Geotextiles and Geomembranes xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Geotextiles and Geomembranes

journal homepage: www.elsevier.com/locate/geotexmem

Magnitude and significance of tensile strains in geomembrane landfill

liners☆
R. Kerry Rowea,∗, Yan Yub
a
GeoEngineering Centre at Queen's-RMC, Department of Civil Engineering, Queen's University, Kingston, ON, K7L 3N6, Canada
b
Key Laboratory of Geotechnical and Underground Engineering of Ministry of Education, Department of Geotechnical Engineering, Tongji University, Shanghai, 200092,
China

A R T I C LE I N FO A B S T R A C T

Keywords: The implications of the tensile stress/strain developed in high density polyethylene (HDPE) geomembranes
Geosynthetics (GMB) is explored in the context of a reduction in stress crack resistance due to ageing in contact with leachate in
Geomembranes a municipal solid waste (MSW) landfill. The experimental evidence of GMB cracking and ultimately failure when
Strains subject to excessive tensile strains is discussed to highlight the need to limit the maximum tensile strain sus-
Landfills
tained by an HDPE GMB to an acceptable level if good long-term performance is to be ensured. The effect of both
Indentations
Side slopes
local GMB indentations induced by gravel in an overlying drainage layer or an underlying clay liner on tensile
strain is reviewed. In addition, the tensile strains caused by down-drag in the GMB on side slopes with settlement
of the waste is examined. The key research related to tensile strains developed in GMBs from these sources is
reviewed and new data presented. It is shown that an appropriate protection layer over the GMB can limit local
GMB tensile strains to less than 3% and that the selection of a suitable slope inclination and stiffness of a
geotextile reinforcement layer can limit the GMB strains due to down-drag to less than 2% and geotextile strains
to less than 4% after long-term waste settlement.

1. Introduction The aging of a HDPE GMB (simply referred to as a GMB hereafter) is

A geomembrane (GMB) liner is a key component of the barrier
system in many modern engineered landfills. In combination with a
clay liner, the GMB minimizes contaminant migration to groundwater
and surface water and therefore to reduce the potential impacts on the
human health and the surrounding environment (Rowe et al., 2004).
GMBs in landfill applications are mostly made from high-density
polyethylene (HDPE). A barrier system consists of a highly permeable
leachate collection system (LCS) and a low permeability liner system.
As part of a landfill composite liner, HDPE GMBs are excellent barriers
for harmful inorganic substances (e.g., heavy metals) typically found in
landfills, and when combined with an underlying geosynthetic clay
liner (GCL) or compacted clay liner (CCL) can perform their intended
functions extremely well for both organic and inorganic substances
(Rowe, 2012). However, with time, HDPE GMBs will experience a loss
of their mechanical properties (e.g., Hsuan and Koerner, 1998; Rowe
and Sangam, 2002; Sangam and Rowe, 2002; Müller and Jacob, 2003;
Rowe et al., 2008, 2009, 2010).
typically considered to have three stages to its degradation (Hsuan and
Koerner, 1998): (I) antioxidant depletion; (II) an induction period be-
tween antioxidant depletion and the onset of polymer degradation; and
(III) oxidative degradation of polymer resulting in a decrease of the
physical properties. When GMB degradation is such that it can no
longer resist the tensile strains/stresses, fully penetrating cracks de-
velop (Abdelaal et al., 2014; Ewais et al., 2014) that can allow the es-
cape of leachate. Once this escape exceeds allowable design values, the
GMB is considered to have reached the end of its service-life (Rowe
et al., 2004).
There are two key design-related sources of potential significant
tensile strains in GMBs used in landfill bottom liners, viz: (a) local in-
dentations of GMBs induced by the overlying drainage materials and/or
by gravel in the underlying clay liner, and (b) down-drag load for GMBs
on side slopes generated by waste settlement. Considerable full scale
experimental work has been conducted to examine the local strains
associated with the indentations of the GMB liners caused by overlying
granular drainage layer with and without a protection layer (e.g.,

☆
This is an expanded version of the Keynote Lecture: “Tensile strains in geomembrane landfill liners” in the conference proceeding and presented at GeoShanghai,
May 2018.
∗
Corresponding author.
E-mail addresses: kerry.rowe@queensu.ca (R.K. Rowe), yan.yu@queensu.ca (Y. Yu).

https://doi.org/10.1016/j.geotexmem.2019.01.001
Received 26 July 2018; Received in revised form 4 November 2018; Accepted 7 December 2018
0266-1144/ © 2019 Elsevier Ltd. All rights reserved.

Please cite this article as: Rowe, R.K., Geotextiles and Geomembranes, https://doi.org/10.1016/j.geotexmem.2019.01.001
R.K. Rowe, Y. Yu
Brachman and Gudina, 2008a; 2008b; Dickinson and Brachman, 2008;
Brachman and Sabir, 2010, 2013; Hornsey and Wishaw, 2012; Sabir
and Brachman, 2012; Rowe et al., 2013a,b) and by the granular par-
ticles in an underlying compacted clay liner (Brachman and Sabir,
2010). Physical and numerical analyses have also been undertaken to
estimate the GMB tensile strains on side slopes under waste settlement
(e.g., Jones and Dixon, 2005; Thusyanthan et al., 2007; Fowmes, 2007;
Fowmes et al., 2008a; Arab, 2011; Sia and Dixon, 2012; Wu, 2013;
Zamara et al., 2014; Kavazanjian and Gutierrez, 2017; Yu and Rowe,
2018a; 2018b; Kavazanjian et al., 2018). The research outcomes from
the papers cited above allow designers to estimate the GMB strains that
may be anticipated for the different design scenarios and to optimize
the GMB liner design under the site-specific conditions.
Short-term punctures can generally be minimized by providing
sufficient protection to the GMB liner (e.g., Koerner et al., 1996; Narejo
et al., 1996; Tognon et al., 2000; Gudina and Brachman, 2006; Koerner
et al., 2010; Dickinson and Brachman, 2008), the magnitude of tensile
strains that a GMB can sustain without compromising their intended
long-term performance reported in the literature varies. To avoid pre-
mature GMB failure due to stress cracking, Seeger and Müller (2003)
indicated that the GMB strain should be less than 3%. In contrast, Peggs
et al. (2005) recommended an allowable tensile strain of 6–8% for a
smooth GMB and 4% for a textured GMB to reduce the potential for
long-term GMB brittle rupture. Rowe et al. (2019) considered tensile
strains for a low level radioactive waste facility with a 550-year design
life and for the 2 mm-thick textured candidate GMBs recommended
maximum allowable strains (from all sources) of 3% on the base, 4% on
side slopes, and 5% in the cover where repair and replacement was
possible. Based on the GMBs examined under the simulated field con-
ditions in the geosynthetic liner longevity simulator (GLLS) cells (Rowe
et al., 2013a,b; Abdelaal et al., 2014; Ewais et al., 2014), it can be
inferred (Rowe, 2018a) that sustained tensions that induce tensile
strains greater than 4% should be avoided by the use of a suitable
protection layer between the GMB and LCS (Rowe et al., 2013a,b;
Abdelaal et al., 2014; Ewais et al., 2014), eliminating potentially pro-
blematic stones from the upper layer of a clay liner or other subgrade,
and designing to limit strains from other sources such as down-drag by
waste placement and subsequent settlement/degradation.
Giroud et al. (1995) reported that strain concentrations in the vi-
cinity of seams can give rise to failure adjacent to seams. Laboratory
testing reported by Kavazanjian et al. (2017) indicated that the strain
magnification induced by a seam was even greater than estimated by
Giroud et al. (1995) and that the allowable tensile strain for GMBs
should likely be less than 4% when the GMB is loaded in tension per-
pendicular to the seam.
The objective of this paper is to draw together and summarize the
research related to the generation of tensile strains in GMBs used in
landfill liners due to indentations, to present new results related to
tensile stresses/strains induced by down-drag of GMBs on side slopes
due to waste settlement, and to address the magnitude and significance
of these tensile strains.
2. GMB tensile strains from indentations
2.1. GMB indentations and laboratory testing methods
A GMB overlain by a gravel drainage layer can experience strains/
stresses in the GMB due to local indentations caused by the gravel
particles during construction and when subjected to overburden
stresses (e.g., the weight of the waste body). A protection layer is re-
quired between the GMB and the drainage layer to prevent the short-
term puncture of the GMB (Koerner et al., 1996; Narejo et al., 1996)
and to minimize the long-term tensile strains in the GMB (Rowe et al.,
2013a,b; Abdelaal et al., 2014; Ewais et al., 2014). However, quanti-
fying the long-term GMB tensile strains due to indentations in the field
remains challenging. Laboratory testing to establish the short-term
2
Geotextiles and Geomembranes xxx (xxxx) xxx–xxx
(typically in a 10- to 100-h sustained pressure test) GMB tensile strains
developed with a proposed protection layer over the GMB is currently
the most feasible way to qualify the efficiency of protection layers in
term of reducing/limiting the GMB tensile strains (e.g., Sabir and
Brachman, 2012; Brachman and Sabir, 2013).
Standard laboratory test methods (e.g., EN 13719; ASTM D5514)
can be used to examine the short-term GMB tensile strains with and
without a protection layer. Index tests using steel plates with fabricated
protrusions to simulate gravel particles have also been performed to
quantify the efficiency of geosynthetic protection layers for GMBs (e.g.,
Brummermann et al., 1994; Narejo et al., 1996). Large-scale laboratory
test apparatus has also been developed to examine the influence of
different protection layers and interactions between the different layers
on the GMB strains (e.g., Brachman and Gudina, 2008a; 2008b;
Dickinson and Brachman, 2008; Sabir and Brachman, 2012) in a full
barrier system. A thin lead sheet (0.4-mm thick) is used beneath the
GMB to record the GMB deformations in these laboratory tests.
2.2. Strain calculation methods
The vertical displacements of the GMB surface recorded by the thin
lead sheets (e.g., as measured by a laser scanner) can be used to cal-
culate the GMB tensile strains (e.g., Gallagher et al., 1999; Zanzinger,
1999; Tognon et al., 2000; Hornsey and Wishaw, 2012). However, the
magnitude of the calculated GMB strains based on the indentations
recorded in the metal sheet is highly dependent on the strain calcula-
tion methods (e.g., BAM, 2015; ASTM D5514; LEF-2 2014; Tognon
et al., 2000; Eldesouky and Brachman, 2018) as discussed below.
In the BAM and ASTM D5514 approaches, the local membrane
strain is calculated by fitting a circular segment to the indentation in
the lead sheet used (noting that there is an error in the equation given
in ASTM D5514 2014). The LEF-2 (2014; the landfill engineering gui-
dance by the United Kingdom Environmental Agency) provides both
local and incremental strain calculation methods based on the 3-mm
segments of the measurement axes of the indentation. Recognizing that
a 1.5 mm-thick (or greater) GMB has bending as well as membrane
strains, an improved alternative approach to calculate the incremental
strains was proposed in Tognon et al. (2000) using the vertical de-
formed GMB profile recorded in the lead sheet to assess both membrane
and bending incremental strains. All these methods err in under-
estimating the strains since they neglect the horizontal displacements.
The Tognon et al. (2000) method is currently the best of the method
based on vertical displacement. Eldesouky and Brachman (2018) have
recently proposed an alternative method to calculate the GMB incre-
mental strains that considers both the vertical and horizontal dis-
placements of the GMB under the axisymmetric conditions. They used a
best-fit polynomial for the measured vertical displacements together
with an Airy's stress function (without a need to measure the radial
displacements) to estimate the radial strains for the midsurface. How-
ever, at present, the Eldesouky and Brachman (2018) method is only
suitable for the axisymmetric conditions and is not yet in a form that
can be used to calculate strains in GMBs below protection layer and
gravel drainage layer in landfill applications.
Table 1 shows the calculated maximum strains for the same in-
dentation (the Profile B reported by Hornsey and Wishaw, 2012, Fig. 1)
based on the different strain calculation methods. Where reported va-
lues obtained by Hornsey and Wishaw (2012) are given together with
values from the authors’ independent calculation. The indentation has a
sharp turn at the real point of maximum strain. However, the capacity
of the various methods to capture this varies substantially. The calcu-
lated local strain based on the ASTM arch elongation method is 0.016%
(remarkably low due to the error in the equation given in ASTM D5514
2014). The use of the BAM arch elongation method (with a correct
equation) results in the local strain of 4.1%; this is in essence an average
over the indention and does not capture the critical location. Based on
the measured vertical displacements at the 3-mm intervals, the LFE-2
R.K. Rowe, Y. Yu Geotextiles and Geomembranes xxx (xxxx) xxx–xxx

Table 1 nominal grain size of 25 mm and the other with a nominal 50 mm,
Calculated maximum GMB strains based on different strain calculation methods placed on a 1.5 mm-thick GMB without a protection layer under an
for the same indentation. Note: all calculated maximum GMB strains are based applied pressure of 250 kPa (sustained for 10 h at 21 ± 2 °C). The GMB
on the Profile B reported by Hornsey and Wishaw (2012). rested on a clay liner compacted at 16% gravimetric moisture content
Strain calculation method Calculated maximum GMB strain (%) (i.e., 4% wet of optimum for this clay with a plastic limit of 16%). Five
types of contacts were identified; viz., point, edge, area, perimeter, and
Hornsey and Wishaw (2012) This papera composite (Brachman and Gudina, 2008a). Based on the GMB in-
ASTM arch elongation method 0.016 0.016b dentations and strains reported, the point contacts were the most fre-
BAM arch elongation method NA 4.1b quent (approximately 40% of the contacts). The point and edge contacts
LFE-2 local strain (3 mm) 3.9 3.7c resulted in the deepest indentations and the largest tensile strains.
LFE-2 incremental strain (3 mm) 13.6 13.1c These tests demonstrated that the nominal grainsize of the gravel in-
LFE-2 local strain (1 mm) NA 4.5
fluenced both average spacing between the gravel contacts but also on
LFE-2 incremental strain (1 mm) 20d 42d
Tognon et al. (2000) (1 mm) NA 75 the maximum tensile strain recorded in the GMB. For example, the
average spacing between the gravel contacts was 37 mm for the 25-mm
a
The Profile B from Hornsey and Wishaw (2012) was digitized first ap- gravel and this increased to 55 mm for the 50-mm gravel. The max-
proximately at 1-mm intervals within the horizontal distance between −22.5 imum GMB tensile strains were 16 and 32% for the 25-mm and 50-mm
and 22.5 mm, and then the interpolation method was used to get the profile gravel, respectively (Fig. 3; Table 2). There was no short-term puncture
exactly at 1-mm intervals. of the GMB but these strains are far too large for GMBs in the landfill
b
Based on the width of the indentation = 45 mm and the depth of the in-
applications. Thus, a protection layer must be included between the
dentation = 5.6 mm.
c GMB and overlying gravel layer to limit the GMB tensile strains.
Data points at 3-mm intervals were selected according to the digitized
profile at 1-mm intervals.
d
The strain values are very sensitive to the locations where the vertical
displacements are taken and the method of processing the data (e.g., inter- 2.4. Influence of geotextile protection on maximum GMB tensile strain
polation or polynomial fit).
Brachman and Gudina (2008b) reported the results of tests with a
local and incremental strain methods estimated the maximum GMB 50-mm gravel drainage layer with and without a geotextile protection
strains of 3.7% and 13.1%, respectively (similar to values calculated by layer (Fig. 4) above a 1.5-mm thick HDPE GMB overlying a needle-
Hornsey and Wishaw, 2012). Reducing the measurement interval from punched geosynthetic clay liner (GCL) which in turn rested on a firm
3 mm to 1 mm increases the maximum strain from 3.7% to 4.5% for the foundation layer. The barrier system was subjected to an applied load of
LFE-2 local strain method and from 13.1% to 42% for the LFE-2 in- 250 kPa for 10 h at 21 ± 1 °C.
cremental strain method. However, these methods do not capture the In contrast to an otherwise similar test (discussed above) which had
bending strain which is substantial at such a tight point of curvature at a maximum GMB tensile strain from 32% with no protection layer for
the corner of the indentations. Consideration of both the membrane and the GMB over the compacted clay liner (Brachman and Gudina, 2008a),
bending strains using the Tognon et al. (2000) method results in a in this case the maximum strain was reduced to 17% (Fig. 5; Table 2)
maximum GMB strain of 75%. when the GMB was underlain by a GCL on the firm foundation
(Brachman and Gudina, 2008b). Thus, increasing the stiffness of the
foundation from a compacted clay liner at wopt+4% to a GCL on a firm
2.3. Influence of gravel size on maximum GMB tensile strain foundation almost halved the maximum strain demonstrating that the
nature of the material below the GMB can have a significant effect on
Brachman and Gudina (2008a) performed laboratory experiments local strains and hence the need to simulate the entire barrier system
(Fig. 2) using two poorly graded and angular gravels, one with a when assessing potential GMB strains on important projects.

Fig. 1. Cross-section of indentation examined in Table 1 (“Profile B”; modified from Hornsey and Wishaw, 2012).

3
R.K. Rowe, Y. Yu
Fig. 2. Cross-section of test cell (modified from Brachman and Gudina, 2008a).
Fig. 3. Distribution of GMB strains with 25-mm and 50-mm gravels (modified from Brachman and Gudina, 2008a).
A geotextile protection layer between the GMB and gravel drainage
layer reduced the maximum GMB strain for the case with a GMB over a
GCL and firm foundation layer (Fig. 5; Table 2; Brachman and Gudina,
2008b) and a CCL (Table 2). But the extent to which the GMB strain can
be reduced by a geotextile protection layer is dependent on its prop-
erties. The mass per unit area is the most commonly specified property
of a protection layer. For similar needle punched geotextiles, the larger
the mass per unit area of the geotextile, the smaller the GMB strain
(Fig. 5). When using a geotextile with the mass per unit area of 2200 g/
m2 (the highest among three geotextiles tested by Brachman and
Gudina, 2008b), the GMB strains were just below 6% compared to 17%
without a geotextile protection layer. The limitation of using only unit
weight to specify a protection geotextile was noted in Jones et al.
(2000).
Mass per unit area is only part of the story and the tensile stiffness
and creep characteristics of geotextile may also be expected to play a
role. While the tensile stiffness may be correlated with mass per unit
area this only applies to otherwise similar geotextiles as was illustrated
by the performance of the composite geotextile examined by Dickinson
and Brachman (2008) discussed below.
Dickinson and Brachman (2008) examined a GMB-GCL landfill
4
Geotextiles and Geomembranes xxx (xxxx) xxx–xxx
barrier separated from the 50-mm gravel by a geocomposite with
layered geotextiles under an applied vertical pressure of 250 kPa for
10 h at 22 ± 2 °C. The 1100 g/m2 geocomposite (GCTX) halved the
maximum GMB tensile strains to 3.9% (Table 2) compared to 8% ex-
pected for traditional 1100 g/m2 nonwoven geotextile (Fig. 5) and even
to 2.6% with a mass per unit area of about 3000 g/m2 when a thicker
nonwoven core was installed between two thinner and stiffer outer
layers. However, even with a 1100 g/m2 composite geotextile protec-
tion layer, a 3.9% GMB strain is still too large and even the 2.6% for the
3000 g/m2 may not be adequate because these strains are only estab-
lished for short-term (10 h) loading conditions at room temperature.
For long-term field conditions, especially with elevated temperatures
and chemical exposure encountered in MSW landfills, GMB strains
greater than these short-term values are expected as discussed in Sec-
tion 2.6.
2.5. Influence of alternative protection layers on maximum GMB tensile
strain
In addition to the geocomposite (GCTX) discussed above, Dickinson
and Brachman (2008) reported laboratory experiments at 250 kPa for
R.K. Rowe, Y. Yu Geotextiles and Geomembranes xxx (xxxx) xxx–xxx

Table 2
Maximum 1.5 mm HDPE GMB strains in full scale (GLLS) tests at 250 kPa, room temperature, and 10 h sustained loading (except as noted).
Type of foundation (thickness) below GMB Gravel D60 (mm) Protection layer Maximum strain (%)

a
CCL (100 mm) 25 None 16
a
CCL (100 mm) 25 570 g/m2 GTX 13
a
CCL (100 mm) 25 2200 g/m2 GTX 4.5
a
CCL (100 mm) 50 None 32
a
CCL (100 mm) 50 390 g/m2 GTX 19
a
CCL (100 mm) 50 540 g/m2 GTX 15
a
CCL (100 mm) 50 1200 g/m2 GTX 9.1
a
CCL (150 mm) 50 1200 g/m2 GTX 13
a
CCL (100 mm) 50 2200 g/m2 GTX 8.2
a
CCL (100 mm) 50 Sand (150 mm) < 0.5
b
GCL (w = 7%) + CCL (150 mm) 50 None 22
b
GCL (w = 7%) + SP (150 mm) 50 None 13
b
GCL (w = 128%) + SP (150 mm) 50 None 17
c
GCL (w = 128%) + SP (150 mm) 50 None 19
b
GCL (w = 170%) + SP (150 mm) 50 None 20
c
GCL (w = 127%) + SP (150 mm) 50 270 g/m2 GTX 15
b
GCL (w = 129%) + SP (150 mm) 50 390 g/m2 GTX 10
c
GCL (w = 118%) + SP (150 mm) 50 570 g/m2 GTX 11
d
GCL (w = 120–130%) + SP∗(150 mm) 50 580 g/m2 GTX at 55 °C for 1 month 15
c
GCL (w = 128%) + SP (150 mm) 50 1200 g/m2 GTX 7.6–7.7
c
GCL (w = 148%) + SP (150 mm) 50 2200 g/m2 GTX 5.1–5.5
c
GCL (w = 124%) + SP (150 mm) 50 790 g/m2 GCTX 6.7
c
GCL (w = 118%) + SP (150 mm) 50 1000 g/m2 GCTX 5.4
c
GCL (w = 121%) + SP (150 mm) 50 1100 g/m2 GCTX 3.9
c
GCL (w = 114–133%) + SP (150 mm) 50 3000 g/m2 GCTX 2.0–2.6
c
GCL (w = 126%) + SP (150 mm) 50 Sand (150 mm) 0.1–0.2
d
GCL (w = 120–130%) + SP∗(150 mm) 50 SP∗(150 mm) at 55 °C for 40.6 months <1
c
GCL (w = 125–135%) + SP (150 mm) 50 Geonet 13–16
c
GCL (w = 130–135%) + SP (150 mm) 50 Tire shreds (150 mm) 5.2–6.3
c
GCL (w = 117–124%) + SP (150 mm) 50 Tire shreds (150 mm) + 570 g/m2 GTX 2.3–2.8

Note: 25 mm-gravel: D90 = 35 mm, D85 = 30 mm, D60 = 25 mm, D50 = 23 mm, D10 = 15 mm.
50 mm-gravel: D90 = 63 mm, D85 = 55 mm, D60 = 45–50 mm, D50 = 47 mm, D10 = 32–40 mm.
SP (poorly graded sand): D100 = 5.5 mm, D90 = 3 mm, D60 = 0.58 mm, D50 = 0.45 mm, D10 = 0.15 mm.
SP∗(poorly graded sand): D85 = 2 mm, D60 = 0.54 mm, D50 = 0.46 mm, D10 = 0.16 mm.
CCL: compacted clay liner; GCL: geosynthetic clay liner; GTX: geotextile (mass per unit are rounded to 2 significant figures); GCTX: Composite GTX; Geonet: 5.4 mm-
thick under 20 kPa, solid ribs with a mass of 611 g/m2, and 9.5 mm-spacing between the parallel ribs in one direction and 10.5 mm in the other direction.
a
Brachman and Gudina (2008a).
b
Brachman and Gudina (2008b).
c
Dickinson and Brachman (2008).
d
Rowe et al. (2013a).

10 h at 22 ± 2 °C using a geonet, rubber tire shreds, and sand as al-
ternative protection layers between a 50-mm gravel drainage layer and
a GMB-GCL composite landfill liner. The use of a biaxial geonet
(Table 2) as the protection layer resulted in maximum GMB strains
ranging from 13% to 15% in multiple tests. Thus, the geonet was unable
to provide adequate protection to the GMB-GCL liner from excessive
strains. A 150-mm thick layer of tire shreds limited the maximum
strains to 5.2–6.3% when used directly over the GMB and even lower
(2.3–2.8%) if a single layer of geotextile with a mass per unit area of
570 g/m2 was between the tire shreds and the GMB. Thus a thicker
geotextile and tire shreds may have potential for a protection layer.
However, a geotextile would also likely be needed between the gravel
and tire shreds (as a separator) and careful consideration would need to
be given to the effect of compressibility of the tire sheds (Rowe and
McIsaac, 2005) on differential movement of the pipes in the drainage
layer. Generally, tire shreds cannot be considered a replacement for

Fig. 4. Cross-section of test cell for GMB with a geotextile protection layer (modified from Brachman and Gudina, 2008b).

5
R.K. Rowe, Y. Yu Geotextiles and Geomembranes xxx (xxxx) xxx–xxx

Fig. 5. Influence of protection layers on reducing GMB strains (modified from Brachman and Gudina, 2008b). Arrows show strain inferred for a normal needle
punched nonwoven geotextile at 570 and 1100 g/m2.

gravel in a MSW leachate collection system (Rowe and McIsaac, 2005; Table 3
McIsaac and Rowe, 2005). Averaged maximum GMB strains reported by Sabir and Brachman (2012) for
Brachman and Gudina (2008b), Dickinson and Brachman (2008), idealized stone at 250 kPa (based on nominal 50 mm gravel; CCL compacted at
and Rowe et al. (2013a) in separate experiments demonstrated that a wopt+4% = 16%).
150-mm thick sand layer limited the GMB tensile strain to less than Protection Temperature (oC) Averaged maximum GMB strain (%) after
0.2% making it by far the most effective protection layer examined for a different periods of sustained loading
GMB under 50-mm gravel at 250 kPa. It can be anticipated that these
10 h 100 h 1000 h 10000 h
strains would not be significantly affected by either temperature or
time, unlike those with a geotextile protection layer discussed in the None 22 10.9 12.0 13.8
next section. Unfortunately, the sand layer cannot be considered part of None 35 12.6 13.6
the leachate drainage layer in a MSW landfill (Rowe, 2005). None 55 14.9 15.3 17.2 18.0
None 70 15.5 16.7
None 85 16.7 19.1 20.5
2.6. Influence of time and temperature on maximum GMB tensile strain 540 g/m2 GTX 22 8.7 9.7 10.8
540 g/m2 GTX 35 9.8 10.9
540 g/m2 GTX 55 11.7 12.2 13.7
The laboratory experiments conducted in Brachman and Gudina 540 g/m2 GTX 70 13.5
(2008a), Brachman and Gudina (2008b), and Dickinson and Brachman 540 g/m2 GTX 85 14.1 16.0 > 17.2a
(2008) were based on the applied pressure of 250 kPa held for a
a
duration of 10 h at room temperature (21–22 °C). However, under the 800 h.
field conditions in the landfill, the GMB is expected to be loaded much
longer and subjected to temperatures of at least 35–40 °C in the me- GMB tensile strains increased from 14.9 to 18.0% when time was in-
thanogenic phase for normal landfills but also in some cases to 50–60 °C creased from 10 to 10000 h (1.14 years). Increasing the temperature
and in other cases ≥85 °C (Rowe, 2012). from 22 to 85 °C increased the GMB tensile strain observed after 1000 h
Sabir and Brachman (2012) explored the effect of temperature and from 13.8% (22 °C) to 20.5% (85 °C).
time for an isolated particle (simulated by a 28-mm-diameter machined The significance of the foregoing becomes even more apparent in
steel probe) indenting a 1.5-mm thick HDPE GMB overlying a com- full scale tests carried to the point where the GMB ruptures. For ex-
pressible clay liner (compacted at 16% moisture content which was 4% ample, Ewais et al. (2014) examined a 1.5-mm thick GMB aged in a
wet of optimum and at the plastic limit for this soil). Two cases were typical landfill composite liner configuration under a pressure of
examined (i) no protection, and (ii) a 540 g/m2 (4-mm thick) nonwoven 250 kPa at 85 °C where a 560 g/m2 geotextile protection layer was used
geotextile (Table 3). The particle was subject to a sustained vertical between the GMB and the 50-mm gravel drainage layer containing
force corresponding to that expected with a gravel layer subject to an leachate. The strains at the rupture locations were reported to be
average applied stress of 250 kPa. The laboratory results indicated that 24 ± 6% after about 2 years (17,500 h) of sustained load when the
the machined probe was able to closely reproduce the average strains stress crack resistance (SCR) of the GMB was about 760 h and tensile
from real 50-mm gravel. For both cases the strain increased with both break strength averaged 42.5 ± 12 kN/m (Ewais et al., 2014, Table 4).
the period of sustained loading and increased temperature. At room Based on Table 2, the expected short term (10 h) strain would be about
temperature the strain increased from the 10-h value by around 1.4% 11% at room temperature. Considering a projection from Sabir and
per log cycle of time without protection and by 1% per log cycle with Brachman (2012) and Table 3, one might expect a strain of about 22%
the 540 g/m2 GTX (Table 3). At 85 °C the strain increased from the 10-h after 20–26 months (14000–19000 h); this falls within the range of
value by 1.9% per log cycle of time both without and with the 540 g/m2 24 ± 6% that was observed (Ewais et al., 2014, Table 4).
GTX (Table 3). Ewais et al. (2014) reported that the observed ruptures were pre-
For a GMB without a protection layer, at the temperature 55 °C the dominately in the machine direction of the GMB due to a lower initial

6
R.K. Rowe, Y. Yu Geotextiles and Geomembranes xxx (xxxx) xxx–xxx

Table 4 after 10 h at 250 kPa for 25 mm-gravel (Table 2), 19 mm-gravel gave a
Maximum GMB strains at rupture locations reported in literature for a range of strain of 23–25% after 100 h at 2000 kPa when the unprotected GMB
cases in full scale (GLLS) tests involving GCL (hydrated) + SP (150 mm thick) rested on a CCL at 16% water content. Thus while the conditions are not
foundation, 1.5 mm HDPE GMB overlain by 560 g/m2 GTX and 50 mm gravel at directly comparable, it does indicate that increasing compressive stress
250 kPa. Note: each test was terminated and strains evaluated 24 h after a leak
does increase the tensile strain. Some have argued that one cannot get
(signaling the first rupture) was detected.
tensile stress in a GMB liner under compressive stress – the results in
Temp. (oC) Time to rupture Minimum Maximum Average (SD) Tables 2–5 show the error in that argument, which fails to recognize the
(hours) strain at strain at strain at effects of non-uniformity in the distribution of the vertical compressive
rupture (%) rupture (%) rupture (%)
stresses at the level of the GMB.
a
55 770–794 7.9 7.9 7.9 At 2000 kPa for 100 h and for 18-19 mm-gravel (Table 5), the
a
65 240–268 7.4 10.8 9.0 ± 1.5 maximum GMB tensile strain due to indentations for a GMB was (i)
a
70 100–124 7.0 11.8 9.9 ± 2.0 15–18% when the GMB was on a compacted silty sand, (ii) 12–18%
a
75 87–111 6.0 15.7 10 ± 2.9
a when on an air dried GCL on sand, (iii) a few percent higher at 14–21%
85 24–48 9.0 15.3 12 ± 2.0
b
85 14000–19000 14.9 58.9 24 ± 6 when on a GCL at 86% on sand, (iv) a little higher at 18–25% when on a
150 mm-thick CCL at optimum water content, wopt; and (v) 23–25%
Note: GCL: geosynthetic clay liner; SP: poorly graded sand layer; GTX: geo- when on a 150 mm-thick CCL at wopt + 4%. Thus the subgrade made
textile; SD = standard deviation. some difference with the silts and air dried GCL being, as one might
a
Abdelaal et al. (2014); GMB pre-aged to 75 h stress crack resistance (SCR). expect, at the lower end of the range and the CCL compacted 4% wet of
b
Ewais et al. (2014); GMB OIT-depleted to 760 h SCR. optimum (the plastic limit) the highest end. That said the range in
maximum tensile strain was relative narrow (from 18 to 25%) and 18 to
SCR in the cross-machine direction than in machine direction. Although 19 mm-gravel in direct contact can have a very significant effect on
there were considerably more ruptures at indentation slopes (i.e., at the tensile strain in all cases without a suitable protection layer. When a
side slope of indentations with 48% of 123 ruptures) and beneath sand protection layer was used the strain dropped to 2% even when the
gravel contacts (34%) than between indentations (18%), the ruptures stress was increased to 3000 kPa with the strains in this case being due
between indentations were both longest (up to 100 mm) and widest to subtle differences in the surface and density of the subgrade.
making them especially hydraulically significant. Depending on the The effect of stress for otherwise similar conditions (Table 5) can
orientation of the strains, the rupture may or may not occur at the point increase the GMB maximum tensile strain significantly as illustrated by
of maximum strain. Thus for the Ewais et al. (2014) tests, while the an almost ideal foundation (compacted sand) when 18 mm-gavel is in
average strain at ruptures agreed with the projected maximum strain direct contact with the GMB. It increased (Table 5) from 5 to 6% at
quite well (with a large number of ruptures), there were ruptures at 500 kPa to 7–8% at 1000 kPa, 15–18 at 2000 kPa and 18–27% at
strains of 18% (well below the projected maximum strain of about 3000 kPa.
22%). The potential for ruptures to occur at strains less than the max- For poorly graded material, the grainsize (D60) of the drainage
imum strain becomes even more evident as the SCR of the GMB is re- medium is of critical importance (Table 5), at 2000 kPa and with a GMB
duced with ageing as in the Abdelaal et al. (2014) experiments on the directly on a sand subgrade, with 2 mm giving 8–9% strain, 8 mm
same GMB under these same conditions except that the GMB had been giving 11–12% strain, and 18 mm giving 15–18% when no protection
aged to about 75 h SCR (i.e., 10% of that tested by Ewais et al., 2014). layer is used. At 3000 kPa and other conditions similar, these strains
In this case, at 85 °C the GMB ruptured after 1 day (24 h) of sustained increase to 11–12% (2 mm), 13–15% (8 mm), and 18–27% (18 mm),
loading (at an average rupture strain of 12% but with ruptures for respectively when no protection layer is used.
strains between 9% and 15%) compared to the projected maximum
strain of 12.3% for 24 h when the first crack appeared and 13.4% after 2.8. Influence of isolated stones in a CCL on maximum GMB tensile strain
48 h when the test was terminated (based on a 10-h strain of 11% at
room temperature from Table 2 and the same rate used for Ewais et al. The foregoing all related to the drainage gravel being the primary
(2014) as inferred from Sabir and Brachman 2012 data in Table 3). source of the indentions that cause tensile strain in a GMB. However,
There were a large number of cracks when the test was terminated 24 h this is not necessarily the case, especially if the material below the GMB
after the first leak was detected and so the projection based on 10-h is a CCL. Many specifications allow granular material up to 100 mm in
data in Table 2 and rates inferred from Sabir and Brachman (2012) size in a compacted clay liner. Brachman and Sabir (2010) demon-
appear consistent with both the Abdelaal et al. (2014) data after strated that for a CCL compacted at a typical 2–4% wet of optimum, a
24–48 h and Ewais et al. (2014) data after 600 days for the first rupture single 35-mm gravel particle in the upper lift but below the surface of
when SCR = 760 h (Ewais et al., 2014). the CCL (not visible) could induce significant strains in the GMB
At 55 °C, Abdelaal et al. (2014) reported that the GMB ruptured (Table 6). They also showed that even for the CCL compacted at a water
after ⁓ 800 h of sustained loading (at a tensile strain of 8%). There was content of 16% (= wopt + 4%) and a 1000 kPa load there was negli-
one crack that occurred well below the projected maximum strain of gible strain in the GMB due to the gravel particle when a GCL was
13%. At temperatures between 55 and 85 °C, cracks occurred at strains between the CCL and the GMB (Table 6). There are a variety of reasons
as low as 6% (Table 4). Thus it is especially important to (a) project the for including a GCL between the GMB and CCL (Rowe, 2018b); one of
strain for the anticipated temperature and time of exposure, and (b) these is to mitigate the effect of gravel in the CCL on GMB tensile
limit the maximum projected strain from all sources (those above were strains.
only for indentations from the drainage layer) to below the maximum
allowable strain. 3. GMB tensile strains on side slope

2.7. Influence of stress on maximum GMB tensile strain 3.1. GMB tears on side slope

The results presented in Tables 2–4 and discussed above were for an
applied stress of 250 kPa. This is at the lower end of the likely range for
a landfill and well below the stresses in applications like heap leach
pads. Table 5 documents strains after 100 h of sustained loading at a
range of stresses from 500 to 3000 kPa. Compared to the 16% strain
An appropriate protection layer between the GMB and gravel drai-
nage layer can be very effective in minimizing GMB punctures and
limiting the GMB strains as discussed above. However, the GMB can
still fail on side slopes due to inadequate attention to stability and/or
down-drag loads from waste settlement. A well-documented slope

7
R.K. Rowe, Y. Yu Geotextiles and Geomembranes xxx (xxxx) xxx–xxx

Table 5
Maximum GMB strains for a range of cases in full scale (GLLS) tests involving 1.5 mm HDPE GMB at different applied vertical pressures, room temperature with 100 h
of sustained loading.
Type of foundation (thickness) below GMB Gravel D60 (mm) Protection layer Applied pressure (kPa) Maximum strain (%)

a
CCL (w = 12%, 150 mm) 19 None 2000 18–25
a
CCL (w = 16%, 150 mm) 19 None 2000 23–25
a
GCL (w = 7%) + SP (150 mm) 19 None 2000 12–18
a
GCL (w = 85–87%) + SP (150 mm) 19 None 2000 14–21
b
CSL (150 mm) 18 None 500 5.0–6.0
b
CSL (150 mm) 18 None 1000 7.0–8.0
b
CSL (150 mm) 18 None 2000 15–18
b
CSL (150 mm) 18 None 3000 18–27
b
CSL (150 mm) 8 None 2000 11–12
b
CSL (150 mm) 8 None 3000 13–15
b
CSL (150 mm) 2 None 2000 8.0–9.0
b
CSL (150 mm) 2 None 3000 11–12
b
CSL (150 mm) 18 Silty sand (150 mm) 3000 2.0

Note: CCL: compacted clay liner; GCL: geosynthetic clay liner; SP: poorly graded sand layer; CSL: Compacted soil (silty sand) liner with w = 11%.
a
Rowe et al. (2013b).
b
Brachman et al. (2014).

Table 6
Maximum GMB strains reported for a CCL with a 35 mm gravel particle initially
flush with the surface of 150 mm-thick CCL below a 1.5 mm HDPE GMB at
different applied vertical pressures at room temperature (adapted from
Brachman and Sabir, 2010).
Liner CCL compacted with Applied vertical Maximum strain
w (%) pressure (kPa) (%)

GMB/CCLa 14 (wopt + 2%) 600 3.6

GMB/CCLa 14 (wopt + 2%) 1000 6.3
GMB/CCLa 16 (wopt + 4%) 200 5.2
GMB/CCLa 16 (wopt + 4%) 600 10
GMB/CCLa 16 (wopt + 4%) 1000 14
GMB/CCLb 16 (wopt + 4%) 800 12
GMB/CCLb 16 (wopt + 4%) 2000 18
GMB/GCL/ 16 (wopt + 4%) 1000 < 0.1%
CCLa

a
With a 540 g/m2 protection layer and overlying sand protection (as per
Table 2).
b
With a 540 g/m2 protection layer and overlying 50 mm-gravel (as per
Table 2) above GMB.

failure of the waste at the Kettleman Hills Landfill (Mitchell et al.,

1990) reported that tears in a GMB liner on a side slope were associated
with the failure developed by sliding along the interfaces between the
underlying liner system beneath the waste fill. This case resulted in
more attention being addressed to interface strength. While ensuring
stability is critical, measures that ensure stability but do not also con-
sider the implications of the strains developed in the GMB can avert
short-term failure but result in longer-term failure as discussed in this
section. Fig. 6 shows, the field exhumation of a large landfill in South
East Asia (Fowmes, 2007) which had experienced a failure of the GMB Fig. 6. GMB failure at crest of side slope near bench observed after field ex-
at the crest of the side slope near a bench. humation from a large landfill in South East Asia (modified from Fowmes,
2007).
Field observations of GMB failures on side slopes, such as the two
examples above, highlight the importance of protecting the GMBs not
just from the indentations caused by gravel particles but also carefully systems (Villard et al., 1999; Zamara et al., 2014). However, it is gen-
considering the implications of interface friction with respect to both erally not feasible to conduct the field-scale tests because of the prac-
stability and the strains in the GMB induced by down-drag forces acting tical difficulties and associated costs of performing these tests. Thus,
on the GMB due to waste settlement (e.g., Fowmes et al., 2007). It is there is a paucity of field measurements associated with GMB liner
important to ensure both stability and acceptable long-term strains. strains due to waste settlement (a notable exception being Yazdani
et al., 1995). The waste after buried in landfills exhibits both the pri-
mary and secondary settlements (e.g., Dixon and Jones, 2005; Gourc
3.2. Numerical modelling
et al., 2010). The primary settlement is associated with the applied load
on the waste due to its self-weight and subsequent placement of over-
The failures of the GMB liners observed in the field are very valu-
lying waste layers, while the secondary settlement includes two com-
able in terms of recognizing the limitations of the design practice for the
ponents (a) mechanical settlement (e.g., creep and particle physical
geosynthetic liner systems. Field observations improve the under-
distortion, bending, crushing and reorientation under applied load;
standing of failure mechanisms associated with geosynthetic liner

8
R.K. Rowe, Y. Yu
Bareither et al., 2008): and (b) biodegradation-induced settlement (e.g.,
anaerobic biodegradation of organic matter resulting in loss of solid
mass; Gourc et al., 2010). In this investigation, the primary settlement
is considered as the short-term waste settlement (i.e., immediately after
landfill closure; Case 1 in Yu and Rowe, 2018b) and the total settlement
from both the primary and secondary settlements is termed as the long-
term waste settlement (Case 2 in Yu and Rowe, 2018b). Numerical
models are currently the only practical tools for engineers to explore the
different design scenarios and to gain confidence when designing the
geosynthetic liner systems. Both the finite element method (FEM;
Villard et al., 1999; Filz et al., 2001) and finite difference method (FDM;
Jones and Dixon, 2005; Fowmes, 2007; Fowmes et al., 2008a,b; Arab,
2011; Sia and Dixon, 2012; Wu, 2013; Zamara et al., 2014; Thiel et al.,
2014; Yu and Rowe, 2018a, 2018b; Kavazanjian et al., 2018) have been
used to numerically model the performance of geosynthetic liner sys-
tems. All these numerical models have assumed that the slopes had
planar surfaces and that, according to good practice, the GMBs were not
welded across the side slopes. The influence of the slope geometric ir-
regularity (e.g., Mitchell et al., 2016) and geomembrane seams (e.g.,
Giroud, 2005; Fowmes et al., 2008b; Kavazanjian et al., 2017; Rowe
and Shoaib, 2017) on the interaction along interfaces and the geo-
membrane tensile strains has not been considered.
The following sections summarized numerical results and presented
new data based on a numerical model developed by Yu and Rowe
(2018a, 2018b). The interactions between the dissimilar materials were
simulated using a zero-thickness interface model with a Mohr-Coulomb
failure criterion. The strain-softening interface behavior was not con-
sidered in the current version of the model. The numerical results from
Zamara et al. (2014) indicated that both the strain-softening and non-
strain-softening interface models were unable to predict the field per-
formance of a GMB liner with an overlying geotextile. They suggested
the use of reduced values for interface shear strength parameters in
design when the geosynthetic liner systems were expected to be un-
covered for a prolonged time. Thiel (2001) discussed the decision of
using the peak or residual interface strength parameters for the design
of geosynthetic liner systems and recommended the use of residual
interface shear strength parameters on the side slopes. Compared to the
strain-softening interface model, the non-strain-softening model with
residual interface shear strength parameters can generally yield con-
servative (thus safer) performance predictions of geosynthetic liner
systems.
3.3. Centrifuge testing
Centrifuge modelling has been used to examine the performance of
GMB liners under waste settlement (Thusyanthan et al., 2007;
Kavazanjian and Gutierrez, 2017). These centrifuge tests used the
scaled models and increased the body stresses by centrifugal accelera-
tion. A FDM model (Yu and Rowe, 2018a) was used to examine the
GMB strains/loads developed in the large-scale centrifuge test of the
Fig. 7. Prototype geometry based on a large-scale centrifuge test reported by Kavazanjian and Gutierrez (2017).
9
Geotextiles and Geomembranes xxx (xxxx) xxx–xxx
GMB-lined landfill with benches on side slopes similar to those en-
countered in a canyon landfill subject to waste settlement (Fig. 7) as
reported by Kavazanjian and Gutierrez (2017). The results showed that
the calculated GMB strains on benches and waste surface settlement at
the landfill centre were generally in encouraging agreement with the
measured data (Yu and Rowe, 2018a).
The numerical analyses presented in Yu and Rowe (2018a) in-
dicated that the 3-mm thick prototype GMB with an axial tensile stiff-
ness J = 2000 kN/m resulted in a maximum prototype tensile load
equal to the yield strength (i.e., Ty = 120 kN/m). If the GMB had a
higher stiffness (e.g., J = 4000 kN/m) and strength (e.g., Ty = 240 kN/
m), the GMB maximum tensile load was 205 kN/m (i.e., less than the
yield strength 240 kN/m; Fig. 8a) and the maximum tensile strain was
5.1% (< yield strain; Yu and Rowe, 2018a). Thus a GMB with axial
tensile stiffness J = 4000 kN/m could prevent the GMB from yielding
and reduce the maximum strain to about 5% (Fig. 8b) with other things
being equal. However, this would imply the need for an unrealistically
(6 mm) thick GMB liner to control the maximum tensile strain and thus
an alternative approach is needed.
3.4. Influence of GMB upper and lower interface shear strength on GMB
tensile strains
Yu and Rowe (2018b) numerically examined a full-scale landfill
profile with a slope inclination of 1H:1V and two 4-m wide inter-
mediate benches below the ground surface (Fig. 9). The foundation was
competent rock and the GMB was 1.5 mm-thick (Yu and Rowe, 2018b).
For the base case (a slope inclination of 1H:1V) with the interface
friction angle between the LCS (i.e., leachate collection system) and
GMB ϕgmw = 20° (GMB upper interface) and that between the GMB and
GCL ϕgmf = 10° (GMB lower interface), the numerical results (Table 7)
showed that the calculated maximum GMB tensile strain was 8.6% for
the short-term waste settlement (Case 1 from the primary settlement;
using Mohr-Coulomb model with waste unit weight γ = 10.7 kN/m3,
friction angle ϕ = 33°, poisson's ratio ν = 0.33, Young's modulus
E = 6 MPa, and cohesion c = 5 kPa) and increased to 19.8% for the
long-term waste settlement (Case 2 from both the primary and sec-
ondary settlements; using Mohr-Coulomb model by decreasing the
Young's modulus from E = 6 MPa in Case 1 to E = 2 MPa in Case 2
while keeping all other parameter values unchanged). A decrease in the
GMB upper interface friction angle from ϕgmw = 20° to 15° reduced the
calculated maximum GMB tensile strain from 8.6% to 7.5% for Case 1
and from 19.8% to 18.2% for Case 2 when other conditions were un-
changed. The use of the GMB upper interface friction angle ϕgmw = 15°
and GMB lower interface friction angle ϕgmf = 15° (Table 7) resulted in
the calculated maximum GMB tensile strain of 3.4% for the short-term
waste settlement (Case 1) and of 4.9% for the long-term waste settle-
ment (Case 2). A further reduction in the GMB upper interface friction
angle to ϕgmw = 11° (for 1H:1V and ϕgmf = 15°) reduced the calculated
maximum GMB tensile strains to 2.3 and 3.1% for Case 1 and 2,
R.K. Rowe, Y. Yu
Fig. 8. Influence of GMB stiffness on GMB (a) tensile loads and (b) tensile strains. Note: ϕf is the foundation friction angle, c f is the foundation cohesion, ϕgf is the
interface friction angle between the GMB and foundation, ϕgw is the interface friction angle between the GMB and waste, k n is the interface normal stiffness, and k s is
the interface shear stiffness.
respectively. For the slope inclination of 1H:1V, the maximum GMB
strain levels are acceptable if the GMB upper interface friction angle
ϕgmw ≤ 11° and the GMB lower interface friction angle ϕgmf ≥ 15°. Thus
the base case with high friction above and lower friction below the GMB
represents the worst combination of friction angles and a decrease in
the GMB upper interface friction angle or increase in the GMB lower
interface friction angle, or both, can reduce the calculated maximum
GMB tensile strains for both short- and long-term waste settlement.
Lower friction angles above the GMB are good for strains but increase
stability concerns; both need to be addressed.
3.5. Influence of slope inclination on GMB tensile strains
When the GMB upper and lower interface friction angles were kept
the same as those from the base case, changing the slope inclination
from 1H:1V to 2H:1V decreased the maximum GMB tensile strain from
10
Geotextiles and Geomembranes xxx (xxxx) xxx–xxx
8.6 to 4.4% for Case 1 and from 19.8 to 10.7% for Case 2 (Yu and Rowe,
2018b). A further reduction in slope inclination to 3H:1V resulted in the
maximum tensile strains of 2.0 and 2.1% for Case 1 and Case 2, re-
spectively (Yu and Rowe, 2018b). Thus reducing the slope inclination
had a very positive effect in terms of reducing maximum GMB tensile
strains for both short-term and long-term waste settlement, and the use
of a slope inclination of 3H:1V limited the maximum GMB tensile
strains to acceptable design levels for the (challenging friction angle
combination) conditions examined.
For slope inclination of 2H:1V, reducing the GMB upper interface
friction angle from ϕgmw = 20° to 15° and increasing GMB lower in-
terface friction angle from ϕgmf = 10° to 15° (Table 7) resulted in a
decrease of the calculated maximum GMB tensile strain from 4.4 to
1.6% for the short-term waste settlement (Case 1) and from 10.7 to
3.1% for the long-term waste settlement (Case 2). Thus, a slope in-
clination of 2H:1V with the GMB upper interface friction angle ϕgmw ≤
R.K. Rowe, Y. Yu Geotextiles and Geomembranes xxx (xxxx) xxx–xxx

Fig. 9. Landfill profile with a slope inclination of 1H:1V and two 4-m wide intermediate benches below the ground surface (modified from Yu and Rowe, 2018b).

Table 7 decreasing the Young's modulus from E = 6 MPa to E = 2 MPa (while

Calculated maximum GMB tensile strains without a geotextile protection layer keeping other conditions unchanged) increased the maximum GMB
on rock foundation at different slope inclinations for both short-term (Case 1) tensile strain from 3.6% (Fig. 10a) to 8.5% (Fig. 10b). Thus, the waste
and long-term (Case 2) waste settlement (based on Yu and Rowe, 2018b but with mechanical properties different from those in this investigation
with additional cases added). can result in different maximum GMB tensile strains for both short- and
Slope LCS-GMB interface GMB-GCL interface Maximum GMB strain long-term waste settlements.
inclination friction angle, ϕgmw friction angle, ϕgmf (%)
( o) (o)
Case 1 Case 2 3.7. Influence of foundation stiffness on GMB tensile strains

1H:1V (base 20 10b 8.6 19.8

The foundation modelled by Yu and Rowe (2018b; Table 7 above)
case)
1H:1V 15a 10b 7.5 18.2 was competent rock with a very high Young's modulus. When a firm
1H:1V 15a 15b 3.4 4.9 clay foundation was assumed for the 3H:1V slope (modelled using the
1H:1V 11a 15b 2.3 3.1 Mohr-Coulomb model with Young's modulus Ef = 10 MPa, friction
1H:1V (base 20 10b 8.6 19.8
angle ϕf = 26° and cohesion cf = 1 kPa), the maximum GMB strain was
case)
2H:1V 20 10b 4.4 10.7
reduced to the 1.1% (Fig. 11a) compared to 2.0% (Table 7) with a rock
2H:1V 15a 15b 1.6 3.1 foundation for Case 1 (immediately after landfill closure). For Case 2
3H:1V 20 10b 2.0 2.1 (after long-term waste settlement), the maximum GMB strain was 1.5%
(Fig. 11b) for a firm clay foundation (compared to 2.1% for a rock
Note: Results for GMB axial tensile stiffness Jgm = 726 kN/m, the interface foundation; Table 7). Thus the maximum GMB strains from the rock
friction angle between the LCS and GMB ϕgmw and interface friction angle be-
foundation were larger (thus conservative) compared to those calcu-
tween the GMB and GCL ϕgmf as indicated in the table.
a
lated for a firm clay foundation (other things being equal).
Based on residual friction angles ranging from 11° to 16° between the
nonwoven GTX and textured GMB reported by Stark et al. (1996).
b
Based on residual friction angles ranging from 9.3° to 18.5° between the 3.8. Influence of geotextile reinforcement layer on GMB tensile strains
textured GMB and needle-punched GCL with nonwoven GTX up reported by
Triplett and Fox (2001). To reduce the GMB tensile strains for the slope inclination steeper
than 3H:1V, a high stiffness/strength geotextile reinforcement was used
15° and lower interface friction angle ϕgmf ≥ 15° is likely to meet the over the GMB (Yu and Rowe, 2018b) based on the approach proposed
maximum GMB strain requirements under both short- and long-term by Thiel et al. (2014). The geotextile-LCS interface was assumed to have
waste settlement for conditions similar to those examined here. a friction angle ϕgtw = 30°, the interface friction angle between the
geotextile and smooth upper surface of the underlying GMB ϕgtgm
3.6. Influence of GMB thickness on GMB tensile strains = 10°, and the interface friction angle between the GMB and the un-
derlying GCL ϕgmf was a variable (Table 8).
The 1.5 mm-thick GMB considered by Yu and Rowe (2018b) had a As shown in Table 8, for the assumed combination of friction angles
plane strain axial tensile stiffness Jgm = 726 kN/m. If the GMB tensile at the various interfaces, the geotextile reinforcement layer reduced the
stiffness was increased to Jgm = 1210 kN/m (by increasing the GMB maximum GMB strain to less than 2.0% for both Case 1 (immediately
thickness to 2.5 mm), then for a slope inclination of 2H:1V (Fig. 10) the after landfill closure) and Case 2 (after long-term waste settlement) for
maximum GMB tensile strain was 3.6% (versus 4.4% for Jgm = 726 kN/ all cases with GMB-GCL interface ϕgmf ≥ 10°. However, the geotextile
m; Table 7) for Case 1 (immediately after landfill closure) and 8.5% itself experienced large strains for the slope inclination of 1H:1V. For
(versus 10.7% for Jgm = 726 kN/m; Table 7) for Case 2 (after long-term example, the maximum geotextile strains were 5.0% immediately after
waste settlement). Thus, while the maximum GMB strains were sub- landfill closure (Case 1) and 9.1% after long-term waste settlement
stantially reduced for the thicker GMB, they are still too large for the (Case 2) when the geotextile had an axial tensile stiffness of
slope inclination of 2H:1V with the 2.5-mm thick GMB (other condi- Jgt = 4200 kN/m and ϕgmf ≥ 10°. Increasing the geotextile tensile
tions unchanged). It should be noted that the calculated GMB strains stiffness from Jgt = 4200 kN/m to Jgt = 8000 kN/m reduced the
under waste settlement reported in this investigation are dependent on maximum geotextile strains from 5.0 to 3.7% for Case 1 and from 9.1 to
the selected values for the mechanical properties of the waste (Yu and 6.1% for Case 2 (Table 8). These geotextile tensile strains were too large
Rowe, 2018b). For example, when using the Mohr-Coulomb model, to be acceptable for landfill applications. When using a geotextile with

11
R.K. Rowe, Y. Yu
Fig. 10. Influence of GMB thickness (2.5 mm) on calculated GMB tensile strains on rock foundation from landfill profile with a slope inclination of 2H:1V and waste
parameter values from: (a) Case 1 and (b) Case 2.
Jgt = 10000 kN/m (Mirafi® PET high strength geotextile, the highest
stiffness value currently on market), the maximum geotextile tensile
strains were 3.3 and 5.3% for Case 1 and Case 2 (Fig. 12; Table 8),
respectively. Thus just increasing the geotextile tensile stiffness may not
be sufficient to reduce the geotextile tensile strains to acceptable strain
level when the slope was 1H:1V.
A further consideration was to decrease the slope inclination from
1H:1V to 2H:1V and to use a geotextile with Jgt = 8000 kN/m, which
resulted in the maximum geotextile tensile strains of 2.5% for Case 1
(immediately after landfill closure) and 3.7% Case 2 (after long-term
waste settlement), respectively (Table 8). These geotextile strains are
likely acceptable even after long-term waste settlement.
The numerical results suggest that the use of a geotextile (with
12
Geotextiles and Geomembranes xxx (xxxx) xxx–xxx
overlaps to accommodate differential slippage and to ensure a con-
tinuous protection to the underlying GMB) with a low interface shear
strength between the geotextile and GMB can limit the GMB strains to
less than 2.0% for both short-term and long-term waste settlement. If
the slope inclination, geotextile tensile stiffness, and interface friction
angles are all properly selected, the geotextile strains can also be lim-
ited to acceptable level (e.g., less than 3.0% immediately after landfill
closure and 4.0% after long-term waste settlement in this case).
3.9. Influence of GMB lower interface shear strength on GMB and geotextile
tensile strains
When the GMB was overlain by geotextile reinforcement, it was
R.K. Rowe, Y. Yu
Fig. 11. Influence of foundation stiffness on calculated GMB tensile strains from landfill profile with a slope inclination of 3H:1V and waste parameter values from:
(a) Case 1 and (b) Case 2.
conservative to assume that the GMB was smooth with the same upper
and lower interface shear strength ϕgmf = ϕgtgm = 10° (Yu and Rowe,
2018b, Table 8; Fig. 12) for a single composite design. However, if the
GMB had a smooth upper surface, with ϕgtgm = 10°, and a textured
lower surface with ϕgmf = 20°, then the calculated maximum GMB
tensile strains for a geotextile with Jgt = 4200 kN/m (Fig. 13) were the
same as those obtained with ϕgmf = 10° in (Table 5), namely, 1.1% for
Case 1 (immediately after landfill closure) and 1.9% for Case 2 (after
long-term waste settlement). The maximum geotextile tensile strains
(Fig. 13) were also the same as those obtained with ϕgmf = 10° at 5.0%
for Case 1 and 9.1% for Case 2 (Table 5). Thus, increasing the GMB
lower interface shear strength from ϕgmf = 10° to 20° such that it ex-
ceeded that at the upper surface of the GMB (ϕgtgm =10°) had negligible
13
Geotextiles and Geomembranes xxx (xxxx) xxx–xxx
influence on both the GMB and geotextile strains (other things being
equal).
In contrast to the negligible effect of increasing the GMB lower in-
terface shear strength to exceed that at the upper surface of the GMB
(i.e., ϕgmf ≥ ϕgtgm = 10°) decreasing the GMB lower interface shear
strength below that at the upper surface of the GMB (i.e.,
ϕgmf < ϕgtgm = 10°) had a significant effect. For example, decreasing
ϕgmf from 10° to 5° increased the calculated maximum GMB tensile
strain from 1.1% to 3.6% (Fig. 14a; Table 5) for Case 1 (immediately
after landfill closure) and from 1.9% (Table 8) to 8.7% (Fig. 14b;
Table 8) for Case 2 (after long-term waste settlement). Also increased
were the calculated geotextile tensile strains from 5.0% to 6.4%
(Fig. 14a; Table 8) for Case 1, and from 9.1% to 12.9% (Fig. 14b;
R.K. Rowe, Y. Yu Geotextiles and Geomembranes xxx (xxxx) xxx–xxx

Table 8 Tognon et al. (2000) method which is the most appropriate method
Calculated maximum GMB and geotextile tensile strains on rock foundation at available based on vertical deformations and considers both mem-
different slope inclinations and geotextile tensile stiffness values for both short- brane and bending strains based on sound mechanics. An alternative
term (Case 1) and long-term (Case 2) waste settlement (based on Yu and Rowe, method that also considers radial deformation presently under de-
2018b).
velopment (Eldesouky and Brachman, 2018) shows promise.
Slope
inclination
GTX axial
stiffness, Jgt
GMB-GCL
Interface
Maximum GMB
strain (%)
Maximum GTX
strain (%)
• Without a protection layer between the GMB and overlying gravel,
GMBs over a compacted clay liner may experience short-term tensile
(kN/m) friction angle,
strains of 16% for the 25 mm gravel and 32% for the 50 mm gravel
ϕgmf (o) Case 1 Case 2 Case 1 Case 2
under the short-term physical loading condition. These GMB strains
1H:1V 4200 5 3.6 8.7 6.4 12.9 are too large to be acceptable for landfill applications. Thus, a sui-
1H:1V 4200 20 1.1 1.9 5.0 9.1 table protection layer must be included between the GMB and
1H:1V 4200 10a 1.1 1.9 5.0 9.1 overlying gravel layer to limit the GMB tensile strains.
1H:1V
1H:1V
8000
10000
10a
10a
1.0
1.0
1.6
1.6
3.7
3.3
6.1
5.3 • Without protection and with 50 mm gravel the tensile strains were
2H:1V 8000 10a 1.0 1.5 2.5 3.7
almost twice higher (32%) for a GMB over compacted clay com-
pared to those for a GMB over a hydrated (water content 128%)
Note: Assumed GMB axial tensile stiffness Jgm = 726 kN/m, the interface fric- geosynthetic clay liner (GCL) resting on a firm foundation (17%).
tion angle between the LCS and GTX ϕgtw = 30°, the interface friction angle • For GMBs over a needle-punched GCL with a geotextile protection
between the GTX and GMB ϕgtgm = 10° (based on residual friction angles ran- layer, none of the geotextiles with the mass per unit area up to
ging from 9.8° to 12.0° between the smooth GMB and nonwoven GTX reported 2200 g/m2 were able to reduce the short-term GMB strains to ac-
by Akpinar and Benson, 2005), and the interface friction angle between the ceptable level for 50 mm gravel. The 2200 g/m2 geotextile protec-
GMB and GCL ϕgmf as indicated in the table. tion layer between the GMB and 50 mm gravel particles reduced the
a
Based on residual friction angles ranging from 8.0° to 11.1° between the short-term GMB tensile strains to just below 6%.
smooth GMB and needle-punched (or stitch-bonded) GCL with woven GTX up
reported by McCartney et al. (2009).
• A 150-mm-thick sand protection layer limited the maximum GMB
tensile strain to less than 0.2%.

Table 8) for Case 2. This highlights the importance of having a GMB

• A mutilayered geotextile (1100 g/m2) with needle punched non-
woven core between two thinner and stiffer outer layers which en-
lower interface shear strength equal to or greater than the GMB upper hanced tensile properties was effective in reducing the short-term
interface shear strength to reduce both the GMB and geotextile tensile GMB strains generated by 50 mm gravel to 3.9% and even to 2.6%
strains. when increasing the mass per unit area to 3000 g/m2.

3.10. Influence of partial reinforcement on GMB and geotextile tensile

strains
In the previous two subsections, the GMB on side slopes was fully
covered by a high strength/stiffness geotextile reinforcement. The nu-
merical results (Yu and Rowe, 2018b) showed that the geotextile tensile
strains peaked at the crest of the slope near each bench and gradually
decreased down the slope. Thus an alternative design with partial re-
inforcement over the GMB considered in this section was to have a high
strength/stiffness geotextile reinforcement (Jgt = 8000 kN/m) on the
bench and only over one third of the way down the slope. A geotextile
protection layer (Jgt = 200 kN/m) underlying the LCS was used below
the geotextile reinforcement and between the LCS and the GMB over
the remaining two thirds of the slope. The numerical analysis indicated
that the use of the partial reinforcement with an axial tensile stiffness
value of Jgt = 8000 kN/m for the slope inclination of 2H:1V (Fig. 15)
resulted in the calculated maximum GMB tensile strains of 1.8% for
both Cases 1 and 2. The calculated maximum high-stiffness geotextile
(Jgt = 8000 kN/m) tensile strains were 2.4% for Case 1 and 3.6% for
Case 2. However, for the geotextile protection layer with Jgt = 200 kN/
m the calculated maximum tensile strain was 2.2% for Case 1 and in-
creased to 6.3% for Case 2. Thus when using partial geotextile re-
inforcement over GMB, both GMB and geotextile reinforcement ex-
perienced maximum strains that are within the acceptable levels, while
the geotextile protection layer attracts a larger strain that will need
consideration.
• A geonet was unable to limit the GMB strains due to 50 mm gravel to
acceptable levels and GMB strains of 13–15% were observed.
• The 150-mm thick layer of tire shreds limited the maximum GMB
strains due to 50 mm gravel to 5.2–6.3%, and even lower to
2.3–2.8% when used with a 570 g/m2 single layer of geotextile be-
tween the tire shreds and the GMB.
• The measured GMB tensile strains are dependent on the time period
of loading and the temperature of laboratory tests. Relative to the
10-h short-term strain, the strain in the GMB increased at a rate of
about 1% per log cycle of time (in hours) when the GMB was at 22
and 55 °C and about 1.9% per log cycle of time (in hours) when the
GMB was at 85 °C where the drainage layer had 50-mm gravel and
the geotextile protection was about 500–600 g/m2. While these rates
are not generalizable to all conditions, they show the importance to
project the strain for the anticipated temperature and time of ex-
posure, especially when the GMB is in a composite liner.
• Abdelaal et al. (2014) and Ewais et al. (2014) reported cases where
stress cracking occurred at strains as low at 6% and well below the
maximum strains mobilized by 50 mm gravel separated from the
GMB by a 560 g/m2 nonwoven geotextile protection layer. Thus it is
especially important to limit the maximum projected strain from all
sources (those above were only for indentations from the drainage
layer) to below the maximum allowable strain.
The foregoing conclusions were for a gravel drainage layer and an
average applied vertical stress of 250 kPa.

4. Summary
• For a CCL compacted at a typical 2–4% wet of optimum, a single
35 mm gravel particle in the upper lift but below the surface of the
CCL (not visible) can induce significant strains in the GMB. It was
The findings associated with local GMB indentations induced by the also shown that the effect of an isolated stone in the CCL was largely
gravel used in a modern leachate collection system under a 250 kPa mitigated when a GCL was placed between the CCL and the GMB.
vertical pressure are summarized below:
• The paper has also shown that for poorly graded material, the
grainsize (D60) of the drainage medium is of critical importance with
• The method of assessing strains is critical and the inferred strains respect to the strains induced at a given stress level and that for any
can vary by orders of magnitude depending on the approach grainsize the strains increase with increasing applied stress.
adopted. The strains reported in this paper were obtained using the

14
R.K. Rowe, Y. Yu
Fig. 12. Influence of geotextile reinforcement layer (Jgt = 10000 kN/m) on calculated GMB on rock foundation from landfill profile with a slope inclination of 1H:1V
and waste parameter values from: (a) Case 1 and (b) Case 2.
For the cases and conditions examined, the key findings associated
with the down-drag load for GMBs on side slopes are:
• For both short- and long-term waste settlement, decreasing the GMB
upper interface friction angle or increasing the GMB lower interface
friction angle or both reduced the calculated maximum GMB tensile
strains.
• Decreasing the slope inclination from 1H:1V to 3H:1V reduced the
maximum GMB tensile strains for both short-term (e.g., immediately
after landfill closure) and long-term waste settlement. For GMBs on
side slopes without a geotextile protection layer, slope inclinations
steeper than 3H:1V resulted in unacceptable maximum GMB strains
for landfill applications unless measures (e.g. geotextile reinforce-
ment or careful control of friction angles above and below the GMB)
15
Geotextiles and Geomembranes xxx (xxxx) xxx–xxx
were taken to mitigate the effect.
• A geotextile reinforcement layer (with overlaps to accommodate
differential slippage and to ensure a continuous protection) reduced
the maximum GMB tensile strains to less than 2.0% even on slopes
of 1H:1V. However, the geotextile itself could become an en-
gineering concern for the steep slopes and consideration must be
given to the allowable strain in the reinforcement; very high
strength reinforcement may be needed on steep slopes.
• It was shown to be extremely important to have a GMB lower in-
terface shear strength equal to or higher than the GMB upper in-
terface shear strength to limit both the GMB and geotextile re-
inforcement tensile strains for landfill applications.
• The partial geotextile reinforcement could have the same benefit of
the full geotextile reinforcement in reducing the maximum GMB and
R.K. Rowe, Y. Yu Geotextiles and Geomembranes xxx (xxxx) xxx–xxx

Fig. 13. Influence of GMB lower interface shear strength (ϕgmf = 20°) on calculated GMB (Jgm = 726 kN/m) and geotextile (Jgt = 4200 kN/m) tensile strains on rock
foundation from landfill profile with a slope inclination of 1H:1V and waste parameter values from: (a) Case 1 and (b) Case 2.

geotextile reinforcement tensile strains to acceptable low levels, but
the geotextile protection layer may become an engineering concern.
For cases with gravel drainage layer on side slopes, the increase in
maximum GMB tensile strain can be expected as the gravel particles
under the weight of waste can cause indentations of the underlying
GMB resulting in additional GMB tensile strains. As the GMB tensile
strain increases, the susceptibility of the GMB to stress cracking also
increases.
5. Conclusions
To ensure a long service-life of a high-density polyethylene geo-
membrane (GMB) exposed to leachate in a landfill it is necessary to
limit the tensile strains/stresses in the GMB to an acceptable low level.
The potential sources of strain have been considered herein, namely, (i)
strains due to local GMB indentations induced by overlying coarse
gravel in a leachate collection system, (ii) strain induced by an isolated
gravel particle in a CCL, and (iii) the down-drag load due to waste
settlement for GMBs on side slopes. The key research related to limiting
both sources of GMB strains have been discussed.
Some have argued that one cannot get tensile stress in a GMB liner
under compressive stress while others have argued that stress relaxation
will prevent stress cracking due to tensile strain. The results summar-
ized in this paper for 250 kPa vertical applied stress show the error in
that argument, which fails to recognize the effects of non-uniformity in
the distribution of the vertical compressive stresses at the level of the
GMB. Results summarized herein indicate that increasing compressive

16
R.K. Rowe, Y. Yu Geotextiles and Geomembranes xxx (xxxx) xxx–xxx

Fig. 14. Influence of GMB lower interface shear strength (ϕgmf = 5°) on calculated GMB (Jgm = 726 kN/m) and geotextile (Jgt = 4200 kN/m) tensile strains on rock
foundation from landfill profile with a slope inclination of 1H:1V and waste parameter values from: (a) Case 1 and (b) Case 2.

stress increases the tensile strains for an otherwise similar situation.
They also show that stress cracking can occur at strains as low as 6%
(and possibly lower). Thus, in design it is important to (a) project the
strain for the anticipated temperature and time of exposure, and (b)
limit the maximum projected strain from all sources (those above were
only for indentations from the drainage layer) to below the maximum
allowable strain.
Although not addressed in detail, the implications of the results
summarized herein are that the level of protection needed between a
gravel drainage layer will depend on: (i) the type and characteristics
nature of the clay liner below the GMB, (ii) the nature of the drainage
layer above the GMB and especially its particle size, (iii) GMB thickness,
(iv) applied stress, (v) expected liner temperature, (vi) design life and
maximum allowable strain.
It has been confirmed that significant tensile strains can also be
developed in side slopes (especially those steeper than 3H:1V) based on
a numerical modelling technique by using the residual interface shear
strength and without modelling the strain-softening interface behavior.
Currently the only practical means to compare the different design
scenarios of geosynthetic liner systems is to use a numerical model that
is capable of modelling the interaction and potential slippage between
dissimilar materials in a multilayered system. Consideration needs to be
given to the effect of down-drag on GMB strains due to both short- and
long-term waste settlement. In addition to slope angle, the friction
angle between the GMB and the materials above and below the GMB
had a significant effect on the GMB strains generated. Decreasing the
GMB upper interface friction angle or increasing the GMB lower in-
terface friction angle or both reduced the calculated maximum GMB

17
R.K. Rowe, Y. Yu
Fig. 15. Influence of partial reinforcement on calculated GMB (Jgm = 726 kN/m) and geotextile (Jgt1 = 8000 kN/m on bench and over one third of the way down the
slope and Jgt2 = 200 kN/m over the rest two thirds of the slope) tensile strains on rock foundation from landfill profile with a slope inclination of 2H:1V and waste
parameter values from: (a) Case 1 and (b) Case 2.
tensile strains. While addressing the development of down-drag strains
this may induce stability problems. Installing geotextile reinforcement
above the GMB is one means of addressing both strain and stability.
However, for steep slopes, high strength reinforcement may be required
to sustain the loads induced.
The effects of sustained long-term strains from any of the three
sources examined are unlikely to be manifest for considerable time after
landfill is closed; however that does not negate their significance and
failure to address this in design may mean we are creating avoidable
significant problems for future generations.
18
Geotextiles and Geomembranes xxx (xxxx) xxx–xxx
Acknowledgements
The work reported in this paper was supported by a grant (A1007)
from the Natural Sciences and Engineering Research Council of Canada
(NSERC).
References
Abdelaal, F.B., Rowe, R.K., Brachman, R.W.I., 2014. Brittle rupture of an aged HDPE
geomembrane at local gravel indentations under simulated field conditions.
Geosynth. Int. 21 (1), 1–23.
Akpinar, M.V., Benson, C.H., 2005. Effect of temperature on shear strength of two geo-
membrane-geotextile interfaces. Geotext. Geomembranes 23 (5), 443–453.
Arab, M.G., 2011. The Integrity of Geosynthetic Elements of Waste Containment Barrier
Systems Subject to Seismic Loading. Ph.D. Thesis. School of Sustainable Engineering
R.K. Rowe, Y. Yu
and the Built Environment, Arizona State University, Tempe, Arizona, USA.
ASTM D5514, 2014. Standard Test Method for Large Scale Hydrostatic Puncture Testing
of Geosynthetics. ASTM International, West Conshohocken, PA, USA.
Bareither, C.A., Breitmeyer, R.J., Erses, A.S., Benson, C.H., Edil, T.B., Barlaz, M.A., 2008.
Relative contributions of moisture and biological activity on compression of muni-
cipal solid waste in bioreactor landfills. In: Proceedings of Global Waste Management
Symposium, Copper Mountain Conference Center, Colorado, USA, September 7-10,
2008, pp. 9.
Bundesanstalt für Materialforschung und -prüfung (BAM), 2015. Guidelines for the
Certification of Protection Layers for Geomembranes in Landfill Sealing Systems.
BAM, Berlin, Germany.
Brachman, R.W.I., Gudina, S., 2008a. Gravel contacts and geomembrane strains for a GM/
CCL composite liner. Geotext. Geomembranes 26 (6), 448–459.
Brachman, R.W.I., Gudina, S., 2008b. Geomembrane strains and wrinkle deformations in
a GM/GCL composite liner. Geotext. Geomembranes 26 (6), 488–497.
Brachman, R.W.I., Sabir, A., 2010. Geomembrane puncture and strains from stones in an
underlying clay layer. Geotext. Geomembranes 28 (4), 335–343.
Brachman, R.W.I., Sabir, A., 2013. Long-term assessment of a layered-geotextile protec-
tion layer for geomembranes. J. Geotech. Geoenviron. Eng. 139 (5), 752–764.
Brachman, R.W.I., Rowe, R.K., Irfan, H., 2014. Short-term local tensile strains in HDPE
heap leach geomembranes from coarse overliner materials. J. Geotech. Geoenviron.
Eng. 140 (5), 4014011.
Brummermann, K., Bluemel, W., Stoehwahse, C., 1994. Protection layers for geomem-
branes: effectiveness and testing procedures. In: Proceedings of the Fifth International
Conference on Geotextiles, Geomembranes and Related Products, vol. 3. pp.
1003–1006 (Singapore).
Dickinson, S., Brachman, R.W.I., 2008. Assessment of alternative protection layers for a
GM/GCL composite liner. Can. Geotech. J. 45 (11), 1594–1610.
Dixon, N., Jones, D.R.V., 2005. Engineering properties of municipal solid waste. Geotext.
Geomembranes 23 (3), 205–233.
Eldesouky, H.M.G., Brachman, R.W.I., 2018. Calculating local geomembrane strains from
a single gravel particle with thin plate theory. Geotext. Geomembranes 46 (1),
101–110.
EN 13719, 2016. Geosynthetics - Determination of the Long Term Protection Efficiency of
Geosynthetics in Contact with Geosynthetic Barriers. European committee for stan-
dardization (CEN), Brussels.
Ewais, A.M.R., Rowe, R.K., Brachman, R.W.I., Arnepalli, D.N., 2014. Service-life of a
HDPE GMB under simulated landfill conditions at 85 °C. J. Geotech. Geoenviron. Eng.
140 (11), 04014060.
Filz, G.M., Esterhuizen, J.J.B., Duncan, J.M., 2001. Progressive failure of lined waste
impoundments. J. Geotech. Geoenviron. Eng. 127 (10), 841–848.
Fowmes, G.J., 2007. Analysis of Steep Sided Landfill Lining Systems. Ph.D. Thesis.
Department of Civil and Building Engineering, Loughborough University,
Loughborough, UK.
Fowmes, G.J., Dixon, N., Jones, D.R.V., 2007. Landfill stability and integrity: the UK
design approach. Proc. Inst. Civ. Eng. Waste Resour. Manag. 160 (2), 51–61.
Fowmes, G.J., Dixon, N., Jones, D.R.V., 2008a. Validation of a numerical modelling
technique for multilayered geosynthetic landfill lining systems. Geotext.
Geomembranes 26 (2), 109–121.
Fowmes, G.J., Dixon, N., Jones, D.R.V., 2008b. The influence of horizontal welded seams
of geomembrane tension in a steep sided landfill lining system. In: Proceedings of the
4th European Geosynthetics Congress. International Geosynthetics Society, Jupiter,
FL, USA.
Gallagher, E.M., Darbyshire, W., Warwick, R.G., 1999. Performance testing of landfill
geoprotectors: background, critique, development and current UK practice.
Geosynth. Int. 6 (4), 283–301.
Giroud, J.P., 2005. Quantification of geosynthetic behavior. Geosynth. Int. 12 (1), 2–27.
Giroud, J.P., Tisseau, B., Soderman, K.L., Beech, J.F., 1995. Analysis of strain con-
centration next to geomembrane seams. Geosynth. Int. 2 (6), 1049–1097.
Gourc, J.P., Staub, M.J., Conte, M., 2010. Decoupling MSW settlement into mechanical
and biochemical processes – modelling and validation on large-scale setups. Waste
Manag. 30 (8–9), 1556–1568.
Gudina, S., Brachman, R.W.I., 2006. Physical response of geomembrane wrinkles over-
lying compacted clay. J. Geotech. Geoenviron. Eng. 132 (10), 1346–1353.
Hornsey, W.P., Wishaw, D.M., 2012. Development of a methodology for the evaluation of
geomembrane strain and relative performance of cushion geotextiles. Geotext.
Geomembranes 35, 87–99.
Hsuan, Y.G., Koerner, R.M., 1998. Antioxidant depletion lifetime in high density poly-
ethylene geomembranes. J. Geotech. Geoenviron. Eng. 124 (6), 532–541.
Jones, D.R.V., Dixon, N., 2005. Landfill lining stability and integrity: the role of waste
settlement. Geotext. Geomembranes 23 (1), 27–53.
Jones, D.R.V., Shercliff, D.A., Dixon, N., 2000. Difficulties associated with the specifica-
tion of protection geotextiles using only unit weight. In: Cancelli, A., Cazzuffi, D.,
Soccodato, C. (Eds.), Proceedings of 2nd European Geosynthetics Conference, pp.
581–586 (Bologna, Italy).
Kavazanjian, E., Andresen, J., Gutierrez, A., 2017. Experimental evaluation of HDPE
geomembrane seam strain concentrations. Geosynth. Int. 24 (4), 333–342.
Kavazanjian, E., Gutierrez, A., 2017. Large scale centrifuge test of a geomembrane-lined
landfill subject to waste settlement and seismic loading. Waste Manag. 68, 252–262.
Kavazanjian, E., Wu, X., Arab, M., Matasovic, N., 2018. Development of a numerical
model for performance-based design of geosynthetic liner systems. Geotext.
Geomembranes 46 (2), 166–182.
Koerner, R.M., Hsuan, Y.G., Koerner, G.R., Gryger, D., 2010. Ten year creep puncture
study of HDPE geomembranes protected by needle-punched nonwoven geotextiles.
Geotext. Geomembranes 28 (6), 503–513.
Koerner, R.M., Wilson-Fahmy, R.R., Narejo, D., 1996. Puncture protection of
19
Geotextiles and Geomembranes xxx (xxxx) xxx–xxx
geomembranes, part III: examples. Geosynth. Int. 3 (5), 655–675.
LFE-2, 2014. Cylinder Testing Geomembranes and Their Protective Materials. A
Methodology for Testing Protector Geotextiles for Their Performance in Specific Site
Conditions. Environment Agency, UK.
McCartney, J.S., Zornberg, J.G., Swan, R.H., 2009. Analysis of a large database of GCL-
geomembrane interface shear strength results. J. Geotech. Geoenviron. Eng. 135 (2),
209–223.
McIsaac, R., Rowe, R.K., 2005. Change in leachate chemistry and porosity as leachate
permeates through tire shreds and gravel. Can. Geotech. J. 42 (4), 1173–1188.
Mitchell, J.K., Seed, R.B., Seed, H.B., 1990. Kettleman Hills waste landfill slope failure, I:
liner system properties. J. Geotech. Eng. 116 (4), 647–668.
Mitchell, T., Fowmes, G.J., Wackrow, R., Dixon, N., Scotland, I., 2016. Assessing the
planarity of containment liner systems using a terrestrial laser scanner. In:
Proceedings of the 6th European Geosynthetics Congress. International Geosynthetics
Society, Jupiter, FL, USA.
Müller, W., Jacob, I., 2003. Oxidative resistance of high-density polyethylene geomem-
branes. Polym. Degrad. Stabil. 79 (1), 161–172.
Narejo, D., Koerner, R.M., Wilson-Fahmy, R.F., 1996. Puncture protection of geomem-
branes, part II: experimental. Geosynth. Int. 3 (5), 629–653.
Peggs, I.D., Schmucker, B., Carey, P., 2005. Assessment of maximum allowable strains in
polyethylene and polypropylene geomembranes. In: Proceedings of the Geo-Frontiers
2005. American Society of Civil Engineers, Reston, VA.
Rowe, R.K., 2005. Long-term performance of contaminant barrier systems. The 45th
rankine lecture. Geotechnique 55 (9), 631–678.
Rowe, R.K., 2012. Short and long-term leakage through composite liners. The 7th arthur
casagrande lecture. Can. Geotech. J. 49 (2), 141–169.
Rowe, R.K., 2018a. Protecting the Environment with Geosynthetics. 53rd Karl Terzaghi
Lecture, Orlando, USA March 2017.
Rowe, R.K., 2018b. Environmental geotechnics: looking back, looking forward, 16th
Croce Lecture. Ital. Geotech. J (in review).
Rowe, R.K., McIsaac, R., 2005. Clogging of tire shreds and gravel permeated with landfill
leachate. J. Geotech. Geoenviron. Eng. 131 (6), 682–693.
Rowe, R.K., Sangam, H.P., 2002. Durability of HDPE geomembranes. Geotext.
Geomembranes 20 (2), 77–95.
Rowe, R.K., Shoaib, M., 2017. Long-term performance HDPE geomembrane seams in
MSW leachate. Can. Geotech. J. 54 (12), 1623–1636.
Rowe, R.K., Quigley, R.M., Brachman, R.W.I., Booker, J.R., 2004. Barrier Systems for
Waste Disposal Facilities. Taylor & Francis Books Ltd. (E & FN Spon, London.
Rowe, R.K., Islam, M.Z., Hsuan, Y.G., 2008. Leachate chemical composition effects on OIT
depletion in an HDPE geomembrane. Geosynth. Int. 15 (2), 136–151.
Rowe, R.K., Rimal, S., Sangam, H., 2009. Ageing of HDPE geomembrane exposed to air,
water and leachate at different temperatures. Geotext. Geomembranes 27 (2),
137–151.
Rowe, R.K., Islam, M.Z., Hsuan, Y.G., 2010. Effects of thickness on the aging of HDPE
geomembranes. J. Geotech. Geoenviron. Eng. 136 (2), 299–309.
Rowe, R.K., Abdelaal, F.B., Brachman, R.W.I., 2013a. Antioxidant depletion from an
HDPE geomembrane with a sand protection layer. Geosynth. Int. 20 (2), 73–89.
Rowe, R.K., Brachman, R.W.I., Irfan, H., Smith, M.E., Thiel, R., 2013b. Effect of under-
liner on geomembrane strains in heap leach applications. Geotext. Geomembranes
40, 37–47.
Rowe, R.K., Priyanto, D., Poonan, R., 2019. Factors affecting the design life of HDPE
geomembranes in an LLW disposal facility. In: WM2019 Conference, Phoenix,
Arizona, USA, 15p.
Sabir, A., Brachman, R.W.I., 2012. Time and temperature effects on geomembrane strain
from a gravel particle subjected to sustained vertical force. Can. Geotech. J. 49 (3),
249–263.
Sangam, H.P., Rowe, R.K., 2002. Effects of exposure conditions on the depletion of an-
tioxidants from HDPE geomembranes. Can. Geotech. J. 39 (6), 1221–1230.
Seeger, S., Müller, W., 2003. Theoretical approach to designing protection: selecting a
geomembrane strain criterion. In: Dixon, N., Smith, D.M., Greenwood, J.H., Jones,
D.R.V. (Eds.), Geosynthetics: Protecting the Environment. Thomas Telford, London,
pp. 137–152.
Sia, A.H.I., Dixon, N., 2012. Numerical modelling of landfill lining system waste inter-
action: implications of parameter variability. Geosynth. Int. 19 (5), 393–408.
Stark, T.D., Williamson, T.A., Eid, H.T., 1996. HDPE geomembrane/geotextile interface
shear strength. J. Geotech. Eng. 122 (3), 197–203.
Thiel, R., 2001. Peak vs residual shear strength for bottom liner stability analyses. In:
Proceedings of the GRI-15 Conference, Houston, pp. 40–70.
Thiel, R., Kavazanjian, E., Wu, X., 2014. Design considerations for slip interfaces on steep-
wall liner systems. In: Proceedings of the 10th International Conference on
Geosynthetics, Deutsche Gesellschaft Für Geotechnik and International Geosynthetics
Society German Chapter, P. 6, Berlin, Germany.
Thusyanthan, N.I., Madabhushi, S.P.G., Singh, S., 2007. Tension in geomembranes on
landfill slopes under static and earthquake loading - centrifuge study. Geotext.
Geomembranes 25 (2), 78–95.
Tognon, A.R., Rowe, R.K., Moore, I.D., 2000. Geomembrane strain observed in large-scale
testing of protection layers. J. Geotech. Geoenviron. Eng. 126 (12), 1194–1208.
Triplett, E.J., Fox, P.J., 2001. Shear strength of HDPE geomembrane/geosynthetic clay
liner interface. J. Geotech. Geoenviron. Eng. 127 (6), 543–552.
Villard, P., Gourc, J.P., Feki, N., 1999. Analysis of geosynthetic lining system (GLS) un-
dergoing large deformations. Geotext. Geomembranes 17 (1), 17–32.
Wu, X., 2013. Effect of Waste Settlement and Seismic Loading on the Integrity of
Geosynthetic Barrier Systems. Masters' Thesis. School of Sustainable Engineering and
the Built Environment, Arizona State University, Tempe, Arizona, USA.
Yazdani, R., Campbell, J.L., Koerner, G.R., 1995. Long-term in situ strain measurements
of a high density polyethylene geomembrane in a municipal solid waste landfill. In:
R.K. Rowe, Y. Yu
Proceedings of Geosynthetics ’95. Industrial Fabrics Association International,
Roseville, MN, pp. 893–906.
Yu, Y., Rowe, R.K., 2018a. Modelling deformation and strains induced by waste settle-
ment in a centrifuge test. Can. Geotech. J. 55 (8), 1116–1129.
Yu, Y., Rowe, R.K., 2018b. Development of geomembrane strains in waste containment
facility liners with waste settlement. Geotext. Geomembranes 46 (2), 226–242.
20
Geotextiles and Geomembranes xxx (xxxx) xxx–xxx
Zamara, K.A., Dixon, N., Fowmes, G., Jones, D.R.V., Zhang, B., 2014. Landfill side slope
lining system performance: a comparison of field measurements and numerical
modelling analyses. Geotext. Geomembranes 42 (3), 224–235.
Zanzinger, H., 1999. Efficiency of geosynthetic protection layers for geomembrane liners:
performance in a large-scale model test. Geosynth. Int. 6 (4), 303–317.