ARTICLE IN PRESS

Geotextiles and Geomembranes 23 (2005) 205–233

www.elsevier.com/locate/geotexmem

Engineering properties of municipal solid waste

Neil Dixona,, D. Russell V. Jonesb
a
Department of Civil and Building Engineering, Loughborough University, Leicestershire, LE11 3TU, UK
b
Golder Associates (UK) Ltd, Attenborough House, Browns Lane Business Park, Stanton-on-the-Wolds,
Nottinghamshire, NG12 5BL, UK
Received 3 February 2004; received in revised form 24 November 2004; accepted 26 November 2004

Abstract

Mechanical behaviour of the waste body controls many aspects of landfill lining system
design and performance, including stability issues and integrity of the geosynthetic and
mineral lining components. Knowledge of the likely ranges of waste mechanical properties is
required to assess potential modes of failure and hence to design the landfill engineering
measures. This paper provides a summary of measurement and interpretation issues for the
key engineering parameters used to define: unit weight, compressibility, shear strength, lateral
stiffness, in situ horizontal stress and hydraulic conductivity. The topic of waste mechanics is
developing rapidly and many papers have been published on waste mechanics, reporting
results from both laboratory and in situ studies. Although waste is heterogeneous, many of the
studies show that municipal solid waste has mechanical properties that vary in a consistent
and predictable way (e.g. with respect to stress state and method of placement). An
internationally agreed classification system and test standards are required to allow
interpretation of published results. This will lead to development of appropriate constitutive
models for waste and hence to optimization of landfill designs by considering waste/lining
system interaction in full.
r 2005 Elsevier Ltd. All rights reserved.

Keywords: Landfill engineering; Waste mechanics; Lining systems

Corresponding author. Tel.: +1 44 0 1509 228542; fax: +1 44 0 1509 223945.

E-mail addresses: n.dixon@lboro.ac.uk (N. Dixon), RJones@golder.com (D.R.V. Jones).

0266-1144/$ - see front matter r 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.geotexmem.2004.11.002
 ARTICLE IN PRESS
206 N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233

1. Introduction

1.1. Why is knowledge of waste mechanics important?

Geosynthetics form an integral part of modern landfill lining systems.

Geomembranes and Geosynthetic Clay Liners (GCL) are used as barriers to
leachate and gas, often in composite behaviour with compacted clay mineral layers.
Geotextiles and geocomposite drainage materials are used as protection, separation
and drainage layers. While the use of geosynthetics brings many benefits (e.g. ease of
placement, uniform quality controlled predictable properties) they are also
susceptible to damage leading to loss of function and their use introduces potential
failure planes at interfaces between components (e.g. geosynthetic/geosynthetic and
geosynthetic/soil). These can result in large-scale instability of the lining system and
contained waste body. Behaviour of the waste body is a controlling factor in both the
stability, ultimate limit state, of engineered landfill structures (e.g. large-scale
movements leading to collapse) and integrity, serviceability limit state, of lining
components (e.g. geomembrane and mineral barriers can be punctured/sheared and
geotextile protection layers/geocomposite drains can be damaged and/or become
discontinuous). Fig. 1 summarizes modes of landfill failure in which the waste body
plays a role. Knowledge of engineering properties of waste is required to assess each
mode and hence to design against their occurrence. Design of lining systems,
including geosynthetic components, cannot be carried out without considering the
lining system/waste body interaction.
Failures of landfills, although not common, occur on a regular basis in countries
around the world. They include landfills incorporating geosynthetic materials and
designed using common practices. Koerner and Soong (2000) and Jones and Dixon
(2003) provide information on a range of landfill failures. High profile failures
include: Kettleman Hills, USA (Seed et al., 1988; Byrne, 1994), Bulbul Drive, South
Africa (Brink et al., 1999), Cincinnati, USA (Eid et al., 2000; Stark et al., 2000),
Dona Juana, South American (Hendron et al., 1999) and Payatas, Philippines
(reported in international media, 2000). There are many other failures that do not get
reported. In addition to concerns about large-scale stability failures involving the
waste body, designers must also consider interaction between the lining system and
waste body. To do this, designers are increasingly using numerical modelling
techniques. These require a constitutive model to represent the waste body behaviour
(e.g. stiffness and shear strength). A common use of numerical modelling is to
investigate the potential development of post-peak shear strengths on side slope
lining component interfaces resulting from waste settlement (e.g. Long et al., 1995;
Filz et al., 2001; Jones and Dixon, 2005). These types of analyses aid the selection of
interface shear strength parameters for use in conventional stability assessments and
allow investigation of potential integrity failure mechanisms (e.g. loss of geotextile
protection to a geomembrane).
While it is not possible to fully characterize the engineering properties of waste due
to its heterogeneous nature, it is important that its basic behaviour is understood and
that likely ranges of the key engineering properties are known. Table 1 lists the
 ARTICLE IN PRESS
N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233 207

(a) Subgrade stability. (b) Subgrade integrity.

Waste
Waste Shearing of liner
Failure
into waste
Liner Liner
slope failure including waste Differential settlement
∗ Waste provides loads ∗ Waste provides loads

(c) Waste slope stability. (d) Shallow slope lining stability

Waste Waste

Cover soil layer` Lining system

∗ Shearing in waste ∗ Shearing in waste and provides load

(e) Shallow slope lining integrity. (f ) Steep slope lining stability.

Settlement

Waste
? Waste
Loss of protection
Shear failure of
material liner
Geotextile
Liner
∗ Stresses in settement of waste Geomembrane
∗ Stresses in waste, stiffness and strength

(g) Steep slope lining integrity. (h) Cover system integrity.

Waste

Differential
Bulging of mineral liner settlement
due to lack of support Shearing of
mineral liner
Shearing of liner Waste

∗ Stresses in water and stiffness ∗ Settlement of waste

(i) Drainage system integrity.

Collapse of drainage ( j) Leachate/gas well integity.
pipes and geocomposite
drains
Settlement
Waste Waste
Non-uniform
horizontal stresses

∗ Waste provides loads ∗ Settlement in waste and stresses

Fig. 1. Potential landfill infrastructure failure modes: stability and integrity.

properties required to perform an analysis of each of the failure modes summarized

in Fig. 1. This paper reviews engineering behaviour of municipal solid waste (MSW)
with respect to the information required for design. A brief summary is provided for
 ARTICLE IN PRESS
208 N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233

Table 1
Engineering properties of MSW required for design

Design Unit Vertical Shear Lateral Horizontal Hydraulic

case weight compressibility strength stiffness in-situ conductivity
stress

Subgrade stability X X X
Subgrade integrity X X X X
Waste slope stability X X X X
Shallow slope liner stability X X X X
Shallow slope liner integrity X X X X X
Steep slope liner stability X X X X
Steep slope liner integrity X X X X X
Cover system integrity X X X
Drainage system integrity X X
Leachate/gas well integrity X X X X X X

each of the main engineering properties. References are made to key publications,
methods of measurement and calculation are summarized, and where possible,
typical ranges of values are given.

1.2. Material description

MSW is a mixture of wastes that are primarily of residential and commercial

origin. Typically, MSW consists of food and garden wastes, paper products, plastics,
rubber, textiles, wood, ashes, and soils (both waste products and material used as
cover material). A wide range of particle sizes is encountered ranging from soil
particles to large objects such as demolition waste (reinforced concrete and
masonry). The proportion of these materials will vary from one site to another
and also within a site. Life style changes, legislation, seasonal factors, pre-treatment
and recycling activities result in a changing waste stream over time. Examples are
increasing plastic and decreasing ash content over the past few decades in developed
countries (e.g. Fig. 2). In addition, member states of the European Union will
see a reduction in biodegradable waste in landfills through the introduction of
Biodegradable Municipal Waste diversification targets in the Council of
European Community (1999). It should be noted that the composition of MSW
varies from region to region and country to country. For example, developing
countries often have waste streams that contain more biodegradable material and
less plastics, and countries such as Germany with well developed re-cycling and pre-
treatment policies (e.g. the use of mechanical and biological pretreated waste), have
wastes with less biodegradable content and a more uniform and consistent grading.
These variations produce fundamental and significant differences in waste
engineering behaviour and they must be taken into consideration when using results
from the literature.
 N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233 209

100%

90%
Part of composition [weight-%]

80%

70%

60%

50%

40%

30% Miscellaneous (incl. plastics)

Metal and glass
20%
Paper (incl. rag)
10% Vegetable
Dust and cinder
0%
1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000
Year

Fig. 2. Composition of UK MSW 1935–2000 (after Watts et al., 2002).

1.3. Waste mechanics

The current understanding of waste behaviour is far from being complete. For
engineering the disposal of waste, researchers and practitioners have relied on their
knowledge of the behaviour of soils. Although this has been helpful to some extent,
there is an increasing realization that behaviour of waste should be considered in the
context of a separate discipline of waste mechanics (also referred to as waste
geotechnics). As a starting point, it is appropriate to compare some of the results
from preliminary and novel studies on waste properties available in the literature
with those of geological and engineered materials, e.g. granular soils, peat, reinforced
soil. Similarities and differences in measured behaviour can then lead to the
development of laboratory and field tests specifically for obtaining engineering
properties of MSW. An increasing number of international researchers are
investigating engineering behaviour of waste and its interaction with engineered
containment systems.
Evaluating the engineering properties and hence behaviour of MSW bodies is
challenging due to the variety of materials present. It is preferable to undertake
testing on real materials in an undisturbed state. However, this is not always
possible. Undisturbed samples cannot be taken and therefore laboratory tests have
to be on disturbed material that is re-compacted into test apparatus. MSW can be
highly structured material resulting from the method of placement and this structure
will be destroyed. In addition, variation in composition between samples can be
extreme, making it difficult to quantify the contribution to behaviour of the different
components of waste or mechanisms of behaviour. It is also difficult to
systematically change the proportion of waste constituents in order to investigate
 ARTICLE IN PRESS
210 N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233

the role each plays. This is required in order to evaluate the impact of future changes
in waste composition. Additional considerations are the very large size of test
apparatus required to accommodate large particles, and health and safety
requirements that dictate tests on real waste have to be carried out in a controlled
laboratory environment. These are both expensive to construct and operate. An
additional major factor is that engineering properties of waste vary with time due to
the degradation process. At present there are no internationally accepted standard
sampling and testing procedures for waste materials. In addition, there is presently
no accepted guidance on selection of appropriate values of the engineering
parameters for use in design, or agreed approaches for assessing waste behaviour
as part of the design process.

1.4. Waste classification

There are a number of general waste classification systems in common use, and
these have been developed to provide information for specific end uses, e.g. re-
cycling/waste minimization, assessment of biodegradation potential and calorific
value. However, for assessment of engineering behaviour a classification is required
that groups waste constituents in terms of their mechanical properties. In a typical
landfill there will be three distinct phases present—solid, liquid and gas. There may
also be a need to distinguish between mobile liquid in large drainable pores and
liquid in small pores (inter-particle), and liquid that is trapped, absorbed or
otherwise bound to the solid fraction (intra-particle). The information required to
classify waste components can be summarized as:

Knowledge of component shape to distinguish between soil-like (three-dimen-

sional, e.g. granular) and non-soil-like (two-dimensional, e.g. foils) components.
This allows classification of components in relation to their potential for
influencing mechanical behaviour of the waste mass (e.g. shear strength).

Grading by size for each group of components.

A distinction between the material groups (i.e. based on material properties—
metal, paper, plastic), with dominant groupings established, in conjunction with
information on the proportion (e.g. by weight) of the material groups in relation
to the size and mechanical properties of components.

An assessment of compressibility and potential for components to change shape.

An assessment of degradation potential for both organic and inorganic
components.

A number of the existing engineering classification systems are simply based on

material groups (Siegel et al., 1990) or on the distinction between soil-like and non-
soil-like, or fibrous, appearance (Manassero et al., 1996; Thomas et al., 1999). These
existing classification systems do not fulfil the requirements of a rigorous
classification framework. Table 2 provides a summary of classification systems
including the parameters defined. A new classification framework for MSW that
fulfils the requirements listed above has been proposed by Langer and Dixon (2004).
 ARTICLE IN PRESS
N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233 211

Table 2
Overview of existing classification systems

Author Key classification criteria Parameters

Turczynski (1988) Waste type Density, shear parameters,

liquid/plastic limit,
permeability
Siegel et al. (1990) Material groups Part of composition
Landva and Clark (1990) Organic and inorganic Degradability (easily, slowly,
materials non) Shape (hollow, platy,
elongated, bulky)
Grisolia et al. (1995a) Degradable, inert and de- Strength, deformability,
formable material groups degradability
Kölsch (1996) Material groups Size, dimension
Manassero et al. (1996) Soil-like and other Index properties
Thomas et al. (1999) Soil-like and non-soil-like Material groups

Further work is required to trial this classification on a range of waste types and to
relate classifications to mechanical behaviour of the waste body.

1.5. Literature on MSW engineering properties

There is a growing body of literature on the measurement of engineering

properties of MSW. Unfortunately, due to the lack of both agreed classification
system and test standards it is difficult to interpret published results. Often the nature
of the waste tested is not described in any detail and the test boundary conditions are
rarely given. This makes it difficult to amalgamate the results into a common
framework or to apply findings to other sites. This review concentrates on the key
parameters of unit weight, compressibility, shear strength, lateral stiffness, in situ
horizontal stress and hydraulic conductivity. Where possible, the variation of these
parameters with time is also considered. This review has benefited from publications
by Landva and Clark (1990), Fassett et al. (1994), Manassero et al. (1996), Eid et al.
(2000), Chapter 6 of Qian et al. (2002) and Kavazanjian (2003), each of which was
written to summarize the state-of-the-art.

2. Unit weight of MSW

As shown in Table 1, knowledge of unit weight is required for all aspects of landfill
design, and therefore it is surprising that so few detailed studies have been
conducted. Unit weight values vary significantly both between sites and within a
single site. MSW has highly variable components, types and amounts of cover soil
differ between sites, the percentage of inert and industrial wastes varies and
placement procedures play an important role, as do environmental conditions (e.g.
rainfall). Common difficulties in assessing MSW unit weight have been summarized
 ARTICLE IN PRESS
212 N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233

by Fassett et al. (1994) as: separation of the contribution of daily soil cover; assessing
the changes in unit weight with time and depth; the majority of reported values
reflect waste near or at the surface; and obtaining data on the moisture content of the
waste. Fassett et al. (1994) considered that the following factors should be recorded
along with measured unit weights: MSW composition including daily cover and
moisture content; method and degree of compaction; the depth at which the unit
weight was measured; and the age of the waste. The form of the unit weight
measurement should also be recorded and noted by those using the data. Values can
be given as dry unit weight (sample could have been artificially dried), bulk unit
weight (some moisture present but waste not saturated) and saturated unit weight. In
most studies it is the bulk unit weight that is measured and reported.

2.1. Unit weight estimation methods

Unit weight can be estimated and measured using several techniques. These are
summarized in Table 3. Methods based on direct field measurement are considered
to be the most reliable as the influence of waste placement is considered as is the
overburden stress in some methods. Most of the information in the literature relates
to recently placed waste at low overburden stresses.

2.2. Factors affecting unit weight of waste

As with soils, the unit weight is affected by compaction effort and layer thickness,
the depth of burial (i.e. overburden stress) and the amount of liquid present
(moisture content). Unlike soils, the unit weight also varies significantly because of
large variations in the waste constituents (e.g. size and density), state of
decomposition and degree of control during placement (such as thickness of daily
cover or its absence). It is generally believed that initially the unit weight of waste is
very much dependent on waste composition, the daily cover and the degree of
compaction during placement. But as the waste becomes older the unit weight
becomes more dependent on the depth of burial, the degree of decomposition and
climatic conditions. Although unit weight can vary significantly over short distances,
this is not necessarily a major concern in design. Unit weight is used to calculate
vertical and horizontal stresses. Average values of unit weight are acceptable in most
design scenarios (e.g. average values of vertical stress acting on a basal geomembrane
are used to design geotextile protection layers).

2.3. Waste components

Waste components have a controlling influence on the average unit weight of the
waste mass. Individual waste components have a wide range of particle unit weights
and these can change with time. Components may have voids within them in
addition to those between components. This results in a significant percentage of
waste particles behaving differently to soil particles due to their high compressibility.
Degradation of components with organic content will result in a loss of mass,
 N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233
Table 3
Methods for measuring unit weight

Location Method of measurement Comments References

ARTICLE IN PRESS
Field
Large-scale replacement density measurements
Reliable but tests are all at low vertical stresses Gotteland et al. (2002)
from waste surface

Replacement density measurement in boreholes
Reliable and data obtained for a range of vertical Kavazanjian et al., (1995)
stresses

Gamma ray logging of boreholes
Variable results due to range of particle types —

Direct measurement of vertical stresses within
Reliable and shows changes related to vertical Gourc et al. (2001)
waste body stress

Calculation from landfill volume and weight of
Average values obtained are of little use —
waste materials
Laboratory
Measurement of large-size samples
Disturbed sample but large size and range of Powrie and Beaven (1999)
vertical stresses possible

Measurement of small-size samples
Disturbed often pre-treated and sorted, —
unreliable

Measurement of individual component weights
Time consuming and often inaccurate —
and percentages present in sample

213
 ARTICLE IN PRESS
214 N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233

changes in size and alteration of the mechanical properties (i.e. compressibility and
shear strength). It will also change the density of the component. As a waste body
degrades, void ratio reduces and hence a volume reduction occurs. Although there
are few field measurements in degraded waste it is generally believed that
degradation results in an increase in waste density, and hence unit weight.

2.4. Compaction

Since MSW is a particulate material and a large proportion of the components

have a high void ratio and a high compressibility, compaction processes will reduce
the voids within an individual component as well as voids between various
components. The unit weight of compacted waste will depend upon the waste
components, thickness of layer, weight and type of compaction plant and the
number of times equipment passes over the waste. A layer thickness of 0.5–1.0 m will
facilitate the achievement of good compaction and hence high unit weights, however
it is not untypical for waste to be placed in layers of 2–3 m thick. This results in poor-
to-moderate compaction. Fassett et al. (1994) conducted a detailed survey of bulk
unit weight data from the international literature. A statistical analysis of the data is
shown in Table 4. The degree of compaction was derived from an assessment of
individual site practices. Poor relates to little or no compaction, moderate to ‘old’
practices and good to ‘current’ (1994) practices. The assessment was in most cases
subjective but provides a useful guide. An important result is the large variation in
unit weight when little or no compaction is used. Landva and Clark (1990)
and Oweis and Khera (1986) report similar ranges of bulk unit weights. A summary
of measured values from recent studies conducted in four countries is provided in
Table 5. These are also consistent with the Fassett et al. (1994) reported values and
indicate that current practice is still only achieving ‘moderate’ levels of compaction
for placement of fresh MSW.

2.5. Depth

Unit weight of waste varies with effective stress, which is a function of depth.
Fig. 3 produced by Powrie and Beaven (1999) shows the variation in dry density,
saturated density and density at field capacity with vertical effective stress. The data
was obtained by compressing samples of waste in a large diameter cylindrical tests

Table 4
Statistical summaries of bulk unit weight data for fresh waste (Fassett et al., 1994)

Poor compaction Moderate compaction Good compaction

Range (kN/m3) 3.0–9.0 5.0–7.8 8.8–10.5

Average (kN/m3) 5.3 7.0 9.6
Standard deviation (kN/m3) 2.5 0.5 0.8
Coefficient of variation (%) 48 8 8
 ARTICLE IN PRESS
N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233 215

Table 5
Bulk unit weights from international literature

Country Measured bulk unit Comments References

weights (kN/m3)

United 6
Compacted in 2 m lifts using steel Watts and Charles

Kingdom wheeled 21 tonne compactor (1990)

8
0.6 m lifts using same compactor as

above

Belgium 5–10
Common compaction practice Manassero et al.

(1996)
France 7
Upper layers of fresh (non- Gourc et al. (2001)
degraded) MSW

USA 6–7
Fresh MSW after initial placement Kavazanjian (2001)

14–20
Degraded waste with high % of soil
like material

Fig. 3. Relationships between density and average vertical stress. Trend lines shown are based on average
measured values (after Powrie and Beaven, 1999).

chamber. One of the implications of this work, in terms of the waste density
achieved, is that compaction at the tipping face can have a similar effect to the burial
of the waste by several meters of overburden (Powrie et al., 1998). Due to the
difficulties and costs involved there are few field measurements of unit weight
variation with depth. Gourc et al. (2001) present initial data obtained during filling
of the Torcy landfill in France. The results shown in Fig. 4, as bulk unit weight
against overburden stress, demonstrate a clear trend of increasing unit weight with
stress level.
 ARTICLE IN PRESS
216 N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233

Unit weiht (kN/m )

3

Band of measured
values for point A

Band of measured
values for point B

Vertical stress (kPa)

Fig. 4. Bands of measured bulk unit weights plotted against vertical stress (after Gourc et al., 2001).

2.6. Moisture content

Moisture content of waste depends on the initial waste composition, local climatic
conditions, operating conditions, rate of decomposition and organic content. On
exposure to water, the unit weight of any constituent absorbing water would increase
(e.g. that of food waste, garden refuse, paper, textiles) due to increased moisture
content of the intra-particle voids. These increases in individual particle unit weight
are added to the increase in bulk unit weight resulting from increased leachate in the
void spaces between particles of waste to produce increases in the bulk unit weight of
the waste mass. Therefore, older waste would be expected to have a higher bulk unit
weight than fresh waste. Although there is limited field evidence to support this
proposed mechanism, the data from investigations such as those described by
Kavazanjian (2001) provide some corroboration. Daily cover soils play an important
role in controlling the amount and distribution of precipitation that enters waste.
They result in highly structured waste bodies (i.e. horizontal layers of waste bounded
by often low permeability layers of cover soils) and this can cause large spatial
variations in the moisture content of waste. The phasing of final cap construction
also influences the evolution of moisture content changes. Addition of liquid wastes
and re-circulation of leachate will both have a fundamental influence on the
magnitude and distribution of moisture contents, and hence on the magnitude and
distribution of bulk unit weight.

3. Compressibility

The compressibility of MSW has been studied for many decades. Studies of
settlement have been conducted to improve the efficiency of waste placement, predict
final settlement profiles for the cap and to enable assessment of interaction between
 ARTICLE IN PRESS
N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233 217

side slope barrier systems and the settling waste body (Section 1.1). This section
provides a brief summary of methods used to calculate settlements and, where
possible, ranges of typical values. Readers interested in more detailed information on
settlements are recommended to review the key technical papers by Fassett et al.
(1994), Manassero et al. (1996), Oweis and Khera (1998) and Gourc et al. (1998).

3.1. Calculation of vertical stress

It is usually assumed that traditional principles of soil mechanics theories of

settlement can be applied to solid waste. The unit weight, g; of a deposit increases
with depth as discussed previously. The overburden pressure, s; at a given depth, z, is
Z z
s¼ g dz. (1)
0

To take into account stress-dependent unit weight, the overburden pressure can be
calculated using
X
n
s¼ gi zi , (2)
i¼1

where unit weight, g; is assumed to be constant within a given layer i and n is the
number of layers.

3.2. Compression mechanisms

Mechanisms resulting in compression of waste have been summarized by

Manassero et al. (1996) and are listed in Table 6, as are the interrelated factors
affecting the magnitude of settlement. It can be assumed that the total settlement, dt,

Table 6
Waste compression mechanisms (Manassero et al., 1996) and factors controlling magnitude of settlement

Mechanisms resulting in settlement Factors controlling magnitude of settlement

Physical compression and creep (mechanical
Initial composition of waste (grading, particle
distortion, bending, crushing and re-orientation shape, material properties of components, e.g.
of particles) metal, paper)

Raveling settlement (migration of small particles
Initial density and voids ratio
into voids between large particles)
Layer thickness

Collapse of containers and bridging components
Type, thickness and number of cover soil layers
(physical/chemical changes such as corrosion
Stress history (pre- and post-filling mechanical
oxidation) treatment)

Decomposition settlement (biodegradation of
Leachate levels and fluctuations
organic components)
Environmental controlled factors (moisture

Settlement of subgrade under applied waste content, temperature, gas generation)
loading
Compressibility of subgrade
 ARTICLE IN PRESS
218 N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233

(excluding any contribution from the subgrade) is made up from two main
components; primary compression (dp) and secondary compression (ds)
dt ¼ dp þ ds . (3)
Primary compression includes physical compression of particles (distortion, bending,
crushing and particle orientation) and consolidation (significant for saturated waste
bodies). In most wastes, physical compression will occur immediately on application
of load (i.e. in response to placement of overlying layers of waste). Therefore,
primary compression will occur in a period of a few days to a few weeks and hence
can be considered to be short term. Incrementally linear compression models can be
used to calculate primary settlements (Section 3.3).
Secondary compression includes all creep effects (i.e. mechanical compression
under constant stress) and those relating to degradation (both chemical and
biological). Creep effects include time-dependent particle distortion (i.e. bending,
crushing), particle reorientation and raveling. Degradation includes collapse of
containers due to a change in strength (e.g. corrosion) and degradation of organic
compounds. Degradation potential of components is discussed by Landva and
Clarke (1990) and is a key element of any classification system. Biodegradation is the
main component of secondary compression in MSW landfills. Many methods have
been proposed to characterize and predict secondary compression. The degradation
process is influenced by a range of interrelated factors, all of which vary spatially
within a landfill and with time (e.g. moisture content, temperature and stress level).
Present methods of prediction are simplistic and many rely on curve fitting
techniques. A brief introduction to calculation methods is provided in Section 3.4.
Secondary compression occurs throughout the active life of the landfill and is usually
the main component of total settlement.

3.3. Primary compression

The principal source of loading is self-weight, which results in waste settlement

during construction. Waste placement can be considered to be a one-dimensional
compression problem (e.g. waste is placed over a large area in relation to the
thickness of the deposit). An increment of vertical effective stress Ds0 v, produces an
increase in vertical strain Dv. Stresses are assumed to be effective for fresh waste due
to its typical low-moisture content and hence strains are assumed to occur
immediately on application of stress. A constrained modulus D, can be defined as
Ds0v
D¼ ðunits kN=m2 or MN=m2 Þ. (4)
Dv
The settlement during construction can be computed using
Xn
H i Ds0v
dp ¼ , (5)
i¼1
Di
where Ds0 v is the change in vertical effective stress, Hi is the thickness of the sublayer
i of waste, Di is the constrained modulus of layer i.
 ARTICLE IN PRESS
N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233 219

7000
Drai ned co nstrained modulus, D' (kPa) Gotteland et al. (2001)
6000 Powrie & Beaven (1999);
Beaven & Powrie (1995)
Landva et al. (2000)
5000
Dixon et al. (2004)

4000

3000

2000

1000

0
0 50 100 150 200 250 300 350 400 450 500
Mean vertical effective stress (kPa)

Fig. 5. Constrained modulus vs. stress level (after Dixon et al., 2004). J Gotteland et al. (2001); & Powrie
and Beaven (1999); Beaven and Powrie (1995); W Landva et al. (2000) and E Dixon et al. (2004).

The compression index (Cc) can also be used to relate increments of strain to
increments of stress change (see Fassett et al., 1994). Primary compression will occur
during waste placement. As the thickness of the waste increases the stiffness of the
waste will also increase with depth. Constrained modulus is therefore not a constant
but depends upon the level of mean stress in the layer under consideration. The
compression of each layer is calculated separately using the relevant D value and the
total primary compression is calculated as the sum of the individual layers (Eq. (5)).
If the waste layer is saturated, the final primary compression will still be calculated
using D but it will take place over an extended period, controlled by the permeability
of the waste layer and length of the drainage path (i.e. standard consolidation
theory). Note that D ¼ 1=mv where mv is the coefficient of compressibility in m2/kN.
Fig. 5 shows a summary of constrained moduli values for MSW related to stress level
(Dixon et al., 2004). This data can be used to estimate primary compression.

3.4. Secondary compression

As discussed above, long-term settlement is mainly due to biodegradation and

mechanical creep compression. It is common practice to model secondary
compression using the following linear relationship on a settlement vs. log-time graph
t
ds ¼ C a H log , (6)
tp
where t is the time at which settlement due to secondary compression is required
(t4tp); tp is the time for completion of primary compression and Ca is the secondary
compression ratio given by
D
Ca ¼ . (7)
log t2  log t1
 ARTICLE IN PRESS
220 N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233

Table 7
Secondary compression parameters for MSW material (after Oweis and Khera, 1998)

Material Ca

Ten year old landfill 0.02

Fifteen year old landfill 0.24
Fifteen to twenty year old landfill 0.02
Old landfill 0.04
Old landfill with high soil content 0.001–0.005

There is some field data obtained from long-term settlement monitoring studies
to support this approximation. Oweis and Khera (1998) published values of Ca
for a range of waste materials obtained from the literature. Table 7 shows selected
values from their summary and demonstrates the problems of trying to use one Ca
value for the entire period of secondary compression. As the rate of degradation is
unlikely to be constant with time, it is not surprising that Ca is not a constant.
Gourc et al. (1998) provide a comprehensive review of available calculation methods.
Fassett et al. (1994) and Manassero et al. (1996) both give useful summaries
of secondary compression data. Settlement prediction techniques based on modelling
the biodegradation process are under development and appear promising
(e.g. McDougall and Pyrah, 2001).

3.5. Total compression

Actual computations of total compression (i.e. settlement) can be complex. For

example, to estimate the total settlement of a waste deposit, the following
considerations will be necessary for each of the layers in the landfill:

settlement of waste from self-weight (primary compression);

settlement from the weight of each subsequent layer (including final cover) that
overlies the given layer (primary compression);

settlement due to secondary compression, taking into account that Ca is likely to
decrease with age;

settlement of the mineral basal liner (if present) due to primary and secondary
compression; and

settlement of any compressible subgrade.

Due to the heterogeneous nature of landfill constituents and their varied rates of
decomposition, differential settlements occur. The problem is further complicated by
the fact that adjacent cells are completed at different times and filling often takes
place on top of older waste deposits. Differential settlements are important as they
can jeopardize the stability of the final cap and integrity of geosynthetic elements
(e.g. geomembranes) and mineral layers (e.g. compacted clay barriers).
 ARTICLE IN PRESS
N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233 221

4. Shear strength

At present there is limited information on in situ MSW shear behaviour. Shear

strength of MSW is usually defined using the Coulomb failure criterion. This is
commonly used in soil mechanics and in studies of other particulate materials. The
shear strength parameters that define the failure envelope are the slope of the failure
envelop (f) and intercept on the y-axis (c). The intercept, c, can denote real cohesion
between particles, but is often a function of one or more of the following: curvature
of the failure envelope, variation between samples, measurement errors, or an
indication of tensile strength. Therefore, it is common to define it as the ‘apparent
cohesion’ or ‘cohesion intercept’. Care should be exercised when applying experience
of shearing in soils to the study of MSW. Waste contains particles that are
compressible, can sustain large tensile strains (e.g. plastic) and can change with time
(e.g. through degradation). An outcome of using the Coulomb criterion is that it
gives an increase in shear strength with increasing stress level, based on the slope of
the failure envelop, and hence with depth of burial. This is consistent with waste
being considered as a frictional material.

4.1. Measurement of MSW shear strength

Table 8 provides a summary of the advantages and disadvantages associated with

currently available approaches for obtaining information on shear behaviour of
MSW. For the reasons outlined in Section 1.3 it is both difficult and costly to obtain
representative and hence reliable strength parameters for MSW (i.e. large particle
size, heterogeneity, control of structure, etc.). Therefore, it is preferable to obtain
values from field studies (Kavazanjian, 2003). Back analysis of landfill slope failures,
cut slope trials and existing stable slopes can provide information on the shear
strength of a large mass of waste, but poor quality input data makes such analyses
problematic and often unreliable. In situ techniques for measuring shear strength are
presently inadequate and unreliable. An in situ technique for measuring the shear
strength of MSW at a range of depths and for material with varying degrees of
degradation is urgently required. Results from laboratory tests should be viewed
with skepticism. The waste will have been disturbed, and hence the structure will
have been lost, large particles may have been removed or processed and the in situ
density and stress conditions may not have been reproduced. Many of the studies in
the literature have used triaxial compression tests. Kavazanjian (2001) provides a
detailed assessment and concludes that triaxial compression testing is not an
appropriate technique for measuring the shear strength of MSW. Inability to cause
shear failure in laboratory tests has led to shear strength being related to levels of
strain (i.e. different shear strength parameters are given for each strain level). While
this approach has some merit if used in design to try and control strains in the waste
body, it can lead to confusion and hence great care should be taken in applying such
values. The most appropriate laboratory technique is the direct shear test, although
the general concerns regarding the applicability of laboratory tests discussed above
still apply.
 222
N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233
Table 8
Review of methods for measuring shear behaviour of MSW

Location Method of measurement Comments References

Field
Back analysis of slope failures
Adequate information seldom available (e.g. pore Koerner and Soong (2000)
pressures, shape and position of shear surface)

ARTICLE IN PRESS

Back analysis of cutting slope
Large deformations observed but not shear failure Singh and Murphy (1990),
experiments Cowland et al. (1993)

Back analysis of existing stable slopes
Changing waste composition means past experience is Gotteland et al. (2002)
not a guide to future performance

In situ direct shear tests
Difficult to perform and results relate to low levels of e.g. Jessberger and Kockel
stress (1993)

SPT, CPT and vane tests
No clear relationship between penetration resistance
and MSW shear strength, could provide useful
information in degraded more soil like materials

Laboratory
Triaxial compression
Disturbed samples, peak shear strength not obtained Jessberger (1994), Grisolia
due to compression and densification of sample et al. (1995b)

Direct shear
Large device required (e.g. 1  1  1 m), disturbed Kolsch (1995), Gotteland
samples, large displacements required to mobiles peak et al. (2001)
shear strength

Simple shear
Large device required, disturbed samples, useful Kavazanjian et al. (1999)
information on shear stiffness (used in seismic analyses)
 ARTICLE IN PRESS
N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233 223

4.2. Measured shear strength values

The majority of the studies included in this review obtained strengths from direct
shear box tests and back-analyses of failures. Those requiring more detailed
information should read Manassero et al. (1996), Jones et al. (1997) and
Kavazanjian (2001) all of which provide summaries of shear strength parameters
from the literature. Table 9 gives waste shear strength parameters summarized by
Jones et al. (1997) although it is by no means a comprehensive summary. It is
included to demonstrate the wide variation in values that can be obtained. Given the
large range of possible wastes and the difficulties involved in measuring shear
strength, the large scatter is not surprising.
Manassero et al. (1996) suggested that the failure envelope shown in Fig. 6 could
be used as a starting point in design if no site-specific data is available. Design values
of c and f are defined according to three distinct zones:

Zone A: corresponding to very low stress (0 kPapsvo20 kPa) where the MSW
behaviour can be described as being only cohesive. In this case, c ¼ 20 kPa:

Table 9
Examples of measured shear strength parameters from the literature (Jones et al., 1997)

Reference Shear strength Method Comments

parameters

c0 (kPa) f0 (1)

Jessberger (1994) 7 38 Not stated Reporting Gay and Kaiser (1981)

Jessberger (1994) 10 15 Back analysis Reporting Spillman (1980)
Jessberger (1994) 10 17 Back analysis Reporting Spillman (1980)
Jessberger (1994) 0 30 Estimate From field observations
Jessberger (1994) 0 40 Estimate From field observations
Jessberger (1994) 7 42 Simple shear Reporting Gay and Kaiser
(1981). Nine month old MSW
Jessberger (1994) 28 26.5 Simple shear Fresh MSW. Reporting Gay and
Kaiser (1981)
Fassett et al. (1994) 10 23 Suggested values Suggested by authors
Kolsch (1995) 15 15 Suggested values Suggested by author
Kolsch (1995) 18 22 Suggested values Suggested by author
Cowland et al. (1993) 10 25 Back analysis Deep trench cut in waste.
Suggested values by authors
Del Greco and Oggeri (1993) 15.7 21 Direct shear Tests on baled waste
Lower density bales
Del Greco and Oggeri (1993) 23.5 22 Direct shear Tests on baled waste
Higher density bales
Landva and Clark (1986) 19 42 Direct shear Old refuse
Landva and Clark (1986) 16 38 Direct shear Old refuse
Landva and Clark (1986) 16 33 Direct shear Old refuse+1 year
Landva and Clark (1986) 23 24 Direct shear Fresh, shredded refuse
Landva and Clark (1986) 10 33.6 Direct shear Wood waste/refuse mixture
Golder Associates (1993) 0 41 Direct shear Project specific testing
 ARTICLE IN PRESS
224 N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233

400
Envelope from literature reported by
Jones et al. (1997)
Suggested design line c' = 5 kPa
φ'= 25˚ reported by Jones et al. (1997)
Suggested design line based on
North American studies (Kavazanjian 2001)
300 Suggested design (Manassero et al. 1996)
Shear Stress [kPa]

200

100

0
0 100 200 300 400
Normal Stress [kPa]

Fig. 6. Suggested MSW shear strength envelopes for design (after Jones et al., 1997).

Zone B: corresponding to low to moderate stresses (20 kPapsvo60 kPa). In this

case, c ¼ 0 kPa and fE381.

Zone C: corresponding to higher stresses (svX69 kPa). In this case, cX20 kPa and
fE301.

In a similar approach, and based on data from North American studies,

Kavazanjian (2001) suggested c ¼ 24 kPa and f ¼ 0 for normal stress below
30 kPa and c ¼ 0 and f ¼ 331 for higher normal stresses. This envelope is also shown
in Fig. 6. It is believed that some of the Kavazanjian (2001) data was considered by
Manassero et al. (1996) and therefore contributed to development of their envelop.
Based on the data in Table 9, Jones et al. (1997) suggested a design line defined by
c ¼ 5 kPa and f ¼ 251 (Fig. 5). The three ‘suggested’ design conditions differ and
therefore caution should be exercised when using the literature to obtain values for
use in assessment of specific site and waste conditions. It would be considered non-
sensical to suggest that a single failure envelope could be used for all soil types and
suggesting the same for waste is equally ridiculous.
Kolsch (1995) has investigated the tensile strength of MSW using a modified
version of the large shear box. This research was aimed at assessing the contribution
to shear strength from the reinforcement provided by waste fibres (e.g. plastic
sheets). The results obtained help explain the stability of steep cut faces in waste and
of deep tension cracks that have been observed to form in waste masses under certain
circumstances. However, further investigations to quantify the contribution of
tensile strength are required before it should be used routinely in design.
 ARTICLE IN PRESS
N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233 225

4.3. Use of waste shear strength in landfill design

Knowledge of shear strength is required in order to assess waste and lining system
slope stability. Landfill failures tend to be controlled by shear surfaces forming along
interfaces within the liner system or within weak underlying soils. However, failures
do occur entirely within the waste mass, and those that are controlled by weak zones
and interfaces still often have a section of the shear surface forming in the waste.
Therefore, while it is important to evaluate weak interfaces and/or poor foundation
materials it is also necessary to estimate the strength properties of waste when
conducting stability analyses. Waste slope design and assessment is presently based
substantially on experience (i.e. ‘‘x’’ degree angle slopes have been stable for ‘‘y’’ years
therefore this angle can be used for new slopes). Summaries of results from
international research are also presently used in design as discussed above. An
approach based on past experience is flawed for two reasons: (i) MSW slope failures
do occur; and (ii) The constituents, and hence mechanical properties, of new MSW
are constantly changing. In addition, mechanical properties of MSW change with
time due to the degradation process and a slope could become unstable tens of years
after its formation. The design of safe waste slopes in both the short- and long term is
critical to the management of sites and hence optimization of the landfill construction
process, and ensuring the required performance of the lining components. Many of
the failures recorded in the literature are of temporary waste slopes.

5. Lateral stiffness

Information on the lateral stiffness of MSW is required to assess the performance

of steep side slope lining systems that rely in part on the waste for their stability and
integrity. To date, Dixon and his co-workers have published the only information on
lateral (i.e. horizontal) waste stiffness (Dixon and Jones, 1998; Dixon et al., 2000).
This section provides a brief summary of the results obtained by carrying out
pressuremeter tests at different depths in MSW of varying age. This research was
conducted as part of a project to investigate the interaction between steep slope
lining systems and adjacent waste.

5.1. Stiffness parameters of MSW

Elastic parameters such as shear modulus (G), Young’s modulus (E) and Poisson’s
ratio (n) can be used to quantify the response of a material to a change in stress (i.e.
calculate strains). The parameters are related, as is the constrained modulus (D)
introduced previously, and an example of their interdependence is given by
E
G¼ . (8)
2ð1 þ nÞ
In situ measurement of such parameters is required. Waste placement methods,
waste type and depth of burial will have a fundamental influence on the measured
 ARTICLE IN PRESS
226 N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233

values. Tests on disturbed samples will not provide representative results. The
measured values are dependent on, and therefore can be related to, other physical
properties such as density and stress level.

5.2. Lateral shear stiffness parameters obtained from pressuremeter tests

Pressuremeter testing is a standard technique used in soil and rock mechanics to

measure stiffness parameters (e.g. in situ lateral shear stiffness) and other ground
properties such as in situ horizontal stress and, in certain materials, shear strength
parameters. Dixon and Jones (1998) and Dixon et al. (2000) described a novel
method of obtaining in situ stresses and shear stiffness values using the
pressuremeter test in MSW. The test takes the form of inflating a membrane to
expand a preformed cylindrical test pocket. The pressure required to expand the
pocket (and hence deform the surrounding material) is related to the magnitude of
radial expansion. Tests have been carried out in both fresh and partly degraded
MSW at depths to 17.0 m below ground level. In excess of 30 individual tests have
been conducted.
Stiffness values are obtained by calculating the slope bisecting small cycles of
unloading and reloading. Fig. 7 shows a typical pressuremeter test result for MSW.
Unload/reload loops can be used to obtain values for elastic shear modulus. Fig. 8
shows the measured relationship between shear modulus and maximum cavity
pressure (i.e. horizontal stress in the waste adjacent to pressuremeter) at the start of
the unload/reload loop, for wastes of different ages. General trends of increasing
stiffness with stress level are observed, and the older waste (12–15 year old partly
degraded) appears to be less stiff than fresh waste (1–5 year old) for comparable
stress levels. The observed scatter of data is unsurprising and results from using
different strain amplitudes for the unload/reload loops and by variability of the
waste composition (i.e. some tests are in pockets of stiff construction rubble rich
waste and some in weak wet waste). It should be noted that the degraded waste may
have a different initial composition than current fresh waste, and hence this observed
lower stiffness could be a result of waste composition rather than degree of
degradation. The trend of increasing stiffness with mean stress, and hence depth, is

800
700
Total Pressure at Cavity

600 Secant shear moduli from

unload/reload loop
500
Wall (kPa)

400
300
200 2Gur
100
Unload/reload loops
0
0 4 8 12 16 20
100 -
Average Radial Dispalcement (mm)

Fig. 7. Example result of a pressuremeter test in MSW.

 ARTICLE IN PRESS
N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233 227

1st Loop - Fresh MSW Burntstump

Shear Modulus (MPa)
2nd Loop - Fresh MSW Burntstump
0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 3rd Loop - Fresh MSW Burntstump
0 4th Loop - Fresh MSW Burntstump
1st Loop - Fresh MSW Calvert (SBP)
Maximum Cavity Stress Before Start of Loop (kPa)

100 2nd Loop - Fresh MSW Calvert (SBP)

3rd Loop - Fresh MSW Calvert (SBP)
200 4th Loop - Fresh MSW (SBP)
5th Loop - Fresh MSW Calvert (SBP)
300 1st Loop - Fresh MSW Calvert (HPD)
2nd Loop - Fresh MSW Calvert (HPD)
400 3rd Loop - Fresh MSW Calvert (HPD)
4th Loop - Fresh MSW Calvert (HPD)
500
5th Loop - Fresh MSW Calvert (HPD)
1st Loop - 12 year old MSW (SBP)
600
2nd Loop - 12 year old MSW (SBP)
3rd Loop - 12 year old MSW (SBP)
700
4th Loop - 12 year old MSW (SBP)
1st Loop - 15 year old MSW
800
2nd Loop - 15 year old MSW
3rd Loop - 15 year old MSW
900

Fig. 8. Shear modulus vs. maximum stress at start of unload/reload loops for fresh and partially degraded
MSW-SBP are self-boring pressuremeter test and HPD are high-pressure dilatometre tests.

as expected for a drained particulate material. Although there are no other studies in
the literature to corroborate these results, the systematic and consistent behaviour
observed provides confidence in the validity of the measured trends.

6. Horizontal in situ stress

Knowledge of horizontal in situ stress is required in order to aid assessment of

stability and integrity of both shallow and steep slope lining system components and
the performance of structures buried in the waste body such as leachate and gas
wells. Measurement of horizontal stress in a particulate material such as waste is
difficult because the act of introducing a measuring instrument will alter the stress
being measured. For a body at rest, horizontal stresses (sh) can be related to vertical
stresses (sv) by the coefficient of each pressure at rest (K0) where
sh
K0 ¼ . (9)
sv
6.1. Laboratory study

Laboratory measurement of horizontal stress in a waste body can only provide an

indication of possible field behaviour. Laboratory samples cannot easily replicate the
field conditions, especially particle size and method of placement, and hence the
structure of the sample cannot be reproduced. These factors play important roles in
 ARTICLE IN PRESS
228 N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233

the generation of horizontal stresses. Landva et al. (2000) have produced the only
results from a laboratory study of MSW. They conclude that K0 values in the order
of 0.35–0.4 would be typical for fresh MSW and that K0 would be expected to
increase towards a value of 0.5 if less reinforcing material was present. If the
degradation process destroys reinforcing material, these results indicate that K0
values, and hence horizontal stresses, will increase with time. To date, this has not
been substantiated by field measurements.

6.2. Field measurements

An estimation of K0 values have been made using results from pressuremeter tests.
The preliminary results from the study are shown in Fig. 9 (Dixon and Jones, 1998).
It can be seen that there is no clear relationship between K0 and depth. This is due to
disturbance caused by insertion of the pressuremeter (i.e. changing the values of

Pressure Coefficient (Kw) for MSW

0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40
0

4
Height of Waste Above Measuring Point (m)

10

12

14

16
Measurement made using a
pair of horizontal and vertical
18 pressure cells
Calculated from SBP tests

20

Fig. 9. Coefficient of earth (waste) pressure at rest measured using pairs of pressure cells (Dixon et al.,
2004) and calculated from self-boring pressuremeter tests (Dixon and Jones, 1998) vs. depth of burial.
 ARTICLE IN PRESS
N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233 229

horizontal stress being measured) and the heterogeneous nature of the waste tested.
However, the results do appear to be suggesting that higher values than those
obtained by Landva et al. (2000) might be applicable for in situ material. Dixon et al.
(2004) report direct measurements of horizontal stresses in MSW. Pairs of pressure
cells were buried in waste at a range of depths to measure vertical and horizontal
stresses as part of a study of steep slope lining system performance. The preliminary
measurements have been used to calculate waste pressure coefficient values (Kw),
which can be interpreted as K0 values (Dixon et al., 2004), and these are also shown
in Fig. 9. As with the pressuremeter test results, values consistently higher than those
proposed by Landva et al. (2000) are indicated.

7. Hydraulic properties

Waste hydraulic conductivity is important to landfill designers because of the

influence it has on leachate pressure distributions in the waste body and hence on the
magnitude and distribution of effective stresses and therefore on shear strength. The
heterogeneous nature and placement controlled structure of MSW result in widely
varying permeability values in a given deposit. There is limited information in the
literature on MSW hydraulic conductivity measured in situ and therefore the current
understanding is incomplete. However, field observations of fluid flow in waste
bodies and extensive large-scale laboratory experiments provide some information
for use in design.
Placement of waste in layers and the use of daily cover soil (often of relatively
low permeability) result in waste bodies having a structure of sub-horizontal
layers and with foil like particles (e.g. paper and plastic) orientated horizontally.
This produces anisotropic hydraulic properties with higher permeability in a
horizontal direction. Perched leachate is often found above cover soil or low-
permeability layers, and horizontal seepage of perched leachate day-lighting onto
temporary waste slopes has been observed in many landfills. In addition to waste
structure, a second major control on permeability is stress level. A comprehensive
study of waste hydraulic conductivity was reported by Powrie and Beaven (1999)
using a large size compression chamber. They found that the hydraulic conductivity
of non-degraded MSW could reduce by over three orders of magnitude
to approximately 108 m/s between placement and burial to a depth of 60 m
due to compression. Absorption of fluid by waste particles, flow through partially
saturated material and the influence of gas generation on hydraulic conductivity
all require further investigation. The stress dependency of waste hydraulic
conductivity has major implications for the operation of leachate extraction and
recirculation systems, and basal and side slope drainage design. These all influence
the pore water pressure distributions within the waste body, and hence the effective
stresses and shear strength. Generation of high leachate pressures has been a
significant factor in a number of large landfill failures (e.g. Brink et al., 1999;
Hendron et al., 1999).
 ARTICLE IN PRESS
230 N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233

8. Summary of key issues

The current understanding of waste behaviour is far from complete. Evaluating

the engineering properties and hence behaviour of MSW is very difficult due to the
variety of materials present and the influence of waste structure. Knowledge of unit
weight of MSW is required for all aspects of design. Initially the unit weight of waste
is dependent on waste composition, the daily cover and the degree of compaction
during placement. As the waste becomes older, the unit weight becomes more
dependent on the depth of burial, the degree of decomposition and climatic
conditions. Mechanisms resulting in settlement of waste include physical compres-
sion and creep, raveling, collapse of containers and bridging components and
decomposition due to biodegradation of organic components. For simplicity, the
total settlement of a MSW landfill can be taken as the combination of primary and
secondary compression. Primary compression includes the physical compression of
components and consolidation. Secondary compression includes all creep effects and
those relating to degradation.
Knowledge of shear strength is required in order to assess waste slope stability. In
situ measurement of waste shear strength is at present not possible. Back-analysis of
failures provides the most reliable way of obtaining data, although this method is not
without difficulties due to problems obtaining adequate detailed field information.
Laboratory methods have been used widely but results from such studies should be
interpreted carefully due to their association with disturbed samples. Of the methods
available, the direct shear box produces the more reliable information. Although
various strength envelopes have been suggested for design, a conservative approach
should be taken due to the heterogeneity of the waste.
Information on the lateral stiffness of MSW is required to assess the performance
of steep side slope lining systems that rely in part on the waste for their stability and
integrity. The most comprehensive study to date has involved conducting
pressuremeter tests in both fresh and partly degraded MSW at a range of depths.
General trends of increasing stiffness with stress level are observed, and the older
waste (partly degraded) is shown to be less stiff than fresh waste (little degradation)
for the same stress level. Knowledge of in situ horizontal stresses is required to assess
lining component performance post waste placement. Obtaining representative
values is very difficult and this accounts for the small amount of information in the
literature. Knowledge of waste hydraulic conductivity is required in order to
understand leachate pressure distributions, and hence effective stresses, within the
waste body. Structure and stress dependency are the controlling factors.
Measuring and interpreting MSW engineering properties are extremely difficult
tasks. However, knowledge of unit weight, vertical compressibility, shear strength,
lateral stiffness, in situ stresses and hydraulic conductivity is fundamental to the
assessment of landfill stability and integrity of both geosynthetic and mineral lining
components. An internationally agreed classification system and test standards are
required to allow interpretation of published results. This will lead to development of
appropriate constitutive models for waste and hence to optimization of landfill
designs by considering waste/lining system interaction in full.
 ARTICLE IN PRESS
N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233 231

Acknowledgements

This paper has been based on part of a Golder Associates (UK) Ltd/
Loughborough University report entitled: ‘Stability of landfill lining systems:
Report No. 1 literature review, R&D Technical Report P1-385’ produced for the
England and Wales Environment Agency, dated January 2003. The support of Steve
Storey, Richard Moss and Darren Legge, who comprised the Environment Agency
steering committee for the R&D project, is gratefully acknowledged. Thanks are due
to Andrew Connell for his help in producing the report.

References

Beaven, R.P., Powrie, W., 1995. Hydrogeological and geotechnical properties of refuse using a large-scale
compression cell. Proceedings Sardinia ’95, Fifth International Landfill Symposium, Cagliari, Italy, pp.
744–760.
Brink, D., Day, P.W., Preez, Du, L., 1999. Failure and remediation of Bulbul Drive landfill: Kwazlu-natal,
South Africa. Proceedings Sardinia ’99, Seventh International Waste Management and Landfill
Symposium, Cagliari, Italy, pp. 555–562.
Byrne, R.J., 1994. Design issues with strain-softening interfaces in landfill liners. Proceedings Waste
Technology ’94, Charleston, South Carolina, USA, Session 4, Paper 4.
Council of European Community, 1999. Council Directive of 26 April 1999 on the landfill of waste (1999/
31/EC). Official Journal of the European Communities, L182/1.
Cowland, J.W., Tang, K.Y., Gabay, J., 1993. Density and strength properties of Hong Kong refuse.
Proceedings Sardinia ’93, Fourth International Landfill Symposium, Cagliari, Italy, pp. 1433–1446.
Del Greco, O., Oggeri, C., 1993. Geotechnical parameters of sanitary wastes. Proceedings Sardinia ’93,
Fourth International Landfill Symposium, Cagliari, Italy, pp. 1421–1431.
Dixon, N., Jones, D.R.V., 1998. Stress States In, and Stiffness Of, Landfill Waste. Geotechnical
Engineering of Landfills, Thomas Telford, London, pp. 19–34.
Dixon, H., Ng’ambi, S.C., Jones, D.R.V., Connell, A.K., 2000. The role of waste deformations on landfill
steep side wall lining stability. Proceedings Wastecon 2000, September, Cape Town, vol. 2,
pp. 379–388.
Dixon, N., Ng’ambi, S., Jones, D.R.V., 2004. Structural performance of a steep slope landfill lining
system. Proceedings of the Institution of Civil Engineers Geotechnical Engineering 157, 115–125.
Eid, H.T., Stark, T.D., Evans, W.D., Sherry, P.E., 2000. Municipal soil waste slope failure. I: waste and
foundation soil properties. Journal of Geotechnical and Geoenvironmental Engineering, ASCE,
397–407.
Fassett, J.B., Leonardo, G.A., Repetto, P.C., 1994. Geotechnical properties of municipal solid waste and
their use in landfill design. Waste Tech ’94, Landfill Technology Technical Proceedings, Charleston, SC
(USA), January 13–14.
Filz, G.M., Esterhuizen, J.J.B., Duncan, J.M., 2001. Progressive failure of lined waste impoundments.
Journal of Geotechnical and Geoenvironmental Engineering, ACSE, 841-848.
Gay, G., Kaiser, R., 1981. Bestimmung der scherfestigkeit an hausmüll in verschiedenen variationen.
Bericht der Forschungs- und Materialprüfungsanstalt Baden-Würtemberg, Stutgart.
Golder Associates, 1993. Unpublished internal report.
Gotteland, P., Gachet, C., Vuillemin, M., 2001. Mechanical study of a municipal solid waste landfill.
Proceedings Sardinia ’01, Eighth International Waste Management and Landfill Symposium, Cagliari,
Italy, pp. 425–433.
Gotteland, P., Gourc, J.P., Alboura, A., Thomas, S., 2002. On site determination of geomechanical
characteristics of waste. Proceedings GeoEng 2000, The Institute of Engineers, Australia, Published
on CD.
 ARTICLE IN PRESS
232 N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233

Gourc, J.P., Thomas, S., Vuillemin, M., 1998. Proposal of a waste settlement survey methodology.
Proceedings Environmental Geotechnics, Seco e Pinto (Ed.), Balkema, pp. 195–200.
Gourc, J.P., Olivier, F., Thomas, S., Chatelet, L., Denecheau, P., Munoz, M.L., 2001. Monitoring of
waste settlements on five landfills: comparison of the efficiency of different devices. Proceedings
Sardinia ’01, Eighth International Waste Management and Landfill Symposium, Cagliari, Italy,
pp. 515–524.
Grisolia, M., Napoleoni, Q., Tancredi, G., 1995a. Contribution to a technical classification of MSW.
Proceedings Sardinia ’95, Fifth International Landfill Symposium, Cagliari, Italy, pp. 761–767.
Grisolia, M., Napoleoni, Q., Tancredi, G., 1995b. The use of triaxial tests for the mechanical
characterization of MSW. Proceedings Sardinia ’95, Fifth International Landfill Symposium, Cagliari,
Italy, pp. 703–710.
Hendron, D.M., Fernandez, G., Prommer, P.J., Giroud, J.P., Orozco, L.F., 1999. Investigation of 27th
September 1997 slope failure at the Dona Juana landfill. Proceedings Sardinia ’99, Seventh
International Waste Management and Landfill Symposium, Cagliari, Italy, pp. 545–554.
Jessberger, H.L., 1994. Geotechnical aspects of landfill design and construction, Part 2: materials
parameters and test methods. Institution of Civil Engineers Geotechnical Engineering Journal 107,
105–113.
Jessberger, H.L., Kockel, R., 1993. Determination and assessment of the mechanical properties of waste
materials. Proceedings Sardinia ’93, Fourth International Landfill Symposium, Cagliari, Italy,
pp. 167–177.
Jones, D.R.V., Dixon, N., 2003. Stability of landfill lining systems: report No. 1 literature review. R&D
Technical Report P1-385, Environment Agency, UK, p. 219.
Jones, D.R.V., Dixon, N., 2005. Landfill lining stability and integrity: the role of waste settlement.
Geotextiles & Geomembranes 23 (1), 27–53.
Jones, D.R.V., Taylor, D.P., Dixon, N., 1997. Shear strength of waste and its use in landfill stability
analysis. In: Yong, R.N., Thomas, H.R. (Eds.), Proceedings Geoenvironmental Engineering
Conference. Thomas Telford, pp. 343–350.
Kavazanjian, E., 2001. Mechanical properties of municipal solid waste. Proceedings Sardinia ’01, Eighth
International Waste Management and Landfill Symposium, Cagliari, Italy, pp. 415–424.
Kavazanjian, E., 2003. Evaluation of MSW properties using field measurements. In: Koerner, R.M.,
Koerner, G.R., Hsuan, Y., Ashley, M.V. (Eds.), Proceedings 17th Geosynthetic Research Institute
Conference, Hot Topics in Geosynthetics—IV, pp. 74–113.
Kavazanjian, E., Matasovic, R., Bachus, R.C., 1999. Large-diameter static and cyclic laboratory testing of
municipal solid waste. Proceedings Sardinia ’99, Seventh International Waste Management and
Landfill Symposium, Cagliari, Italy, pp. 437–444.
Kavazanjian, N., Matasovic, R., Bonaparte, G.R., Schmertmazin, E., 1995. Evaluation of MSW
properties for seismic analysis. Geoenvironment 2000, Geotechnical Special Publication No. 46,
ASCE, pp. 1126–1141.
Koerner, R.M., Soong, T.-Y., 2000. Stability assessment of ten large landfill failures. Advances in
Transportation and Geoenvironmental Systems using Geosynthetics, Proceedings of Sessions of Geo
Denver 2000, ASCE Geotechnical Special Publication No. 103, pp. 1–38.
Kolsch, F., 1995. Material values for some mechanical properties of domestic waste. Proceedings Sardinia
’95, Fifth International Landfill Symposium, Sardinia, vol. 2, p. 20.
Kölsch, F., 1996. Einfluss der Faserbestandteile auf die Scherfestigkeit von Siedlungsabfall (The influence
of fibrous constituents on shear strength of municipal solid waste). Ph.D. Thesis, Leichtweiss-Institut
für Wasserbau der TU Braunschweig, Germany.
Landva, A., Clark, J.I., 1986. Geotechnical testing of waste fill. Proceedings of Canadian geotechnical
Conference, Ottawa, Ontario, pp. 371–385.
Landva, A., Clark, J.I., 1990. Geotechnics of waste fills. Geotechnics of Waste Fills—Theory and Practice,
ASTM STP 1070, pp. 86–106.
Landva, A.O., Valsangkar, A.J., Pelkey, S.G., 2000. Lateral earth pressure at rest and compressibility of
municipal solid waste. Canadian Geotechnical Journal 37, 1157–1165.
 ARTICLE IN PRESS
N. Dixon, D.R.V. Jones / Geotextiles and Geomembranes 23 (2005) 205–233 233

Langer, U., Dixon, N., 2004. Mechanical properties of MSW: development of a classification system.
In: Yong, R.N., Thomas, H.R. (Eds.), Proceedings Fourth British Geotechnical Association
Geoenvironmental Engineering Conference. Integrated Management of Groundwater and Contami-
nated Land, pp. 267–274.
Long, J.H., Gilbert, R.B., Daly, J.J., 1995. Effect of waste settlement on sloped lining systems. Proceedings
of Geosynthetics ’95, Nashville, Tennessee, USA, February, vol. 2, pp. 729–744.
Manassero, M., Van Impe, W.F., Bouazza, A., 1996. Waste disposal and containment. Proceedings
Second International Congress on Environmental Geotechnics, Osaka, vol. 3, pp. 193–242.
McDougall, J.R., Pyrah, I.C., 2001. Settlement in landfilled waste: extending the geotechnical approach.
Proceedings Sardinia ’01, Eighth International Waste Management and Landfill Symposium, Cagliari,
Italy, pp. 481–490.
Oweis, I.S., Khera, R., 1986. Criteria for geotechnical construction on sanitary landfills. In: Fang, H.Y.
(Ed.), International Symposium on Environmental Geotechnology, Lehigh University Press,
Bethlehem, PA, vol. 1, pp. 205–222.
Oweis, I.S., Khera, R., 1998. Geotechnology of Waste Management, second ed. PWS Publishing
Company.
Powrie, W., Beaven, R.P., 1999. Hydraulic properties of household waste and implications for landfills.
Institution of Civil Engineers Geotechnical Engineering Journal 137, 235–247.
Powrie, W., Richards, D.J., Beaven, R.P., 1998. Compression of waste and implications for practice.
Geotechnical Engineering of Landfills, Thomas Telford, 3–18.
Qian, X., Koerner, R.M., Gray, D.H., 2002. Geotechnical Aspects of Landfill Design and Construction.
Prentice-Hall, Englewood Cliffs, NJ.
Seed, R.B., Mitchell, J.K., Seed, H.B., 1988. Slope stability failure investigation: Landfill Unit B-19, Phase
1-A, Kettleman Hills, California. Report No. UCB/GT/88-01. Department of Civil Engineering,
University of California, Berkley, July.
Siegel, R.A., Robertson, R.J., Anderson, D.G., 1990. Slope Stability Investigations at a Landfill in
Southern California. Geotechnics of Waste Fill—Theory and Practice; ASTM STP 1070.
Singh, S., Murphy, B., 1990. Evaluation of the stability of sanitary landfills. Geotechnics of Waste Fills—
Theory and Practice, ASTM STP 1070, pp. 240–258.
Spillmann, P., 1980. Beitrag zur langzeitig standsicheren konstruktion hoher abfalldeponien, Teil 1 und 2,
Müll und Abfall 12, pp. 311–317, 337–339.
Stark, T.D., Eid, H.T., Evans, W.D., Sherry, P.E., 2000. Municipal soil waste slope failure. II: stability
analyses. Journal of Geotechnical and Geoenvironmental Engineering, ASCE, pp. 408–419.
Thomas, S., Aboura, A.A., Gourc, J.P., Gotteland, P., Billard, H., Delineau, T., Gisbert, T., Ouvry, J.F.,
Vuillemin, M., 1999. An in-situ waste mechanical experimentation on a French landfill. Proceedings
Sardinia ’99, Seventh International Waste Management and Landfill Symposium, Cagliari, Italy, pp.
445–452.
Turczynski, U., 1988. Geotechnische Aspekte beim Aufbau von Mehrkomponentendeponien. Ph.D.
Thesis, Fakultät der technischen Wissenschaften, Bergakademie Freiberg, Germany.
Watts, K.S., Charles, J.A., 1990. Settlement of recently placed domestic refuse landfills. Proceedings
Institution of Civil Engineers, Part 1 88, 971–993.
Watts, K.S., Charles, J.A., Blaken, N.J.R., 2002. Settlement of landfills: measurements and their
significance. Waste 2002, Integrated Waste Management and Pollution Control: Research, Policy and
Practice, pp. 673–682.