent installations.
•  The CQA consultant should be trained in standard CQA practices, such as
   inspecting the entire installed GM liner for damage prior to placement of overlying
   materials.  In the case history described above, four punctures and  slits were found
   in the GM panels during the leak location survey.  If they were visible to a person
   standing by the hole, the CQA consultant should have identified them.
•  The potential for GM damage during placement of a soil layer over a GM can be
   reduced by protecting the GM.  Measures for GM protection include placing a
   protection layer (e.g., thick GT cushion or GC drainage layer) over the GM, using a
   greater initial lift thickness of soil above the GM, and using construction equipment
   with low ground pressure to place soils over the GM.  These options are consistent
   with EPA guidance (Daniel  and Koerner, 1993).
•  Special care should be taken to protect GMs in areas  subjected to high static and
   dynamic stresses from construction equipment, such as beneath  temporary access
   roads.  Measures for GM protection include placing a protection layer over the GM
   and  increasing the thickness of the soil layer over the  GM in these sensitive areas.

F-A.2.3   L
Problem Classification: landfill liner construction/construction
Region of U. S. :  southeast
Waste Type:  MSW
References: Silva, M., "Some Landfills Starting to Leak, State Study Says", Miami
    Herald, 18 Jun 1995, p. 6B-7B.
    Tedder, R.B., "Evaluating the Performance of Florida Double-Lined Landfills",
    Proceedings of Geosynthetics '97, Long Beach, California,  1997, Vol. 1, pp. 425-
    438.
                                      F-92
-------
Problem Summary: leakage through holes in HOPE GM primary liner

Problem Description: As described by Silva (1995) and Tedder (1997), a survey of LDS
flow rates from double-lined MSW landfills in Florida was performed by the Florida
Department of Environmental Protection (FDEP).  The highest average LDS flow rate
was occurring at a landfill cell in Dade County. The cell was constructed in 1991 with a
single-composite liner system on the 5.8 ha side slope and double-liner system on the
0.97 ha base. The double-liner system includes a GN LCRS drainage layer, HOPE GM
primary liner, GN LDS, and HOPE GM/CCL composite secondary liner. Leachate
collected in the LCRS and LDS sumps of the cell was pumped from the sumps when
the sump leachate level  reached 0.6  m.

Until March 1994, problems with the LCRS and LDS flow measuring system (e.g.,
flowmeters, check valves) made the measured flows unreliable. From April 1994 to
March 1995, the average LDS flow rate from the cell was about 4,660 Iphd.  All of this
flow was attributed to primary liner leakage, rather than other sources, based on
chemical constituents in  the LDS flow and the fact that the primary liner is a GM and the
LDS drainage layer is a GN. Tedder (1997) found that the measured flow rate was
greater than the calculated primary liner leakage rate for this cell of about 3,600 Iphd for
leakage through 2.5 11-mm diameter GM holes/ha under a head of 4 mm (i.e., the
thickness of the GN LCRS drainage layer).  Interestingly, Tedder showed that, of the 24
landfill cells considered in his survey, only one cell, the cell discussed in this section,
exhibited LDS flow rates attributable to primary liner leakage that were greater than
calculated primary liner leakage rates.

Tedder calculated the primary liner leakage rate assuming that the head on the liner
was 4 mm. However, the head of leachate in the sump area of this cell can be  much
greater, up to 0.6 m. Using Bernoulli's equation for free flow through an orifice, the
authors of this appendix calculated that the observed primary liner leakage rate of 4,660
Iphd could be caused by one 5.8 mm diameter GM hole under a 0.6 m head of leachate.
Resolution: In April 1995, the "pump on" level in the LCRS sump was lowered from 0.6
m to 0.45 m, and the LDS flow rates decreased. This decrease was attributed to the
reduction of primary liner leakage at the sump.  Over a two-month period, the "pump on"
level was further decreased, resulting in even lower LDS flow rates.  Currently, leachate
is pumped from the LCRS sump when the sump leachate level reaches 0.25 m. The
average LDS flow rates is now on the order of 2,000 Iphd.  GM primary liner holes have
not been located and repaired at this landfill because: (i) there is no anticipated
environmental impact of the primary liner leakage given the expected performance
capabilities of both the LDS and the composite secondary liner; (ii) repair of liner
systems after waste placement would be extremely difficult and expensive; and (iii)
additional liner system damage could occur in any attempt to excavate the waste and
repair the liner system.
                                     F-93
-------
Lessons Learned for Future Projects:  Based on the available information, the following
lessons can be learned from this case history:

•  With respect to the potential for leakage, leachate sumps are generally the most
   critical locations in landfills with internal sumps.  Leachate heads are typically
   sustained and at higher levels in sumps than at other locations.  In addition, GM
   liners in sumps often have seamed corners to fit the sump geometry.  These seams
   may contain  holes. As described above, even one GM hole at a sump can cause
   relatively high leakage rates due to the relatively high head of leachate in the sump.
•  To decrease the rate of leakage through GM holes at internal sumps, the sump
   design should include additional liner components, such as a GCL, beneath the GM
   liner in the sump area, even if the GM is already underlain by a CCL.  Bonaparte
   (1995) provides additional discussion of this design approach. The GM primary liner
   in the next group of cells constructed at the landfill described above was underlain
   by a GCL.
•  The GM panel layout should be configured to minimize seams in sumps.
•  The "pump on" levels  in sumps should be kept as low as practicable to reduce
   leakage if there are holes in the GM liner in the sump, especially if the GM is not
   underlain by  a GCL.

F-A.2.4   L
Problem Classification: landfill liner construction/operation
Region of U. S. :  southcentral
Waste Type:  HW
Reference:  Anderson, K.A., "Leachate Collection Management at a Hazardous Waste
    Landfill", Hydrological Science and Technology, Vol. 9, No. 1-4, 1993, pp. 30-53.
Problem Summary: leakage through holes in HOPE GM primary liner

Problem Description:  Beginning in March  1990, about six months after the start of
waste placement in a  new double-lined landfill cell located in Louisiana, average
monthly LDS flow rates from the cell increased from  less than 140 Iphd to up to 1,200
Iphd.  The relatively high LDS flow rates coincided with high rainfall and high LCRS flow
rates. The primary liner for the cell is a 1 .5-mm thick HOPE GM on the side slope and a
GM/CCL composite on the base. The LDS drainage layer is a GN.  Based on the
correspondence of LDS flow rates with LCRS flow rates and on the liner system
components, the LDS flow from the cell was  primarily attributed to leakage through the
GM primary liner on the side slope. The GM primary liner on the side slope is overlain
only by a GN LCRS drainage  layer and a GT filter. There is no soil protection layer on
the side slope; waste was placed directly on  the GT. It is not known if the GM primary
liner was damaged during construction or operation. However, without a thick
protection layer, the potential for liner damage during landfill operation increases. After
visual inspection of the exposed liner system at the site, the owner concluded that the
increased LDS flow was likely due to a GM hole in an area of the cell where runoff from


                                     F-94
-------
the waste was collected and pumped out for disposal.  This was confirmed after
rhodamine dye was added to the impounded leachate and detected in LDS flow less
than 24 hours later. Anderson (1993) did not give the depth of leachate that was
temporarily impounded in the cell. He did indicate, however, that the  leachate was
contained on one side of the cell by a 4-m high intermediate berm, and that the leachate
level varied.

Resolution: The ponded leachate was pumped out of the cell, waste was excavated
from the side slopes of the ponded area, and the liner system was inspected. Four
small tears in the GM primary liner were found and patched. After the repairs were
made, LDS flow rates decreased to previous levels, less than 140 Iphd.

Lessons Learned for Future Projects'. Based on the available information, the following
lessons can be learned from this case history:

•  Holes in GM liners should be anticipated, even in liners installed with CQA. If there
   is a head of leachate over a liner hole, leakage occurs.
•  GM holes can be located during landfill operation using dye or other methods.
   However, waste relocation may be required to repair these GM holes.
•  It is not clear if the GM holes found in the landfill cell described above were caused
   during construction or operation.  A soil protection layer was not placed between the
   waste and the liner system geosynthetics on the 2.5H:1 V side slopes, increasing the
   chances for the side slope liner system to be damaged by landfill equipment (e.g., a
   bulldozer) during waste placement. The potential for GM damage can be reduced
   by installing a protection layer (e.g., thick GT cushion or GC drainage layer) over the
   GM, using spotters to direct equipment operators during placement of waste over
   the GM, and placing only select waste over the GM.

F-A.2.5   LJ_

Problem Classification:  landfill liner construction/operation
Region of U. S.:  northeast
Waste Type: MSW
Reference: Loewenstein, D.L., and Smrtic, M.J., "Primary Liner System Repair Made
    During Operation of the  City of Albany Landfill: A Case History", Modern Double Lined
    Landfill Management Seminar, An Operation and Maintenance Perspective,
    sponsored by New York State Association for Solid Waste Management in
    cooperation with New York State Department of Environmental Conservation Division
    of Solid Waste, Saratoga Springs, New York, Jan 1994, 6 p.

Problem Summary: leakage though  HOPE GM/CCL composite primary liner at pipe
penetration

Problem Description: In March 1992, about four months after waste placement began,
the average LDS flow rate from a double-composite lined landfill cell located in New
                                     F-95
-------
York increased from about 10 to 400 Iphd. At the time, only a small amount of waste
had been placed in the cell. The sharp increase in LDS flow rate could not be attributed
to leakage through the composite primary liner since flow rates through a composite
liner for a landfill should be very small (i.e., less than 1 Iphd).  It could also not be
attributed to water expelled  from the CCL component of the primary liner as it
compressed under the weight of the overlying waste since little waste had been placed.
Shortly before the  flow rate  increased, landfill operations personnel had been
performing minor regrading of the 0.45-m thick sand LCRS drainage layer on the base
of the cell. When the  base of the cell was inspected in the area of regrading activities,
deep tire tracks were observed in the sand over the LCRS pipe penetration of the
primary liner at the perimeter berm. The tracks had been caused by a rubber-tired
loader. The 150-mm diameter Schedule 80 PVC LCRS pipe was positively projecting
into the sand LCRS drainage layer. With this design, the specified separation between
the top of the pipe and equipment trafficking on the sand is only 0.3 m.

A closed-circuit television was used to inspect the LDS pipe to evaluate where leachate
was entering the pipe.  However, due to the sediment buildup in the pipe, it was not
clear where leachate was entering.  The LDS pipe was flushed and the camera was
reintroduced into the pipe. This time, sediment was observed accumulating in the LDS
pipe at a distinct location near the pipe penetration of the composite secondary liner.
The GM boot at the LCRS pipe penetration of the primary liner was cut so the
penetration could be inspected.  It was found that the LCRS pipe was broken between
the penetration and where it was connected to the boot. Additionally, the CCL
component of the composite primary liner was rutted adjacent to the LCRS pipe. The
damage at the penetration had allowed leachate to flow out of the broken LCRS pipe
within the pipe boot, between the LCRS pipe and the damaged CCL, and into the GC
LDS drainage layer.

Resolution: The softened, rutted CCL was removed around the LCRS pipe and
replaced with new compacted clay.  All cracked or broken LCRS pipe was replaced.
Then a new GM boot was constructed around the pipe penetration of the GM primary
liner.  About one year after the repair, LDS flow rates were  approximately 50 Iphd.

Lessons Learned for Future Projects'.  Based on the available information, the following
lessons can be learned from this case history:

•  Landfill operations manuals should include  limitations on the types of equipment that
   may traffic over the liner system before the first lift of waste is placed. These
   limitations are enforced during construction when CQA is implemented; they should
   also be applied during operation.
•  Landfill operations personnel should be aware of sensitive areas of a liner system,
   such as at pipe penetrations or sumps, and should protect these areas. In the case
   history described above, the loader operator repeatedly trafficked on the sand
   drainage layer  above the pipe penetration until the sand had deep tire tracks. If the
   operator had been directed to protect this sensitive area, he may have used a
                                     F-96
-------
   different path over the liner system.  Sensitive areas can be identified with cones,
   flags, or other markers to reinforce this. They can also be isolated from traffic by
   berms, bollards, or other means.

F-A.2.6   L-8

Problem Classification:  landfill liner construction/design
Region of U. S.:  southwest
Waste Type:  MSW
Reference: unpublished

Problem Summary: landfill gas migrated beyond liner system and into vadose zone
resulting in groundwater contamination

Problem Description: A single-composite liner system for a landfill was constructed in
1990 in  a canyon underlain by sedimentary marine bedrock and alluvial deposits. Along
the perimeter of the landfill, the composite liner was extended horizontally  and the GM
was secured by covering it with a 0.9-m thick layer of relatively permeable soil. At the
toe of the landfill, near the mouth of the canyon, an asphalt parking lot was constructed
over the relatively permeable soil layer and natural ground. Upgradient of the landfill, in
the upper reaches of the canyon, the shallow aquifer is located about 8 m  below ground
surface  in  bedrock. Downgradient of the landfill, near the mouth of the canyon, the
shallow aquifer is located about 22 m below ground surface in alluvium and bedrock.
Groundwater flow velocities range from less than 0.1 m/yr in the bedrock to about 250
m/yr in the alluvium.

Four years after start of landfill operations, the waste near the toe of the landfill was at
intermediate grades and covered with a relatively low-permeability soil intermediate
cover layer that graded  into the relatively permeable soil layer anchoring the edge of the
liner.  At this time, several VOCs (i.e., trichlorofluoromethane, dichlorofluoro-methane,
benzene, toluene, xylenes, ethylbenzene, and dichlorobenzene) were detected in
groundwater from a monitoring well located about 60 m from the downgradient edge of
the landfill. The VOCs were generally at concentrations less than EPA maximum
contaminant levels (MCLs) for drinking water.  Based on groundwater modeling, the
VOC plume downgradient of the  landfill was estimated to be approximately 60 m wide,
100 m long, and up to 3 m deep  in the shallow aquifer.  The source of VOCs was
thought to be landfill gas for the following  reasons: (i) inorganic landfill leachate
constituents were not detected in samples from the downgradient well at concentrations
above background levels; (ii) the VOCs detected in groundwater were also found in
leachate and gas samples; and (iii) relatively high concentrations of methane (i.e.,
greater than 30%) were detected in the headspace of downgradient groundwater
monitoring wells and groundwater subdrains beneath the landfill liner system.

To verify the source of VOCs was landfill gas, a soil gas survey was conducted at 300
m intervals along the perimeter of the landfill. Based on the results of the survey, gas
                                      F-97
-------
appeared to be migrating out of the landfill and into the vadose zone near the landfill
toe.  Here, the asphalt served as a confining layer, blocking landfill gas moving from the
waste into the relatively permeable soil layer from venting to the atmosphere. Instead,
some gas migrated through the relatively permeable soil layer and then beyond the limit
of the liner system. Around the remainder of the landfill, where there was no asphalt,
gas migrating into the relatively permeable soil layer essentially vented to the
atmosphere and had no observable impact on groundwater.  In addition to the soil gas
survey, gas samples were collected from the headspace of groundwater monitoring
wells, groundwater subdrains, and vadose gas monitoring probes. The chemical
signatures of the gas samples indicated that landfill gas was the likely source of the
VOCs in groundwater downgradient of the landfill.

Resolution: A fate and transport analysis of the VOC plume was performed that
demonstrated the plume would be retarded, primarily by biodegradation, before
reaching the property line if gas was controlled. The property line is located about 460
m from the downgradient edge of the landfill.  To improve VOC source control, two
additional gas extraction wells were installed in the waste near the landfill toe.  If these
wells do not provide adequate source control,  the liner system under the asphalt at the
perimeter of the landfill may have to be reconstructed to eliminate the gas migration
pathway and a groundwater remedy may need to be implemented.
Lessons Learned for Future Projects'.  Based on the available information, the following
lessons can be learned from this case history:

•  Landfill gas may migrate into the vadose zone at the perimeter of a landfill if gas is
   not well controlled and there is a pathway.
•  In the case history described above, the landfill would not have impacted
   groundwater if the landfill gas had been better controlled, the soil layer anchoring the
   edge of the liner system had been less permeable, the asphalt parking lot had not
   been constructed over the edge of the liner system, or the edge of the liner system
   had been modified  so that the end of the liner extended back to the ground surface
   and into the asphalt. With the latter detail,  gas is blocked from moving laterally or
   vertically into the vadose zone.

F-A.2.7   LA

Problem Classification: landfill liner construction/construction
Region of U. S.: northeast
Waste Type:  MSW
Reference:  unpublished

Problem Summary: leakage though HOPE GM primary liner at pipe penetration

Problem Description:  After a heavy rainfall that ponded up to 0.5 m of water in the
LCRS of three newly constructed  cells, one of the cells exhibited primary liner leakage
near the LCRS pipe penetration of the primary liner. The LDS flow rate from this cell
                                      F-98
-------
was about 250 Iphd. The pipe penetration detail had been carefully constructed and
had two special features designed to minimize the potential for leakage through the
HOPE GM primary liner: (i) the GM primary liner was underlain by a GCL at the
penetration; and (ii) the penetration was sealed with two HOPE boots, creating a space
that could be pressure tested and later filled with foam. The double-boot system can be
described as follows: (i) starting at about 1.5 m into the cell from  the LCRS pipe
penetration, a 250-mm diameter pipe was placed around the 150-mm diameter LCRS
pipe to provide secondary containment of leachate as the LCRS  pipe penetrated the
secondary liner and the perimeter berm around the landfill; (ii) an HOPE plug was
placed at the end of the outer pipe in the cell to prevent inflow of leachate into the
annular space between the two pipes; (iii) two  prefabricated HOPE GM boots were
installed around the outer pipe where the pipe  penetrated the primary liner; and (iv) the
boots were clamped to the pipe and extrusion seamed to the primary liner. The seams
were vacuum tested where there was space for the vacuum box.  Then the integrity of
the boots was tested by applying an air pressure of about 170 kPa to the space
between the boots. This pressure was too high, and the outer boot was overstressed
and failed where it was seamed to the primary liner.  The outer boot was subsequently
reseamed to the liner.  The boots were retested at a pressure of  20 kPa and found to
hold the air pressure without noticeable leakage. Then the space between the boots
was filled with expanding foam sealant.  After a heavy rain, which resulted in ponded
rainwater at the pipe penetration, primary liner leakage occurred  at the penetration, as
described above.

Resolution: The pipe bedding gravel around the pipe penetration was removed, and the
outer pipe boot was inspected.  A small hole at the extrusion seam of the outer boot to
the primary liner was found and repaired. Again rainwater ponded at the pipe
penetration, and primary liner leakage was observed.  The outer boots was uncovered
again and inspected, but no obvious GM hole was observed. Since the pathway for this
leakage could not be identified during construction, the problem was not remedied.  The
environmental impact from the leakage, however, is expected to  be negligible given that
the landfill has a composite secondary liner. Interestingly, at another landfill in the same
region, holes in a double-booted pipe penetration of the primary liner were identified
during construction using an ammonia colorimetric leak test (ASTM E 1066). The holes
were repaired, and the penetration was found to be essentially nonleaking when
rainwater ponded around it.

Lessons Learned for Future Projects'. Based on the available information, the following
lessons can be learned from this case history:

•  With respect to the potential for leakage, pipe penetrations are generally the most
   critical locations within landfills without internal sumps. As demonstrated  by the
   case history described above, even when extra measures are taken to enhance the
   integrity of the primary liner at the penetration, it  is difficult to  construct the
   penetration to be defect free. Methods for constructing better connections between
   GMs and ancillary structures are needed.
                                      F-99
-------
•  Since pipe penetrations are critical locations, designs without pipe penetrations
   should be preferred whenever possible.

F-A.2.8   L-11

Problem Classification: landfill liner construction/construction
Region of U. S.:  northeast
Waste Type:  MSWash
Reference: unpublished

Problem Summary:  construction debris in CCL with initially smooth surface protruded
from CCL after CCL was left exposed and subsequently eroded

Problem Description: During construction of the soil-bentonite CCL component of a
composite secondary liner, relatively large pieces of construction debris (e.g., bricks,
rebars) were found in the soil delivered to the site for use in the CCL. While the project
specifications precluded the contractor from using borrow soil containing "deleterious"
material, debris-free soil was not locally available.  Additionally, the soil, as delivered to
the site, had particles larger than the specified maximum particle size of 19  mm.  In
spite of being notified by the owner that the soil did not meet specifications,  the
contractor indicated that the soil would  be acceptable after it had been screened. The
contractor then proceeded with CCL construction.  After the soil had been screened, the
total amount of construction debris remaining in the soil was about 0.02% by weight. In
addition, laboratory permeability and field Boutwell permeability tests performed on the
CCL demonstrated that the CCL met the  hydraulic conductivity criterion of the
specifications (i.e., hydraulic conductivity  no greater than 1 x 10~9 m/s).  Construction of
the CCL was completed in the fall. At this time, the surface of the CCL was smooth and
debris was not visible. The contractor left the CCL unprotected over the winter.  When
construction resumed the next spring, the surface of the CCL had eroded and some
pieces of construction debris were protruding from the CCL.  The CCL in this state was
not suitable for placement of the overlying GM liner.

Resolution:  The state regulatory agency  required the CCL on  the base of the landfill to
be covered with a GCL and the CCL on the side slope to be covered by a 80-mm thick
layer of debris-free CCL to protect the overlying GM from puncture by protruding debris.
A GCL was not required on the 2.5H:1V side slope due to concerns with the effect of
the GCL on liner system  slope stability.

Lessons Learned for Future Projects: Based on the available information, the following
lessons can  be learned from this case history:

•  CCLs  should  not be  constructed with materials containing construction debris or
   large particles, even if prior to GM installation the CCL has a smooth surface and
                                      F-100
-------
Figure F-A.2.1. Construction debris in a CCL with an initially smooth surface
protruded from the CCL after the CCL was left exposed and subsequently eroded.
   meets the hydraulic conductivity criterion. The debris may adversely impact the
   hydraulic conductivity of the CCL and/or damage an overlying GM.  In the case
   history described above, the constructed CCL had a smooth surface and met the
   hydraulic conductivity criterion after construction. However, left unprotected, it
   developed, due to erosion, a surface with some protruding particles.
•  CCLs should not be left unprotected for an extended period of time. They can
   desiccate and crack due to evaporation of water in the CCL, crack when exposed to
   freezing and thawing actions, and be eroded by wind and water.

F-A.2.9  L-11

Problem Classification: landfill  liner construction/construction
Region of U. S.: northeast
Waste Type: MSWash
Reference:  unpublished

Problem Summary: leakage though HOPE  GM primary liner at pipe penetration
                                     F-101
-------
Problem Description: After rainwater ponded in three newly-constructed landfill cells,
the LDS flow rates from the cells were about 200, 200, and 1,000 Iphd.  These flows
were primarily attributed to primary liner leakage since the LDS drainage layer is a GN
and the primary liner on the base of the cells is a composite consisting of an HOPE GM
over a GCL and structural fill  layer.  The owner suspected that the leakage was
occurring at defects around the LCRS pipe penetrations of the primary liner as it is
difficult to construct a defect-free connection between HOPE GM and pipes. To test
primary liner integrity at the penetrations, ponding tests were conducted. In each cell,
dye added to water ponded in the LCRS around the pipe penetration was rapidly
detected in LDS liquid, indicating a direct connection between the LCRS and LDS at the
penetration.  Subsequently, the LCRS pipe penetrations were inspected to locate the
defects. In two cells, including the cell with the highest LDS flow rate, the GM pipe
boots were not securely fastened to the pipes; a pencil could be inserted between the
boots and the pipes.  In the remaining cell, the GM pipe boot appeared  adequate
though rather short.

Resolution: In the cell with the highest LDS flow rate, the faulty GM boot was
reconstructed.  Silicon sealant was placed at the intersection of the pipe and GM boot in
the other cell with an obvious boot problem.  In the remaining cell, a new pipe boot was
installed over the old short boot. In  each cell, the space between the boot and the
primary liner was filled with bentonite slurry.  Based on a subsequent ponding test with
dye, primary liner leakage into the LDS sumps, if any, was very small and could not be
distinguished from earlier primary liner leakage or construction water. By two months
after the repairs, the LDS flow rates had dropped to less than 100 Iphd for the cell with
the reconstructed boot and less than 10 Iphd for the other two cells.

Lessons Learned for Future Projects:  Based on the available information, the following
lessons can be learned from this case history:

•  With respect to the potential for leakage, pipe penetrations are generally the most
   critical locations within landfills without internal sumps.
•  Since pipe penetrations are critical locations, designs without pipe penetrations
   should be preferred whenever possible.  If pipe penetrations are used, they  should
   be carefully constructed and inspected.
•  The integrity of pipe penetrations can be tested during construction and, if
   necessary, the penetrations can be improved. In the case history presented above,
   leakage at a pipe penetration was detected  during construction and  reduced by
   filling the space between the boot and the GM primary liner with bentonite slurry.

F-A.2.10  L-15

Problem Classification: landfill liner construction/construction
Region of U. S.:  southeast
Waste Type: MSW
                                      F-102
-------
Reference:  unpublished

Project Summary,  sand bag under installed GM liner approved by CQA consultant

Problem Description: About 2 ha of GM liner was installed and inspected by site CQA
personnel. The liner was found to be acceptable. Regulatory personnel who
subsequently inspected the liner observed a bump beneath the liner on the slope of the
LCRS pipe trench. When the liner was cut at a bump, a sand bag was found beneath
the liner.

Resolution:  The sand bag was removed, and the liner was repaired.

Lessons Learned for Future Projects:  Based on the available information, the following
lessons can be learned from this case history:

•  Prior to deploying a GM, all extraneous objects (e.g., tools, sand bags) should be
   removed from the surface on which the GM is to be placed. If sand bags are used to
   secure the GM panels until the panels are seamed, care should be taken by the
   installer to ensure that the sandbags, and all other extraneous objects, are not
   trapped beneath the GM after seaming.
•  The CQA consultant should be trained in standard CQA practices, such as
   inspecting the subgrade for extraneous objects and gravel that may damage the
   liner prior to liner deployment and inspecting the installed liner for underlying objects.

F-A.2.11 L-17

Problem Classification: landfill liner construction/construction
Region of U. S.: northeast
Waste Type: HW
Reference: Bonaparte,  R. and Gross, B.A., "LDCRS Flow Rates from Double-Lined
   Landfills and Surface Impoundments", EPA/600/SR-93/070, EPA Risk Reduction
   Research Laboratory, Cincinnati, Ohio, 1993, 65 p. (Landfill AB)

Problem Summary: leakage through holes in HOPE GM primary liner

Problem Description: Shortly after waste placement began in a double-lined landfill  cell,
the black color of the LDS  liquid and the results of chemical analysis of the liquid
indicated that primary liner leakage was occurring. The LDS flow rates at this time were
approximately 1,030 Iphd.  The side slope liner system includes a 0.3-m thick gravel
protection layer, GN LCRS drainage layer, HOPE GM primary liner, and GN LDS
drainage layer. The base liner system includes an HOPE GM/CCL composite primary
liner and a gravel LDS drainage layer.  Based on the liner system components, leakage
was suspected to be occurring through the GM primary liner on the side slope. Waste
was excavated off the side slope where it was present, and the liner system was
inspected. Two 10-mm diameter GM holes and five GM fusion seam holes were found
                                     F-103
-------
on the side slopes. At the fusion seam holes, the outside track of the dual track seam
had separated allowing leachate to flow through the air channel between the tracks and
potentially through the liner if the inside track had holes.  Though the seam had passed
air pressure testing during construction, only the  strength of the inside track, which was
not visible, had been tested.  The project specifications did not require testing of both
the inside and outside tracks.

Resolution: The GM holes were repaired.  The average LDS flow rate for the next two
months after the repairs was about 320 Iphd.

Lessons Learned for Future Projects: Based on the available information, the following
lessons can be learned from this case history:

•  Construction-related holes in GM liners should be anticipated,  even in liners installed
   with CQA.  If there is a head of leachate over a liner hole, leakage occurs.
•  Frequently, both the inside and outside tracks of a fusion seam are tested in  peel
   and shear. In the case history described above, if the outside  track had also been
   destructively tested, the separation of the outside track after construction may have
   been avoided. If both tracks are required to have integrity, both tracks  should be
   tested.
•  Even though it may not be required for a project, it may be beneficial to test both
   tracks of the dual track fusion seam to gather more information on overall seam
   quality.  Dual track fusion seaming machines  are designed to make high quality
   seams along two tracks. Holes periodically occurring in one track may also be
   occurring in the other track. By testing both tracks, seaming problems  may be
   identified and corrected quicker.  Also, as shown by Giroud et  al. (1995b), the failure
   of one track of the seam increases the bending strains in the GM next to the
   adjacent track, simply due to seam geometry. These strain concentrations cause
   stress concentrations, since stress and strain are linked, and may lead to GM failure
   under certain conditions.  The increase in stress concentrations can be avoided by
   having two intact seams.  Furthermore, the costs associated with destructive testing
   of both tracks of GM seam samples is small relative to the total cost of CQA.

F-A.2.12  L-19

Problem Classification: landfill liner construction/construction
Region of U. S.: northeast
Waste Type:  MSW
Reference: unpublished

Problem Summary:  wind uplifted and tore HOPE GM liner during  construction

Problem Description: During installation of a 1.5-mm thick HOPE  GM liner, the GM
installer arrived at the site one morning to find that about 3 ha of GM had been uplifted
by wind, torn from the installed GM, and blown into a twisted, folded pile.  The wind
                                      F-104
-------
uplift occurred even though the GM had been weighted down with sand bags.
Interestingly, the GM primarily tore through extrusion seams and along, but outside of,
fusion seams.

Resolution: None of the wind-blown GM in the pile could be salvaged because the GM
contained too many folds and creases where the yield strain of the HOPE had been
exceeded. The damaged GM was replaced with new GM of the same type and
thickness.

Lessons Learned for Future Projects'.  Based on the available information, the following
lessons can be learned from  this case history:

•  The leading edge of an uncovered GM liner should be secured to prevent wind from
   flowing beneath the GM and uplifting it. This is typically accomplished by seaming
   adjacent panels of GM liner shortly after deployment and placing a row of adjacent
   sandbags along the edge of the GM.
•  GM liners should be covered with a soil layer as soon as practicable after installation
   to prevent GM uplift by wind.

F-A.2.13  L-19

Problem Classification: landfill liner construction/construction
Region of U. S.: northeast
Waste Type: MSW
Reference: unpublished

Problem Summary: severe wrinkling of HOPE GM due to thermal expansion during
construction

Problem Description:  Construction of two 12-ha single-composite lined landfill cells
began in the fall.  Deployment and seaming of the 1.5-mm thick HOPE GM component
of the composite liner were conducted through the winter, when temperatures were near
or below freezing.  At this time the GM was relatively taut.

Placement of the sand LCRS drainage layer was delayed because the sand initially did
not pass the conformance tests. As a result, the GM was left exposed. In the spring
and summer, due to the warmer weather, the exposed GM expanded and became very
wrinkled.  Then, the sand layer could not be placed because of concern that the GM
wrinkles would  fold over beneath it.

Resolution: Several thousand linear meters of wrinkles were required to be cut and
repaired prior to placing the overlying sand LCRS drainage layer.  In  addition, the sand
layer was required to be placed during the cooler nighttime hours, i.e., when GM
wrinkles are smaller. Wrinkle repair and sand placement took several months.
                                     F-105
-------
Lessons Learned for Future Projects: Based on the available information, the following
lessons can be learned from this case history:

•  GM liners should be covered with a soil layer as soon as practicable after installation
   to reduce GM wrinkles and to protect the liner from damage. Placement of the soil
   layer should not cause the GM wrinkles to fold over; to that end, it may be necessary
   to place the soil during the cooler nighttime and  early morning hours when the
   wrinkles are smaller.
•  Liner systems should be constructed in manageable increments that ensure
   protection of the liner system materials under seasonal weather changes.  In the
   case history described above, if the expansion had been constructed in several
   increments, the GM liner constructed in the winter would have likely been covered
   with the sand LCRS drainage layer by spring and would have been less wrinkled.

F-A.2.14  L-30

Problem Classification: landfill liner construction/construction
Region of U. S.: northeast
Waste  Type: MSW
Reference: Koerner, G.R., Eith, A.W., and Tanese, M., "Properties of Exhumed HOPE
    Field Waves and Selected Aspects of Wave Management", Proceedings of the 11th
    GRI Conference on Field Installation of Geosynthetics, Geosynthetic Research
    Institute, Philadelphia, Pennsylvania, 1998, pp.  152-162e.

Problem Summary: large folded wrinkles in HOPE GM primary liner at two exhumed
leachate sumps

Problem Description:  When about 0.4 ha of previously-constructed liner system was
exposed at two leachate sumps in an active double-lined  landfill during construction of a
landfill  expansion,  the 1.5-mm thick HOPE GM top and secondary liners on the side
slope in the vicinity of the sumps were found to have large, folded wrinkles. The GM
liners on the side slope were separated by a GN LDS drainage layer.  The landfill liner
system had been constructed in 1988, approximately eight years prior to the
exhumation.  GM wrinkles on the landfill side slope  above the sumps were orientated
diagonal to the crest of the 31-1:1 V side slope and converged in the sumps. Wrinkles
were also more numerous and larger near the slope toe and near the sump than away
from the slope toe and the sump. These wrinkles likely began to develop when the GM
expanded during the day as temperatures increased and  became folded as gravel was
placed  over the GM. As the wrinkles propagated down the slope during several days  of
temperature cycling, the wrinkles became more numerous and larger at the slope toe.
Koerner et al. (1998) hypothesized that the diagonal wrinkle orientation was due to the
original backfilling  operation in the sump area. The backfill was placed from the toe of
the slope upwards and outwards. They also suggested that the two riser pipes in the
sump complicated backfilling and may have contributed to some of the localized GM
movement and the resulting wrinkle pattern. The 50 to 800 mm high wrinkles had three
                                     F-106
-------
main configurations: (i) a vertically folded prayer wrinkle; (ii) a horizontally folded S-
shaped wrinkle; and (iii) a mushroom-shaped wrinkle (i.e., a prayer wrinkle confined
laterally and loaded vertically). At some of the wrinkle folds, the GM was observed to
be yielding, as evidenced by the change in its color (i.e., it became lighter). These
wrinkles were present even though the liner system had been carefully constructed with
third-party CQA.  The large wrinkles were not noticed prior to placement of a 540 g/m2
needlepunched nonwoven GT cushion over the GM primary liner on the side slope.
Furthermore, the sump and leachate riser pipe bedding gravel and the first 0.3 m of the
0.6-m thick gravel LCRS drainage layer were placed over the GT by a crane with a
large bucket (typically used to place concrete) to reduce GM wrinkle development and
propagation.

Samples of unwrinkled GM and wrinkled GM  at folds were taken for laboratory testing.
The wrinkled GM had yielded and was noted to be thinner than adjacent unyielded GM.
Both wide width tensile tests (ASTM D 4885)  and single point notched constant tensile
load tests (ASTM D 5397) were conducted to assess the effect of the GM folding on GM
integrity.  Interestingly, all of the wrinkled GM samples failed at folds. Both the wrinkled
and unwrinkled GM samples had wide width tensile properties at yield that met the
project specifications and were not significantly different from one another. Wide width
tensile properties at break could not be evaluated due to equipment limitations.  All GM
samples also had acceptable times to failure (i.e., greater than 200 hours) in the
notched constant tensile load test.  However, the wrinkled GM samples had a
somewhat lower time to break than the unwrinkled samples (i.e., average of 1,033 to
>1,823 hours for wrinkled samples and >2,300 hours for all unwrinkled samples).

It should be noted that wrinkling of GM liners  may also be caused by downdrag of the
GM by waste as it settles. However, the wrinkles associated with downdrag should be
shorter and more numerous than those associated with thermal expansion. This can be
explained based on wrinkle theory developed by Giroud (1994b). As a result of the
bending associated with a GM wrinkle, there are two opposite forces at the base of the
wrinkle.  These forces must be balanced by the shear force that results from the settling
waste and the  interface friction between the GM and the material beneath it. For a
given amount of GM elongation, as the shear force increases, the wrinkles become
shorter and more numerous. The shear forces associated with settling waste (i.e.,
downdrag and the weight of the GM and waste) are greater that those associated with a
GM alone.  Thus, the difference in wrinkle appearance between wrinkles associated
with thermal expansion (that develop under very small compressive stresses) and the
wrinkles that develop under high compressive stresses.

Resolution: The wrinkled GM was removed when the sumps were reconstructed for the
landfill expansion. Textured GM was installed on the side slope to  increase the
interface shear strength between the GM and underlying CCL, for the composite
secondary liner, and between the GM and underlying GC LDS drainage layer, for the
primary liner. As a result of increasing the interface shear strength, the wrinkles that
developed on the side slope were less likely to propagate downslope.  Interestingly,
                                     F-107
-------
Giroud (1994b) has shown analytically that GM wrinkles are shorter and spaced closer
together when the shear strength between the GM and the underlying material is
increased, which may result from using a textured GM and/or from increasing the
normal stress.

Lessons Learned for Future Projects'.  Based on the available information, the following
lessons can be learned from this case history:

•  GM liners  should be covered with a soil layer as soon as practicable after
   installation, but not during the hottest part of the day if the GM is significantly
   wrinkled, to reduce GM wrinkling as a result of the following two mechanisms: (i)
   thermal insulation provided by the soil  layer reduces GM temperature variations that
   cause wrinkling; and (ii) the increase in interface shear strength resulting from the
   weight of the soil layer decreases the size of the wrinkles. A thick GT cushion is
   generally not sufficient to protect the GM from thermal effects.
•  Placement of the soil layer should not cause the GM wrinkles to fold over.  Prior to
   placing soil over a GM, the GM should be inspected for wrinkles.  Excessive GM
   wrinkles and wrinkles that may fold over should be removed by waiting to backfill
   until the GM cools and contracts during the cooler nighttime and early morning
   hours, pulling the wrinkles out, or cutting the wrinkles out. The latter method is less
   desirable than the former methods because it requires intact GM to be cut, and it
   results in more GM seaming and subsequent testing.
•  Using a textured GM on the side slopes has two beneficial effects: (i) it results in
   wrinkles that are smaller than in the case of a smooth GM; and (ii) it reduces the
   propagation of wrinkles down the side  slopes.  These two effects decrease the risk
   of the formation of very large wrinkles at the toe of side slopes.

F-A.3     Landfill Liner Degradation

F-A.3.1   L£

Problem Classification:  landfill liner degradation/design
Region of U. S.:  southeast
Waste Type:  MSW
Reference: Basnett,  C.R. and Bruner, R.J., "Clay Desiccation of a Single-Composite
    Liner System", Proceedings of Geosynthetics '93, Vancouver,  British Columbia,
    1993, Vol. 3, pp. 1329-1340.

Problem Summary: desiccation cracking of CCL in exposed  HOPE GM/CCL composite
liner

Problem Description: Basnett and Bruner (1993) described the severe desiccation
cracking of the CCL component of the composite liners for two sections (Section 1 and
2) of a landfill in Florida. The liner systems consist of the following components, from
top to bottom:
                                      F-108
-------
•  0.6-m thick sand drainage layer;
•  1.5-mm thick HOPE GM liner;
•  0.3-m thick clayey sand CCL (maximum hydraulic conductivity of 1 x 10~9 mis); and
•  GT reinforcement layer.

Prior to installing the GM liner, desiccation cracking of the CCL was controlled by
spraying it with water. Due to the potential for sliding and erosion of the sand on the 3
horizontal:! vertical (3H:1V), approximately 14-m high, liner system side slopes, the
sand drainage layer was only placed on the landfill base during construction.  The
design required that sand be placed incrementally up the side slopes during landfill
operation, with the sand advancing ahead of the waste.

In 1988, near the end of construction of the Section 1 liner system, water was observed
to be trapped between the GM liner and CCL at the slope toe. At that time, the water
was attributed to rainwater that had flowed beneath the GM during installation. In  1991,
construction  of the Section 2 liner system began. When the GM panel in Section 1 was
rolled back to allow the Section 2 CCL to be constructed into the  CCL in Section 1, the
Section 1 CCL on the side slope was observed to be severely desiccated. Cracks were
about 25 mm wide and penetrated the full thickness of the CCL.  To  evaluate whether
the  desiccation was localized, the Section 1 CCL was inspected at 30 m from Section 2,
about 2 m from the slope crest.  Here the desiccation was also severe and the bottom
surface of the overlying GM was moist.  Construction of Section 2 continued and, similar
to the Section 1 construction, water became trapped between the GM and CCL near the
slope toe.  About six weeks after liner installation, the water mounding in Section 2 was
so significant that small weep holes were cut in the GM to drain the water. These  holes
were subsequently patched.

The observed phenomenon can be explained as follows.  During the day, as the
temperature  of the liner system increases, water evaporates from the CCL and the air
entrapped between the CCL and the GM becomes saturated with water vapor. During
the  night, as  the temperature decreases, the ability of the entrapped  air to contain water
vapor decreases.  As a result, a fraction of the vapor condenses into water and droplets
of water appear against the lower face of GM since the GM is, then the coldest element
of the system. If the droplets are small,  surface tension prevents them from moving.  If
the  droplets are large (i.e., if the evaporation-condensation mechanism is significant),
gravity overcomes surface tension and water migrates downslope and accumulates at
the  toe of the slope. Therefore, progressively, the CCL moisture content decreases in
the  upper part of the slope and increases near the toe. To evaluate the mechanism,
Basnett and  Bruner (1993) collected CCL samples along  a transect extending down the
side slope in Section 1 and Section 2, and the samples were analyzed for moisture
content.  The Section 1 CCL, which was constructed with an average moisture content
of about 22% along the transect, had average moisture contents  of about 7% nearest
the  slope crest, 11 % near midslope, and 24% near the slope toe. The moisture content
was also observed to increase with depth at sampling locations on the upper portion of
                                     F-109
-------
the slope.  Similar effects, though less extreme, were observed for the Section 2 CCL
when it was sampled and tested at about two months after construction of the
composite liner.

Resolution: At the time of the field  study described by Basnett and Bruner (1993),
Section 1 was filled to about 70% of its waste capacity, and its composite liner was
covered with a sand drainage layer. Section 2 had an exposed liner system and had
not yet accepted waste.  No actions were required for Section 1 by the regulatory
agency presumably because: (i) the older cell was almost filled  and would be closed
shortly afterwards; (ii) the repair would require that the waste be removed from the cell,
which is extremely difficult, and costly; (iii) the CCL was only observed to be desiccated
on the side slope;  all other thing being equal, side slope liner holes are less detrimental
than base liner holes because the head of leachate on the side slope is less than the
head on the base slope; and (iv) environmental impacts from the potential for increased
liner leakage are expected to be negligible given that the liner includes a GM.  Until it is
covered and thermally insulated with the sand drainage layer, the Section 2 CCL on the
side slopes will likely continue  to lose moisture and crack.

Lessons Learned for Future Projects:  Based on the available information, the following
lesson can be learned from this case history:

•  Composite liners constructed with a CCL should be covered to prevent heating and
   desiccation of the CCL. EPA cautions that temporarily using a GM alone over a
   CCL may be problematic and indicates that a light-colored GM may be preferable
   (Daniel and Koerner,  1993).

F-A.3.2    L-4

Problem Classification: landfill liner degradation/operation
Region of U. S.:  northeast
Waste Type: HW
Reference: Adams, FT.,  Overmann, L.K., and Cotton, R.L., "Evaluation and
    Remediation of a Fire Damaged Geosynthetic Liner System",  Proceedings of
    Geosynthetics '97, Long Beach, California,  1997, Vol. 1, pp. 379-392.

Problem Summary:  HOPE GM/CCL composite liner damaged by waste fire

Problem Description: Adams et al. (1997) described the impact of a waste fire on a
double-liner system for an active HW landfill containing industrial waste sludges and
other chemical manufacturing  by-products. The fire was caused by a chemical reaction
of one of the materials disposed of in the landfill and located within several meters of
the liner system. Based on temperature measurements made near the fire, Adams et
al. (1997) concluded that the temperature in the vicinity of the liner system may have
approached 800°C.  After the fire was discovered, it took about 11 months to contain
and finally extinguish it. The fire was controlled by installing hundreds of 50-mm
                                      F-110
-------
diameter steel pipes into the affected waste at approximately 0.8 m spacings and
recirculating leachate through the pipes to cool the waste.

After the fire was put out, waste was excavated over the portion of the landfill believed
to be affected by the fire. About 300 m2 of liner system on the landfill side slope was
obviously damaged based on visual inspection. Nearer the center of the affected area,
damage was most severe, as evidenced by the melting and disintegration  of the liner
system geosynthetics and desiccation cracking of the CCL component of the composite
secondary liner.  Nearer the boundary of the affected area, the geosynthetics appeared
rippled and stretched.

Resolution: The damaged liner system components were identified, removed, and
replaced with new materials.  The extent of damage to the CCL component of the
composite secondary liner was defined by visual inspection and hydraulic conductivity
testing of CCL samples. Near the center of the impacted area, the entire 0.9-m
thickness of the CCL was desiccated.  Here, the hydraulic conductivity of the CCL
samples was about two to three orders of magnitude greater than that measured for
samples collected during CCL construction.  As the distance from the center of the
impacted area increased, the thickness of CCL affected by the fire decreased, defining
a bowl-shaped region of CCL impact.  The desiccated CCL was removed and replaced.
With respect to the HOPE GM primary liner and the HOPE GM component of the
composite secondary liner, the extent of damage was defined by visual inspection and
the results of laboratory testing of GM samples. GM samples were taken along the
perimeter of the area that appeared to be damaged and tested for thickness, density,
melt flow index, tensile strength at break, and tensile elongation at break.  Samples
were required to meet the original project specifications for these tests.  If a sample did
not meet the specifications, another sample was cut near the failing sample, but further
from the area with visible damage. The process was repeated until an area bounded  by
samples that met the specifications could be defined. The GM within this area was
removed and replaced.

Lessons Learned for Future Projects:  Based on the available information,  the following
lessons can be learned from this case history:

•  Geosynthetics and CCLs can be severely damaged by intense heat.  In the case
   history described above, the estimated maximum liner system temperature of 800°C
   was high enough to melt the geosynthetics  and desiccate the entire thickness of
   CCL.
•  Landfills should be operated to minimize the potential  for waste fires. Measures to
   be taken could include not depositing loads of hot waste in a landfill and covering
   waste with a soil cover to decrease waste access to oxygen.

A.3.3  L-12

Problem Classification:  landfill LCRS or LDS operation/design


                                     F-111
-------
Region of U. S.:  northeast
Waste Type:  MSW
Reference:  unpublished

Problem Summary:  leachate extraction well installed in landfill appeared to puncture
GM primary liner

Problem Description: A deep, 100-mm diameter leachate extraction well was installed
in a double-lined landfill after the LCRS appeared to be clogged.  The well design called
for the well  to extend into the sand LCRS drainage layer over the GM primary liner (i.e.,
extend within 0.3 m of the liner), but the elevation of the top of the borehole was not
surveyed immediately before well installation.  With waste settlement under its own
weight and  the weight of the drill rig, the target borehole depth may have been too deep.
Following well installation, average LDS flow rates increased from about 300 Iphd to 400
Iphd, and it was suspected that the well had penetrated the GM primary liner. The well
design called for the well to be installed within 0.3 m of the liner, but the elevation of the
top of the well boring was not surveyed immediately before well installation.
Considering waste settlement (under its own weight and the weight of the drill rig) since
the previous survey of the landfill, the target boring depth may have been too deep.

Resolution: No action has been taken with respect to the potential puncture of the GM
primary liner because it is not clear if the primary liner was actually punctured and the
LDS flow rates have remained relatively low. Environmental impacts from this possible
GM hole are expected to be negligible.

Lessons Learned for Future Projects:  Based on the available information, the following
lessons can be  learned from this case history:

•  Care should be taken to not damage the liner when drilling into landfilled waste.
   Settlement of the waste surface must be taken into  account when selecting the
   depth of drilling, and boreholes should not  extend close (e.g., within 1 m) to the liner.

F-A.3.4    L-14

Problem Classification: landfill liner degradation/construction
Region of U. S.:  northeast
Waste Type:  MSW
Reference:  unpublished

Problem Summary:  HOPE GM liner damaged  by fire believed to be started by lightning
strike

Problem Description: As the CQA consultant was completing paperwork in a trailer at a
landfill construction site, he observed black smoke coming from the landfill. At the time,
there was a thunderstorm,  and the contractor had left the site. When the CQA
                                      F-112
-------
consultant arrived at the landfill, he found rolls of GC drainage layer (HOPE GN and
polyester GT) on fire. The GC rolls had been unbagged and were lined up on the
installed HOPE GM liner at the top of the side slope, parallel to the slope crest, in
preparation for deployment the next day. The fire was attributed to lightning,  and
presumably propagated from one roll to another. Though there were no eyewitnesses
to this, the thunderstorm was large and was said to have more lightning strikes than any
previous storm in the area.
Figure F-A.3.1. GC rolls at the top of a liner system side slope caught fire
(potentially due to a lightning strike) and damaged the underlying GM.

The authors are not aware of other instances where a geosynthetics fire was started by
lightning. Part of this is likely due to the properties of the polymers themselves. Many
polymers do not burn easily and some are self-extinguishing.

Resolution: The fire was extinguished and the GM liner and GC rolls were inspected for
damage. GC  rolls that had caught on fire and GM that was melted, scorched, or rippled
was removed  and  replaced with new materials.

Lessons Learned for Future Projects:  Based on the available information, the following
lessons can be learned from this case history:
                                     F-113
-------
•  The igniting of geosynthetics by lightning is a rare occurrence.  The authors of this
   appendix are not aware of other instances of this.
•  It is not surprising that the fire spread after the GN ignited. The GN and GM were
   manufactured from polyethylene, a polymer that burns easily and is not self-
   extinguishing.

F-A.3.5  L-20

Problem Classification: landfill liner degradation/construction
Region of U. S.:  northeast
Waste Type:  MSW
Reference:  unpublished

Problem Summary: saturation of GCL beneath GM liner when rainwater ponded on
tack-seamed patch over GM hole

Problem Description: During construction of a  2.4 ha double-lined landfill cell, water
ponded over the GM/GCL composite primary liner on the base of the cell after a
significant storm (i.e., more than 50 mm). At one patch, which was tack seamed over a
GM hole, water flowed beneath the patch, through the hole, and beneath the GM liner.
About 1.4 ha of GCL beneath the GM became  saturated. Within this zone of saturation,
the GCL was swollen and soft. (Note: a tack seam is intended to hold a patch into place
prior to seaming; a tack-seamed patch is only bonded to the GM in some spots and is
not sealed at its periphery.)

Resolution:  The water was pumped out of the  cell, the GM liner was cut to expose the
swollen, saturated GCL, and the damaged GCL was removed. New GCL was placed,
and the GM primary liner was repaired.

Lessons Learned for Future Projects: Based on the  available information, the following
lessons can be learned from this case history:

•  To the extent practicable, holes in GM liners installed over GCLs should be repaired
   as soon as possible to avoid hydration of the GCL due to rainfall.  Holes located in
   areas where rainwater may pond should be patched first. The patches should be
   sealed with a permanent seam and not only tack-seamed.
•  When a GM is placed over a GCL, the GM  should be covered with soils as soon as
   possible to minimize swelling of the GCL in case  of hydration.

F-A.3.6  L-43

Problem Classification: landfill liner degradation/construction
Region of U. S.:  northeast
Waste Type:  HW
Reference:  unpublished


                                     F-114
-------
Problem Summary: water ponded between HOPE GM and CCL components of
composite secondary liner and was contaminated from a source other than the landfill

Problem Description: Construction of a 3.0 ha double-lined landfill cell was completed
in July 1993. The cell was constructed in an excavation through an approximately 14 m
deep glacial till layer and into an approximately 4 m thick clay layer that overlies
bedrock.  The shallow groundwater table is contained within the glacial till layer, and the
excavation had to be dewatered during construction.  The constructed cell has
approximately 12 m high, 3H:1V side slopes. The cell liner system on the base consists
of the following components from top to bottom:

•  0.3-m thick sand LCRS drainage layer;
•  HOPE GM/0.9-m thick CCL composite primary liner;
•  0.3-m thick sand layer LDS drainage layer; and
•  HOPE GM/3-m thick CCL composite  secondary liner.

On the side slopes, the liner system consists of the following components from top to
bottom:

•  sacrificial GT;
•  GC LCRS drainage  layer;
•  HOPE GM primary liner;
•  GN LDS drainage layer; and
•  HOPE GM/2-m thick CCL composite  secondary liner.

The clay used to construct the CCLs was classified as a CH material  in accordance with
the Unified Soil Classification System and had a liquid limit in the range of 29 to 31%
and a plasticity index in the range of 13 to 15%.

By September 1994, a large isolated bubble of water had developed between the GM
and CCL components of the composite secondary liner at the slope toe at the
southwest corner of the cell, which had not yet received waste.  The bubble was  located
along the southern exposure of the cell, which receives the most solar radiation
throughout the day, and extended about 3 m along the toe. Although the cell base by
the bubble was graded  inwards towards  the center of the cell, the bubble did not  extend
into the base liner system because of the overburden pressure provided by the soil
layers above the secondary liner. The water was pumped from the bubble, chemically
analyzed for organic constituents and metals, and found to be clean.  About 16,000 L of
water were pumped from the bubble, and the liner system perforations made on the
side slope to examine the bubble were repaired.

In the spring of 1995, a small water bubble had developed at the same location.  About
300 L of water were removed from the bubble, and the liner system was repaired.
Since the bubble was relatively small and the water removed the prior year from the

                                     F-115
-------
large bubble was clean, the water removed from the small bubble was not chemically
analyzed.

By June 1996, a water bubble had developed at the same location as the earlier
bubbles and a smaller bubble had developed at the slope toe about 2 m from the first
bubble.  By this time, about 70% of the cell had been filled with waste and the toe of the
intermediate waste slope was located about 3 m from the bubbles. About
4,000 L of water were pumped from the bubbles and chemically analyzed for organic
constituents and metals. While the metal concentrations of the bubble water were
consistent with those for clean water, the bubble water contained organic constituents,
including benzene. Subsequently, an extensive testing program was undertaken to
determine the source of the water and the contamination.

Resolution: A testing program was developed to evaluate the following potential
sources of the bubble water and/or the contamination:

•  leachate from the landfill;
•  groundwater;
•  surface water;
•  fuel from construction equipment; and
•  water used to construct the CCL.

Each of these sources is discussed below.

The source of the water and contamination is not leachate from the landfill. The landfill
leachate chemistry is different from the bubble water chemistry.  The leachate has
higher metal concentrations and more organic constituents than the bubble water. Also,
the benzene concentration of the bubble water  is higher than that of the landfill
leachate.  Also, no organic constituents have been detected in water from  the LDS of
the landfill, indicating that leachate has not migrated through the cell primary liner.

Groundwater does not appear to be the source of the bubble water or contamination.
The inorganic and organic chemistry of the shallow groundwater collected from three
wells installed in the glacial till layer upgradient of the bubbles is different from the
bubble water chemistry.  The groundwater does not contain organic constituents, so it
does not appear to be the source of the organic contamination. Furthermore, when the
upper 150-mm of the CCL at the bubble was tested for contamination, the highest
degree of contamination was detected in the top 50 mm, and no contamination was
found in the bottom 50 mm.  This suggests that the contamination is migrating outward
rather than inward with groundwater.

Surface-water runoff flowing over the CCL after the CCL was constructed may be a
source of the contamination.  The organic constituents found in the bubble liquid were
also historically present at a nearby former oil facility. However, the pathway for runoff
flow from the oil facility to the landfill has not yet been established. If runoff


                                     F-116
-------
contaminated the CCL during construction, the contamination was not widespread since
the liquid found in the initial bubble in 1994 was clean.  The CCL contamination may
have been upslope of the bubble area and may have migrated downslope into bubble
area over time as the CCL lost water due to thermal effects.

Another possible source of the bubble liquid contamination  is fuel from  construction
equipment. It may be that some fuel from construction equipment or a generator
dripped on the CCL upslope of the bubble area during construction or during the repair
of the liner system after the first bubble was found.  The CCL contamination may have
migrated downslope into bubble area over time as the CCL lost water due to thermal
effects.

The source of the water in the bubble appears to be from the CCL. As waste was
placed on the CCL, the CCL consolidated under the weight of the overlying soil layers
and waste, squeezing water from the CCL.  Some of this water may have flowed
between the GM and CCL along wrinkles to southeast and southwest corners of the
cell.  Very large 1 m  high wrinkles were observed in the GM liner on the cell base at the
slope toe prior to placement of the LDS drainage layer.  In addition, since the CCL
component of the secondary liner on the side slope was not thermally insulated by an
overlying soil layer, the CCL may have lost water due to thermal effects.  Furthermore,
the inorganic chemistry of the LDS liquid and the bubble water are similar, suggesting
that source of the bubble water,  like the LDS liquid,  was the CCL.  The consolidation
characteristics and pore water chemistry of the CCL component of the  composite
secondary liner are currently being investigated so that more definitive  conclusions on
the source of the bubble water can be drawn.

The bubble water was tested in September 1996 and found to be clean. The liner
system will be repaired at the conclusion of the investigation.

Lessons Learned for Future Projects'. This case history is complex, and there are
several conditions that may have contributed to the development of bubbles of
contaminated water at the slope toe. Based on the available information, the following
lessons can be learned from this case history:

•  Composite liners constructed with a CCL should be covered to prevent heating and
   desiccation of the CCL, which can lead to the ponding of water between the CCL
   and overlying GM. EPA cautions that temporarily using a GM alone over a CCL may
   be problematic, and that a light-colored GM may be necessary (Daniel and Koerner,
   1993).
•  GMs should be covered with soil  layers as soon as practicable after installation, but
   not during the hottest part of the day if the GM is significantly wrinkled, to reduce GM
   wrinkles.
•  During construction of liner systems, runon should be controlled so  that it does not
   contact, and potentially contaminate, the liner.
                                     F-117
-------
•  Construction equipment should be inspected for fuel and oil leaks, and those leaks
   should be repaired prior to using the equipment in liner construction to avoid liner
   contamination.

F-A.3.7   L-44

Problem Classification:  landfill liner degradation/design
Region of U. S.:  northcentral
Waste Type: MSW
Reference: unpublished

Problem Summary: landfill gas well punctured GM component of composite liner

Problem Description:  During installation of gas extraction wells in a active landfill, one
of the 0.9-m diameter boreholes for the wells was advanced into the composite liner. At
the location of this localized damage, the top of the liner is approximately 18 m below
the waste surface. The problem was identified when portions of the liner system were
observed in the cuttings from the bucket auger. Based on observations by field
personnel, the bucket auger extended through the 0.6-m thick soil protection layer, GC
LCRS drainage layer, and 2.0-mm thick HOPE GM liner and into the upper 0.15-m of
the 0.9-m thick CCL.  Upon observing these components in the auger cuttings, field
personnel poured bentonite pellets into the bottom of the borehole to create an
approximately 0.9-m thick bentonite seal at the  borehole base.  The top of the borehole
was then temporarily covered with plywood until the potential environmental impact of
the liner damage and the need for additional liner remediation could be evaluated.
When the cause of the problem was investigated, it was discovered that a typographic
error had been made on the design drawing for the gas extraction system: the specified
borehole depth at the location of the liner damage was greater than the depth to the top
of the liner.

Resolution: An evaluation of the  liner damage found the potential environmental impact
from the damage to be negligible. The damaged liner is located in an upgradient
portion of the cell, and the portion of the LCRS that might drain to this area is very
small. Due to the small drainage area  and the high transmissivity of the GC LCRS
drainage layer, the leachate head on the damaged liner should be small. Also, the
damaged area has been filled with low-permeability bentonite and is underlain by a
relatively thick natural low-permeability soil layer.  Irrespective of this, the landfill owner
has proposed to repair the liner system. The proposed remedy for this involves
advancing a 3-m  diameter steel access shaft to the top of the protection layer over the
damaged liner, excavating the waste in the shaft, exposing the damaged portion of the
liner system, and repairing each damaged liner system component. This remedy has
not yet been implemented.

Lessons Learned for Future Projects:  Based on the available information, the following
lessons can be learned from this  case  history:
                                      F-118
-------
•  Care should be taken to not damage the liner system components when drilling into
   landfilled waste. To prevent damage, boreholes should not extend close to the liner.

F-A.4     Landfill LCRS or LDS Construction

F-A.4.1   L-10

Problem Classification: landfill LCRS or LDS construction/construction
Region of U. S.:  northeast
Waste Type:  MSW
Reference: unpublished

Problem Summary: rainwater entered LDS through anchor trench

Problem Description:  Flow rates from the LDS of a newly constructed landfill cell
ponded with rainwater were higher than expected, on the order of 500 Iphd. The
double-liner system for the cell includes a GM primary liner on the side slope,  GM/GCL
composite primary liner on the base, and GC LDS drainage layer on the side slope and
base. With these components, the LDS flow from the cell is primarily attributed to
leakage through the GM primary liner on the side slope.  An electrical leak location
survey and a gas tracer leak location survey were performed to locate GM primary liner
holes on the base and side slope of the cell, respectively. Several small GM punctures
and an approximately 1.1-m long tear were found on the side slopes near the  toe. The
tear was located under a rubsheet placed beneath the gravel side slope toe drain. The
CCL beneath the tear had a 25 to 50 mm deep depression. The large hole is believed
to have been caused by a bulldozer blade during construction of the gravel toe drain.
The holes were repaired; however, the LDS flow rate did not show significant decrease.
Subsequently, the liner system anchor trench at the top of the approximately 9-m high
and 4.6-m wide perimeter berm was inspected and found to be full of water. The
anchor trench had been backfilled with the sandy site soil, in accordance with  the plans.
This  soil allowed significant water to infiltrate and pond on the geosynthetics, including
the GC LDS drainage layer.  The GC was conveying water that had infiltrated  the
anchor trench into the LDS.

Resolution:  Sections of the back of the anchor trench were excavated to the outer
slope of the perimeter berm at 8  m intervals along the length of the trench, and the ends
of the geosynthetics in the trench sections were laid horizontal. The perimeter berm
was reconstructed to grade with gravel. This allows water infiltrating the trench to drain
to the outside slope of the perimeter berm. To minimize infiltration of rainwater into the
anchor trench, a GM was placed over the top of the berm and covered with a 0.3-m
thick layer of site soil. After construction of this redesign was complete, the LDS flow
rate decreased to less than 100 Iphd.
                                      F-119
-------
Lessons Learned for Future Projects: Based on the available information, the following
lesson can be learned from this case history:

•  Geosynthetic anchor trenches should be backfilled with low-permeability soil and the
   soil should be well compacted. If this is not practicable, the anchor trenches should
   be designed to drain freely and/or covered with a barrier,  such as the GM used in
   the case history described above. In addition, the ground surface should be graded
   away from the trenches to reduce runon from infiltrating into the trenches.

F-A.4.2   L-15

Problem Classification:  landfill LCRS or LDS construction/construction
Region of U. S.:  southeast
Waste Type: MSW
Reference:  unpublished

Project Description: sand bags in LCRS drainage layer and  debris in LCRS pipe trench
approved by CQA consultant

Summary: During construction of a landfill liner system, the contractor used sand bags
to secure the GM liner. The sand used to fill the bags met the project specifications for
the LCRS drainage layer material. After GM installation was completed, the contractor
placed the sand LCRS drainage layer over the sand bags. The contractor also
occasionally disposed of debris, such as food waste and aluminum  cans, in the LCRS
pipe trench gravel.  The LCRS was inspected  by site CQA personnel and found to be
acceptable.

Resolution: Regulatory personnel who inspected the site required the sand bags and
debris to be removed from the LCRS.

Lessons Learned for Future Projects: Based on the available information, the following
lesson can be learned from this case history:

•  CQA personnel should be trained in standard CQA practices,  such as keeping the
   LCRS free of items and debris that may potentially impede the flow of leachate.  In
   the case history described above, the sand in the sand bags met the project
   specifications for the LCRS drainage layer material.  However, the sand was
   wrapped in a woven bag that may impede flow.  In addition, there were numerous
   bags in the LCRS.  While leachate in the LCRS drainage layer can easily flow
   around one bag, flow may be impeded if there are numerous bags.  Consequently,  it
   is good practice to keep sand bags out of the LCRS. Alternatively, if the sand in the
   bags meets the project specifications for the overlying drainage  layer material, the
   bags can be cut and the sand left in place.
                                     F-120
-------
F-A.4.3   L-16

Problem Classification: landfill LCRS or LDS construction/construction
Region of U. S.:  southcentral
Waste Type:  HW
Reference: Bonaparte, R. and Gross, B.A., "LDCRS Flow Rate from Double-Lined
    Landfills and Surface Impoundments", EPA Risk Reduction Research Laboratory,
    Cincinnati, OH, EPA/600/SR-93/070, 1993, 65 p. (Landfill cells T-7 and T-8)

Problem Summary: rainwater entered LDS through anchor trench

Problem Description:  About two years after construction, the LDS flow rates from two
double-lined cells increased from about 30 and 120 Iphd to 220 and 840 Iphd,
respectively. The side slope liner system for the cells includes GN LCRS and LDS
drainage layers and a GM primary liner. The base liner system includes a gravel LCRS
drainage layer, GN LDS drainage layer,  and GM/CCL composite primary liner.  Based
on the liner system components and the results of chemical analyses of LDS flow, water
in addition to that squeezed from the CCL as it compresses appeared to be entering the
LDS.  The anchor trench for the cells was inspected and found to contain ponded water.
The anchor trench was not initially well compacted.  Over time, the anchor trench soil
settled and a depression developed over the anchor trench.  The depression trapped
runoff, which subsequently infiltrated into the trench. The GN LDS drainage layer in the
anchor trench was conveying this water into the LDS.

Resolution: The anchor trench soil was removed, and the trench was allowed to dry.
The soil was recompacted into the trench and graded to drain away from the trench.
Shortly afterwards, LDS flow rates for the two cells decreased to 70 and 120 Iphd,
respectively.

Lessons Learned for Future Projects: Based on the available information, the following
lesson can be learned from this case history:

•  Geosynthetic anchor trenches should be backfilled with low-permeability soil and the
   soil should be well compacted. If this is not practicable, the anchor trenches should
   be designed to drain freely and/or covered with a barrier. In addition,  the ground
   surface should be graded away from the trenches to reduce runon from infiltrating
   into the trenches.

F-A.4.4   L-28

Problem Classification:  landfill LCRS or LDS construction/construction
Region of U. S.:  southeast
Waste Type:  MSW
Reference: unpublished
                                     F-121
-------
Project Summary, excessive needle fragments in manufactured needlepunched
nonwoven GT

Problem Description:  During construction of an 8 ha single-composite lined landfill cell
in 1993, numerous broken needles were found in the needlepunched nonwoven GT
filter placed over the GN LCRS drainage layer. The quantity of needle fragments was
abnormal (i.e., about 150/ha) and resulted from a GT production problem at the
manufacturing plant. Though the project specification for the GT did not require the
manufacturer to check for needles, it was the state of practice for manufacturers at the
time.  By the time the needle problem was discovered by the CQA consultant, about 5
ha of GT had been deployed. About 3 ha of this GT had already been covered by a 0.3-
m thick soil protection layer.

Resolution: The contractor initially tried to fix the GT that had been deployed and not
covered with soil by manually searching for and removing needle fragments. However,
most of this GT was replaced with GT from  a different manufacturer because the needle
fragment density was very high. The contractor also tried to locate broken needles  in
the GT beneath the soil layer using a metal detector. However, this method of needle
fragment detection proved to be unreliable.  At this point the contractor was unsure how
to proceed: both removing the soil to expose the defective GT and leaving the defective
GT in place could potentially result in GM damage.  To evaluate the potential for needle
fragments to extend through  the GN  LCRS  and puncture the underlying 1.5-mm thick
HOPE GM, laboratory tests were conducted to evaluate GM puncture by needle
fragments of different lengths and  orientations. Of the ten laboratory tests with 192
needle fragments, only one needle fragment punctured the GM. These results were
used, along with the probability distribution of needle fragment sizes and orientations
observed in a sample of the  GT, to estimate the probability of GM puncture.  Based on
the results of the analysis, the expected GM holes caused by the needle fragments
would be very small and would occur infrequently. The defective GT was left in place
beneath the soil layer because it appeared that there was less potential for GM damage
by leaving the GT in place than by excavating the overlying soil to remove the GT. The
manufacturer of the defective GT installed magnets in the manufacturing plant to
remove broken needles from GTs  produced in  the future.

Lessons Learned for Future  Projects: Based on the available information, the following
lessons can be learned from  this case history:

•  Project specifications for  needlepunched nonwoven GTs should require that the GTs
   be needle-free and should require a certification from the manufacturer attesting to
   this.
•  The CQA Plan should require that deployed GTs near GMs be inspected for needles
   before the GTs are covered with overlying materials.
•  Even with excessive needles in the GT,  if the  GM is separated from the GT by a GN,
   few needles  are expected to puncture the GM.
                                     F-122
-------
Figure F-A.4.1. Manually searching for needle fragments in an installed GT.
Figure F-A.4.2. Many needle fragments were found in the deployed GT.
                                   F-123
-------
F-A.4.5   L-32

Problem Classification:  landfill LCRS or LDS construction/construction
Region of U. S.:  southeast
Waste Type:  MSW
Reference: unpublished

Problem Summary: HOPE  LCRS pipe separated at joints

Problem Description:  The condition of the LCRS pipes at an active 19 ha single-lined
landfill cell is surveyed for clogging and damage on an annual basis by running a small
video camera through the pipes.  The pipes are 200-mm diameter smooth wall HOPE.
The pipes were supplied in  6.1 m lengths and connected in the field by fusion seaming
the ends together. The  CQA of the pipe installation consisted primarily of visual
inspection.  The pipes are bedded in gravel wrapped with a GT filter.

During the initial video conducted after the landfill was constructed, several pipe joints
were found to be separated less than 10 mm and one joint appeared to be crushed over
a length of about 0.6 m.  Two subsequent annual surveys have revealed no further
separation in the pipe joints or pipe crushing over time.  The reason for this pipe
separation is unclear. It may be that the pipes were never seamed together during
construction or that the quality of some of the pipe seams was so poor that the seams
failed during construction. The crushed pipe may have been damaged by equipment
trafficking over it during  construction or operation.

Resolution: No action has been required by the regulatory agency presumably
because: (i) leachate flowing out of the pipe at an open pipe joint can still flow to the
leachate sump (though the  localized head at the open joint may be somewhat higher
that those upgradient and downgradient of the open joint); (ii) the pipe condition has
remained unchanged during subsequent annual videos; (iii) repair of LCRS pipes after
waste placement would  be  extremely difficult and expensive; and (iv) environmental
impacts from having a localized higher head on the liner at the open pipe joints are
expected to be negligible. The pipes will continue to be surveyed on an annual basis.

Lessons Learned for Future Projects:  Based on the available information, the following
lesson can  be learned from this case history:

•  The CQA consultant should verify that all connections required for adjacent LCRS
   and LDS pipes have been made. When the pipe is connected by butt fusion
   seaming, the seam should be inspected for holes.

There is little available information for this case history; additional lessons might have
been learned  if the information had been complete.
                                     F-124
-------
F-A.4.6   L-33

Problem Classification:  landfill LCRS or LDS construction/construction
Region of U. S.:  southeast
Waste Type:  MSW
Reference: unpublished

Problem Summary: HOPE LCRS pipe separated at joints

Problem Description:  The condition of the LCRS pipes at an active 32 ha single-lined
landfill cell is surveyed for clogging  and damage on an annual basis by running a small
video camera through the pipes.  The pipes are 200-mm diameter smooth wall HOPE.
The pipes were supplied in 6.1 m lengths and connected in the field by fusion seaming
the ends together. The  CQA of the pipe installation consisted primarily of visual
inspection.

During the initial video conducted after the landfill was constructed, about 10 pipe joints
were found to be separated less than 10 mm. Subsequent annual surveys have
revealed no additional pipe joint separations over time.  The reason for the separation of
some pipe segments is  unclear.  It may be that the quality of some of the pipe seams
was so poor that the seams failed during construction.

Resolution: No action has been required by the regulatory agency presumably
because: (i) leachate flowing out of the pipe at an open pipe joint can still flow to the
leachate sump (though the localized head at the open joint may be somewhat higher
that those upgradient and downgradient of the open joint); (ii) the pipe condition has
remained unchanged during subsequent annual videos; (iii) repair of LCRS pipes after
waste placement would  be extremely difficult and expensive; and (iv) environmental
impacts from having a localized higher head on the liner at the open pipe joints are
expected to be negligible. The pipes will continue to be surveyed on an annual basis.

Lessons Learned for Future Projects: Based on the available information, the following
lesson can be learned from this case history:

•  The CQA consultant should verify that all connections required for adjacent LCRS
   and LDS pipes have been made. When the pipe is connected by butt fusion
   seaming, the seam should be inspected for holes.

There is little available information for this case history; additional lessons might have
been learned  if the information had  been complete.
                                     F-125
-------
F-A.5     Landfill LCRS or LDS Degradation

F-A.5.1   LA

Problem Classification:  landfill LCRS or LDS degradation/design
Region of U. S.:  northeast
Waste Type:  MSW
Reference: unpublished

Problem Summary: erosion of sand LCRS drainage layer on liner system side slopes

Problem Description:  Portions of the 0.6-m thick sand LCRS drainage layer on the
approximately 100-m  long, 41-1:1 V side slopes of this landfill were progressively eroded
by rain.  The sand has a specified minimum hydraulic conductivity of 1 x 10~4 m/s and
maximum  fines content of 5%. Based on the particle size analyses conducted on the
sand as part of CQA conformance testing,  the areas with the most erosion tended to
have been covered with sand containing the most fines (i.e., almost 5%). The erosion
has resulted in 0.3-m  wide gullies that propagate from the toe of the side slopes
upward. Sand has also washed into the exposed gravel around the LCRS pipes and in
the sump area.  By one year after construction, the sand had been pushed back up the
side slopes with low ground-pressure bulldozers more than six times and the gravel in
the sump area of two  cells that had not yet received waste had been replaced twice. It
is anticipated that the two cells will not receive waste for at least another two years.

Resolution: A plastic  tarp has been purchased to place over the sand on the
approximately 2.4 ha  of side slopes in the two inactive cells. However, the tarp has not
yet been installed because of concerns with how to anchor the tarp and protect it from
uplift by wind. In the meantime, the owner is considering other option: covering the
sand with yard compost. Waste has been  placed on the side slope of the one active
cell, and protection of the sand is not needed.

Lessons Learned for Future Projects:  Based on the available information, the following
lessons can be learned from this case history:

•  Erosion of soil layers on liner system side slopes should be anticipated and dealt
   with in  design. In areas where the potential for erosion is relatively high, erosion
   control structures (e.g., runoff diversion berms, silt fence) can be used to reduce the
   need for intensive  maintenance of the soil layers. Alternatively, the soil layers can
   be covered with a  tarp or temporary erosion control mat.
•  Better methods for protecting exposed soil layers on liner system side slopes or
   alternatives to these soil  layers are needed.
•  Though it may be  less costly for the owner to construct several landfill cells at once,
   this can leave new cells exposed to the environment for a significant time period.
   These  cells will experience more erosion than  cells filled sooner and will have more
   opportunity for liner damage. Additionally, every time an eroded soil layer is pushed
                                      F-126
-------
   back up the side slopes there is an opportunity for the underlying liner system
   materials to be damaged by construction equipment.
•  Post-construction plans should be developed for portions of landfills that may sit idle
   for an extended period of time.  The plans should include procedures describing how
   the liner system should be maintained prior to operation.

F-A.5.2   L-11

Problem Classification: landfill LCRS or LDS degradation/design
Region of U. S.:  northeast
Waste Type:  MSWash
Reference: unpublished

Problem Summary:  erosion of sand protection layer on liner system side slopes

Problem Description: Portions of the 0.45-m thick sand protection layer on the side
slopes of a landfill liner system were progressively eroded by runon and runoff. The
side slopes of the landfill were constructed over an existing MSW landfill; the base
slopes were constructed on natural ground. With each significant rainfall, the sand on
Figure F-A.5.1. Erosion of sand protection layer on liner system sideslopes.
                                      F-127
-------
the side slopes was eroded by runon from the adjacent existing MSW landfill cover
system and runoff at locations on the slopes where flow was concentrated.  Runon from
the existing landfill occurred because the contractor had not completed or maintained
the surface-water drainage system. Runon from the existing landfill also carried topsoil
onto the sand protection layer.  The 2.5H:1V cell side slopes are broken into
approximately 18 m long segments by benches that slope into the landfill. Water
collected on a bench primarily infiltrates through the sand protection layer to the GC
LCRS drainage layer.  However, because the sand hydraulic conductivity is not high
enough to allow all of the water to infiltrate (i.e., the specified minimum hydraulic
conductivity is 1 x10~5 mis), some of the water also flows across the bench and along
the bench to a low point. At this low point, concentrated runoff flows across the bench
and down a side slope, and erosion of the sand was most pronounced. In two areas
where the protection layer had eroded, the underlying GC LCRS drainage layer, GM
primary liner,  GN LDS drainage layer, and GM component of the composite secondary
liner were uplifted by landfill gases.

Resolution: The remedy developed required: (i) control of runon from the adjacent
MSW landfill;  and (ii) control of runoff on the side slopes of the landfill under
construction.  To control runon from the adjacent MSW landfill, small diversion berms
were constructed around the landfill side slopes. In addition, hay bales, silt  fence, and
erosion mat were used to control erosion of cover system soils and subsequent
sedimentation and overtopping of runon-control swales. To manage runoff,  a riprap-
lined downchute was constructed over the sand at the low area on the side  slopes
where concentrated runoff occurred.  The sand protection layer was restored on the
side  slopes.

Lessons Learned for Future Projects:  Based on the available information, the following
lessons can be learned from this case history:

•  Erosion of soil protection layers on liner system side slopes should be anticipated
   and dealt with in design. The potential for erosion can be reduced by grading the
   liner system to avoid concentrated runoff and using a relatively permeable soil in the
   protection layer.  In areas where the potential for erosion is relatively high, erosion
   control structures (e.g.,  runoff diversion berms, silt fence) can be used to reduce the
   need for intensive maintenance of soil protection layers. Alternatively, protection
   layers can be covered with a tarp or temporary erosion control mat.
•  Runon into active waste containment systems from adjacent areas must be
   controlled.  In fact, this is a regulatory requirement for MSW landfills and HW landfills
   and impoundments.
•  When a landfill is constructed on top of an existing landfill (vertical expansion),
   exposed GM liners can  be uplifted by gases from the underlying  landfill.  Therefore,
   in the case of a vertical  expansion, unless gases from the underlying landfill are well
   controlled, GMs must be covered by a layer of soil to prevent GM uplift and
   precautions must be taken to prevent erosion of this soil layer.
                                      F-128
-------
F-A.5.3   L-13

Problem Classification: landfill LCRS or LDS degradation/construction
Region of U. S.:  southeast
Waste Type:  MSW
Reference: Tisinger, L.G., Clark, B.S., Giroud, J.P., and Christopher,  B.R., "Analysis of
    an Exposed Polypropylene Geotextile", Proceedings of Geosynthetics '93,
    Vancouver,  British Columbia, 1993, pp. 757-771.
    Tisinger, L.G., Clark, B.S., Giroud, J.P., and Schauer, D.A., "Performance of
    Nonwoven Geotextiles Exposed to a Semi-Tropical Environment", Proceedings of
    Fifth International Conference on Geotextiles, Geomembranes, and Related
    Products, Singapore, 1994,  pp. 1223-1226.

Problem Summary: polypropylene continuous filament nonwoven GT filter degraded
due to outdoor exposure

Problem Description: Tisinger et al. (1993, 1994) described the degradation of an
exposed 540 g/m2 polypropylene continuous filament nonwoven GT filter over a GN
LCRS drainage  layer. The site is located in a semi-tropical environment with high
ambient temperatures up to 38°C, humidity, sun, wind, and rain. The design called for
the GT to be covered with a sand protection layer just before waste placement. This
required the GT to be exposed for at least several months. This strategy was selected
because the local sand proposed for the protection layer was very erodable and would
require significant maintenance  if left exposed.  The manufacturer's recommendation for
maximum outdoor exposure time of the GT was 500 hours (20 days).  Though a 270
g/m2 GT met the project specifications, a heavier 540 g/m2 GT was selected,
anticipating that this GT would retain enough  strength after several months of exposure
to meet the specifications.

During construction, it became apparent that waste placement would  be delayed and
the GT exposure would be on the order of six months or  more.  Because some
deterioration was expected until the sand protection layer could be placed, samples of
the GT were periodically tested to verify that the strength properties of the GT still met
specifications. Samples tested after 4.5 months of exposure exhibited no significant
changes in properties.  However, after 6.5 months of exposure,  significant degradation
of the GT was found. Grab, tear, and puncture strengths had decreased by 22.9, 34.0,
and 24.1%, respectively; the change in burst strength was insignificant. Even with this
degradation, the GT still met the specifications and approval was given to start
placement of the sand protection layer.  A few days later, before sand was placed, holes
were observed in the GT in two  areas near the side slope crest of one of the perimeter
berms. The holes ranged in size from 20 to 200 mm. No holes were observed on the
side slopes or base. GT samples collected at this time had grab, tear, puncture, and
burst strengths that were 62.6, 64.9, 48.2 and 57.4%, respectively, less than their pre-
exposure values. The rate of GT degradation had increased substantially, and the GT
burst strength of 1,793 kPa did not meet the specified value of 2,000 kPa.  Based on
                                      F-129
-------
differential scanning calorimetry, infrared spectrophotometry, and microstructural
analyses of the GT, GT degradation was attributed to heat and ultraviolet radiation. All
possible mechanisms of hole formation were reviewed and it was concluded that the
holes had probably developed in the degraded GT due to fiber breakage, removal, and
abrasion  by wind action since all the holes were on the perimeter berm exposed to the
prevailing winds.
Figure F-A.5.2. Holes developed in a polypropylene GT at the side slope crest
after 6.5 months of exposure to the environment.

Interestingly, a 540 g/m2 polyester continuous filament nonwoven GT filter was
substituted for the polypropylene GT in part of the landfill. While the mechanical
properties of the polyester GT decreased with time, the rate of degradation was slower
than that for the polypropylene GT and appeared to  be decreasing with time. After 14.5
months of exposure, samples of the polyester GT had grab, puncture, and burst
strengths that were 30.3, 20.1, and 13.0%, respectively, less than their pre-exposure
values and a tear strength that was 24.2% greater than its pre-exposure value.

Resolution: The heavily degraded polypropylene GT was replaced with a 270  g/m2
polypropylene GT and covered with a sand protection layer soon after installation.
                                      F-130
-------
Lessons Learned for Future Projects:  Based on the available information, the following
lessons can be learned from this case history:

•  GTs should be covered as soon as possible after installation to protect them from
   the environment.
•  If a GT is to be exposed to the environment for an extended time period after
   installation, a GT that initially far exceeds the project specifications and will meet the
   specifications after some degradation can be selected. As shown in the case history
   presented above, some types of GTs perform better than others. The potential
   degradation of the selected GT should be evaluated under all the anticipated
   environmental conditions. EPA recommends that the  effect of ultraviolet light on GT
   properties be evaluated using ASTM D 4355 (Daniel and Koerner, 1993).  The test is
   typically run for 500 hours; however, it can be run for longer time periods to meet
   project-specific conditions. In any case,  prior to covering the GT, the condition of
   samples of the exposed GT taken from the field should be evaluated by laboratory
   testing to verify that the exposed GT is still satisfactory.
•  If test results indicate that the GT will not have the required properties after exposure
   (typically a specified strength retention),  the GT should be protected with a sacrificial
   opaque waterproof plastic tarp, soil layer, or other means. Tisinger et al. (1993)
   suggest that this may be the best strategy since a heavy degraded GT that meets
   the specifications is more sensitive to stress concentrations than a new lighter GT
   that meets the same specifications.

F-A.5.4   L-18

Problem Classification: landfill LCRS or LDS degradation/construction
Region of U. S.:  southcentral
Waste Type:  remediation waste
Reference: Paulson, J.N., "Veneer Stability Case Histories: Design  Interactions
    Between Manufacturer/Consultant/Owner", Proceedings of the  7th GRI Seminar
    Geosynthetics Liner Systems: Innovations, Concerns, and Designs, 1993, pp. 235-
    241.

Problem Summary:  polypropylene staple-fiber needlepunched nonwoven GT filter
degraded due to outdoor exposure

Problem Description: Paulson (1993) described the degradation of the 350 g/m2
polypropylene staple-fiber needlepunched nonwoven GT  component of a GC  LCRS
drainage layer.  On the base of the landfill cell, the GC  is  overlain by a soil protection
layer. On the side slopes, the GC was  initially exposed;  a soil protection layer was to
be placed incrementally over the GC on the side slopes during filling operations.
Regulatory approval to place  waste in the cell was not received on schedule after the
cell was constructed, leaving  the GC exposed to the environment.  By about one year
after construction, the GT component of the GC was falling apart, exposing the GN and
underlying GM primary liner.  Samples of the GT were collected for strength testing, but
                                      F-131
-------
could not be tested due to the amount of GT degradation.  The GT also had a strong
rotten egg-like odor. Paulson noted that the site is locally in a heavy industrial area
known locally as having acidic precipitation. He attributed the odor to the fallout of
industrial emissions that generated sulfuric acid on the GT.  The exposure of the GT to
ultraviolet light, sulfuric acid from industrial emissions, water, and high ambient
temperature caused its severe degradation.

Resolution: The GC LCRS drainage layer on the side slopes was replaced, and waste
placement in the cell began soon afterwards.

Lessons Learned for Future Projects:  Based on the available information, the following
lessons can be learned from this case history:

•  GTs and GCs should be covered as soon as possible after installation to protect
   them from the environment.
•  If a GT is to be exposed for an extended time  period after installation, the potential
   degradation of the GT should be evaluated under all the anticipated  environmental
   conditions. EPA recommends that the effect of ultraviolet light on GT properties be
   evaluated using ASTM D 4355 (Daniel and Koerner, 1993). The test is  typically run
   for 500 hours; however, it can be run for longer time periods to meet project-specific
   conditions. In any case, prior to covering the GT, the condition of samples of the
   exposed GT taken from the field should be evaluated by laboratory testing to verify
   that the exposed GT is still satisfactory.
•  If test results indicate that the GT will not have the required properties after exposure
   (typically a specified strength retention),  the GT should be protected with a sacrificial
   opaque waterproof plastic tarp, soil layer, or other means.

F-A.5.5   L-30

Problem Classification:  landfill LCRS or LDS degradation/construction
Problem Cause: operation
Waste Type:  MSW
Reference: unpublished

Problem Summary:  HOPE LCRS pipe crushed during construction

Problem Description: The valve on the 150-mm diameter HOPE LCRS  pipe draining a
newly constructed cell was kept closed until just before the start of waste placement.
During this time, a significant amount of water (i.e., more than one meter deep) ponded
in the cell.  When the valve on the pipe was opened so water could drain, drainage
occurred only very slowly.  With no other on-site location to dispose of waste, the baled
waste was placed in the ponded water. Processed C&DWwas placed over the bales to
keep the bales from floating.  The crushed condition of the pipe was only identified
when an attempt was made to flush the pipe to increase the water flow rate from the
cell.  The location of the pipe  damage relative to the landfill  cell was not evaluated.
                                      F-132
-------
The processed C&DW contained relatively high concentrations of sulfate. As the waste
decomposed, the sulfate was reduced to hydrogen sulfide gas. The hydrogen sulfide
concentrations were very high, about 2,200 ppm, in the air at the waste surface.  Offsite,
hydrogen sulfide concentrations were about 2 ppm, and the air had a rotten egg smell.

Resolution: The LCRS pipe was buried under waste and water and was not repaired
since it still allowed water to drain, albeit slower than as designed. To control the
hydrogen sulfide gas through chemical reaction, hydrogen peroxide was pumped into
the waste. However, significant hydrogen sulfide was still formed. Due to the gas
problem, the landfill was  closed early, after only 1.5 years of filling. A gas extraction
system with a flare was installed in the landfill, and gas emissions from the facility are
successfully being controlled.
Lessons Learned for Future Projects: Based on the available information, the following
lessons can be learned from this case history:

•  Care should be taken to not damage leachate pipes during construction. The
   contractor should maintain sufficient soil cover between construction equipment and
   the liner system during construction. Equipment operators should be aware of pipe
   locations, since pipes can be crushed by trafficking equipment. Also, soil around
   pipes should be compacted using hand operated or walk-behind compaction
   equipment.
•  After construction of a cell with an external sump, the pipe from the cell to the sump
   should be inspected to verify that the pipe is functioning as designed. The
   inspection may be performed by surveying the pipe with a video camera, pulling a
   mandrel through  the pipe, flushing the pipe with water, or other means.

F-A.6     Landfill LCRS or LDS Malfunction

F-A.6.1   L-12

Problem Classification: landfill LCRS or LDS malfunction/operation
Region of U. S.:  northeast
Waste Type:  MSW
Reference: unpublished

Problem Summary:  LCRS pipes were not regularly cleaned and became partially
clogged and LCRS drainage layer may be partially clogged

Problem Description: As waste was placed into a double-lined landfill and then covered
with soil intermediate cover, the LCRS flow rates from the landfill decreased. The LDS
flow rates, however, increased from less than 100 Iphd to about 300 Iphd.  Because
LDS flow rates did not decrease with decreasing LCRS flow rates and LDS flow rates
                                      F-133
-------
were higher than those typical of nearly filled landfills in that region of the country, the
LCRS was believed to be partially clogged.  The LCRS drainage layer is a sand with a
specified minimum hydraulic conductivity of 1 x 10~5 m/s. The LCRS and LDS pipes in
the landfill were not flushed annually as is common practice in the region. Rather than
performing maintenance on the LCRS pipes, the landfill owner decided to install a deep,
100-mm diameter leachate extraction well in the landfill.  After the well was installed
through the waste, LDS flow rates increased to about 400 Iphd, and it was suspected
that the well had penetrated the GM primary liner.

Resolution:  The LCRS pipes were cleaned out and are scheduled to be flushed
annually. Insufficient time has past to determine if cleaning the pipes solved the
problem.
Lessons Learned for Future Projects'.  Based on the available information, the following
lessons can be learned from this case history:

•  LCRS and LDS pipes should be maintained by cleaning the pipes at least annually
   and more frequently, if warranted.
•  Landfills with external sumps could also include riser pipes at the low point of
   leachate collection systems as a precautionary measure to allow for leachate
   removal from the landfill,  if necessary.

F-A.6.2  L-22

Problem Classification: landfill LCRS or LDS malfunction/design
Region of U. S.:  northeast
Waste Type:  industrial
Reference: Koerner, G.R., Koerner, R.M., and Martin, J.P., "Field Performance of
    Leachate Collections Systems and Design Implications",  Proceedings of 31st
   Annual SWANA  Conference, San Jose, California, 1993, pp. 365-380.

Problem Summary: waste  fines clogged needlepunched nonwoven GT filter wrapped
around perforated LCRS pipes

Problem Description: Koerner et al. (1993) described the clogging of a 540 g/m2
needlepunched nonwoven GT filter wrapped around LCRS pipes perforated with about
20 13-mm diameter holes/m.  The pipes are bedded within a 0.3-m thick pea gravel
LCRS drainage layer. The gravel is overlain by the same type of GT as that used to
wrap the pipe and then a 0.3-m thick sand protection layer. The apparent opening size
(AOS) of the GTs is 0.19 mm. The landfill was used for disposal of industrial plant
waste, lime-stabilized waste, and slurried fines. About 75% by weight of the slurried
fines particles pass the 0.15 mm sieve and about 45% pass the 0.074 mm sieve.
Koerner et al. (1993) did not indicate  if GT filter design was performed as part of the
LCRS design.
                                      F-134
-------
By about one year after construction, it was apparent that the LCRS was not functioning
adequately.  Rainwater ponds developed on the waste surface, grew with time, and
required pumping to remove them.  In addition, the amount of leachate removed from
the LCRS sump was less than expected. The LCRS was excavated near the sump and
the following observations were made:

•  the LCRS gravel drainage layer was relatively clean and full of leachate;
•  based on piezometer measurements, the leachate in the gravel drainage layer near
   the sump was under pressure indicating it was confined below the waste and not
   able to freely drain into the pipes;
•  the GT wrapping the LCRS pipes was clogged at the pipe perforations; and
•  the GT between the sand protection and gravel drainage layers (upper GT) was not
   clogged.

To evaluate the effect of clogging on the transport of leachate though the  GT,
permittivity tests were conducted on the following samples: (i) uncleaned upper GT; (ii)
cleaned upper  GT; and (iii) upper GT conditioned in the laboratory with site-specific
slurried fines for six months to model the GT around the pipe.  The cleaned upper GT
had a permittivity of about 1.8 s"1.  The permittivities of uncleaned upper GT and
laboratory-conditioned GT about two and five magnitudes lower, respectively.  Koerner
et al. (1993) concluded that the upper GT was performing well, but the GT around the
LCRS pipe had poor performance.

As described by Giroud (1996), the purpose of a GT is to retain the material behind the
filter, not capture particles in motion.  The upper GT described by Koerner et al. (1993)
retains the sand protection layer over the gravel layer. Based on filter design
calculations performed by the authors of this appendix, the upper GT also captures
some of the slurried fines particles if they move with the leachate. These  fines, as  well
as biological particles, reduced the permittivity of the upper GT.  The GT around the
pipe serves no purpose.  It is not needed to prevent the gravel from falling through  the
pipe perforations.  In fact, this GT proved to be detrimental  as it captured fines and
biological particles at the small flow areas at the pipe perforations.

Resolution:  Koerner et al. (1993) do not indicate how this problem was resolved.

Lessons Learned for Future  Projects: Based on the available information, the following
lesson can be learned from this case history:

•  Perforated pipes bedded in gravel should not be wrapped with a GT because the GT
   is useless, and, in some cases,  even detrimental.  Furthermore, EPA recommends
   that perforated pipes generally not be wrapped with a GT (Bass, 1986).
                                     F-135
-------
A.6.3  L-36

Problem Classification: landfill LCRS or LDS malfunction/design
Region of U. S.:  southeast
Waste Type:  MSWash
Reference:  unpublished

Problem Summary: waste fines clogged needlepunched nonwoven GT filter around
LCRS pipe  bedding gravel

Problem Description:  The LCRS for a single-lined landfill cell consists of a sand
drainage layer (specified minimum hydraulic conductivity of 1 x 10~5 m/s) and a piping
system. The pipes are bedded in gravel wrapped with a needlepunched nonwoven GT.
The specified maximum apparent opening size of the GT is reportedly significantly lower
than that necessary to retain the sand.  Leachate collected in the pipes drains to an
internal sump and is removed by pumping.  Significantly less leachate than expected
flowed to the sump. In addition, leachate ponded in the landfill and seeped from the
landfill side slopes. When a video camera was run through the LCRS pipes, the pipes
were found to be full of the incinerated MSW ash placed in the landfill. The pipes were
flushed, but the sump still recharged very slowly even though the landfill was full of
leachate. Based on this observation, it was concluded that the LCRS is clogged. From
the gradation  of the sand LCRS drainage layer and the apparent opening size of the
GT, it is expected that the clogging is most significant in the GT around the pipe
bedding gravel.  The sand has larger openings than the GT and passes fine ash
particles. The fine particles may have become trapped on  and in the GT.  However, this
does not explain the large quantities of ash in the LCRS pipes.  It may be that the GT
around the  pipe bedding gravel has opened at some locations, allowing leachate to
bypass the  GT and flow directly into the pipe bedding gravel and the pipe.

Resolution: An underdrain system was constructed around the downgradient edge of
the landfill to collect leachate migrating from the landfill.  The underdrain consisted of a
collection pipe in  a gravel-filled trench. The top of the gravel was exposed. Due to the
slow draining  of leachate to the sump, leachate overtopped the landfill cell and flowed
into the underdrain. The leachate carried ash  particles that eventually clogged the
gravel in the underdrain.  Currently, it is proposed that an HOPE manhole be installed
on the sand drainage  layer on the upgradient side of the landfill to access the  landfill
leachate. The manhole will be installed on the upgradient side of the landfill because it
is anticipated  that the clogging will be less severe upgradient and leachate will recharge
the manhole faster.

Lessons Learned for Future Projects:  Based on the available information, the following
lesson can  be learned from this case history:

•  When the waste in a containment system contains some fine particles that may
   migrate  to the LCRS, the potential for LCRS clogging may be reduced by allowing
                                      F-136
-------
   those fine particles to pass though the LCRS to the leachate collection pipes, which
   can subsequently be cleaned. The fine particles will pass more easily through the
   drainage system if no GTs are used in the drainage system or if the drainage system
   contains relatively thin open nonwoven GTs rather than thicker nonwoven GTs with
   a smaller apparent opening size.  Note that the above does not apply to an LCRS
   with only a GN drainage layer. Though a GN drainage layer has a high
   transmissivity, it is thin and is, therefore, generally more susceptible to clogging by
   sedimentation than a granular drainage layer.
•  Drainage system pipes should be maintained by cleaning the pipes at least annually
   and more frequently, if warranted.

A.6.4  L-37

Problem  Classification:  landfill LCRS or LDS malfunction/operation
Region of U. S.:  southeast
Waste Type: MSW
Reference:   unpublished

Problem  Summary:  leachate seeped out landfill side slopes in the vicinity of chipped
tire layers

Problem  Description: As part of a site cleanup, about 1.2 million chipped tires were
disposed of in a 3.2 ha MSW landfill cell.  The average particle size of the tire chips is
about 100 mm.  Subsequently, leachate seeped out of the landfill side slopes in the
vicinity of chipped tire layers.  The coarse tire chips have a higher hydraulic conductivity
than the MSW and, apparently, promote lateral drainage within the waste. The LCRS
for the single-lined cell consists of a sand drainage layer (specified minimum hydraulic
conductivity of 1 x 10~5 m/s) and a piping system. Leachate collected in the pipes drains
to an internal sump and is removed by pumping.

Resolution:  A 0.9-m diameter bucket auger was advanced through the waste to the top
of the sand LCRS drainage layer at six locations near the seeps. Perched leachate in
the tire chips was found in several boreholes and several boreholes were dry above the
LCRS. The depth of perched leachate was up to 3 m.  The boreholes with perched
leachate  were completed as leachate wells.  The wells allow some of the leachate
collected in the tire chip layers to readily drain to the LCRS.  Leachate levels in the wells
are inspected weekly.  If there is leachate in a well, the well is pumped.

Lessons  Learned for Future Projects:  Based on the available information, the following
lesson can  be learned from this case history:

•  Leachate may seep from landfill side slopes if the leachate can perch on less
   permeable layers within the waste that are relatively close to the side  slope. The
   potential for seepage can  be decreased  by not placing  layers of these less
   permeable materials near the side slope, sloping less permeable layers away from
                                      F-137
-------
   the side slopes, distributing the materials throughout the waste, constructing
   leachate chimney drains to the LCRS around these layers, removing perched
   leachate from wells installed over these layers, or other means.

F-A.7     Landfill LCRS or LDS Operation

F-A.7.1    L£

Problem Classification: landfill LCRS or LDS operation/operation
Region of U. S.:  southeast
Waste Type:  MSW
Reference: unpublished

Problem Summary:  overestimation of LDS flow quantities due to problems (e.g.,
clogging) with automated LDS flow measuring and removal equipment

Problem Description/Resolution: The quantities of liquid pumped from the LCRS and
LDS sumps of a MSW landfill cell are monitored by landfill personnel. From 1991, when
operation began, to  March 1994, problems with the LCRS and LDS flow measuring
system made the measured flows unreliable.  During 1991, LDS flow rates were
estimated by multiplying the time the pump was on by the flow capacity of the pump.
However, due to problems with the control system, the pump sometimes stayed on
even when there was no more liquid to be removed (i.e.,  it pumped air),  leading to the
overestimation of flow rates. The control system that measured the liquid levels in the
sumps and operated the pumps was prone to compressor failure and clogging of air
lines. This lead to inaccurate measurements of leachate levels in the sumps and
caused pumps to run for too long of an interval or even continuously until they burned
out.  Mechanical flowmeters were installed into the cells in January 1992 to solve the
flow rate measurement problem.  Flow rates measured using the flowmeters were
several times lower than flow rates calculated using the "pump on" time. However, the
measured LDS flow rates remained high, and the flow measuring system underwent
frequent repair.  The impeller and filter screen in the flowmeters frequently became
clogged, making the flowmeters inoperable. In March 1994, the mechanical flowmeters
were replaced with customized venturi flowmeters that were less prone to clogging.
However, these flowmeters were damaged in July 1994 by an electrical storm.  The
meters were  subsequently repaired. In December 1994,  it was discovered that a failed
check valve in the leachate riser house allowed LDS liquid that had been metered to
flow back into the LDS of the cell and be remetered. The check valve was replaced.
Even with the above repairs, the measured LDS flow rates were still  relatively high.  In
early 1995, the leachate level measurement system  in the LCRS sump experienced drift
due to the buildup of landfill gas pressures in the sump, though the gases could
passively vent through riser pipes. To correct this problem, the "pump on" levels in the
LCRS and LDS sumps were lowered so the gases could vent more freely.
                                     F-138
-------
Lessons Learned for Future Projects: Based on the available information, the following
lessons can be learned from this case history:

•  The potential for clogging of water-level indicators, pumps, and flowmeters must be
   considered when selecting the types of equipment to use at a MSW landfill.  In the
   case history described above, the venturi flowmeters were less prone to clogging
   than the mechanical flowmeters with filter screens and impellers.
•  Leachate quantity measurement systems should be calibrated and adjusted as
   needed at least annually to ensure that the quantities measured are accurate.
•  Due to the potential for problems in automated leachate metering and pumping
   equipment, landfill operations plans should include a verification and contingency
   method for estimating the quantities of liquid removed from the LCRS and LDS.
F-A.7.2   L-23

Problem Classification:  landfill LCRS or LDS operation/operation
Region of U. S.: northeast
Waste Type:  MSW
Reference: unpublished

Problem Summary: valves on LCRS pipes were not opened and leachate could not
drain, and waste and leachate flowed over a berm into a new unapproved cell

Problem Description: An 8 ha landfill expansion was constructed with three single-
composite lined cells, separated from each other and from adjacent older cells by
intercell berms. Leachate collected in the expansion cells was conveyed to sumps in
the adjacent cells by three pipes, each fitted with a valve to be opened prior to
placement of waste in the cells.  In two of the expansion cells, the valves were not
opened before waste placement began and leachate collected in these cells could not
drain. Eventually, the waste became buoyant due to rising  leachate levels. After about
1.5 years of operation, a bulldozer operating at the active face sunk in the waste and
had to be removed with  a crane.  By this  time, about 12 m of waste had been placed
over the valves.

The intercell berms had  an exposed HOPE GM primary liner.  The design called for the
sand LCRS drainage layer to be placed incrementally over the 31-1:1 V berm side slopes,
advancing the sand with waste placement. In the one expansion cell with an open valve
on the LCRS pipe, waste was placed too close to an intercell berm between  it and a
new cell that had not yet been approved for waste.  Sufficient space between the waste
and the intercell berm should have been  maintained to temporarily store runoff from the
waste. After a storm, leachate and waste washed over the  berm and into the new cell.
A temporary access road made out of waste was constructed over the GM primary liner
on the intercell berm to access the new cell and clean out the waste that had washed
into it. The waste placed directly on the GM primary liner damaged the  GM.
                                     F-139
-------
Resolution: The corrective measures for the above problems have not yet been
implemented. However, it is anticipated they will cost on the order of $1,000,000 (about
40% of the original construction cost).  It is proposed that the impounded leachate be
pumped from the two cells and treated. Then, the waste will be excavated from around
the valves, and the valves will be opened.  The temporary waste access road will be
removed, and the underlying GM liner will be inspected for damage and repaired.  The
sand LCRS drainage layer will be placed on the berm slopes incrementally with waste
placement.

Lessons Learned for Future Projects: Based on the available information, the following
lesson can be learned from this case history:

•  The procedures to be followed by landfill operations personnel should be
   documented in an operations manual. Special operation procedures required for a
   specific design should be emphasized in the manual.  Periodically, an audit should
   be conducted to verify that the specified operation procedures are being practiced.

F-A.7.3   L-34

Problem Classification: landfill LCRS or LDS operation/operation
Region of U. S.: southeast
Waste Type: HW
Reference: unpublished

Problem Summary:  LCRS leachate pump moved air and liquid causing pump airlock
and underestimation of leachate quantities

Problem Description: The quantity of leachate pumped from the LCRS sump of a 5 ha
cell was measured by multiplying the number of times the pump turned on by a fixed
"pump on" time. The "pump on" time setting at the pump controller tended to drift
causing the pump to operate too long.  As a result, air was pulled into the pump, and the
pump tended to become airlocked and shut down. The pump did not reprime as the
leachate levels rose.  When the landfill operator noticed this, the pump was removed
from the sump and adjusted and the "pump on" time setting was reset.  With the pump
coming on longer but less frequent,  the quantity of leachate removed from the landfill
was underestimated. An accumulating flowmeter was  installed to  provide a better
measurement of leachate quantities. However, when air pulled into the pump moved
through the flowmeter, the flowmeter overestimated the quantity of leachate removed.
For example, in one month the accumulating flowmeter indicated that the leachate
volume removed from the cell was about 1.2 million liters. In comparison, the LCRS
flow rates from the adjacent cells of a similar size were about 10 times less.
                                     F-140
-------
Resolution: The pumps were replaced with self-priming pumps from a different
manufacturer. Leachate flow quantities calculated using the "pump on" counter
compared well with quantities measured with the flowmeter.

Lessons Learned for Future Projects'. Based on the available information, the following
lessons can be learned from this case history:

•  Leachate sump pumps should be self priming.
•  Leachate quantity measurement systems should be calibrated and adjusted as
   needed at least annually to ensure that the quantities measured are accurate.
•  Due to the potential for problems in automated leachate metering and pumping
   equipment, landfill operations plans should include a verification and contingency
   method for estimating the quantities of liquid removed from the LCRS and LDS.

F-A.7.4   L-35

Problem Classification: landfill LCRS or LDS operation/design
Region of U. S.: northeast
Waste Type: MSW
Reference:  unpublished

Problem Summary: LCRS leachate pumps and flowmeters continually clogged and
LDS leachate pumps turned on too frequently and burned  out prematurely

Problem Description: The LCRS and LDS of two landfill cells were designed with large,
shallow sumps. State regulations at the time the landfill was permitted required that the
head of leachate on the liner system, including in the sump, be no more than 0.3 m. The
LCRS drainage layer is sand with a specified minimum hydraulic conductivity of
1 x 10~4 m/s, and the LDS drainage layer is a GN. In the LCRS, the submersible pumps
and magnetic flowmeters continually become clogged with a white precipitate. In the
LDS, the flow rates into the sumps are less than the capacity of the submersible pumps.
To keep liquid levels in the sump less than 0.3 m but above the pump intake, the pump
cycle was very short. The pump motor overheated from turning on and off so quickly
and burned out.

Resolution: The LCRS pumps and flowmeters are disassembled and cleaned with citric
acid about every month. A spare pump is used to pump a sump when a pump is being
cleaned.  The LDS pumps were replaced with smaller models to increase cycle times.

Lessons Learned for Future Projects: Based on the available information, the following
lessons can be learned from this case history:

•  Chemicals may precipitate within leachate sump pumps and flowmeters and
   interfere with their operation.  In the case history described above, clogging was only
   a problem for LCRS pumps and flowmeters. LDS pumps and flowmeters were not
                                     F-141
-------
   adversely affected by clogging.  However, the leachate in the LDS has been diluted
   by water from consolidation of the CCL component of the primary liner.
•  Leachate sump pumps should be selected to be compatible with sump geometries
   and anticipated leachate recharge rates.
F-A.8     Landfill Liner System Stability

F-A.8.1   L-21

Problem Classification:  landfill liner system stability/design
Region of U. S.:  northeast
Waste Type:  coal ash
Reference: unpublished

Problem Summary: sliding along PVC GM/CCL interface during construction

Problem Description:  A single-composite liner system being constructed for a canyon
landfill underwent slope failure during construction. The liner system consists of the
following components, from top to bottom:

•  0.45 to 0.6-m thick lime-stabilized sludge (28-day compressive strength of about 550
   kPa) or bottom ash protection layer;
•  0.3-m thick sand LCRS drainage layer (minimum hydraulic conductivity of 1 x 10~4
   m/s);
•  0.75-mm thick PVC GM liner; and
•  0.45-m thick CCL (maximum hydraulic conductivity of 1 x 10~9 m/s).

The inclination of the landfill side slopes are, on average, about 3.5H: 1V for the upper
60 m of slope length and 10H: 1V for the bottom 90 m of slope length. The maximum
side slope inclination is 3H:1V.  The compaction criteria for the clay liner were 95% of
the standard Proctor maximum dry unit weight and wet of the optimum moisture
content. During construction, about 1 ha of the liner system on the upper 3.5H:1V side
slopes slid downslope along the GM/CCL interface. The slide zone represented about
50% of the upper slope area that had been covered.  Sliding occurred both after
placement of the sand drainage layer and during placement of the protection layer. The
slide zone was identified by cracking of the sand layer or stabilized sludge layer near
the crest of the side slope and wrinkling of the GM  liner near the slope toe. When the
GM in the slide zone was exposed, it was taut and, in some cases, torn near the slope
crest. Analyses of liner system slope stability had not been conducted as part of the
landfill design. However, over 30 ha of liner system had been successfully constructed
previously using the same liner system components and geometry and similar site soils
to construct the CCL.
                                     F-142
-------
After sliding of the liner system occurred, a forensic investigation was conducted to
identify the cause of this failure. The CCL surface beneath the GM liner in the slide
zone was visually inspected and found to be relatively wet. However, this wet zone was
only at the surface of the CCL (i.e., in the upper 10 mm).  Samples of the CCL material
in the slide zone at about 5 mm and 150 mm beneath the top surface of the CCL had
measured moisture contents of about 7 and 2 percentage points wet of standard Proctor
optimum, respectively. The average moisture contents of two deeper samples of CCL
material were equal to the average CCL moisture content measured during construction
using a nuclear density gauge and an oven.  The average moisture content of the
surface sample of CCL material was about 7 percentage points wet of optimum,
indicating saturation.  The increase in moisture content at the surface of the CCL
between compaction and sliding is believed to have resulted from condensation of water
on the lower face of the GM due to thermal effects and spraying of the CCL surface to
prevent desiccation prior to placement of the GM.  The GM liner was placed on the CCL
in the fall during days with large diurnal temperature fluctuations.  During the day, as the
temperature of the liner increased, water evaporated from the CCL and the air
entrapped between the CCL and the GM became saturated with water vapor.  During
the night, as the temperature decreased, the ability of the entrapped air to contain water
vapor decreased.  As a result, a fraction of the vapor condensed into water and water
droplets formed on the lower face  of the GM.  In the sliding zones, the GM liner and
CCL were exposed to these temperature fluctuations for up to two weeks before being
covered with the sand drainage layer.  In the previous phase of construction, which was
completed without incident, the GM liner was placed in the summer when  the diurnal
temperature fluctuations were less and the GM was only exposed to these fluctuations
for about three days before it was  covered with soil.

Subsequently, direct shear tests were performed on the GM/CCL interface in
accordance with ASTM D 5321 to  investigate the effect of CCL moisture content on
interface shear strength under  low normal stress. When the CCL was compacted at
about 2 percentage points wet  of its optimum moisture content,  the peak and large-
displacement secant interface friction angles of the GM/CCL interface were about 19°
and 18°, respectively. The peak and large-displacement secant interface friction angles
decreased with increasing CCL moisture content and were about 14° and 12°,
respectively, when the interface was tested with the CCL at about 7 percentage points
wet of optimum. Slope stability analyses performed by the owner indicate that on the
steepest 3H:1V portions of the  lining system, the liner system is just stable, with a factor
of safety of 1.03, for a CCL moisture content 2 percentage points wet of optimum. The
liner system is unstable, with a factor of safety of 0.64, for a CCL moisture content 7
percentage points wet of optimum. The owner finds a minimum factor of safety of 1.0
acceptable for the following reasons: (i) most slopes are  less steep than 3H:1V and
have a factor of safety significantly greater than 1.0 if the 31-1:1 V slopes have a factor of
safety of 1.0;  (ii) prior to placement of ash in the landfill, the protection layer will have
set and gained strength, increasing the calculated factor of safety to over 1.5; (iii) the
ash placed in the landfill will buttress the slopes; and (iv) liner system instability
                                     F-143
-------
occurring during construction before the protection layer has gained sufficient strength
will be detected and repaired before construction is completed.

Resolution:  The following corrective measures were implemented.

•  A temporary protective cover was placed over the GM liner and CCL in the slide
   zone to protect the CCL from frost damage until the GM and overlying soil layers
   could be reconstructed in the spring.
•  A polypropylene monofilament woven GT reinforcement layer was installed between
   the GM liner and overlying soils to carry the load of the soils  for the approximately 1
   ha of liner system constructed on the upper side slopes after the failure occurred.
   The large-displacement secant interface friction angles of the sand/GT and GT/GM
   interfaces were about 25° and 15°, respectively. The strain measured in the
   installed GT near the crest of the side slope was less than 2%, which was expected
   based on the calculated GT tension. Even if the GM/CCL interface is very wet and
   is the critical interface, i.e., the interface with the lowest strength, slope stability
   analyses performed by the owner indicate that most of the load associated with the
   overlying soils will be carried  in tension by the GT reinforcement, which is much
   stiffer than the GM.
•  The owner developed new construction procedures to reduce the potential for liner
   system sliding in the future: (i) the surface of the CCL must not be wet with
   supplemental moisture (e.g.,  rain, dew, spraying) when the overlying GM is installed;
   (ii) the GM liner must be covered with the sand drainage layer within five working
   days after it is placed to reduce the  potential for moisture migration to the GM/CCL
   interface; and (iii) GT reinforcement must be used if the CCL surface has a moisture
   content greater than 3 percentage points wet of optimum prior to placement of the
   sand.

Lessons Learned for Future Projects: Based on the available information, the following
lessons can be learned from this case history:

•  Liner system slopes should always be evaluated using rigorous slope stability
   analysis methods that  consider the actual shear strengths of the liner system
   materials. Actual interface strengths can only be assessed by project-specific
   testing.  Such testing is recommended.
•  The effect of construction on  moisture conditions  at the GM/CCL interface should be
   considered when developing  the specification for CCL construction and selecting the
   strength of liner system interfaces for slope stability analyses.  The CCL construction
   specification should generally include limitations on maximum compacted moisture
   content, restrictions on applying supplemental moisture, and requirements for
   covering the CCL and  overlying GM as soon as practical to minimize the moisture
   migration to the GM/CCL  interface.
                                      F-144
-------
A.8.2  L-24

Problem Classification:  landfill liner system stability/operation
Region of U. S.:  northeast

Waste Type:  MSW
Reference:  unpublished
Problem Summary:  sliding along GN/GCL (HOPE GM side) and GCL (bentonite
side)/CCL interfaces during operation

Problem Description: A single-composite liner system for one cell of a landfill expansion
was constructed in a 17-m deep excavation. The cell was shaped like a triangle, with
3H:1V side  slopes on one side and the excavation base on the other two sides. The
liner system on the base slope consists of the following components, from top to
bottom:

•  0.45-m thick sand protection  layer;
•  nonwoven GT filter;
•  GN LCRS drainage layer;
•  GCL composed of a 1.5-mm  thick HOPE GM (textured on side slopes and smooth
   on base) and a bentonite layer glued to one side of the GM; and
•  0.9-m thick CCL.

On the side slopes, the GT and GN  are replaced by a GC.  The GCL was installed with
the bentonite side down and was seamed by fusion seaming the GM component of
adjacent panels.

When construction of a new adjacent cell began, waste from an old unlined landfill in
the footprint of the new cell was  required to be relocated to the existing lined active cell.
The existing cell did not have adequate capacity for the waste, and was temporarily
overfilled to hold the waste.  With the additional waste, the height of the waste in the
existing cell was about 17 m and the intermediate waste slopes were at about 2.5H:1V,
significantly steeper than the maximum slope of 41-1:1 V specified in the operations plans.
The top of the waste extended from the crest of the side slopes up to about 80 m from
slopes, in the direction of the apex of the triangular-shaped cell. A 0.9-m high sand
berm was located  at the toe of the intermediate waste slopes to increase slope stability
by providing a buttress for the waste.

As part of the landfill permit, the  design engineer conducted a slope stability analysis of
the landfill assuming that the secant friction angles for the waste and the weakest liner
system interface were 25° and 7°, respectively. No project-specific interface shear
strength tests of the liner system interfaces were performed. The stability analysis
conducted for design did not consider the liner system with the waste at intermediate
grades;  there was no regulatory  requirement to include this in the permit. The minimum
calculated static factor of safety for landfill stability was 1.4. This factor of safety was for


                                      F-145
-------
the 31-1:1 V waste slopes (between benches) at their final grades.  The minimum static
factor of safety calculated for failure through the liner system, with waste at final grades,
was 5.0.  When the landfill owner later decided to overfill the cell with waste, the stability
of the liner system with the relatively steep waste slopes was not analyzed.

Shortly before the liner system in the active cell was connected to the liner system in the
new cell,  the sand berm at the toe of the active cell began to heave.  A crack on the
order of 50 mm wide developed on top of the waste near the slope crest. The crack
extended for about 20 m along the crest, on one side of the triangular-shaped cell, and
was quasi parallel to the side slope anchor trench.  Smaller cracks developed parallel to
the larger crack and extended back into the slope from the larger crack. All visible
cracks ended by 7 m from the side slope crest As the liner system for the new cell was
being tied into that for the old cell, GCL folds were observed near the tie-in.  In one
instance, about 3 m of GCL was folded into a 1.2-m long section.  When the sand was
removed from over the LCRS pipe, the pipe popped up out of the ground.  Under the
moving waste mass, the pipe had bent and doubled-up.

Resolution:  About 270,000 m3 of waste was removed from the failure zone to inspect
and repair the damaged liner system. The damaged liner system extended up to about
60 m into the waste,  and was defined by the area where the GCL was taut and torn.
Within the failure zone, the liner system slid along the hydrated bentonite,  tearing the
GCL in two locations: (i) four 3-m long parallel tears about 2 m apart and parallel to the
toe of the waste slope; and (ii) a 0.15-m long tear parallel to the toe of the  waste slope.
When waste and liner system materials were removed from over the tears, repairs had
to be quickly made since the GCL was so taut that the tears opened further. The GCL
on the side of the tear furthest into the waste pulled back into the waste. When GCL
samples were cut to  define the area of damage, the GCL at the sample locations also
tended to split along the cuts. The GM component of the GCL in the samples was
generally about 10% thinner than the thicknesses measured during GCL conformance
testing and the samples were thinner than the minimum allowable thickness given in the
project specifications. However, the samples met the required strength requirements.
The GCL was repaired by patching it. Interestingly, no damage to the GT  and GN over
the GCL was found.  These materials moved up to about 3 m less than the GCL due to
sliding between the GN  and GM component of the GCL.  The liner system at the anchor
trench has not yet been inspected. To increase the stability of the liner system, the
intermediate waste slopes have been regraded to 4H:1V.

The design engineer backanalyzed the stability of the liner system at failure and
concluded that with an assumed secant friction angle for the waste of 25°, the minimum
secant friction angle for the liner system interfaces at failure was about 2°.

Lessons Learned for Future Projects: Based on the available information, the following
lessons can be learned from this case history:
                                     F-146
-------
   Liner system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the actual shear strengths of the liner system
   materials. Actual interface strengths can only be assessed by project-specific
   testing.  Such testing is recommended.
   Proposed changes to the landfill filling sequence should be reviewed by the design
   engineer to ensure that these changes will not adversely affect the landfill.
Figure F-A.8.1. When waste was excavated from the failure zone, large wrinkles
in the GCL were evident.

A.8.3  L-25

Problem Classification:  landfill liner system stability/design
Region of U. S.:  southwest
Waste Type:  HW
References: Byrne, R.J., Kendall, J., and Brown, S., "Cause and Mechanism of Failure
    Kettleman Hills Landfill B-19, Phase 1A", Stability and Performance of Slopes and
    Embankments-II, ASCE Geotechnical Special Publication No. 31, 1992, pp. 1188-
    1215.
                                      F-147
-------
    Mitchell, J.K., Seed, R.B., and Seed, H.B., "Kettleman Hills Waste Landfill Slope
    Failure. I: Liner-System Properties", Journal of Geotechnical Engineering, Vol. 116,
    No. 4, Apr 1990, pp. 647-668.
    Seed, R.B., Mitchell, J.K., and Seed, H.B., "Kettleman Hills Waste Landfill Slope
    Failure. II: Stability Analyses", Journal of Geotechnical Engineering, Vol. 116, No.
    4, Apr 1990, pp. 669-690.

Problem Summary: sliding along HOPE GM/polyester needlepunched nonwoven GT
and HOPE GM/CCL interfaces during operation

Problem Description:  Landfill B-19 at the Kettleman Hills hazardous waste treatment,
storage, and disposal facility, located in Kettleman City, California, is a 15 ha double-
composite lined landfill.  The landfill was constructed in a 30 m deep excavation. The
liner system on the landfill base consists of the following components, from top to
bottom:

•  0.3-m thick soil protection layer;
•  GT;
•  0.3-m thick granular material/GT/GN LCRS drainage layer;
•  1.5-mm  thick HOPE GM/0.45-m  thick CCL composite primary liner;
•  GT;
•  0.3-m thick granular LDS drainage layer;
•  GT;
•  1.5-mm  thick HOPE GM/2-m thick CCL composite secondary liner;
•  GT/drainage rock/GT/2-mm thick HOPE GM vadose zone monitoring system.

The liner system on the side slopes consists of the following components, from top to
bottom:

•  0.6-m thick soil protection layer;
•  1.5-mm  thick HOPE GM protection layer (i.e., sacrificial GM);
•  GT;
•  GN LCRS drainage layer;
•  1.5-mm  thick HOPE GM primary liner;
•  GT;
•  GN LDS drainage layer; and
•  1.5-mm  thick HOPE GM/2-m thick CCL composite secondary liner.

The GT in the liner system is a polyester needlepunched nonwoven type. The landfill
was designed using presumed interface shear strengths for the liner system. The
presumed strengths were based on published information available at the time. The
plasticity  index of the CCL material ranged from 22 to 46%.  On average, the CCL
material was compacted to 94% of its Standard Proctor maximum dry unit weight at a
moisture  content 5 percentage points wet of optimum.  There was no limit on the
                                     F-148
-------
maximum CCL moisture content in the specifications. Waste placement began in the
first landfill cell, Phase IA, in early 1987. This cell occupies about 6 ha and was
constructed against the 2H:1V excavation side slopes on the southwest and northwest
and the 31-1:1 V side slopes on the northeast. The southeast side of the cell was
constructed on the relatively flat base and separated from the adjacent cell by a berm.
By 15 March 1988,  up to 27 m of waste had been placed over the liner system. The
slope of the waste, dipping  toward the southeast, was about 3H:1V.

Around 6:30 am on 19 March 1988, an approximately 10-mm wide crack was observed
across the landfill access road at the north corner of the landfill. This access road was
used to descend into the cell. By about 9:30 am, a 80 to 100-mm wide crack with a
vertical offset  of 160 to 200 mm was observed along the top of the waste slopes where
they intersected the northwest and southwest side slopes of the landfill. By noon, the
cracks in the waste had opened wider to several decimeters. The main failure, with
sliding of the waste mass from the northwest to the southeast, occurred by about 1:30
p.m. Horizontal displacements of the waste mass were up to 11  m and vertical
displacements along the side slopes were up to 4 m. The vertical riser from the LCRS
and LDS sumps, located in  the center of the northeast side,  also appeared to have
translated about 11  m with the waste. Around the side slopes, the soil cover over the
waste and the waste was cracked and, in some locations, the liner system was torn.
Based on the  nature of the  movements and the pattern of cracking, it was concluded
that the waste moved as a mass and sliding likely occurred within the landfill liner
system.

Excavation of waste from Phase IA began in October 1989.  As the waste was removed
and relocated into other phases of B-19, the condition of the waste and the liner system
was mapped.  The field investigation revealed that,  on the base of the cell,  the 31-1:1 V
northeast side slopes, and the lower portions of the 2H:1V northwest and southwest
side slopes, sliding occurred  along the GM/CCL interface of the secondary liner. The
materials above the CCL component of the  secondary liner appeared to have moved
monolithically with little movement between or within the liner system components.
Striations on the surface of  the CCL were consistent with the southeastern direction of
the slide.  There was no evidence of sliding within the CCL, except at the location of the
LDS sump. On the northeast side slopes, all of the geosynthetics remained fixed in the
anchor trench at the top of the slope and had been  pulled along the slope in the
direction of sliding.  On the  northwest and southwest side slopes, the GM primary liner
pulled out of the anchor trench and slide downslope with the waste about 8 m and 5 m,
respectively, over the underlying GT.  Further downslope, the failure surface moved to
the GM/CCL interface of the secondary liner and this GM tore. The tear extended
across the northwest slopes and into the southwest slopes and was on the order of 8  to
9 m wide.  At the toe of the  northwest and southwest slopes, the failure mechanism
appeared more complex, due to the kinematic constraints posed by the relatively sharp
change in orientation of the liner and the changes in liner system components.
Immediately upslope of the  toe area, wrinkles and folds were observed in some of the
geosynthetics. A corresponding pattern of tension tears was found downgradient of the
                                     F-149
-------
toe area. Based on relative movements between liner system geosynthetics and
striations on the surface of the CCL component of the primary liner, at least some of the
sliding at the toe occurred at the GM/CCL primary liner interface.

After the failure occurred,  interface direct shear and pullout tests were conducted to
evaluate the shear strength of critical liner system interfaces and stability analyses were
performed using  actual interface strengths.  Unconsolidation-undrained triaxial tests,
consolidated-undrained triaxial tests,  and one-dimensional consolidation tests were also
performed to evaluate the shear strength and consolidation characteristics of the CCL.
Based on the results of the tests, only a small amount of displacement (5 mm or less) is
required to mobilize the peak shear strength along an interface. At greater
displacements, the shear strength decreased and approached the large-displacement
value. The most critical interfaces with  the lowest large-displacement shear strengths
are GM/GT,  GM/GN, and GM/CCL (saturated). The mean large-displacement secant
interface friction angles of the GM/GT and GM/GN interfaces under submerged
conditions are about 8°. The mean large-displacement undrained  interface shear
strength of the GM/CCL interface is 33  kPa  with the CCL at its average as-placed
moisture content and overburden pressures corresponding to the maximum depth of
waste at failure.  The undrained interface shear strength decreases with increasing CCL
moisture content. At lower stresses,  the GM/GT and GM/GN interfaces are more
critical; at higher stresses, the GM/CCL interface  is more critical. This is generally
consistent with the observed  Phase IA failure surface.

The two-dimensional and three-dimensional slope stability factors  of safety for the
landfill cell were calculated by Byrne  et  al. (1992) using peak and large-displacement
shear strengths.  Only the three-dimensional analyses are discussed herein because
they better represent the complex cell geometry.  The factors of safety for the pre-failure
cell configuration using peak and large-displacement strengths were 1.25 and 0.85,
respectively.  From a comparison of the load-displacement curves for the critical
interfaces, peak shear strength would not be expected to be simultaneously mobilized
at the interfaces. Assuming that peak shear strengths were  mobilized on the landfill
base and 3H:1V side slope and large-displacement shear strengths were mobilized on
the 21-1:1 V side slopes, the calculated factor of safety was 1.08.  Thus, the measured
interface shear strengths and the rapid  decrease  in shear strength with displacement
after peak strength has been  reached can explain the Kettleman Hills Landfill failure.

Resolution:  The waste was relocated into other phases of B-19, the liner system was
repaired, and the cell was refilled with waste.

Lessons Learned for Future Projects'. Based on the available information, the following
lessons  can  be learned from this case history:

•  Liner system slopes should always be evaluated using  rigorous slope stability
   analysis methods that consider the actual shear strengths of the liner system
                                      F-150
-------
   materials. Actual interface strengths can only be assessed by project-specific
   testing.  Such testing is recommended.
•  The effect of displacement on interface shear strength must be considered when
   selecting the strength to use for design.
•  The effect of construction on moisture conditions at the GM/CCL interface should be
   considered when developing the specification for CCL construction and selecting the
   strength of liner system interfaces for slope stability analyses. The CCL construction
   specification should generally include limitations on maximum compacted moisture
   content, restrictions on applying supplemental moisture, and requirements for
   covering the CCL and overlying GM as soon  as practical to minimize the moisture
   migration to the GM/CCL interface.

A.8.4  L-26

Problem Classification: landfill liner system stability/design
Region of U. S.:  southwest
Waste Type:  MSW
References: Anderson, R.L., "Earthquake Related Damage and Landfill Performance",
    Earthquake Design and Performance of Solid Waste Landfills,  Yegian, M.K. and
    Liam Finn, W.D., eds., ASCE Geotechnical Special Publication No. 54,  1995, pp. 1-
    16.
    Augello, A.J., Matasovic, N., Bray, J.D.,  Kavazanjian, E., Seed, R.B., "Evaluation of
    Solid Waste Landfill Performance During the Northridge Earthquake", Earthquake
    Design and Performance of Solid Waste Landfills, Yegian, M.K. and Liam Finn,
    W.D., eds., ASCE  Geotechnical Special Publication No. 54,  1995, pp. 17-50.
    Geosynthetic Research Institute, "Evaluation of an High Density Polyethylene
    Geomembrane Exhumed from Canyon C", Geosynthetic Research Institute, Drexel
    University, Philadelphia, Pennsylvania, 29 Apr  1994.
    Matasovic, N., Kavazanjian, E.,  Jr., Augello,  A.J., Bray, J.D., and Seed, R.B., "Solid
    Waste Landfill Damage Caused by 17 January 1994 Northridge Earthquake", The
    Northridge, California, Earthquake of 17 January 1994, Woods, M.C. and Seiple,
    W.R., eds., California Department of Conservation, Division of Mines and Geology
    Special Publication 116, 1995, pp. 221-229.
    Matasovic, N. and  Kavazanjian, E., Jr., "Observations of the Performance of Solid
    Waste Landfills During Earthquakes", Eleventh World Conference on Earthquake
    Engineering, Acapulco, Mexico, Elsevier Science Ltd., 1996, CD-ROM Paper No.
    341.
    Matasovic, N., Kavazanjian, E.,  Jr., and Anderson, R.L., "Performance of Solid
    Waste Landfills in  Earthquakes", Earthquake Spectra, Vol. 13,  No. 5, May 1998.
    Stewart, J.P., Bray, J.D.,  Seed,  R.B.,  and Sitar, N., "Preliminary Report on the
    Principal Geotechnical Aspects of the January 17, 1994 Northridge Earthquake",
    Report No. UCB/EERC-94/08, College of Engineering, University of California at
    Berkeley, Berkeley, California, 1994, 238 p.
                                      F-151
-------
Problem Summary: two tears in HOPE GM liner and cracks in soil intermediate cover
after Northridge earthquake

Problem Description:  The Chiquita Canyon Landfill is located at the western edge of the
Santa Clara Valley in Valencia, California. The site is underlain by alluvium, cemented
sandstone with interbedded conglomerate and siltstone, poorly to moderately indurated
sandstone, siltstone, mudstone, and conglomerate.  No active faults are known to cross
the site. The landfill is divided into five units:  Primary Canyon, and Canyons A, B, C,
and D. Canyons A and D are separated by a soil fill wedge. The Primary Canyon is not
modern or geosynthetically lined and is not discussed further.  Canyons A to D have
single liners overlain by a sand LCRS drainage layer on the base and a soil protection
layer on the side slope. The liner for Canyons A and D is a 1.5-mm thick HOPE GM on
the side slope and a soil-bentonite CCL on the base.  Canyon  B has a 1.5-mm thick
HOPE GM on the side slope and western base and a soil-bentonite CCL on the eastern
base. Canyon C has a single composite liner on the base and a 1.5-mm thick HOPE
GM liner on the side slopes.  The liner system side slope inclinations range from
2.5H:1Vto1.5H:1V.

The 17 January 1994 Northridge earthquake (moment magnitude Mw 6.7) occurred on a
blind thrust fault at a depth of approximately 15 km at the northern end of the San
Fernando Valley in the greater Los Angeles area. The Chiquita Canyon Landfill is
located about 12 km from the zone of energy release (i.e., the fault plane). Strong
motion stations located on rock outcrops in the area recorded peak horizontal
accelerations on the order of 0.4g. The estimated rock peak horizontal acceleration at
the landfill resulting from  the earthquake is 0.33g. At the time of the Northridge
earthquake, Canyon B was inactive and awaiting closure, Canyons A and D were
partially filled and used only for landfilling in wet weather, and Canyon C was active.

After the earthquake, GM liner tears were found in Canyons C and D on benches above
the waste.  The tears were located near and parallel to liner system anchor trenches.
The GM tear located in Canyon  C  was 4.3 m long and opened up to 0.25 m wide.
Longitudinal cracks were present in the soil intermediate cover at the top of the waste
below the tear.  The cracks were up to 0.3 m wide, with vertical offset of 0.15 to 0.3 m
and extended along the entire side slope.  In Canyon D, there were three parallel GM
tears with a total length of about 23 m. Longitudinal cracks in the soil intermediate
cover at the top of the waste below the tears and across the side slope were about 0.3
m wide with 0.2 m of vertical offset.  At some locations, the cracks exposed the
underlying GM liner. The subgrade beneath the GM at both tear locations did not
appear to have been impacted.  Forensic analyses indicated that the GM tears initiated
from locations where GM seam samples were cut for destructive testing.  In fact, the
tears appear to have occurred from a corner of the rectangular hole left where the seam
was cut out. Both the stress concentrations around the hole (which had been patched)
and the high pullout capacity of the anchor trench appear to have been factors in  the
initiation and propagation of the tears.  As the GM liner moved during the earthquake, it
was constrained at the anchor trench and subsequently tore at locations with
                                      F-152
-------
concentrated stresses. Furthermore, in Canyons C and D, it appears that the slope
stability factor of safety of the waste at intermediate grades was relatively low under the
seismic loading of the Northridge earthquake. Cracks in the soil intermediate cover
were also observed in Canyon A near the contact of the landfill and the soil fill wedge
and in Canyon B along the perimeter of the side slope liner system.  The cracks in
Canyon A were about 0.15 m wide with 0.13 m of vertical offset. At some locations,
waste was exposed. The soil intermediate cover in Canyon B had some minor cracking
with no waste exposed.
Figure F-A.8.2. Longitudinal cracks in soil intermediate cover at Chiquita Canyon
Landfill after the Northridge earthquake.

Resolution: The GM tears were patched.  The liner system anchor trenches in Canyons
C and D above the GM tears were abandoned.  The GM liner was removed from
trenches, laid horizontal over the side slope bench, and covered with a soil berm.  The
cracked soil intermediate cover on the entire landfill was regraded and revegetated.

Lessons Learned for Future Projects: Based on the  available information, the following
lessons can be learned from this case history:
                                     F-153
-------
•  Composite liner systems on both base and side slopes, if properly designed and
   constructed, can sustain relatively strong earthquake shaking without experiencing
   significant damage.
•  Liner system anchor trenches should be designed to secure geosynthetics during
   construction, but release the geosynthetics before they are damaged during
   earthquakes.  For example, the liner system anchor may be designed with a
   relatively low factor of safety that is greater than 1.0 under static loading, but less
   than 1.0 under seismic loading that would damage an anchored liner system.  An
   alternative is to unanchor the liner system after construction and secure it on a
   bench with an overlying soil layer.
•  Stress concentrations at or near the liner system side slope crest should be avoided.
   In particular, GM seams should generally not be sampled near the slope crest.
•  Seismic design of composite liner systems can not prevent seismic displacements.
   However, good seismic design should limit seismic displacements to tolerable
   amounts.  To do this, designs may incorporate predetermined slip surfaces to
   confine movements to locations where they will cause the least damage (i.e., above
   the GM liner), soil covers that will resist cracking, and inverted liner system keyways.
•  Surficial cracking of soil cover layers during seismic loading, especially near
   locations with contrast in seismic response characteristics (e.g., top of waste by rock
   canyon walls),  should be anticipated and dealt with as an operation  issue through
   post-earthquake inspection and repair.

F-A.8.5  L-27

Problem Classification:  landfill liner system stability/design
Region of U. S.:  southwest
Waste Type:  MSW
References: Anderson, R.L., "Earthquake Related Damage and Landfill Performance",
    Earthquake Design and Performance of Solid Waste Landfills, Yegian, M.K. and
    Liam  Finn, W.D., eds., ASCE Geotechnical Special Publication No. 54, 1995, pp. 1-
    16.
    Augello, A.J., Matasovic, N., Bray, J.D., Kavazanjian, E., Seed, R.B., "Evaluation of
    Solid Waste Landfill Performance During the Northridge Earthquake", Earthquake
    Design  and Performance of Solid Waste Landfills,  Yegian, M.K. and Liam Finn,
    W.D., eds., ASCE Geotechnical Special Publication No. 54, 1995, pp.  17-50.
    Chang,  S., Bray, J.D., and Seed, R.B., "Engineering Implications of Ground Motions
    from Northridge Earthquake", Bulletin of the Seismological Society of America, Vol.
    86, No.  1, Part B Supplement, 1996, pp. S270-S288.
    Matasovic, N., Kavazanjian, E., Jr., Augello, A.J., Bray, J.D., and Seed, R.B., "Solid
    Waste Landfill Damage Caused by 17 January 1994 Northridge Earthquake", The
    Northridge, California, Earthquake of 17 January 1994, Woods, M.C. and  Seiple,
    W.R., eds., California Department of Conservation, Division of Mines and  Geology
    Special Publication 116, 1995, pp.  221-229.
                                      F-154
-------
Problem Summary: further tearing of GT cushion and extensive cracks in soil
intermediate cover from Northridge earthquake

Problem Description: The Bradley Avenue Landfill is located in a former gravel quarry
in the San Fernando Valley in southern California.  The site is underlain by alluvium,
which overlies crystalline and metamorphic basement complex rocks.  The older
alluvium consists of silty, subangular sand, cobbles and boulders. These deposits,
which are more than 150 m deep, are crudely horizontally stratified, unweathered, and
contain less than 1% clay. The more recent 15 to 23 m thick alluvial deposits are
mainly accumulations of subangular boulders, gravels, and sands.  The deposits are
uncemented but are tightly packed and stand at slopes of 1H: 1V or steeper.  The landfill
is divided into three units: Bradley East, Bradley West, and Bradley West Extension.
Bradley East and Bradley West are not modern or geosynthetically lined and are not
discussed further. Bradley West Extension has a single composite liner system on the
base and up to 1 H:1V side slopes. The liner system consists of the following
components, from top to bottom:

•  gravel LCRS drainage layer;
•  nonwoven GT cushion;
•  1.5-mm thick HOPE GM liner; and
•  CCL.

The 17 January 1994 Northridge earthquake (moment magnitude Mw 6.7) occurred  on a
blind thrust fault at a depth of approximately 15 km at the northern end of the San
Fernando Valley of the greater Los Angeles area. The  Bradley Avenue Landfill is
located about 11  km from the zone of energy release (i.e., the fault plane). Peak
horizontal accelerations recorded at nearby rock stations ranged from 0.2g to 0.4g.  The
estimated rock peak horizontal acceleration at the landfill resulting from the earthquake
is 0.36g.  At the time of the Northridge earthquake, the  Bradley West Extension was
receiving waste.

After the earthquake, extensive cracks were observed in the soil intermediate cover
near its contact with the side slope liner system. The cracks, which had up to 25 mm of
vertical offset, may have been the result of limited downslope movement of the GT
cushion over the  GM liner or the GM liner over the  CCL. No tears were observed in the
GM liner. However, GT tears reportedly  initiated prior to the earthquake appeared to
have increased in size as a result of the earthquake strong shaking.

Resolution: The GT tears were repaired.  The cracked  soil intermediate cover was
regraded and revegetated.

Lessons Learned for Future Projects:  Based on the available information, the following
lessons can be learned from this case history:
                                      F-155
-------
•  Seismic design of composite liner systems can not prevent seismic displacements.
   However, good seismic design should limit seismic displacements to tolerable
   amounts. To do this, designs may incorporate predetermined slip surfaces to
   confine movements to locations where they will cause the least damage (i.e., above
   the GM liner), soil covers that will resist cracking, and inverted liner system keyways.
•  Surficial cracking of soil cover layers during seismic loading, especially near
   locations with contrast in  seismic response characteristics (e.g., top of waste by rock
   canyon walls), should be  anticipated and dealt with as an operation issue through
   post-earthquake inspection and repair.

F-A.8.6   L-38

Problem Classification: landfill liner system stability/design
Region of U. S.:  northeast
Waste Type: MSW
Reference:  Boschuk, J. Jr., "Landfill Covers An Engineering Perspective", Geotechnical
    Fabrics Report, IFAI, Mar 1991, pp. 23-34.
    Giroud, J.P., "Lessons Learned From Studying the Performance of Geosynthetics",
    Proceeding of Geotextiles-Geomembranes Rencontres 93, Vol. 1, Joue-les-Tours,
    France, September 1993, pp.  15-31.

Problem Summary: sliding along needlepunched nonwoven GT/HDPE GM primary liner
interface after rainfall

Problem Description:  Boschuk (1991) and Giroud (1993) described the slope stability
failure of a double-lined landfill cell. The liner system consists of the following
components, from top to bottom:

•  0.6-m thick sand LCRS drainage layer (specified minimum hydraulic conductivity of
   1 x1Q-5m/s);
•  200 g/m2 needlepunched nonwoven GT;
•  1.5-mm thick HOPE GM primary liner;

•  0.3-m thick granular LDS drainage layer;  and
•  1.5-mm thick HOPE GM/0.6-m thick CCL composite secondary liner.

The 3H:1V side slopes of the cell were up to 30 m long.  Project-specific  interface direct
shear testing was not performed.  After analyzing the stability of the liner system side
slope, the design engineer added the GT between the sand LCRS drainage layer and
the HOPE GM primary liner.  The stability analyses showed the sand drainage layer
would be unstable if it were placed over the GM, but stable if placed over a
needlepunched nonwoven GT layer. Apparently, the potential for sliding  between the
GT and GM was not evaluated.
                                     F-156
-------
After several rainfall events, the sand drainage layer became wet and the GT began
creeping over the GM on the side slopes. About one month after sliding started, a
relatively heavy rainfall occurred at the site and the liner system slopes failed due to
excessive creep and tearing of the GT and cracking of the sand layer at the slope crest.
When the rain ended, the liner system was inspected and the sand at the toe of the
slope was saturated. At locations where the GT tore and slid downslope, the underlying
GM was abraded.  In several areas, higher "ridges" in the GM, which was not installed
perfectly flat, were sanded off.  Since the failure coincided with rainfall,  seepage
pressures in the sand probably contributed to the failure.

An infinite slope analysis was conducted by the authors of this appendix.  In this
analysis, the secant friction angle for the nonwoven GT/GM interface was assumed to
be 10°, which is within the range of friction angles reported for this interface in the
technical literature.  The calculated slope stability factors of safety are 0.52 and 0.25
without and with full seepage pressures in the sand layer, respectively.  These values
do not take into account tension in the geosynthetics, and,  being an infinite slope
stability analysis, do not take into account the toe buttressing effect.

Resolution: The liner system was repaired at a cost of over $800,000.  The method of
repair was not given by Boschuk.

Lessons Learned for Future Projects: Based on the available information, the following
lesson can be learned from this case history:

•  Liner system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the anticipated seepage pressures in the  liner
   system and the actual shear strengths of the liner system materials  assessed by
   project-specific testing. Such testing is recommended.

There is little available information for this case history;  additional lessons may have
been learned if the information was complete.

F-A.8.7   L-39

Problem Classification:  landfill liner system stability/design
Region of U. S.:  unknown
Waste Type:  MSW
Reference: Soong, T.Y. and Koerner, R.M.,  "The Design of Drainage Systems Over
    Geosynthetically Lined Slopes", GRI Report #19,  Geosynthetic Research Institute,
    Philadelphia, PA, 1997, 88 p.

Problem Summary: sliding along needlepunched nonwoven GT/HDPE  GM liner
interface after rainfall
                                      F-157
-------
Problem Description: Soong and Koerner (1997) described the slope stability failure of
a liner system on a 45 m long, 3H:1V side slope that occurred in 1992.  The liner
system consists of the following components, from top to bottom:

•  0.45-m thick AASHTO #57 crushed limestone LCRS drainage layer (average particle
   size  of 15 mm);
•  needlepunched nonwoven GT cushion; and
•  HOPE GM liner.

Soong and Koerner (1997) do not indicate if project-specific interface direct shear
testing or slope stability analyses were performed.

About one to two years after the liner system was constructed, a portion of the GT tore
at the crest of the slope and slid to the slope toe after a heavy rainfall. A number of
successive slides occurred during several subsequent rainfalls.  Soong and Koerner do
not indicate if the GM liner was damaged by the slides. They attributed the failure to
seepage pressures that developed in the drainage layer during heavy rainfall after it had
become contaminated with fines.  The source of the fines was not discussed.

An infinite slope analysis was conducted by the authors of this appendix.  In this
analysis, the secant friction angle for the nonwoven GT/GM interface was assumed to
be 10°, which is within the range of friction angles reported for this interface in the
technical literature. The calculated slope stability factors of safety are 0.52 and 0.25
without and with full seepage pressures in the sand layer, respectively.  These values
do not take into account tension in the geosynthetics, and, being an infinite slope
stability  analysis, do not take into account the toe buttressing effect. Based on these
factors of safety, the GT cushion would have been in tension  even without seepage
pressures.  Soong and Koerner do not indicate whether the GT was designed to be in
tension.

Resolution: The method of repair was not given by Soong and Koerner (1997).

Lessons Learned for Future Projects: Based on the available information,  the following
lesson can  be learned from this case history:

•  Liner system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the anticipated seepage pressures in the liner
   system and the actual shear strengths of the liner system  materials assessed by
   project-specific testing. Such testing is recommended.

There is little available information for  this case history; additional lessons might have
been  learned if the information had been complete.
                                      F-158
-------
F-A.8.8   L-40

Problem Classification: landfill liner system stability/design
Region of U. S.:  unknown
Waste Type:  MSW
Reference: Soong, T.Y. and Koerner, R.M., "The Design of Drainage Systems Over
    Geosynthetically Lined Slopes", GRI Report #19, Geosynthetic Research Institute,
    Philadelphia, PA, 1997, 88 p.

Problem Summary: sliding along gravel/HDPE GM liner interface after rainfall

Problem Description: Soong and Koerner (1997) described the slope stability failure of
a liner system on a 30 m long, 3H:1V side slope that occurred in 1993.  The liner
system consists of the following components, from top to bottom:

•  0.45-m thick AASHTO #3 crushed gravel LCRS drainage layer (average particle size
   of 37 mm); and
•  HOPE  GM liner.

Soong and Koerner (1997) do not indicate if project-specific interface direct shear
testing or slope stability analyses were performed.

About three to four years after the liner system was constructed, the  gravel slid over the
GM to the slope toe.  Soong and Koerner do not indicate if the GM liner was  damaged
by the slides.  They attributed the failure to seepage pressures that developed in the
gravel layer during heavy rainfall after the gravel had become contaminated with fines.
The source of the fines was not discussed. The fines apparently inhibited drainage of
water from the gravel and allowed the seepage pressures to develop.

An infinite slope analysis was conducted by the authors of this appendix.  In this
analysis, the secant friction  angle for the gravel/GM  interface was assumed to be 18°,
which is within the range of friction angles reported for this interface in the technical
literature.  The calculated slope stability factors of safety are 0.97 and 0.47 without and
with full seepage pressures in the gravel layer, respectively. Being an infinite slope
stability analysis, these values do not take into the toe buttressing effect.  Based on
these factors  of safety, the liner system was probably, at best, only marginally stable
after construction.

Resolution: The method of repair was not given by Soong and Koerner (1997).

Lessons Learned for Future Projects: Based on the available information, the following
lesson can be learned from  this case history:

•  Liner system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the anticipated seepage pressures in the  liner
                                      F-159
-------
   system and the actual shear strengths of the liner system materials assessed by
   project-specific testing. Such testing is recommended.

There is little available information for this case history; additional lessons might have
been learned if the information had been complete.

F-A.8.9   L-41

Problem Classification:  landfill liner system stability/design
Region of U. S.:  unknown
Waste Type:  MSW
Reference: Soong, T.Y. and Koerner, R.M., "The Design of Drainage Systems Over
    Geosynthetically Lined Slopes", GRI  Report #19, Geosynthetic Research Institute,
    Philadelphia, PA, 1997, 88 p.

Problem Summary: sliding along very flexible GM liner/needlepunched nonwoven GT
interface after rainfall

Problem Description: Soong and Koerner (1997) described the slope stability failure of
a liner system on a 20 m long, 2.51-1:1 V side slope that occurred in 1994.  The liner
system consists of the following components, from top to bottom:

•  0.3-m thick sand LCRS drainage layer;
•  very flexible polyethylene GM  liner; and
•  needlepunched nonwoven GT.

Soong and Koerner (1997) do not indicate if project-specific interface direct shear
testing or slope stability analyses were performed.

About two to six months after the  liner system was constructed, the GM tore along
about 30 m of the slope crest and slid downslope over the GT after a light rainfall.
Soong and Koerner attributed the failure to  seepage pressures that developed in the
sand layer during the rainfall.

An infinite slope analysis was conducted  by the authors of this appendix.  In this
analysis, the secant friction angle for the  GM/GT interface was assumed to be 18°,
which is within the range of friction angles reported for this interface in the technical
literature.  The calculated slope stability factor of safety is 0.81 without and with full
seepage pressures in the sand layer. This value does not take into account tension in
the GM, and, being an infinite slope stability analysis, does not take into account the toe
buttressing effect.

Thus, the authors found  that seepage pressures above the GM would not significantly
affect the stability of a failure surface along  the GM/ GT interface below the GM.  This
effect can be explained as follows (Giroud et al., 1995a).  The slope stability factor of
                                      F-160
-------
safety equals the shear strength divided by the shear stress. The shear stress acting
above the GM is the same as that acting below the GM, with or without seepage above
the GM. The shear strength of an interface is dependent on the interface adhesion,
interface friction angle, and effective normal stress at the interface.  The effective
normal stress at an interface decreases as the seepage pressures at that interface
increase.  With seepage pressures above a GM, the effective stress on top of the GM
decreases and, consequently, the shear strength at the interface above the GM
decreases. This decrease in shear strength leads to a decrease in  the slope stability
factor of safety.  Seepage pressures above a GM,  however, do not  impact the effective
stresses at the interface below the GM. Thus, the shear strength of this interface and
the slope stability factor of safety are unchanged with these seepage pressures.

Based on  the infinite slope analysis conducted by the authors of this appendix, the  liner
system was, at best, only  marginally stable after construction and the rainfall had the
effect of "triggering" the slide.  The light rain described  by Soong and Koerner (1997)
may have increased the weight of the overlying sand, which increased the tension in the
underlying GM.

Resolution: The method of repair was not  given by Soong and Koerner (1997).

Lessons Learned for Future Projects'. Based on the available information, the following
lesson can be learned from this case history:

•  Liner system slopes should always be evaluated using rigorous  slope stability
   analysis methods that  consider the anticipated seepage pressures in the liner
   system and the actual  shear strengths of the liner system materials assessed by
   project-specific testing. Such testing is recommended.

There is little available information for this case history; additional lessons might have
been learned if the information had been complete.

F-A.8.10  L-42

Problem Classification:  landfill liner system stability/operation
Region of U. S.:  unknown
Waste Type:  MSW
Reference: Soong, T.Y. and Koerner,  R.M., "The Design of Drainage Systems Over
    Geosynthetically Lined Slopes", GRI Report #19, Geosynthetic  Research Institute,
    Philadelphia, PA, 1997, 88 p.

Problem Summary: sliding along needlepunched nonwoven GT/PVC GM liner interface
after a thaw

Problem Description: Soong and Koerner (1997) described the slope stability failure of
a liner system on 30 m long, 4H:1V side slope segments between benches that
                                      F-161
-------
occurred in 1995.  The liner system consists of the following components, from top to
bottom:

•  0.45-m thick crushed gravel LCRS drainage layer (average particle size of 25 mm);
•  needlepunched nonwoven GT; and
•  PVC GM liner.

Soong and Koerner (1997) do not indicate if project-specific interface direct shear
testing or slope stability analyses were performed.

About one to two years after the liner system was constructed, the GT tore and slid
downslope, exposing approximately 3 ha of GM.  Soong and Koerner attributed the
failure to seepage pressures that developed in the gravel layer as frozen water in the
gravel melted.  The water could not flow freely out of the gravel at the toe of the slope
because of the ice  at the toe.

An infinite slope analysis was conducted by the authors of this appendix.  In this
analysis, the secant friction angle for the GT/GM  interface was assumed to be 18°,
which is within the  range of friction angles reported for this interface in the technical
literature. The calculated slope stability  factors of safety are 1.30 and 0.62 without and
with full seepage pressures in the gravel layer. These values do not take into account
tension in the GT, and, being an infinite slope stability analysis, do not take into account
the toe buttressing effect.

Resolution: The method of repair was not given by Soong and Koerner (1997).

Lessons Learned for Future Projects'.  Based on the available information, the following
lesson can be learned from this case history:

•  Liner system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the anticipated seepage pressures in the liner
   system and the actual shear strengths of the liner system materials assessed by
   project-specific testing. Such testing is recommended.
•  Outlets of drainage layers should be  kept free of snow and ice.

There is little available information for this case history; additional lessons might have
been learned if the information had been complete.

F-A.8.11  L-45

Problem Classification:  landfill liner system stability/operation
Region of U. S.: unknown
Waste Type: unknown
Reference: Hullings, D.E. and Sansone, L.J., "Design Concerns and Performance of
    Geomembrane Anchor Trenches", Proceedings of the 10th GRI Conference Field


                                      F-162
-------
    Performance of Geosynthetics and Geosynthetic Related Systems, Geosynthetic
    Research Institute, Philadelphia, PA, 1996, pp. 219-233.

Problem Summary: sliding along needlepunched nonwoven GT/HDPE GM liner
interface after erosion of soil anchoring geosynthetics

Problem Description: Mailings and Sansone (1996) described the slope stability failure
of a liner system after the 0.5-m thick soil layer anchoring the geosynthetics beyond the
slope crest was eroded by landfill traffic.  The runout length of the geosynthetics beyond
the crest was 3 m.  At the time of the failure the 1.5-mm thick HOPE GM liner on the
side slope was only covered by a needlepunched nonwoven GT protection layer.
Assuming the secant interface friction angle was 10° between the GT and underlying
GM and 15°  between the GM and underlying soil, Hullings and Sansone concluded that
the anchorage resistance provided by the soil layer was greater for the GM than for the
GT.

During the wet winter months, landfill traffic traveled over the anchored geosynthetics,
rather than over an adjacent muddy road. The soil layer covering the geosynthetics
quickly began to erode and, eventually, was not thick enough to anchor the
geosynthetics.  The GT and the overlying soil layer slid downslope over the GM.

Resolution: The method of repair was not given by Hullings and Sansone (1996).

Lessons Learned for Future Projects: Based on the available information, the following
lesson can be learned from this case history:

•  Soil layers anchoring geosynthetics should be maintained during landfill construction
   and operation.  The limits  of liner system can be identified with cones, flags, or other
   markers and can also be isolated from traffic by berms, bollards, or other means.

There is little available information for this case history; additional lessons might have
been learned if the information had been complete.

F-A.8.12  L-46

Problem Classification:  landfill liner system stability/operation
Region of U. S.: northwest
Waste Type: MSW
Reference: unpublished

Problem Summary: sliding along GN/HDPE GM primary  liner interface during
construction
                                      F-163
-------
Problem Description: A double-composite liner system underwent slope failure during
construction.  The side slope liner system consists of the following components, from
top to bottom:

•  0.3-m thick soil protection layer;
•  needlepunched nonwoven GT filter;
•  GN LCRS drainage layer;
•  1.5-mm thick HOPE GM primary liner;
•  GN LDS drainage layer;
•  1-mm thick HOPE GM secondary liner; and
•  0.6-m thick CCL.

The side slopes are  11 m high and have an inclination of 3H:1V.  Project-specific
interface direct shear testing was not conducted.  Slope stability analyses performed as
part of the liner system design used assumed interface shear strengths and relied on
the strength of the GT to carry part of the load of the overlying soil layer. The critical
interface for slope stability was assumed to occur between the GN LCRS drainage layer
and GM primary liner. The secant friction angle for this interface was taken as 11°. For
the analyses, it was  also assumed that the GT mobilized its wide-width tensile strength
at 10% strain.  The resulting factors of safety against tearing of the GT and sliding along
the GN/GM primary  liner interface were 2.0 and 1.6, without and with construction
loads, respectively.  The designer concluded that even if the strain were greater, the
GM should not tear "because of its ability to elongate 750% before tearing". The effect
of the strain in the GT and GN  LCRS drainage layer on the hydraulic properties of these
materials was not considered.

As the soil protection layer was being placed in the cell, the soil layer began to crack
near the slope crest.  A small test pit was excavated through the soil layer to inspect the
geosynthetics. The  GT was torn in this area and the GN was stretched and damaged.
The cause of this damage was initially unclear and was considered to be potentially
caused by slope instability or construction equipment. Subsequently, a slope failure
initiated at the excavation area and expanded laterally as it progressed downslope. The
GT tore near the slope crest and the GN panels separated and both slid downslope
over the GM in a localized area of the cell, approximately 15 m wide and 9 m  long. The
sides of the slide zone ran along sewn seams of GT. The GN panel separations along
the slope were about 5 cm.  Compression ridges in the soil protection layer were
present near the slope toe. Construction ceased  until the case of the failure could be
investigated.  About  three months later, after a significant precipitation event reportedly
deposited several decimeters of wet snow at the site, several more slides occurred.
The original slide zone moved further downslope, and slides similar in character to the
original one occurred in three new areas.  The slide zones covered approximately 75%
of the north slope  of the cell, the only slope covered with the soil protection layer.  The
original slide zone moved 6 to 9 m further downslope. The other three slide zones,
ranging from about 15 to 30  m wide, slid from 6 to 18 m downslope.  At one location
near the slope crest, the  GN was observed to have slide along the GM and GN panels


                                      F-164
-------
were separated along the slope by about 0.2 m. In general, the slide zones appeared to
have moved somewhat diagonal across the slope, perhaps sliding parallel to the top
ridge of the GN.  Tears in the GM liner were not observed.

A forensic investigation was conducted to evaluate the cause of the slope failure. The
investigation including performing interface shear strength tests and reviewing the slope
stability analyses conducted by the design engineer.  The measured secant friction
angle for the GN/HDPE GM interface was about 8°, significantly lower than the
assumed value of 11 °.  In addition, the strengths of the GT used in the original analyses
were too high: the GT secant modulus at 10% strain rather than the GT strength at 10%
strain was used in the calculations.  Further, the low ground pressure bulldozer actually
placing the soil protection layer weighed about twice as much as that used  in design.
Using the measured critical interface shear strength, the GT strength at 10% strain, and
the actual weight of the construction equipment, the investigators calculated factors of
safety of 0.63 and 0.48 without and with construction loads, respectively. The
investigators also indicated that GT filters probably should not be used to carry
significant tensile loads since they have very low tensile moduli and, thus, require
significant strain to mobilize moderate strengths.

Resolution: In areas where the soil protection layer had been placed, the soil was
removed and the damaged GT and GN were repaired. The placement of the soil
protection layer over the GT was subsequently limited to increase slope stability: the
soil layer was required to be placed in 6 m increments along the slope, advancing
upslope with waste placement.

Lessons Learned for Future Projects: Based on the available information, the following
lesson can be learned from this case history:

•  Liner system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the anticipated seepage pressures in the liner
   system and the actual shear strengths of the liner system materials assessed by
   project-specific testing. Such testing is recommended.

F-A.9     Landfill Liner System Displacement

F-A.9.1   L£

Problem Classification:  landfill liner system displacement/design
Region of U. S.: northeast
Waste Type: MSW
Reference: unpublished

Problem Summary: uplift of HOPE GM by landfill gas after erosion of overlying sand
LCRS drainage layer
                                      F-165
-------
Problem Description: During construction of a landfill liner system over the side slopes
of an existing active MSW landfill, the GM component of the single-composite liner was
uplifted by gas emanating from the waste. The liner system on the approximately
4H:1V 25-m high side slopes was underlain by a 0.3-m thick structural fill foundation
layer.  The foundation layer was not permeable enough to freely convey gas beneath
the liner system, and a gas collection layer was not provided.  Instead, the design called
for gas beneath the liner system to be collected in an approximately 0.3 m wide by 1.5
m high gravel-filled trench located at the crest of the side slopes. This design required
the gas to migrate through the existing landfill waste or liner system foundation layer to
the trench and ultimately to a flare. The design was not based on site-specific
considerations, rather it was based on designs that had been used successfully at other
landfills.  Landfill gas generation rates beneath the liner system were not estimated, and
field measurements of gas production beneath the side slopes were not made.

Prior to construction of the liner system, the existing waste slope was regraded. The
exposed waste was wet and contained pockets of gas that subsequently vented.
Landfill gas was smelled by residents up to approximately 2 km from the existing landfill.
The liner system was subsequently constructed over the waste.  As the HOPE GM was
seamed on the side slopes, gas emitted from the waste periodically ignited;  one seamer
was injured. After a 0.6-m thick sand LCRS drainage layer was placed over the GM,
there were no more gas problems until the sand began to erode after a rainfall. At the
eroded areas there was less overburden pressure on the liner system, allowing bubbles
of gas to form beneath the GM.  Ten 6-m diameter gas bubbles developed over the 3.6
ha landfill side slopes and uplifted the GM to about 1.5m. In some areas the  GM was
observed to be yielding, as evidenced by the change in its color (i.e., it became lighter).
When the GM was subsequently cut to release the gas, the yielded GM appeared
thinner than adjacent GM. The biaxial strain in the uplifted GM was calculated by the
authors of this appendix to be about 16%, assuming that the uplifted area has a
spherical curved shape. In comparison, the biaxial yield  strain of HOPE GMs  is on the
order of 10%.

Resolution: The GM liner was cut at the ten gas bubbles, and temporary gas  venting
pipes were  installed through the liner system and into the underlying waste.  These gas
vents will be removed and the liner repaired before waste is placed over the liner
system in these areas.

Lessons Learned for Future Projects'. Based on the available information, the following
lessons can be learned from this case history:

•  When liner systems (or cover systems) are constructed over existing waste, the
   potential for the waste to generate gas must be considered. Calculation methods
   are available to estimate the gas generation rates for MSW landfills, based on the
   age of waste and other factors. Some landfills may be generating little or no gas at
   the time of construction of an overlying liner system.  For these landfills,  a  gas
   collection layer may not be needed.  Other landfills, like the one described above,
                                      F-166
-------
   may be generating significant quantities of gas.  For these landfills, a gas collection
   layer may be required beneath the entire liner system.
•  When a landfill is constructed on top of an existing landfill (vertical expansion),
   exposed GM liners can be uplifted by gases from the underlying landfill. Therefore,
   in the case of a vertical expansion, unless gases from the underlying landfill are well
   controlled, GMs must be covered by a layer of soil to prevent GM uplift and
   precautions must be taken to prevent erosion of this soil layer.
•  Uplift of a GM liner by landfill gas may generate a GM strain that exceeds the yield
   strain of the GM, which may weaken the GM, thereby reducing its ability to withstand
   subsequent stresses.

F-A.9.2   L-11

Problem Classification:  landfill liner system displacement/design
Region of U. S.:  northeast
Waste Type: MSWash
Reference: unpublished

Problem Summary: uplift of geosynthetics by landfill gas after erosion of overlying sand
protection layer

Problem Description:  A landfill double-liner system was constructed over the slopes of
an existing unlined MSW landfill. On the 2.5H:1V side slopes of the existing landfill, the
double-liner system was underlain by a 0.15-m thick foundation layer. The foundation
layer was not permeable enough to freely convey gas beneath the liner system, and a
gas collection layer was not provided. Instead, the design called for gas beneath the
liner system to be collected in the gas extraction wells for the existing landfill.

Portions of the 0.45-m thick sand protection layer on the side slopes were progressively
eroded by runon and runoff. In two areas where the protection layer had eroded, the
underlying GC  LCRS drainage layer, HOPE GM primary liner, GN LDS drainage layer,
and HOPE GM component of the composite secondary liner were uplifted  about 0.1 m
by landfill gases.  The diameters of the uplifted areas were about 10 and 20 m.  The
biaxial strain in the uplifted GM was calculated by the authors of this appendix to be
about very small (i.e., less than  0.002%), assuming that the uplifted area has a
spherical curved shape. In comparison, the biaxial yield strain of HOPE GM is on the
order of 10%.

Resolution: The GM liner was cut at the two gas bubbles to release the gas and the
liner system was repaired.  The sand protection layer was replaced on the side slopes.

Lessons Learned for Future Projects:  Based on the available information, the following
lessons can be learned from this case history:
                                      F-167
-------
•  One waste containment system problem can lead to a number of other problems. In
   the case history described above, erosion of the sand protection layer led to uplift of
   the unballasted liner system geosynthetics by landfill gas.  Erosion of the protection
   layer could have also lead to damage of the underlying geosynthetics by
   construction equipment replacing the protection layer over the liner system and by
   exposure to the environment (e.g., ultraviolet light, freezing temperatures).
•  In the case of a vertical expansion, unless gases from the underlying landfill are well
   controlled, GMs must be covered by a layer of soil to prevent GM uplift and
   precautions must be taken to prevent erosion  of this soil layer.
•  From the viewpoint of gas uplift, erosion of the soil protection layer is generally less
   detrimental if it occurs over a large area than if it occurs over a small area because
   the GM strain is likely to be larger if the uplifted area is limited.

F-A.9.3   L-25

Problem Classification:  landfill liner system displacement/design
Region of U. S.:  northeast
Waste Type:  MSW
Reference: unpublished

Problem Summary: uplift of composite liner by surface-water infiltration during
construction

Problem Description:  A single-composite liner system for two cells of a  landfill
expansion was constructed with 16.7-m high side slopes. The bottom 7.6 m of the side
slope liner system was constructed in an excavation against native soil having a
specified maximum hydraulic conductivity of 1  x 10~7 m/s. The upper 9.1 m of the side
slope liner system was constructed against a more permeable rocky mine spoil berm.
The ground surface outside of the berm was graded towards the berm, allowing runoff
to pond at the toe of the exterior berm slope. Prior to construction of the liner and after
rainfall, portions of the excavation base near the toe of the side slopes seemed soft and
moist.  The soft native soil was removed and replaced with compacted mine spoil.  After
several days, this mine spoil became moist. However, it  did not soften  as much
material as the native soil and was considered acceptable for liner installation. The liner
for the cells consists of the following components, from top to bottom:

•  GCL composed of a 1.5-mm thick HOPE GM and a bentonite layer glued to one side
   of the GM; and
•  0.9-m thick CCL.

The GCL was installed with the bentonite side down and was seamed by fusion
seaming the GM component of adjacent panels.

As the liner was being constructed, there were several storms that left runoff ponded at
the toe of the exterior berm slope. This water was not removed by the contractor, but


                                      F-168
-------
was left to evaporate and infiltrate into the soil.  Subsequently, water pooled under the
GM and saturated the GCL and CCL near the toe of the side slope. When the GM was
cut for the water to drain and the damaged GCL was removed, the underlying CCL was
very soft.  From inspection of the cell side slopes, the water appeared to be originating
about halfway up the side slopes at the interface of the berm mine spoil and the native
soil.  Apparently, runoff was seeping through the more permeable berm soils into cell.

Resolution: Prior to  reconstructing the damaged liner, the runoff infiltration problem was
resolved as follows:

•  Three vertical wells were installed into the berm to drain the water.  The wells were
   pumped for about one week before construction resumed. Pumping continued at a
   lower rate throughout construction.
•  The ground outside of the berm was regraded so that runoff did not collect at the toe
   of the berm slope.
•  A gravel underdrain was constructed beneath the liner in the wet areas.  Water
   collected in the underdrain flows through a water collection pipe that penetrates the
   berm.

The damaged liner was reconstructed.

Lessons Learned for Future Projects:  Based on the available information, the following
lesson can be learned from this case history:

•  Surface-water runoff should be managed to reduce foundation problems during and
   after construction. Runoff should  not be allowed to pond near the cell, where it can
   infiltrate into the cell.

F-A.9.4   L-31

Problem Classification:  landfill liner system displacement/design
Region of U. S.:  northeast
Waste Type: MSW
Reference: unpublished

Problem Summary: uplift of composite liner by surface-water infiltration during
construction

Problem Description: A single-composite liner system for one cell of a landfill expansion
was constructed with 16.7-m high side slopes.  The bottom 7.6 m of the side slope liner
system was constructed in an excavation against native soil having a specified
maximum  hydraulic conductivity of 1 x 10~7 m/s. The upper 9.1 m of the side slope liner
system was constructed against a more permeable rocky mine spoil berm.  The liner
consists of the following components, from top  to bottom:
                                      F-169
-------
•  GCL composed of a 1.5-mm thick HOPE GM and a bentonite layer glued to one side
   of the GM; and
•  0.9-m thick CCL (specified maximum hydraulic conductivity of 1 x 10~9 m/s).

The GCL was installed with the bentonite side down and was seamed by fusion
seaming the GM component of adjacent panels.

A surface-water diversion ditch was located in the mine spoil on top of the berm. The
ditch was used to convey runoff and rainwater that ponded in the cell during
construction.  Some of the water collected in the ditch infiltrated into the mine spoil
berm, pooled under the GCL, and saturated the GCL and CCL.  When the GCL was cut
for the water to drain and the damaged GCL was removed, a 0.6-m diameter cavity was
found in the berm soils and CCL.  The bottom of the cavity was located near the
interface of the mine spoil and native soil.  Apparently, the relatively high rate of water
infiltration through the ditch and into the mine spoil caused erosion of the mine spoil and
CCL where the water exited the soil and flowed beneath the GCL.

Resolution: The liner was repaired, and the ditch was lined with clay to reduce
infiltration.

Lessons Learned for Future Projects: Based on the available information, the following
lessons can be learned from this case history:

•  Surface-water runoff must be managed to reduce foundation problems during and
   after construction. Temporary and permanent surface-water diversion structures
   located near a cell may need to be lined to reduce infiltration, especially if the
   structures are located on relatively permeable soils and convey relatively large
   amounts of water.
•  Elementary geotechnical engineering principles must be followed for the design of
   berms. For example, adequate filters should be used to prevent internal erosion
   ("piping") in berms.

F-A.10    Cover System Construction

F-A.10.1  C-2

Problem Classification:  cover system construction/construction
Region of U. S.:  northeast
Waste Type: MSW
Reference: unpublished

Problem Summary: portion of topsoil from off-site source was contaminated with
chemicals
                                     F-170
-------
Problem Description: During placement of the topsoil surface and protection layer for a
landfill cover system, several truckloads of topsoil brought to the site by the contractor
had an aromatic odor.  The project specification for topsoil prohibited deleterious
material in the topsoil, so topsoil hauling was ceased until the affected soil could be
tested. Samples of the affected soil were collected and analyzed for VOCs and metals.
Based on the results of the testing, the soil was found to contain unacceptably high
concentrations of lead.

Resolution: Topsoil  that smelled aromatic or contained chemicals ionized by a
photoionization  meter was removed from the site.  Each truckload of topsoil
subsequently brought to the site was screened  using the above criteria.

Lessons Learned for Future Projects'.  Based on the available information, the following
lessons can be  learned from this case history:

•  EPA recommends that soil borrow sources be investigated by the owner unless the
   materials are supplied by a commercial materials company (Daniel and Koerner,
   1993).  In the case history described above, topsoil was excavated by the contractor
   from an off-site property. If the owner had required that testpits be excavated so the
   topsoil  could be inspected prior to construction, the topsoil contamination may have
   been identified earlier. The soil contamination also might have been identified earlier
   if the contractor had been required to submit chemical analyses on samples of the
   topsoil  brought to the site.
•  On-site CQA personnel should be trained to identify signs of soil  contamination.

F-A.10.2  C-16

Problem Classification: cover system construction/construction
Region of U. S.:  northeast
Waste Type:  MSW
Reference: Calabria, C.R. and Peggs, I.D., "Investigation of Geomembrane Seam
    Failures: Landfill Cover", Proceedings of the 10th GRI Conference Field
    Performance of  Geosynthetics and Geosynthetic Related Systems, Geosynthetic
    Research Institute, Philadelphia, PA, 1996, pp. 234-257.

Problem Summary:  high failure rate of HOPE GM  seam samples during destructive
testing

Problem Description: A 1-mm thick textured HOPE GM barrier was installed over an
MSW landfill between November 1994 and March  1995. The project specifications
required that destructive testing of the GM seams be performed by the installer; the
CQA consultant was only to monitor the installation.  Initially, only the inside track of
fusion seam samples was destructively tested in shear and peel by the installer.  The
project specifications, however, required both tracks of the fusion seam samples be
destructively tested.  After about 50% of the GM had been approved, based on passing
                                      F-171
-------
destructive tests, and this GM had been covered with a topsoil surface and protection
layer, the CQA consultant realized that the installer had not tested both seam tracks.
Archived fusion seam samples were subsequently obtained and tested. About 60%
(i.e., 25 of 42) of the archived seam samples and 49% (i.e., 44 of 90) of the seam
samples for the entire GM failed the peel test, primarily due to seam separation
exceeding the minimum specified value of 10%. Most of the failures were associated
with four of nine seaming machines and two of nine operators. Fifty percent (i.e., 6 of
12) of the extrusion seam samples taken from the section of GM not covered with
topsoil  also failed. These failure frequencies for fusion and extrusion seam samples do
not include samples collected and tested to isolate poor quality seams. The installer
attributed the high seam sample failure frequency to BTEX in landfill gas being
absorbed by the HOPE and inhibiting the formation of good seams.  However, after the
installer sent a new supervisor to the site, the failure rate for extrusion seams dropped.
In the section of GM covered with topsoil, only about 7% (i.e., two of 30) of the extrusion
seam samples failed destructive testing.  Interestingly, the same installer had placed an
HOPE  GM barrier over an adjacent section of the landfill about one year earlier and only
had about 10% of seam samples failing destructive testing.

Calabria and Peggs (1996) described their investigation to determine if the amount of
BTEX absorbed by the HOPE GM impacted the seam quality at the site. The
investigation included obtaining archived seam samples for destructive testing and
microstructural examination and analyzing GM from the site for BTEX constituents.
They also exposed site-specific GM samples to BTEX, seamed them, and tested them
in peal  and shear. Calabria and Peggs found that most of the archived fusion seam
samples showed rippling along the seam tracks and extensive warping. They attributed
the ripples to GM overheating from setting the seaming machine temperature too high
and speed too  low. They attributed the warp to manual adjustment of the seaming
machine to change its direction. They also noted that the GM at the outer edge of the
seam tracks was notched, creating a location where stresses could be concentrated,
which could potentially lead to stress cracking.  Over 50%  of the archived portions of
seam samples that had previously failed destructive testing failed the peel test when
tested by Calabria and Peggs. Some of these seams had  linear features oriented along
the length of the seam  in areas of the seams where the GM was shiny and not heated
sufficiently to melt its surface.  Calabria and Peggs attributed these linear features to
soil particles being dragged along the seam by the hot wedge of the seaming machine.
A number of the extrusion seam samples could be separated by hand. Some of the
these samples contained soil particles in  the extrusion seams.  Others showed evidence
of GM overheating and lack of bonding of the extruded bead to the top or bottom GM.

Selected seam samples from the installed GM were collected and analyzed for BTEX
constituents and subjected to peel testing. None of the constituents was detected at a
concentration greater than 1 mg/L.  No relationship was found between constituent
concentration and seam failure rate.  Site-specific GM samples exposed to BTEX,
seamed, and then tested them in peal were found to have  good quality seams.  Only
four of  50 individual samples tested in peel did not meet the project specification.
                                     F-172
-------
Calabria and Peggs noted that these few failures appeared to be related to the
cleanliness of the GM in the area being seamed.

Based on their investigation, Calabria and Peggs concluded that the high failure rate for
GM seam samples was predominantly caused by soil in the seams (i.e.,  inadequate
cleaning prior to seaming).  Other causes of failure were overheating and, for extrusion
seams, inadequate grinding. The BTEX absorbed by the GM had no apparent impact
on seam quality.

Resolution: The failed seams were isolated and repaired.

Lessons Learned for Future Projects:  Based on the available information, the following
lessons can be learned from this case history:

•  It is important to store archive portions of GM seam samples taken for destructive
   testing.
•  The absorption of relatively low concentrations of BTEX by HOPE GM does  not
   appear to affect the quality of seams subsequently constructed.
•  HOPE GM must be thoroughly cleaned along a seam path before the seam  is
   constructed since dirt in the seam  adversely impacts seam integrity.
•  Dual track fusion seaming machines are designed to make high quality seams along
   two tracks.  Both tracks should be destructively tested since failure of one track is
   generally indicative of overall seaming problems, and failure of one track can
   increase the stress concentrations in the adjacent track.  In addition, seaming
   problems may be identified  and corrected quicker.

F-A.11     Cover System Degradation

F-A.11.1   C-1

Problem Classification: cover system degradation/design
Region of U. S.:  northcentral
Waste Type:  MSW
Reference: Harris, J.M., Rivette, C.A., and Spradley, G.V., "Case Histories of Landfill
    Erosion Protection Using Geosynthetics", Geotextiles and Geomembranes, No. 11,
    1992, pp. 573-585.

Problem Summary: failure of geosynthetic erosion mat-lined downchute on 3H:1V side
slope

Problem Description:  Harris et al. (1992) described the failure of a geosynthetic erosion
mat-lined downchute on the cover system of a landfill in Missouri.  An erosion mat was
used to line one downchute that conveyed runoff from approximately 2 ha of cover
system and 8 ha of adjacent property; riprap was used to line the remaining three
downchutes that drained a total of about 10 ha. The erosion mat was a polyethylene,


                                     F-173
-------
three-dimensional, grass reinforcement type (i.e., turf reinforcement and revegetation
mat (TRM)). The mat-lined downchute was installed on the top slope about 3 m from
the slope crest, down the side slope, and along a perimeter section of the landfill.  At the
start, the downchute slope is about 5%, and runoff is diverted into the downchute by
small diversion berms. The downchute grade increases to 33% on the side slope.  Near
the slope toe, the downchute has a more gentle inclination of about 8%.  Riprap was
placed in the downchute at this transition for energy dissipation.  Mat was supplied in
rolls that were typically 1.5 m wide and 30 m long. Adjacent rolls were overlapped at
least 75 mm and secured to the underlying soil with 0.2 m long staples installed at  0.75
m spacings. Roll ends overlapped a minimum of 0.45 m and were shingled downward.
Mat was also  anchored in 0.3 m deep trenches at the top of each roll and along the
sides of the downchute. After the mat was placed, it was seeded and covered with
about 13 mm  of topsoil. Within one  month after construction, following a series of
significant rainfall events, the channel was unserviceable.  Soil had raveled along the
sides of the downchute, soil had eroded underneath the mat and along mat panel
overlaps, and the mat had moved downslope about 2 m.  Failure of the mat appeared to
have started at the top of the slope and progressed downward. Though grass was
becoming established across the cover system by this time, there was little grass in the
downchute at the time of failure.

The most severe damage to the downchute is believed to have occurred after a peak
rainfall intensity of about 64 mm/hr, estimated to represent the peak rainfall from a 1-hr
storm with a 5-year recurrence interval.  The peak runoff from this storm in the
downchute on the side slope was estimated by Harris et al. (1992) to be 1.33 m3/s. The
corresponding peak velocity in the downchute was calculated to be 2.9 m/s. A design
chart used by Harris et al. (1992) indicated that limiting velocities in bare and fully
grassed mat-lined channels were about 3 and 5 m/s, respectively, for a flow duration of
1 hour.  Subsequently, a detailed  laboratory testing program was conducted to evaluate
the relationship between flow velocity and erosion of a  mat-lined  surface for a simulated
flow duration of 0.5 hr. The results of the study  indicated that fully-grassed mat-lined
channels had noticeable erosion at flow velocities of about 5 m/s. However, without
grass, the velocity required to develop noticeable erosion was about 3 m/s. Harris et al.
(1992) concluded that the combination of large drainage area, steep slope, and the
inability of grass to sprout quickly in  the channel lead to failure of the downchute.

Resolution: The downchute was relined with  riprap, and topsoil was replaced in the
eroded areas.

Lessons Learned for Future Projects:  Based on the available information, the following
lessons can be learned from this case history:

•  Flow velocities in drainage channels under the design storm should be calculated so
   the appropriate channel lining can be selected. If an erosion mat is selected for a
   channel and  the erosion mat cannot withstand the design flow velocities until grass
                                      F-174
-------
   is established, significant maintenance and/or failure of the downchute should be
   anticipated.
•  If the downchute had been constructed earlier, within the plant growing season, the
   grass may have become established faster and erosion of the downchute may have
   been less severe. The erosion mat was  installed and seeded in the fall, when plant
   growth is relatively low, resulting in an extended period with poor to no grass cover
   in the downchute. The average plant growing season at the site starts in April and
   ends in October, the month in which construction of the downchute was completed.

F-A.11.2  C-12

Problem Classification: cover system degradation/design
Region of U. S.:  northeast
Waste Type: MSW
Reference:  unpublished

Problem Summary:  erosion of topsoil layer  on 60 m long, 31-1:1 V side slope

Problem Description: A 16 ha landfill cover  system was constructed with 60 m long,
3H:1V side slopes that were unbenched.  The design called for sand berms to divert
surface-water runoff from the top of the landfill to six riprap-lined downchutes on the
side slopes.  Sand diversion berms were also located about midway down the side
slopes on the east side of the landfill and  near the toe of the side slopes on the west
side of the landfill. The cover system consists of the following components, from top to
bottom:

•  vegetated topsoil layer, 0.2 m thick on top slopes and 0.3 m thick on side slopes;
•  sand drainage layer with a specified minimum hydraulic conductivity of 1 x 10~5 m/s,
   0.2 m thick on top slopes and 0.4 m thick on top slopes; and
•  1-mm thick HOPE GM barrier.

Within three  years after construction, about  0.8 ha of the cover system was severely
eroded and 0.07 ha of cover system soils had slid downslope.  Sixteen deep gullies
developed on the landfill side slopes in the vicinity of the riprap-lined downchutes and in
areas where the sand berms at the side slope crest were breached. Gullies typically
started near  the slope crest and propagated downslope. The gullies extended through
the topsoil and sand drainage layers down to the GM barrier.  In several locations, the
GM was damaged by punctures and tears, and the subgrade beneath the GM was
irregular. EPA HELP model simulations conducted after the erosion was observed
indicated that the sand drainage layer on  the landfill top slopes and side slopes had
insufficient capacity to convey surface-water infiltration  from the 25-year, 24-hour storm.
Under this condition, the lateral drainage that could not be conveyed within the drainage
layer flowed  through the  overlying topsoil  layer and as surface flow.  Seepage pressures
in the sand drainage layer and topsoil layer  increase surface erosion. Other project
details that contributed to the development of erosion and gullies at the site include: (i)
                                      F-175
-------
sand diversion berms and downchutes did not intercept lateral flow in the sand drainage
layer; (ii) runoff collected by berms and downchutes could infiltrate through the topsoil
layer and enter the drainage layer; and (iii) a lack of access control resulted in
unauthorized trafficking of four-wheel drive vehicles or dirt bikes on the landfill.

In two areas, the cover system topsoil moved downslope, causing longitudinal cracks
through the topsoil layer and into the sand drainage layer at the crest of the slide areas
and compression ridges at the toe of the areas. The GM beneath the slide areas was
undamaged. The slides are attributed to the development of high seepage pressures in
the sand drainage layer. One area of sliding was located near the top of the east side
slope and was approximately 20 m wide x 3 m long; the other area was located near the
middle of the west side slope and was about 45 m wide x 15 m long.

Resolution:  The following corrective measures were implemented:

•  vegetated swales, underlain by HOPE GM, were constructed along the side slope
   crest to collect runoff and water in the sand drainage layer at the top of the landfill
   and direct it to the downchutes;
•  the cover system soils were replaced and the damaged GM was repaired within the
   gullies and  slide areas; and
•  a chain link fence was installed around the perimeter of the  landfill to limit vehicle
   access.

Lessons Learned for Future Projects'. Based on the available information, the following
lessons can  be learned from this case history:

•  The surface-water runoff management strategy for this landfill, which did not
   effectively divert the flow of water from the  top of the landfill to the downchutes and
   allowed uninterrupted sheet flow over the 60-m  long, 3H:1V side slopes, proved
   inadequate to prevent erosion. A design that incorporated runoff interceptors on the
   side slopes, such as benches or swales, would  likely have been more effective in
   limiting erosion.
•  Cover system drainage layers should be designed to handle the total anticipated
   flow. Hydraulic requirements for the drainage layer should be evaluated using water
   balance calculations or other appropriate analyses (e.g., Giroud and Houlihan,
   1995).  Soong and Koerner (1997) recommend  using a short-duration intensive
   storm in the water balance and do not recommend the EPA HELP computer model
   for this purpose.  The drainage layer flow rates output from the HELP model are an
   average for a 24-hour period and are less than the peak flow rates.  Thus, the HELP
   model values are somewhat unconservative.
                                      F-176
-------
F-A. 12    Cover System StabiIity

F-A.12.1  C-3

Problem Classification:  cover system stability/construction
Region of U. S.:  northeast
Waste Type:  MSW
Reference: Paulson, J.N., "Veneer Stability Case Histories: Design Interactions
    Between Manufacturer/Consultant/Owner", in Proceedings of the 7th GRI Seminar
    Geosynthetics Liner Systems: Innovations, Concerns, and Designs, Geosynthetic
    Research Institute, Philadelphia, PA, 1993, pp. 235-241.

Problem Summary: sliding along nonwoven GT/GM interface during construction

Problem Description: Paulson (1993) described a cover system slope failure. The
cover system  consists of the following components, from top to bottom:

•  topsoil surface and protection layer;
•  geosynthetic reinforcement;
•  nonwoven GT cushion; and
•  GM barrier.

The side slopes are about 9 to 27 m long; the slope inclination was not given by
Paulson. The design specified that the geosynthetic reinforcement be secured on the
top of the landfill by extending the reinforcement onto the top and covering it with a 0.9-
m thick topsoil layer.  Slope stability analyses were conducted assuming topsoil would
be placed over the reinforcement from the bottom of the slopes upward.  However, this
condition was not incorporated into the construction specifications. When construction
began, access to the bottom of the side slopes was not available. So the contractor
started placing topsoil from the crest of the slope downwards.  Shortly afterwards, about
a 50 m wide by 20 m long section of soil covered cover system slid along the GT/GM
interface.

Resolution: The cover system in the failed area was repaired  by placing new
geosynthetic reinforcement and GT layers over the GM barrier, and placing the topsoil
over the GT from the bottom of the side slopes upward.

Lessons Learned for Future Projects: Based on the available information, the following
lessons can be learned from this case history:

•  EPA recommends that soil layers on side slope geosynthetics generally be placed
   from the toe of slope upward to avoid tensioning the geosynthetics (Daniel and
   Koerner, 1993).   If the construction specifications for the cover system described
   above had included this requirement, the cover system may not have slid.
                                      F-177
-------
•  The construction-related assumptions made during design should be incorporated
   into the construction specifications.
•  All anticipated loads should be considered and incorporated into the design.

F-A.12.2  C-4

Problem Classification:  cover system stability/design
Region of U. S.:  southeast
Waste Type:  MSW
References: Bonaparte, R., Othman, M.A., Rad, N.R., Swan, R.H., and Vander Linde,
    D.L., "Evaluation of Various Aspects of GCL Performance", Appendix F in Report of
    1995 Workshop on Geosynthetic Clay Liners, Daniel, D.E. and Scranton, H.B.,
    Editors, EPA National Risk Management Resource Laboratory, Cincinnati, OH,
    EPA/600/R-96/149,  1996, pp. F-1-F-34.
    Vander Linde, D.L., Luettich, S.M., and Bonaparte, R., "Lessons Learned From the
    Failure of a  Landfill  Cover System", Geosynthetics: Lessons Learned From Failures,
    IFAI, to be published in 1998.

Problem Summary: sliding along topsoil/GCL interface after rainfall

Problem Description: Bonaparte et al. (1996) and Vander Linde et al. (1998) described
the slope stability failure of a cover system in northern Georgia that occurred shortly
after construction. The  cover system consisted of the following components:
   0.3-m thick silty sand topsoil surface and protection layer (hydraulic conductivity in
   the range of 10~6 to 10~5 m/s); and
   stitch-bonded reinforced GCL barrier.
The 3H:1V cover system side slopes are up to 54 m long.  During design, the slope
stability factor of safety for the cover system was calculated to be greater than 1.3
assuming no seepage pressures in the topsoil layer and a topsoil/GCL secant interface
friction angle of 24°. This interface shear strength was selected by the engineer based
on information provided by the GCL manufacturer.

Within three months after construction, the cover system experienced two major
episodes of downslope movement of the topsoil layer over the GCL. The first major
slide occurred about one month after the end of construction; the movement occurred
after a three-day rainfall of 58 mm. The slide was limited to a relatively small area on
the bottom half of the side slopes. The design engineer believed that the movement
was the result of erosion.  Consequently, the contractor was directed to repair the area,
but reportedly was not able to complete the work due to the wet/soft condition of the
cover system.  The next major slide occurred about six weeks later, after a 41 mm
rainfall occurred at the site over a two-day period.  Numerous slides occurred on the
upper and lower portions of the side slopes. (A slide was characterized by a section of
topsoil moving monolithically over the GCL.)  Horizontal cracks through the topsoil layer


                                      F-178
-------
and  parallel to the slope crest were apparent near the top of the section and
compression ridges, up to about 0.45 m high, were present at the bottom of the section.
The total downslope movement during these two episodes exceeded 1 m at some
locations.  By the end of these two sliding episodes, the topsoil on about 50% of the
landfill side slopes had slid downslope.
Figure F-A.12.1. Cover system at the end of the second sliding episode.

A forensic investigation was conducted to identify the cause of the cover system failure.
The topsoil was removed in part of the slide zone and the GCL was inspected.  Though
the GCL was hydrated and under a low overburden stress, the bentonite in the  GCL
was not visibly extruded through the upper GT. The GCL was intact and did not appear
to have moved. The soil over the GCL was saturated and did not drain freely. A back
analysis of the cover system slope stability was conducted which accounted for the
effect of rainfall-induced seepage pressures. The hydraulic heads of water in the topsoil
layer during the two episodes  of downslope movement were estimated by several
different calculation methods to range from 0.15 to 0.3 m. These heads were used to
                                     F-179
-------
calculate seepage pressures in the topsoil layer. The back analysis was initially
conducted using secant interface friction angles of 20°, 24°, and 26° for the topsoil/GCL
interface. A slope stability analysis was conducted and the following results were
obtained using the equations presented by Giroud et al. (1995a):

                                        Factor of Safety
    Friction Angle        Head = 0 m      Head = 0.1  m       Head = 0.2 m
        20°                 1.09           0.84                  0.60
        24°                 1.35           1.04                  0.73
        26°                 1.47           1.13                  0.80

Shear strength tests performed on the topsoil/GCL interface after the completion of the
back analyses resulted in peak and large-displacement secant friction angles of 23° and
21°.  Based on these results and on the calculated slope stability factors of safety
presented above, the episodes of cover system sliding were primarily attributed to
seepage pressures in the topsoil.

Resolution: The cover system was redesigned to consist of the following components:

•  0.6-m thick topsoil surface and protection layer;
•  GC drainage layer; and
•  HOPE GM barrier.

This redesign has not yet been approved by the landfill owner and may be modified. As
of almost three years since the failure, no improvements have been made to the cover
system.

Lessons Learned for Future Projects'.  Based on the available information, the following
lessons can be learned from this case history:

•  Cover system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the anticipated  seepage pressures in the cover
   system and the actual shear strengths of the cover system materials and interfaces.
   Both of these considerations probably contributed to the cover system failure
   described above. The hydraulic head of water  in the topsoil layer should be
   evaluated using water balance calculations or other appropriate analysis method
   (e.g., Giroud and Houlihan, 1995).  Soong and  Koerner (1997) recommend using a
   short-duration intensive storm in the water balance and do not recommend the EPA
   HELP computer model for this purpose.  The drainage layer flow rates output from
   the HELP model are an average for a 24-hour period and may be much less than
   the peak flow rates calculated using other methods if the precipitation data used in
   the HELP model are not carefully selected.
•  Cover systems incorporating a low-permeability barrier layer should include a
   drainage layer above the barrier. A drainage layer is typically included when the
                                      F-180
-------
   cover system side slopes are steeper than 5H:1V (EPA,  1994). For the case history
   described above, cover system failure would likely have  been prevented if the cover
   system design had included an adequate drainage layer.
•  The hydraulic head of rainwater in the cover system topsoil layer is a function of
   drainage path length.  A design that incorporated topsoil seepage interceptors, such
   as benches or swales, on the 54 m long, 3H:1V side slopes would likely have
   reduced the potential for slope failure.
•  Though the topsoil/GCL interface shear strength assumed for design was within the
   range of strengths reported in the technical literature, it was higher than the strength
   measured in the laboratory after the failure. An error or one or two degrees on the
   interface friction angle can have a significant impact on the factor of safety. Actual
   interface strengths can only be assessed by project-specific testing.  Such testing is
   recommended.

F-A.12.3  C-5

Problem Classification:  cover system stability/design
Region of U. S.:  northeast
Waste Type:  MSW
Reference: Boschuk, J. Jr., "Landfill Covers An Engineering Perspective", Geotechnical
    Fabrics Report, IFAI,  Mar 1991, pp. 23-34.

Problem Summary: sliding along sand/woven GT interface after rainfall

Problem Description: Boschuk (1991) described a cover system slope failure.  The
cover system consists of the following components, from top to bottom:

•  topsoil surface and protection layer;
•  medium-coarse sand drainage layer;
•  woven GT reinforcement layer; and
•  GM barrier.

Project-specific interface direct shear testing was not performed.  For design, the
sand/GT interface secant friction angle was assumed to be 24°.  After a significant
rainfall, the soil layers slid off the GT on the 3H:1V side slopes.

Boschuk (1991) did not indicate if a slope stability analysis was performed as part of the
cover system design. Since the failure appeared to coincide with rainfall, seepage
pressures in the cover system soils probably contributed to the failure. An infinite slope
analysis was  conducted by the authors of this appendix.  Using the assumed sand/GT
interface friction angle of 24°, which is at  the lower end of the range of secant friction
angles reported for this interface in the technical literature, the calculated slope stability
factor of safety is 1.34 with no seepage pressures in the cover system, 0.98 with
seepage in the sand layer, and 0.63 with  full seepage in the sand and topsoil layers.
                                      F-181
-------
Resolution: The method of repair was not given by Boschuk.

Lessons Learned for Future Projects: Based on the available information, the following
lessons can be learned from this case history:

•  Cover system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the anticipated seepage pressures in the cover
   system and the actual shear strengths of the cover system materials. Both of these
   considerations probably contributed to the cover system failure described above.
•  Cover system drainage layers should be designed to handle the total anticipated
   flow. Hydraulic requirements for the drainage layer should be evaluated using water
   balance calculations or other appropriate analysis method (e.g., Giroud and
   Houlihan, 1995).  Soong and Koerner (1997) recommend using a short-duration
   intensive storm in the water balance and do not recommend the EPA HELP
   computer model for this purpose.  The drainage layer flow rates output from the
   HELP model are  an average for a 24-hour period and may be much less than the
   peak flow rates calculated using other methods if the precipitation data used in the
   HELP model are  not carefully selected.  It is not clear if the drainage layer hydraulic
   requirements were evaluated for the cover system described above.
There is little available information for this case history; additional lessons might have
been learned if the information had been complete.

F-A.12.4  C-6

Problem Classification: cover system stability/design
Region of U. S.:  northeast
Waste Type: MSW
Reference:  Boschuk, J. Jr.,  "Landfill Covers An  Engineering Perspective", Geotechnical
    Fabrics Report, IFAI, Mar 1991, pp. 23-34.

Problem Summary: sliding along sand/GM interface after rainfall

Problem Description: Boschuk (1991) described a cover system slope failure. The
cover system consists of the following components, from top to bottom:
•  topsoil surface and protection layer;
•  sand drains
•  GM barrier.
sand drainage layer (specified minimum hydraulic conductivity of 1 x 10~4 m/s); and
The 3H:1 V cover system side slopes are over 60 m long. After three days of rainfall,
the lower portions of the slopes became saturated and the soil layers slid downslope
along the sand/GM interface.

Boschuk (1991) did not indicate if interface direct shear testing or a slope stability
analysis were performed as part of the cover system design.  Since the failure appeared


                                      F-182
-------
to coincide with rainfall, seepage pressures in the cover system soils probably
contributed to the failure.  An infinite slope analysis was conducted by the authors of
this appendix.  In this analysis, the secant friction angle for the sand/GM interface was
assumed to be 20°, which is within the range of friction angles reported for this interface
in the technical literature.  The calculated slope stability factor of safety is 1.09 with no
seepage pressures in the soil  layers and about 0.80 with full seepage in the sand layer.
Being an infinite slope stability analysis,  these values do  not take into account the toe
buttressing effect.

Resolution: The cover system was reconstructed with benches, so that the maximum
3H:1V cover system slope length was reduced.  At each  bench, collection pipes were
installed to drain the water from the sand drainage  layer onto the bench. It is not known
if a slope stability analysis or drainage layer design calculations were performed for this
redesign.

Lessons Learned for Future Projects:  Based on the available information, the following
lessons can be learned from this case history:

•  Cover system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the anticipated seepage pressures in the cover
   system drainage layers and the actual shear strengths of the cover system
   materials.  Both of these considerations probably contributed to the cover system
   failure described above.
•  Cover system drainage layers should be designed to  handle the total anticipated
   flow. Hydraulic requirements for the  drainage layer should be evaluated using water
   balance calculations or other appropriate analysis method (e.g., Giroud and
   Houlihan, 1995). Soong and Koerner (1997) recommend  using a short-duration
   intensive storm in the water balance  and do not recommend the EPA HELP
   computer model for this purpose.  The drainage layer flow rates output from the
   HELP model are an average for a  24-hour period and may be much less than the
   peak flow rates calculated  using other methods if the  precipitation data used in the
   HELP model are not carefully selected.  It is not clear if the drainage layer hydraulic
   requirements were evaluated for the  cover system described above.
•  The hydraulic head of rainwater in the cover system topsoil layer  is a function of
   drainage path length.  A design that incorporated topsoil seepage interceptors, such
   as benches or swales, on  the over 60 m long, 3H:1V side slopes would likely have
   reduced the potential for slope failure.

There is little available information for this case history; additional lessons might have
been learned if the information had  been complete.

F-A.12.5  C-7

Problem Classification: cover system stability/design
Region of U. S.:  northeast
                                      F-183
-------
Waste Type:  MSW
Reference: Boschuk, J. Jr., "Landfill Covers An Engineering Perspective", Geotechnical
    Fabrics Report, IFAI, Mar 1991, pp. 23-34.

Problem Summary: sliding along gap-graded sand/GM interface after rainfall

Problem Description: Boschuk (1991) described a cover system slope failure that
appeared to be related to clogging of the sand drainage layer. The cover system
consists of the following components, from top to bottom:

•  topsoil surface and protection layer;
•  gap-graded sand   drainage  layer (specified minimum hydraulic  conductivity  of
   1 x10-4m/s); and
•  GM barrier.

The cover system side slopes are about 50 to 90 m long.  In less than one year after
construction, the entire  lower third of the cover system slope slid downslope along the
sand/GM interface over several rainfall events.  The sand drainage layer in the slide
zone contained significant fines, presumably washed into  the sand from the topsoil layer
and the sand  in the upper two-thirds of the side slope.

Boschuk (1991) did not indicate if filter design calculations or laboratory tests were
performed to evaluate whether the topsoil would be retained by the sand.  He also did
not indicate if a slope stability analysis or interface direct shear testing were performed
as part of the cover system design.  Since the failure appeared to coincide with clogging
of the sand drainage layer and rainfall and since failure occurred at the upper interface
of the GM, seepage pressures in the cover system soils probably contributed to the
failure.  An infinite slope analysis was not conducted by the  authors of this appendix
because the slope inclination was not given by Boschuk.

Resolution: The cover system drainage layer on the side  slope was reconstructed with
an improved drainage layer consisting of, from top to bottom:

•  uniformly-graded sand drainage layer;
•  GT filter; and
•  GN drainage layer.

No information is available on the method used to design  the improved drainage layer.

Lessons Learned for Future Projects: Based on the available information, the following
lessons can be learned from this case history:

•  Gap-graded soils are more prone to  migration of finer-sized particles (i.e., internal
   instability) than continuously-graded soils. As shown in the case history described
   above, this particle migration may result in clogging of the soil.  Therefore, if gap-


                                      F-184
-------
   graded soils are used as drainage materials, the effect of particle migration should
   be evaluated during design.  For example, filter design can be performed to assess if
   the coarser particles will retain the finer particles or laboratory tests can be
   performed to assess the effect of particle migration on hydraulic conductivity.
•  In the case history described above, the sand should act as a filter for the overlying
   topsoil.  Filter layers should be designed to be compatible with the upgradient soil
   using filter criteria and/or laboratory testing.  It is not clear if this was done for the
   cover system described above.
•  Cover system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the anticipated seepage pressures in the cover
   system drainage layers and the  actual shear strengths of the cover system
   materials. Both of these considerations probably contributed to the cover system
   failure described above.

There is little available information for this case history; additional lessons might have
been learned if the information had  been complete.

F-A.12.6  C-8

Problem Classification:  cover system stability/design
Region of U. S.:  northeast
Waste Type:  MSW
Reference:  Boschuk, J. Jr., "Landfill Covers An Engineering Perspective", Geotechnical
    Fabrics Report, IFAI, Mar 1991, pp. 23-34.

Problem Summary: sliding along gravel/GM interface during construction

Problem Description: Boschuk (1991) described a cover system that could not be
constructed due to slope failure. The cover system, as designed, consisted of the
following components, from top to bottom:

•  topsoil surface and protection layer;
•  pea gravel drainage layer; and
•  GM barrier.

During construction, the gravel placed on the 3H:1V side slope continually slid down the
slope, eventually damaging the GM. The Contractor had tried to place the gravel by
pushing it up the slope with a bulldozer and by placing it on the slope using a clamshell
bucket, but neither method worked. The length of the side slope was not given by
Boschuk.

Boschuk (1991) did not indicate if interface direct shear testing or a slope stability
analysis were performed as part of  the cover system design. However, from an infinite
slope analysis, the cover system would  be unstable if the secant friction angle for the
                                      F-185
-------
gravel/GM interface was less than the slope angle, or 18.4°.  This value is within the
range of friction angles reported for a gravel/GM interface in the technical literature.

Resolution: The method of repair was not given by Boschuk.

Lessons Learned for Future Projects'. Based on the available information, the following
lesson can be learned from this case history:

•  Cover system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the actual shear strengths of the cover system
   materials.

There is little available information for this case history; additional lessons might have
been learned if the information had been complete.

F-A.12.7  C-9

Problem Classification: cover system stability/design
Region of U. S.:  northeast
Waste Type:  MSW
Reference: Boschuk, J. Jr., "Landfill Covers An Engineering Perspective", Geotechnical
    Fabrics Report, IFAI,  Mar 1991, pp. 23-34.

Problem Summary: sliding along sand/calendered nonwoven GT interface during
construction

Problem Description: Boschuk (1991) described a cover system slope failure.  The
cover system consists of the following components, from top to bottom:

•  topsoil surface and protection layer;
•  sand drainage layer;
•  calendered nonwoven GT cushion; and
•  GM barrier.

Project-specific interface direct shear tests between the sand and GT resulted in a
secant interface friction angle of about 21°. These tests were reportedly conducted at
confining stresses significantly greater than those  representative of field conditions.  As
topsoil was being placed over the already-installed sand drainage layer on 3H:1V
slopes, the cover system  slid along the sand/GT interface.

Boschuk (1991) did not indicate if a slope stability  analysis was performed as part of the
cover system design.  However, he noted that tilt table tests subsequently performed at
low stresses representative of those at failure gave an secant friction angle for the
sand/GT interface of about 18°. The differences in secant interface friction angles may
be attributed to variation in the tested geosynthetics, accuracy of the test methods, and

                                      F-186
-------
differences in the test conditions. An infinite slope analysis was conducted by the
authors of this appendix.  For an interface secant friction angles of 18°and 21 °, the
calculated slope stability factors of safety are 0.98 and 1.15, respectively, with no
seepage forces. Being an infinite slope stability analysis, these values do not take into
account the toe buttressing effect.

Resolution: The calendered nonwoven GT was replaced with a needlepunched
nonwoven GT.  The results of tilt table tests between the sand and needlepunched
nonwoven GT indicated that the needlepunched GT would be suitable.

Lessons Learned for Future Projects: Based on the available information, the following
lessons can be  learned from this case history:

•  Cover system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the  actual shear strengths of the cover system
   materials.
•  Actual  interface shear strengths can only be assessed by project-specific testing.
   The effects of variation in the tested geosynthetics, accuracy of test methods, and
   test conditions must be considered when selecting the interface shear strength to
   use in design.

There is little available information for this case history; additional lessons might have
been learned  if the information  had been  complete.

F-A.12.8  C-10

Problem Classification: cover system stability/design
Region of U. S.:  northeast
Waste Type:  MSW
Reference: Boschuk, J. Jr., "Landfill Covers An Engineering Perspective",  Geotechnical
    Fabrics Report, IFAI, Mar 1991, pp. 23-34.

Problem Summary: sliding along sand/GM interface after rainfall

Problem Description: Boschuk (1991) described a cover system slope failure.  The
cover system  consists of the following components,  from top to bottom:

•  topsoil surface and protection layer;
•  sand drainage layer; and
•  GM barrier.

Perforated pipes wrapped with  a nonwoven GT filter were installed in the sand drainage
layer to drain the collected water out of the  cover system. Eventually, fines clogged the
GT at the pipe perforations and water became trapped in the drainage layer.  Pore
pressures increased in the soils, and the  soils slid downslope. Failure occurred at the


                                      F-187
-------
sand/GM interface, primarily on the lower third of the slope. After the failure, the pipes
were dry and the surrounding sands were saturated.

Boschuk (1991) did not indicate if filter design, interface direct shear testing, or a slope
stability analysis were performed as part of the cover system design. The GT should
have been designed to be compatible with the sand using filter criteria calculations
and/or laboratory testing. Hydraulic gradients tests performed on the sand and GT after
the slope failure found that the  sand fines clogged the GT. A thinner, more open
nonwoven GT that let the fines pass but retained the sand, would have performed
better. An infinite slope analysis was conducted by the authors of this appendix. In this
analysis, the secant friction  angle for the sand/GM interface was assumed to be 20°,
which is within the range of friction angles reported for this interface in the technical
literature.  The slope  stability factor of safety is 1.09 with no seepage pressures in the
soil layers and about  0.80 with  full seepage in the sand layer.  Being an infinite slope
stability analysis, these values  do not take into account the toe buttressing effect.

Resolution: The pipes were removed and replaced with perforated pipes bedded in
gravel wrapped in a GT. The cover system was reconstructed  in the failed areas.
Boschuk did not indicate if the same type of GT was used. If it was,  clogging of the GT
may occur again.

Lessons Learned for Future Projects: Based on the available information, the following
lessons can be learned from this case  history:

•  GT filters should be designed to be compatible with the upgradient soil using filter
   criteria calculations and/or laboratory testing.
•  Cover system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the anticipated seepage pressures in the cover
   system drainage layers and the  actual shear strengths of the cover system
   materials. Both of these considerations probably contributed to the cover system
   failure described above.

There is little available information for this case history; additional lessons might have
been learned if the information  had  been complete.

F-A.12.9  C-11

Problem Classification: cover system stability/design
Region of U. S.: northeast
Waste Type:  MSW
Reference: Boschuk, J. Jr., "Landfill Covers An Engineering Perspective", Geotechnical
    Fabrics Report, IFAI, Mar 1991, pp. 23-34.

Problem Summary: sliding  along topsoil/nonwoven GT interface after rainfall
                                      F-188
-------
Problem Description: Boschuk (1991) described a cover system slope failure. The
cover system consists of the following components, from top to bottom:

•  topsoil surface and protection layer;
•  nonwoven GT filter;

•  gravel  drainage layer; and
•  GM barrier.

Over time, the GT became clogged by the topsoil and rainwater infiltrating the topsoil
did not drain freely into the underlying  gravel.  After rainfall, pore pressures increased in
the topsoil layer, and the topsoil slid downslope over the GT.  Failure occurred primarily
on the lower third of the  slope.

Boschuk (1991) did not indicate if filter design, interface direct shear testing, or a slope
stability analysis were performed as part of the cover system design.  The GT should
have been designed to be compatible with the topsoil using filter criteria calculations
and/or laboratory testing. Compatibility between a soil and a GT is more of a concern
when the soil is topsoil rather than clean sand because topsoil generally has a lower
degree of  internal stability than clean sand.  Internally unstable soils are likely have a
large amount of finer particles that move through the soil and  may result in clogging of
downgradient filters.  Topsoil is sometimes susceptible to this piping of fines because  it
generally has a significant amount of fines and it is placed loose. However,  this  effect
may be  counterbalanced by the cohesiveness of topsoils.  Soils with cohesive fines are
more  likely to be internally stable. An infinite slope analysis was conducted  by the
authors of this appendix. In this analysis, the secant friction angle for the topsoil/GT
interface was assumed to be 24°, which is within the range of friction angles reported  for
this interface in the technical literature.  The slope stability factor of safety is 1.34 with
no seepage  pressures in the topsoil layer and about 0.63 with full seepage in the topsoil
layer. Being an infinite slope stability analysis, these values do not take into account
the toe buttressing effect.

Resolution:  The method of repair was not given by Boschuk.

Lessons Learned for Future Projects'. Based on the available information, the following
lesson can be learned from this case history:

•  Cover  system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the anticipated seepage pressures in the cover
   system drainage layers and the actual shear strengths of the cover system
   materials. Both of these considerations probably contributed to the cover system
   failure  described above.
•  When  a filter layer is  used, the filter should be designed to be compatible with the
   upgradient soil using filter criteria calculations and/or laboratory testing.
                                       F-189
-------
There is little available information for this case history; additional lessons might have
been learned if the information had been complete.
F-A.12.10 C-13

Problem Classification: cover system stability/operation
Region of U. S.:  northeast
Waste Type:  MSW
Reference:  unpublished

Problem Summary: sliding along PVC GM/CCL interface after a thaw

Problem Description:  A landfill cover system was constructed in the fall.  It consisted of
the following components, from top to bottom:

•  0.1-m thick topsoil surface layer;
•  0.2-m thick sand protection layer;
•  needlepunched nonwoven GT filter;
•  0.3-m thick sand drainage layer;
•  0.75-mm thick PVC GM barrier; and
•  0.45-m thick CCL.

The side slope grade  is 3H:1V in some areas and 4H:1V in other areas.

During the winter, the cover system was covered with snow and the ambient
temperature was below freezing temperature until the spring.  In April, a slide occurred
at the GM/CCL interface, on  a 4H:1V slope, but no failure occurred in the  3H:1V slope
and in other parts of the 41-1:1 V slope. The slide area was approximately 26 m long in
the slope direction and 60 m wide.  Rupture of the GM occurred about 2 m downslope
of the slope crest, and the GM slid about 3 m downslope from the tear. Above the tear,
the GM remained secured in an anchor trench. When  the soil layers were removed
from the GM in the slide area so the GM could be inspected, the upper 6 m of GM was
torn into strips that covered only about 50% of the underlying CCL surface (i.e., there
were spaces between the strips). The middle 6 m of GM in the slide zone was flat and
not in tension. The lower 14 m of GM, near the slope toe, was very wrinkled,  occupying
a slope length of only 9 m. Based on the coverage of the GM strips in the upper 6 m of
the slope, the average GM strain at failure in this portion of the slope was  less than
100%. The average GM  strain in slide area was  12%.

Temperature records  showed that the slide occurred a few days after a first thaw
following the winter. The investigation showed that water could not exit from the sand
drainage layer because the lower end of the  drainage layer was blocked by ice and
snow accumulated at  the edge of the road located at the toe of the  landfill  cover slope.
As a result, the cause of the  slide was assumed to be the seepage  pressures that


                                      F-190
-------
developed when flow started after melting of the ice at the lower end of the drainage
layer. However, slope stability analyses showed that seepage pressures above a GM
have little effect on the factor of safety with respect to a slide that occurs at an interface
located beneath the GM (Giroud et al., 1995a).  This can be explained as follows. The
slope stability factor of safety equals the shear strength divided by the shear stress.
The shear stress acting above the GM is the same as that acting below the GM, with or
without seepage above the GM. The shear strength of an interface is dependent on the
interface adhesion, interface friction angle, and  effective normal stress at the interface.
The effective normal stress at an interface decreases as the seepage pressures at that
interface increase. With seepage pressures above a GM, the effective stress on top of
the GM decreases and, consequently, the shear strength at the interface above the GM
decreases. This decrease in shear strength leads to a decrease in the slope stability
factor of safety. Seepage pressures above a GM, however, do not impact  the effective
stresses at the interface below the GM. Thus, the shear strength of this interface and
the slope stability factor of safety are  unchanged with these seepage pressures.

With seepage forces identified as only a minor contributor to the slope failure, an
additional investigation was conducted to evaluate the effect of temperature fluctuations
on GM/CCL interface shear strength.  Interface  shear tests simulating the conditions
during the winter (-7°C) followed by thaw (+0.5°C) showed that the formation of ice
lenses at the GM/CCL interface at below-freezing temperature increased the water
content at the GM-CCL interface, resulting in a  marked decrease of the interface shear
strength after a thaw, compared to the interface shear strength before freezing. Slope
stability calculations showed that the cover system  on a 4H:1V slope at the site would
be unstable if the initial water content of the CCL was greater than 23%. Systematic
measurements of the water content of the CCL  showed that this condition was met by
the CCL water content in the area where the slide occurred, and showed that the CCL
water content was greater in the area where the slide occurred than in other areas.
This was attributed to the  heavy rainfall that preceded the installation of the GM in the
area where the slide eventually occurred.  The analyses also showed that the conditions
fora slide to occur in the 3H:1V slope were not  met.  The softening of the toe of the
soils overlying the GM  resulted in a decrease of the factor of safety due to partial loss of
toe support.  This effect was evaluated and was found to be less significant than the
effect of the decrease in interface shear strength.

The fact that a PVC GM had ruptured with an apparently small strain,  compared to the
typical 300% strain at break of PVC GMs, was investigated. Tests on the PVC GM  at
23°C showed that the PVC GM after six months in the ground was identical to the
original GM, and tensile tests conducted at 0.5°C showed that, under the conditions
after a thaw, the PVC GM had a yield strain of 9%,  which is much less than the strain at
break at 23°C.  This 9% yield strain explains that the observed rupture of the GM is
consistent with the observed displacements.

Resolution: In the slide area, the cover soil and  the GM were removed. The CCL was
reworked, and covered by soil layers  and GT as in the original design. However, no GM
                                      F-191
-------
was used in the repair of the slide area. This decision was made by the owner and the
reason for this decision is not documented.

Lessons Learned for Future Projects'. Based on the available information, the following
lessons can be learned from this case history:

•  Seepage forces,  even when they exist, should not be considered responsible for all
   slope stability failures. Seepage forces may have a marked effect on slides that
   occur along an interface located above the GM; however, they generally have no
   significant effect on slides that occur on interfaces located below the GM.
•  Freeze-thaw of CCLs can have a significantly detrimental impact on GM/CCL
   interface shear strength.
•  Excessive CCL water content, due for example to rainfall prior to GM placement, can
   have a detrimental impact on interface shear strength. The CCL construction
   specification should generally include limitations on maximum compacted moisture
   content and restrictions on applying supplemental moisture.
•  Outlets of drainage layers should be kept free of snow and ice.
•  At low temperatures, PVC GMs may break at a low strain (e.g., on the order of
   10%).

F-A.12.11 C-14

Problem Classification:  landfill liner system stability/construction
Region of U. S.:  unknown
Waste Type:  MSW
Reference: unpublished

Problem Summary: sliding along geogrid/HDPE GM interface during construction

Problem Description: During the construction of a cover system  over a landfill with
approximately 32 m  long, 31-1:1 V side slopes, a portion of the cover system failed. The
cover system  consists of the following components, from top to bottom:

•  0.15-m thick topsoil protection layer;
•  0.45-m thick sand drainage layer;
•  geogrid reinforcement layer;
•  smooth HOPE GM barrier;
•  GN gas venting layer; and
•  sand bedding layer.

The design specified that the reinforcement be secured on the top of the landfill by
extending the reinforcement onto the top and covering it with the 0.6-m thick topsoil and
sand layers. Slope stability analyses were conducted assuming  that the soil layers
                                      F-192
-------
would be placed over the reinforcement from the bottom of the slope upward.  However,
this condition was not incorporated into the construction specifications.

When construction began, not all of the geogrid rolls were secured at the top of the
slope because landfill gas wells were in the way. Where the gas wells interfered with
geogrid installation, the geogrid was only tied to adjacent geogrid on the side slope and
did not extend far beyond the slope crest. Access to the bottom of the side slopes was
limited  at some locations due to wetlands near the slope toe. As a consequence of
these conditions, the contractor placed a stockpile of sand over the geogrid on the side
slope near the crest and began placing the sand from the crest downward. Shortly after
sand placement began, the reinforcement snapped at the slope crest beneath the sand
stockpile and construction equipment placing the sand. The GM then tore near the
slope crest and along outward diagonals down the length of the GM on both sides  of the
stockpile. Several rolls of geogrid in the slide area had not been anchored. The torn
GM was subsequently inspected and found to have been damaged by the GN. The GN
abraded the  GM and in some areas broke through it.

Resolution:  The cover system was redesigned and reconstructed successfully. The
redesigned cover system consists of the following components, from top to bottom:

•  0.6-m thick gravel surface and protection layer (specified maximum particle size of
   50 mm);
•  GT cushion;
•  geogrid reinforcement layer;
•  textured HOPE GM barrier;
•  GC gas venting layer; and
•  sand bedding layer.

Lessons Learned for Future Projects: Based on the available information, the following
lessons can be learned from this case history:

•  EPA recommends that soil layers  on side slope geosynthetics generally be placed
   from the toe of slope upward to avoid tensioning  the geosynthetics (Daniel and
   Koerner,  1993). If the construction specifications for the cover system described
   above had included this requirement, the cover system may not have slid.
•  The construction-related assumptions made during design should be incorporated
   into the construction specifications. In the case history presented above, the
   specifications should have required that the geogrid reinforcement be anchored prior
   to placing the soil layer to be reinforced.
F-A.12.12 C-17

Problem Classification:  cover system stability/design
Region of U. S.:  unknown

                                      F-193
-------
Waste Type:  MSW
Reference: Soong, T.Y. and Koerner, R.M., "The Design of Drainage Systems Over
    Geosynthetically Lined Slopes", GRI Report #19, Geosynthetic Research Institute,
    Philadelphia, PA, 1997, 88 p.

Problem Summary: sliding along sand/CCL interface during rainfall

Problem Description: Soong and Koerner (1997) described the slope stability failure of
a cover system on a 40 m long, 2.51-1:1 V side slope that occurred in 1995. The cover
system consists of the following components, from top to bottom:
   0.75-m thick silty sand surface and protection layer (approximate hydraulic
   conductivity <
   CCL barrier.
conductivity of 1 x 10"5 m/s); and
Soong and Koerner (1997) do not indicate if slope stability analyses were performed for
design.  About two to three years after the cover system was constructed, the sand slid
downslope over the CCL during a storm.  The slide was relatively small and localized.
Soong and Koerner attributed the failure to seepage pressures that developed in the
sand layer.

An infinite slope analysis was conducted by the authors of this appendix. In this
analysis, the friction angle for the sand was assumed to be 30°. The calculated slope
stability factors of safety are 1.44 and 0.66 without and with full seepage pressures in
the sand layer.  Being an infinite slope stability analysis, these values do not take into
account the toe buttressing effect.

Resolution: The method of repair was not given by Soong and Koerner (1997).

Lessons Learned for Future Projects: Based on the available information, the following
lesson can be  learned from this case history:

•  Cover system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the anticipated seepage pressures in  the cover
   system drainage layers.
•  Cover system drainage layers should be designed to handle the total  anticipated
   flow.  Hydraulic requirements for the drainage layer should be evaluated using water
   balance calculations or other appropriate analysis method (e.g., Giroud  and
   Houlihan, 1995). Soong and Koerner (1997) recommend using a short-duration
   intensive storm  in  the water balance and do not recommend the EPA HELP
   computer model for this purpose.  The drainage layer flow rates output from the
   HELP model are an average for a 24-hour period and may be much less than the
   peak flow rates calculated using other methods if the precipitation data used in the
   HELP model are not carefully selected. It is not clear if the drainage layer hydraulic
   requirements were evaluated for the cover system described above.


                                      F-194
-------
There is little available information for this case history; additional lessons might have
been learned if the information had been complete.

F-A.12.13 C-18

Problem Classification:  cover system stability/design
Region of U. S.: unknown
Waste Type:  MSW
Reference: Soong, T.Y. and Koerner, R.M., "The Design of Drainage Systems Over
    Geosynthetically Lined Slopes", GRI Report #19, Geosynthetic Research Institute,
    Philadelphia, PA, 1997, 88 p.

Problem Summary: sliding along sand/CCL interface immediately after rainfall

Problem Description: Soong and Koerner (1997) described the slope stability failure of
a cover system on  a 50 m long, 3H:1V side slope that occurred in 1996. The cover
system consists of the following components, from top to bottom:

•  0.6-m thick topsoil surface and protection layer;
•  0.3-m thick sand drainage layer (approximate hydraulic conductivity of 1 x 10~4 m/s);
   and
•  CCL barrier.

Soong and Koerner (1997) do not indicate if slope stability analyses were performed for
design.  About five to six years after the cover system was constructed, the sand slid
downslope over the CCL immediately after a storm.  At least four localized slides
occurred.  Soong and Koerner attributed the failure to relatively high seepage pressures
that developed in the cover system because the  drainage layer hydraulic conductivity
was too low.

An infinite slope analysis was conducted by the authors of this appendix. In this
analysis, the friction angle for the sand was assumed to be 30°.  The calculated slope
stability factors of safety are 1.73 and 1.40 without and with full seepage pressures in
the sand layer. With seepage pressures in the sand and topsoil layers (i.e., the sand
drainage layer has insufficient capacity to convey all infiltration), the calculated factor of
safety is 0.77.  Being an infinite slope stability analysis, these values do not take into
account the toe buttressing effect.

Resolution: The method of repair was not given  by Soong and Koerner (1997).

Lessons Learned for Future Projects: Based on the available information, the following
lesson can be learned from this case history:
                                      F-195
-------
•  Cover system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the anticipated seepage pressures in the cover
   system drainage layers.
•  Cover system drainage layers should be designed to handle the total anticipated
   flow. Cover system drainage layers should be designed to handle the total
   anticipated flow.  Hydraulic requirements for the drainage layer should  be evaluated
   using water balance calculations or other appropriate analysis method  (e.g., Giroud
   and Houlihan, 1995). Soong and Koerner (1997) recommend using a short-duration
   intensive storm in the water balance and do not recommend the EPA HELP
   computer model for this purpose. The drainage layer flow rates output from the
   HELP model are an average for a 24-hour  period and may be much less than the
   peak flow rates calculated using  other methods if the precipitation data used in the
   HELP model are not carefully selected. It is not clear if the drainage layer hydraulic
   requirements were evaluated for the cover system described above.

There is little available information for this case history; additional lessons might have
been learned if the information had been complete.

F-A.12.14 C-19

Problem Classification: cover system stability/design
Region of U. S.:  unknown
Waste Type:  MSW
Reference:  Soong, T.Y. and Koerner, R.M., "The Design of Drainage Systems Over
    Geosynthetically Lined Slopes",  GRI Report #19, Geosynthetic Research Institute,
    Philadelphia, PA, 1997,  88 p.

Problem Summary:  sliding along sand/CCL interface after rainfall

Problem Description: Soong and Koerner (1997) described the slope stability failure of
a cover system on a 45 m long, 3H:1V side slope that occurred in 1996. The cover
system consists of the following components, from top to bottom:
•  0.75-m thick topsoil surface and protection layer;
•  0.3-m thick sand drainage layer; and
•  CCL barrier.

The design called for water collected  in the sand drainage layer to drain to the toe, be
collected in a gravel toe drain, and exit the cover system through a pipe.  The gravel
toe drain was not wrapped with a GT.  Soong and Koerner (1997) do not indicate if filter
design of the topsoil, sand, and gravel or slope stability analyses were performed for
design.
                                      F-196
-------
By about five to six years after the cover system was constructed, a number of localized
slides of the sand over the CCL had occurred.  When the gravel toe drain was
exhumed, the gravel was found to be very contaminated with fines, which presumably
migrated into the gravel from the overlying sand and topsoil.  Soong and Koerner
attributed the failure to relatively high seepage pressures that developed in the cover
system after the gravel toe drain became clogged.

An infinite slope analysis was conducted by the authors of this appendix.  In this
analysis, the friction angle for the sand was assumed to be 30°. The calculated slope
stability factors of safety are 1.73 and 1.45 without and with full seepage pressures in
the sand layer.  With seepage pressures in the sand and topsoil layers (i.e., the sand
drainage layer has insufficient capacity to convey all infiltration), the calculated factor of
safety is 0.76.  Being an infinite slope stability analysis, these values do not take into
account the toe buttressing effect.

Resolution: The method of repair was not given by Soong and Koerner (1997).

Lessons Learned for Future Projects'. Based on the available information, the following
lesson can be learned from this case history:

•  Adjacent soils conveying water should  be designed to be compatible with the
   upgradient soil using filter criteria and/or laboratory testing.
•  Cover system  slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the anticipated seepage pressures in the cover
   system drainage layers.
•  Cover system  drainage layers should be designed to handle the total anticipated
   flow.  Cover system  drainage layers should be designed to handle the total
   anticipated flow. Hydraulic requirements for the drainage layer should be evaluated
   using water balance calculations or other appropriate analysis method (e.g., Giroud
   and Houlihan,  1995). Soong and Koerner (1997) recommend  using a short-duration
   intensive storm in the water balance and do not recommend the EPA HELP
   computer model for this purpose. The  drainage layer flow rates output from the
   HELP model are an  average for a 24-hour period and may be  much less than the
   peak flow rates calculated using other methods  if the precipitation data used in the
   HELP model are not carefully  selected. It is not clear if the drainage layer hydraulic
   requirements were evaluated for the cover system described above.

There is little available information for this  case history; additional lessons might have
been learned if the information had been complete.

F-A.12.15 C-20

Problem Classification:  cover system stability/design
Region of U. S.:  unknown
Waste Type: MSW
                                      F-197
-------
Reference'. Soong, T.Y. and Koerner, R.M., "The Design of Drainage Systems Over
    Geosynthetically Lined Slopes", GRI Report #19, Geosynthetic Research Institute,
    Philadelphia, PA, 1997, 88 p.

Problem Summary: sliding along sand/CCL interface after rainfall

Problem Description: Soong and Koerner (1997) described  the slope stability failure of a
cover system on a 45 m long, 2.5H:1V side slope sections between benches that
occurred in 1996.  The cover system consists of the following components, from  top to
bottom:

•  0.6-m thick topsoil surface and protection layer;
•  0.2-m thick sand drainage layer; and
•  CCL barrier.

The design called for water collected in the sand drainage layer to drain to the toe, be
collected in a gravel toe drain, and exit the cover system through a pipe.  The pipe was
wrapped with a GT.  Soong and Koerner (1997) do not indicate if filter design or slope
stability analyses were performed for design.

By about four to five years after the cover system was constructed, a number of small
localized slides of the sand over the CCL had occurred. When the gravel toe drain was
exhumed,  the GT was found to be clogged with fines at pipe perforations. The fines
presumably migrated to the GT from the sand and topsoil.  Soong and Koerner
attributed the failure to relatively high seepage pressures that developed in the cover
system after the GT around the pipe became clogged.

An infinite slope analysis was conducted by the authors of this appendix. In this
analysis, the friction angle for the sand was assumed to be  30°. The calculated slope
stability factors of safety are 1.44 and 1.24 without and with full seepage pressures in
the sand layer. With seepage pressures in the  sand and  topsoil layers (i.e., the sand
drainage layer has insufficient capacity to convey all infiltration), the calculated factor of
safety is 0.63.  Being an infinite slope stability analysis, these values do not take into
account the toe buttressing effect.

Resolution: The method of repair was not given by Soong and Koerner (1997).

Lessons Learned for Future Projects: Based on the available information, the following
lesson can be learned from this case history:

•  Perforated pipes bedded in gravel should not be wrapped with a GT because the  GT
   is  useless,  and, in some cases, even detrimental (Giroud, 1996). Furthermore, EPA
   recommends that perforated pipes generally not be wrapped with a GT  (Bass,
   1986).
                                      F-198
-------
•  Cover system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the anticipated seepage pressures in the cover
   system drainage layers.
•  Cover system drainage layers should be designed to handle the total anticipated
   flow. Cover system drainage layers should be designed to handle the total
   anticipated flow. Hydraulic requirements for the drainage layer should be evaluated
   using water balance calculations or other appropriate analysis method (e.g., Giroud
   and Houlihan, 1995). Soong and Koerner (1997) recommend using a short-duration
   intensive storm in the water balance and do not recommend the EPA HELP
   computer model for this purpose. The drainage layer flow rates output from the
   HELP model are an average for a 24-hour period and may be much less than the
   peak flow rates calculated using other methods if the precipitation data used in the
   HELP model are not carefully selected. It is not clear if the drainage layer hydraulic
   requirements were evaluated for the cover system described above.

There is little available information for this case history;  additional lessons might have
been learned if the information had been complete.

F-A.12.16 C-21

Problem Classification: cover system stability/design
Region of U. S.:  southwest
Waste Type:  MSW
References: Anderson, R.L., "Earthquake Related Damage and Landfill Performance",
    Earthquake Design and Performance of Solid Waste Landfills, Yegian, M.K. and
    Liam Finn, W.D., eds., ASCE Geotechnical Special Publication No. 54, 1995, pp. 1-
    16.
    Augello, A.J., Matasovic,  N., Bray, J.D., Kavazanjian, E.,  Seed, R.B., "Evaluation of
    Solid Waste Landfill Performance During the Northridge Earthquake", Earthquake
    Design and Performance of Solid Waste Landfills, Yegian, M.K. and Liam Finn,
    W.D., eds., ASCE Geotechnical Special Publication No. 54, 1995, pp. 17-50.
    Chang, S., Bray, J.D., and Seed, R.B., "Engineering Implications of Ground Motions
    from Northridge Earthquake", Bulletin of the Seismological Society of America, Vol.
    86, No. 1, Part B Supplement, 1996, pp. S270-S288.
    Matasovic, N., Kavazanjian, E., Jr., Augello, A.J., Bray, J.D., and Seed, R.B., "Solid
    Waste Landfill Damage Caused by 17 January 1994 Northridge Earthquake", The
    Northridge, California, Earthquake of 17 January 1994, Woods, M.C. and Seiple,
    W.R., eds., California Department of Conservation,  Division of Mines and Geology
    Special Publication 116, 1995, pp.  221-229.
    Stewart, J.P., Bray,  J.D.,  Seed, R.B., and Sitar, N.,  "Preliminary Report on the
    Principal Geotechnical Aspects of the January 17, 1994 Northridge Earthquake",
    Report No. UCB/EERC-94/08, College of Engineering,  University of California at
    Berkeley, Berkeley,  California, 1994, 238 p.

Problem Summary: minor cracks in soil intermediate cover from Northridge earthquake


                                      F-199
-------
Problem Description: The Lopez Canyon Landfill is located in the foothill region of the
San Gabriel Mountains, approximately 50 km northwest of downtown Los Angeles.  The
site is underlain by siltstones, sandstones, and conglomerates. Terrace deposits are
present locally near the southeastern boundary of the property. Three known faults are
located to the northwest and through the southeast corner of the site. The landfill is
divided into four units: Areas A, B, AB+, and C.  Areas A,  B, and AB+ are not modern or
geosynthetically lined and are not discussed further.  Area C has a single composite
liner system on the base and  the 1.51-1:1 V to 11-1:1 V side slopes.  The side slope liner
system consists of the following components, from top to  bottom:

•  soil protection layer;
•  nonwoven GT filter;
•  GN LCRS drainage layer;  and
•  2-mm thick HOPE GM  liner (textured on one side, textured side down);
•  GCL.

The side slope liner system was constructed over a reinforced concrete veneer. The
predetermined liner system slip surface on the side slope is at the GN/GM interface.

The 17 January 1994 Northridge earthquake (moment magnitude Mw 6.7) occurred on a
blind thrust fault at a depth of approximately 15 km at the northern end of the San
Fernando Valley of the greater Los Angeles area. The Lopez Canyon Landfill is located
about 8 km from the zone  of energy  release (i.e., the fault plane). Strong motion
stations located on rock outcrops in the area recorded peak horizontal accelerations on
the order of 0.4g to 0.45g. The estimated rock peak horizontal acceleration at the
landfill resulting from the earthquake is 0.42g.  At the time of the Northridge earthquake,
Area C was active. Phase I of Area  C was fully lined and filled to a height of about 30
m. Phase II was fully lined across the base, but only partially lined on the side slope.

Post-earthquake damage inspection was carried out immediately after the earthquake.
There was no sign of permanent relative displacement between the waste and liner
system in Area C.  However,  minor cracking was observed in the soil intermediate
cover.  At one location on side slope of Phase II where the composite liner was not
completed (i.e., GN LCRS drainage  layer was just placed and anchored at the top of the
slope with sand bags), it appeared that the GN had slid up to +/- 20 mm over the
underlying GM liner during the earthquake. This sliding occurred at the predetermined
slip surface. The GN was not damaged.

Resolution: Portions of the soil intermediate cover with the widest cracks (i.e., 100 mm)
were regraded.

Lessons Learned for Future Projects: Based on the available information, the following
lessons can be learned from this case history:
                                      F-200
-------
•  Surficial cracking of soil cover layers, especially near locations with contrast in
   seismic response characteristics (e.g., top of waste by side slopes), should be
   anticipated and dealt with as an operation issue through post-earthquake inspection
   and repair.

F-A.12.17 C-22

Problem Classification: cover system stability/design
Region of U. S.:  southwest
Waste Type:  MSW
References: Anderson, R.L., "Earthquake Related Damage and Landfill Performance",
    Earthquake Design and Performance of Solid Waste Landfills, Yegian, M.K. and
    Liam Finn, W.D., eds., ASCE Geotechnical Special Publication No. 54, 1995, pp. 1-
    16.
    Augello, A.J., Matasovic,  N., Bray, J.D., Kavazanjian, E., Seed, R.B., "Evaluation of
    Solid Waste Landfill Performance During the Northridge Earthquake", Earthquake
    Design and Performance of Solid Waste Landfills, Yegian, M.K. and Liam Finn,
    W.D.,  eds., ASCE Geotechnical Special Publication No. 54, 1995, pp. 17-50.
    Matasovic, N., Kavazanjian, E., Jr., Augello, A.J., Bray, J.D., and Seed, R.B., "Solid
    Waste Landfill Damage Caused by 17 January 1994 Northridge Earthquake", The
    Northridge,  California, Earthquake of 17 January 1994, Woods, M.C. and Seiple,
    W.R.,  eds.,  California Department of Conservation, Division of Mines and Geology
    Special Publication 116, 1995, pp. 221-229.
    Stewart, J.P., Bray, J.D.,  Seed, R.B.,  and Sitar, N., "Preliminary Report on the
    Principal Geotechnical Aspects of the January 17, 1994 Northridge Earthquake",
    Report No. UCB/EERC-94/08, College of Engineering, University of California at
    Berkeley,  Berkeley, California, 1994, 238 p.

Problem Summary: 215-m long crack in soil intermediate cover from Northridge
earthquake

Problem Description:  The Calabasas Landfill is a canyon fill located in the Santa
Monica Valley in Agoura, California.  The  site is underlain by landslide deposits,
interbedded sandstone and shale, and interbedded sandstone and conglomerate.
Three inactive faults have been identified  on site. The landfill is divided into a number
of lined and unlined cells.  The lined cells  have  a CCL or a single-composite liner
system.

The 17 January 1994 Northridge earthquake (moment magnitude  Mw 6.7) occurred on a
blind thrust fault at a depth of approximately 15 km at the northern end of the San
Fernando  Valley of the greater Los Angeles area.  The Calabasas Landfill is located
about 23 km from the zone of energy release (i.e., the fault plane). The estimated rock
peak horizontal  acceleration at the landfill resulting from the earthquake is 0.20g. At the
time of the Northridge earthquake, the landfill had only one geosynthetically-lined cell
(Cell P). Cell  P  was partially constructed and receiving waste.
                                      F-201
-------
After the earthquake, a 215-m long crack in the soil intermediate cover was observed
near and parallel to the liner system anchor trench in Cell P. The crack was up to 150
mm wide and vertically offset up to 100 mm.  No waste was exposed.

Resolution: The cracked soil intermediate cover was regraded and revegetated.

Lessons Learned for Future Projects:  Based on the available information, the following
lesson can be learned from this case history:

•  Surficial cracking of soil cover layers, especially near locations with contrast in
   seismic response characteristics (e.g., top of waste by side slopes), should be
   anticipated and dealt with as an operation issue through post-earthquake inspection
   and repair.

F-A.12.18 C-23

Problem Classification: cover system stability/design
Region of U. S.:  southwest
Waste Type:  MSW
References: Anderson, R.L., "Earthquake Related Damage and Landfill Performance",
    Earthquake Design and Performance of Solid Waste Landfills, Yegian, M.K. and
    Liam Finn, W.D., eds., ASCE Geotechnical Special Publication No. 54, 1995, pp. 1-
    16.
    Augello, A.J., Matasovic, N., Bray, J.D., Kavazanjian, E., Seed, R.B., "Evaluation of
    Solid Waste Landfill Performance During the  Northridge Earthquake", Earthquake
    Design and Performance of Solid Waste Landfills, Yegian, M.K. and Liam Finn,
    W.D.,  eds., ASCE Geotechnical Special Publication No. 54, 1995, pp. 17-50.
    Matasovic, N., Kavazanjian, E., Jr., Augello, A.J., Bray, J.D., and Seed, R.B., "Solid
    Waste Landfill Damage Caused by 17 January 1994 Northridge Earthquake", The
    Northridge,  California, Earthquake of 17 January 1994, Woods, M.C. and Seiple,
    W.R.,  eds.,  California Department of Conservation, Division of Mines and Geology
    Special Publication 116, 1995, pp. 221-229.
    Stewart, J.P., Bray, J.D., Seed, R.B., and Sitar, N., "Preliminary Report on the
    Principal Geotechnical Aspects of the January 17, 1994 Northridge Earthquake",
    Report No. UCB/EERC-94/08, College of Engineering, University of California at
    Berkeley,  Berkeley, California, 1994, 238 p.

Problem Summary: minor cracks in soil intermediate cover from Northridge earthquake

Problem Description: The Simi Valley Landfill is a canyon fill located in Ventura County,
California. The  site is underlain by alluvium, which overlies crystalline and metamorphic
basement complex rocks. The landfill has a single-composite liner system.
                                      F-202
-------
The 17 January 1994 Northridge earthquake (moment magnitude Mw 6.7) occurred on a
blind thrust fault at a depth of approximately 15 km at the northern end of the San
Fernando Valley of the greater Los Angeles area. The Simi Valley Landfill is located
about 22 km from the zone of energy release (i.e., the fault plane). The estimated rock
peak horizontal acceleration at the landfill resulting from the earthquake is 0.21g. At the
time of the Northridge earthquake, the landfill had two lined cells,  one active and one
inactive.

After the earthquake, minor cracking of the soil intermediate cover was reported.

Resolution: The cracked soil intermediate cover was regraded and revegetated.

Lessons Learned for Future Projects'.  Based on the available information, the following
lesson can be learned from this case history:

•  Surficial cracking of soil cover layers, especially near locations with contrast in
   seismic response characteristics (e.g., top of waste by side slopes), should be
   anticipated  and dealt with as an operation issue through  post-earthquake inspection
   and repair.


F-A.13    Cover System Displacement

F-A.13.1  C-12

Problem Classification: cover system displacement/design
Region of U. S.: northeast
Waste Type: MSW
Reference: unpublished

Problem Summary:  cover system settlement caused tearing of GM boots around gas
well penetrations of GM barrier

Problem Description: A landfill cover system with a 1-mm thick HOPE GM barrier was
constructed in 1991 and 1992. By late 1992, a gas collection system, including vertical
HOPE gas collection wells that penetrated the GM barrier, had been installed in the
landfill.  At each penetration, an HOPE GM boot was clamped to the well and extrusion
seamed to the  GM barrier to seal the barrier around the well.

When several of the GM boots around the wells were inspected in 1995, the boots were
observed to be torn from the GM barrier.  The boots were not designed to
accommodate settlement of the waste, which would cause downward displacement of
the GM barrier relative to the wells. Since the cover system had been constructed, the
landfill top had settled from 0.3 to 0.9 m and the side slopes had settled less than 0.3 m.
                                      F-203
-------
Resolution: The gas extraction well boots were replaced with new expandable boots
that can elongate up to 0.3 m.  These boots can also be periodically moved down the
well to accommodate landfill settlement.

Lessons Learned for Future Projects'. Based on the available information, the following
lesson can be learned from this case history:

•  Gas  extraction well boots should be designed to accommodate the anticipated
   landfill settlements.

F-A.13.2  C-15

Problem Classification:  cover system displacement/construction
Region of U. S.:  northeast
Waste Type: paper mill sludge
Reference: Badu-Tweneboah, K., Williams, N.D., and Haubeil, D.W., "Assessment of a
    PVC Geomembrane Used in a Landfill Cover System", Proceeding of Fifth
    International Conference on Geotextiles, Geomembranes and Related Products,
    Singapore, 1994, pp. 1029-1032.

Problem Summary: localized cover system settlement during construction stretched,
but did not damage, PVC GM barrier and opened GCL joints

Problem Description:  A cover system was constructed over saturated,  highly
compressible paper mill sludge. The cover system consisted of the following
components, from top to bottom:

•  0.75-m thick soil surface and protection layer;
•  GC drainage layer;
•  0.5-mm thick PVC GM barrier;
•  GCL; and
•  GC gas collection layer.

To facilitate construction, a 3 to 6-m thick stabilized sludge working surface with a
minimum undrained shear strength of 24 kPa was spread over the in-place  sludge.
After the geosynthetics were installed, they were covered with a soil layer. The soil was
hauled from the perimeter of the landfill and spread over the geosynthetics by low-
ground pressure bulldozers. The specifications required that the ground pressure of
this equipment be less than 34 kPa.

The repeated trafficking of bulldozers over portions of the cover system resulted in
pumping of the underlying sludge into the stabilized sludge. This  pumping progressively
reduced the shear strength of the stabilized sludge layer,  resulting in localized bulges
and, at times, placement of excessive thickness of soil.  When subjected to the stresses
of this excess soil, the weakened stabilized sludge layer underwent a localized bearing


                                      F-204
-------
capacity failure in a 60-m long by 18-m wide area. Tests pits dug to the top of the cover
system geosynthetics showed that they had been subjected to settlements of 0.1 to 2.4
m. Based on calculations performed by the authors of this appendix, if the
geosynthetics within this trough have a circular curved shape across the width of the
trough, the average strain in the geosynthetics in the case where the settlement is 2.4 m
is about 4.7%.
Figure F-A.13.1. Localized bearing capacity failure of stabilized sludge caused a
depression in the liner system.

The soil was removed from over the geosynthetics in the affected area.  None of the
geosynthetics appeared to have been damaged by the straining, though the GCL
seams had separated at two locations along the length of the panels. Adjacent GCL
panels had been overlapped 0.15 m along the roll length; however, based on
calculations by the authors of this appendix, the seam would open if the average strain
exceeded 3.2%. The GM showed some lateral wrinkling, indicating it was in tension in
the other direction.  However, it had no tears, scratches, or separated seams, except for
the few locations damaged by the backhoe excavating the soil off of the geosynthetics.
Samples of the GM panels and seams taken from the affected area and tested in the
laboratory indicated that the GM still met the project specifications and was not
                                     F-205
-------
adversely affected.  This is not surprising given that the PVC GM underwent a relatively
low level of strain as calculated above. In comparison, the strain at break of a PVC GM
is on the order of 300% at normal temperature.

Resolution: The affected area was repaired by removing the cover system materials in
this area, restabilizing and regrading the sludge, and reinstalling the cover system with
new geosynthetic materials.  The bulldozers used to spread the soil surface and
protection layer had ground pressures less than that used previously.  Additional grade
control measures were implemented to ensure that no more than 0.75 m of soil was
placed over the geosynthetics.

Lessons Learned for Future Projects: Based on the available information, the following
lessons can be learned from this case history:

•  PVC GMs are acceptable for construction over very compressible, low shear
   strength waste materials, since they can stretch (if the temperature is above 0° C).
•  Cover systems with soil layers placed over compressible, low shear strength waste
   should use lightweight construction equipment and have good control of the
   thickness of soil placed over the waste
•  When GCL is used over soft subgrade, seam overlaps should be wider than normal.

F-A.14    Impoundment Liner Construction

F-A.14.1  S-3

Problem Classification: impoundment liner construction/construction
Region of U. S.:  southeast
Waste Type:  MSW leachate
Reference:  unpublished

Problem Summary: large wrinkles in HOPE GM primary liner at two leachate ponds

Problem Description:  Two double-lined leachate ponds were constructed during the
winter. The liner system for the ponds consists of the following components, from top to
bottom:

•  1.5-mm thick HOPE GM primary liner;
•  GN drainage layer;
•  1.5-mm thick GM secondary liner; and
•  0.15-m thick CCL

The ponds have 3H:1V side slopes and are about 3 m deep. At the end of construction,
the GM primary liner in the ponds was noticeably wrinkled.  However, the ponds
seemed acceptable to the CQA consultant.  As temperatures increased  before the
ponds were put into service,  the GM became more wrinkled. On the side slope,

                                     F-206
-------
wrinkles were oriented parallel to the slope crest. Wrinkles became more numerous
and larger near the slope toe as they propagated down the slope during several months
of temperature cycling.  By the following summer, wrinkles were, on average, about 100
mm high.  In both cells there were several large wrinkles, located near the slope toe,
that folded over.

Resolution: The wrinkles in the GM will be cut out, and the GM will be seamed.

Lessons Learned for Future Projects'.  Based on the available information, the following
lesson can be learned from this case history:

•  During construction, excessive GM wrinkles and wrinkles that may fold over should
   be removed by waiting to backfill until the GM cools and contracts during the cooler
   nighttime and early morning hours, pulling the wrinkles out, or cutting the wrinkles
   out. The latter method is less desirable than the former methods because it requires
   intact GM to be cut, and it results in more GM seaming and subsequent testing.
•  Enough slack should be left in GMs so they are essentially stress-free at their lowest
   expected temperatures.
•  GM liners should be covered with a soil layer or insulated by other means as soon
   as practicable after installation, but not during the hottest part of the day if the GM is
   significantly wrinkled, to reduce GM wrinkling.

F-A.14.2    4
Problem Classification:  impoundment liner construction/construction
Region of U. S. :  southwest
Waste Type:  HW
Reference:  Bonaparte, R. and Gross, B.A., "LDCRS Flow Rates from Double-Lined
    Landfills and Surface Impoundments", EPA Risk Reduction Research Laboratory,
    Cincinnati, OH, EPA/600/SR-93/070, 1993, 65 p. (Impoundment C-2)

Problem Summary: leakage through holes in HOPE GM component of composite
primary liner

Problem Description:  A double-composite lined pond was placed into service in
September 1986 and slowly filled with waste. The primary liner for the pond consists of
a 2.5-mm thick HOPE GM over a 0.45-m thick CCL (specified maximum hydraulic
conductivity of 1 x 10~9 m/s). The LDS drainage layer is a  GN on the side slope and a
gravel layer over a GN on  the base.  The GM component of the primary liner is
protected on the side slope by a GM that is tack seamed to it.  The primary liner is not
protected by a soil protection layer.

After seventeen months of operation, when the liquid  level in the  pond had reached its
highest level of about 3 m, monthly LDS flow rates increased from 0 to 250 Iphd to 520
to 1 ,380 Iphd. The pond liquid level was allowed to drop by evaporation, and the GM
                                     F-207
-------
protection layer was inspected for holes along the pond perimeter at an elevation
corresponding to the maximum liquid level. One small hole was found in the GM
protection layer and primary liner.  The holes were repaired and the pond was slowly
filled with liquid again.  At thirty months after operation, again the liquid level in the pond
again reached a level of about 3 m and monthly LDS flow rates increased from 0 to 50
Iphdto800to1,020lphd.

In both cases where LDS flow rates increased when the pond liquid level was raised to
3 m, the increase occurred over a relatively short time period (less than one month).
Presumably at least part of this flow was due to primary liner leakage.  It  is not clear
how leakage entered the LDS in such a short time period given that the primary liner is
a composite.  It may be that the CCL underlying the GM on the side slope has become
desiccated due to thermal effects. This desiccation would be worst in the upper portion
of the side slope, where the primary liner is not insulated  from ambient temperature
cycling by the pond liquid.  The hydraulic conductivity of a desiccated CCL may be
several orders of magnitude greater than that of a protected CCL.

Resolution: When the protection layer was removed and the GM primary liner was
inspected at an elevation corresponding to the  maximum  liquid level, a GM  patch
previously seamed  over an old tear was found to have failed at the seam. A new patch
was seamed over the tear, and the pond was slowly filled to a liquid level of about 3 m.
Monthly LDS flow rates remained relatively low, in the range of 0 to 130 Iphd.

Lessons Learned for Future Projects'. Based on the available information, the following
lessons can be learned from this case history:

•  Holes in GM liners installed with CQA should be anticipated.  If there is a head of
   leachate over a  hole, primary liner leakage can occur.
•  Primary liner leakage can occur in impoundments  with GM/CCL primary liners that
   are not insulated from the environment with a soil  protection layer or other material.
•  It can be difficult to locate primary liner holes in operating impoundments without
   taking the  pond  out of service and inspecting the primary liner components.

F-A.14.3    5
Problem Classification:  impoundment liner construction/construction
Region of U. S. :  northeast
Waste Type:  HW
Reference:  Bonaparte, R. and Gross, B.A., "LDCRS Flow Rates from Double-Lined
    Landfills and Surface Impoundments", EPA Risk Reduction Research Laboratory,
    Cincinnati, OH, EPA/600/SR-93/070, 1993, 65 p. (Impoundments H-2 and H-3)

Problem Summary: leakage through holes in HOPE GM primary liners at two ponds
                                     F-208
-------
Problem Description: Two double-lined shallow ponds (H-2 and H-3) became operable
in November 1988. The ponds have a 2-mm thick HOPE GM primary liner.  The LDS
drainage layer is a GN on the side slopes and sand on the  base. The ponds are
designed to have a maximum liquid level of 1.4 m. Prior to operation, leak location
surveys were performed in both ponds, and identified primary liner holes were repaired.

Chemical analyses of LDS flows from ponds H-2 and H-3 indicated that primary liner
leakage was occurring by seven and one months, respectively, after start of operation.
LDS flow rates from the ponds also increased with increasing pond liquid level.  The
average LDS flow rates from the ponds H-2 and H-3 from seven to 25 months of
operation were 310 and 5,150 Iphd, respectively.  It is unclear why the LDS flow rates
from pond H-3 are so high, given that a leak location survey was performed  prior to
pond operation.  Since primary liner leakage was evident shortly after filling started, a
primary liner hole was probably  located on the base or near the slope toe of this pond.

Resolution: After 25 months of operation, GM primary liner holes were located in the
ponds and repaired. No information is available on the size or location of the holes.
The average LDS flow rates from the ponds  H-2 and H-3 from 26 to 31 months of
operation were 400 and 440 Iphd, respectively.  For pond H-2, the LDS flow rates
increased after the  repairs when the liquid level in the pond was increased.

Lessons Learned for Future Projects'. Based on the available information, the following
lessons can be learned from this case history:

•  Holes in GM liners should  be anticipated, even in liners  installed with CQA.  If there
   is a head of leachate over a  liner hole, leakage occurs.
•  It can be difficult to locate  primary liner holes in operating impoundments without
   taking the  pond  out of service and inspecting the primary liner.

F-A.15    Impoundment Liner Degradation

F-A.15.1   S-1

Problem Classification: impoundment liner degradation/construction
Region of U. S.:  northcentral
Waste Type:  process water, ash, and other
Reference: Peggs,  I.D., Winfree, J.P., Giroud, J.P., "A Shattered Geomembrane Liner
    Case History: Investigation and Remediation", Proceedings of Geosynthetics '91,
    Atlanta, 1991, Vol. 2, pp. 495-505.

Problem Summary: slow crack growth stress cracks and shattering cracks in exposed
HOPE GM liner at five ponds

Problem Description: Peggs et al. (1991) described the shattering of HOPE GM liners
in five of thirteen ponds at a coal-fired power plant facility.  These liners were exposed
                                     F-209
-------
and, therefore, subject to significant thermally induced tensile stresses under the wide
range of ambient temperatures at the site (i.e., -30 to 40°C). At temperatures near
freezing, the GM on the side slope was taut. By four years after installation, the five
pond liners exhibited relatively short (i.e., 20 mm) slow crack growth (SCG) stress
cracks in and adjacent to some seams. There were no cracks below water level.  All of
the seams had been constructed by lapping the panels and applying a bead of HOPE
extrudate (extrusion flat seam).  The stress cracks generally occurred in the lower GM.

During the winter, record low temperatures appeared to precipitate the cracking of some
seams from the side slope crest to toe, a length of about 22 m.  The maximum crack
opening occurred at mid-slope and was about 0.2 m. These long cracks were
surrounded by branching rapid crack propagation (RCP) shattering cracks. The
shattering cracks propagated into the GM panels at an angle to the long crack (upward
in the top half of the slope and downward in the bottom half of the slope).  Shattering
cracks also extended from some of the short stress cracks into the GM  panels,
fracturing the GM. The mechanics of shattering crack development have been
described by Giroud (1994a). The liner in the pond containing the fewest visible stress
cracks had been installed with compensation panels to allow for liner contraction at low
temperatures; designed compensation panels had not been installed  in the other pond
liners.  There was also some indication that the most seriously damaged liners had
been installed at high ambient temperatures and would, therefore, require the largest
amount of compensation in order to be stress-free at the lowest operating temperature.

Subsequent laboratory studies conducted by Peggs et al. (1991), which included
examination of microtome sections and measurement of mechanical and physical
properties, found that many stress cracks were initiated where the original seam had
been repaired with an extruded fillet bead.  Other stress cracks were  initiated where
there was evidence of overheating during seaming.  A few stress cracks initiated where
there was no visible cause for initiation. Peggs et al. (1991) also showed that the
shattering cracks were all initiated at stress cracks located along seams. They
concluded that if stress cracks are not present in HOPE GM, shattering cracks will not
occur.  Peggs et al. (1991) hypothesized that the observed stress cracks were caused
primarily by heating of the HOPE during seaming. However, Giroud (1994b)  later
concluded that normal  heating during seaming does not seem to make HOPE GMs
more susceptible to stress cracking. Rather, HOPE  GM stress cracks are primarily
associated with stress concentrations caused by seams. The shattering cracks that
occurred in the pond liners can be explained by the conjunction of the following:

•  the HOPE resin  used in these GMs did not have a low stress-cracking susceptibility;
•  tensile stresses  caused by thermal contraction;
•  stress concentrations caused by the seams;
•  decreased allowable yield strain at the low temperatures at which GM shattering
   occurred; and
•  increased crystallinity of the HOPE next to the seam.
                                      F-210
-------
Resolution: Seams and panels were visually inspected for cracks. Panels and seams
with shattering cracks were replaced with new GM containing an adequate amount of
slackness calculated using the temperature of the GM at the time of the repair. The
new panels were installed during the coolest time of the day so the minimum amount of
slackness would be required.  In addition, every third seam was cut from the anchor
trench to the toe of slope and the GM was allowed to relax.  Compensation panels,
typically 1 m wide, were installed at each cut seam. Stress  cracks in seams were
repaired using a wide (i.e., 100 mm) bead extrusion technique developed specifically for
this purpose.

Lessons Learned for Future Projects:  Based on the available information, the following
lessons can be learned from this case history:

•  Resin used to manufacture HOPE GM should be resistant to stress cracking. This is
   currently evaluated using the notched constant tensile load test (ASTM D 5397).
   This test was not performed for the GM in the case history described above.
•  HOPE GMs should be installed so that they are essentially stress-free at their lowest
   expected temperatures.  This is consistent with EPA guidance (Daniel and Koerner,
   1993).
•  For HOPE GMs, fusion seams are preferred over extrusion seams because fusion
   seams have higher seam integrity and lower stress concentrations at seams
   (Giroud, 1994b).  Extrusion fillet seams are preferred over extrusion flat seams
   because fillet seams have  lower stress concentrations at seams.  In the case history
   described above, stress cracking may have been less severe if fusion seams, rather
   than extrusion flat seams, had been used.
•  In general, holes in HOPE  GM seams should not be repaired by seaming over the
   hole. This reheating of seams can embrittle the HOPE at the repair and make it
   more susceptible to stress cracking.
•  In general, GMs should be covered with a thermal insulation layers at very low
   temperatures (e.g., -20°C for HOPE GMs) since GM strain at break decreases with
   decreasing temperature.

F-A.16    Impoundment Liner System Stability

F-A.16.1  S-2

Problem Classification:  impoundment liner system stability/design
Region of U. S.: unknown
Waste Type:  HW
Reference:  Paulson, J.N., "Veneer Stability Case Histories:  Design Interactions
   Between Manufacturer/Consultant/Owner", Proceedings of the 7th GRI Seminar
   Geosynthetics Liner Systems: Innovations, Concerns, and Designs, Geosynthetic
   Research Institute, Philadelphia, PA, 1993, pp. 235-241.
                                     F-211
-------
Problem Summary: sliding along polypropylene needlepunched nonwoven GT/HDPE
GM interface during waste placement

Problem Description:  Paulson (1993) described the slope stability failure of components
of a sludge impoundment liner system that occurred during sludge placement. The
double-liner system consists of the following components, from top to bottom:

•  530 g/m2 polypropylene needlepunched nonwoven GT cushion layer;
•  HOPE GM primary liner;
•  GN LDS drainage  layer;
•  GM secondary liner; and
•  CCL.

The 2H:1V side slopes are about 10 m high. Sludge was placed in the impoundment by
dumping it on the GT  cushion at the slope crest and allowing it to flow down the GT to
the slope toe.  On several occasions when the sludge adhered to the GT,  a low-ground
pressure bulldozer was used to push the sludge downslope. This method of sludge
placement caused tension to develop in the GT.  Eventually, the GT progressively tore
at the slope crest and slid over the GM to the slope toe.

Paulson (1993) does not indicate if slope stability analyses or interface direct shear
testing were performed as part of the lining system design. An infinite slope analysis
was conducted by the authors of this appendix. In this analysis,  the secant friction
angle for the polypropylene GT/HDPE GM interface was assumed to be 10°, which is
within the range of friction angles reported for this interface in the technical literature.
The slope stability factor of safety  thus calculated is 0.35 with no seepage pressures.
Based on this factor of safety, the  GT would be in tension if sludge were placed on it.
Paulson (1993) does not indicate whether the GT was designed to be in tension.

Resolution: The damaged GT was replaced. Additionally, a thin GM slipsheet was
placed over the GT in the sludge dumping area to facilitate the sliding of sludge down
the slope.

Lessons Learned for Future Projects: Based on the available information,  the following
lessons can be learned from this case history:

•  Liner system slopes should always be evaluated using rigorous slope stability
   analysis methods that consider the actual shear strengths of the liner system
   materials and the method of waste placement. Both of these considerations
   probably contributed to the liner system failure described above.
•  Waste should generally be placed over geosynthetics from the toe of slope upward
   to avoid tensioning the geosynthetics.  Methods of waste placement that are not toe
   to top must be pre-approved by the engineer who analyzed the stability.
                                     F-212
-------
F-A.17    References

Bass, J.M., "Avoiding Failure ofLeachate Collection and Cap Drainage Layers",
    EPA/600/2-86/058, EPA Hazardous Waste Engineering Research Laboratory,
    Cincinnati, Ohio, 1986, 142 p.
Bonaparte, R., "Long-Term Performance of Landfills'" Proceedings of the ASCE
    Specialty Conference Geoenvironment 2000, ASCE Geotechnical Special
    Publication No. 46, 1995, Vol. 1, pp. 415-553.
Bonaparte, R. and Gross, B.A., "LDCRS Flow Rate from Double-Lined Landfills and
    Surface Impoundments",  EPA/600/SR-93/070, EPA Risk Reduction Research
    Laboratory, Cincinnati, OH, 1993, 65 p.
Daniel and Koerner "Technical Guidance Document: Quality Assurance and Quality
    Control for Waste Containment Facilities", EPA/600/R-93/182, EPA Risk Reduction
    Research Laboratory, Cincinnati, OH, 1993, 305 p.
Environmental Protection Agency, "Seminar Publication: Design, Operation, and Closure
    of Municipal Solid Waste Landfills",  EPA/625/R-94/008, EPA Center for
    Environmental Research Information, Cincinnati, Ohio, 1994, 86 p.
Giroud, J.P., "Impermeability: The Myth and a Rational Approach", Proceedings of
    International Conference on Geomembranes, Denver, Colorado, Vol. 1, 1984, pp.
    157-162.
Giroud, J.P., "Lessons Learned From Studying the Performance of Geosynthetics",
    Proceeding of Geotextiles-Geomembranes Rencontres 93, Vol. 1, Joue-les-Tours,
    France, Sep 1993, pp. 15-31.
Giroud, J.P., "The Mechanics of GM Stress Cracking", Computer Methods and
    Advances in Geomechanics, Siriwardane, H.J. and Zaman, M.M. (editors),
    Balkema, Proceedings of the Eighth International Conference on Computer
    Methods and Advances in Geomechanics, Morgantown, West Virginia, USA,  1994a,
    Vol. 1, pp. 177-188.
Giroud, J.P., "Quantification of Geosynthetics Behavior", Proceeding of the Fifth
    International Conference on Geotextiles, Geomembranes, and Related Products,
    Singapore, 1994b, Vol. 4, pp. 1249-1273.
Giroud, J.P., "Granular Filters and Geotextile Filters", Proceedings ofGeofilters '96,
    Lafleur, J., and Rollin, A.L., Editors, Montreal, 1996, pp. 565-690.
Giroud, J.P. and Houlihan, M.F., "Design of Leachate Collection Layers",  Proceedings
    of Fifth International Landfill Symposium, Sardinia, Italy, Vol. 2,  1995, pp. 613-640.
Giroud, J.P., Bachus, R.C., and  Bonaparte, R., "Influence of Water Flow on the Stability
    of Geosynthetic-Soil Layered Systems on Slopes", Geosynthetics International,
    IFAI, Vol. 2,  No. 6, 1995a, pp. 1149-1180.
Giroud, J.P., Tisseau, B., Soderman, K.L.,  and Beech, J.F., "Analysis of Strain
    Concentrations Next to Geomembrane Seams", Geosynthetics International,  IFAI,
    Vol. 2, No. 6, 1995b, pp. 1049-1097.
Soong, T.Y. and Koerner, R.M.,  "The Design of Drainage Systems Over
    Geosynthetically Lined Slopes", GRI Report #19, Geosynthetic Research Institute,
    Philadelphia, Pennsylvania,  1997, 88 p.
                                     F-213
-------
                       Appendix G
       Long-Term  Landfill  Management
Rudolph Bonaparte, Ph.D., P.E.
GeoSyntec Consultants
Atlanta, GA 30342
      by

David E. Daniel, Ph.D., P.E.
University of Illinois
Urbana, IL 61801
Robert M. Koerner, Ph.D., P.E.
Drexel University
Philadelphia, PA 19104
                           performed under

                   EPA Cooperative Agreement Number
                           CR-821448-01-0
                            Project Officer

                         Mr. David A. Carson
               United States Environmental Protection Agency
                   Office of Research and Development
               National Risk Management Research Laboratory
                         Cincinnati, OH 45268
-------
                              DISCLAIMER

The ideas presented in this appendix are the authors' and do not represent EPA
policy.  EPA considers some of these approaches developmental.  If the Agency
initiates any changes with respect to post-closure, it will be noticed in the Federal
Register and only after review and discussions with the stakeholders involved.
-------
                                 Appendix G
                      Long-Term Landfill Management

G-1   Introduction
The performance data for operating landfills presented in this research report demonstrate
that landfills can be designed, constructed, and operated/maintained to achieve very high
levels of leachate and landfill gas containment and collection.  The report has also
demonstrated that design,  construction, and operation/maintenance issues and problems
persist at many landfills. In Chapter 6, the authors attempted to provide guidance to design
engineers on how to avoid the most significant issues and problems that may typically
arise. Information on the anticipated service lives of the various engineered components of
a landfill waste containment system was also given.

The ultimate degradation of any individual waste containment system component of a
landfill after the completion of that component's useful service life may or may not lead
to a release of leachate or gas and contamination of groundwater.  Furthermore,  a
release may, or may not, result in a significant environmental impact.  In evaluating the
consequences of ultimate degradation, the design engineer must consider a wide range
of factors including: the climatological and hydrogeologic setting; the composition, age,
and level of degradation of the waste; the potential for leachate and gas generation after
the component has completed its service life; the potential to maintain, rehabilitate,  or
install other systems to achieve leachate and gas containment;  and collection, cost, and
social and institutional factors. These various factors should be considered within an
overall decision-making framework, herein referred to as a long-term landfill
management strategy.

G-2   Strategies for Long-Term Landfill Management
Seven strategies originally developed by Bonaparte (1995) for long-term landfill
management are summarized in Table G-1 and presented below. In this report, the
discussions of the strategies are focused towards municipal solid waste (MSW) landfills.
Each strategy has implications for long-term landfill maintenance and monitoring  (with
advantages and disadvantages).  It is noted that variations on the strategies described
in this report are possible  and alternative strategies could also be developed.

Strategy 1 - Standard Landfill with Perpetual Post-Closure Period'. This approach
embodies the original United States Environmental Protection Agency (EPA) "liquids
management strategy".  With this approach, waste is disposed of at its' "as received"
moisture content, and efforts are made to minimize the ingress of moisture into the landfill.
This type of landfill does not receive supplemental moisture through, for example, leachate
recirculation or water addition. Liquid/gas containment and collection during the active life
of this type of landfill are achieved through operation of the liner system and liquid and gas
removal systems.  Ideally, the leachate collection and removal system (LCRS) should be
                                       G-1
-------
  Table G-1. Summary of Strategies for Long-Term Landfill Management (from Bonaparte, 1995).
Strategy
1
2
3
4
5
Landfill
Classification
Standard Landfill
- Perpetual
Post-Closure
Period
Standard Landfill
- Limited
Post-Closure
Period
Standard Landfill
- Clean Closure
Recirculation
Landfill -
Perpetual
Post-Closure
Period
Recirculation
Landfill -
Limited
Post-Closure
Period
Moisture
Conditions
No
supplemental
moisture
addition
No
supplemental
moisture
addition
No
supplemental
moisture
addition
Supplemental
moisture is
added to the
landfill
Supplemental
moisture is
added to the
landfill
Requirements for Long-Term
Protection of Groundwater and Air
Containment and collection systems must
be maintained for as long as leachate and
gas are generated and through the post-
closure period. Final cover system must
be maintained in perpetuity.
Containment and collection systems must
be maintained for as long as leachate and
gas are generated and through the post-
closure period. Impacts of any leachate or
gas generated after the post-closure
period must be within acceptable limits.
Containment and collection systems must
be maintained until clean closure.
Containment and collection systems must
be maintained for as long as leachate and
gas are generated and through the post-
closure period. Final cover must be
maintained in perpetuity. An enhanced
liner system may be required to gain
regulatory approval. Recirculation
hardware is needed.
Containment and collection systems must
be maintained for as long as leachate and
gas are generated, and through the post-
closure period. An enhanced liner system
may be required to gain regulatory
approval. Recirculation hardware is
needed. Impacts of any leachate or gas
generated after the post-closure period
must be within acceptable limits.
Advantages
• This strategy can prevent significant gas and
leachate migration in perpetuity, assuming
adequate maintenance
• This strategy complies with regulations and is
consistent with EPA's liquids management
strategy
• Capital costs are lower than for other
strategies
• This strategy may be applied to virtually all
sites, including sensitive sites
• This strategy has the same advantages as
Strategy 1 , except that, due to limited post-
closure period, this strategy may not be
acceptable for sensitive sites
• Clean closure eliminates contaminant source
• Concerns about long-term performance of
landfill waste containment system are
eliminated
• Waste is used as a resource
• This strategy has the same advantages as
Strategy 1 , except that the capital costs are
higher
• Recirculation improves gas generation
• Biostabilized waste has a lower latent
potential to pollute groundwater and air than
Strategies 1 and 2
• Post-closure maintenance costs are lower
than dry landfill strategies
• This strategy combines the advantages of
Strategies 2 and 4
• This strategy may be suitable for the great
majority of sites, including sensitive sites
Disadvantages
• Final cover system must be maintained, and
monitoring must be performed in perpetuity
• This strategy results in perpetual
maintenance and monitoring costs
• Waste retains a latent potential to generate
leachate and gas and release pollutants
• Limited gas or leachate migration could
occur in the long term, after the post-closure
period
• The design must consider the potential for
long-term migration, and it must be
established that potential migration rates
are within acceptable limits
• Clean closure is not presently cost effective
for most projects
• Environmental impacts of clean closure
operations must be addressed
• Waste residuals must be managed
• This strategy has the same disadvantages
as Strategy 1 , except that the waste retains
a lower latent potential to pollute
groundwater and air
• Recirculation hardware and operations
result in additional capital and operations
costs; if required, an enhanced liner system
will results in additional capital costs
• Design and operational experience with
recirculation is currently limited
• This strategy has the same disadvantages
as Strategy 4, except that perpetual care of
the final cover system is not required
• The design must consider the potential for
long-term migration, and it must be
established that potential migration rates
are within acceptable limits
o
ro
-------
   Table G-1.  Summary of Strategies for Long-Term Landfill Management (from Bonaparte, 1995) (cont).
    Strategy
Landfill
Classification
Moisture
Conditions
Requirements for Long-Term
Protection of Groundwater and Air
Advantages
Disadvantages
               Recirculation
               Landfill -
               Clean Closure
                  Supplemental
                  moisture is
                  added to the
                  landfill
               Containment and collection systems must
               be maintained until clean closure. An
               enhanced liner system may be required to
               meet regulations.  Recirculation hardware
               is needed.
                                         This strategy combines the advantages of
                                         Strategies 3 and 4, except that capital costs
                                         are higher
                                         May be able to reuse liner system after clean
                                         close a cell
                                             This strategy has the same disadvantages
                                             of Strategies 3 and 4, except that long-term
                                             maintenance and monitoring is not required
               Inward-Gradient
               Landfill
                  This strategy
                  may be
                  developed
                  with or without
                  supplemental
                  moisture
                  addition
               Inward gradient design must not allow
               leachate diffusion through liner system.
               Depending upon the design strategy,
               inward gradient must be maintained in
               perpetuity, for a limited post-closure
               period, or until clean closure.
                                         An inward gradient provides active (not
                                         passive control of contaminant advection and
                                         diffusion
                                         An inward gradient approach can be
                                         incorporated into the framework of Strategies
                                         1 to 6
                                             This strategy is not compatible with current
                                             U.S. regulations
                                             Inward gradient will not exist in unsaturated
                                             zone above water table, unless an
                                             engineered hydraulic control system (e.g., a
                                             double-liner system) is constructed
                                             Large liquid volumes must likely be
                                             collected from  LCRS and treated
9
CO
-------
designed to allow rapid removal of leachate generated by the landfill.  In this way,
the potential for significant advective or diffusive transport through the liner system
can essentially be eliminated.  Both  transport mechanisms  require time periods on
the order of years for leachate breakthrough of modern geomembrane/compacted
clay liner (GM/CCL) composite liners.  Rapid leachate removal can be achieved by
having a highly permeable, adequately sloped LCRS that precludes both a leachate
head buildup on the liner and a sustained leachate head on the liner.

After closure  of such a landfill, the final cover system (which will typically contain a
GM to satisfy present-day federal regulations) will essentially eliminate infiltration
into the landfilled waste. Consequently, over time, leachate and gas generation will
progressively decrease and eventually cease. A nominal stabilization of waste (i.e.,
partial conversion of the decomposable organic constituents and leaching or fixation
of hazardous waste constituents) will probably occur by the end of the post-closure
period, although this process will be incomplete due to a deficit of moisture in the
landfill  (Tchobanoglous et al., 1993). The waste will retain  a latent capacity to
generate leachate and gas and release pollutants should moisture be reintroduced
into the landfill in the future. The goal with this strategy is to prevent a reoccurrence
of leachate or gas generation in the  future.  This goal  can be achieved through
perpetual maintenance of the final cover system. In a modern landfill with no
supplemental moisture addition,  leachate and gas generation will eventually cease
and, thus, the need for ongoing maintenance of systems other than the final cover
system will, in the long term, not  be  necessary.

Strategy 2 - Standard Landfill with Limited Post-Closure Period: This strategy is the
same as Strategy  1, except  it is assumed that the post-closure maintenance period will
be of limited duration. After the post-closure period, requirements for ongoing
maintenance and monitoring are suspended. This approach  is consistent with current
regulations that prescribe a minimum post-closure period of 30 years. For this strategy
to be effective in the long term, events that occur after the end of the post-closure
period must not result in unacceptable groundwater or air pollution.  These events may
include degradation of the final cover system, renewed  leachate and gas generation,
accumulation of leachate and gas in the unit, and, eventually, migration of these waste
by-products from the unit. An assessment of the eventual  impacts requires evaluation
of:  (i) the potential for long-term degradation of the final cover system; (ii) water
infiltration through  the final cover system in its long-term condition;  (iii) leachate and gas
generation resulting from the water infiltration; (iv) the potential for long-term
degradation of the liner system; (v) gas migration through the final cover system in its'
long-term condition and leachate and gas migration through the liner system in its' long-
term condition; and (vi) impacts to groundwater resulting from the leachate and gas
migration and impacts to air resulting  from gas migration. Evaluations of the type just
                                     G-4
-------
described have been performed by Rowe and Fraser (1993a,b) considering the
potential for groundwater contamination for several hypothetical landfill scenarios.

Potential groundwater impacts associated with this type of strategy depend on a
variety of factors.  Any impacts may be acceptable if long-term contaminant
migration rates are low and/or if the landfill is located in a favorable hydrogeologic
setting. Conversely, if anticipated contaminant migration  rates are high and the
landfill is in a vulnerable hydrogeologic setting, the long-term consequences of this
strategy may not be acceptable.  In the future, when regulators and design
engineers assess the need to extend the post-closure period of a particular landfill
beyond the 30-year regulatory minimum, they will need to perform an assessment of
the type described in the preceding paragraph.  The authors recommend that, for
new landfills, this assessment process be included at the initial design stage, within
the framework of a long-term management strategy.

Strategy 3 - Standard Landfill with Clean Closure:  This strategy is one that has post-
closure care until, at some future date, the landfill will undergo clean closure. Clean
closure would involve mining the  landfilled waste and subjecting it to one or more
resource recovery operations, possibly including materials recovery and waste-to-
energy. Landfill mining has been tried at a few locations on a small scale (Nutting,
1994). However, regulatory, economic, and technology constraints make broad-
based application of this strategy infeasible at the present time.  Future technology
advancements and changing markets may change this situation to a degree where
resource recovery and clean closure become viable.  The advantages of this
strategy include removal of the contaminant source (i.e., clean closure) and
beneficial use of waste materials.  Since the landfill is clean closed, concerns about
the long-term performance of landfill waste containment system components are
also eliminated. The disadvantages of this strategy include cost, potential
environmental impacts associated with resource recovery operations, and the need
to redispose of waste residuals.

Strategy 4 - Recirculation Landfill with Perpetual Post-Closure Period:  The fourth
strategy is similar to Strategy 1, except that, instead of avoiding supplemental
moisture addition, leachate recirculation (or another form  of moisture addition) is
implemented for purposes of enhancing gas generation and waste stabilization.
With this technique, the landfill is viewed as an anaerobic bioreactor that both
accelerates the "biostabilization" of waste and the treatment of the recirculated
leachate (Pohland and Harper, 1986; Pohland et al. 1992).  Leachate recirculation
has been a technique under development for twenty years. Reinhart (1993) reports
that full-scale recirculation landfills are presently in operation or under construction in
twelve states, and that state regulations allow for the use of recirculation in all but
                                     G-5
-------
seven states.  Information on leachate recirculation and landfill bioreactor design has
been summarized by EPA (EPA, 1995).

The advantages of leachate recirculation include enhanced/accelerated gas
production and waste stabilization and removal or biofixation of leachable
constituents.  Pohland and Harper (1986) have suggested that leachate recirculation
reduces the time period for landfill stabilization from several decades to a few years.
This technique involves higher initial capital costs (e.g., recirculation hardware and,
possibly, an enhanced  liner system beneath the landfill),  but cost advantages are
gained by having less need for post-closure maintenance.  This technique also
results in a waste mass having a much lower "latent" threat of pollution than the
waste mass in a dry landfill. While this strategy requires perpetual care, it is
reasonable to assume that the necessary level of care will be less than with Strategy
1.  A current disadvantage of all of the recirculation strategies is the limited long-term
and large-scale field experience with this technique.  This situation will improve in
the coming years.

Strategy 5 - Recirculation Landfill with Limited Post-Closure Period: This strategy is
similar to the  preceding one except that the landfill has a limited post-closure care
period  (e.g., 30 years).  The same factors that affect the  long-term effectiveness  of
Strategy 2 apply to this strategy.  However, this strategy  has an advantage in that
recirculation results in more complete conversion of decomposable organic materials
than occurs in a standard landfill, as well as less potential for the  leaching of
hazardous constituents should renewed  infiltration occur after the end of the post-
closure period. It is envisioned that with leachate recirculation and adequate design
of the landfill  liner and final cover systems, Strategy 5 will be acceptable in most
hydrogeologic settings  and it will not be necessary to implement Strategy 4 (i.e.,
perpetual post-closure care).

Strategy 6 - Recirculation Landfill with Clean Closure:  This strategy is similar to
Strategies 4 and 5, except that after the waste has undergone biostabilization,  it is
removed from the landfill and processed as with Strategy 3. Processing may involve
recovery of any recyclable materials and incineration or land application of stabilized
waste residuals.  The main advantages this  option are the  same as with Strategy 4,
plus the benefits of resource recovery and clean closure of a site. Disadvantages
include higher capital costs and any potential health risks associated with clean closure.

One intriguing aspect of this strategy is that after the waste  is mined, the liner
system can be inspected and remediated if necessary and the site re-permitted to
accept new waste materials.  Even further, by proper sequencing of a large site,  a
perpetual placement/mining/replacement scheme can  be envisioned, see Figure G-
1.  Some countries with critical shortage of space are discussing this concept.
                                     G-6
-------
     A
B
D
                                                   Fill A,  B, C,  ...  H
                                                   Mine A, repairs fill
                                                   Mine B, repairs fill
                                                         etc.
                   Plan View

Figure G-1. Sequential mining of individual waste cells to provide for a
              perpetual landfill of a given footprint area.
Strategy 7 - Inward-Gradient Landfill: The seventh strategy can be similar to any of the
preceding six, except that an inward hydraulic gradient is included across all, or a
portion of, the liner system.  Inward gradients can be maintained by creating a hydraulic
pressure in the leak detection system (LDS) of a double-liner system (Rowe and
Fraser, 1993a,b) or triple-liner system (Giroud, 1984b), or by installing a single-liner
system in an excavation below the groundwater table.  This latter type of facility was
used in the early 1980s at landfills in the northcentral U.S., prior to the time when U.S.
regulations mandated the use of GMs.  Inward gradient landfills have also been termed
"zone of saturation landfills" (Oakley, 1987).

The inward gradient design concept may be used with CCLs, but the concept is not
directly applicable to GM/CCL composite liners (which is the typical configuration
used in modern U.S. landfills). Inward gradient landfills  appear to be of growing
interest in  Canada, but the regulatory framework in the U.S. precludes their use in
this country in most situations. Inward gradient facilities can be designed such that
inward flow velocities through the liner exceed outward chemical diffusion rates
(Rowe and Fraser, 1993a,  b).  However, the goal of preventing significant chemical
diffusion can also be achieved without an inward gradient by using a composite liner
and by designing the  LCRS above the composite liner to rapidly convey leachate to
a sump, thereby preventing the buildup of leachate head on the liner and a
sustained  leachate head on the liner.  It would be prudent to consider a GM/CCL/GM
three-component barrier system  for a liner system with an inward gradient design.
                                    G-7
-------
The main advantage of this strategy is that it provides an inward gradient that
prevents leakage or diffusion out of the unit.  It may also provide a setting in which
long-term maintenance of final cover systems,  liquid collection systems, etc., is
unnecessary. Disadvantages include the difficulty in gaining regulatory acceptance
for this type of unit, the need to use an engineered (rather than natural) hydraulic
control system at many sites, uncertainly associated with long-term groundwater
levels (at sites with natural hydraulic control systems), and the need to manage/treat
the large volume of liquid likely to require removal from the LCRS.

Each management strategy described above has minimum requirements for
providing short- and long-term protection of groundwater quality. These  minimum
requirements are summarized in Table G-1.  The strategies all require that for at
least some period of time, engineered systems be used for the containment and
collection of landfill leachate and gas. The strategies vary, however, in the required
service lives of the various systems and in detailed design performance criteria (e.g.,
contaminant transport rates, breakthrough times, mass fluxes, etc.).

G-3   Incorporating Management Strategies into Design
The authors believe that  insufficient attention is given to defining a long-term
management strategy during the permitting and design stages of most new landfills.
Both regulators and designers typically focus their efforts on achieving regulatory
compliance, developing an adequate detailed design, and developing specifications
and a construction quality assurance (CQA) plan to achieve satisfactory initial
construction.  Rarely do the project participants raise questions such as "what is our
strategy for long-term management of this facility, what are the design, construction,
operating, monitoring,  and maintenance actions required to implement the strategy,
and what funding should be set aside at the beginning of the project to enable
implementation?"  These questions should receive more attention during the landfill
design and permitting process.  Each new facility should have a well-defined
management strategy, a  well-defined set of design and performance criteria that will
achieve the strategy objectives, and financial assurances consistent with the
strategy. Figure G-2 provides a flow chart, developed by Bonaparte (1995), that
indicates the manner that long-term landfill management planning can be
incorporated into the landfill permitting and design process.

G-4   Landfill Maintenance, Monitoring, and Response Actions
Long-term maintenance is also a critical element of many of the landfill management
strategies described previously.  Thus, any plan for new landfill development should
contain the following elements:  (i) a program for long-term maintenance that is
consistent with the requirements of the long-term management strategy (the
maintenance  program will need to account for each waste containment system
                                    G-8
-------
component that contributes to achieving the objectives of the landfill management
strategy); (ii) financial funding mechanisms for long-term maintenance consistent
with the requirements of the strategy; and (iii) provisions for quality control and
quality assurance of required maintenance operations.

The monitoring programs for a landfill should also be developed to be consistent
with the long-term management strategy for that facility.  For example, for Strategy
1, a program should be established to monitor the final cover system on a
permanent basis. As another example,  if a strategy assumes that leachate
generation will cease after a certain period of time, monitoring should be used to
confirm this aspect of system performance. As a final example, if the strategy
involves a recirculation  landfill designed to obtain a high  degree of biostabilization of
the waste mass,  leachate and gas quantities and compositions should be monitored
during the recirculation  process to confirm that this objective is being achieved.
Monitoring results should be compared to design-phase  analysis and modeling
information to confirm that the behavior of the facility is as predicted.

Groundwater monitoring wells will continue to be a necessary element of landfill
monitoring programs, if for no other reasons than to meet the requirements of
existing regulations and the customs of most practicing professionals. Depending
on the management strategy, other types of monitoring should  also be considered,
including:  (i) vadose zone moisture, chemistry or pressures, for the early detection
of a release of either leachate or gas; (ii) LCRS and LDS flow rates  and  liquid
quality; (iii) liquid heads in LCRS sumps; (iv) LCRS and LDS hydraulic conductivities,
which may be assessed by injecting a tracer through a test pipe and monitoring the
travel time to a detection point (i.e., sump); (v) liner or final cover material integrity,
using test coupons or other means; and (vi) gas extraction system flow,  pressure,
and temperature. Finally, monitoring  programs should also describe response
actions, or at least steps to be taken to establish appropriate response actions,
should monitoring results reveal a problem. As indicated in Figure G-2,  potential
response actions should be evaluated during the initial design phase of a landfill
project. Design details  should be developed to enable their implementation should
the need for response actions ever arise.

G-5   Conclusion
Landfill containment and control systems must perform satisfactorily during the
entire period of significant leachate and gas generation.  This period may be
different for different types of facilities and for facilities in different climatic and
hydrogeologic settings.  In addition, standard landfills will retain a "latent" capacity to
generate leachate or gas for long periods of time.
                                    G-9
-------
                 Define site/climate
                   characteristics
      Define regulatory
        framework
 Define owner criteria (e.g.,
waste quantities, types, etc.)
                                             Identify potentially applicable long-term
                                              management strategies and develop
                                                  conceptual designs for each
                  Perform life-cycle
                cost-benefit analysis
                                         Define and evaluate
                                    preliminary performance criteria
                                                    Select long-term landfill
                                                    management strategy
                                                 Define data needs for strategy
                                                 and perform site investigation
                                                  and other required studies
                                           Perform life-cycle leachate and gas migration
                                           potential assessments and establish detailed
                                          performance criteria (e.g., maximum acceptable
                                          leachate head, minimum time of travel, minimum
                                          design life, etc.) to achieve strategy objectives
                                            Prepare detailed design (e.g., required liner
                                               thickness, liner and LCRS hydraulic
                                             conductivities, base grade, etc.) to satisfy
                                                     performance criteria
                                           Provide design redundancy and overdesign to
                                                  compensate for uncertainty
                Identify maintenance
            requirements to achieve strategy
                       goals
 Identify monitoring programs
required to assess progress of
   strategy implementation
  Establish response action plan
  to address problems identified
         by monitoring
                                          Establish quality control and quality assurance
                                          programs covering not only initial construction,
                                           but also operation, closure, and post-closure
                                                        maintenance
                                            Establish funding mechanism to fully cover
                                                    strategy requirements
Figure G-2.  Flow chart for incorporating  long-term management strategy into
landfill  permitting and design process (from Bonaparte, 1995).
                                                         G-10
-------
Seven strategies for addressing long-term landfill management were described in
this appendix. These strategies provide a framework for assessing the time periods
for leachate and gas generation, and the resulting minimum requirements for long-
term groundwater quality protection. Conservative designs, developed within the
framework of an appropriate long-term management strategy, can prevent both
short- and long-term leachate and gas impacts to groundwater.  Achievement of
these objectives is also dependent on the use of appropriate quality control and
assurance of not only landfill construction, but also landfill operations, closure, and
post-closure maintenance.  It is recommended that new landfills be designed and
permitted within the framework of a long-term management strategy.  The flow chart
presented in this appendix (Figure G-2) can be used for this purpose.

G-6    References
Bonaparte, R. (1995), "Long-Term Performance of Landfills," Proceedings of the
  ASCE Specialty Conference Geoenvironment 2000, ASCE Geotechnical Special
  Publication No. 46, Y.B. Acarand D.E. Daniel, eds., Vol. 1, pp. 514-553.
EPA (1995),  "Seminar Publication, Landfill Bioreactor Design and Operation,"
  EPA/600/R-95/146, Office of Research and Development, Washington, DC, 230
  P-
Giroud, J. P. (1984b), "Impermeability: The Myth and a Rational Approach,"
  Proceedings International Conference on Geomembranes, Vol.  1, Denver, pp.
  157-162.
Nutting, L. M. (1993),  "The Economics of Landfill Mining and Reclamation,"
  Proceedings Waste Tech '93, National Solid Waste Management Association,
  Marina Del Ray, 17 p.
Oakley, R. E. (1987),  "Design and Performance of Earth-Lined Containment
  Systems," Geotechnical Practice for Waste Disposal '87, Geotechnical Special
  Publication No. 13, R.  D. Woods, ed., pp. 117-136.
Pohland, F. G., Cross, W. H., Gould, J. P. and Reinhart, D. R. (1992), "Behavior and
  Assimilation of Organic and Inorganic Priority Pollutants Codisposed with
  Municipal Waste," Report for cooperative agreement CR-812158, EPA Risk
  Reduction Research Laboratory, Cincinnati, 197 p.
Pohland, F. G. and Harper,  S. R. (1986), "Critical Review and Summary of Leachate
  and Gas Production from Landfills," EPA/600/2-87/044, EPA Hazardous Waste
  Engineering Research Laboratory,  Cincinnati, 186 p.
Reinhart, D.  R. (1993), "Wet vs. Dry Landfilling: Does it Really Make a Difference,"
  Proceedings Waste Tech '93, National Solid Waste Management Association,
  Marina Del Ray, 9 p.
Rowe, R. K. and Fraser, M. J. (1993a), "Long-Term Behavior of Engineered Barrier
  Systems," Proceedings, IV International Landfill Symposium, Sardinia, pp. 397-
  406.
                                   G-11
-------
Rowe, R. K. and Fraser, M. J. (1993b), "Service Life of Barrier Systems in the
  Assessment of Contaminant Impact," Geotechnical Research Center Report,
  GEOT-5-93, The University of Western Ontario, 8 p.
Tchobanoglous, G., Theisen, H. and Vigil, S. (1993), "Integrated Solid Waste
  Management," McGraw-Hill, Inc., New York, 978 p.
                                   G-12
-------