Sustainable Practices For Landfill Design and Operation
Sustainable Practices For Landfill Design and Operation
Sustainable Practices For Landfill Design and Operation
Sustainable
Practices
for Landfill Design
and Operation
Waste Management Principles and Practice
Series editor
Timothy G. Townsend
Sustainable Practices
for Landfill Design
and Operation
Timothy G. Townsend Jon Powell
Department of Environmental Engineering Innovative Waste Consulting Services, LLC
Sciences Gainesville, FL, USA
Engineering School of Sustainable
Infrastructure and Environment Qiyong Xu
University of Florida Peking University Shenzhen
Gainesville, FL, USA Graduate School
Shenzhen, China
Pradeep Jain
Innovative Waste Consulting Services, LLC Debra Reinhart
Gainesville, FL, USA Department of Civil, Environmental
and Construction Engineering
Thabet Tolaymat University of Central Florida
U.S. Environmental Protection Agency Orlando, FL, USA
Cincinnati, OH, USA
Springer Science+Business Media LLC New York is part of Springer Science+Business Media
(www.springer.com)
Preface
This book was developed for waste and materials management practitioners all over
the world, including researchers, practicing engineers and scientists, municipality
staff and management, landfill operators, and regulatory agencies. The body of
work presented here results from the development of bioreactor landfill design
guidelines for the US Environmental Protection Agency’s Office of Research and
Development, along with the combined knowledge and experience of the authors
pertaining to sustainable practices for design and operation of sanitary landfills. We
presume the reader has a basic understanding of landfills, although the initial chap-
ters attempt to set the stage by providing introductory commentary and a discussion
of fundamental landfill concepts (both traditional and sustainable).
Following the introductory Chaps. 1–3, we provide a series of case studies that
highlight the state of the practice of sustainable landfilling throughout the world.
We sequenced the book so that readers could obtain a practical view of historic and
current practices at operating facilities and how approaches to sustainable landfill-
ing can differ from one location to another. Subsequent chapters are broken up to
present discrete, focused discussion on the various infrastructure components,
design practices, operational considerations, and monitoring elements that pro-
mote the more sustainable use of landfills as a component of integrated solid waste
management systems.
This book was not written from a position of advocacy. Although the idea of
accelerating decomposition in landfills has been around for decades, we felt that the
opportunity to present the current state of science, including benefits and concerns,
as well as current limitations and uncertainties, was appropriate, particularly in light
of the significant amount of research and full-scale operational experience in the last
20 years. Furthermore, this text was not intended to be a rigorous design manual
sufficient to completely design landfill-integrating sustainable technologies. Rather,
the book was developed to serve as a tool for designers, regulators, and other parties
interested in sustainable landfill practices and to be used in conjunction with funda-
mental design methodologies, location-specific regulations, new and emerging
research results, and good engineering judgment. Dozens of graphs, figures, and
tables provided throughout the text provide the designer an excellent foundation to
v
vi Preface
begin their analysis and apply the principles from this book to their site or facility.
In like fashion, operational experiences are provided throughout, tying in the impor-
tant underlying fundamental concepts (e.g., accelerated gas production after liquids
addition) to critical operational considerations (e.g., how to effectively collect the
additional gas that is produced).
This book was designed to be assimilated by the reader in two ways. First, for the
novice on the topic of sustainable landfills, a back-to-front reading through the
chapters in sequence will provide an excellent background on sustainable landfill-
ing practices since the chapters are presented in a progressive order; planning con-
siderations are followed by detailed design and operational considerations, which
are in turn followed by end-of-facility-life considerations. Second, for the more
advanced reader, individual chapters may be examined with enough context so that
the reader can apply the information presented in the book to their particular prob-
lem without heavy reliance on previous or subsequent chapters.
The efforts and contributions of numerous people that helped to make this book a
reality must be acknowledged. At the University of Florida, Shaun Alvarado, Jim
Wally, and Longsheng Jiang helped tremendously in organizing the chapters, refer-
ences, and figures. Dr. Hwidong Kim provided valuable expertise in the develop-
ment of many of the book’s figures. At Innovative Waste Consulting Services in
Gainesville, FL, Justin Smith, Lizmarie Maldonado, Ali Bigger, and Saraya Pleasant
all gave their time to conduct analysis, provide editing, and address miscellaneous
formatting requests. Steve Laux at Jones Edmunds and Associates provided excel-
lent practical insight on multiple chapters. Photos and other case study information
were generously provided by Fred Doran at SAIC; Ramin Yazdani at Yolo County,
CA; Marco Ritzkowski at Hamburg University of Technology; Professor Matsufuji
of Fukuoka University in Japan; Scott Merry of the University of the Pacific; Jones
Edmunds and Associates, Gainesville, Florida; and Waste Management Inc.
A special acknowledgement must be given to the numerous collaborators,
researchers, and funding agencies that allowed for such significant development of
our understanding of sustainable landfilling and the benefits and challenges that
sustainable landfilling can provide. Much of the work presented in this book comes
as a result of funding from the Hinkley Center for Solid and Hazardous Waste
Management, the Florida Department of Environmental Protection, the New River
Solid Waste Association, the Alachua County Public Works Department, and the
Polk County Waste & Recycling Division.
Finally, the authors also wish to acknowledge each of their families for tolerating
the many long nights and early mornings spent developing this book.
vii
Contents
ix
x Contents
xvii
xviii Abbreviations, Acronyms, and Initialisms
Over the past 50 years, much of the world has witnessed a remarkable evolution in
the management of municipal solid waste (MSW; the garbage and refuse resulting
from household, commercial, and institutional activities), from uncontrolled dump-
ing on the land and indiscriminate burning, to integrated systems incorporating
waste processing, recycling, and treatment. This progress parallels society’s grow-
ing awareness of the need to protect human health and the environment, and the
importance of resource and energy conservation. Governments, businesses, and
individuals now recognize, and in many cases embrace, the adoption of sustainable
practices in many aspects of daily life, including the management of solid waste.
While many definitions have been proposed, sustainability can be broadly defined
as the ability to meet current societal needs without compromising the anticipated
needs of future generations. The US Environmental Protection Agency further
describes sustainability as follows: “Everything that we need for our survival and well-
being depends, either directly or indirectly, on our natural environment. Sustainability
creates and maintains the conditions under which humans and nature can exist in pro-
ductive harmony, which permits fulfilling the social, economic and other requirements
of present and future generations” (US EPA 2008). Meeting present and future envi-
ronmental, social and economic demands constitute the three pillars of sustainability.
The terms “landfill” and “sustainability,” as linked together in the title of this
book, may suggest a contradiction to some, as landfills—places set aside for the
Fig. 1.1 Estimated MSW management in (a) middle and low income countries (total of 195 mil-
lion Mg) and (b) high income countries (total of 588 million Mg); Source Hoornweg and Bhada-
Tata (2012)
at dumps throughout the developing world, and the environmental and human health
challenges they present.
Economic realities in many nations result in a large human presence at landfill
sites, scavengers who are not officially associated with the daily operation of waste
disposal. People, often including young children, sort through incoming waste for
recovery of salable materials. It is not uncommon for waste scavengers and their
families to live on or adjacent to the landfill itself. Potential immediate health risks
include those posed by working in close proximity to waste vehicles and heavy
equipment, exposure to harmful materials or chemicals, exposure to disease vec-
tors, and explosions or fires that can occur because of gases produced from the
decomposition process or incoming reactive wastes. In some cases, waste slides
(slope failures) have occurred, burying and killing scavengers and their families.
Pollution of water and air resources commonly results from uncontrolled land-
filling of waste. Leachate is the term used to describe the liquid resulting from water
coming into contact with waste. Chemicals disposed of in the waste or byproducts
from reactions in the landfill, dissolve (leach) into the water, and when this leachate
emerges from the waste and, enters groundwater or a surface water stream, a risk is
posed to those consuming or coming in contact with the affected water resource.
Gases and particulate matter can also be released to the environment. Gases pro-
duced from the waste decomposition process, primarily methane, pose a potential
risk of explosions and fires, and also act as a carrying mechanism for other chemi-
cals in the landfilled waste, many of which may be toxic to humans. Particulates can
be released from landfill fires or as dust disturbed as part of landfill operations.
Uncontrolled landfilling can pose a threat to ecological resources. Surface water
resources contaminated as a result of waste disposal often have reduced dissolved
oxygen levels, thus diminishing the ecological health of the water body and poten-
tially resulting in the growth and spread of disease-carrying organisms. Without
forethought in appropriate locations for landfills, important ecologic areas are
destroyed as a result of waste disposal. A common example is the filling of wetlands
as means of reclaiming land. Lastly, indiscriminant disposal of waste through land
disposal represents a less than desirable practice from a materials and resource man-
agement perspective. Recovery of materials does take place by those sequencing the
waste stream, but much more material recovery potential remains buried in the land-
fill, both in terms of resources and energy.
The first step in the evolution of modern landfills from uncontrolled dumps was the
development of sanitary landfill practices designed to address immediate human
health concerns. The implementation of sanitary landfilling involves several
changes to operational practices that focus on minimizing the spread of disease
and the occurrence of landfill fires. The placement of waste into defined cells,
often constructed in distinct units and compacted in place with heavy equipment
6 1 The Landfill’s Role in Sustainable Waste Management
Fig. 1.5 Waste compaction in organized cells is a fundamental component of sanitary landfill
operation
Fig. 1.6 Cover soil application at sanitary landfills aids in reducing odor, vectors, fires and helps
in the control of storm water and leachate
(see Fig. 1.5), allows more contained and controlled disposal. A critical element in
sanitary landfill operation is the routine placement of cover soil on top of recently
placed waste (see Fig. 1.6) to minimize fires, odors and disease vectors. Another
key sanitary landfill feature includes site access control, which helps to discourage
1.3 The Evolution of Modern Landfills 7
Fig. 1.7 A barrier layer being placed as part of the construction of a landfill liner and leachate
collection system
waste scavenging and properly define the facility’s boundary through fencing or
similar means.
While the evolution of sanitary landfill practices reduced many of the direct
human health concerns associated with open dumps, it did not address the two
major pollutant emissions associated with landfilled MSW: leachate and landfill gas
(LFG). As regulators and scientists began to monitor groundwater quality surround-
ing landfills, the body of evidence indicating that leachate negatively affected
groundwater quality grew (Sawney and Kozloski 1984; Reinhard et al. 1984;
Schultz and Kjeldsen 1986). This resulted in many governments requiring MSW
landfill construction to include barrier layers for preventing leachate migration out
of the landfill (see Fig. 1.7) and drainage systems allowing the removal of accumu-
lated leachate for treatment before disposal. Many of these technical requirements
followed those previously developed for the management of hazardous wastes, a
regulatory system designed upon the principle of cradle-to-grave management of
wastes that posed an increased risk to human health and the environment. In lined
landfill systems, leachate is removed from the landfill and treated prior to its return
to the environment. Groundwater surrounding the lined landfill is monitored to
assess whether the containment system functions properly.
Early LFG concerns focused largely on controlling subsurface migration into
adjacent buildings and enclosed spaces, where methane produced from anaerobic
waste decomposition could result in explosive conditions. This concern was partly
addressed through the requirement of a bottom liner, and was often accompanied by
soil vapor monitoring probes surrounding the lined landfill to assess gas migration.
Another early gas concern arose from locations with regional air pollution concerns
(e.g., California), and ultimately these and other issues with LFG (odor, toxic
8 1 The Landfill’s Role in Sustainable Waste Management
Fig. 1.8 In foreground, a gas collection well is used to extract and control landfill gas for older
waste, while in the background, new waste disposal continues
1.5 P
ractices and Technologies for More
Sustainable Landfilling
The evolution of landfills from disposal systems to treatment systems through land-
fill bioreactor and similar technologies, while motivated by multiple drivers, repre-
sents a major step toward more sustainable landfilling. The objective of such
10 1 The Landfill’s Role in Sustainable Waste Management
Fig. 1.9 Internal combustion engine producing electricity from landfill gas
1.6 Scope and Organization of Book 11
Fig. 1.10 Landfill mining allows reclamation of metals, soil, and degraded organic matter
Additional sustainable landfill practices address how to best manage the landfill
after waste disposal operations have ceased. Some operators, for example, add air
to the landfill as a method of providing a final aerobic curing step and thus minimize
future environmental emissions. The practice of reclaiming stabilized landfills
through mining has been proposed; through this practice, resources can be recov-
ered and the land requirement for future disposal is reduced (see Fig. 1.10).
Technologies for utilizing landfill space for additional purposes, ranging from
human enjoyment to recovery of solar and wind energy, also fall within the scope of
technologies consistent with sustainable landfill practices.
This book provides design and operation guidance for engineers and operators to
implement sustainable landfilling practices at their facility. Methods for promoting
rapid waste stabilization in a safe and controlled manner are a major focus. The
methods described apply to facilities operated as landfill bioreactors, sites practic-
ing leachate recirculation but not operated as bioreactors, landfills with high waste
moisture content, and owners and operators that desire to collect LFG as efficiently
as possible.
This book begins with an introduction of waste and landfill fundamentals, includ-
ing a detailed discussion of the practice of landfilling and bioreactor landfills.
Planning considerations for implementing more sustainable landfills, along with a
12 1 The Landfill’s Role in Sustainable Waste Management
review of the state of the practice of such facilities, are presented. Next, the focus
turns toward liquids addition systems and other liquids management considerations.
Later chapters of the book address concerns such as slope stability, LFG, and
operation. The book ends with a discussion of final landfill disposition, the econom-
ics of sustainable landfilling practices, and the integration of other sustainable com-
ponents such as landfill reclamation and energy recovery at integrated solid waste
management facilities including a landfill as a component.
References
Buivid MG, Wise DL, Blanchet MJ (1981) Fuel gas enhancement by controlled landfilling of
municipal solid waste. J Resour Conserv 6:3–20
Hoornweg D, Bhada-Tata P (2012) What a waste: a global review of solid waste management.
Urban development series knowledge papers no. 15. The World Bank, Washington, DC
Leckie JO, Pacey JG, Halvadakis C (1979) Landfill management with moisture control. J Environ
Eng-ASCE 105:337–355
McDonough W, Braungart M (2003) Cradle to cradle: remaking the way we make things, 1st edn.
North Point, New York
Pohland FG (1980) Leachate recycle as landfill management option. J Environ Eng-ASCE
106:1057–1069
Reinhard M, Goodman NL, Barker JF (1984) Occurrence and distribution of organic chemicals in
two landfill plumes. Environ Sci Technol 18:953–961
Reinhart DR, Townsend TG (1997) Landfill bioreactor design and operation, 1st edn. CRC, Boca
Raton
Reinhart DR, McCreanor PT, Townsend TG (2002) The bioreactor landfill: its status and future.
Waste Manage Res 20(2):172–186
Reinhart DR, Amini H, Bolyard SC (2012) The role of landfills in us sustainable waste manage-
ment. Environmental Engineer: Applied Research and Practice, vol 15, Winter 2012
Sawney BL, Kozloski RP (1984) Organic pollutants in leachate from landfill sites. J Environ Qual
13(3):349–352
Schultz B, Kjeldsen P (1986) Screening of organic matter in leachates from sanitary landfills using
gas chromatography and mass spectrometry. Water Res 20(8):965–970
US EPA (2008) Municipal solid waste generation, recycling, and disposal in the United States:
facts and figures for 2008. US EPA-530-F-009-021. United States Environmental Protection
Agency, Washington, DC
Chapter 2
Waste and Landfill Fundamentals
Fig. 2.2 Typical MSW composition in Chinese cities (Based on composition studies from 12 cit-
ies reported in Zhang et al. 2010)
tion rates strongly correlate with income level (Hoornweg and Bhada-Tata 2012),
with greater average generation rates occurring in high income nations (2.1 kg/capita-
day) compared to upper middle, lower middle, and lower income nations (1.2, 0.8 and
0.6 kg/capita-day, respectively). Figure 2.1 presents estimated MSW composition in
the US in 2010 (US EPA 2011). The largest component of the US MSW stream is
paper (29 %), with yard trimmings (13 %), food scraps (14 %), and plastics (12 %)
also contributing heavily. This composition is representative of much of the devel-
oped world, with an abundance of packaged products and a greater quantity of dis-
carded goods. This differs from many parts of the world, as indicated in Fig. 2.2,
2.2 Landfill Components 15
Fig. 2.3 Global MSW composition (a) High income countries (b) Lower income countries
(Hoornweg and Bhada-Tata 2012)
which presents the typical composition of MSW in major Chinese cities (Zhang et al.
2010). Organic waste is the dominant component (58 %), with paper a much smaller
contributor (9 %).
Waste composition is a critical factor when considering sustainable landfilling
practices, as many of the potential problems with MSW landfills (e.g., water
pollution potential, atmospheric emissions) result from the dominance of biodegrad-
able materials in the waste. In higher income nations paper dominates, with appre-
ciable amounts of food waste and plant trimmings also contributing (Fig. 2.3a). In
lower income nations, food scraps and related organic materials dominate (Fig. 2.3b).
Landfills managing either waste stream require sustainable practices to promote safe
waste stabilization and control of emissions, although the manner in which some of
this control will be achieved may differ. For example, landfills dominated by greater
amounts of paper will need more liquids added to encourage stabilization, while in
landfills dominated by food waste, sufficient moisture may already exist.
the later sections of this book, several major landfill components are discussed
below, including liners, systems for leachate and gas management, and landfill man-
agement after waste disposal has stopped. Figure 2.4 provides an overview illustra-
tion of major landfill components.
Fig. 2.6 (a) Construction of compacted earthen liner. (b) Geomembrane liner panels. (c) Thermal
fusion welding of geomembrane panels. (d) Thermal extrusion welding of geomembrane (Photo
courtesy of Jones Edmunds)
Fig. 2.6 (continued)
20 2 Waste and Landfill Fundamentals
where overlapping straight lengths of geomembranes is not available and when the
geomembranes are attached to other plastic components (Fig. 2.6d). Rigorous test-
ing of liner materials and seams must be included as part of geomembrane liner
construction.
Fig. 2.7 (a) Rounded stone used as LCRS drainage material. (b) Sand placement for LCRS.
(c) Geonet installation in LCRS. (d) Leachate pump station
22 2 Waste and Landfill Fundamentals
Fig. 2.7 (continued)
Given the large amount of highly degradable organic matter in most MSW landfills,
this material decomposes soon after waste disposal, which results in the production
of biogas—a more detailed discussion of this process is provided in Sect. 2.5.
2.2 Landfill Components 23
Under the anaerobic conditions that normally develop in landfills (due to the com-
bination of compacting and covering the waste, and lining the bottom), large fractions
of components such as food waste, paper, and yard trash are biologically decomposed
to a gas that consists primarily of methane and carbon dioxide. The extent and rate at
which this conversion takes placed is dictated by the waste type (e.g., the amount of
food waste versus the amount of paper) and conditions such as moisture content, pH,
and temperature. A focus of this book is on controlling the conversion process, but
these reactions occur in all MSW landfills, and thus a common design component of
many modern landfills is a gas collection and control system (GCCS).
The primary driving force for gas produced within an MSW landfill to migrate
from the disposal unit is pressure. As gas is produced in the constrained volumes
within the waste, pressure builds and the gas moves toward lower pressures outside
the landfill. Thus, basic elements of most GCCS are extraction points that provide
controlled pathways for gas escape from the landfill. These extraction points are
most commonly vertical wells within the waste, though other configurations have
been used. At some facilities, wells are naturally vented to the atmosphere (and pos-
sibly a flare), but when maximum gas recovery efficiency is desired, the wells are
tied together using a series of connected manifold pipes, and this piping network is
connected to mechanical blowers or fans to induce a vacuum in the well-field. The
combined gas is then either flared or utilized in some beneficial fashion.
An important element of the GCCS is the extraction points where the operator
has the ability to control the degree of vacuum placed at a given location. Figure 2.8
illustrates a typical gas extraction wellhead, which includes the gas well penetration
through the surface of the landfill; a wellhead to allow measurement of flow, pres-
sure, and temperature; a control valve for adjusting pressure and flow; a flexible
connection to the gas manifold; and appropriate connections to the surface cap that
minimize air entrance into the landfill. Another important design and operation con-
sideration for a GCCS is the management of condensate that forms in the pipes; this
liquid must be appropriately removed or else it will interfere with gas transmission.
Since gas condensate typically includes dissolved chemicals such as volatile organic
compounds (VOCs) that may have deleterious health or environmental impacts, it is
normally managed in a similar fashion to landfill leachate.
Landfill operation not only includes the daily activities associated with the place-
ment of waste in the landfill, it includes the execution of a variety of specialized
tasks such as those related to leachate management and gas extraction (Bolton
1995). A modern landfill site will include a number of elements beyond the disposal
unit, including a scale-house for weighing incoming materials, a system of roads for
routing trucks to and from the waste disposal area, and facilities for employees and
maintenance of equipment and vehicles. Other large areas may be devoted to sur-
face water management systems, cover soil excavation and processing, and buffers
from neighboring property. Many landfill sites also house other dedicated waste
management operations such as yard trash processing, composting, recycling, and
storage of appliances, tires, or other bulky material. In short, landfill facility opera-
tion is a multi-faceted endeavor.
Trucks carrying waste that enter the site for disposal are first weighed using
scales and appropriate information is recorded for billing (Fig. 2.10a). Some land-
fills may simply have a receiving area where truck counts or truck load volume is
recorded in lieu of scales. Waste vehicles are directed to dedicated disposal areas
within the waste containment unit, commonly known as the working or active face
(Fig. 2.10b). As waste vehicles unload their contents, landfill employees using a
variety of equipment push the material to the desired location and compact the
waste. Most landfill operators utilize large steel-wheeled compactors designed to
maximize density after three or four passes over a layer of waste, (typically less than
1 m). As waste is unloaded from collection and transport vehicles, “spotters” exam-
ine the waste for improper materials; this is especially important for the first lift of
waste placed in a new landfill to exclude materials that pose a puncture risk to the
liner (Fig. 2.10c). Cover soil is hauled to the working face and then placed over a
finished lift of waste by the end of the working day. In some cases, alternative cover
materials to soil will be used, included mulch, tarps or foam.
Waste placement and compaction follows a predetermined filling sequence
designed to fill the containment area in an organized manner than meets desired site
objectives (e.g., slopes for stormwater control, placement of internal hauling roads).
Strategic waste filling results in a final landfill configuration that meets designed
targets for elevation, side slopes, stormwater control structures, and grading of the
landfill top deck.
26 2 Waste and Landfill Fundamentals
Fig. 2.10 (a) Trucks carrying waste that enter the site for disposal are first weighed using scales
and appropriate information is recorded for billing. (b) Waste vehicles directed to dedicated dis-
posal areas within the waste containment unit. (c) Spotting incoming waste. (he ) Monitoring the
gas system
2.3 Landfill Operation 27
Fig. 2.10 (continued)
In addition to waste tipping, compaction, and soil placement, the landfill opera-
tor is responsible for other operational features of the site such as operation and
maintenance of the leachate removal and gas control systems. Leachate system
operations includes ensuring proper operation of pumps, providing for appropriate
maintenance, recording system data, and any operational needs of the leachate treat-
ment and discharge components. In a similar manner, mechanical landfill gas
extraction blowers must be maintained and the well field must be appropriately bal-
anced to ensure efficient collection and to minimize possible risk of landfill fires
(Fig. 2.10d).
28 2 Waste and Landfill Fundamentals
Landfill operation does not end at closure. Throughout the life of the facility and
after closure, groundwater and soil vapor samples must be collected and analyzed to
meet regulatory permit requirements. Leachate collection and gas collection sys-
tems must be maintained. Post closure refers to the period following closure when
necessary operation and monitoring of the landfill continues. Regulations typically
mandate a minimum period of post-closure care; in the US, this period is 30 years.
In addition to necessary monitoring, post-closure activities include operating the
leachate and gas systems, maintaining the landfill cap and related features, and
ensuring the integrity of all critical site features.
The importance of biological activity in landfilled MSW has been discussed multi-
ple times in the introductory material provided thus far. Given that a major element
of sustainable landfill operations is the control of the waste stabilization process and
the byproducts resulting from it, as part of a discussion of landfill fundamentals, it
is useful to describe the process in greater detail. Several researchers have provided
descriptions of a progression of phases that a landfill will undergo after waste place-
ment (Senior 1995; Palmisano and Barlaz 1996), with descriptions of changes in
leachate and gas composition that result from each phase. A generalized depiction
of these landfill phases is presented in Fig. 2.11.
Once waste is landfilled, the void spaces within the waste mostly contain air, and
thus the initial phase of waste decomposition is often described as aerobic.
Placement of daily cover, additional waste, and waste compaction may limit oxy-
gen transfer, resulting in the termination of aerobic decomposition within a short
period of time. For this reason, the portion of waste decomposed under aerobic
conditions is relatively small with respect to the entire landfill stabilization process.
The major gas components observed in the aerobic phase are oxygen, nitrogen
(entrapped from the atmosphere) and carbon dioxide generated as a byproduct of
aerobic decomposition.
As the oxygen trapped within the waste is depleted, the landfill conditions may
change to anaerobic. With a lack of oxygen, waste may be decomposed by the bac-
teria that can use nitrate and sulfate (rather than oxygen) as an electron acceptor. In
the acid phase, hydrolysis of macromolecules such as cellulose and protein enhances
organic acid production and results in a decrease in pH. Although these organic
acids can be consumed by methanogenic microorganisms, a great amount of organic
acid can be accumulated due to the low growth rate of methanogens in comparison
with the growth of acid formers. These cumulative organic acids and CO2 (a byprod-
uct of waste degradation) can depress the pH of the landfill. In addition, hydrogen
gas can be produced as a byproduct of the degradation of butyric and propionic acid.
Figure 2.12 provides an overview of anaerobic waste stabilization microbiology
As the redox potential of a landfill decreases, the growth of methanogenic
microorganisms increases. Organic acid and hydrogen gas produced from waste
2.4 Waste Stabilization Processes 29
Table 2.1 Major leachate quality classes and changes during stabilization
Leachate
constituent Changes with stabilization
Organic In the early phases of landfill stabilization, the concentration of organic matter is
matter largely a result of the volatile fatty acids and other easily biodegradable
chemicals. As the landfill progresses into an active methane-forming phase, most
of the easily degradable organic matter is consumed within the landfill, and
concentrations decrease. As activity progresses toward stabilization, leachate
organic matter becomes dominated by large molecular weight chemicals that are
recalcitrant to biodegradation
Inorganic As the landfill ages, the ionic concentration tends to increase as leachate becomes
ions less influenced by rainwater dilution and more wastes become exposed to moisture.
Many inorganic ions such a chloride and sodium will be conserved in the system so
when leachate is recirculated, concentrations will increase with time. Eventually, as
more moisture flushes through the landfill, concentrations will decrease
Nutrients Ammonium will exist at the dominant nutrient chemical and will behave in a
similar nonreactive manner as other inorganic ions as the long as the environment
remains anaerobic. At the points when air enters the landfill again, some of the
ammonia may be biologically transformed to other nitrogen species
Trace Trace pollutant concentrations are often sufficiently low that trends will be hard to
chemicals observe, but the long-term trend with stabilization will be chemical specific.
Some chemical constituents may biodegrade and others may be entrained with the
waste (e.g., sorption, precipitation). Other trace elements will behave similar to
inorganic ions
oxygen, can cause waste combustion within the landfill and the formation c onditions
leading to a fire at or near the landfill surface.
As the waste decomposes, whether under anaerobic or aerobic conditions, the
volume occupied by waste decreases resulting in a recovery of landfill air space
may be realized (estimates of a 15–30 % gain in landfill air space upon stabilization
are common). However, the additional disposal capacity is only gained if the landfill
operator structured the filling sequence to utilize the recovered air space. If a landfill
is operated as a bioreactor after a final cap has been placed and no additional waste
is added, the air space likely will not be re-gained.
With accelerated waste decomposition, where primarily organic waste decom-
poses, the LFG generation rate increases. Therefore, in bioreactor landfills and
similar facilities, gas generation rates are much higher than in conventional land-
fills; consequently LFG can potentially be recovered and used economically. If not
properly controlled through design and operation of a LFG collection system, the
enhanced LFG production rates may result in increased emissions to the environ-
ment. Under anaerobic conditions, both methane (CH4) and carbon dioxide (CO2)
are generated, while under aerobic conditions nitrogen gas (N2) and CO2 dominate.
A gas extraction system can be utilized within the life of the landfill and for years
after closure, to collect and control the landfill, including potential conversion to
energy. Since bioreactor landfills increase the rate of LFG generation, the increased
quantity in a shorter time period can improve the practicality of the beneficial use of
the gas (e.g., electricity generation).
Ultimately, a properly designed, operated, and maintained bioreactor landfill, or
a facility operated in a similar manner that enhances waste decomposition, potential
for offers considerable reductions in environmental impacts relative to conventional
landfills. The waste is stabilized over a reduced timespan, when the landfill is still
being monitored and when the landfill infrastructure is in its best condition.
A means of leachate management can be provided, additional air space can be
gained (potentially decreasing the necessity to construct a new landfill), and the
viability of collecting and beneficially using the LFG is increased. However, this is
only feasible if the bioreactor landfill is properly designed, operated, and main-
tained. Most of the rest of this book focuses on the technologies that can be used to
meet such objectives.
References
Bolton N (1995) The handbook of landfill operations. Blue Ridge Services, Atascadero, CA
Buivid MG, Wise DL, Blanchet MJ (1981) Fuel gas enhancement by controlled landfilling of
municipal solid waste. J Resour Conserv 6:3–20
El-Fadel M (1999) Leachate recirculation effects on settlement and biodegradation rates in MSW
landfills. Environ Technol 20:121–133
Hoornweg D, Bhada-Tata P (2012) What a waste: a global review of solid waste management.
World Bank urban development series; knowledge paper no. 15, Washington, DC
34 2 Waste and Landfill Fundamentals
Knox K, De Rome L, Caine M, Blakely NC (1999) Observation from a review of the Brogborough
and landfill 2000 test cell data. In: Proceedings of 7th international waste management and
landfill symposium, Sardinia, Italy. Environmental Sanitary Engineering Centre, Cagliari,
Italy, pp 45–52
Koerner R (2005) Design with geosynthetics, 5th edn. Pearson Prentice Hall, Upper Saddle River
Komilis DP, Ham RK, Stegmann R (1999) The effect of landfill design and operation practices on
waste degradation behavior: a review. Waste Manag Res 17:20–26
Leckie JO, Pacey JG, Halvadakis C (1979) Landfill management with moisture control. J Environ
Eng-ASCE 105:337–355
Palmisano AC, Barlaz MA (1996) Microbiology of solid waste, 1st edn. CRC, Boca Raton
Pohland FG (1980) Leachate recycle as landfill management option. J Environ Eng-ASCE
106:1057–1069
Qian X, Koerner R, Gray D (2002) Geotechnical aspects of landfill design and construction.
Prentice Hall, Upper Saddle River, NJ
Reinhart DR, Al-Yousfi AB (1996) The impact of leachate recirculation on municipal solid waste
landfill operating characteristics. Waste Manag Res 14:337–346
Reinhart DR, Townsend TG (1997) Landfill bioreactor design and operation, 1st edn. CRC, Boca
Raton
Reinhart DR, McCreanor PT, Townsend TG (2002) The bioreactor landfill: its status and future.
Waste Manag Res 20(2):172–186
Senior E (1995) Microbiology of landfill sites, 2nd edn. Lewis, Boca Raton
Solid Waste Association of North America (2002) Request for comment on bioreactor definition.
Submitted to the United States Environmental Protection Agency, 29 June 2001
Townsend TG (1995) Leachate recycle at solid waste landfills using horizontal injection. Ph.D.
dissertation, University of Florida, Gainesville
US EPA (2011) Municipal solid waste generation, recycling, and disposal in the United States:
facts and figures for 2010. United States Environmental Protection Agency, Washington, DC
Zhang DQ, Tan SK, Gersberg RM (2010) Municipal solid waste management in China: status,
problems and challenges. J Environ Manag 91:1623–1633
Chapter 3
Planning for Sustainable Landfilling Practices
Proper planning is critical for any project the magnitude of a solid waste landfill.
The introduction of sustainable design and operational elements to such a project
demands additional emphasis on up-front planning. Preliminary considerations
include decisions on the desired objectives of the landfill facility and the extent of
additional components and technologies to be implemented. Some sustainable land-
fill practices may be limited by applicable regulations governing the facility, thus a
strong understanding of the regulatory and permitting process is critical to planning.
The facility must be designed and operated with due consideration of regulatory
requirements, as well as other design and operational features necessary to safe and
successful fulfillment of desired project goals.
While the objective of sustainable landfill practices and technologies may be
greater long-term environmental protection, those considering these approaches
should recognize that improper application of many of these technologies could result
in deleterious impacts. For example, the addition of liquids to promote rapid waste
stabilization is a major sustainable landfill technology and a major focus of this book,
but uncontrolled liquids addition has the potential to result in greater emissions to the
environment. If liquids are added at a flow rate or pressure greater than that which the
landfill’s containment infrastructure can accommodate, this can result in leachate out-
breaks and waste side slope failures. At landfills where waste stabilization is acceler-
ated, an inappropriately designed or operated LFG collection system may result in
greater gas-phase emissions to the environment. These considerations must be
planned for during the design process, even if they are not addressed specifically in
the regulations.
In addition to the required engineering design of the facility and its components,
implementation of sustainable landfill practices requires more demanding opera-
tional and monitoring considerations. These facilities require more control; this
greater control is provided through a combination of greater operational attention,
added control infrastructure, and additional collection of data used in the operation.
As an example, a landfill operator who would normally only be required to monitor
the safe and effective removal and disposal of leachate may be required to manage
and monitor a system for pumping liquids back into the landfill. Other potential
operational duties may include gas extraction, air addition, additional site or facility
inspections, interfacing with new technology and equipment, and care and mainte-
nance of energy conversion units.
The intent of this chapter is to introduce the planning elements required when
pursuing sustainable landfill practices. A discussion of these considerations pro-
vides a good introduction to the detailed technical presentations in later chapters.
In addition to discussing planning objectives, typical regulatory requirements, and
design and operational issues, upfront considerations regarding the long-term fate
of facilities integrating sustainable landfill technologies and economic consider-
ations are discussed. Methods to examine sustainability of different waste manage-
ment practices (e.g., life-cycle assessment) are also introduced.
Facility owners and operators must identify project objectives as part of the process
of planning implementation of sustainable landfill technologies. Identified objec-
tives may be constrained by a number of considerations, including regulatory limi-
tations, specific site features, local infrastructure and markets, and economics.
Whether the planning is for an existing landfill facility or a new operation may also
greatly dictate which objectives are reasonable to address. Table 3.1 summarizes a
list of potential sustainable landfill project objectives along with planning consider-
ations. The considerations are discussed in greater detail, both later in this chapter
and elsewhere in the book.
3.3 Regulatory Constraints and Considerations 37
The location, design and operation of modern engineered landfills are regulated by
national, regional or local government agencies. The specificity of the regulations
with respect to sustainable landfilling practices varies by jurisdiction and project
planners and developers must consult the appropriate regulatory agency to deter-
mine necessary requirements. The following section highlights major landfill regu-
latory requirements in the US and Europe to provide context as to typical regulatory
requirements and how these might impact the implementation of sustainable landfill
practices.
MSW landfills in the US currently fall under several federal regulations. Foremost
among these are the RCRA Subtitle D landfill regulations found in 40 CFR 258 (US
Government 2012a). In addition to other requirements, these rules contain location
38 3 Planning for Sustainable Landfilling Practices
practicing accelerated waste stabilization could produce a large volume of gas prior
to a regulatory trigger for capture, the National Emissions Standards for Hazardous
Air Pollutants (NESHAP) provided requirements for bioreactor landfills to capture
gas sooner than conventional municipal landfills (US Government 2012c). In this
rule, a bioreactor was defined as:
An MSW landfill or a portion of a MSW landfill where any liquid, other than leachate or
landfill gas condensate, is added in a controlled fashion into the waste mass (often in com-
bination with recirculating leachate) to reach a minimum average moisture content of at
least 40 % by weight to accelerate or enhance the anaerobic biodegradation of the waste.
Although the NESHAP rules for bioreactors differ slightly depending on whether
the landfill is a new or existing source, generally bioreactors as defined under
NESHAP must have LFG collection components installed before initiating operation
of the bioreactor and must begin collecting either within 180 days of bioreactor opera-
tion or after the waste moisture contents reaches 40 % (by weight), whichever is later.
Directives and policies in the EU have been put in place that are similar to US regu-
lations, with technical requirements such as liner systems and other protective mea-
sures to minimize risk to human health and the environment as a result of land
disposed waste. While many EU nations have migrated away from landfilling as a
primary method of managing MSW, as a whole, landfilling remains a common prac-
tice. As of 2010, more than half of EU member states landfilled greater than 50 %
of the municipal waste generated in their country (EEA 2013).
With a goal of reducing reliance on landfills, the EU has passed several directives to
promote resource recovery and landfill diversion. For example, the EU Landfill
Directive of 1999 (Council of the European Union 1999) provided a timeline for mini-
mizing the amount of biodegradable waste disposed of in landfills. Those nations meet-
ing this directive first process their waste through waste-to-energy (WTE) systems or
mechanical biological treatment (MBT) prior to landfill disposal. The EU’s Waste
Framework Directive of 2008 established a target to recycle 50 % of municipal waste
by the year 2020 (European Parliament and Council of the European Union 2008).
Table 3.2 Potential impacts of sustainable landfill practices on standard landfill design elements
Landfill design
element Impact of incorporating sustainable landfill practices on design element
Foundations The increased unit weight of the waste created by the introduction of
liquids and by the more rapid stabilization of the MSW can impact the
earthen foundation upon which the landfill is constructed. The designer
should factor this unit weight into the design of the landfill foundation.
Greater slopes in the leachate collection and removal system may be
required to ensure gravity drainage can still occur since greater differential
settlement of the foundation may be predicted based on the increased unit
weight of the landfill
Liner systems Liner systems are normally comparable to standard engineered landfills,
though possible increases in temperatures resulting from accelerated
biological activity (especially if air is added) may need to be considered
Leachate The leachate collection system needs to be designed to accommodate the
collection systems larger volumes of leachate that are expected as a result of liquids
introduction. Other design elements, such as foundation settlement and gas
extraction systems, should be considered in tandem with leachate collection
system design. A well-designed and constructed leachate collection system
is one of the most critical features of a sustainable landfill
Stormwater The possibility of surface seeps as a result of liquids introduction should be
control systems considered in the design of stormwater collection systems. Systems
designed to mitigate and control seeps can minimize the mixing of leachate
with stormwater
Slope stability The addition of liquids to landfills can impact the pore water pressure
existing within the waste mass, which in turn can lead to changes in the
shear stresses within the landfill mass and cause slope stability concerns.
Waste characteristics may also change as waste decomposes due to liquids
introduction. Designers should factor added water pressures into slope
stability analyses.
Leachate The recirculation of leachate will be a part of a site’s liquids management
management system. Leachate storage volumes should be examined as part of a water
systems balance that considers leachate production and recirculation rates. Leachate
treatment technologies that complement sustainable landfill technologies
should be considered based on site-specific factors
Gas extraction Liquids introduction not only increases the rate of gas production, it also
systems may impact the efficacy of many of the standard landfill gas collection
techniques. Gas collection systems need to be designed to accommodate
both enhanced gas production from liquids addition and the increased
volume of liquid within the waste
Capping and The approach to capping and closing a landfill using sustainable landfilling
closure system technologies during active filling and/or after closure should consider the
liquids introduction and other sustainable landfilling infrastructure and
impacts from waste settlement
storage unit, a conveyance mechanism to deliver liquids from the storage unit to the
landfill unit, and a scheme to apply liquids to the landfilled waste mass—collectively
referred to herein as the liquids addition system. Possible storage systems for the
liquids include ponds, tanks, or other storage units that are located outside the lined
landfill area. Liquids can be delivered from the storage system to the landfill in
a variety of fashions. In the simplest form, liquids can be hauled to the landfill in a
tanker truck and discharged directly to the surface (to infiltrate at the working face)
or to an impoundment area (e.g., a pond). Liquids can also be delivered to points of
interest through a piping network.
The design of a liquids introduction system includes the estimation of the vol-
ume of liquids that need to be added to increase the moisture content of the waste
from an initial value to a target value, identifying sources of liquids available, selec-
tion of the type of liquids introduction system, developing detailed specifications on
sizing and configuration of the liquids introduction devices, selecting spacing
between individual devices, and identifying materials of construction. Liquids can
be applied to the landfilled waste using a multitude of surface and subsurface tech-
niques. Surface applications include drip irrigation, spray irrigation, infiltration
ponds, and trenches, while subsurface applications consist of buried horizontal
injection trenches, planar or blanket systems, or vertical wells. The systems that add
liquids via surface application are less complicated to design than those that add
liquids via subsurface application. The design process for surface liquids introduc-
tion systems involve the specification of a liquids application rate, the area of liquids
application, and a piping and pumping system to accomplish liquids introduction.
Conversely, the design process for subsurface liquids introduction systems entails
the specification of the sizing and configuration of individual injection devices,
spacing between these devices, injection pressure (or flow rate), material selection
(e.g., trench bedding media, pipe diameter and thickness), and pumping system
design. Design methods and considerations for a variety of liquids addition strate-
gies are presented in Chaps. 6 through 9.
The design of a leachate collection and removal system (LCRS) is one of the
most important design elements for all landfill designs, especially for landfills with
high moisture contents or those where liquids are deliberately added into the waste.
A well-functioning LCRS can effectively reduce the potential for groundwater
impacts resulting from leachate leakage and slope failure due to increased pore-
water pressure and changes in waste characteristics. The main components of a
LCRS include a liner system sloped to promote gravity drainage, a perforated col-
lection pipe network, drainage media to route the collected liquids to targeted con-
veyance points, and pumping systems to remove leachate from the landfill. More
liquids are expected to be collected by the LCRS in bioreactor landfills and similar
operations because of the added liquids. Therefore, the LCRS must have adequate
drainage capacity to handle the increased leachate flow; Chap. 10 focuses on
LCRS design.
As part of the design and permitting process of a bioreactor landfill, other design
elements may also need to be considered and integrated into the design elements
discussed above, such as seepage control and leachate management. Leachate seeps
3.4 Engineering Design Considerations 43
are usually observed as wet spots on the surface of landfill side slopes, especially in
landfills where liquids are added under pressure. Leachate seepage can be odorous
and attract vectors, in addition to causing other environmental issues, such as leach-
ate migration beyond the lined limits of the landfill, storm water contamination,
cover soil erosion, gas emissions through the cover, and potential slope stability
issues. Design engineers for landfills practicing liquids addition need to balance the
use of pressurized liquids addition for moisture distribution with the need to mini-
mize leachate seepage problems.
A leachate treatment and management system is another primary design element
that needs to be considered. One objective for operating a landfill as a bioreactor is
leachate treatment. Leachate recirculation, to some extent, can reduce the organic
chemicals in the leachate through biological degradation. It is important for a design
engineer to understand the degree of treatment if external leachate treatment is
needed to meet desired treatment limits, particularly in cases where the leachate
production rates at the site exceed the design leachate recirculation rates. In addi-
tion, the amount of leachate produced at a bioreactor is generally greater than a
conventional landfill when outside liquids are added. Therefore, a leachate storage
system of sufficient capacity is critical to bioreactor landfill leachate management.
The enhanced leachate production rate due to bioreactor operations should be fac-
tored into the leachate management design process. Leachate management tech-
niques, from seep control to storage and treatment, are presented in Chap. 11.
Promoting rapid waste stabilization increases the LFG generation rate, and planning
for such an outcome is a major consideration in sustainable landfill project develop-
ment. Operating a landfill as a conventional landfill or as a bioreactor landfill gener-
ates the same amount of landfill gas over the long term, as the total amount of gas
that can be produced is a function of the waste mass and its characteristics. The
increased LFG generation rate associated with landfills practicing enhanced stabili-
zation techniques can be beneficial to the landfill owner because the accelerated gas
generation may make beneficial use of the gas more economically feasible, and
provides an opportunity to collect gas in the early years of a site’s active gas collec-
tion system. However, the increased gas generation rate presents design and opera-
tional challenges.
The main design elements for a gas collection system include gas extraction
devices, larger gas conveyance pipes, condensate collection, storage and convey-
ance system, and a vacuum source. If the generated landfill gas is not efficiently
collected, the accelerated gas generation rate will increase landfill gas emissions to
the atmosphere. The efficiency of a landfill gas collection system depends on design
elements such as the density and type of collection devices (e.g., horizontal, verti-
cal, surface collection, leachate collection system integration), the presence or
absence of a bottom liner system, landfill cover characteristics, applied vacuum, and
44 3 Planning for Sustainable Landfilling Practices
While liquids addition to promote anaerobic stabilization has been the most widely
discussed and implemented technique for sustainable landfill operation, other
opportunities may need to be considered in the design and planning process. Other
considerations include waste processing and placement objectives, the types of
wastes accepted for disposal, and waste mixing considerations (e.g., mixing wet
wastes with dry wastes). A sizable impediment to waste stabilization through liq-
uids addition is the inability to uniformly wet the waste. Thus waste processing
through shredding prior to disposal and the deliberate reduction in waste compac-
tion have both been proposed as techniques to promote even moisture distribution.
The co-disposal of some wastes may limit effectiveness of sustainable practices
(e.g., when ash layers limit liquids reaching MSW) and in other cases enhance it
(e.g., when biosolids are mixed with MSW to provide moisture and nutrients).
Air addition has been proposed as a tool for more sustainable landfill operation.
While it poses a greater risk with respect to landfill fires, air addition provides for
more rapid waste stabilization. Air addition has been selectively used for targeted
benefits, such as warming cold landfills to prime them for subsequent anaerobic
stabilization and for stabilizing landfills at their end of life. The use of air addition
at landfills is the focus on Chap. 14.
Planning in the design process must also include necessary engineering to
ensure necessary facility integrity in the short and long term, and to provide future
opportunities to maximize sustainable landfill practices. With the addition of liq-
uids, the formation of elevated gas pressures, and the changing nature of stabilized
3.5 Operation and Monitoring 45
waste, the engineer must assess and address potential concerns with respect to
slope stability (discussed in Chap. 12). Some future opportunities may be maxi-
mized through upfront planning. For example, future waste reclamation and reuse
of landfill cells will be much easier if the liner system is designed with this process
in mind (landfill reclamation issues are presented in Chap. 17). Feasibility of future
energy recovery opportunities, such as landfill gas, solar energy, and wind power,
may be greater if site infrastructure is designed from the beginning with these
objectives considered (Chap. 19).
Under normal landfill operation, the facility is operated such that waste is deposited
and compacted to reach a final configuration and then closed with an engineered
capping system. Engineers and operators must also assess at what point to close
3.7 Economic Considerations 47
distinct areas of the landfill and what types of cover systems to use. These decisions
become even more important in facilities operated with sustainable practices such
as rapid waste stabilization, reclamation and energy recovery systems. The focus of
Chap. 17 is on landfill management at end-of-life and opportunities for more sus-
tainable practices.
The standard approach of closing once a predetermined design elevation is
reached may not be the best choice for landfills practicing accelerated stabilization.
One objective of rapid stabilization is to recover additional disposal capacity (air-
space), thus premature construction of a cap may prevent utilization of this addi-
tional capacity. In addition, closure systems that rapidly settle will be more subject
to damage and thus necessitate repair. Thus as part of planning, the engineer and
operator must consider whether to overfill the waste anticipating future settlement
or planning a temporary cover or capping system that later will be removed to allow
addition placement—the planning process should assess whether overfilling is a
practice permitted under state or local regulatory rules.
While premature capping of a landfill area may be disadvantageous for facilities
undergoing rapid stabilization, an engineered cover or cap has benefits for such
systems. Gas collection can be enhanced, as can the control of leachate seeps from
the side of the landfill. Thus, options such as temporary capping systems and partial
closure of target areas should also be considered.
A targeted benefit of sustainable landfill operations is to minimize the environ-
mental, economic, and social impacts as much as possible. Planning for the future
of the facility early in the process allows the engineer and operator to maximize
future use of the site and to minimize future cost and impact. An alternative route to
closing the facility may be to reclaim all or part of the stabilized waste and cover
soil through a large-scale mining operation. The mining plan should consider poten-
tial quality and use of the excavated materials and likely outlets for their reuse.
Consideration must be given to the design and construction of the system to best
allow such recovery and possible reuse of landfill infrastructure.
Other costs are more difficult to quantify. For example, additional disposal
capacity gained through accelerated waste stabilization depends largely on whether
the landfill is operated and configured in a manner to recoup the airspace. Potential
savings from lower monitoring and operating costs in the future and reduced liabil-
ity are much more difficult to quantify. Table 3.5 presents a series of economic fac-
tors and a discussion of considerations for each. Economic considerations of
sustainable landfilling practices are discussed in Chap. 18.
Fig. 3.1 Generalized depiction of key aspects of LCA for waste management systems, which are
often represented by charts showing different processes, material and energy flows, and relation-
ship of different processes to one another
(typically materials and energy), processes and actions that occur during the life
cycle of a given material, emissions, and potential sinks or offsets that decrease the
impact of the emissions. Life-cycle models can examine entire systems such as that
shown in the figure, or models can allow very close examination of one or a few
elements (e.g., landfilling processes only). Advancements in computing, analytical
capability, and availability of site monitoring data have led to rapid expansion of the
study of LCA in waste management in recent years, which are available to landfill
owners, operators, and engineers to evaluate potential impacts based on a series of
site-specific inputs to assess potential environmental, economic, and social impacts
over short to long time horizons.
Different of computer-based tools have been developed over the years to exam-
ine the life-cycle impacts of waste management systems. A summary of recently-
developed or recently-updated models is provided in Table 3.6.
Detailed procedures and approaches to conduct an LCA are beyond the scope of
this book, but a discussion about major factors related to sustainable landfills that
can impact LCA results is warranted. For example, methane has a large greenhouse
gas potential compared to carbon dioxide, so uncontrolled methane emissions can
have a substantial impact on life-cycle greenhouse gas emissions for a given land-
fill. Implementation of sustainable landfilling technologies (including early and
effective gas control) can show very favorable lifetime greenhouse gas emissions
compared to a traditional landfill, but the reverse may also be true if early or
50 3 Planning for Sustainable Landfilling Practices
Table 3.6 Listing of LCA computer models with a focus on waste management processes
Model name Developer Description
EASEWASTE (now Technical University Allows LCA of integrated waste management
EASETECH) of Denmark operations including transportation,
composting, resource recovery, and landfilling
based on resources consumption and
environmental emissions from these
operations for municipal solid waste
Municipal solid US Environmental LCA tool designed to aid solid waste planners
waste—decision Protection Agency, in evaluating the economic and environmental
support tool RTI International, aspects of integrated municipal solid waste
(MSW-DST) and NC State management operations including collection,
University transfer, materials recovery, composting,
waste-to-energy, and landfill disposal
WRATE (Waste and Environment LCA of integrated municipal solid waste
Resources Assessment Agency UK management operations including collection,
Tool for the materials recovery, composting, waste-to-
Environment) energy, and landfill disposal
References
Council of the European Union (1999) Council Directive 1999/31/EC of 26 April 1999 on the
landfill of waste. Official Journal of the European Communities
EEA (2013) Managing municipal solid waste—a review of achievements in 32 European coun-
tries. European Environmental Agency, EEA report no. 2/2013
References 51
European Parliament, Council of the European Union (2008) Directive 2008/98/EC of the
European Parliament and of the Council of 19 November 2008 on waste and repealing certain
directives. Official Journal of the European Union
Gentil E, Damgaard A, Hauschild M, Finneveden G, Eriksson O, Thorneloe S, Kaplan P, Barlaz
M, Muller O, Matsui Y, Li R, Christensen T (2010) Models for waste life cycle assessment:
review of technical assumptions. Waste Manag 30:2636–2648
Jain P, Powell JT, Smith JL, Townsend TG, Tolaymat TM (2014) Life-cycle inventory and impact
evaluation of mining municipal solid waste landfills. Environ Sci Technol 48(4):2481–2487
Reinhart DR, Townsend TG (1998) Landfill bioreactor design & operation. CRC, Boca Raton
US Government (2012a) US code of federal regulations: title 40 Part 258. Criteria for Municipal
Solid Waste Landfills
US Government (2012b) US code of federal regulations: title 40 Part 60. Subpart www. Standards
of performance for municipal solid waste landfills
US Government (2012c) US code of federal regulations: part 63. Subpart AAAA. National
emission standards for municipal waste landfills
Chapter 4
State of Practice
4.1 T
he Evolution of Sustainable Landfill Research
and Application
Several full-scale landfills have been operated using sustainable landfilling techniques
in North America. Investigators have reported on different key aspects of sustainable
landfill operation and science, including gas production (Faour et al. 2007), biological
and chemical aspects of leachate and gas (Barlaz et al. 2010), and bioreactor practice
or performance in general (Benson et al. 2007; US EPA 2007; Bareither et al. 2010;
Kumar et al. 2011). The focus of this section is to present summaries of some of the
better documented North American case studies in the technical literature.
Reviewers tend to agree that increased moisture content, to near field capacity, is
the dominant factor in the promotion of the accelerated waste degradation observed
in sustainable landfills (Benson et al. 2007; Kumar et al. 2011). Accelerated decom-
position can be characterized by greater waste decay rates, which is often described
through the landfill gas decay constant; normally accelerated stabilization occurs
after some lag period following liquids addition initiation. Summary reports includ-
ing leachate quality data tend to agree that over time, as sustainable landfill opera-
tion progresses, the strength of leachate tends to decrease over time (as represented
by a decreasing ratio of BOD:COD), while ammonia levels tended to stay elevated,
even after many years in some cases (Benson et al. 2007; Barlaz et al. 2010).
Enhanced and/or accelerated settlement has been observed in most reported studies
that collected routine measurements of surface elevations. Overall, sustainable land-
fill sites in North America have shown an ability to perform within landfill regula-
tory limits and guidelines (Benson et al. 2007; US EPA 2007; Bareither et al. 2010).
The Delaware Solid Waste Authority (DSWA) has practiced leachate recircula-
tion at several of its operating landfills since the early 1980s (Watson 1987).
Operators employed multiple methods of leachate recirculation at the Central
Solid Waste Management Center (CSWMC) in Sand Town, Delaware, which
accepted waste until 1996. Morris et al. (2003) summarized results of long-term
monitoring at the site.
4.2 Full-Scale Case Studies: North America 57
s ystem from a submersible pump placed in the leachate storage tanks. The manifold
system was designed to distribute the liquids uniformly over the application area,
but once initiated it was recognized that the rate of leachate application was larger
than the achievable rate of liquids uptake into the waste, resulting in excessive
leachate ponding and runoff.
The limited rate of leachate infiltration through the surface of the landfill from
the drip irrigation system led to the use of surface infiltration ponds (Townsend
et al. 1995), which provided a greater storage volume and more consistent infiltra-
tion rate into the waste. The ponds were constructed by excavating a depression into
the waste on the surface of the landfill (Fig. 4.3) in combination with constructed
perimeter walls comprised of newly-compacted waste. Liquids were added via a
piped connection to the leachate tanks; a constant depth of liquid was then main-
tained in the ponds during operation (Fig. 4.4). Hydraulic performance of the ponds
was closely monitored by tracking the water balance on a daily basis (Townsend
1992); leachate infiltration rates ranged from 6 × 10−6 to 9 × 10−6 cm/s (5,500–8,300
gallons per acre-day). Using the infiltration data and the liquid depths in the infiltra-
tion ponds, the vertical hydraulic conductivity of the underlying waste was esti-
mated to range from 3 × 10−6 to 4 × 10−6 cm/s (Townsend et al. 1996).
Throughout the research, waste samples were collected by augering into the waste
with a solid-shaft open-flight auger, both before and after liquids addition (Jain et al.
2005a). The results demonstrated that the infiltration pond technique resulted in
favorable moisture distribution and waste decomposition (Townsend et al. 1996;
Kim and Townsend 2012; see Chap. 16). Observed disadvantages of surface ponds
included floating waste (typically occurring after several months of operation as a
result of biogas becoming trapped under plastic film), and the requirement of a large
area of landfill surface. Given the relatively high amount of rainfall in Florida
(approximately 50 in. per year), additional moisture entered the ponds over time,
60 4 State of Practice
especially when stormwater from surrounding areas entered the ponds. When ponds
were constructed or modified by compacting waste to extend the perimeter berms,
seeps would sometimes occur at the base of the newly added berms. Experience with
these infiltration ponds is discussed further in Chap. 7.
The next phase of leachate recirculation at the site was performed using buried
horizontal trenches containing perforated pipe and a bedding material of shredded
tires (Fig. 4.5). Most of the trenches were constructed using an excavator to
4.2 Full-Scale Case Studies: North America 61
d imensions of approximately 1 m deep by 1 m wide (3 ft by 3 ft), with lengths from
110 to 240 m (360–780 ft). Shredded tires were placed on the bottom half of the
trench, followed by the placement of perforated PVC pipe (in most cases 7.7-cm
(3-in.) diameter pipe), and the remaining trench volume was filled with shredded
tires. The top of the trench was then covered by waste excavated from the trench
and topped with cover soil. This buried trench system, referred to as horizontal
injection lines (HIL), was used for leachate recirculation for the remaining life of
the landfill and after closure (in 1998), but was studied in greatest detail from 1992
to 1995 (Townsend 1995; Townsend and Miller 1998).
Leachate was added to individual injection lines at a rate ranging from 0.003 to
0.005 m3/s (50–100 gpm) and the resulting injection pressure was recorded.
Leachate recirculation rates were highest at the beginning of daily injection cycles,
and as leachate recirculation progressed, achievable flow rates decreased and injec-
tion backpressure increased. After a non-operational rest period, leachate recircula-
tion flow rates would return to higher levels, but soon returned to previous lower
rates and higher back pressures; this trend continued, with increasing cumulative
injection volume over time. Townsend and Miller (1998) described the hydraulic
performance of individual injection lines; results of this work are discussed further
in Chap. 9.
Gas collection infrastructure was installed at the site in 1994. Large gas pressures
were observed in the HILs and the injection system was reconfigured to operate
with the dual purposes of leachate recirculation and gas collection (Fig. 4.6). The
HIL pipes were configured so that leachate could be added as desired, such that
when no leachate was recirculated, gas could be extracted using an independent gas
manifold connected to the landfill’s blower flare station. Although leachate recircu-
lation activity resulted in high gas production rates, flooded conditions surrounding
the trenches precluded effective gas extraction (Townsend et al. 1994). In many
cases, when connected to the gas collection manifold, leachate would periodically
surge into the gas collection line, resulting in larger amounts of liquids to manage
than would typically result from extracted gas condensate alone.
Another observation at the site was the presence of large gas pressures in the
LCRS. This led to the retrofit of the LCRS for gas collection, a technique which was
found to be much more effective than gas collection from the buried injection
trenches (Townsend and Miller 1997). The operation of the leachate lift stations was
modified to minimize gas escape through the manholes, thus promoting a greater
gas capture rate from the LCRS.
The landfill was closed in 1998 and capped with a final cover system that
included a geomembrane. Leachate recirculation into the landfill using the buried
horizontal lines and a surface trench system continued as of 2014. Early research
efforts at the site examined the use of membrane treatment to create a diluted leach-
ate stream that could be spray irrigated on the site and a concentrated stream that
would be recirculated back to the landfill (Townsend 1992). Leachate is currently
treated using reverse osmosis (RO) following this approach; with the RO permeate
land-applied to the vegetated final cover system and the concentrate recirculated.
62 4 State of Practice
Fig. 4.6 Combined liquids addition and gas extraction system connected to horizontal trenches at
ACSWL (white pipe is gas extraction manifold and gray pipe is liquids addition manifold)
The Yolo County Planning and Public Works Department followed up a successful
demonstration of bioreactor landfill concepts in a side-by-side pilot-test cell com-
parison begun in 1994 (see Table 4.1) with full-scale implementation of several
sustainable landfill technologies. Three different landfill cells were constructed to
operate with liquids addition, with 6- and 3.5-acre (2.4 and 1.4 ha) cells operated
anaerobically and a 2.5-acre cell operated aerobically (see Fig. 4.7 for an overall site
schematic). Leachate recirculation systems in the full-scale cells were constructed
as they were built with horizontal injection lines buried in the waste (four injection
line layers in the 3.5- and 6-acre anaerobic cells and three injection line layers in the
2.5 acre aerobic cell) (Yazdani et al. 2002, 2006). Instrumentation layers were
integrated as part of a supervisory control and data acquisition (SCADA) system
for improved air and leachate injection control capabilities and access to
4.2 Full-Scale Case Studies: North America 63
Fig. 4.7 Configuration of aerobic and anaerobic bioreactor cells at the Yolo County Landfill
instrumentation data from the sensors embedded within the waste (Yazdani et al.
2006). When completed, the anaerobic cells were covered with a geomembrane
cap (Fig. 4.8).
Instruments were placed throughout the landfill to measure temperature, mois-
ture content, and fluid pressure. As illustrated in Fig. 4.9, instruments were placed
adjacent to liquids addition and gas extraction manifolds. Pressure data from sen-
sors installed at the bottom of the full-scale bioreactor cell indicated that the maxi-
mum head on the liner (HOL) was within regulatory limits (peaking typically
around 0.9 in. (2.3 cm)). Sampling tubes were also installed for the collection of
gases within the landfill for measurement of major gas components. Field-scale gas
tracer tests were performed to characterize moisture content of the waste and gas
flow patterns. Liquids addition was performed using horizontal trenches installed
64 4 State of Practice
Fig. 4.8 View of Yolo County Landfill northeast anaerobic bioreactor cell; upon completion the
slopes were covered with a geomembrane cap ballasted by tires (Photo courtesy of Ramin Yazdani)
The New River Regional Landfill (NRRL) located in Raiford, FL, US, is owned and
operated by the New River Solid Waste Association (NRSWA) and receives a mix
of residential and commercial waste from surrounding municipalities. The NRRL
site occupies approximately 500 acres (202 ha) in total area and consists of six con-
tiguous lined landfill cells totaling approximately 82 acres in size. In 2001, the
NRSWA retrofitted approximately 10 acres (Cell 1 and part of Cell 2) with sustain-
able landfilling infrastructure (Fig. 4.11) including leachate recirculation, air injec-
tion, landfill gas extraction, and monitoring equipment. Several research experiments
were performed in this area of the site and NRSWA has continued implementation
of sustainable landfill practices in other areas of the site.
The liquids addition system of the original bioreactor landfill area consisted of
45 vertical well clusters (Fig. 4.12) that were used to recirculate leachate (and
groundwater) and add air to the landfill (Jain et al. 2005a). Each cluster consisted
of three wells with approximate depths of 20, 40, and 60 ft. The injection wells
within each cluster were approximately 2 ft apart, and each cluster approximately
50 ft from other clusters (Fig. 4.13; additional construction photos of the wells are
provided in Chap. 8). Pumps located in the facility’s lined leachate ponds pro-
vided liquids to the injection well-field. Added liquids included leachate collected
from the lined landfill units on site as well as groundwater, pumped into the ponds
as needed to fulfill the liquids addition requirements. Approximately one-half of
the well field was constructed with aid addition infrastructure. Two positive dis-
placement blowers located at the landfill gas blower flare station provided the
pressurized air.
Thirty-one 3-ft deep by 3-ft wide horizontal gas collection trenches were con-
structed at 120-ft spacing on the landfill surface (including side slopes) beneath a
4.2 Full-Scale Case Studies: North America 67
Fig. 4.12 Cluster of vertical wells at NRRL after construction and prior to liquids or air addition
68 4 State of Practice
Fig. 4.14 Recently augered cluster of boreholes with instruments placed at bottom of borehole.
Each instruments bundle was temporarily attached to the end of a pipe
Fig. 4.15 The pipes were dislodged from the instrument bundles using a small diameter pipe to
the provide separation force required
70 4 State of Practice
Air was added to 134 vertical wells installed at three different depths at flow rates
ranging from 5 to 50 scfm, and the corresponding steady-state pressures were
recorded and used in an analytical fluid flow model to estimate air permeabilities of
1.6 × 10−13 to 3.2 × 10−11 m2. The estimated air permeability decreased significantly
with increasing waste depth, which was attributed to the lower porosity of waste in
deeper sections caused by higher overburden pressures, moisture contents, and
landfill gas pressures.
Leachate recirculation tests were conducted in 2003 and 2004 to estimate the
saturated hydraulic conductivity (Jain et al. 2006; see Fig. 4.17). The tests were con-
ducted at 23 locations using the borehole permeameter test and the saturated hydrau-
lic conductivity (Ks) of the landfilled waste was estimated to range from 5.4 × 10−6 to
6.1 × 10−5 cm/s. Similar to air permeability, the hydraulic conductivity of the waste
decreased with depth, the likely result of greater overburden pressures associated
with increasing waste depth in the landfill. The decrease in hydraulic conductivity
with depth suggested that a single screened well was sufficient to achieve uniform
distribution and that a cluster of multi-depth wells was unnecessary.
Jain et al. (2014) reviewed the performance of the vertical well system for liquids
addition. Over a 5-year period, 25,000 m3 of leachate was added to the well field.
The performance was evaluated in terms of fluid conductance (defined as flow rate
per unit well screen length per unit liquids head above the well bottom), which was
found to range from 5.6 × 10−8 to 3.6 × 10−6 m/s. Liquid depths within the well had to
be maintained below the landfill surface to avoid surface seepage; therefore, the
system operation was labor intensive, especially for wells installed at the shallow
depth. Concrete collars to minimize seeps under pressurized addition of liquid were
tested, but leachate surface seeps were still problematic.
4.2 Full-Scale Case Studies: North America 71
Fig. 4.17 Addition of liquids into a vertical well cluster at NRRL and measurement of flow rates
and liquid depths
The ability of the resistivity sensors and TDR clusters to monitor moisture
c ontent was evaluated (Kumar et al. 2009; Jonnalagadda et al. 2010). Sensors used
to detect landfill moisture showed that the extent of lateral moisture movement
ranged from 8 to 10 m. When the spatial average moisture content of the landfill
following the experimental period was calculated, the resultant value was very high,
suggesting that obtaining true moisture content magnitude from in-situ sensors can
be complicated by various factors (e.g., channeling of liquids). From these observa-
tions, it was concluded that in-situ moisture monitoring devices are well suited to
detect the presence of moisture, but not necessarily to calculate an exact in-situ
moisture content. The resistivity sensors, which were less expensive to construct
and install compared to the TDR sensors, performed comparably to the TDR sen-
sors and in general proved to be more reliable.
In 2004 and 2005, air injection tests were conducted to examine the change in
landfill gas quality upon initiation of aerobic decomposition conditions (Powell
et al. 2006). The concentrations of CH4, CO2, O2, and trace chemicals (nitrous oxide
(N2O), a suite of volatile organic compounds, CO, and H2S) were measured both
before and during air addition. A significant increase in CO was observed in 9 out
of 14 monitoring points after initiating air addition, and this increase was concurrent
with a decrease in the ratio of CH4 to CO2. A significant decrease in H2S was
observed at 6 of 14 monitoring points, but no noticeable effect on N2O and volatile
organic concentrations was observed. The results suggested that aerobic decompo-
sition conditions can be accomplished within compacted MSW and that certain
problematic gases (e.g., H2S) can be controlled.
72 4 State of Practice
Fig. 4.18 Installation of a vibrating wire pressure transducer into a buried vertical well at NRRL
maintenance problem. Kadambala et al. (2011) evaluated the use of buried vertical
wells as a method to avoid seepage issues; vertical wells were constructed on the
surface of the landfill and connected via a buried manifold in a surface trench
(Fig. 4.18). Another lift of waste was then placed above the top of the manifold,
which allowed for the successful addition of liquids into vertical wells under pres-
sure without resultant surface seeps (see Chap. 8 for more details).
Leachate recirculation at the Crow Wing County Landfill in North Central Minnesota
started 1998 (Doran 2007; US EPA 2007). The site, which began waste acceptance
in 1991, receives approximately 40,000 tons of MSW per year, has four lined land-
fill units (Fig. 4.19), and introduces leachate to the landfill by spray application to
the working face, spray application to intermediate landfill cover, and buried hori-
zontal trenches (Doran 2007; US EPA 2007; Burns and McDonnell Engineering
Company Inc. 2014). All cells are equipped with a composite bottom liner consist-
ing of compacted clay and a 60-mil HDPE geomembrane. The recirculation pipe
design consists of alternating 4-in. and 5-in. perforated pipes bedded in shredded
tires. Approximately 3.5 million gallons (13,200 m3) of leachate are recirculated
annually, with a range of 1.9 million to 5.0 million gallons (7,200–18,900 m3)
(Burns and McDonnell Engineering Company Inc. 2014). The waste moisture
74 4 State of Practice
content prior to initiating leachate recirculation activities was 19 %. The moisture
balance at the site is updated annually and was 22 % in 2013, with a maximum of
25 % observed in some locations (Burns and McDonnell Engineering Company Inc.
2014).
Leachate is stored at the site using four treatment ponds that are configured in
series and provide a total storage capacity of approximately 3.9 million gallons.
Leachate is recirculated into the buried horizontal trenches, at rates which range
from 25 to 50 gallons per linear ft of trench per year. Recirculation is practiced
between March and October to avoid the colder winter months and potential issues
with freezing conditions (Burns and McDonnell Engineering Company Inc. 2014).
The leachate management hierarchy at the site consists of, in terms of decreasing
preference, treated leachate spray application to an on-site spray field, leachate
recirculation into the landfill, and off-site hauling. Historical measurements of the
collected raw leachate quality show that the BOD:COD ratio has dramatically and
rapidly decreased in each of the four cells to <0.2 within 3–5 years of initiating
leachate recirculation (Burns and McDonnell Engineering Company Inc. 2014). An
ex-situ leachate treatment system via ponds serves to reduce ammonia by nitrifica-
tion (and subsequent conversion of nitrate to nitrogen gas via denitrification after
leachate is recirculated back into the landfill). As of 2013, no leachate has been
managed via off-site hauling since 2002, demonstrating that the combined system
4.2 Full-Scale Case Studies: North America 75
of leachate recirculation and on-site treatment has been effective over a long period
as the primary leachate management options.
An active GCCS was constructed at the site in 2008 and consists of gas collectors
within the waste as well as plumbing to capture gas from leachate recirculation
devices and the LCRS (Burns and McDonnell Engineering Company Inc .2014).
Collected LFG is managed via a flare and an on-site boiler and the quality has his-
torically ranged from 45 to 50 % CH4 at a flow rate of 200 standard cubic feet per
minute (5.7 m3/min).
As for operational observations, some seepage occurred during early stages of
recirculation activities near locations of former access roads. A robust leachate col-
lection toe drain around the perimeter of the cells was found to alleviate seepage
issues, although the drains were found to intercept stormwater during large rain
events. To mitigate leachate seeps, sandy soils were used for an intermediate cover
and a recirculation line perforation setback distance of 15 m was maintained (US
EPA 2007).
Historical airspace monitoring at the site found that the airspace utilization fac-
tors for the cells increased over time, for cell 1, from 1,004 to 1,341 lb/yd3, an effect
attributed to the leachate recirculation activities and subsequent waste degradation
and settlement, which was calculated at 20 % of total waste height after 5 years of
recirculation operations (US EPA 2007). Given the high degree of settlement
observed, flexibility in infrastructure piping (including flexible stainless steel and
excess 5 ft. engagements at pipe ends) was key in reducing operational issues as
leachate recirculation and subsequent settlement progressed (US EPA 2007).
The Polk County North Central Landfill (PCNCL) in Winter Haven, FL, US was the
site of intensive research on sustainable landfill operation, particularly as related to
the controlled addition of liquids into buried horizontal trenches and galleries. The
county had historically operated several MSW landfills and had on occasion prac-
ticed leachate recirculation through surface ponding as a means of leachate manage-
ment. Motivated in large part by rising leachate disposal costs, the county modified
its existing lined Phase II landfill unit to operate as a sustainable landfill (Fig. 4.20).
Landfill gas from the site is conveyed to a neighboring industrial facility for direct
beneficial use.
Beginning in 2000, the landfill began installing a series of horizontal trenches for
liquids addition. The majority of these trenches were constructed using an excavator
to approximate dimensions of 1 m deep by 1 m wide (3 ft by 3 ft), with lengths up
to 220 m (720 ft). Ten-centimeter (4-in.) high density polyethylene (HDPE) pipes
were used; 0.95-cm diameter pipe perforations (0.375-in.) were placed at a fre-
quency of 2 for every 0.6 m (2 ft) of pipe and were each oriented 45° from either
side of the vertical and placed in the downward direction. A variety of bedding
materials were used, including shredded automobile tires and crushed glass; in
76 4 State of Practice
Fig. 4.20 Plan view of Polk County North Central Landfill (PCNCL)
s everal cases no bedding material was used and excavated waste was placed back
into the trench after pipe installation. In a few cases, injection lines were installed
using a trenching device that pulled the pipe in place directly into the waste with no
bedding (see Chap. 9). Figure 4.21 shows a segment of HDPE pipe being thermally
welded after placement on top of the shredded tire bedding media. Figure 4.22
shows the segmented construction of a horizontal injection trench; in some cases,
only sufficient lengths of trench were installed to keep up with the incoming waste
placement needs, and construction was continued at a later time.
More than 100 buried horizontal trenches were installed at the site. Perforations
for the HDPE pipes stopped at least 30 m (100 ft) before the pipe exited the side
slope of the landfill. At the transition from perforated to non-perforated pipe, a plug
of clayey soil was placed as bedding around the trench to prevent short-circuiting of
leachate to the side slopes. After exiting the landfill, the pipes were routed to the
base of the landfill where they were connected to a manifold system via hydrants
(Fig. 4.23). The site’s leachate tanks served as the source of liquids; a variable speed
pump system was installed specifically for the purpose of liquids addition and per-
mitted the addition of liquids into aspecfied hydrant at a constant flow rate. The
pumping system was integrated with a SCADA system. The SCADA system, along
with the pump station, flow meters, and pressure transducers, allowed continuous
control (e.g., opening and closing valves for specific recirculation lines) and record-
ing of operational data (e.g., liquids addition pressure, flow rate, added volume, and
run time). The permit conditions for the site mandated a maximum injection pressure
4.2 Full-Scale Case Studies: North America 77
Fig. 4.22 Partially constructed horizontal trench for liquids addition at PCNCL (shredded tires
used as a bedding media)
78 4 State of Practice
Fig. 4.23 Delivery pipes transmitted liquids from manifold system to individual horizontal
trenches at PCNCL
Fig. 4.24 Installation of buried pore water pressure transducers surrounding horizontal injection lines
cut off to avoid slope stability concerns and the system was configured so that liq-
uids flow into a given line would cease when the pressure threshold was reached.
In the vicinity of selected injection trenches a series of vibrating wire pressure
transducers were installed, both within the trench, below the trench, and within
the waste at various radial distances from the trench (Fig. 4.24). Each pressure
4.2 Full-Scale Case Studies: North America 79
transducer was inserted into a sand bag saturated with water to prevent damage and
to allow pressure from the waste to be transmitted to the transducer quickly. The
transducer wires were encased in PVC pipes filled with polyurethane expanding
foam to prevent damage and preferential liquid flow. Larson et al. (2012) conducted
air addition tests using 13 pressure transducers to measure the vertical air permea-
bility of landfilled waste overlain by 3–6 m of waste plus a cover soil layer. The
vertical air permeability was determined to range from 2 × 10−13 to 8 × 10−13 m2 for
the topmost 3–6 m of compacted waste.
Liquids addition was performed for a period of 5 years and more than 100,000 m3
(25 million gallons) were added to the landfill. Larson (2007) found that fluid con-
ductance values (flow rate per length of pipe per unit pressure head; see Chap. 9)
were similar for shredded tires and crushed glass. In the first stages of liquids addi-
tion, fluid conductance values for trenches with bedding were greater than in
trenches without bedding media, and were greater for trenches closer to the surface
than those deeper in the landfill. As liquids addition proceeded, these differences
became less pronounced. In follow-up work incorporating additional injection lines,
Kumar (2009) observed that fluid conductance increased with increasing cumula-
tive injection volume and decreased with increased overburden waste depth.
Kumar (2009) also examined the spatial variation of pore water pressure in the
waste as a result of pressurized liquid addition using the buried transducers. At a
constant flow rate of 0.057 m3/min, liquids were intermittently added through hori-
zontal lines in trenches filled with bedding media of shredded tires, crushed glass,
or excavated waste. Within the trench, pressure distributions were more uniform for
trenches with bedding media compared to those without. For instruments in the sur-
rounding waste, pore pressures were found to dissipate a short distance from the
trenches; a drop of 4 psi approximately 25 ft from the trenches was observed (Kumar
2009). Cho (2010) conducted additional evaluations at the site with similar results.
Additionally, flow and pressure data were used to estimate waste properties; esti-
mated horizontal and vertical hydraulic conductivities ranged from 3.0 × 10−4 to
7.0 × 10−4 cm/s and 1.0 × 10−5 to 1.9 × 10−5 cm/s, respectively, resulting in corre-
sponding anisotropy values from 37 to 280.
Two large liquids addition horizontal blankets were also installed at the
PCNCL. Each blanket was constructed with a bedding depth of approximately
0.5 m over a 30 m by 60 m area. One blanket utilized shredded tires as a bedding
material (Fig. 4.25) and one used crushed glass. Prior to installation of the blanket,
landfill surface soils were scraped. Then, two liquids addition pipes and multiple
pressure transducers were placed within the bedding. After bedding material place-
ment, no additional cover was added, and the subsequent waste lift was placed
directly on the bedding. The horizontal blankets allowed large amounts of liquids to
be added with little pressure buildup. Both liquids addition pipes were damaged for
one of the blankets, so liquids addition was not possible through the blanket, dem-
onstrating the critical need for robust construction, pipe redundancy, and opera-
tional care to avoid damage that would preclude liquids addition to large-area
drainage blankets.
80 4 State of Practice
Fig. 4.25 Permeable blanket of shredded tires used for liquids addition (under construction)
In addition to the technical lessons learned at PCNCL with regard to the perfor-
mance of horizontal liquids addition systems, other observations were made in sup-
port of the design and operation of future systems. The system was designed so that
one hydrant would support individual trenches or several trenches at different
depths. As high backpressures often limited the volume of liquid that could be
added, an improved design would allow single hydrants to operate multiple trenches
at the same depth. Fluid conductance values of trenches located deep within the
landfill are relatively low and this limits the introduction of liquids at high pres-
sures. Adding liquids to these lines early in operation before waste depths grow
large would allow for more efficient use. When high pressures are used in these
deep lines for a continued operational period, they cause leachate seepage because
the waste under the side slopes will be more permeable due to decreased overburden
waste depths. Leachate seeps at the base of the landfill was a major issue at PCNCL;
this necessitated construction of a toe drain to assist in collecting leachate and trans-
mitting it to the LCRS (Fig. 4.26). Some of the design and operational recommen-
dations with respect to managing seeps presented in Chap. 11 are a result of lessons
learned at this site.
The Outer Loop Recycling and Disposal Facility (OLRDF) is located in Louisville,
KY, USA, and is owned and operated by Waste Management of Kentucky, Inc. The
facility consists of several different waste management units, including several
landfill cells operated as bioreactor landfills (Fig. 4.27). One landfill area was
4.2 Full-Scale Case Studies: North America 81
Fig. 4.26 Toe drain beneath liner at the base of the PCNCL
constructed with recirculation infrastructure after the bulk of waste filling was
complete (retrofit cell). In another unit, liquid and air addition devices were incorpo-
rated into the waste mass as the routine landfill operation and waste filling took
place (as-built cell). These units, along with a reference cell (Control cell) where no
enhancement techniques were practiced, were studied for more than a decade (Hater
and Green 2003; US EPA 2006; Abichou et al. 2013a, b). Each type of cell (retrofit,
as-built, and control) was built with a quasi-duplicate to facilitate data comparison.
Liquids addition to the retrofit cell was accomplished using deep surface trenches
bedded with tire chips; leachate was treated via external nitrification prior to injec-
tion into the retrofit cell to counteract ammonia buildup (US EPA 2006). Added
liquids included leachate and industrial liquid waste streams. Industrial liquids were
frequently added at or near the working face of the landfill using mobile drip irriga-
tion systems that could be connected to incoming tanker trucks (Fig. 4.28). In the
as-built cells, perforated pipes were placed in horizontal trenches constructed on
top of waste lifts as the landfill was filled (Fig. 4.29). After a lift of waste was placed
on a piping layer, air addition commenced continuously for 30–90 days. The intent
of air addition was to shorten the acid phase of anaerobic waste decomposition by
consuming easily degradable organic waste aerobically. Both leachate and indus-
trial liquids were introduced after air addition ceased (Hater and Green 2003; US
EPA 2006).
Monitoring at the site, documented in detail for periods of operation up to 2006 in
several site reports, included solid waste composition (including moisture content),
leachate volume and constitution, gas production potential of solids, collection rate,
and quality, and surface settlement analysis.
Fig. 4.28 Surface drip liquids addition system at Outer Loop landfill (Photo courtesy of Waste
Management Inc.)
4.2 Full-Scale Case Studies: North America 83
Waste samples collected from 2000 through 2005 showed that the median ratio
of cellulose and hemicellulose to lignin (i.e., C+H/L, degradable to non-
biodegradable) in Retrofit Cells decline was correlated to waste age falling from
approximately 1.5 (1–2 year old waste) to 0.7 (waste age 10 years), in comparison
to C+H/L in waste from Control cells where median levels ranged from approx. 1.3
to 2.4 with no observable correlation to waste age. A decreasing temporal trend for
median C+H/L was observed in As-Built cell A (US EPA 2006). Moisture content
measurements were found to exhibit large spatial variability, particularly within
bioreactor cells and temporal variation with respect to waste age was inconclusive
(US EPA 2006).
Leachate quality and quantity data reported by US EPA (2006) showed that bio-
reactor cells operated with leachate head-on-liner less than the regulatory limits.
The BOD:COD ratio in leachate from as-built cells was indicative of accelerated
organics decomposition. The pre-recirculation leachate treatment provided effec-
tive removal of ammonia as observed in leachate collected from Retrofit cells (US
EPA 2006).
Several years of LFG collection data from the operational period of the bioreac-
tor cells showed a statistically significantly greater gas production rate in the as-
built cell compared to the control cell (Tolaymat et al. 2010). Mean methane yield,
estimated based on the measured BMP of freshly buried waste was 54.8 m3 CH4/Mg
wet waste; this Lo was used to calculate methane first-order generation rates of
0.11 year−1 for the as-built bioreactor cells and 0.06 year−1 for the control cell by
Tolaymat et al. (2010). The enhanced gas generation rate calculated for the As-Built
84 4 State of Practice
cells has important implications for active operations and post-closure operations.
In the case of active operations, the result underscores the importance of early gas
collection for sustainable landfills to reduce fugitive emissions and enhance the via-
bility of beneficial landfill gas use projects if infrastructure and capacity are avail-
able. As for post-closure care implications, the measured decay rate of the As-Built
cells would produce 90 % of the total CH4 production potential in 22 years com-
pared to the Control cell which would take 41 years to produce 90 % of the total
CH4 potential. The implications of enhanced gas productions for landfill operation
are further discussed in Chap. 3.
Measured settlement for the as-built cell was more pronounced than for the con-
trol cell, which was consistent with solids composition data that suggested acceler-
ated decomposition of waste in the as-built cells occurred relative to the control
cells (US EPA 2006; Abichou et al. 2013a). Abichou et al. (2013a) reported liquids
introduction to the retrofit cell produced overall settlement rates ranging from 5 to
8 % of the column height (actual settlement depths of 60–100 cm across the landfill
surface of the Retrofit cell) over 8 years of operation. In comparison, the average
long-term total settlements of as-built and control cells were considerably higher at
37 % and 19 %, respectively.
4.3.1 Europe
Fig. 4.30 Off-gas treatment system at German landfill undergoing aeration (Photo courtesy of
Marco Ritzkowski)
20 m3/ton MSW (wet weight), lower than those peak values reported in small,
lab-scale studies by approximately an order of magnitude (Knox 1999).
A full-scale investigation was conducted at the Sanitary Landfill of Lingen,
Germany; two cells (each approximately 1 ha) were operated as bioreactors to
examine in-situ leachate treatment and leachate recirculation (Kumar et al. 2011).
One cell was constructed with a layer of pre-composted waste placed at the bottom,
and the other was used as a control without the pre-composted layer. Leachate recir-
culation was performed in both cells, and a comparison of leachate composition
over a 4-year monitoring period showed that the use of a compost layer was much
more effective in reducing leachate strength (BOD and COD).
4.3.2 Asia
Like Europe, several countries in Asia have moved largely away from managing
unprocessed domestic waste in landfills (e.g., Japan, Taiwan). One notable sustain-
able landfilling technology from Asia has been the semi-aerobic landfill concept
developed in Japan (Lee et al. 1994; Matsufuji 2004). This technique will be dis-
cussed in detail in Chap. 14, but in brief, the technique includes air venting intro-
duced at the base layer of the landfill, and at times into the waste itself, as a means
of providing leachate treatment within the landfill and promoting waste stabilization
(Matsufuji 2004). Figure 4.31 shows the liner and LCRS of a semi-aerobic landfill
in Japan.
4.3.3 Australia
July 1996 and October 1997, increasing the overall volumetric moisture content
from 27 to 31 % in the test section, while the moisture content in the control cell did
not vary (Yuen et al. 2001). The researchers observed that achieving uniform mois-
ture distribution in the waste was difficult because of the heterogeneous nature of
the landfilled waste, and pointed out limitations associated with applying recircula-
tion test results observed at small-scale test cells to a full-scale landfill (Yuen 2001).
References
Abichou T, Barlaz MA, Green R, Hater G (2013a) The outer loop bioreactor: a case study of settle-
ment monitoring and solids decomposition. Waste Manag 33(10):2035–2047
Abichou T, Barlaz MA, Green R, Hater G (2013b) Liquids balance monitoring inside conven-
tional, retrofit, and Bio-reactor landfill cells. Waste Manag 33(10):2006–2014
EMCON Associates (1975) Sonoma County solid waste stabilization study. Prepared for
Environmental Protection Agency Report No. EPA/530/SW-65D.1 Grant No G06-EC-00351
Augenstein DC, Wise DL, Wentworth RL (1976) Fuel Gas recovery from controlled landfilling of
municipal wastes. Resource Recovery and Conservation 2:103–117
Bareither CA, Benson CH, Barlaz MA, Edil TB, Tolaymat TM (2010) Performance of North
American bioreactor landfills. I: Leachate hydrology and waste settlement. J Environ Eng
136:824–838
Barlaz MA, Milke MW, Ham RK (1987) Gas production parameters in sanitary landfill simulators.
Waste Manag Res 5:27–39
Barlaz MA, Bareither CA, Hossain A, Saquing J, Mezzari I, Benson CH, Tolaymat TM, Yazdani
R (2010) Performance of North American bioreactor landfills. II: Chemical and biological
characteristics. J Environ Eng 136:839–853
Benson CH, Barlaz MA, Lane DT, Rowe JM (2007) Practice review of five bioreactor/recircula-
tion landfills. Waste Manag 27:13–29
Bilgili MS, Demir A, Ozkaya B (2007a) Influence of leachate recirculation on aerobic and anaero-
bic decomposition of solid wastes. J Hazard Mater 143(1–2):177–183
Bilgili MS, Demire A, İnce M, Ozkaya B (2007b) Metal concentrations of simulated aerobic and
anaerobic pilot scale landfill reactors. J Hazard Mater 145:186–194
Bilgili MS, Demire A, Akkaya E, Ozkaya B (2008) COD fractions of leachate from aerobic and
anaerobic pilot scale landfill reactors. J Hazard Mater 158:157–163
Burns and McDonnell Engineering Company Inc. (2014) Leachate management summary, pre-
pared for Crow Wing County Sanitary Landfill, Brainerd. Project No. 77272
Caine M, Campbell D, Van Santen A (1999) The landfill gas timeline: the Brogborough test cells.
Waste Manag Res 17:430–442
Cho Y (2010) Investigation of geotechnical and hydraulic aspects of landfill design and operation.
Ph.D. Dissertation, University of Florida, Gainesville
Clarke WP (2000) Cost-benefit analysis of introducing technology to rapidly degrade municipal
solid waste. Waste Manag Res 18:510–524
Doran, F. (2007). Better life through chemistry- Leachate treatment at the Crow Wing County
Landfill. In association with RW Beck, Wastecon
Eliasen R (1945) Decomposition in land-fills. Am J Public Health 32:1029–1037
EPA Victoria. (2010) EPA Publication 788.1, Best practice environmental management- siting,
design, operation and rehabilitation of landfills
Faour AA, Reinhart DR, You H (2007) First-order kinetic gas generation model parameters for wet
landfills. Waste Manag 27:946–953
References 89
Farquar GF, Rovers FA (1973) Gas production during refuse decomposition. Water Air Soil Pollut
2:483–495
Gawande NA, Reinhart DR, Thomas PA, McCreanor PT, Townsend TG (2003) Municipal solid
waste in situ moisture content measurement using an electrical resistance sensor. Waste Manag
23:667–674
Hater G, Green R (2003) Landfills as bioreactors: research at the outer loop landfill, Louisville,
Kentucky. Waste Management Inc., EPA/600/R-03/097
Hudson AP, White JK, Beaven RP, Powrie W (2004) Modelling the compression behaviour of
landfilled domestic waste. Waste Manag 24(3):259–269
Jain P, Kim H, Townsend T (2005a) Heavy metal content in soil reclaimed from a municipal solid
waste landfill. Waste Manag 25:25–35
Jain P, Powell J, Townsend T, Reinhart D (2005b) Air permeability of waste in a municipal solid
waste landfill. J Environ Eng 131:1565–1573
Jain P, Farfour WM, Jonnalagadda S, Townsend T, Reinhart DR (2005c) Performance evaluation
of vertical wells for landfill leachate recirculation. In: Proceedings of Geo Frontier 2005,
ASCE conference, Austin
Jain P, Powell J, Townsend T, Reinhart D (2006) Estimating the hydraulic conductivity of land-
filled municipal solid waste using the borehole permeameter test. J Environ Eng 132:645–652
Jain P, Ko JH, Kumar D, Powell J, Kim H, Maldonado L, Townsend T, Reinhart DR (2014) Case
study of landfill leachate recirculation using small-diameter vertical wells. Waste Manag
34(11):2312–2320
Jonnalagadda S, Kumar D, Jain P, Gawande N, Townsend TG, Reinhart D (2010) Comparison of
resistivity and time domain reflectometry sensors for assessing moisture content in bioreactor
landfills. Geotech Test J 33(3):183–191
Kadambala R, Townsend TG, Jain P, Singh K (2011) Temporal and spatial pore water pressure
distribution surrounding a vertical landfill leachate recirculation well. Int J Environ Res Public
Health 8:1692–1706
Kim H (2005) Comparative studies of aerobic and anaerobic landfills using simulated landfill
lysimeters. Doctoral thesis, University of Florida, Gainesville
Kim H, Townsend T (2012) Wet landfill decomposition rate determination using methane yield
results for excavated waste samples. Waste Manag 32:1427–1433
Kim H, Jang Y-C, Townsend T (2011) The behavior and long-term fate of metals in simulated
landfill bioreactors under aerobic and anaerobic conditions. J Hazard Mater 194:369–377
Knox K (1998) Practical benefits for the waste industry from the UK’s landfill test cell programme.
Waste Management, p 18–19
Knox K (1999) A review of the Brogborough and Landfill 2000 test cell monitoring data.
Environment Agency R&D Technical Report P231
Ko JH, Powell J, Jain P, Kim H, Townsend T, Reinhart D (2013) Case study of controlled air
addition into landfilled municipal solid waste: design, operation, and control. J Haz Toxic
Radioactive Waste 14(4):351–359
Kumar S (2009) Study of pore water pressure impact and fluid conductance of a landfill horizontal
liquids injection system. Master of Engineering Thesis, University of Florida, Gainesville
Kumar D, Jonnalagadda S, Jain P, Gawande NA, Townsend TG, Reinhart DR (2009) Field evalu-
ation of resistivity sensors for in situ moisture measurement in a bioreactor landfill. Waste
Manag 29:1547–1557
Kumar S, Chiemchaisri C, Mudhoo A (2011) Bioreactor landfill technology in municipal solid
waste treatment: an overview. Crit Rev Biotechnol 31(1):77–97
Larson J (2007) Investigations at a bioreactor landfill to aid in the operation and design of horizon-
tal injection liquids addition systems. Master of Engineering Thesis, University of Florida,
Gainesville
Larson J, Kumar S, Gale SA, Jain P, Townsend T (2012) A field study to estimate the vertical gas
diffusivity and permeability of compacted MSW using a barometric pumping analytical model.
Waste Manag Res 30:276–284
90 4 State of Practice
Townsend TG, Miller WL, Lee H-J, Earle JFK (1996) Acceleration of landfill stabilization using
leachate recycle. J Environ Eng 122(4):263–268
US EPA (2006) Landfill bioreactor performance: second interim report outer loop recycling &
disposal facility Louisville, Kentucky. EPA/600/R-07/060. U.S. Environmental Protection
Agency, Cincinnati
US EPA (2007) Bioreactor performance summary paper. EPA530-R-07-007. US EPA Office of
Solid Waste, Municipal and Industrial Solid Waste Management Division, Washington, DC
Watson RP (1987) A case study of leachate generation and recycling at two sanitary landfills. In:
Proceedings from the technical sessions of the GRCDA 25th annual international seminar,
equipment, services, and systems show 1987. Saint Paul, MN
Yazdani R, Kieffer J, Akau H (2002) Full scale landfill bioreactor project at the Yolo County cen-
tral landfill: final report. Yolo County, Planning and Public Works Department. EPA Project XL
progress report
Yazdani R, Kieffer J, Sananikone K, Augenstein D (2006) Full scale bioreactor landfill for carbon
sequestration and greenhouse emission control: final technical progress report. Yolo County,
Planning and Public Works Department. D.O.E. Award Number DE-FC26-01NT41152
Yazdani R, Mostafid ME, Han B, Imhoff PT, Chiu P, Augenstein D, Kayhanian M, Tchobanoglous
G (2010) Quantifying factors limiting aerobic degradation during aerobic bioreactor landfill-
ing. Environ Sci Technol 44:6215–6220
Yuen STS (2001) Bioreactor landfills: do they work? 2nd ANZ Conference on Environmental
Geotechnics, Newcastle, Australia
Yuen STS, Styles JR, Wang QJ, McMahon TA (1997) The design, construction and instrumenta-
tion of a full-scale bioreactor landfill. GeoEnvironment 97, 1st Australia-New Zealand
Conference on Environmental Geotechnics, p 345–352, Melbourne, November 1997
Yuen STS, Wang QJ, Style JR, McMahon TA (2001) Water balance comparison between a dry and
a wet landfill—a full-scale experiment. J Hydrol 251:29–48
Zhang H, He P, Shao L (2008) Methane emission from MSW landfill with sandy soil covers under
leachate recirculation and subsurface irrigation. Atmos Environ 42(22):5579–5588
Chapter 5
Landfill Constituent Relationships
and Dynamics
MSW landfills, by their very nature, are dynamic systems. Liquids move through
the waste mass in response to moisture infiltration from rainfall, drainage of liquids
already entrained within the waste, or liquids purposely introduced to manage
leachate or promote rapid waste stabilization. Solid waste components are con-
verted to various gases as a result of biological decomposition, and the gases move
through the waste mass in response to internal pressure buildup and external atmo-
spheric changes. The waste mass itself changes over time as waste mass is lost,
mechanical stresses redistribute, and liquid and gas content changes and redistrib-
utes. This chapter provides a discussion of the fundamental relationships among
solid, liquid and gas phases. As appropriate, magnitudes of pertinent waste charac-
teristics are summarized. A basic discussion of liquid, gas, and solid movement at
landfills is also described. The objective of this discussion is to provide sufficient
fundamentals to prepare the reader for subsequent, more detailed discussions on
these phase dynamics presented later in the book, and to provide the reader with
necessary background to understand the potential applications of some of the more
advanced constituent dynamics simulation tools that are available.
Many of the basic characteristics of landfilled waste that play a role in the design of
sustainable landfill technologies relate one landfill constituent to another. Borrowing
from classic soil mechanics literature, it is helpful to illustrate the relationship of
these different phases in a constituent phase diagram (see Fig. 5.1).
The three primary phases are solid, liquid and gas. The solid phase consists of the
waste components and cover soil. Water represents the liquid phase, although this
water (i.e., leachate) will contain dissolved and suspended materials. At some point
in a landfill’s life, the gas phase may be predominantly represented by chemicals
similar to atmospheric air, but over much of the landfill’s existence, gaseous prod-
ucts of biological reactions will dominate the gas phase (e.g., methane, carbon diox-
ide). The phase diagram provides nomenclature for mass components on the left
side, while volume components are represented on the right side. In the following
sections, some of the more important material properties are described, with many
of these defined by the mass and volume components in Fig. 5.1.
5.2.2 Density
Density relates the mass of a material to its respective volume, and following
conventions illustrated in Fig. 5.2, is defined as:
MT
rT = (5.1)
VT
Fig. 5.2 Dry density of processed, unprocessed and landfill excavated MSW as a function of
applied stress as measured by Beaven (2000)
where ρT refers to the total density (i.e., bulk density) of the media (kg/m3 or lb-m/ft3),
MT is the total (wet) mass of the media (kg or lb-m) (the mass of gaseous phase is
insignificant compared to solid and liquid phases), and VT is the total volume of the
media (m3 or ft3). A related term, specific weight, relates the weight of material to
its respective volume:
WT
gT = (5.2)
VT
where γT is the total specific weight of the material (kN/m3 or lb/yd3) and WT is the
total (wet) weight of the media (kN or lb). Density and specific weight relate to one
another as follows:
g T = g rT (5.3)
where g is the gravitational constant. The term density will be more commonly used
in this book as a generic tem referring to both parameters. In engineering calcula-
tions involving force or weight, specific weight is the correct parameter to use.
The density of waste plays an important role in many landfill design procedures.
It is especially important in sustainable landfill operation where a primary objective
is waste transformation (stabilization), a process through which density changes.
It is also important to recognize the role that cover soil plays in determining the
density of the landfilled mass (waste plus soil). Soil is substantially denser than
most municipal wastes. As the relative amount of soil increases in a landfill (which
will happen as waste decomposition proceeds), the density of the landfill increases.
96 5 Landfill Constituent Relationships and Dynamics
Table 5.1 Densities and specific weights of the primary materials in a landfill under common
landfill conditions
Material ρT (kg/m3) ρT (lb-m/ft3) γT (kN/m3) γT (lb/yd3)
Water 998 62.4 9.77 1,685
Soil 1,680–1,920 105–120 16.4–18.8 2,835–3,240
Newly compacted waste (no soil) 610–710 38–44 6–7 1,030–1,198
(Kavazanjian 2001)
Excavated landfill waste (near the 1,017–1,220 64–76 10–12 1,716–2,059
surface of the landfill) (Zornberg
et al. 1999)
Excavated landfill waste (deep 1,330–1,530 83–96 13–15 2,245–2,582
in the landfill) (Cowland et al. 1993)
Crude, domestic waste, retrieved 500–1,180 31.2–73.7 4.90–11.6 842.8–1,989
from the tipping face of a landfill
(Powrie et al. 1998)
Landfilled waste and cover soil 1,150 71.8 11.3 1,938
(Hull et al. 2005)
Landfilled MSW most recently 240–1,260 15–78.7 2.35–12.4 404.5–2,124
deposited 2 years prior to sampling
(sample depths: 1.5–6 m)
(Chiemchaisri et al. 2007)
Landfilled samples extracted via 1,088–2,350 67.9–147 10.67– 1,834–3,961
borehole every 1 m depth from 23.05
a landfill closed in 1985
(Al-Yaqout et al. 2007)
Transfer station collected MSW 906.5–1,071 56.8–66.8 8.89–10.5 1,528–1,805
(Penmethsa 2007)
Landfilled MSW (Alaska), soil, 530 33.1 5.2 893.7
and subgrade soils (Hanson
et al. 2008)
Landfilled MSW (Michigan), soil, 999 62.4 9.8 1,685
and subgrade soils (Hanson
et al. 2008)
Landfilled MSW extracted 1,326–1,785 82.8–111 13–17.5 2,235–3,009
via borehole (Machado et al. 2010)
Large-scale MSW bioreactor 510–714 31.8–44.6 5.0–7.0 860–1,204
lysimeter (Bareither et al. 2012a)
Landfilled MSW in Spain 1,530–2,141 95.5–134 15–21 2,579–3,609
(Yu et al. 2012)
Landfilled MSW densified in-place 1,590–2,130 99.3–133 15.6–20.9 2,680–3,590
via rolling and dynamic compaction
(Yu-xin et al. 2013)
5.2 Fundamental Properties of Landfill Waste 97
and ultimately settles, density will increase as the moisture content increases, the
fraction of less dense materials is reduced (e.g., paper), and the relative abun-
dance of cover soil increases.
Waste density is heavily influenced by overburden pressure, that is, the pressures
imposed by overlying materials (waste and cover soil) due to their weight. It is well
understood that landfilled material deep in the disposal unit, especially after a
degree of stabilization has occurred, will be much denser than recently-compacted
waste on the surface. Several researchers have measured the relationship between
density and applied stress; the experimental devices used for these measurements,
often described as an oedometer or a compression cell, allow measurement of waste
density (and other parameters of interest) under conditions of different applied ver-
tical stresses. Beaven (2000) conducted a comprehensive evaluation of waste char-
acteristics resulting from the application of various pressures in a large-scale
compression cell (2 m diameter), with the capability to convey leachate. The cell
holds approximately 3,000–6,500 kg of waste (at field capacity) and is equipped
with piezometers to monitor the leachate head within the cell. Three different waste
streams were assessed using this device: a new sample of MSW (i.e., unprocessed),
a processed sample of new MSW (shredded), and a waste sample excavated from a
landfill (partially degraded). Waste samples were either loaded at a specified in situ
density (0.5 ton/m3) or at a low pressure (approximately 5 bars), at a waste height of
2.5 m, and then subjected to increasing applied pressure (up to five successive load
increases from the initial 40 kPa) to characterize waste dry density in response to
increased applied pressure. The results of this characterization are presented in
Fig. 5.2; units of force are presented in both kPa and lb/ft2 (psp), while units of
density are presented in both Mg/m3 and lb-m/ft3 (pcf). Beaven (2000) also mea-
sured other waste parameters as a function of applied pressure (e.g., effective poros-
ity, leachate volume extracted from waste during compression) and several of these
will be presented later in this chapter.
Dry density is a useful way of presenting density data, as wet weight densities
can be subsequently estimated for different moisture contents; it is important to
note, however, that the degree of compaction that can be achieved will be influenced
by the moisture conditions during testing. Figure 5.3 presents the wet-weight densi-
ties at different applied loads measured by Beaven (2000) under conditions where
the waste was at field capacity. McKnight (2005) used a smaller-scale oedometer (a
hydraulic press (10-ton cylinder jack and hand pump) coupled with a steel cylinder
to contain the waste sample (0.43-m diameter)) to measure the relationship between
density and applied pressure for samples of differing ages and states of decomposi-
tion excavated from a landfill in Florida, US. Results from these experiments are
presented in Fig. 5.4, and include for comparison data points for soil, cardboard,
compost, and lines from Beaven’s experiments.
Density estimates such as those presented in Table 5.1 and Figs. 5.1, 5.2, and 5.3,
along with waste and facility-specific density information, can be used to provide
better inputs for estimating landfill capacity, designing LCRS and landfill
foundations (Chap. 10), and projecting materials recovery amounts in landfill min-
ing operations (Chap. 17).
98 5 Landfill Constituent Relationships and Dynamics
Fig. 5.3 Wet density (at a moisture content of field capacity) of processed, unprocessed and land-
fill excavated MSW as a function of applied stress as measured by Beaven (2000)
Fig. 5.4 Dry density of excavated landfill waste samples (McKnight 2005) as a function of applied
load. For comparison purposes, data from Beaven (2000) and measured for compost and cardboard
included
5.2.3 Porosity
Porosity is the fraction of the volume of the void space to the bulk volume of a
porous medium, and is defined as:
VV
h= (5.4)
VT
5.2 Fundamental Properties of Landfill Waste 99
Fig. 5.5 Drainable porosity measured as a function of applied stress as measured by Beaven
(2000)
Several attempts have been made to measure the porosity of MSW. Like density,
porosity will be influenced by applied stress (overburden pressure), with an observed
decrease as stress increases. Korfiatis et al. (1984) constructed a column of waste to
investigate infiltration rates and porosity by measuring the volume of water required
to saturate the waste and found porosity values to vary from 50 to 60 %. Zeiss and
Major (1993) investigated porosity variations as a function of the degree of compac-
tion and found that porosity ranged from 47 % at high compaction to 58 % at lower
compaction. Zornberg et al. (1999) investigated porosity in relation to confining pres-
sure in a landfill undergoing vertical expansion and estimated a range of 49–62 %
based on the specific weight of the waste and the applied overburden pressure.
The term drainable porosity is also used in some cases to describe void space,
and as defined by Beaven (2000) is the volume of water released from a unit of
volume of fully saturated material that is allowed to drain freely under gravity. Also
referred to as effective porosity, the drainable porosity is analogous to the concept
of specific yield in hydrology. Figure 5.5 presents the range of drainable porosity
measurements reported by Beaven (2000) as a function of applied pressure.
The term moisture content refers to the amount of water (mass or volume) contained
in a matrix (soil, waste) relative to the total mass of that medium. Moisture content
may be defined differently depending on the application or discipline, thus it is
100 5 Landfill Constituent Relationships and Dynamics
important to verify the definition of any moisture content value provided. Most
engineers and facility operators, when discussing the moisture content of solid
waste, refer to a weight-based definition, relative to the total weight of the media:
MW
MC = (5.5)
MT
where MC = gravimetric moisture content (wet weight basis), MT = bulk weight of
the landfilled waste, and MW = weight of moisture. This value is measured by
weighing a wet sample of waste, drying the waste in an oven and measuring the
weight of moisture (normally water) that evaporated, and dividing the moisture
weight by the total weight. Typical values of wet-weight moisture content for MSW
(as disposed) are provided later in this section. In some applications it may be com-
mon to encounter a moisture content that relates the weight of water in a medium to
the dry weight of solids. This parameter is referred to herein as water content, and
is defined as
MW
w= (5.6)
MS
Where w = water content, MW is the mass of water, and MS is the mass of dry solids.
This term is routinely used to characterize soil relationships such as soil
compaction.
The volumetric moisture content (θ) refers to the volume of water occupied by
the volume of a medium. Volumetric moisture content is the parameter used when
modeling liquids flow through a porous media, and is defined as:
VW
q= (5.7)
VT
where θ = volumetric moisture content; VT = total volume of the landfilled waste;
VW = volume of water (or liquids). The following relationship allows conversion
between MC and θ.
rW
MC = qw (5.8)
rT
Where ρT = bulk density of the landfilled waste and ρW = density of water.
The moisture content of waste when disposed depends on the composition of the
waste, climatic conditions, and landfilling practices such as surface water manage-
ment. Table 5.2 presents initial MC of MSW from several locations as reported in
the literature.
Table 5.2 Reported moisture content of as-disposed MSW
Location and reference Moisture content (wet weight basis) Remarks
New Jersey (Korfiatis et al. 1984) 44.3 % Samples collected from a local landfill
United States (Tchobanoglous 15–40 % Municipal waste in compactor truck or normally or well-compacted
et al. 1993) in a landfill
Pennsylvania (Gabr and Valero 1995) 23.1 % (near surface) to 56.5 % (20 m deep) Samples collected from a waste profile from a landfill opened
in 1940
Florida (Townsend et al. 1996) 31.3 % (13), 29.7 % (6), 27.6 % (11) Samples were collected from control and bioreactor area prior
to leachate recirculation
California (El-Fadel 1999) 26–46 % Samples collected before leachate recirculation
California (Zornberg et al. 1999) 3.5–50 % (average 28 %) 80 Samples collected from a landfill
South Korea (Jang et al. 2002) 36.0 % Borehole extracted samples from a municipal landfill
California (Mehta et al. 2002) 11.8–26.7 % 11 Samples
Florida (Jonnalagadda 2004) 23 % (11.5–36.8 %) 51 Samples from a lined MSW landfill before leachate recirculation
5.2 Fundamental Properties of Landfill Waste
New Jersey (Hull et al. 2005) 28.3 % (18.8–41.6 %) 98 Samples from 13 gas extraction well borings
Florida (McKnight 2005) 24.1–43.4 % 17 Samples from a part of 40-ha unlined landfill
New York (Harris et al. 2006) 26.0–44.2 % Eight samples, collected via hollow-stem auger
36.4–76.4 % 12 Samples, collected via bucket auger
Kuwait (Al-Yaqout et al. 2007) 2.1–37.9 % (mean approx. 13.5 %) 13 Borehole extracted samples from a landfill
Thailand (Chiemchaisri et al. 2007) 39.6–60.1 % Landfilled MSW most recently deposited 2 years prior to sampling
Mean of three samples collected
Illinois (Reddy et al. 2009a, b) 30.6 % Four 1.5-year-old samples collected from the landfill with
boreholes at 20 m depth (samples >5 kg)
France (Stoltz et al. 2010a, b) 35.8 % Fresh MSW (sample 150 kg)
California (Zekkos et al. 2010a, b) 21.1 % Samples extracted from landfill via bucket auger (samples 5–10 kg)
Texas (Hossain and Haque 2012) 15 % MSW collected from a transfer station
China (Yu-xin et al. 2013) Compacted: 5.3–22.7 % In situ landfilled MSW from boreholes
Non-compacted: 6.0–35.9 %
101
102 5 Landfill Constituent Relationships and Dynamics
Fig. 5.6 Field capacity (% volume) measured for processed, unprocessed and landfill excavated
MSW as a function of dry density (Beaven 2000)
Fig. 5.7 Waste field capacity as a function of dry waste density (based on several different studies)
fundamentals and predictive techniques is beyond the scope of this book, and the
reader is referenced to the many excellent works on fluid flow through porous media
(e.g., Bear 1972; Fetter 2001; Pinder and Celia 2006; Todd and Mays 2005). The
objective of the following section is to provide sufficient background so the designer
of sustainable landfill technologies understands the underlying physics of how
moisture moves within a landfill and the tools that can be used in design.
Conditions within a landfill with high moisture levels, either from added liquids or
inherent with waste and precipitation, will, in some places, likely be saturated. The
designer must understand, however, that unsaturated conditions may dominate much
of the landfill. A discussion of fluid flow fundamentals, however, will relate to most
engineers best by first starting with a description of saturated fluid dynamics.
Darcy’s law states that the saturated flow rate through a porous media (Q) is
proportional to the cross-sectional area of flow (A) and the hydraulic gradient
(∂h/∂l), which in turn relates the change in potentiometric head over a unit length.
As such, Darcy’s law for one-dimensional flow may be written as:
Q ¶h
q= =- K (5.9)
A ¶l
where Q is the total flow rate (L3/T) through a cross-sectional area A (L2); ∂h/∂l is
the pressure gradient (unitless, potentiometric pressure head, h/path length, l); q is
the specific discharge (L/T; also commonly referred to as the Darcy flux or Darcy
velocity); and K is a proportionality constant called the hydraulic conductivity (L/T).
The specific discharge does not correspond to the actual velocity of the moving
fluid. The true velocity (v) would be greater, and may be determined from specific
discharge as follows:
q
v= (5.10)
h
¶h
qx = - K x (5.11)
¶x
5.3 Moisture Movement 105
¶h
qy = - K y (5.12)
¶y
¶h
qz = - K z (5.13)
¶z
where qx, qy, and qz represent the specific discharge in the x, y and z directions, and
Kx, Ky, Kz represent the hydraulic conductivity in the x, y and z directions. As will
be discussed later in this section, the hydraulic conductivity of landfilled waste dif-
fers depending on the direction considered (i.e., waste is anisotropic with respect to
hydraulic conductivity).
The law of conservation of mass can be used to derive a governing equation for
the fluid flow scenarios of interest. For example, when conserving mass around a
three-dimensional control volume, the following equation can be written:
é ¶r q x ¶r q y ¶r qz ù ¶rq
-ê + + ú= +w (5.14)
ë ¶x ¶y ¶z û ¶t
where ∂(ρθ)/∂t refers to the increase in storage in the control volume, the combined
é ¶r q x ¶r q y ¶r qz ù
term - ê + + ú represents net flow of the fluid into the control vol-
ë ¶x ¶y ¶z û
ume, w refers to the fluid generation rate by the control volume (e.g., gas generation
by waste), and t is time.
There are three modes by which the fluid storage of a porous media can change.
First, as the name suggests, a porous media possesses void spaces from which fluid
can be stored or withdrawn. Second, porous media is compressible (i.e., undergoes
a change in volume with a change in pressure). An increase in the fluid pressure in
the pores causes the pores to expand and store more fluid volume. An increase in
pore pressure results in media compression and a consequential reduction in poros-
ity. Finally, the storage can change because of the compressibility of fluid. Under
larger pressure, fluid compresses (i.e. density increases) and occupies a smaller pore
space volume and, as a result, more fluid mass can be stored in the same volume of
pore space (Bear 1972, 1979). The significance of each of the three modes for fluid
storage depends on the pressure driving fluid flow. For flow of a compressible liquid
in compressible media (e.g., water flow in a confined aquifer system), ∂(ρθ/∂t) = ρSs
∂(h/∂t) (Ss is the specific storage). Substituting Darcy’s equation in (5.14), this rela-
tionship for flow of compressible fluid in compressible media (assuming variation
in ρ with respect to x, y, and z are negligible) may be further described as:
¶ æ ¶h ö ¶ æ ¶h ö ¶ æ ¶h ö ¶h
Kx + ç Ky ÷+ Kz = Ss +w (5.15)
¶x çè ¶x ÷ø ¶y è ¶y ø ¶z çè ¶z ÷ø ¶t
106 5 Landfill Constituent Relationships and Dynamics
Fig. 5.8 Hydraulic conductivity measured as a function of dry density (Beaven 2000)
As indicated earlier, much of the moisture movement in a landfill will not be satu-
rated, and thus tools and techniques that incorporate unsaturated flow principles
must be discussed. First, a discussion of the differences between saturated and
unsaturated flow is warranted. While the potentiometric head (h) in saturated flow
includes both a liquid pressure (P) and elevation (z), liquid pressure will not be posi-
tive under unsaturated flow conditions. In fact, a suction head (ψ) resulting from
capillary forces will serve along with the elevation head as the components of
108 5 Landfill Constituent Relationships and Dynamics
¶ (y + z )
qz = - K (q ) (5.16)
¶z
Inserting this into the governing equation for Darcy’s law presented earlier, the
equation can be written as:
¶ æ ¶y ö ¶ æ ¶y ö ¶ æ ¶y ö ¶K z ¶q
¶x çè
Kx
¶x ÷ + ¶y ç K y ¶y ÷ + ¶z ç K z ¶z ÷ + ¶z = ¶t (5.17)
ø è ø è ø
This is commonly referred to as the Richards equation. Both K and ψ vary as a
f unction of the moisture present in the unsaturated media.
Several different relationships have been developed to describe ψ and K as a
function of volumetric moisture content (θ). Two of the more common ones, the
Brooks-Corey equation and the van Genuchten equation, are presented in Table 5.6.
The retention of soil water and subsequently porous media water content in both
equations is a function the of matric potential of the porous media. In the Brooks-
Corey model, the relationship between these two variables plots as a straight line on
a log–log plot, with the slope of the line represented by λ (the pore size distribution
factor). In the van Genuchten model, the relationship between volumetric water
content and the log of the matric potential yields a more complex plot that tends to
work well for most soils. The van Genuchten equation is generally applicable to a
larger range of environmental data, while the Brooks-Corey model tends to fit
coarse soils within a narrow pore size distribution range (high λ values); however,
the Brooks-Corey equation yields functions that are mathematically easier to
manipulate (Stankovich and Lockington 1995); several authors have documented
parameter equivalencies for moving from one framework to the other (Lehnard
et al. 1989; Stankovich and Lockington 1995).
Unsaturated flow is a special case of simultaneous flow of two immiscible fluids
(air and water), where the non-wetting phase (air) is assumed to be stagnant and its
pressure is assumed to be zero everywhere in the porous media. More detailed dis-
cussion of the wetting phase, non-wetting phase and unsaturated flow can be found
in other sources (Bear 1979; Stephens 1995) and is briefly described at the end of
this chapter.
The Richards equation has been used in the past for predicting leachate genera-
tion from MSW landfills (Ahmed et al. 1992; Korfiatis et al. 1984; Schroeder et al.
1994; Straub and Lynch 1982). However, the Richards equation does not account
for changes in fluid storage in the medium due to deformation of the medium and
compressibility of the fluid that can result from high pore fluid pressure. The fluid
pressure in conventional dry landfills is expected to be low, thus changes in fluid
storage due to deformation of media and compressibility of fluid realistically can be
neglected. The fluid pressure that would be encountered in landfills where liquids
are actively added into the waste mass will likely be high, especially in the vicinity
5.3 Moisture Movement 109
K (q ) æ q - qr ö
n
j = porosity
÷ = ( Se )
n
=ç
KS è j - q r ø
of liquids introduction devices. The change in fluid storage due to media deformation
and fluid compressibility, therefore, might be considerable at such facilities.
Stephens (1995) proposed a modified version of the Richards equation (5.22)
that also accounts for liquids storage due to deformation of the media and compress-
ibility of fluid. The modified equation is more appropriate for modeling the fluid
flow as part of landfill facilities where sustainable practices such as liquids addition
are implemented. McCreanor (1998) and Jain et al. (2005a, b) used (5.22) for simu-
lating fluid flow from liquids introduction systems in bioreactor landfills:
¶ æ ¶y ö ¶ æ ¶y ö ¶ æ ¶y ö ¶K z ¶y
¶x çè
Kx
¶x ÷ + ¶y ç K y ¶y ÷ + ¶z ç K z ¶z ÷ + ¶z = ( C + b Ss ) ¶t (5.18)
ø è ø è ø
where β is a constant equal to 1 when p ≥ 0 and 0 when p < 0.
110 5 Landfill Constituent Relationships and Dynamics
Once a governing equation for moisture flow has been established, it can be solved
for the resultant moisture movement for specific scenarios of interest and associated
boundary conditions. While analytical solutions to some problems may be avail-
able, in most cases, complex problems require the use of a numerical flow model.
Table 5.7 summarizes several fluid flow modeling studies related to landfills and
their application. The remainder of this section focuses on two specific, commonly
used models: the Hydrologic Evaluation of Landfill Performance (HELP) and
SEEP/W. SEEP/W is focused upon because much of the work presented elsewhere
in this book was developed using this program; similar software packages are
expected to yield comparable results.
SEEP/W, a commercial program from Geo-Slope International (Alberta, Canada),
has been used to numerically simulate subsurface liquids addition. SEEP/W is a
finite element model and numerically solves the Richards equation for saturated and
unsaturated water flow. The model has been used to analyze groundwater seepage
and excess pore-water pressure dissipation problems within porous media (Jain et al.
2010a, b). SEEP/W can model various material types and boundary conditions such
as unsaturated soils, different injection pressures, groundwater seepage, and excess
pore-water pressure dissipation problems within porous media (Hughes and Sanford
2004). The governing differential equation used in SEEP/W is as follows:
¶ æ ¶H ö ¶ æ ¶H ö ¶q
ç Kx ÷ + ç Ky ÷ +Q = (5.19)
¶X è ¶X ø ¶Y è ¶Y ø ¶t
Where H is the total head (L); Kx and Ky are the hydraulic conductivities in hori-
zontal and vertical directions, respectively (L T−1); Q is liquid flux (L3 T−1); θ is
the volumetric water content; and t is time (T). The van Genuchten function for
soil-water characteristic curves and the van Genuchten-Mualem function for the
calculation of relative permeability are one option in SEEP/W and were used for
the simulations presented in this book. Additional details on SEEP/W can be
found in Krahn (2007).
The HELP model is a computerized water-budget program for landfills that the
US Army Corps of Engineers developed for the United States Environmental
Protection Agency (USEPA). The HELP model is the predominant tool used by
engineers in North America to estimate the leachate production rate from lined
landfills as a function of weather data (precipitation, evapotranspiration, wind, and
temperature) and landfill design and operating parameters (daily cover, area etc.).
HELP is a quasi-two-dimensional model that simulates one-dimensional flow in
cover soil and waste layers and two-dimensional flow in the drainage layer
(Schroeder et al. 1994). HELP serves as a tool for the rapid modeling a landfill’s
water balance and uses a mixture of empirical and numerical modeling to estimate
moisture inputs, moisture retention, and moisture transport into, within, and out of
a landfill. Most state landfill permitting agencies in the US require that the HELP
model be performed as part of a landfill permit application.
Table 5.7 Moisture flow models used to simulate specific landfill scenarios
Study Model Application
McCreanor and Reinhart SUTRA—USGS unsaturated and saturated SUTRA was used to model leachate distribution around vertical and
(1996, 2000) flow model with capabilities for density horizontal leachate recirculation devices for a range of liquids addition
variation as well as solute and energy rates, waste and cover permeability, waste anisotropy, waste
transport modeling heterogeneity, and dosing frequency
Beaven (2000) MODFLOW—USGS, 3-D groundwater flow MODFLOW was used to model infiltration into landfill with
model, used for simulating steady state and hydraulic conductivity varied as a function of effective stress
transient flow
Oldenburg (2001), Kling and TOUGH(2)—Transport of unsaturated TOUGH(2) incorporates gas/liquid partitioning and was applied to a
Korkealaakso (2006) groundwater and heat-landfill gas migration bioreactor landfill with a module devoted to bioreactor option (T2LBM)
which uses a Monod kinetic rate for biodegradation (acetic acid used
as a proxy for degradable waste) either by aerobic or anaerobic means
(Oldenburg 2001). Kling and Korkealaakso used the model to plan a
bioreactor monitoring system (sensors and located in Finland)
Bachus et al. (2002) VS2DI—a finite difference model that Bachus et al. (2002) used VS2DI to simulate the liquid flow from
simulates liquid flow and solute or energy horizontal trenches in a homogenous and isotropic media to estimate
transport in saturated-unsaturated lateral and vertical extents of the zone of impact as a function
porous media of injection pressure for intermittent system operation
Haydar and Khire (2007), Khire and HYDRUS—A finite element model for Khire and Haydar (2005) numerically modeled steady-state fluid flow from
Mukherjee (2007), Haydar and Khire simulating the two-dimensional movement horizontal trenches as a function of injection pressures, trench geometry
(2007), Khire and Kaushik (2012) of water, heat, and multiple solutes and size, hydraulic conductivity of the trench backfill, and horizontal and
in variably saturated media. The model vertical trench spacing for isotropic waste. Khire and Mukherjee (2007)
includes a parameter optimization simulated the impact of the leachate injection rate on the steady-state
algorithm for inverse estimation of a injection pressure, lateral extent of moisture movement, and head on the
variety of soil hydraulic and/or solute bottom liner for isotropic waste. The impact of well radius, well depth,
transport parameters and screen length, and dosing frequency were also investigated
Jain et al. (2010a, b) SEEP/W simulates liquid flow in saturated Jain et al. (2010a, b) used SEEP/W for modeling liquids distribution
and unsaturated porous media using a around horizontal and vertical sources for a range of injection pressure,
combination of finite element and finite waste permeability and anisotropy, and source dimensions. The modeling
difference methods results were used to develop generalized design charts to estimate flow
rate, lateral and vertical extents of the zone of impact as a function of
injection pressure and the added liquids volume
112 5 Landfill Constituent Relationships and Dynamics
In HELP, the landfill’s vertical cross section is defined in the soil properties input
file as layers, classified in one of the four predefined types in the model (Fig. 5.9).
The layer types identified by the model are vertical percolation, lateral drainage,
barrier soil, and geomembranes. The user has the option to specify the initial mois-
ture of different layers or use model default values for the initial moisture content
for each layer based on the wilting point, field capacity, and total porosity for each
layer of material. In the HELP model, surface water runoff is forecasted using the
US Soil Conservation Service’s curve number method, infiltration, evapotranspira-
tion, and percolation are simulated using the Darcy equation adapted for unsatu-
rated conditions. Lateral drainage is determined using the Boussinesq equation
adapted for landfill conditions. Liner leakage is estimated using Darcy’s equation
for saturated conditions. Evapotranspiration is modeled by a plant growth and decay
model for perennial and annual crops. Vertical flow is modeled in the vertical direction
5.3 Moisture Movement 113
with the gravitational and hydrostatic potentials considered as driving forces. While
nowhere near as powerful for simulating liquids flow in landfills as the numerical
software packages described in Table 5.7, HELP does provide several opportunities
for the designer to incorporate liquids addition (Xu et al. 2012), and these will be
discussed in detail in Chap. 10.
Future chapters illustrate the application and output of fluid flow models as part of
the landfill design process. A number of operating conditions (added flow rates,
pressures) and landfill properties (initial moisture content, hydraulic conductivity)
must be defined as part of the specific application simulated. In this section, exam-
ple results from several simulations of liquids addition into landfills (using SEEP/W
model runs for horizontal trenches, vertical wells, and horizontal blankets) are pre-
sented to provide the reader with sense of the expected influence of several major
inputs. Figure 5.10 depicts the saturated zone that may result under steady-state
conditions to illustrate how moisture distribution progresses under steady-state con-
ditions in a landfill, depending on the liquids addition devices. Transient zones are
also depicted, displaying the manner in which moisture movement progresses to
steady-state conditions.
Waste hydraulic conductivity has a significant impact on moisture distribution in
a landfill. Figure 5.11 shows the results of simulations where liquids are added to
waste with different hydraulic conductivities (1 × 10−4 cm s−1, 1 × 10−5 cm s−1,
1 × 10−6 cm s−1) and compares the saturated zones formed. Liquids are added into
waste through devices under a constant pressure and the waste is assumed to be
isotropic (i.e., anisotropy is equal to 1). Under the same injection pressure, the
higher the waste conductivity is, the larger the wetted zone will be. Due to the
assumed isotropic property of waste, the saturated zone formed by a horizontal
injection trench is initially a circle with the center at the location of the injection
pipe. Once it reaches the steady state, its lateral spread will remain constant, and the
wetted zone will only expand vertically downward by gravity. If a constant flow rate
is used, as opposed to a constant injection pressure, the saturated zone formed by
the liquid addition will be essentially the same because the total amount of added
liquids is the same. However, to achieve the same flow rate, a higher injection pres-
sure is required for the waste with low hydraulic conductivity.
Figure 5.12 illustrates the effect of waste anisotropy (Kx/Kz) moisture distribu-
tion under a constant injection pressure. The waste was simulated with the same
vertical hydraulic conductivity, 1 × 10−5 cm s−1, but with different anisotropy, rang-
ing from 1 to 100. The waste with high anisotropy has a larger lateral expansion
because of a higher lateral hydraulic conductivity (Kx). Under the same injection
pressure, the saturated zone formed in highly anisotropic waste is larger than that
114 5 Landfill Constituent Relationships and Dynamics
Fig. 5.10 Illustration of two transient zones as well as the saturated zone under steady state condi-
tions for: (a) horizontal trench; (b) horizontal blanket; and (c) vertical well
for a waste with low anisotropy. With the same vertical hydraulic conductivity,
higher anisotropy means higher lateral hydraulic conductivity, which results in
larger lateral liquid distribution. When the vertical hydraulic conductivity is the
same, the liquid has the same vertical distribution as horizontal.
5.3 Moisture Movement 115
Fig. 5.11 The effect of hydraulic conductivity on liquids distribution: (a) vertical well; (b) hori-
zontal trench; (c) infiltration gallery
The hydraulic conductivity of waste may change with depth. Waste deeper in the
landfill has been further degraded and compacted than waste near the surface.
Therefore, it is reasonable to assume that the hydraulic conductivity of waste
decreases with depth, impacting the distribution of liquids in the subsurface.
Figure 5.13 shows a vertical well, a horizontal trench, and a horizontal gallery in
116 5 Landfill Constituent Relationships and Dynamics
Fig. 5.12 The effect of waste anisotropy on liquids distribution: (a) vertical well; (b) horizontal
trench; (c) infiltration gallery
two scenarios. The first shows the distribution of liquids assuming that hydraulic
conductivity is constant. The second shows hydraulic conductivity decreasing from
10−5 to 10−6 cm s−1. A decreasing hydraulic conductivity as a function of depth
results in more lateral movement as the moisture moves deeper into the landfill.
5.4 Gas and Air Movement 117
Fig. 5.13 The effect of hydraulic conductivity decreasing with depth on liquids distribution:
(a) vertical well; (b) horizontal trench; (c) infiltration gallery
The typical practice of landfill gas production estimation assumes that average dis-
posed mass of waste has some associated gas production potential. This production
potential may be explained in terms of total gas produced (CH4 and CO2), or often
simply in terms of CH4. Thus, the total volume of CH4 produced from a given mass
of waste may be written as:
( )
G = 2 Lo k M waste e - kt (5.22)
where G is gas production rate (m3/year) and k is the first-order decay rate coeffi-
cient (year−1). The use of this relationship for design and prediction, especially as
related to landfills practicing liquids addition to increase waste decomposition rates,
will be discussed in Chap. 13.
When gas is produced in the confined pore spaces of the landfilled waste mass, gas
pressures increase. Similarly, when air is added to a landfill under pressure, gas
pressure in the waste pore space increase. Gas will move in the landfill in response
to the resulting pressure gradient (gas will migrate from high pressure zones within
the landfill toward lower pressure zone such as a gas wells or the landfill surface),
5.4 Gas and Air Movement 119
and a concentration gradient (high gas constituent concentration inside the landfill
compared to low concentrations in the surrounding atmosphere). In a similar man-
ner as previously discussed for moisture movement in the landfill, gas movement
within the landfill can be simulated by using an appropriate governing equation.
Important to developing such a governing equation, however, is the fundamental
relationship between gas pressure and flow through a porous media. In the earlier
presentation of Darcy’s law for moisture movement, hydraulic conductivity was
used, but this relates to water movement through a porous media. Permeability is a
property of the media itself and is independent of the fluid being transmitted.
Equation (5.23) provides for the relationship between hydraulic conductivity and
porous media permeability.
krg
K= (5.23)
m
Fig. 5.14 Landfill gas pressure distribution within a landfill with different boundary conditions
(Townsend et al. 2005)
¶æ p ö ¶ æ k z p ¶p ö
qg ÷ = ¶z ç m RT ¶z ÷ + M (5.25)
¶t çè RT ø è ø
Where θg is the gas constant of the media (i.e., waste) (unitless, volume of gas
per volume of bulk material), p is gas pressure (ML−1 T−2), T is absolute tempera-
ture, R is the gas constant for landfill gas (L2T−2 K−1), kz is the vertical intrinsic
permeability of the waste (L2), μ is the dynamic fluid viscosity (ML−3 T−1), and M is
the gas generation rate (ML−3 T−1). The equation considers partial gas pressures as
well as Darcy’s law for the gas phase. Because the horizontal extents of waste layers
within a landfill are much greater than the vertical extents, the vertical profile is the
only considered direction for variation in pressure distribution. Figure 5.14 illus-
trates a potential outcome of the solution developed; the pressure profiles within the
landfill are presented for conditions when all of the gas exits through the surface
(no flow boundary at the base), all of the gas is collected through the LCRS (no flow
boundary at the top) and when pressures are equal at the base and surface.
capacity and that settlement is due to the compression of void space from overlying
waste. The governing equation formulated coupled settlement to landfill gas pres-
sure as follows:
¶p ¶ æ Z ö æ ¶p ¶ 2 p RT ö
+ ( Patm + p ) ( ln Z g ) = ç ÷÷ ç k g +D 2 + G÷ (5.26)
¶t ¶t çZ
è g ø è ¶z ¶z m ø
Where p is gauge pressure, Patm is atmospheric pressure, R is the universal gas con-
stant (J mol−1 K−1), T is temperature (K), Z is waste height (m), kg is the landfill
unsaturated gas conductivity (m day−1), D is the diffusion coefficient (m3 day−1),
m is the molar mass of the landfill gas (kg mol−1), G is the rate of generation of gas
per unit volume of waste (kg m−3 day−1), t is the time (day).
Shear strength properties of MSW are important in the landfill system as these
properties play a key role in an engineering evaluation of the stability of the waste
mass at the design stage (see Chap. 12 for a detailed discussion of the application
of shear strength to slope stability analysis in sustainable landfill design). The
shear strength properties of soils and MSW are evaluated since the landfill system
includes components such as the final cover system, the waste itself, and the bot-
tom liner system. In contrast to most soils, MSW is heterogeneous which can make
the selection of a single set of key variables that impact shear strength difficult
(e.g., cohesion (c) which is the non-frictional part of shear resistance and is inde-
pendent of normal stress, and internal friction angle (φ), which is the angle on the
Mohr’s Circle of the shear stress and normal effective stress at which shear failure
occurs). Not only does the heterogeneity of the waste itself complicate parameter
selection, but other factors such as the amount and type of cover soil used in opera-
tions, moisture content of the waste, decomposition effects, and waste placement
methods also create difficulties for the designer in selecting a defensible set of
values to examine shear strength.
With both c and φ, the designer has the flexibility to choose parameter values
based on engineering judgment—the design may incorporate the use of both values,
although sometimes c is disregarded for a more conservative analysis (see discussion
by Thiel (2009) for more information on interpreting direct shear testing data). MSW
shear strength can be measured using standard measurement tests such as direct
shear—in some cases, specialized equipment may be used for direct shear testing to
accommodate the large particle size that will be encountered in MSW. Several inves-
tigators have conducted experiments and testing to estimate these factors, which
could serve as a basis for parameter selection. In addition, several authors have
reviewed literature pertaining to cohesion and friction angles for MSW (Dixon and
Jones 2005; Zekkos 2005; Gabr et al. 2007). In general, MSW has a reported cohesion
124 5 Landfill Constituent Relationships and Dynamics
range from 0 to 50 kPa and a typical friction angle ranging from 20° to 35°. Table 5.9
summarizes reported values for c and φ from the technical literature. Figure 5.15
graphically presents the result of many of the shear tests conducted on MSW; the
cohesion is presented as a function of the internal angle of friction.
Table 5.9 Reported values of MSW cohesion (c) and internal friction angle (φ)
Internal friction
Study summary and reference Cohesion angle (φ)
Geotechnical testing on aged solid Under consolidated, Under consolidated,
waste removed from a landfill which undrained conditions undrained conditions
began accepting waste in 1940 in Effective strength: Effective strength:
Pennsylvania (US). Strength 16.8 kPa 34°
parameters evaluated at 20 % strain Cohesion remained Increased with
level (Gabr and Valero 1995; relatively constant with increasing
Kavazanjian et al. 1995) changing horizontal displacement
displacement
24 kPa 0–3°
Samples were collected from two KY Landfill: 11.6 kPa KY Landfill: 23.5°
landfill sites in the US (Kentucky and NY Landfill: 9.3 kPa NY Landfill: 28°
New York), shredded, and processed
(Harris et al. 2006)
Landfilled samples extracted via Effective strength: Effective strength:
borehole every 1 m depth from a landfill 7.43–35 kPa 26.7–50° (mean
closed in 1985 (Al-Yaqout et al. 2007) 33.4°)
Large scale direct shear testing 15 kPa (low moisture 36° (at 1 atm normal
(30 cm × 30 cm) of MSW collected from content MSW) stress, low moisture
a San Francisco Landfill (US), 109 shear MSW)
strength tests were performed (Zekkos
et al. 2010a, b)
Synthetic MSW was examined for 1 kPa (fresh MSW) 35° (fresh MSW)
geotechnical properties. Leachate 16 kPa (anaerobic acid) 34°
recirculation, which causes enhanced 18 kPa (accelerated CH4) 29°
waste degradation, was performed
34 kPa (decelerated CH4) 29°
(Reddy et al. 2011)
40 kPa (CH4 stabilized) 28° (CH4 stabilized)
Under consolidated, Under consolidated,
undrained conditions undrained
conditions:
Total strength: 21–57 kPa Total strength: 1–9°
Effective strength: Effective strength:
18–56 kPa 1–11°
Geotechnical testing on fresh and Waste extracted from a Waste extracted
aged solid waste from a long-running bioreactor experiment: from a bioreactor
bioreactor experiment was performed experiment:
(Bareither et al. 2012b) 22.3 kPa (Initial 40.0° (Initial
condition) condition)
21.7 kPa (aged 2.9 year) 42.6° (aged
2.9 year)
Transfer station MSW: Transfer station
8.9 kPa MSW: 31.5°
5.5 Solids Movement 125
Fig. 5.15 Ranges of measured waste cohesion and internal friction angle reported from the literature
Fig. 5.16 Internal friction angle as a function of food waste content measured using a direct shear
test (from Cho et al. 2011)
Food waste has been suspected of contributing to lower MSW friction angles,
which is a critical design consideration in areas with comparatively less packag-
ing wastes in the discard stream (e.g., developing countries, many Asian coun-
tries). Cho et al. (2011) examined this relationship in both a small- and large-scale
direct shear apparatus. Figure 5.16 presents the results of this experiment.
At lower food waste contents typical of MSW in the western part of the world, the
126 5 Landfill Constituent Relationships and Dynamics
friction angle was similar to those values presented in Table 5.9 (between 30° and
40°). However, at a food waste content of 40 % (by weight), more reflective of wet
landfills of Asian countries and developing nations, the internal angle of friction
was shown to decline markedly.
In reality, fluid flow in a MSW landfill is a multiphase phenomenon, and the predic-
tion of moisture and gas movement using the techniques described so far can be
seen as approximations. One approach to better simulate the combined interaction
of moisture and fluid with respect to the dynamics of these constituents is to develop
separate governing equations for each phase, and to relate them with characteristics
that scale existing properties (e.g., permeability) to the relative amount of each
phase present. For example, the flow of moisture movement could be described by
the following governing equation:
¶ æ k x krl rl ¶pl ö ¶ æ k y krl rl ¶pl ö ¶ æ k z krl rl ¶r k ö ¶ ¶
÷ + ( k z krl rl ) = ( rl Slh )
ç ÷+ ç ÷+ ç
(5.27)
¶x è ml ¶x ø ¶y è ml ¶y ø ¶z è ml ¶z ø ¶z ¶t
While the flow of gas could be described by a separate and distinct governing
equation:
é æ ¶p ö
2
¶ æ ¶p ö æ ¶p ö
2
¶ æ ¶p ö ù
ê k x krg ç g ÷ + pg ç k x krg g ÷ + k y krg ç g ÷ + pg ç k y krg g ÷ ú
1 ê è ¶x ø ¶x è ¶x ø è ¶x ø ¶y è ¶y ø ú
ê ú
mg ê æ ¶pg ö
2
¶æ ¶pg ö gwg ¶pg gwg ¶ ú
ê + k z krg ç
¶z
÷ + pg ç k x krg
¶z ¶z
÷+
RT
pg zkrg
¶z
+
RT ¶z
( kzg pg )ú
ë è ø è ø û
M g RT wg ¶
=
wg
+
RT ¶t
( pg Sgh ) (5.28)
Where krl is relative permeability, kx, y, z is permeability in the respective directions,
pl is liquid pressure, pg is gas pressure, η is porosity, Sl is the degree of saturation for
liquid phase, Sg is the degree of saturation in the gas phase, Mg is gas generation
rate per unit waste volume (kg m−3 s−1), wg is the molecular weight of the gas, μl is
liquid viscosity, μg is gas viscosity. Equations (5.27) and (5.28), along with (5.29)–
(5.33) (Corey 1994), using appropriate boundary conditions, should be solved
simultaneously to simulate gas and liquid flow in solid waste:
Sl + Sg = 1 (5.29)
Pc = Pg - Pl (5.30)
Sl = f1 ( Pc ) (5.31)
krl = f2 ( Pc ) (5.32)
krg = f3 ( Pc ) (5.33)
References 127
This approach has been attempted for MSW landfill systems by several research-
ers (Berglund 1998; Nastev et al. 2001) to meet specific design or research objec-
tives. Future research should continue such efforts with respect to better
understanding and developing tools for sustainable landfill practices.
References
El-Fadel M (1999) Leachate recirculation effects on settlement and biodegradation rates in MSW
landfills. Environ Technol 20:121–133
El-Fadel M, Khoury R (2000) Modeling settlement in MSW landfills: a critical review. Crit Rev
Environ Sci Technol 30(3):327–361
Ettala M (1987) Infiltration and hydraulic conductivity at a sanitary landfill. Aqua Fenn
17:231–237
Fetter CW (2000) Applied hydrogeology, 4th edn. Prentice Hall, Upper Saddle River
Fleming IR (2011) Indirect measurements of field-scale hydraulic conductivity of waste from two
landfill sites. Waste Manag 31:2455–2463
Fungaroli AA (1971) Pollution of subsurface water by sanitary landfills. U.S. Environmental
Protection Agency, SW-12rg, Washington, DC
Fungaroli AA, Steiner RL (1979) Investigation of sanitary landfill behavior. U.S. Environmental
Protection Agency, Cincinnati
Gabr MA, Valero SN (1995) Geotechnical properties of municipal solid waste. Geotech Test J
18(2):241–251
Gabr MA, Hossain MS, Barlaz MA (2007) Shear strength parameters of municipal solid waste
with leachate recirculation. J Geotech Geoenviron Eng 133(4):478–484
Gibson RE, Lo KY (1961) A theory of soils exhibiting secondary compression. Acta Polytech
Scand 296(10):1–15
Hanson JL, Liu WL, Yesiller N (2008) Analytical and numerical methodology for modeling tem-
peratures in landfills. GeoCongress 2008:24–31
Harris JM, Shafer AL, DeGroff W, Hater GR, Gabr M, Barlaz MA (2006) Shear strength of
degraded reconstituted municipal solid waste. Geotech Test J 29(2):1–8
Hashemi M, Kavak HI, Tsotsis TT, Sahimi M (2002) Computer simulation of gas generation and
transport in landfills—I: quasi-steady-state condition. Chem Eng Sci 57:2475–2501
Haydar M, Khire M (2007) Leachate recirculation in bioreactor landfills using permeable blankets.
J Geotech Geoenviron Eng 133(4):360–371
Hettiarachchi CH, Meegoda JN, Tavantzis J, Hettiaratchi P (2007) Numerical model to predict
settlements coupled with landfill gas pressure in bioreactor landfills. J Hazard Mater
139:514–522
Hossain SM, Haque MA (2012) Effects of intermixed soils and decomposition on hydraulic con-
ductivity of municipal solid waste in bioreactor landfills. J Mater Civ Eng 24:1337–1342
Hughes JD, Sanford WE (2004) SUTRA-MS—a version of SUTRA modified to simulate heat and
multiple-solute transport. U.S. Geological Survey Open-File Report 2004-1207
Hull RM, Krogmann U, Strom PF (2005) Composition and characteristics of excavated materials
from a New Jersey landfill. J Environ Eng 131:478–490
Jain P, Farfour WM, Jonnalagadda S, Townsend T, Reinhart DR (2005) Performance evaluation of
vertical wells for landfill leachate recirculation. In: Proceedings of geo frontier 2005, ASCE
conference, Austin
Jain P, Powell J, Townsend T, Reinhart D (2005b) Air permeability of waste in a municipal solid
waste landfill. J Environ Eng 131(11):1565–1573
Jain P, Powell JP, Townsend TG, Reinhart DR (2006) Estimating the hydraulic conductivity of
landfilled municipal solid waste using the borehole permeameter test. J Environ Eng
132(6):645–652
Jain P, Townsend T, Tolaymat TM (2010a) Steady-state design of vertical wells for liquids addition
at bioreactor landfills. Waste Manag 30:2022–2029
Jain P, Townsend T, Tolaymat TM (2010b) Steady-state design of horizontal systems for liquids
addition at bioreactor landfills. Waste Manag 30:2560–2569
Jang YS, Kim YW, Lee SI (2002) Hydraulic properties and leachate level analysis of Kimpo met-
ropolitan landfill, Korea. Waste Manag 22:261–267
Jonnalagadda S (2004) Resistivity and time domain reflectrometry sensors for assessing in-situ
moisture content in a bioreactor landfill. Master’s Thesis, University of Florida, Gainesville
Kavazanjian E Jr (2001) Mechanical properties of municipal solid waste. In: Proceedings Sardinia
2001, eighth international landfill symposium
References 129
Powrie W, Richards DJ, Beaven RP (1998) Compression of waste and implications for practice. In:
Dixon N, Murray EJ, Jones DRV (eds) Geotechnical engineering of landfills. Thomas Telford,
London
Qian X, Koerner RM, Gary DH (2002) Geotechnical aspects of landfill design and construction.
Prentice Hall, New York
Reddy KR, Gangathulasi J, Parakalla NS, Hettiarachchi H, Bogner JE, Lagier T (2009a)
Compressibility and shear strength of municipal solid waste under short-term leachate recircu-
lation operations. Waste Manag Res 27:578–587
Reddy KR, Hettiarachchi H, Parakalla N, Gangathulasi J, Bogner J, Lagier T (2009b) Hydraulic
conductivity of MSW in landfills. J Environ Eng 135:677–683
Reddy KR, Hettiarachchi H, Gangathulasi J, Bogner JE (2011) Geotechnical properties of munici-
pal solid waste at different phases of biodegradation. Waste Manag 31:2275–2286
Rover FA, Farquhar GJ (1973) Infiltration and landfill behavior. J Environ Eng Div 99:671–690
Schroeder PR, Dozier TS, Zappi PA, McEnroe BM, Sjostrom JW, Peyton RL (1994) The hydro-
logic evaluation of landfill performance (HELP) model: engineering documentation for version
3 US EPA/600/R-94/168a. United States Environmental Protection Agency Office of Research
and Development, Washington, DC
Shank KL (1993) Determination of hydraulic conductivity of the Alachua County Southwest
Landfill, Master Thesis. University of Florida, Gainesville
Sowers GF (1973) Settlement of waste disposal fills. Proc., 8th Int. Conf. on Soil Mechanics and
Foundation Engineering, Moscow, 2(2):207–210
Stankovich JL, Lockington DA (1995) Brooks-Corey and van Genuchten soil-water-retention
models. J Irrig Drain Eng 121:1–7
Stephens DB (1995) Vadose zone hydrology. CRC Lewis, Boca Raton
Stoltz G, Gourc JP, Oxarango L (2010a) Characterisation of the physico-mechanical parameters of
MSW. Waste Manag 30:1439–1449
Stoltz G, Gourc J, Oxarango L (2010b) Liquid and gas permeabilities of unsaturated municipal
solid waste under compression. J Contam Hydrol 118(1–2):27–42
Straub WA, Lynch DR (1982) Models of landfill leaching: moisture flow and inorganic strength.
J Environ Eng 108(2):231–250
Tchobanoglous G, Theisen H, Vigil S (1993) Integrated solid waste management: engineering
principles and management issues. McGraw-Hill, New York
Terzaghi K, Peck RB (1967) Soil mechanics in engineering practice, 2nd edn. Wiley, New York
Thiel R (2009) A technical note regarding interpretation of cohesion (or adhesion) and friction
angle in direct shear tests, Geosynthetics, April–May 2009
Tinet A, Oxarango L (2010) Stationary gas flow to a vertical extraction well in MSW landfill con-
sidering the effect of mechanical settlement on hydraulic properties. Chem Eng Sci
65(23):6229–6237
Todd DK, Mays LW (2005) Groundwater hydrology, 3rd edn. Wiley, Hoboken, p 30
Townsend TG (1995) Leachate recycle at solid waste landfills using horizontal injection.
Ph.D. Dissertation, University of Florida, Gainesville
Townsend T, Miller WL, Lee H, Earle JFK (1996) Acceleration of landfill stabilization using
leachate recycle. J Environ Eng 122(4):263–268
Townsend T, Wise W, Jain P (2005) One-dimensional gas flow model for horizontal gas collection
systems at municipal solid waste landfills. J Environ Eng 131(12):1716–1723
Vaidya RD (2002) Solid waste degradation, compaction and water holding capacity. M.A. Thesis,
Virginia Polytechnic Institute and State University, Blacksburg
Walsh JJ, Kinman RN (1982) Leachate and gas production under controlled moisture conditions.
In: Proceedings of the eighth annual research symposium on land disposal of solid waste, EPA-
600-002, US Environmental Protection Agency, Cincinnati
Watts KS, Charles JA (1990) Settlement of recently placed domestic refuse landfill. Proc Inst Civ
Eng Part l 88:971–993
References 131
Wigh RJ (1979) Boone County field site interim report, EPA-600/279/058. U.S. Environmental
Protection Agency, Cincinnati
Wu H, Chen T, Wang H, Lu W (2012) Field air permeability and hydraulic conductivity of land-
filled municipal solid waste in China. J Environ Manage 98:15–22
Wysocki EJ, Nabavi TR, Djafari S (2003) The use of leachate recovery wells to evaluate municipal
solid waste hydraulic characteristics at Fresh Kills Landfill. Paper presented at 8th Annual
Landfill Symp. of SWANA, SWANA, Atlantic City
Xu Q, Kim H, Jain P, Townsend TG (2012) Hydrologic evaluation of landfill performance (HELP)
modeling in bioreactor landfill design and permitting. J Mater Cycles Waste Manag 14:38–46
Yen BC, Scanlon B (1975) Sanitary landfill settlement rates. J Geotech Eng Div
101(GT5):475–487
Yu L, Batlle F, Carrera J (2012) Variations of waste unit weight during mechanical and degradation
processes at landfills. Waste Manag Res 29:1303–1315
Yu-xin J, Wen-jie X, Danzeng D, Yi-feng W, Tao P, Zai-yang Z (2013) Laboratory testing of a
densified municipal solid waste in Beijing. J Cent South Univ 20:1953–1963
Zeiss C, Major W (1993) Moisture flow through municipal solid waste: pattern and characteristics.
J Environ Syst 22:211–231
Zeiss C, Uguccioni M (1995) Mechanisms and patterns of leachate flow in municipal solid waste
landfills. J Environ Syst 23(3):247–270
Zeiss C, Uguccioni M (1997) Modified flow parameters for leachate generation. Water Environ
Res 69(3):276–285
Zekkos DP (2005) Evaluation of static and dynamic properties of municipal solid waste.
Ph.D. Dissertation, University of California, Berkeley
Zekkos D, Kavazanjian E, Bray JD, Matasovic N, Riemer MF (2010a) Physical characterization of
municipal solid waste for geotechnical purposes. J Geotech Geoenviron Eng 136:1231–1241
Zekkos D, Athanasopoulos GA, Bray JD, Grizi A, Theodorateos A (2010b) Large-scale direct
shear testing of municipal solid waste. Waste Manag 30:1544–1555
Zhan TLT, Chen YM, Ling WA (2008) Shear strength characterization of municipal solid waste at
the Suzhou landfill, China. Eng Geol 97:97–111
Zornberg JG, Jernigan BL, Sanglerat TR, Cooley BH (1999) Retention of free liquids in landfills
undergoing vertical expansion. J Geotech Geoenviron Eng 125(7):583–594
Chapter 6
Moisture Supply and Conveyance
Moisture levels can be enhanced through the addition of either liquids or wet wastes
(e.g., sludges). The most common source of supplemental moisture deliberately
added to landfilled waste is leachate collected from the same (or perhaps adjacent)
landfill unit. Leachate recirculation has been practiced as a means of enhanced
waste stabilization and for basic liquids management. However, in some cases, the
target moisture addition requirements specified by the engineer as well as the avail-
able leachate volumes for recirculation necessitate other sources of moisture to
achieve project objectives. These additional moisture sources may be storm water
purposefully retained after rain events or surface water or groundwater extracted
from outside the landfill. In other cases, the moisture sources are waste products
themselves (e.g., industrial wastewater, septage, wastewater sludges), and in this
case the landfill operator is presented with an opportunity to collect revenue for the
disposal of the waste liquids or wet wastes in addition to providing a needed source
of moisture to the landfill itself.
The choice of an additional moisture source will depend on several factors:
availability; difficulty and cost associated with capturing, extracting, or obtain-
ing the source; and limitations imposed by applicable regulations. With regard to
regulatory requirements, planners and engineers must consult the appropriate
regulations and regulatory agencies. In the U.S., for example, federal MSW land-
fill regulations permit the addition of leachate and landfill gas condensate back to
the landfilled waste (US government 2012). As described in Chap. 3, however,
the addition of bulk liquid wastes is prohibited unless special permission is
granted. A solid waste that fails the paint filter test is considered a bulk liquid
waste. In the case of domestic wastewater sludge (e.g., municipal biosolids), for
example, the moisture content must be less than approximately 80–85 % (by
weight) to pass the paint filter test. The subsequent sections provide more detailed
discussion on the following potential supplemental moisture sources: leachate,
water, wastewater, spent aqueous products, and wet wastes.
6.2.2 Leachate
Much of the early work on the subject of wet landfills involved examining the
effects of leachate recirculation on leachate quality (Pohland 1980). Today, leachate
recirculation is commonly practiced both by landfill operators targeting enhanced
waste stabilization and when a lower cost method of managing leachate is sought.
Most leachate recirculation systems involve constructing a pumping system to con-
vey leachate from storage units (tanks, ponds) to the liquids addition devices within
the landfill. Another option sometimes practiced is to utilize the landfill’s existing
6.2 Moisture Sources 135
Fig. 6.1 Leachate storage and aeration pond equipped with a pump for recirculating leachate to
the landfill
pumping system (as part of the LCRS) to convey leachate back to the landfill.
Liquids conveyance strategies are discussed in further detail later in this chapter.
Depending on site-specific design and operation features, the chemical quality of
the leachate recirculated may differ from that emanating at the base of the landfill.
Storage ponds and tanks may be equipped with aeration systems to help control
odors and achieve some rudimentary treatment (Fig. 6.1). The leachate recirculated
in this case may be somewhat lower in organic content (BOD), ammonia, and met-
als that precipitate under oxidation (e.g., iron). Leachate at uncovered ponds or
tanks may also become diluted as a result of large rain events, or concentrated in
areas where evaporation is greater than rainfall.
Leachate treated in a more rigorous manner—as part of a treatment plant, for
example—may in some cases also be recirculated to the landfill, although a benefit
of leachate recirculation is reduction of some of the landfill constituents to mini-
mize external treatment requirements (e.g., BOD). As will be described in greater
depth in Chap. 11, landfill operators may opt to deliberately treat leachate prior to
recirculation to meet specific facility objectives. For example, leachate nitrification
transforms much of the ammonia-nitrogen (NH3/NH4+) to nitrate-nitrogen (NO3−),
which, when recirculated back to the landfill, may undergo denitrification to nitro-
gen gas (N2) (Berge et al. 2005). Advanced oxidation processes have the potential to
convert organic matter otherwise recalcitrant to decomposition to a more biodegrad-
able form (Batarseh et al. 2010). Treatment techniques such as reverse osmosis,
ultrafiltration, and evaporation can be used to dewater leachate (i.e., concentrate the
leachate by removing relatively clean water) prior to return to the landfill.
136 6 Moisture Supply and Conveyance
As will be illustrated in the next section, designers often target the addition of
sufficient liquids to the landfill to reach the waste’s field capacity. Even in wet cli-
mates, the amount of leachate generated (especially when good stormwater man-
agement practices are employed) is often insufficient to reach this target. Thus,
outside water sources are sometimes added, such as stormwater, surface water, or
groundwater. Regulatory allowance of water addition will be a major controlling
factor in the implementation of this procedure, as some regulatory agencies do not
permit the addition of water. As stated earlier, US regulations allow leachate recir-
culation, but prohibit the addition of bulk liquid wastes (US government 2012). The
allowance of water addition, however, has been interpreted differently among US
states. Since clean water is not a liquid waste, some regulators allow water addition,
while others argue that the intent of the bulk liquids waste prohibition is to minimize
moisture entry into the landfill and thus this practice is not permitted.
Groundwater and surface water addition has been practiced at several sites in
the US (Yazdani et al. 2010; Ko et al. 2013) to meet target moisture content. These
operators normally utilize separate pumping systems to deliver water to a leachate
storage tank or pond where it is mixed with leachate prior to addition to the land-
fill (Fig. 6.2). Landfill operators have the ability to retain rainfall depending on
stormwater management practices; this is described in more detail in Chap. 11.
Regulatory operating requirements for landfills, however, typically limit the
Fig. 6.2 Groundwater well used for extracting fresh water to mix with leachate before recirculat-
ing to a landfill
6.2 Moisture Sources 137
Some landfill operators have pursued the disposal of industrial wastewaters or other
spent aqueous products. At the Outer Loop landfill (see Chap. 4), for example,
operators disposed of beverage waste, oily wastewater, paint waste, ink water, and
other industrial wastewaters (US EPA 2006). Such liquid wastes would otherwise
require treatment and disposal at a domestic or industrial wastewater treatment
facility, often at considerable expense. By accepting such wastes for disposal, land-
fill operators can provide supplemental moisture while adding a revenue source. As
stated previously the operator may need to obtain special regulatory permission, as
this practice could be restricted under normal circumstances.
Industrial wastewater or spent aqueous products would typically be hauled
directly to the landfill in a tanker truck and discharged to a designated disposal area.
Surface application techniques such as those described in Chap. 7 would be most
common. Concerns discussed in this chapter and others regarding proper leachate
containment will be magnified when outside liquids are disposed. While subsurface
techniques such as those described in Chaps. 8 and 9 might be feasible if appropri-
ate liquids unloading and conveyance infrastructure is available, the operator should
be cautious when adding liquids with high solids content that might clog or other-
wise limit future liquids addition.
Tolaymat et al. (2004) reviewed the factors that should be considered prior to
addition of industrial wastewater or similar liquid wastes. For example, liquid pH
and its impact on microbial activity should be assessed. Extremes in pH, particu-
larly low pH, might require neutralization prior to disposal. Tolaymat et al. (2004)
suggested conducting limited field tests to determine the ability of the waste to buf-
fer added liquid and to distribute the liquids over large areas to limit possible harm-
ful effects.
Elevated concentrations of chemical constituents (salts, heavy metals, organic
pollutants) have the potential to be toxic to the landfill biota responsible for carrying
out the waste stabilization process. While MSW has an ability to attenuate and
transform many trace pollutants (Reinhart and Pohland 1991; Pohland et al. 1992),
the operator should carefully consider the chemical composition of a new liquid
waste prior to disposal. Some useful toxicity information might be available from
the existing literature, but specific anaerobic toxicity testing may also be warranted.
The BMP assay, for example, was developed in part as an anaerobic toxicity test
(Owen et al. 1979); this methodology is discussed in greater detail in Chap. 16.
Industrial wastewater and similar supplemental liquids with high organic matter
content may provide a potential substrate for the production of methane as a result
of anaerobic decomposition, but as described in Chap. 2, anaerobic biological
138 6 Moisture Supply and Conveyance
systems consist of multiple biotic groups and are subject to upset if the system
becomes unbalanced. If a liquid waste has a large concentration of rapidly ferment-
able organic compounds, this may result in rapid acid buildup, which in turn, sup-
presses methanogenic activity. Similar to concerns with trace chemicals or salts,
specific tests such as the BMP should be considered as a screening technique, at
least for the first time a candidate liquid waste is proposed. Ko et al. (2012) used the
BMP assay to assess the potential effect (toxicity and methane yield) of three indus-
trial liquids (fishery, dairy, and brewery wastewaters) when added to landfills.
Several solid waste streams have inherently high moisture contents. Examples
include wastewater sludge and food processing waste. While some operators are
reluctant to accept such sources (because of operational and nuisance issues
described below), in a similar manner as industrial liquids, receipt of wet wastes
offers a source of supplemental moisture and potential revenue. In some cases, cer-
tain wet wastes such as biosolids may be accepted as a service to a local utility
department, and sometimes may be part of a negotiated deal to accept some or all of
the landfill’s leachate at the utility’s wastewater treatment plant. Similar to the other
moisture sources discussed already, special regulatory permission may be required
if the waste falls outside the bounds of allowable wastes for disposal.
Domestic wastewater sludge (biosolids) is perhaps the most commonly proposed
wet waste added to landfills; it is a waste stream generated in relatively large mag-
nitudes in most developed nations. While biosolids are commonly applied to agri-
culture, forest, mined land, or disposed offshore in the ocean, these practices are
facing growing opposition and restrictions. When disposed in landfills, biosolids
present a source of moisture, methane potential, and nutrients. Many of the early
studies exploring potential sustainable landfill technologies (see Chap. 4) examined
the addition of wastewater sludge as a means of enhancing waste stabilization and
increasing gas production (EMCON Associates 1975; Pohland 1980; Buivid et al.
1981). In a sludge digester at a wastewater treatment facility, the solids content of
biosolids will be on the order of 0.5–2 % (98–99.5 % moisture). Dewatering will
often be practiced prior to disposal to reduce transportation and disposal costs. As
previously described, to pass the paint filter test, domestic biosolids need to be
dewatered to 15–20 % solids content (80–85 % moisture).
The reluctance of many landfill operators to accept biosolids for routine disposal
stems from operational difficulties, odors, and health and safety concerns. Biosolids
possess a strong and often offensive odor, which coupled with the propensity of this
waste stream to attract flies and other disease vectors, and the possible health risks
resulting from pathogenic organisms, demand that disposed biosolids be covered
relatively quickly to minimize concerns to landfill personnel, customers, and the
surrounding community. When disposed at a landfill, biosolids will normally be
transported using a dump truck or similar vehicle (Fig. 6.3). Since this waste stream
6.3 Moisture Addition Targets 139
Fig. 6.3 A truck load of biosolids unloaded at the working face of a landfill
is so wet and has little or no strength in this form, biosolids cannot be moved and
compacted in the same manner as MSW. Biosolids adhere to the tracks of dozers
and the cleats of compactors, and the equipment may become stuck or sink into the
working surface, thus making waste compaction a laborious process. Unless mixed
with other waste, the presence of biosolids may hinder compaction efforts and result
in lower compaction densities. Wet biosolids create an extremely slick working
surface, making it very hard for waste spotters and other landfill personnel to walk
on it. Longer-term operational issues derive from the differential settlement or soft
spots that might occur when biosolids are buried without mixing with other wastes,
and possible slope stability concerns if the biosolids are placed in continuous layers
near the side slopes of the landfill. Methods that operators have successfully used
for disposing of biosolids are reviewed in a later part of this chapter.
Once an objective of increasing the MSW moisture content has been established
and available moisture sources have been identified, the designer can proceed
with estimating the targeted amount of moisture to add to the landfill. As described
in Chap. 4, multiple laboratory, pilot-scale, and full-scale studies have confirmed
140 6 Moisture Supply and Conveyance
that elevating the moisture content of MSW from the original relatively dry conditions
enhances the rate of waste stabilization. Additionally, the movement of moisture
through the landfilled waste serves to transport micro-organisms, substrates,
nutrients, and waste products throughout the landfill. Although some researchers
recommend a specific moisture content desirable to optimize waste stabilization
(e.g., Guijara and Suflita (1993) reported that at least 50 % moisture content
would result in optimal methanogenesis), selection of a target moisture content
for a full-scale operation based on optimized smaller-scale studies is generally not
practical. First, the achievable moisture content of waste is very much related to
the specific weight of the waste (see Chap. 5). Second, waste stabilization will
likely be optimal under saturated or near-saturated conditions where liquids are
moving relatively rapidly, a condition that is not feasible (and questionably safe)
for full-scale landfill operations.
In lieu of selecting a target moisture content needed to achieve optimal waste
stabilization, a common practice in the design of bioreactor or similar landfills is to
target the introduction of at least enough moisture to reach landfilled waste field
capacity. The field capacity concept was described in Chap. 5; in theory, all liquids
added to increase the moisture content to field capacity would be absorbed by the
waste, rather than leaving the landfill as leachate. In reality, because of the funda-
mental processes governing fluid flow in porous media, there is no way to bring all
waste to field capacity without first bringing some of the waste to moisture levels
greater than field capacity (including some at saturation), and then letting the liquids
drain by gravity. Thus targeting field capacity as the desired moisture content is not
a necessary outcome for success of the system, but rather a means to provide a real-
istic target for moisture addition for design purposes.
Chapter 5 provided measured data from several studies reporting the initial mois-
ture content of landfill disposed waste; the reported moisture content depends upon
composition and local climate, and the designer should gather information specific
to the site being planned. Similarly, previous measured data for field capacity were
presented; a very notable observation from these data is the strong relationship
between field capacity and waste specific weight. Tolaymat et al. (2013) present a
recommended approach for determining a target liquid addition volume to achieve
field capacity (or other desired target moisture content) for a given waste. For waste
with an initial moisture content of MCi (% wet weight) and a target moisture content
of MCt (% wet weight), the volume of moisture required to bring per a unit mass of
waste to the target value (Vr) may be determined from (6.1):
MCt - MCi
Vr = C (6.1)
100 - MCt
Where C is a conversion factor for which C = 1,000 L/Mg for metric units and
239.8 gal/ton for customary units. Figure 6.4 presents a plot of Vr as a function of
MCi and MCt.
6.3 Moisture Addition Targets 141
The addition of wet waste (e.g., biosolids) would also increase the overall
moisture content of the landfill. The mass of wet waste (Mwet waste) required per mass
of MSW (Mmsw) required to reach a target moisture content (MCtarget) may be deter-
mined using (6.2):
In the case of liquids addition, when a target moisture addition volume has been
determined and an estimate of the total mass of waste to be wetted is known, the
total target liquid volume to be added to the landfill can be calculated. In this sec-
tion, moisture addition rates are discussed on a longer scale (e.g., monthly, yearly)
and for the entire landfill. In the following three chapters, liquid addition rates
achievable for individual liquids addition devices are discussed. Prior to the
detailed design, however, the engineer must develop a target liquid addition rate.
142 6 Moisture Supply and Conveyance
A specification of flow rate (volume that should be added over a given duration)
is necessary for the detailed design of a liquids introduction system and for devel-
oping an operation plan for the system. Several factors affect the rate at which
liquids should be or can be added, including available liquids volumes, the ability
of the waste to accept the liquids, the desired time to operate the system, impacts
on the leachate collection system, and operational and safety considerations
(e.g., slope stability, seepage).
The time available for system operation has a direct impact on the liquids intro-
duction rate. The liquids addition system is generally operated during the working
hours of the landfill staff. The system will also be shut down occasionally for main-
tenance and may be prevented from operating during heavy rainfall events or other
inclement weather. The total system operation time also depends on the manner in
which a system is constructed. A liquids addition system constructed as the waste is
placed (often referred to as an as-built system) has a greater potential window of
operation compared to a landfill where the liquids addition system is not constructed
until after waste placement is complete (often referred to as a retrofit system).
The impact of added liquids on the LCRS and the liner also play a major role
with respect to establishing the liquids addition rate. As discussed in Chap. 2, regu-
latory design requirements often limit the depth of leachate ponded on top of the
bottom liner. Chapter 10 provides guidance on how to integrate liquids addition rate
into the LCRS design and performance. While as-built operations with a LCRS
designed to handle liquids addition may not be limited by LCRS performance, ret-
rofit sites where the LCRS was not designed with liquids addition design in mind
may require wetting to be conducted over a longer period of time (i.e., at a lower
rate) to reach desired addition targets.
Even though an engineer may select a desired liquids addition rate, such a rate
might not be achievable within the constraints placed as a result of other consider-
ations (e.g., construction techniques, costs). As described in Chap. 5, compacted
solid waste has a relatively low permeability, especially deep in the landfill. The
distribution of desired moisture volumes into the landfill over a specified time inter-
val may thus require a large number of devices or operational pressures. Both of
these factors have an impact on system cost and operational complexity. In addition,
other concerns with regard to leachate seepage (Chap. 11), slope stability (Chap. 12),
and impact on gas system performance (Chap. 13) may limit an operator’s ability to
achieve target addition rates.
Fig. 6.5 Schematic of two different pumping systems in bioreactor landfills (a) gravity feed
system, and (b) pressurized pumping system
pressurized injection into vertical wells has been attempted (Kadambala et al. 2014).
In the case of a pressure-controlled approach, the pumping system provides not only
the pressure to deliver the liquids to the source, but also enough to achieve the
desired design injection pressure.
In a gravity-controlled system, liquids are delivered to a target source and dis-
charged to an open atmosphere at the surface of or within the landfill. Examples
include surface application (such as spray irrigation), ponds, or a vertical well where
a free water surface exists (no pressure beyond the depth of water column is applied).
The main goal of a gravity-controlled system is to deliver sufficient volume of liq-
uids either directly to the landfill surface or to a secondary storage system.
Controlling the flow rate is a common issue in gravity-controlled systems and may
require manual labor to control or add a sufficient flow rate while avoiding a condi-
tion where the liquid surface rises above the waste surface. Certain mechanical and/or
electrical controls can be installed to reduce manual labor associated with operating
the system. For example, a water level sensor can trigger a control valve once the
water level is below or above a desired level. Another approach that has proven suc-
cessful is placing a storage tank on top of a landfill, as shown in Fig. 6.6. With this
approach, liquids are first delivered to the storage tank by a pump system and then
discharged into the waste by control valves; in this case, lower flow rates are easier
to regulate.
6.4 Conveyance Systems for Liquids Addition 145
Fig. 6.6 Tank container chassis used for the storage and gravity discharge of leachate to a liquids
distribution system on top of a landfill
The type of pump system selected will be dictated in part by the degree of control
and automation desired. Centrifugal wastewater pumps, such as those commonly
used as part of most LCRS can be specified, but their control may be limited relative
to other pump types less commonly used for leachate, such as positive displacement
pumps. In addition to specifying on-off conditions (which can be accomplished for
any pump type), some designers may wish to control injection pressures, and for
landfills where pressurized liquids are added at different elevations, such control
might be more easily accomplished with positive displacement or similar pumps.
Smooth-walled plastic pipe is commonly used for pressurized leachate force mains
at landfills, although some regulatory jurisdictions may require double-walled pipe
when placed outside the lined landfill unit. Depending on the degree of control
desired (Chaps. 8 and 9) and the specific data to be monitored and recorded (Chap. 16),
the piping system can be equipped with meters or sensors for measuring flow rate,
cumulative fluid flow, and pressure. Figure 6.7 shows a hydrant system used to dis-
tribute pressurized liquids to individual horizontal liquids addition trenches at a
North American site; pressure measurements in the force main were integrated into
the pumping system’s control logic so valves could be actuated shut when target
pressures were reached.
Chapters 7, 8, and 9 provide more information on the individual liquids addition
devices that might be employed. The devices will be connected to the liquids con-
veyance system by either direct pipe connection or with a flexible hose. Flexible
hoses have the advantage of accommodating differential settlement if this is a con-
cern. Figure 6.8 shows the point where liquids are conveyed from a pressurized
146 6 Moisture Supply and Conveyance
Fig. 6.7 Hydrant system for delivering liquids to specific addition devices in a landfill
force main to a shallow surface trench (described further in Chap. 6). In this
example, the trench was also connected to the landfill’s GCCS, and appropriate
valves (and pressure monitoring devices) were provided to isolate liquids addition
from gas extraction. The opportunities and challenges associated with combining
liquids addition and gas extraction from trenches, wells and similar devices are
discussed in Chap. 13.
The pumping system design for the liquids conveyance system will follow stan-
dard procedures used for water and wastewater. A system curve that plots the sys-
tem pressure as a function of flow rate from the entrance to the exit of the conveyance
system will be developed and compared to candidate pump curves. The following
equation portrays a typical system curve:
V22 - V12
Ph = ( Z 2 - Z1 ) + ( P2 - P1 ) + + hL , minor + hL , friction (6.3)
2g
V -V
2
2
1
2
(6.4)
2g
where, V = velocity head (L); Z = elevation head (L); Pi = pressure head (L);
hL,friction = head loss due to friction (L); hL,minor = head loss due to local disturbances
of flow (e.g., valves, bends, and couplings) (L). The friction loss term (hL,friction),
which accounts for the pressure loss as liquids flow through the pipe, can be esti-
mated using several approaches, such as the Darcy-Weisbach Equation, which is
presented as (6.5):
8 fLQ 2
hL , friction = (6.5)
p 2 gD 2
where L = the length of the pipe section (L); D = the pipe diameter (L); g = the grav-
ity constant (L T−2); Q = the flow rate (L3 T−1); and f = the dimensionless friction
factor and is a function of the Reynolds number (Re) and relative roughness (e/D).
The minor loss, hL,minor, results from in-line fittings, changes in direction, and
changes in flow area. It is usually calculated using the method of loss coefficients.
Each fitting has a loss coefficient, Kminor, associated with it. The minor loss is
obtained by multiplying the loss coefficient by velocity head:
V2
hL,minor = åK minor × (6.6)
2g
A system curve can be created using (6.3), and then used to select an appropriate
pump. The operation point is determined by plotting the system curve and a manu-
facturer’s pump curve.
148 6 Moisture Supply and Conveyance
Q V22
Ph = DZ + + + hL , minor + hL,friction (6.7)
k 2g
where κ is the fluid conductance of the device. The fluid conductance relates flow
rate to pressure; this concept will be discussed with respect to vertical and horizon-
tal liquids addition devices in Chaps. 8 and 9, respectively.
Many operators are reluctant to accept biosolids for disposal because of the
issues described earlier (workability, operational issues, odors, and health and
safety concerns). Since operators cannot move or compact wet wastes in the
same manner as MSW, a variety of techniques may be utilized to bury these
wastes. Some procedures involve burying the biosolids in depressions excavated
on or near the working face. This practice, however, can result in soft spots or
differential settlement. Other techniques therefore focus on mixing the biosolids
with the waste or other materials. Reinhart et al. (2007) evaluated biosolids
disposal techniques at MSW landfills, and a summary of typical operational
techniques is presented in Table 6.1. In most cases, these practices would be
applicable for similar wet wastes.
When increasing the moisture content of the solid waste is a primary objective,
mixing the biosolids (or other wet waste) with the MSW is the preferred option.
Figure 6.9 displays mixing of biosolids with MSW by pushing a load of MSW on
to of a layer of biosolids; this would be followed by making several passes over
the waste with a compactor, with an end result being the mixing of the two materi-
als. Mixing wet and dry waste with available landfill equipment also can also be
used achieve this outcome. Operators can also mix wet waste with other materials,
such as mulch or soil, to improve workability; this may limit some moisture
distribution to MSW, however.
Table 6.1 Techniques for biosolids disposal at MSW landfills (from Reinhart et al. 2007)
Method Description
Direct unloading and The landfill operator practicing co-disposal of biosolids with MSW
mixing with waste would direct the biosolids truck directly to the landfill’s working face,
loads where biosolids are unloaded and disposed of with loads of MSW. This
method requires more coordination of landfilling operations to ensure
that there are a couple of new loads of MSW set aside for the incoming
loads of biosolids. Spreading biosolids in thin layers on the working
face and covering with MSW is another option. This method is similar
to the MSW co-disposal method discussed before in that biosolids are
disposed of directly on the landfill’s working face. However, the
biosolids are handled separately with the blade of the bulldozer by
spreading them in a thin layer over the surface of the landfill and then
covering this layer with regular MSW before compacting
Pit or trench burial Pit (trench) burial of biosolids involves more site preparation and
of biosolids equipment requirements than others because of the need to dig the pit
or trench into the waste before the biosolids are unloaded into the
landfill. Some landfill operators might prefer this method since it
minimizes the need for handling the biosolids. One of the
disadvantages of this method is the potential creation of soft spots on
the surface of the landfill where the biosolids have been placed
Mixing biosolids with The mixing with additives technique includes mixing biosolids with
other materials prior MSW, yard waste, mulch or mulch fines, or soil. One of the main
to disposal or use as advantages of this disposal method is that it provides a more workable
cover material material than biosolids alone. However, this method requires a separate
area of the landfill to be cleared and designated for the mixing process
Another method is to mix biosolids with additives using a
predetermined ratio, as discussed in the previous method, and used as
daily cover It should be noted that the use of materials other than site
soils dirt for daily cover may require regulatory approval
Fig. 6.9 MSW pushed on top of biosolids at the working face of landfill
150 6 Moisture Supply and Conveyance
References
Batarseh E, Reinhart D, Berge N (2010) Sustainable disposal of municipal solid waste: post biore-
actor polishing. Waste Manag 30:2170–2176
Berge ND, Reinhart DR, Dietz J, Townsend T (2005) In situ ammonia removal in bioreactor land-
fill leachate. Waste Manag 26:334–343
Buivid MG, Wise DL, Blanchet MJ (1981) Fuel gas enhancement by controlled landfilling of
municipal solid waste. Resour Conserv 6:3–20
EMCON Associates (1975) Sonoma county solid waste stabilization study. Prepared for
Environmental Protection Agency Report No. EPA/530/SW-65D. 1 Grant No G06-EC-00351
Gurijala KR, Suflita JM (1993) Environmental factors influencing methanogenesis from refuse in
landfills. Environmental Science and Technology, 27: 1176
Kadambala R, Powell J, Singh K, Townsend T (2014) Evaluation of a buried vertical well leachate
recirculation system for municipal solid waste landfills. Unpublished manuscript
Khire M, Haydar M (2005) Leachate recirculation using geocomposite drain-age layers in engi-
neered MSW landfill. In: Proceedings of Geo Frontier 2005, ASCE Conference, Austin
Ko J, Townsend T, Kim H (2012) Evaluation of the potential methane yield of industrial wastewa-
ters used in bioreactor landfills. J Mater Cycles Waste 14:162–168
Ko J, Powell J, Jain P, Kim H, Townsend T, Reinhart D (2013) Case study of controlled air addition
into landfilled municipal solid waste: design, operation, and control. J Hazard Toxic Radioact
Waste 14(4):351–359
Owen WF, Stuckey DC, Healy JB Jr, Young LY, McCarty PL (1979) Bioassay for monitoring
biochemical methane potential and anaerobic toxicity. Water Res 13:485–492
Pohland FG (1980) Leachate recycle as landfill management option. J Environ Eng
106:1057–1069
Pohland F, Cross W, Gloud J, Reinhart D (1992) Behavior and assimilation of organic and inor-
ganic priority pollutants co-disposed with municipal refuse, EPA-600/R-93-137a. US EPA,
Cincinnati
Reinhart D, Pohland F (1991) The fate of selected organic pollutants codisposed with municipal
refuse. Res J Water Pollut Control Fed 63(4):780
Reinhart D, Townsend T, Dubey B, Kim H, Paredes V, Xu Q (2007) Design and operational issues
related to co-disposal of sludges and biosolids in Class-I landfill—Phase III, Report #0432023.
Center for Solid and Hazardous Waste Management, Gainesville
Tolaymat T, Kremer F, Carson D, Davis-Hoover W (2004) Monitoring approaches for landfill
bioreactors. National Risk Management Research Laboratory Office of Research and
Development U.S. Environmental Protection Agency EPA/600/R-04/301, Cincinnati
Tolaymat T, Kim H, Jain P, Powell J, Townsend T (2013) Moisture addition requirements for
bioreactor landfills. J Hazard Toxic Radioact Waste 17(4):360–364
Townsend T, Miller W (1998) Leachate recycling using horizontal injection. Adv Environ Res
2(2):129–138
US Government (2012) US code of federal regulations: Title 40 Part 258. Criteria for Municipal
Solid Waste Landfills, Washington, DC
US EPA (2006) Landfill bioreactor performance. Second interim report outer loop recycling and
disposal facility, Louisville
Yazdani R, Mostafid ME, Han B, Imhoff PT, Chiu P, Augenstein D, Kayhanian M, Tchobanoglous
G (2010) Quantifying factors limiting aerobic degradation during aerobic bioreactor landfill-
ing. Environ Sci Tech 44:6215–6220
Chapter 7
Systems for Surface Addition of Liquids
Abstract One of the earliest forms of liquids addition practiced by landfill operators
was surface addition. Techniques include: direct wetting of the waste, spray or drip
irrigation, and ponding. In contract to subsurface liquid addition systems, construc-
tion requirements for surface systems are minimal. Care must be taken to ensure
that liquids do not migrate outside of the controlled landfill area. The different con-
figurations of surface systems are presented and discussed, along with design
approaches that can be used to identify liquids addition amounts in light of the site’s
moisture addition goals.
Surface systems involve adding liquids to the surface of the landfill, either directly
to uncovered waste or to a layer of high-permeability drainage media on top of the
waste. The liquids migrate downward into the landfilled waste under the influence of
gravity and capillary (suction) forces, although some liquids may be lost as a result
of evaporation (and possibly transpiration). Surface systems are often selected
because of their relative simplicity with respect to construction and operation. Unlike
subsurface systems, surface systems can normally be constructed with existing or
readily available equipment and supplies. Surface systems have been widely utilized
at landfill sites where the primary goal was leachate management via recirculation as
opposed to control of biological reactions (Barber and Maris 1984), and were often
an early method employed at landfill sites that later evolved to using more elaborate
subsurface techniques (Watson 1987; Townsend et al. 1995; Mehta et al. 2002).
While surface systems provide a cost-effective and relatively easy-to-construct
methodology for introducing liquids, this approach presents several potential con-
cerns and limitations. From an environmental, human health, and aesthetic perspec-
tive, these systems often result in an increased potential for leachate exposure.
Exposure of workers to airborne leachate, odor from leachate, and increased oppor-
tunities for contamination of stormwater are all issues that must be assessed as part
Direct application at the working face typically involves a water tanker truck that
carries leachate to the working face and then discharges the leachate by a hose, rear-
mounted spray bar, or spray nozzle to the waste before the application of daily/
intermediate cover. Since many landfills are already equipped with water trucks for
7.2 Surface System Configuration 153
Fig. 7.2 Spray application of leachate at the landfill surface using a tanker truck (Photo courtesy
of John Schert)
dust control, this method of surface application is often the most familiar and direct
method for a landfill operator to implement. Figure 7.2 shows the process of liquids
application on the landfill’s active face via tanker truck application using a front-
mounted spray nozzle.
As with many surface liquid addition techniques, when the primary objective is
to increase moisture content of the waste, tanker truck application is typically lim-
ited to recently-deposited waste that has not yet been covered with soil. While some
evaporation will occur, addition directly to the waste promotes retention of moisture
within the landfill and limits potential for leachate runoff from the desired applica-
tion area. Some operators construct temporary berms of soil around the application
area to minimize leachate run-off to side slopes and other areas that are not targeted
for leachate recirculation. Liquids distribution from the truck using the rear spray
bar may be feasible, but only if the truck has access to the application area; if newly-
deposited waste is the target, access by the water truck may be limited. Thus,
although use of the truck’s spray nozzle is often more practicable, this method
requires more operator control, but allows for better liquids distribution. An alterna-
tive is to introduce the liquids using a hose connected from the truck to the waste,
but since this does not provide as efficient distribution, controls to prevent migration
from the application area are important (e.g., soil berms).
Direct application of liquids to the waste using a water truck can result in effec-
tive moisture distribution in the areas where it is applied, and may aid in waste
compaction. As with many of the surface techniques, application is limited to peri-
ods of dry weather to minimize potential mixing with stormwater and off-site
migration. Possible concerns to landfill operators include exposure to aerosols from
leachate that is sprayed, additional odors that may result from the leachate, and the
wet conditions of the working area.
154 7 Systems for Surface Addition of Liquids
The spray irrigation of vegetated land as a means of managing both raw and treated
leachate has been practiced at landfills around the world (Gordaon et al. 1988;
McBride et al. 1989a, b). This practice has been extended to leachate application on
the landfill surface, both on closed areas where soil and grass cover the waste, and
directly on the waste prior to placement of cover soil. Spray application was one of
the first reported methods for leachate recirculation at landfills. For many operators
utilizing spray irrigation, a primary objective is reduction of leachate volume
through evaporation and transpiration.
Spray irrigation systems for large grassed areas at landfills utilize standard irriga-
tion equipment; issues with reduced spray head performance due to biological or
mineral clogging may necessitate frequent maintenance and repair. The primary
concern for application to covered landfilled areas is limiting the application rate so
that leachate mitigation outside the landfill area does not occur. This typically limits
application during dry weather conditions to rates that do not exceed the liquids
removal through evapotranspiration and infiltration into the landfill. Application
during wet weather will normally be prohibited concerns over potential stormwater
impacts often pose a regulatory hurdle.
Landfill operators also use spray irrigation to introduce liquids to uncovered
waste prior to placement of the cover soil. This is accomplished using portable
spray heads that can be moved around the landfill as the disposal area progresses.
Figure 7.3 shows a spray irrigation system for leachate on the working face of an
Fig. 7.3 Spray irrigation at the landfill surface prior to cover soil placement using portable spray
heads
7.2 Surface System Configuration 155
active landfill; standard irrigation equipment was used. Similar to issues faced with
direct wetting using tanker trucks, control of liquids migration from the waste must
be considered, and may require the use of soil berms at strategic locations around
the active disposal area.
Potential worker and customer exposure to airborne leachate represents a
commonly-voiced concern with leachate spray irrigation systems. Gray et al.
(2005) modeled potential exposure of landfill workers from spray irrigation of
leachate. Based on results from conservative worst-case exposures, they concluded
that the risk posed to landfill workers exposed to several trace organic chemicals
was minimal. Given the variable nature of leachate quality from one landfill to
another, however, a site-specific assessment may be advisable if this technique is to
be employed.
Similar to spray application, drip irrigation, if properly designed, can provide rela-
tively uniform liquids distribution at the surface of the landfill. Drip irrigation does
not pose the same problems with aerosol dispersion as spray application. Two gen-
eral drip irrigation configurations include fixed and portable systems. Fixed systems
utilize a permanent or semi-permanent pipe or tubing configuration that is placed on
the surface of a covered landfill, or more commonly, embedded within soil above
the waste. Orifices in the pipe or tubing are sized and spaced to optimize liquids
distribution when liquids are added under pressure; the drip lines are at times placed
within a bed or trench of gravel or similar permeable medium to optimize
distribution.
Portable systems are used for drip irrigation directly to waste after deposition but
prior to placement of cover material. Perforated pipes, hoses or tubing are dragged
into place by hand or using landfill equipment and connected either to a force main
or to a tanker truck. Liquids are added to the waste and the system is moved within
the targeted waste area as necessary to provide needed distribution and capacity.
Figure 7.4 shows a surface drip system consisting of perforated HDPE pipe on top
of the landfill surface. Periodic monitoring and controls to prevent migration from
the application area, such as soil berms, are important.
Although leachate aerosol concerns are not present with drip irrigation systems,
migration of leachate outside the landfill area and subsequent impacts to stormwa-
ter must be evaluated. Selecting an appropriate rate of application is important, as
the rate of liquids infiltration into compacted waste will be less than infiltration
rates associated with most soils. From an operational perspective, drip irrigation
orifices may require periodic cleaning and maintenance because of clogging from
the leachate.
156 7 Systems for Surface Addition of Liquids
Fig. 7.4 Drip irrigation piping for liquids distribution at the landfill surface (Photo Courtesy of
Waste Management Inc.)
Surface ponding involves the use of infiltration ponds, lagoons, or pits at the surface
of the landfill. Liquids are hauled by truck or pumped directly to the ponds, where
liquids added at an amount that standing liquids levels develop. Because of the
simplicity in construction and operation, surface ponding was one of the first
methods of leachate recirculation used at many landfills. Ponds provide storage
capacity for liquids that permit moisture infiltration into the landfill even when
liquids are not actively added. The standing liquids also provide additional driving
force to enhance the rate of liquids addition into the landfill.
Surface ponds have been constructed in several different manners. In some cases,
a surface layer of waste is excavated and re-compacted to construct perimeter berms
around the excavated area to provide for greater storage capacity (Townsend et al.
1995). Alternatively, a perimeter berm of low permeability soil can be constructed
on the existing landfill surface to form the pond (Warith 2002), or incoming waste
can be compacted in place to form the pond perimeter. Ponds that are excavated into
the waste must be lined with a permeable media (e.g., rock, sand) to prevent waste
from floating (which over periods of prolonged operation can still be a problem
even if the pond bottoms are covered). If berms are used, they should consist of low-
permeability soils to keep leachate from seeping through the walls. To optimize
liquids distribution, pond locations should be staggered and moved.
7.2 Surface System Configuration 157
Surface trenches represent a liquids introduction technique which relies on the use
of excavated trenches on the surface of the landfill to distribute liquids into the
upper layers of the waste mass. A liquids distribution pipe would typically be placed
in the trench and the trench would be backfilled with a permeable protective media
(e.g., stone, shredded tires). Unlike ponds, the trenches are normally covered with
soil so that the liquid surface is not visible at the surface of the landfill. To prevent
soil from migrating into the trench and filling in voids of the permeable media, a
geotextile will normally be placed above the bedding material prior to placement of
soil. Liquids are added to trenches using a pipe manifold or tanker truck through
vertical access pipes that connect the surface of the landfill to the buried pipe or
bedding media.
Reported trench depths used at landfills range in depth from 1 to 4.5 m (3–15 ft)
into the waste, depending on the excavator’s reach. The width of the trench is often
the same as that of the excavator bucket, with 1–1.7 m (3–5 ft) being a typical range.
Two approaches used for surface trench liquids addition are distinguished here as
shallow trenches and deep trenches. Shallow trenches are excavated into the top
layer (approximately 1 m) of the waste and covered with soil. This approach offers
the advantage of a relatively simple construction procedure. The rate of liquids
application will be limited by the maximum depth of liquids that can be safely pon-
ded without exiting the trench and causing leachate migration issues. Figure 7.6
shows a shallow surface trench system under construction. In this system, perforated
HDPE pipe surrounded by whole tires was covered by a geotextile and then by com-
pacted soil. Trenches excavated then ultimately buried within the waste (e.g., subsur-
face horizontal systems; see Chap. 9) are most often constructed as shallow trenches.
Deep trenches are still excavated at the surface of the landfill, but they can extend
into the landfill to depths of 4–5 m or more. Bedding media and pipe are placed in
the bottom of the trench, but the remaining trench volume is backfilled with com-
pacted waste (a geotextile might be used to separate the bedding media from the
waste, but this is less common in deep systems). Each completed trench is covered
with soil and an appropriate inlet pipe for liquids addition provided. Figure 7.7 illus-
trates the construction of a deep trench used for surface application of liquids. The
use of deeper trenches allows a greater volume of liquids to be stored in the trench,
and the greater depth (as well as the backfilled waste) offers the operator the poten-
tial to add liquids under some degree of pressure. Depending on the depth of the
trench from the surface, short-circuiting of leachate back to the surface may be a
problem if enough waste is not added on top of the trench. With respect to design,
deeper surface trenches operated under pressure are more appropriately designed
following the subsurface techniques described in Chap. 9.
The primary design variables associated with the sizing of a surface liquids applica-
tion system include the target liquid addition volume, the application time, and the
application rate. Considerations for selecting the target addition volume were
160 7 Systems for Surface Addition of Liquids
discussed in Chap. 6. The duration of application will be dictated by the target vol-
ume and addition rate, as well as other site-specific constraints such as waste filling
rates, precipitation amounts and frequency, and stormwater control methods. Given
that the primary driver for liquids infiltration into the landfill will be gravity, the rate
of surface liquids addition is largely controlled by the hydraulic conductivity of the
waste, the area of the application, and the depth of liquids ponded on the surface of
the waste. Other design elements (e.g., pumping and storage systems, stormwater
control infrastructure) are described elsewhere in the book.
When designing a system to apply liquids directly to the waste surface by spraying
(or similar application techniques), the application rate should not result in any
excessive ponding or migration from the application zone. The designer could use
software that models unsaturated flow (see Chap. 5); the surface application of liq-
uids where the liquids are not ponded above the waste would likely be an unsatu-
rated flow case. A simple approach, however, is to specify an application rate, q,
equal to the vertical saturated hydraulic conductivity (KZ) at the surface of the land-
fill. At a unit gradient (i = 1), the liquid infiltration rate per unit landfill surface area
into the landfill (L/T or L3/L2 T) would be equal to KZ as shown in (7.1).
q = KZ i = KZ (7.1)
Table 7.1 presents a range of unit gradient infiltration rates for various KZ values
representative of what is typically reported in the literature. As described in Chap. 5,
vertical hydraulic conductivity of compacted waste is greatest at the surface of the
landfill where it is exposed to the least overburden pressure.
Table 7.1 Unit gradient infiltration rates at different vertical hydraulic conductivities
Hydraulic conductivity Infiltration rate at unit gradient
æ cm ö æ m ö æ m3 ö æ gal ö
ç s ÷ ç ÷ ç ÷ ç ÷
è ø è day ø è hectare - day ø è acre - day ø
5 × 10−4 0.432 4,320 469,000
1 × 10−4 0.0864 864 93,900
5 × 10−5 0.0432 432 46,900
1 × 10−5 0.00864 86.4 9,390
5 × 10−6 0.00432 43.2 4,690
1 × 10−6 0.000864 8.64 939
7.3 Design Methodology 161
The design engineer may wish to factor in evaporative losses when calculating
an expected achievable application rate for a spray irrigation system. As spray irri-
gation of wastewater effluent is a common practice, design manuals for these sys-
tems can provide guidance for spray field configuration and application rates that
include evaporation and transpiration (US Environmental Protection Agency 2006).
Evaporative losses from spray irrigation systems are a function of water droplet
size, air temperature, humidity, and air velocity (Kincaid and Longley 1989; Tarjuelo
et al. 1999). As small droplet size will lead to substantial evaporative losses due to
greater amounts of surface area, specification of efficient spray nozzles (as well as
practicing maintenance) is important for maximizing evaporation (if this is an
objective). Estimating evaporation of water from ponded systems (such as surface
ponds described in the next section) can be estimated by applying appropriate pan
evaporation data and a corresponding pan coefficient.
The difference between surface ponding and surface application through spray and
drip irrigation is that a larger amount of liquids are added with ponding such that
infiltration of liquids into the waste occurs continually. The pressure head build-up
associated with the ponding technique has the benefit of providing a greater driving
force for liquids movement into the landfill. Figure 7.8 provides a conceptual illus-
tration of a surface infiltration pond at a landfill.
The infiltration rate of a surface pond can be simply expressed using Darcy’s
Equation:
h+d
q = Kz (7.2)
d
Fig. 7.8 Definition sketch for calculation of liquid infiltration rate into landfill from surface pond
162 7 Systems for Surface Addition of Liquids
Where, q = the vertical infiltration rate per area (L3 L−2 T−1); Kz = the vertical
hydraulic conductivity (L T−1); h = the depth of the surface pond (L); and d = the
depth of the saturated waste under the pond (L) (see Fig. 7.8 for a definition sketch
of the system). Soon after ponding begins, when the depth of liquids is lowest rela-
tive to the depth of the saturated waste, the infiltration rate of liquids into the landfill
is at its greatest. As the saturated zone moves downward into the landfill, the driving
gradient approaches 1, and qZ approaches KZ. Although the vertical hydraulic con-
ductivity is considered a constant, in practice, it decreases with the depth due to
overburden pressure. In this approach, seepage from the pond walls is neglected. An
approach for including liquids migration from the pond walls is provided in the
subsequent section on surface trenches. Depending on local climatic conditions, the
effects of evaporation and precipitation should be considered in sizing the ponds.
Townsend et al. (1996) measured the performance of surface infiltration ponds at
a landfill in Florida. Infiltration was measured by conducting a daily water balance
on four separate ponds and estimating evaporation. The observed surface infiltration
rates (qz) ranged from 0.005 to 0.02 m/day (5,500–17,000 gal/acre-day). These
infiltration estimates were used to assess the waste hydraulic conductivity (see
Chap. 5).
The description and modeling of surface ponds in the previous section was such that
only flow from the bottom of the ponds was considered. In large pond areas, the
exposed infiltration area on the sides of the ponds will be small relative to the bot-
tom area. This will not be the case for surface trenches, however, and given the
anisotropic nature of landfilled MSW, accounting for flow from the trench sides is
important.
Singh (2010) tested surface trenches containing waste tires as a bedding material
and measured infiltration rates. The trenches were 1 m wide and 1.2 m deep. After
16 days of operation, flow rates in each trench (Q) normalized to a unit trench length (L)
ranged from 0.69 to 0.95 m2/day (56–77 gal/ft-day). For comparison purposes,
when these measurements are normalized to only the bottom area of a trench (sides
excluded), the values are considerably larger than infiltration rates measured by
Townsend et al. (1995) for surface ponds; this illustrates the role that flow through
the more permeable trench walls can play.
Jain et al. (2010) developed a method for predicting flow through a horizontal
source. This technique and its utility are presented in greater detail in Chap. 9, but
the technique can be applied to surface trenches and ponds. Figure 7.9 presents a
design chart, which includes a definition sketch. Based on dimensions of the trench
(length and width) and properties of the waste (anisotropy, a = KX/KZ), a dimension-
less value, η, can be obtained. From this, a dimensionless flow rate may be esti-
mated, and using the value of trench width (l) and KZ, the steady-state flow into the
trench can be predicted.
References 163
Fig. 7.9 Design chart for estimating steady state flow into horizontal source (trench or pond) at
the landfill surface (from Jain et al. 2010)
As the trench width (l) becomes greater than the depth of liquids in the trench (w),
the value for qs (the dimensionless flow rate) approaches a minimum value close to
1. This condition represents an infiltration pond and qZ would approach KZ (as
described in the previous section). As the degree of anisotropy gets larger, or as the
trench depth increases, the values for η increases, illustrating the greater impact of
liquid infiltration through the sides of the trench.
References
Barber M, Maris PJ (1984) Recirculation of leachate as a landfill management option: benefits and
operation problems. Q J Eng Geol Hydrogeol 17:19–29
Gordaon AM, McBride RA, Fisken AJ, Voroney RP (1988) Effect of landfill leachate spraying on
soil respiration and microbial biomass in a northern hardwood forest ecosystem. Waste Manag
Res 6:141–148
Gray D, Pollard S, Spence L, Smith R, Gronow J (2005) Spray irrigation of landfill leachate: esti-
mating potential exposure to workers and bystanders using a modified air box model and gen-
eralized source term. Environ Pollut 133:587–599
Jain P, Tolaymat T, Townsend T (2010) Steady state design of horizontal systems for liquids addi-
tion at bioreactor landfills. Waste Manag 30(12):2560–2569
Kim H, Townsend T (2012) Wet landfill decomposition rate determination using methane yield
results for excavated waste samples. Waste Manag 32(7):1427–1433
Kincaid DC, Longley TS (1989) A water droplet evaporation and temperature model. Trans ASAE
32(2):457–463
McBride RA, Gordon AM, Groenvelt PH (1989a) Treatment of landfill leachate by spray
irrigation—overview of research results from Ontario, Canada. I site hydrology. Bull Environ
Contam Toxicol 42:510–517
164 7 Systems for Surface Addition of Liquids
McBride RA, Gordon AM, Groenvelt PH (1989b) Treatment of landfill leachate by spray
irrigation—overview of research results from Ontario, Canada. II soil quality for leachate dis-
posal. Bull Environ Contam Toxicol 42:518–525
Mehta R, Barlaz AM, Yazdani R, Augenstein D, Bryars M, Sinderson L (2002) Refuse decomposi-
tion in the presence and absence of leachate recirculation. J Environ Eng-ASCE 128(3):
228–236
Singh K (2010) Performance evaluation of surface infiltration trenches and anisotropy determina-
tion of waste for municipal solid waste landfills. Master’s Thesis, University of Florida
Tarjuelo JM, Montero F, Honrubia FT, Ortiz JJ, Calvo MA (1999) Analysis of uniformity of sprin-
kle irrigation in a semi-arid area. Agric Water Manage 40:315–331
Townsend T, Miller WL, Earle J (1995) Leachate-recycle infiltration ponds. J Environ Eng-ASCE
121(6):465–471
Townsend T, Miller WL, Lee H, Earle JFK (1996) Acceleration of landfill stabilization using
leachate recycle. J Environ Eng-ASCE 122(4):263–268
US Environmental Protection Agency (2006) Land treatment of municipal wastewater effluents,
EPA/625/R-06/016. National Risk Management Research Laboratory, Cincinnati
Warith M (2002) Bioreactor landfills: experimental and field results. Waste Manag 22:7–17
Watson RP (1987) A case study of leachate generation and recycling at two sanitary landfills. In:
Proceedings from the technical sessions of the GRCDA 25th annual international seminar,
equipment, services, and systems show, vol 2, Saint Paul, 11–13 Aug 1987
Chapter 8
Buried Vertical Systems for Liquids Addition
Abstract Chapter 8 presents the second of three chapters that explore major liquid
additions systems types, with the focus of this chapter being buried vertical systems.
Configuration options, construction options, and materials of construction are dis-
cussed, including small-diameter and large-diameter systems. Design approaches
with vertical wells are presented along with operational experience to inform the
designer of potential opportunities and drawbacks. Coupled with the design discus-
sion is a presentation of several design charts and tools to identify and justify the
selected spacing of liquids addition devices. The chapter finishes with a discussion
of operations, monitoring, and closure consideration related to vertical systems.
The two general configurations for subsurface (buried) liquids addition systems are
vertical wells and horizontal trenches or galleries. This chapter describes the design,
construction, and use of vertical wells for the addition of liquids into landfills. The
concept of these devices, as illustrated in Fig. 8.1, is that a vertical borehole is con-
structed within landfilled waste, allowing liquids to be added to a range of depths within
the waste mass. As a subsurface system, vertical wells avoid many of the issues associ-
ated with the surface systems reviewed in the previous chapter (odors, aerosols, disrupt-
ing surface conditions, and impacts from inclement weather). From a performance
perspective, vertical wells can have an advantage over surface systems in that the poten-
tially large hydrostatic head of water within the well can provide a comparatively larger
driving force (pressure) to encourage liquids distribution within the landfill.
Vertical wells are commonly used in active GCCS, thus many landfill operators
have familiarity with vertical well construction techniques. One advantage of verti-
cal wells is that they can be installed after large depths of waste have been placed
(as we will see in the next chapter, buried horizontal systems are limited to construc-
tion at relatively shallow depths). This approach is thus well-suited to sites where
liquids addition operations are conceived or initiated after most of the planned land-
fill operation or filling has been completed. It may also be a preferred option for
A variety of construction techniques have been attempted for vertical liquids addi-
tion systems. Most techniques involve installing the well after the waste has been
placed and compacted, and thus require specialized equipment for drilling a hole
into the landfill. An alternative, however, is to construct the well as waste is being
8.2 Configuration, Construction and Materials 167
Fig. 8.2 Bucket auger rig for drilling vertical wells in landfill waste
Fig. 8.3 Hollow stem auger equipment for drilling vertical wells in landfill waste
Fig. 8.4 Solid shaft open flight auger for drilling vertical wells in landfilled waste
8.2 Configuration, Construction and Materials 169
Fig. 8.5 Direct push rig installing vertical wells at the surface of a landfill
geotechnical sampling and water well drilling. The use of these devices for small-
diameter liquid addition well installation is discussed in more detail in Sect. 8.2.3.
Another device commonly used for installing small diameter wells in soil is a
direct push rig or direct push technology (DPT), which involves hammering the
well pipe into place. A DPT rig was tested at a Florida landfill as part of small-
diameter (5-cm) liquids addition well installation (Fig. 8.5), but the maximum
reachable depth was 6 m (20 ft). The maximum depth that could be achieved would
be a function of the waste characteristics (the presence of rigid debris that would
cause drill refusal) and those of the rig itself.
In the context presented here, large-diameter wells are differentiated from small-
diameter wells in several ways. First, large-diameter wells are in most respects the
same as wells commonly installed for landfill gas collection, with diameters ranging
from 0.6 to 1 m. Second, a major fraction of the volume of the borehole is filled with
a permeable media, typically rounded stone. A perforated pipe (either HDPE or
PVC) is placed in the center of the rock, with sufficient distance between the landfill
surface and the beginning of pipe perforations to avoid losing liquids to the surface.
While large-diameter wells can be constructed during the progression of waste
170 8 Buried Vertical Systems for Liquids Addition
filling by continuously adding new segments, the use of specialized drilling equip-
ment to construct a well after waste has been placed is more common. Large-
diameter bucket augers are most frequently employed, but large diameter open
flight augers may also be used. With a bucket auger, the bucket is typically removed
from the hole every 0.15 m to remove the cuttings. When open flight augers are
used, drill cuttings emerge from the borehole as the auger stem is rotated. In both
cases, the cuttings must be removed and appropriately disposed of.
Once the boring is completed to the target depth, the well pipe is lowered into the
hole, with an effort made to keep the pipe in the center of the hole. Permeable media
(e.g., non-calcareous rock) is backfilled between the waste mass and the pipe. A seal
of concrete and/or bentonite (clay) is placed as part of the surface completion step
above the top of the rock and below the landfill surface to minimize possible entry
and exit of liquids or gases around the well pipe.
In contrast to large diameter wells, small diameter wells as presented here are those
that involve augering a hole in the landfilled waste using mechanized equipment
and inserting a perforated pipe within the hole without the presence of surrounding
drainage media. In this case, the pipe is in direct contact with the surrounding waste
mass. Both solid shaft and hollow stem augers can be used, but the most commonly-
reported application has been the use of a solid shaft open flight auger and PVC
pipe. The bottom sections of the pipe are slotted or perforated and the top part is
solid to minimize the potential of surface seeps during liquids addition operation.
Figure 8.6 conceptually illustrates the method through which vertical wells were
installed at the New River Regional Landfill in Florida (see Chap. 4). Each solid
shaft drill stem was 1.6 m long, and the first stem was equipped with a tool for cut-
ting into the landfill. As needed, additional lengths of drill stem were added
(Fig. 8.7). Periodically, the drill shaft was rotated in place without advancing the
depth; this action resulted in drill cuttings being brought to the landfill surface and
assisted in enlarging the hole.
When the target well depth was reached, the drill shaft was again rotated without
advancing for an extended period to clean the hole of as many cuttings as possible. At
this point, the drill stem was removed from the hole without rotating; this helped keep
any remaining cuttings on the stem, thus removing them from the hole. As soon as the
final piece of stem was removed, the lowest portion of the well pipe was inserted into
the boring (Fig. 8.8). The pipes were connected as they were lowered into place in the
hole. Threaded or glued connections can be used, although threaded connections
have shown to be more quickly deployed during construction. When drilling small-
diameter wells in landfills, the borehole tends to close back on itself relatively quickly,
requiring mechanical force to push the pipe to the desired depth in some cases.
One of the critical requirements during vertical well installation of any kind is
close monitoring of the length of drill stem in the augered borehole (Fig. 8.9).
Targeted well depths are typically designed to provide at least 3 m (10 ft) of buffer
8.2 Configuration, Construction and Materials 171
Fig. 8.6 Illustration of small diameter liquids addition wells as installed at the New River regional
landfill. (a) Auger into landfill (b) spin auger in place to remove waste and clear hole (c) pull auger
from hole without spinning (d) place pipe immediately in hole and add clay seal
between the bottom of the well and the top of the LCRS. This is needed to both
avoid short-circuiting of the liquids to the LCRS and to avoid damage to the liner
from the drill stem. It is thus critical that the engineer provide specific instructions
for the depth for each specific well location based on accurate landfill surface eleva-
tion measurements and record drawings of the liner system and LCRS.
A common practice when constructing vertical wells is to place a low
permeability seal or collar in the annulus between the pipe and waste somewhere
above the well screen and up to the landfill surface (Fig. 8.10). This helps avoid
172 8 Buried Vertical Systems for Liquids Addition
Fig. 8.9 Careful recording of auger depth into the landfill is critical to avoid damage to the bottom
liner system
Fig. 8.10 A collar of bentonite (clay) being added to the annulus at the surface of a recently con-
structed liquids addition well
174 8 Buried Vertical Systems for Liquids Addition
undesirable liquid entry into the hole (e.g., stormwater intrusion) and liquid or gas
escape from the landfill. Under most designs where vertical liquids addition wells
are used, the depth of liquid is maintained below the surface of the landfill. In
some cases the engineer or operator may want to operate at hydrostatic pressures
above the surface elevation of the landfill. Experience of how these well seals
perform to prevent surface seepage of added liquids is discussed in more detail
later in this chapter.
The design process for vertical wells begins with an assessment of the target volume
of liquids to be added to the landfill and an evaluation of the timeframe and rate at
which that volume is to be added to the landfill. Chapter 6 provides more informa-
tion on these design steps. Once the target liquids addition volume and overall
design flow rate have been established, the engineer proceeds with the design of an
individual well. The design of a single well includes the specification of the well
diameter (both the auger boring and the pipe), construction materials, screen length,
and well depth. The design of a vertical well system involves locating (at proper
spacing) a sufficient number of vertical wells across the landfill and providing a
delivery system to convey liquids to each of the wells. A primary objective in
assigning the number of wells and their location is to provide a system that allows
the operator to efficiently distribute liquids throughout the areas of landfill targeted
for liquids addition. Two major design parameters that must be identified for a given
landfill include the flow rate that can be added to a given well and the shape of the
saturated zone that results from adding that flow rate to that well. It is important for
the designer to understand that the fluid flow patterns predicted with methods out-
lined are idealized, and systems as heterogeneous as landfill should be expected to
be much more variable, both spatially and with time. The engineer should use these
techniques to develop an understanding of likely performance ranges, and couple
this with good engineering judgment and system-specific objectives.
Both the flow and dimension of the saturated zone can be predicted by the engi-
neer using fluid flow modeling as described in Chap. 5. Several authors have pre-
sented examples of such modeling for vertical liquid addition wells in landfills.
McCreanor and Reinhart (1996) used the saturated-unsaturated flow and transport
model (SUTRA) to simulate the saturation profiles that would occur around a
vertical well in homogenous and isotropic waste at several constant flow rates.
Using SUTRA, Jain et al. (2005b) modeled moisture flow through a vertical well
installed in unsaturated waste. He reported pressure at the bottom of the well and the
lateral extent of the zone of impact as a function of waste properties, well dimen-
sions, flow rate, and time. Khire and Mukherjee (2007) simulated the impact of
leachate injection rate on the steady-state injection pressure, the lateral extent of
moisture movement, and head on the bottom liner for an isotropic waste. The impact
of well diameter, well depth, and screen length was also investigated.
8.3 Design Methodology 175
Fig. 8.11 Example output of a seepage model simulation (SEEP/W) of pressurized liquids
addition into a vertical well
Fig. 8.12 Definition sketch for major dimensions associated with estimate of liquids into a verti-
cal well in a landfill
The following sections discuss historic data measured at landfill sites using
v ertical wells for liquids addition and a design method that can be used to predict
achievable flow rates and saturated zone dimensions for vertical liquids addition
wells. The design methods presented allow the engineer to estimate these parame-
ters without performing computer simulations. Figure 8.12 defines the system along
with appropriate dimensions.
Fig. 8.13 Example data from the operation of a vertical liquids addition well at NRRL; liquid
depth and flow rate as a function of time
2005b). Clusters of wells were installed, with each cluster containing three wells
installed at 6, 12 and 18 m (20, 40, and 60 ft) depth. Reported flow rates ranged
from 0.0019 to 0.011 m3/min (0.5–2.91 gal per minute); Fig. 8.13 illustrates typical
well performance for a single well over several days of operation. As part of initial
site operations, Jain et al. (2006) observed that achievable flow rates decreased with
depth in the landfill (hydraulic conductivity was lower at greater waste depths).
Longer-term performance of the NRRL liquids addition well field was reported
in Jain et al. (2014a). More than 25,000 m3 (6,600,000 gal) were added over a 5-year
period. The performance was evaluated based on fluid conductance, defined as flow
rate per unit well screen length per unit liquid head at the well bottom (units = L/T).
Figure 8.14 presents variation in fluid conductance with the liquids volume added.
The median fluid conductance was found to range from 5.6 × 10−8 to 3.6 × 10−6 m s−1
for all wells.
The achievable flow rate into a vertical well can be predicted by the engineer using
a fluid flow simulation technique as described in Chap. 5. Figure 8.11 provided the
result of an example of such a simulation under a defined set of conditions. The flow
rate decreases during the first part of operation as the wetting zone progresses, ulti-
mately reaching steady state.
Several researchers have developed approaches that allow estimation of flow
rates into a vertical well without the need to conduct model simulations. Xu et al.
(2014) conducted a series of SEEP/W simulations for a range of potential operating
conditions and developed a best-fit relationship of the simulation results to produce
a simple equation capable of estimating flow-rate (and several other parameters as
178 8 Buried Vertical Systems for Liquids Addition
Fig. 8.14 Fluid conductance as a function of liquids volume added for (a) deep wells, (b) middle
wells, and (c) shallow wells
discussed later in this chapter). The Xu et al. (2014) assessment simulated the verti-
cal well as a line-source under axisymmetric flow conditions. Using this approach,
the entire length of well screen is assumed to be saturated, with no liquid level
above the well screen. The following relationship for steady state flow into a verti-
cal well was developed.
(
Q = 0.61 × A × K z × L2 ) (8.1)
Where, Q = the flow rate of leachate (L3 T−1); A = waste anisotropy ratio (Kx/Kz;)
Kz = vertical hydraulic conductivity (L T−1); and L = well screen length (L).
Jain et al. (2010) developed an approach to estimate steady-state flow rate into a
vertical well as a function of well dimensions, injection pressure, and waste
hydraulic conductivity and anisotropy using dimensionless parameters and design
charts. SEEP/W model simulations were conducted over a range of operational con-
ditions for vertical well systems. In addition to the parameters assessed by Xu et al.
(2014), Jain et al. (2010) included the radius of the well and conditions where liq-
uids were added at pressures greater than the screen length of the well. A dimen-
sionless variable analysis was conducted to broaden the scope of applications for
the results beyond the range of individual parameter values used for modeling. Use
of the design process proposed by Jain et al. (2010) is described below.
First, the dimensionless variable η is calculated using information on well dimen-
sions and landfill anisotropy:
LV 2
h= A (8.2)
rw2
8.4 Flow Rates 179
Where LV = the screen length of the vertical well, rw = the radius of the vertical
well, and A = the anisotropy ratio (KX/KZ). The dimensionless variable η indicates
the dominant flow direction, vertical or horizontal; a low η value signifies a flow
that is dominant in the vertical direction, whereas a high η value indicates a flow
that is dominant in the horizontal direction.
The designer identifies a target liquid level in the vertical well (hv, measured
from the bottom of the well), which allows for the depth of liquids to be greater than
the screen length. Using the targeted liquid level, hv, and the well screen length, LV,
a dimensionless injection pressure head, pId, is calculated as follows:
hV
pId = (8.3)
LV
Once the values of pId and η have been determined, the steady-state dimension-
less flow rate, qs, for a vertical well can be estimated using the chart presented in
Fig. 8.15.
With a value of qs in hand, the steady state flow rate into the vertical well (QS)
can be calculated as:
Qs = qs p rw2 K Z (8.4)
Fig. 8.15 Design chart for estimating steady state flow (QS) into a vertical source under buried
conditions (from Jain et al. 2010)
180 8 Buried Vertical Systems for Liquids Addition
Jain et al. (2014b) also presented a design chart to estimate average flow rate for
conditions where the system does not reach steady state. The design chart provides
an estimate of the error that might result from the use of steady-state flow rate in the
design process decreases with an increase in the fraction of liquids volume needed
to achieve steady state. The use of the steady-state flow rate for estimation of oper-
ating duration to add designed liquid volume may result in slight overestimation of
the operating time.
The ability to estimate the size and shape of the saturated zone surrounding a verti-
cal well can be of great value when determining the appropriate placement or spac-
ing of liquids addition devices. The engineer and operator must consider numerous
factors that may result in non-idealized flow conditions (e.g., cover soil layers,
waste heterogeneity) and incorporate such conditions into design and operation.
Both Xu et al. (2014) and Jain et al. (2010) used the output of vertical well simula-
tions to develop a method for predicting the wetted zone around a vertical well at
steady state. Refer to Fig. 8.12 for the definition sketch of pertinent dimensions in
this approach.
Xu et al. (2014) examined the lateral spread of liquid away from a vertical well
as a function of the maximum steady-state moisture distribution (Xmax). At steady
state, the maximum lateral spread is reached and the added liquids will only migrate
downward in the vertical direction. According to Darcy’s Law, the maximum lateral
spread for a vertical well (Xmax) injection can be expressed as:
Q
X max = (8.5)
p Kz
A correction factor was developed that allowed for the estimation of lateral
spread at a distance, D, from the top of the well, such that:
æ -1.6
D
ö
X = X max × ç 1 - e LV ÷ (8.6)
ç ÷
è ø
Where, D = the depth measured from the top well screen (L) and LV = the length
of well (L). The lateral distance at the base of the well (where D = L) would thus be:
to cases where the water level is within the screened interval. Using the d imensionless
analysis approach simplified by Jain et al. (2010), the relationship for Xwell was
found to be:
Qs
X well = (8.8)
p KZ
The engineer must specify the number of vertical wells for installation and their
locations. While some measured data are available regarding the success of vertical
wells for distributing moisture (see below), the engineer will need to decide upon an
appropriate well spacing based on site-specific conditions and project objectives
coupled with estimates of likely expected moisture movement within the landfill.
The information presented in Sect. 8.4 allows the engineer to estimate the flow rate
that can be added to a given vertical well. This, coupled with the liquid addition
targets discussed in Chap. 6, can be used to provide a preliminary estimate of the
number of wells needed. The engineer can specify spacing based on previous opera-
tional experience or using methods that allow prediction of the saturated zone sur-
rounding the well.
Several projects have used 17-m (50-ft) spacing for small-diameter vertical wells
(Read et al. 2001; Jain et al. 2005b). Only limited data are available from field mea-
surements, however, regarding the distribution of moisture surrounding vertical
wells. Based on the responses of moisture sensors (Kumar et al. 2009) installed
around the NRRL vertical well clusters (50-ft spacing), the lateral extent of mois-
ture movement was reported to range from 8 to 10 m. Jain et al. (2014a, b) reported
that this system was effective in wetting the waste as the average gravimetric
moisture contents of 272 samples collected in 2007 was 45 % (wet weight basis)
compared to the initial average moisture content of 23 % (wet weight basis) (for 51
samples) collected in 2001.
Engineers often specify device spacing based on the distance needed to provide
adequate moisture distribution within the landfill. The methods in Sect. 8.5 allow
estimation of steady state zones of impact, and thus can be used for device spacing.
For example, assigning a spacing based on Xwell or Xmax, or some desired overlap,
would be a typical approach. However, the engineer may also wish to factor time
into the design. The time needed to reach steady state may be large, and thus in
cases where more rapid coverage is desired, closer spacing may be necessary.
The dimensionless design chart approach described earlier can be extended to
determine the lateral extent of liquid movement at the base of the well (Xwell) at
times prior to reaching steady state. First, η is calculated in the same manner as
presented in Sect. 8.4. Then using Fig. 8.16, the number of pore volumes needed to
182 8 Buried Vertical Systems for Liquids Addition
Fig. 8.16 Design chart to determine the cumulative volume of added liquids needed to reach
steady state (from Jain et al. 2010)
reach steady state (Vn,critical) can be determined for different pID values. The
cumulative volume of liquids to reach steady state (Vt,critical) can be calculated as
follows:
(
Vt ,critical = Vn,critical p rw2 w (q s - q d ) ) (8.9)
Where rw and Lv are as previously defined θs is the porosity and θr is the residual
moisture content.
Figure 8.16 presents fractions of the steady-state lateral extent (ratio of transient
lateral extent (Xwell) to the steady-state lateral extent (Xwell,s)) achieved as a function
of the fraction of steady-state liquids volume (i.e., ratio of the design transient vol-
ume (Vt) to the volume needed to achieve the steady-state condition (Vt,critical) for
vertical well) as published by Jain et al. (2014b). A ratio of 1 represents the steady-
state condition whereas ratios less than 1 represent transient conditions. As can be
seen in Fig. 8.17, coverage of approximately 70–90 % of the lateral extents of the
zone of impact is achieved by addition of only 40 % of the liquids volume needed
to achieve steady state for a vertical well.
With a value of Vt,critical in hand, the value of Xwell can be estimated as a function
of added volume (Vt) using Fig. 8.17. First, the ratio of Vt to Vt,critical is calculated.
Then a value of Xwell / Xwell,s is estimated using Fig. 8.16. As the SEEP/W simulation
results did not converge on a simple relationship, a range is presented and the
designer would need to select an appropriate value of Xwell/Xwell,s.
8.7 Operation, Monitoring and Closure 183
Fig. 8.17 Design chart to determine the fraction of the radial extent of flow from a vertical well
as a function of the cumulative volume added
These design approaches reflect the technical aspects of design required for
v ertical wells. The design engineer and site operator must also consider other fac-
tors such as cost and compatibility with current and future operations when finaliz-
ing the number and configuration of vertical wells. Economics are addressed further
in Chap. 18.
While horizontal systems can be operated under pressure, vertical systems normally
require that liquid levels be maintained below the surface of the landfill and thus
pressure is limited to the depth of the well below the landfill surface. As described
earlier, construction of a vertical well will normally include placement of a low
permeability seal (clay, concrete) to prevent the short-circuiting of leachate (liquid
and gas) in the annulus of the well to the surface of the landfill. At the NRRL proj-
ect, this was found insufficient to prevent liquids migration when the liquid surface
in the well was above the landfill surface. Further attempts at the NRRL to examine
pressurized addition at vertical wells explored the placement of a concrete collar
around the vertical wells (Fig. 8.18). While this was more effective than a simple
clay seal around the well, surface seepage still occurred. Jain et al. (2014a, b)
reported that the liquid depths within the well had to be maintained below the land-
fill surface to avoid seeps at the base of the wellheads; therefore, system operation
was labor intensive, especially for wells installed at shallow depths.
184 8 Buried Vertical Systems for Liquids Addition
Fig. 8.18 Installation of a concrete collar around a nest of vertical injection wells
A challenge of operating vertical well systems, especially those with many wells,
is maintaining sufficient addition rates to achieve liquid levels efficient for driving
moisture distribution, but not large enough to result in surface seeps. This requires
relatively frequent operator monitoring and adjustment. Routine liquid level mea-
surements are necessary (see Chap. 6 for a discussion of monitoring techniques).
Settlement is also an issue that requires ongoing monitoring and maintenance.
The settlement of waste beneath and surrounding a vertical well can result in
“extending” the well to a point above the landfill surface that makes operations and
monitoring difficult. The degree of settlement at any point depends on the
underlying waste thickness. Since the waste thickness below the bottom of the well
is less than the total waste thickness at the well location, the vertical wells settle
less than the surrounding landfill surface. The designer and operator should expect
vertical liquids addition wells to continue extending above the landfill surface in a
similar manner, and at even more pronounced magnitude as a result of enhanced
waste stabilization and consolidation. At NRRL, clusters of wells of different
depths settled at different rates because of different depths of waste beneath them.
The engineer must provide a flexible design that allows the operator to routinely
adjust the connection between the well and the liquids distribution or gas collection
manifold.
Vertical liquids addition wells also present an operational challenge in that waste
will preferentially settle around the well as this is where most of the liquids are
added. Greater liquids addition volumes result in greater weight which increases the
stresses causing settlement and also results in more waste decomposition and more
volume loss. Depressions may form around vertical wells which, if not addressed,
will result in low spots for water to pond, thus making operator access difficult. This
8.7 Operation, Monitoring and Closure 185
Fig. 8.19 Differential settlement around a cluster of vertical liquids addition wells that resulted in
ponding of stormwater
Fig. 8.20 Illustration of the construction of a buried vertical well system employed at the New
River landfill in Florida. (a) Initial installation of the vertical well. (b) Connection of the vertical
well to a horizontal manifold. (c) Placement of a lift of waste on top of the vertical wells and day-
lighting of the manifold on the side slope of the landfill
References 187
References
Jain P, Powell J, Townsend TG, Reinhart DR (2005a) Air permeability of waste in a municipal
solid waste landfill. J Environ Eng 131(11):1565–1573
Jain P, Farfour WM, Jonnalagadda S, Townsend TG, Reinhart DR (2005b) Performance evaluation
of vertical wells for landfill leachate recirculation. In: Proceedings of Geo Frontier 2005,
ASCE conference, Austin
Jain P, Powell J, Townsend TG, Reinhart DR (2006) Estimating the hydraulic conductivity of
landfilled municipal solid waste using the Borehole permeameter test. J Environ Eng 132(6):
645–652
Jain P, Townsend TG, Tolaymat T (2010) Steady-state design of vertical wells for liquids addition
at bioreactor landfills. Waste Manag 30:2022–2029
Jain P, Ko J, Kumar D, Powell J, Kim H, Maldonado L, Townsend TG, Reinhart DR (2014a) Case
study of landfill leachate recirculation using small-diameter vertical wells. Waste Manag
34(11):2312–2320
Jain P, Townsend TG, Tolaymat T (2014b) Transient design of landfills liquids addition system.
Waste Manag. doi:10.1016/j.wasman.2014.05.008
Jonnalagadda S, Kumar D, Jain P, Gawande N, Townsend T, Reinhart D (2010) Comparison of
resistivity and time domain reflectometry sensors for assessing moisture content in bioreactor
landfills. Geotech Test J 33(3):183–191
Kadambala R, Powell J, Singh K, Townsend TG (2014) Evaluation of buried vertical well leachate
recirculation system for municipal solid waste landfills. Unpublished manuscript
188 8 Buried Vertical Systems for Liquids Addition
Kelly R, Shearer B, Jongmin K, Goldsmith CD, Hater G, Novak J (2006) Relationships between
analytics methods utilized as tools in the evaluation of landfill waste stability. Waste Manag
26:1349–1356
Khire MV, Mukherjee M (2007) Leachate injection using vertical wells in bioreactor landfills.
Waste Manag 27(9):1233–1247
Kim H, Townsend T (2012) Wet landfill decomposition rate determination using methane yield
results for excavated waste samples. Waste Manag 32(7):1427–1433
Kumar D, Jonnalagadda S, Jain P, Gawande N, Townsend T, Reinhart D (2009) Field evaluation of
resistivity sensors for in situ moisture measurement in a bioreactor landfill. Waste Manag
29:1547–1557
McCreanor PT, Reinhart DR (1996) Hydrodynamic modeling of leachate recirculating landfills.
Water Sci Technol 34(7–8):463–470
Morris JWF, Vasuki NC, Baker JA, Pendleton CH (2003) Findings from long-term monitoring
studies at MSW landfill facilities with leachate recirculation. Waste Manag 23:653–666
Read A, Hudgins M, Phillips P (2001) Perpetual landfilling through aeration of the waste mass;
lessons from test cells in Georgia (USA). Waste Manag 21:617–629
Watson RP (1987) A case study of leachate generation and recycling at two sanitary landfills. In:
Proceedings from the technical sessions of the GRCDA 25th annual international seminar,
equipment, services, and systems show 2, Saint Paul
Xu Q, Townsend TG, Jain P (2014) Steady-state saturated zone equations for liquids addition
devices at landfills. Unpublished manuscript
Chapter 9
Buried Horizontal Systems
for Liquids Addition
Abstract Chapter 9 is the third and final chapter on liquids addition system types,
with a focus on horizontal systems. Buried trenches, blankets, and combination
systems are discussed as the most common horizontal system types, with a compan-
ion evaluation of potential benefits and drawbacks of each. The latter portion of the
chapter focuses on design techniques and approaches for horizontal systems, includ-
ing tools to help identify and design horizontal systems over a variety of operating
conditions and site constrains. Considerations for operation, monitoring and closure
are presented at the end of the chapter.
The advantages that subsurface methods of liquids addition have over surface addi-
tion (e.g., ability to add liquids during inclement weather, greater capacity for pro-
viding adequate liquids to the bulk of the waste mass) were described in the previous
chapter’s presentation of vertical liquid addition wells, and these advantages are
shared by horizontal subsurface systems. The subsurface liquids addition methodol-
ogy discussed here utilizes buried pipes, trenches, or beds of permeable media con-
structed horizontally in the landfill during the waste filling process. The installation
of these devices differs from vertical wells (which are installed only after a substan-
tial amount of waste has been placed) and thus provides the operator the ability to
add liquids much earlier in the operational life of the landfill.
The placement and use of horizontal liquids addition devices are among the most
common of practices used at large-scale facilities implementing liquids addition.
While this practice requires relatively frequent construction of devices throughout
the life of the landfill, the types of equipment needed for construction are those
often already part of the site’s equipment fleet (e.g., excavators, loaders), and thus
installation may be performed by the landfill staff themselves without the necessity
of an outside contractor with specialized equipment (such as a drilling rig).
Several configurations of buried horizontal systems have been utilized for liquids
addition. For the purposes of discussions herein, these are grouped as horizontal trench
(buried trench) systems and horizontal blanket (buried infiltration gallery or horizontal
gallery) systems. Both types have been utilized to distribute liquids within the landfill
mass. Horizontal systems can expand lengths to of hundreds of feet (or meters) and are
vertically offset with spacing that depends on the dimensions of the horizontal trench
or blanket among other factors (i.e., flow rates, pressures, and operational objectives).
Similar to the previous chapter on vertical systems, this chapter examines the funda-
mentals of horizontal system construction and materials, along with design consider-
ations; existing data from practicing facilities and methods for predicting achievable
flow rates and moisture distribution profiles are both discussed.
The installation of perforated pipes buried within the waste in a horizontal fashion
has been described as horizontal injection lines (HILs), horizontal injection trenches
(HITs), or simply horizontal trenches. Common to all systems is a conduit capable of
distributing liquids placed on top of a lift of waste, with the inlet of that conduit
configured to allow the introduction of liquids when desired (illustrated in Fig. 9.1).
Fig. 9.2 Illustration of the process of constructing a horizontal liquids addition trench. (a) Initial
conditions (b) scrape away cover soil (c) excavate trench (d) install first layer of bedding (e) install
pipe (f) install additional bedding (g) compacted waste over trench (h) place soil over excavation
area
Fig. 9.5 Placement of shredded tires on top of HDPE liquids distribution pipe in horizontal trench
layer to provide some protection to the pipe against stresses from waste and equip-
ment overburden and settling beneath the trench. Most designers and operators
install bedding material on the bottom of the trench followed by the perforated
pipe, followed by more bedding material (Figs. 9.4 and 9.5). Depending on avail-
ability of construction and bedding materials, as well as other site-specific con-
struction constraints, some operators may place the pipe at the bottom of the trench
194 9 Buried Horizontal Systems for Liquids Addition
Fig. 9.6 HDPE liquids distribution pipe in the process of being placed on top of a bottom bedding
layer of crushed glass in a horizontal trench
or the top of the trench. Similarly, the designer will specify the perforation scheme
used in the recirculation pipes, and care must be taken to ensure that pipe perfora-
tions are placed appropriately during installation (e.g., construction specifications
may call for the perforations to be placed vertically downward to avoid soil entrance
into the pipe).
Bedding media can include standard materials used in civil engineering drainage
systems, such as naturally rounded or crushed rock. Given the expense of these
materials, alternative bedding media originating from waste materials have been
used at many landfill sites, including chipped vehicle tires; mulch; crushed con-
crete, brick and other masonry; and crushed glass (Figs. 9.6 and 9.7). Since the
limiting factors to liquids movement into the landfill is most likely the compacted
waste, the bedding media must simply possess a permeability greater than the waste.
Because of its strength and flexibility, HDPE is the most commonly used pipe
material with diameters of 3 or 4 in. being most common. Segments of HDPE pipe
can be thermally welded inside or adjacent to the trench and the welding device
moved as needed (Fig. 9.8). Many operators prefer to weld a long length of pipe at
a central location and to drag the pipe into place (Fig. 9.9). PVC pipe has been suc-
cessfully used for horizontal injection trenches at some sites (Fig. 9.10; Townsend
and Miller 1998), and since it can be solvent welded (glued) with readily available
supplies, it does not require the thermal welding equipment necessary for HDPE
installation. Given the possible stresses the pipe will be exposed to, however, HDPE
is most common in current installations. An alternative to gluing or welding pipe
includes leaving some sections of pipe unconnected and having a segment of larger
9.2 Configuration, Construction and Materials 195
Fig. 9.7 Placement of crushed glass on top of HDPE liquids distribution pipe in horizontal trench
diameter pipe sheathed around two adjacent smaller diameter pipe ends to allow for
future expansion and contraction.
Liquids addition pipes are perforated to allow for liquids distribution into the
waste. At the flow rates commonly used, the size and spacing of perforations may
differ based on site conditions to prevent preferential discharge into certain areas of
196 9 Buried Horizontal Systems for Liquids Addition
Fig. 9.9 Pulling welded HDPE pipe with tractor to proximity of excavated trench
Fig. 9.10 Placement of shredded tires in liquids addition trench on top of PVC pipe
the trench. For example, engineers have employed specific, more complex patterns
of smaller and varying hole sizes along with different spacing to achieve uniform
flow distribution along the length of the pipe; the design procedures for this are
common in manifold design for liquid outfalls. Given the experience that the limita-
tion to liquids addition into the trench will be the waste itself, for larger flow rates,
9.2 Configuration, Construction and Materials 197
the trench will fill regardless of the perforation scheme, and thus most engineers opt
to provide more and larger perforations rather than incorporate a detailed manifold
distribution design. Only in cases where lower flow rates are added (in a manner
where liquid depths are not expected to build up) are more complicated designs war-
ranted. Typical orifice spacing is every 0.6–2 m (2–6 ft) with diameters of 0.5–
1.0 cm (0.25–0.375 in.). Pipes can be purchased pre-perforated, although some
operators choose to drill orifices with landfill personnel using standard drilling tools
(Fig. 9.11).
An important consideration in the construction of a liquids addition system is the
recording of trench and pipe locations. Surveying pipe locations as they are installed
is a recommended practice (Fig. 9.12). Many modern landfills are equipped with
equipment that allows ready measurement of vertical and horizontal coordinates,
and these devices can be used to routinely measure device location and elevation,
with the results incorporated into the site’s record drawings. Detailed record keep-
ing with regard to device location may be required as part of the facility’s operating
permit. Regardless of whether recording locational details is required, this informa-
tion is critical to evaluation of system performance and facilitates future construc-
tion activities and operation of other systems such as those for gas collection.
As described above, the placement of perforated pipes and permeable bedding
material into trenches excavated into the surface of the landfill is the most common
construction technique. Distribution pipes have been placed in trenches without
bedding material (Townsend and Miller 1998), and devices constructed in this fash-
ion can provide liquids addition capability (though not initial liquids storage); the
downside to this approach is potential damage to the pipe greatly limiting liquids
distribution because of the absence of permeable bedding for liquids to flow through
198 9 Buried Horizontal Systems for Liquids Addition
Fig. 9.13 Trenching machine used for installing horizontal injection pipe into the surface of a
landfill lift
Fig. 9.14 Perforated HDPE injection pipe installed using a trenching device
200 9 Buried Horizontal Systems for Liquids Addition
Fig. 9.15 Clay seepage collar construction at the end of a horizontal injection trench where pipe
perforations start
9.2.2 Blankets
Fig. 9.16 A horizontal drainage blanket of crushed glass installed on a landfill lift
Fig. 9.17 A horizontal drainage blanket of shredded tires installed on a landfill lift
Seeps can be problematic with pressurized horizontal liquids addition systems (see
Chap. 11), and a common route for leachate to channel to the landfill side slope is the
pathway created by the trench and the pipe. Some designers and operators thus opt
to connect multiple horizontal trenches and blankets together within the landfill,
which results in fewer penetrations to the side of the landfill. Horizontal systems also
have the potential to be utilized in conjunction with vertical systems. One example
of this approach would be the construction of horizontal trenches or blankets
throughout the progression of the landfill, but without the connection to exit lines
leaving the landfill. At a later time, vertical wells could be drilled into the landfill
with the purposeful intention of intercepting the buried horizontal devices (Fig. 9.18).
This would require careful surveying of device locations, especially trenches, so a
hydraulic connection can be made. Liquid would be added to the vertical entry
points on the surface of the landfill, but the liquids addition capacity would be much
larger than a typical vertical well.
The design process begins with determination of the target volume of liquids to be
added to the landfill and the determination of the rate at which that volume will be
added to the landfill. Chapter 6 provides information for completing these design
steps. Once the target liquids addition volume and flow rate have been determined for
the landfill as a whole, the design must include the individual horizontal devices and
their operating conditions, with ultimate integration into a design of multiple devices
comprising the complete system. The design of an individual horizontal device
includes specification of the trench configuration and materials, the length of perfo-
rated pipe, and the flow and/or pressure at which the device should be operated.
Similar to the design of a vertical system, horizontal system design entails locat-
ing a sufficient number of horizontal trenches (or blankets) throughout the landfill
and designing a delivery system to convey liquids to each of the liquids addition
devices. The design engineer should aim to efficiently distribute liquids throughout
the landfill by systematically locating trenches within a set of established boundaries.
These boundaries might include the anticipated saturated zone of adjacent trenches,
slopes, the landfill surface, and the landfill bottom.
To design within these constraints, two major design parameters must be identi-
fied for a given landfill: (i) the flow rate that can be added to a given horizontal
device and (ii) the shape of the saturated zone that results from adding that flow rate
to the horizontal device. Both the flow and dimensions of a saturated zone can be
predicted by the engineer for a given design configuration, landfill properties, and
operation conditions using a combination of historic performance from similar
facilities and predictive tools resulting from fluid flow modeling techniques; both
are discussed in the remainder of this chapter.
9.3 Design Methodology 203
Fig. 9.18 Illustration of combining vertical and horizontal liquids addition (a) The selected area is
backfilled with permeable media (b) Successive waste lifts have an area backfilled with permeable
media in a fashion similar to that shown in (a) (c) A vertical well is drilled through the horizontally-
constructed permeable media beds and the screened section intersects with each permeable layer
Fig. 9.19 Performance for a pressurized horizontal liquids addition device (a) linear flow rate and
pressure, (b) fluid conductance
Fig. 9.20 Comparing fluid conductance change with time for a pressurized horizontal liquids
addition (a) actual time (intermittent liquids addition), (b) cumulative operating time
The fluid flow modeling techniques discussed in Chap. 5 can be used to examine
the distribution of liquids into the waste surrounding a horizontal liquids addition
device as a function of operating conditions and landfill properties. Several authors
have presented examples of the use of such modeling for horizontal trenches and
blankets in landfills. In an effort to estimate the zone of influence of horizontal
trench, Townsend (1995) developed an equation describing flow through a horizon-
tal line source in a porous medium based on saturated and steady-state conditions.
McCreanor and Reinhart (2000) numerically simulated fluid flow from horizontal
injection trenches using SUTRA; the impacts of waste heterogeneity and anisotropy
were investigated, but operating conditions such as injection pressure, and flow rate
at the trench, which is an important operation variable, was not examined. Haydar
and Khire (2005) numerically modeled fluid flow from horizontal trenches using
HYDRUS-2D and examined the steady-state flow rate as a function of injection
pressures, trench geometry and size, hydraulic conductivity of the trench backfill,
and horizontal and vertical trench spacing. Jain et al. (2010a, b, 2013) modeled
206 9 Buried Horizontal Systems for Liquids Addition
Fig. 9.21 Pressurized horizontal liquids addition performance in response to changing operating
conditions (a) flow rate and pressure, (b) fluid conductance
liquids flow from horizontal trenches as a function of media properties (e.g., waste
hydraulic conductivity, porosity), trench dimensions, and operating pressure and
developed design charts to estimate both steady-state and transient flow rates and
lateral and vertical zone of impact; this approach will be presented in greater detail
in the following sections.
As an illustration of what the typical output results from a simulation of fluid
flow into a horizontal liquids addition devices, Fig. 9.22 and 9.23 present the output
of SEEP/W simulations for a horizontal trench and a horizontal blanket operated
continuously under constant pressure. The data presented in these figures (flow rate,
extent of lateral and vertical wetted front movement with time) illustrate typical
outcomes for such a simulation, and the magnitudes are only reflective of the spe-
cific scenario and conditions modeled. As liquids are added, the flow rate drops
notably in the first part of operation, followed by a relatively steady flow that
decreases slowly with time. The decrease in flow rate corresponds to the expanding
wetted zone around the device.
9.3 Design Methodology 207
Fig. 9.22 Example output of from a SEEP/W simulation of pressurized liquids addition into a
horizontal trench
208 9 Buried Horizontal Systems for Liquids Addition
Fig. 9.23 Example output of from a SEEP/W simulation of pressurized liquids addition into a
horizontal blanket
addition times, or operational characteristics. Miller and Emge (1997), for example,
reported the qualitative performance of horizontal injection trenches in distributing
liquids to MSW in a landfill. In a review of leachate recirculation rates for several
different landfills, Bareither et al. (2010) reported volumetric dosing rates to range
from 0.178 to 0.939 m3 per m of pipe (for dosing periods of less than 1 day).
Townsend and Miller (1998) evaluated the hydraulic performance of horizontal
injection trenches at a lined landfill in FL, US (Alachua County Southwest
Landfill; see Chap. 4). Leachate was recirculated into the waste mass using 7.6-
cm (3-in) diameter perforated horizontal injection lines installed in 1 m by 1 m
(3.3 ft by 3.3 ft) horizontal trenches and at three different depths during landfill
operation; shredded tires were used as a bedding media in most of the lines. The
trenches were approximately 33 m (100 ft) apart horizontally and 4.5 m (15 ft)
apart vertically. The lengths of the injection lines ranged from 100 to 220 m (330–
720 ft). Approximately 30,000 m3 (7.9 million gal) of leachate were recirculated
over a period of 19 months. All the injection lines were characterized in terms of
flow rates and associated leachate back-pressures. The maximum leachate recir-
culation rate per unit length of injection line was reported to be 3.0 × 10−3 m2/min
(0.22 gpm/ft). The fluid conductance ranged from 9.9 × 10−5 to 5.4 × 10−4 m/min
(0.00243–0.0113 gpm/ft2). Trenches without bedding initially had lower fluid
conductance values, but with time, these values approached those in the trenches
with shredded tires. Fluid conductance values were lower for those trenches bur-
ied deeper in the landfill.
Pressurized liquids addition into buried horizontal trenches were closely moni-
tored at the Polk County North Central Landfill (see Chap. 4). Leachate was recir-
culated into more than 100 trenches (approximately 1 m deep by 1 m wide) with
lengths up to 220 m (720 ft). Distribution pipes were constructed of 10-cm (4-in.)
diameter HDPE with 0.95-cm diameter perforations (0.375-in). Bedding materials
used included shredded automobile tires and broken glass. In several cases no bed-
ding material was used (excavated waste was placed back into the trench after pipe
installation). More than 100,000 m3 (25 million gal) of leachate were added to the
landfill. Larson (2007) and Kumar (2009) measured fluid conductance values,
which were found to decrease with the cumulative volume of liquids added. For
trenches where a cumulative 1.24 m3 of leachate per m of pipe (100 gal per ft) were
added, the average κ was 1.2 × 10−4 m/min (0.029 gpm/ft2), ranging from 2.1 × 10−4
to 5.2 × 10−3 m/min (0.005–0.13 gpm/ft2). For trenches where a cumulative 2.48 m3
of leachate per m of pipe (200 gal per ft) were added, the average κ was 4.9 × 10−4 m/
min (0.012 gpm/ft2), ranging from 9.8 × 10−5 to 2.0 × 10−3 m/min (0.002–0.05 gpm/
ft2). Measured κ values were similar for shredded tires and crushed glass. In the first
stages of liquids addition, κ values for trenches with bedding were greater than
those without bedding media, and were greater for trenches closer to the surface
compared to deeper locations in the landfill. As liquids addition proceeded, these
differences decreased.
Doran (1999) reported field experience with leachate recirculation using hori-
zontal injection lines at a landfill in Minnesota. Leachate was recirculated into a
5.2-ha (12.8-acre) landfill using a set of 11 injection lines installed in a 0.6 by 0.6 m
210 9 Buried Horizontal Systems for Liquids Addition
Similar to the design methods for vertical systems (presented in Chap. 8), the
designer can estimate the achievable flow into horizontal devices using a fluid flow-
modeling program. Example modeling output was presented previously in Figs. 9.22
and 9.23 for both a buried horizontal trench and a blanket, each simulated at con-
stant pressure. Again, similar to both surface systems and vertical wells, the flow
rate decreases with time as the wetting front expands into the landfill, ultimately
reaching a steady-state condition.
Several methods have been developed that provide a simplified approach to esti-
mate achievable flow into buried horizontal devices without the need for modeling.
Xu et al. (2014) conducted a series of SEEP/W simulations for a range of operating
conditions and developed best-fit relationships to predict achievable flow into hori-
zontal trenches. The following relationship for flow rate was developed:
Q = 1.82 × K z × P × A (9.1)
where Q = the flow rate per unit blanket length (L3 T−1 L−1); A = waste anisotropy
(Kx/Kz); and P = injection pressure (L). The ratio of Q/KZ represents the fluid con-
ductance (κ).
Jain et al. (2010b) developed an approach using dimensionless parameters and
design charts to estimate steady-state flow rate in a buried horizontal liquid addition
device. The approach was developed to be equally applicable to trench and blanket
systems. SEEP/W simulations were conducted over a range of conditions that would
reasonably be encountered at a landfill site. The parameters evaluated included the
depth and width of the trench or blanket, and the pressure within the device, which
could be a hydrostatic pressure either less than or greater than the thickness of the
trench. The simulation results were used to formulate dimensionless parameters and
design charts to allow for the determination of steady-state flow. The simulation results
were presented in a series of dimensionless charts to broaden the scope of application
for the results beyond the range of individual parameter values used for modeling.
The first step in determining steady-state flow rate is calculation of the dimen-
sionless variable η as follows:
w2
h= ×A (9.2)
l2
9.4 Flow Rates 211
Where w is the depth of the trench, l is the width of the trench, and A is the
anisotropy (KX/KZ). The variable η indicates the dominant flow direction, vertical or
horizontal. A low η value signifies a flow dominant in the vertical direction, whereas
a high η value indicates a flow primarily in the horizontal direction. A trench would
tend toward a greater η value compared to a blanket.
The designer then identifies a target injection pressure head (pI). Trenches are
frequently operated at liquid pressures that exceed the depth of the trench. For blan-
kets, depending on the thickness and area of the blanket, the liquid pressure may be
limited to a liquid depth less than the thickness of the blanket (in this case, the liquid
depth is treated as the depth of the device). Dividing the target injection pressure
head (in units of water column depth) by the depth of the trench, the dimensionless
injection pressure head, pID, is calculated as:
pI
pID = (9.3)
w
Now that the values of pId and η have been determined, the steady-state dimen-
sionless flow rate, qs, for a horizontal source can be estimated using the chart pre-
sented in Fig. 9.24.
Figure 9.24 is similar to Fig. 7.9 presented in the discussion of surface systems
(Chap. 7), but provides PID values greater than 1; this allows for the consideration of
the pressurized addition only possible in a buried system (not a surface source).
Once qs has been estimated, the steady-state flow rate (QS) into the horizontal
device can be calculated as
Qhs = q × l × K z (9.4)
where the terms are the same as previously defined.
Fig. 9.24 Design chart for estimating steady state flow (qs) into a horizontal source (trench or
blanket)
212 9 Buried Horizontal Systems for Liquids Addition
The ability to estimate the size and shape of the saturated zone surrounding a hori-
zontal trench can be a helpful tool for the designer in determining the appropriate
location and spacing of devices. Again, flow patterns are idealized and the designer
must factor this into the final design. Figure 9.20 presents a definition sketch of
critical parameters associated with the system modeled.
Townsend (1995) developed an equation to estimate the steady-state zone of
influence of a horizontal injection trench, as shown in (9.5)
Q æX Kz ö
X= tan -1 ç ÷ (9.5)
2p K z çZ Kx ÷
è ø
This can be expressed in a form such that Z can be solved directly and a saturated
zone profile can be easily plotted.
Kz
X
Kx
Z= (9.6)
æ 2 X p KZ ö
tan ç ÷
è Q ø
This relationship was derived on the assumption that the injection trench could
be treated as a line source and the surrounding media was homogenous. Based on
Townsend’s equation, the maximum upward movement (Zmax) and lateral spread of
moisture from the trench (Xtrench) are presented in (9.7) and (9.8), respectively.
Q
Z max = (9.7)
2p K x K z
Q
X trench = (9.8)
4Kz
Once the injected liquid reaches the maximum lateral distance, gravity and the
saturated zone will only expand in the vertical direction until it reaches the leachate
collection system (neglecting any channeling or preferential lateral flow paths).
Using Townsend’s Equation, the maximum lateral spread distance (Xtrench,max) can be
calculated, as shown in (9.9).
Q
X trench ,max = (9.9)
2Kz
Xu et al. (2013), as part of the work referenced in the previous section, developed
a series of equations based on SEEP/W modeling results to predict the saturated
9.6 Device Spacing 213
zone surrounding a horizontal liquids addition trench. With this approach, the
lateral spread of liquids from a horizontal trench at the maximum distance from the
trench (Xtrench, max) can be determined. Using the estimate for Q for a horizontal
trench (9.1), Xtrench,max can be solved as:
Q 1.82 P × K z × A
X trench ,max = = = 0.91P × A (9.10)
2Kz 2Kz
When Xu et al. (2014) simulated the zone of saturation surrounding a horizontal
trench over a range of typical landfill conditions and used this to develop an equa-
tion for the saturated zone, the results differed somewhat from the solution pre-
sented in (9.5) and (9.6). The lateral spread at the trench was found as:
Kz
X
Kx
Z= (9.12)
æ 2 X p KZ ö
tan ç ÷
è Q R ø
ì1.3 Z <0
ï æ X - X trench ö
R=í (9.13)
ï
1.3 - 0.3 ç ÷Z ³0
î è X max - X trench ø
With this equation, a modified form of the saturated zone equation presented
earlier can be calculated that better reflects the results of modeled porous media
flow simulations.
The engineer must specify the number of horizontal devices for installation and
their locations. The information presented in Sect. 9.4 allows the engineer to esti-
mate the flow rate that can be added to a given horizontal device. This, coupled with
214 9 Buried Horizontal Systems for Liquids Addition
the liquid addition targets discussed in Chap. 6, can be used to provide a preliminary
estimate of the number of devices.
Engineers often specify device spacing based on the distance needed to provide
adequate moisture distribution within the landfill. The methods in Sect. 9.5 allow
for the estimation of steady state zones of impact of a line source, and thus may be
useful for device spacing. However, the engineer should also factor operating time
into the design. The time needed to reach steady state may be large, and thus in
cases where more rapid coverage is desired, closer spacing may be necessary. Jain
et al. (2010b) provides a methodology for determining the time needed to reach
steady-state conditions.
Design charts developed in the Jain et al. (2010b) approach allow Xtrench to be
solved as a function of the steady state dimensionless flow rate, qs. The equations
developed by Townsend (1995) and Xu et al. (2014) corresponded to a line source,
whereas design chart developed by Jain et al. (2010a, b) can be used to estimate the
zone of impact of a horizontal source as a function of not only waste properties but
source (trench or blanket) dimensions as well. The following equation defines xIds,
which is equivalent to the ratio of Xtrench at steady-state (Xtrench,s) and the trench
width, l (see Fig. 9.25).
x trench ,s
xIds = (9.14)
l
Once the designer estimates dimensionless flow rate (qs) for selected source
dimensions, injection pressure, and waste properties, qs can be used to estimate XIds
using the design chart presented in Fig. 9.26.
Using (9.14), Xtrench occurring at steady state can be determined. As discussed
earlier, in order to reach steady state, a certain volume of liquids must be added to
the device to fully saturate the surrounding zone of impact. Figure 9.27 provides the
relationship between the dimensionless parameter, η, the dimension injection
Fig. 9.25 Definition sketch for major dimensions associated with estimate of pressurized liquids
addition into a buried horizontal device in a landfill
9.6 Device Spacing 215
pressure head (PId), and the critical number of pore volumes, Vn,critical, required to
reach steady-state for a single device.
To evaluate the suitability of the pore volume to meet the design and operation
objectives of the system, the total volume of liquids needed to achieve a fully satu-
rated zone at steady state should be calculated. This value, along with the necessary
time required to reach steady state, can be used to determine a suitable design
216 9 Buried Horizontal Systems for Liquids Addition
Fig. 9.28 Design chart for the fraction of maximum lateral extent (Xtrench/Xtrench,s) attained as a
function of Vt/Vt,critical
s pacing. The volume of liquids added to a single device required to form a fully
saturated zone at steady state, Vt,critical, can be determined as follows:
Vt ,critical
Vn,critical = (9.15)
lw (q n - qr )
where (θn − θr) is the drainable porosity of the waste.
The designer can estimate the magnitude of Xtrench at any volume added less than
steady state (Vt) using Fig. 9.28, which presents a design chart the ratio of Xtrench to
Xtrench,s as a function of Vt to Vt,critical. Figure 9.28 shows the range of modeling result
for Jain et al. (2013); this figure demonstrates that approximately 80–90 % of Xtrench,s
is reached by adding only 40 % of Vt,critical. Utilizing the ratio found using Fig. 9.28,
Xtrench can be estimated; an appropriate spacing for horizontal wells would be a
value twice that of Xtrench.
The designer may also need to predict the depth of the saturated zone beneath the
trench as part of evaluating appropriate device spacing (ZI; see Fig. 9.25). Figure 9.29
presents a design chart for estimating the magnitude of ZI that would occur when
Xtrench reaches steady state (ZI,S). Figure 9.29 plots the ZI,S as a function of the ratio of
Vn,critical to qs; determination of both of these values has previously been presented.
In a similar manner as described for Xtrench, the value of the H that occurs prior
to steady state conditions can be estimated. Figure 9.30 presents the XI/XI,S as a
function of Vt to Vt,critical. A line segregating the simulation results into two groups
9.6 Device Spacing 217
Fig. 9.30 Fraction of maximum vertical extent (zIs) attained as a function of fraction of volumes
of moisture needed (Vt, critical) to reach steady state
218 9 Buried Horizontal Systems for Liquids Addition
(η ≤ 10−4 and η > 10−4) is shown; data from simulations with η > 10−4 fell above the
line, whereas data from simulations with η ≤ 10−4 fell below the line. Approximately
40–55 % of the vertical zone of impact can be achieved with the addition of 40 %
of liquids volume needed to achieve the steady-state vertical extent of the zone
of impact.
Finally, the time to reach the steady state (tds) can be estimated with an estimate
of the ratio of Vn,critical to qs. The design chart (presented in Fig. 9.31) can also be used
for estimating the ratio of the average flow rate resulted from an given liquids addi-
tion pressures to flow rate resulting at steady state; this value would be equal to the
value of Vt/Vt,critical divided by the value of t/ts.
An operations plan for the horizontal liquids addition system should be prepared by
the design engineer and included as part of the site’s overall operations plan (see
Chap. 3). Specific operation details will include target injection lines, liquids addi-
tion rates, operation times, and operation pressure. The engineer will specify these
parameters based on the objectives of the system, site-specific constraints, and
using the design methodology outlined in the previous sections. At sites where land-
fill gas collection is employed, the designer must closely integrate construction and
operation considerations of horizontal recirculation systems with the phasing and
operation of gas collection systems (see Chap. 13).
References 219
References
Bareither CA, Benson CH, Barlaz MA, Edil TB, Tolaymat TM (2010) Performance of North
American bioreactor landfills I: leachate hydrology and waste settlement. J Environ Eng
136:824–838
Doran F (1999) Lay leachate lay. Waste Age 30(4):74–79
Haydar M, Khire M (2005) Leachate recirculation using horizontal trenches in bioreactor landfills.
J Geotech Geoenviron Eng 131(7):837–847
Jain P, Townsend T, Tolaymat T (2010a) Steady-state design of vertical wells for liquids addition
at bioreactor landfills. Waste Manag 30:2022–2029. doi:10.1016/j.wasman.2010.02.020
Jain P, Townsend T, Tolaymat T (2010b) Steady-state design of horizontal systems for liquids addi-
tion at bioreactor landfills. Waste Manag 30:2560–2569
Jain P, Townsend TG, Tolaymat T (2013) Transient design of bioreactor landfills liquids addition
system. Waste Manag 34(9):1667–1673
220 9 Buried Horizontal Systems for Liquids Addition
Khire M, Haydar M (2005) Leachate recirculation using geocomposite drainage layers in engi-
neered MSW landfill. In: Proceedings of Geo Frontier 2005, ASCE conference, Austin
Kumar S (2009) Study of pore water pressure impact and fluid conductance of a landfill horizontal
liquids injection system. Master of Engineering Thesis, University of Florida, Gainesville
Larson J (2007) Investigations at a bioreactor landfill to aid in the operation and design of horizon-
tal injection liquids addition systems. Master of Engineering Thesis, University of Florida,
Gainesville
McCreanor PT, Reinhart DR (2000) Mathematical modeling of leachate routing in a leachate recir-
culating landfill. Water Resour 34(4):1285–1295
Miller DE, Emge SM (1997) Enhancing landfill leachate recirculation system performance. Pract
Period Hazard Toxic Radioact Waste Manage 1(3):113–119
Townsend T (1995) Leachate recycle at solid waste landfills using horizontal injection. Ph.D. dis-
sertation, University of Florida, Gainesville
Townsend T, Miller W (1998) Leachate recycle using horizontal injection. Adv Environ Res
2(2):129–138
Townsend T, Miller W, Bishop R, Carter J (1994) Combining systems for leachate recirculation
and landfill gas collection. Solid Waste Technol 8(4):18–24
Xu Q, Townsend TG, Jain P (2014) Steady-state saturated zone equations for liquids addition
devices at landfills. Unpublished Manuscript
Chapter 10
Leachate Collection and Removal
Systems (LCRS)
Abstract The basic function and importance of leachate collection and removal
systems are first presented in Chap. 2, but Chap. 10 significantly expands this intro-
duction and provides a discussion of the major design and operation considerations
of LCRS in the context of sustainable landfilling. Leachate impingement on bottom
liner systems, techniques to predict leachate head on the liner as a result of added
liquids, settlement considerations, and clogging mechanisms and avoidance proce-
dures are presented. Each LCRS concept is presented with a particular focus on how
sustainable landfilling approaches can influence the designer’s and operator’s
approaches to prevent excessive build-up of liquids on the bottom liner system,
which is one of the most critical pieces of infrastructure that protects the environ-
ment from potential impacts of landfills.
The function and general configuration of the leachate collection and removal
system (LCRS) were described in Chap. 2. For landfills with elevated moisture
content, either as a result of purposeful liquids addition, site stormwater manage-
ment practices, or incoming waste properties, the importance of the LCRS for a
lined landfill cannot be overstated; this chapter is thus devoted to this landfill com-
ponent. The LCRS will in most regulatory jurisdictions be required to maintain
leachate depth on the liner to less than a specified threshold (in the US, this depth is
less than 0.3 m). Moreover, a properly functioning LCRS is critical to limit leachate
discharges from seepage at the landfill base (refer to Chap. 11) and to minimize
slope stability problems (refer to Chap. 12). Therefore, the successful removal of
liquids from a LCRS is an important component of all landfill designs, especially at
facilities practicing sustainable operation by accelerating waste stabilization.
The components of a LCRS include a liner system graded (sloped) to promote
gravity drainage, drainage media to route the liquids rapidly off of the liner to tar-
geted conveyance points, drains consisting of perforated pipes, and pumping sys-
tems to remove leachate from the landfill. In essence, the LCRS is a high-permeability
drainage layer placed between the low-permeability liner system and the waste.
Other components of the LCRS include its accompanying pipes and sumps and a
liquids removal system. Leachate removal is accomplished by gravity drainage
from sheet flow over the sloped liner, as well as from rock drains and perforated
pipes that intercept the sheet flow at intervals necessary to minimize the depth of
ponded leachate on the liner. The rock drains and pipes route the leachate to a low
point (i.e., sump) from which the leachate is periodically pumped from the landfill
or allowed to gravity drain to a collection point outside the landfill. In some cases,
the pipes penetrate the liner system and discharge leachate to an external pump sta-
tion (Fig. 10.1a). Alternatively, pumps can be installed within the landfill unit
(Fig. 10.1b).
Fig. 10.1 Illustration of two methods for pumping leachate from the landfill (a) internal pump
station (b) internal pump station
10.1 Leachate Removal Fundamentals 223
Fig. 10.2 Illustration of two typical leachate collection and removal systems (a) planar, (b) saw
tooth
Third, the potential for clogging the LCRS must be considered. If the LCRS clogs,
the hydraulic conductivity of the drainage material decreases and the drainage
performance of the LCRS may be reduced. Greater volumes of leachate passing
through the LCRS may result in additional clogging concerns.
This chapter provides a fundamental overview of LCRS design and operation
issues that should be assessed as part of the planning and implementation for any
landfill, especially when liquids addition is practiced and at landfills with inherently
high moisture levels. Readers are encouraged to consult additional references for
more specific landfill design methodologies that pertain to LCRS design (McBean
et al. 1995; Qian et al. 2002).
Leachate impingement is the rate that leachate percolates from the base of the
landfilled waste into the LCRS and is expressed in units of flow rate per unit area
(L3/T per L2). Where liquids addition is practiced, the impingement rate is expected
to be greater compared to a traditional landfill. Although added liquid will at first
remain in the waste, once field capacity is reached, some fraction of the added
moisture will migrate through the waste and intercept the LCRS. The concept of
field capacity and how it is typically used in estimating moisture addition volumes
was discussed in Chap. 6.
The engineer must make an appropriate assumption for impingement rate as part
of the LCRS design process. Approaches for determining impingement rate include
(i) using leachate flow data collected from similar landfills already in operation, (ii)
conducting a landfill water balance, and (iii) using conservative estimates of
hydraulic conductivity of the waste and factoring possible impacts of any liquid
addition, as appropriate. While using existing leachate flow data from other sites to
estimate impingement (by normalizing leachate flow to the contributing landfill
area) provides a valuable comparison, possible differences in site features and oper-
ation in most cases still necessitates an independent estimate of impingement for
the design proposed.
In Chap. 5, methods for predicting moisture flow in landfills and performing
landfill water balances that forecast leachate generation were reviewed. Several
approaches of differing complexity may thus be employed to estimate impinge-
ment; however, in some locations (e.g., US), regulatory agencies require use of a
standard methodology, the HELP model. A description of HELP was provided in
Chap. 5, and the following section summarizes issues of importance when applying
HELP to predict impingement at landfills where liquids addition is practiced.
10.2 Predicting Leachate Impingement 225
As presented in Chap. 5, the HELP model is a widely used tool for performing a
water balance on landfills and it includes features that allow for simulation of liq-
uids addition (Schroeder et al. 1994). Most engineers utilize HELP output results to
determine (a) the maximum head on the liner and (b) the leachate generation rate
necessary for sizing the various components of the leachate removal and manage-
ment system. Inherent in these data is the impingement rate, which HELP provides
as an output both in LT−1 units and L3T−1 units (the L3 T−1 units correspond to the
input area designated in the HELP simulation). The maximum impingement rate is
the lowest saturated hydraulic conductivity of the profile layers above the liner, but
is not greater than maximum daily infiltration rate.
Xu et al. (2012) presented a discussion of the different techniques that can be
used as part of HELP simulations to account for liquids addition in landfills and to
predict the resulting impingement rate, and these techniques are summarized herein.
The HELP model provides several options for incorporating additional liquids
beyond rainfall, including utilizing the HELP model’s leachate recirculation fea-
ture, the model’s subsurface inflow feature, and a technique referred to as the rain-
fall modification method, where precipitation inputs are manipulated to approximate
liquids addition.
The leachate recirculation feature (LRF) in HELP’s soil and design input screen
allows the user to define a percentage of the leachate generation collected in the
LCRS to be added to a designated landfill layer and is conceptually illustrated in
Fig. 10.3. The LRF method is widely applied when simulating a landfill where the
primary motivation is to manage leachate through recirculation, but not to bring the
landfill to optimal conditions for waste stabilization. The LRF method does not
allow the designer to simulate the addition of a specific volume of liquid to a waste
layer and can only simulate the recirculation of a defined percentage of leachate
Fig. 10.3 Schematic of liquids addition methods used in the HELP leachate recirculation method
226 10 Leachate Collection and Removal Systems (LCRS)
Fig. 10.4 Schematics of liquids addition methods used in the HELP subsurface inflow method
collected from the LCRS. For modeling the scenario where a specific liquid volume
will be added (e.g., to reach field capacity), the ability to select the actual volume of
liquid injected into a layer is necessary.
The subsurface inflow (SSI) feature of HELP simulates lateral groundwater flow
into a defined layer in the simulated landfill (Schroeder et al. 1994). Contrary to the
leachate recirculation feature which allows liquid to be added to just one layer, the
SSI feature permits the addition of a specified volume of liquid to any number of
layers (conceptually illustrated in Fig. 10.4). With this feature, the liquids addition
rate is constant throughout a simulation and liquids are added continuously. Since
liquids addition systems at operating landfills are frequently operated intermittently,
and because flow rates can vary dramatically from day to day, the SSI feature in
HELP (which is based on an average liquids addition rate) might underestimate the
maximum impingement rate.
The third method, referred herein as the rainfall modification (RFM) method,
uses the weather data files in the weather data input screen to simulate liquids addi-
tion by adjusting the evapotranspiration and runoff inputs, and by modifying the
precipitation data files to account for the liquids added to the landfill. It allows for a
defined volume of liquid to be added and distributed daily throughout the year into
any layer, as conceptually illustrated in Fig. 10.5. The RFM method is the only
HELP liquids addition method that can simulate specific changes in the liquids
addition regime throughout the year. To model leachate recirculation using the RFM
approach, layers of a landfill are divided into several groups, depending on the
leachate addition scheme. The surface layers above where liquids are added are
simulated in HELP as normal. The leachate impingement rate migrating from the
surface layer is obtained from the output file and added to the leachate volume for
recirculation, and these values serve as the input rainfall file for the next layer. The
rainfall data and evapotranspiration climate inputs must be modified for underlying
10.2 Predicting Leachate Impingement 227
Fig. 10.5 Schematics of liquids addition method used in the HELP rainfall modification method
waste layers, which are not affected by external weather conditions. Since the RFM
allows for daily control of the added liquid volume, any combination of liquids
addition can be simulated. The RFM can simulate temporal and limited spatial vari-
ation of liquids addition, which allows for a more realistic assessment of potential
impingement changes as a result of liquids addition system operation.
Xu et al. (2012) also provided several observations regarding other aspects of
HELP the designer should consider for landfills where liquids addition is practiced.
Selecting an appropriate model simulation time is important. An insufficient model-
ing period may not capture the leachate entry into the LCRS. Even if the design
objective is for the majority of the added liquid to remain within the landfill as
stored moisture (i.e., utilizing absorption capacity), the LCRS must be designed
with sufficient capacity to handle the increased impingement rate occurring under
fully wetted conditions.
The designer should also closely evaluate the appropriate input waste character-
istics used in the model input. The HELP default for hydraulic conductivity of the
waste, for example, may be too large. An early HELP default for hydraulic conduc-
tivity was 1 × 10−3 cm/s, and the current default is 2 × 10−4 cm/s for compacted MSW
(Schroeder et al. 1994). As described in Chap. 5, this value is likely substantially
greater than true conditions for well-compacted MSW at modern landfills. Also as
indicated in Chap. 5, hydraulic conductivity will decrease with depth in the landfill.
The designer should carefully consider appropriate selection of hydraulic conduc-
tivity, as well as moisture content, field capacity, and porosity, when setting up the
model run. HELP allows the designer to assign different waste characteristics to
distinct layers within the landfill, which permits simulation of changing waste prop-
erties deeper in the landfill.
228 10 Leachate Collection and Removal Systems (LCRS)
The maximum impingement rate predicted by HELP corresponds to the lowest sat-
urated hydraulic conductivity of the waste or soil layer above the LCRS. In other
words, if liquids are added at a rate greater than the hydraulic conductivity of the
waste, the impingement rate will be the same as the hydraulic conductivity and the
excess moisture will be stored within the waste. Thus, an alternative and more con-
servative approach to determine impingement in HELP includes assigning the verti-
cal hydraulic conductivity of the waste as the impingement rate.
As outlined in Chaps. 7, 8, and 9, liquids may be added to distinct landfill areas
at different times. While the impingement rate obtained from the HELP model is
evenly distributed over the LCRS, the impinging liquids may vary spatially at land-
fills where liquids addition is practiced, especially when liquids are added under
pressure. This is conceptually illustrated in Fig. 10.6, where the saturated zones that
could result from a horizontal trench (Fig. 10.6a) and a vertical well (Fig. 10.6b) are
shown along with an indication of enhanced impingement below the liquid addition
devices (relative to that which would be predicted using HELP by treating the
impingement rate as equal to the hydraulic conductivity). For some designs it might
be important to consider how LCRS performance could be affected by pressurized
liquids addition devices located near the base of the landfill.
Fig. 10.6 Conceptual illustration of potential differences in impingement rates as predicted using
HELP and actual distribution for (a) horizontal trench systems and (b) vertical well systems
10.2 Predicting Leachate Impingement 229
Q
e= (10.1)
Ai
As has been described in earlier chapters, the wetted zone resulting from a liq-
uids addition device increases in dimension as the zone moves vertically downward
from the device and ultimately approaches steady state. At distances far from the
device which are close to steady state, the impingement is approximately equal to
the vertical hydraulic conductivity (e = Kz), but closer to the device, the gradient is
greater than 1 and the impingement will be greater than the vertical hydraulic con-
ductivity (e > Kz).
The saturated zone equations (presented in Chaps. 8 and 9) are used to solve for
impingement rate occurring where the saturated zone under a liquids addition
devices intercepted the LCRS (Xu et al. 2014). The impingement rate (e) for a hori-
zontal trench located a distance Z above the LCRS can be approximated as
follows.
pK z
e= (10.2)
æx
-1 Kz ö
tan ç ÷
çz Kx ÷
è ø
Kz
e= (10.3)
( )
2
-1.6 D
1- e L
Where L is the length of the well screen and D is the distance from the top of the
well screen to the top of the LCRS (see Chap. 8).
230 10 Leachate Collection and Removal Systems (LCRS)
The engineer designs the LCRS to prevent the maximum liquid level (leachate
head) ponded on top of a liner from exceeding a design or regulatory threshold. The
design requirement for both MSW and hazardous waste landfills in the US is 1 ft
(0.3 m) or less of head on the liner. Several design methodologies have been devel-
oped to predict leachate head over an impermeable sloped drainage path of an
LCRS. The following sections describe methods for single layer (granular and
geonet) and multi-layer systems.
Several equations have been developed to solve for the maximum depth of leachate
above a sloped liner overlain by a granular drainage media with a drain at the down-
stream end. The maximum depth is calculated by first assuming the maximum
leachate inflow rate occurs under steady-state conditions while the LCRS is operat-
ing properly (e.g., leachate flows freely through the drains and does not back up
onto the primary drainage slopes). The methods for calculating head on the liner
incorporate the following factors: (i) drainage path length (L); (ii) drainage path
slope (L L−1 or degrees); (iii) leachate impingement rate (L T−1); and (iv) hydraulic
conductivity of drainage material (L T−1). These factors are schematically illustrated
in Fig. 10.7, which provides a definition sketch to supplement the design equations
Fig. 10.7 Definition sketch of the four main factors affecting head on a liner
10.3 Predicting Leachate Head on Liner 231
Table 10.1 Summary of analytical equations from Moore, McEnroe, and Giroud to predict the
leachate head on a single liner
Method Analytical equation
Moore (1980) e é k tan 2 a k tana eù
hmax = L ê + 1 - tan 2 a + ú
k êë e e k úû
McEnroe (1993) e
R=
k ×sin 2 a
S = tan a
A= (1 - 4 R )
B= ( 4 R - 1)
1
1
R< 1
é (1 - A - 2 R ) (1 + A - 2 RS ) ù 2 A
4 hmax = L ( S ) ( R - RS + R S2
)
2 2
ê ú
ëê (1 + A - 2 R ) (1 - A - 2 RS ) úû
1 é R (1 - 2 RS ) ù é 2 R ( S - 1) ù
R= hmax = L [ S ] ê ú exp ê ú
4 ë 1 - 2 R û ê
ë (1 - 2 RS ) (1 - 2 R ) úû
1 1
é1 æ 2 RS - 1 ö 1 -1 æ 2 R - 1 ö ù
R> hmax = L ( S ) ( R - RS + R 2 S 2 ) 2 exp ê tan -1 ç ÷ - tan ç ÷ú
4 ëB è B ø B è B øû
Giroud et al. e
(2000) l=
k × tan 2 a
ì é æ 2
ü
æ 8l ö ö ù ï
5/8
ï
j = 1 - 0.12 exp í- ê log ç ç ÷ ÷ ú ý
ç ÷
ïî êë è è 5 ø ø úû ïþ
æ 1 + 4l - 1 tana ö
h max = L ç
ç
´ ÷´ j
è 2 cosa ÷ø
Fig. 10.8 Comparison of maximum head on the liner calculated using the Moore, McEnroe and
Giroud equations
In some LCRS designs, synthetic drainage products (i.e., geonets) are used to
provide leachate conveyance. In these designs, a geonet is placed directly on the
geomembrane liner and is overlain by a layer of soil or other granular media. The
design will include a geotextile to separate the geonet from the overlying soil (or a
geocomposite consisting of a geonet and geotextile bonded together). Because the
thickness of a geonet is small (<0.01 m), when a designer specifies a geonet with
sufficient capacity to handle all of the leachate flow for the required impingement
rate, the depth of leachate will be well below the regulatory requirement.
The flow capacity of a geonet is most often described by its transmissivity, T
(L2/T). The engineer will specify a geonet that provides necessary capacity for the
predicted leachate generation or impingement rate. In the case where the geonet
must provide sufficient capacity to handle all of the leachate flow within its thick-
ness, required transmissivity (TREQ) will be solved as follows:
Q el
TREQ = = (10.4)
iw i
Where Q is the total leachate flow rate collected over an area (width (w) by
drainage path (L)), e is the impingement rate, and i is the slope of the drainage path.
The designer specifies a geonet that delivers needed transmissivity at the anticipated
design load and after applying a series of safety factors (to account for potential
10.3 Predicting Leachate Head on Liner 233
h eL (10.5)
max =
K tan (a )
where,
K =T (10.6)
t
Since the maximum allowable leachate head on the primary liner will be greater
than the thickness of the geonet, some LCRS designs will additionally utilize the
flow capacity provided by the granular media overlying the geonet. This demands
an alternative technique for estimating maximum head on the liner, which has been
provided by Giroud et al. (2004). This technique is illustrated in Fig. 10.9.
This method requires assumptions regarding the hydraulic conductivity of both
drainage layers. Hydraulic conductivity (K) of the geonet can be determined from
the transmissivity (T) and thickness (t) as described in the previous section. The first
step in the method is to calculate the length of geonet that handles all of the leachate
flow (Lu).
t1 K k1 sin (a )
Lu = (10.7)
e
If Lu is greater than the total length of the drainage path (L), the geonet can
handle the complete flow, and similar to the previous case, the maximum head on
the liner will be:
e
hmax = L (10.8)
k1 tan (a )
In the case where Lu < L, leachate extends above the geonet in the granular drain-
age layer, such that:
é
hmax = t1 ( cos a ) + j2 ê
( )
1 + 4l2 - 1 tan a ù é
ú ´ ê L - Tk ,1 ( sin a ) ú
ù
(10.9)
ê 2 ú ë e û
ë û
Where,
e
l= (10.10)
k2 tan 2 (a )
ì é æ 8l ö ù ü
5 /8 2
ï ê è 5 ø ú ï
ç ÷
j2 = 1 - 0.12 exp í- log10 ý (10.11)
ï êë ú ï
û
î þ
Where the terms represent those previously defined. This equation allows the
engineer to predict the combined depth of liquid above the liner in both the geonet
and the drainage media above it.
The engineer designs the LCRS as a combination of sloped liner areas and sloped
pipes or drains to route leachate to designated removal points. As these slopes are
often not great in magnitude (a few percent or less), changes to bottom liner eleva-
tion as a result of differential foundation settlement must be considered. If the soils
beneath the landfill settle in a manner that causes the liner base grade to change, the
performance of a gravity drainage system can be compromised. It is thus standard
10.4 Foundation Settlement Considerations 235
Landfill cell
Soil 1
Landfill cell
Resulting slopes
Soil 2
Fig. 10.10 Conceptual schematic of foundation settlement due to waste load (a) shortly after
placement of the landfill (b) long-term deformation showing the conceptual change in the slope of
the LCRS from the settlement of soils 1 and 2
practice for the design engineer to predict the landfill foundation settlement that is
expected to occur and to develop a grading plan for the liner foundation and pipes
that accommodates long-term foundation settlement. Since landfills where liquids
addition is practiced may be subject to greater differential settlement as a result of
greater loads (due to wetter waste), and because LCRS performance is especially
important at these facilities, incorporation of predicted settlement into a landfill’s
bottom grading design is essential.
The greatest magnitude of foundation settlement will occur toward the interior of
the landfill, where the greatest waste depths and resulting overburden pressures are
placed on the underlying soils. Figure 10.10 conceptually illustrates the process of
landfill foundation settlement. Prior to construction of the landfill, soil layers exist
in an assumed steady-state condition with respect to applied pressure from overly-
ing materials. Upon construction of the landfill unit, stresses are imparted on the
underlying soil layers, which in turn can cause settlement. In Fig. 10.10, a soil layer
in the landfill’s foundation settles non-uniformly in response to the differentially
applied load, and this causes the LCRS to slope inward unless appropriately antici-
pated and accounted for in the design.
236 10 Leachate Collection and Removal Systems (LCRS)
The degree of settlement will depend on the location and properties of soil layers
in the landfill foundation, the configuration (i.e., slope and height) of the landfill,
and the loads produced from the landfill as a result of the weight of the landfilled
materials. The geotechnical design of a landfill foundation is a much more detailed
procedure than presented here and a complete review is outside the scope of this
book; geotechnical engineering references should be consulted (Holtz and Kovacs
1981; Bowles 2001; Coduto 2001; Das 2010). The purpose of this section is to offer
the landfill design engineer a basic overview of the techniques used as part of land-
fill foundation analysis and to emphasize the importance of this design consider-
ation for wet landfills.
The design of a landfill grading plan that accounts for foundation settlement
begins with a detailed geotechnical characterization of the subsurface geology of
the site. This will include conducting soil borings and in-situ tests to characterize
subsurface soils (e.g., standard penetration test (SPT), cone penetration test (CPT))
and to retrieve samples for testing. The engineer will use the field measurement data
and the results provided from samples tested in the laboratory to estimate the amount
of subsurface settlement likely to occur upon completion of the waste fill. A variety
of methods have been developed to examine the settlement of large foundations
(Bowles 2001; Coduto 2001; Das 2010), although most techniques are focused on
building foundations. The engineer must often rely on basic soil deformation prin-
ciples; Table 10.2 provides two fundamental equations to predict the settlement of a
soil layer, one for immediate settlement based on a linear stress–strain relationship,
and one based on classic soil consolidation theory.
Immediate settlement occurs rapidly as a result of a direct strain response to a
stress, and is based on the assumption that the soil deforms in response to a constant
stress–strain modulus over the stress range of interest. The engineer would apply
this method to layers of unsaturated soils or those in the saturated zone not expected
to undergo consolidation. The stress–strain modulus can be estimated from in-situ
Consolidation settlement H0 s*
Ds = Cc log v*2
1 + e0 s v1
Ds v = q = g H (10.12)
Where γ is the specific weight of the landfilled material and H is the thickness of
the landfilled material at that point.
In the second approach, the spatial pattern of waste mass distribution is also
considered. In this case, added loads from the landfill are assumed to be distributed
into the underlying foundation soils. Geotechnical engineers have derived mathe-
matical relationships for stress distribution within an idealized soil, such that the
added vertical stress in the soil under an applied load (ΔσV) is some fraction of the
applied load (qo), which can be solved as:
s z = q0 I (10.13)
Where I is an influence value representing the fraction of qo occurring at a point
of interest. This influence value can be determined mathematically or using influ-
ence charts (Holtz and Kovacs 1981), or can be predicted using numerical solutions
in commercially-available software packages.
Once settlement has been estimated for distinct locations along the landfill foun-
dation, the differential settlement between points can be determined. The engineer
will select points that correspond to critical drainage pathways corresponding to the
designed LCRS. When the calculated settlement causes grade reversal in the
designed LCRS, the design of the LCRS must be revised to accommodate the maxi-
mum expected settlement so that the necessary slopes for gravity drainage will still
be maintained.
238 10 Leachate Collection and Removal Systems (LCRS)
Clogging issues have been suggested to be a greater concern for facilities operated
to enhance biological waste stabilization for the following reasons: (a) the larger
amounts of leachate passing through the LCRS during the landfill’s life; (b) the
potential differences in biological activity that might occur during sustainable land-
fill operations; and (c) the critical role that a successfully operating LCRS plays in
terms of slope stability at bioreactor landfills (Chap. 12). The degree to which clog-
ging occurrs has been demonstrated to be greatest under higher mass loading rates
to the LCRS (Fleming et al. 1999; Rowe et al. 2000). While landfills subjected to
liquids addition result in greater leachate flow, the concentrations of biodegradable
organic matter (BOD) are expected to become lower than conventional landfills
more rapidly.
240 10 Leachate Collection and Removal Systems (LCRS)
The typical method through which an engineer accounts for clogging in the LCRS
design process is to apply a reduction factor to the hydraulic conductivity (granular
media) or transmissivity (geonets) used as part of the head-on-liner design equa-
tions presented earlier in this chapter. Through this approach, the LCRS is designed
to maintain less than the target leachate depth even when clogging occurs. Detailed
design procedures associated with incorporating reduction factors for the LCRS are
provided in Qian et al. (2002).
Other design practices can also be implemented to reduce chances of clogging
problems and provide the operator better ability to address problems if they do
occur. Sufficient cleanout lines should be provided in the design so that all pipes can
be readily inspected and cleaned; pipe lengths should be limited to that which can
be accessed with available equipment. When geotextiles are used as part of the
design, configurations should be avoided where large of amounts of leachate flow
are routed through small geotextile areas. Where feasible, the operations plan
should discourage the occurrence of fine particulate material in waste or cover soil
in proximity to the LCRS. An added redundancy is the placement of a geocompos-
ite layer (geonet bonded to a geotextile) directly above the liner to facilitate more
rapid drainage.
As discussed in Chap. 16, leachate flow and liquid level monitoring should be a
routine practice at landfills where the additions liquids or wet wastes is practiced.
Changes in leachate production over time may provide some indication of less effi-
cient collection due to LCRS clogging. Interpreting this change can be complicated,
however, as several factors affect leachate production (e.g., climate, stormwater
References 241
management practices). More telling will be changes in the locations where leachate
collection occurs and the presence of ponded liquids or seeps in places where these
conditions formerly were not present.
A more direct assessment of LCRS performance is monitoring of LCRS gravity
drainage pipes. Many landfill permits require periodic inspection and cleaning of
LCRS pipes, hence the necessity to install cleanout lines as a means to provide
internal pipe access. This includes inspection with cameras and jetting with high
pressure water and specialized cleaning devices. At sites where calcium carbonate
clogging is evident and cannot be removed with high-pressure jetting, cleaning with
acid solution has been used with some success.
The inspection and cleaning of pipes does not, however, provide direct evidence
of clogging in other LCRS locations such as granular drains, geonets, geotextiles,
or the sand drainage blanket. Monitoring leachate depths on the liner (see Chap.
16), coupled with flow measurements, may provide an indirect measure of internal
LCRS clogging, but access to the LCRS for remediation is difficult because of the
large depth of waste that would need to be removed. In cases where clogging
appears to be a problem, but remediation is not feasible, other actions that the
operator may need to implement include reducing leachate generation in that area
(e.g., early closure), cessation of liquids addition, and increased leachate or ground-
water monitoring.
References
Maliva RG, Missimer TM, Leo KC, Statom RA, Dupraz C, Lynn M, Dickson JAD (2000) Unusual
calcite stromatolites and pisoids from a landfill LCRS. Geology 28(10):931–934
McBean E, Rovers F, Farquhar G (1995) Solid waste landfill engineering and design. Prentice Hall
PTR, Englewood Cliffs
McEnroe BM (1993) Maximum saturated depth over landfill liner. J Environ Eng 119(2):
262–270
Moore C (1980) Landfill and surface impoundment performance evaluations. EPA/530/SW-869.
USEPA, Washington, DC, p 63
Qian X, Koerner R, Gray D (2002) Geotechnical aspects of landfill design and construction.
Prentice Hall, Upper Saddle River
Rittman BE, Fleming I, Rowe RK (1996) Leachate chemistry: it’s implications for clogging. North
American Water and Environment Congress’96, Anaheim
Rohde JR, Gribb MM (1990) Biological and particulate clogging of geotextile/soil filter systems.
In: Koerner RM (ed) Geosynthetic testing for waste containment applications. American
Society for Testing and Materials, Philadelphia
Rowe RK, Fleming IR (1998) Estimating the time for clogging of LCRSs. In: Proceedings of the
third international congress on environmental geotechnics, Lisbon, September, vol 1, pp 23–28
Rowe RK, Quigley RM, Booker JR (1995) Clayey barrier systems for waste disposal facilities. E
& FN Spon, London
Rowe RK, Armstrong MD, Cullimore DR (2000) Mass loading and the rate of clogging due to
municipal solid waste leaching. Can Geotech J 27(2):355–370
Schroeder PR, Dozier TS, Zappi PA, McEnroe BM, Sjostrom JW, Peyton RL (1994) The hydro-
logic evaluation of landfill performance (HELP) model: engineering documentation for version
3. EPA/600/R-94/168b. US Environmental Protection Agency, Cincinnati
VanGulck JF, Rowe RK, Rittmann BE, Cooke AJ (2003) Predicting biogeochemical calcium pre-
cipitation in landfill LCRSs. Biodegradation 14:331–346
Xu Q, Kim H, Jain P, Townsend TG (2012) Hydrological evaluation of landfill performance
(HELP) modeling in bioreactor landfill design and permitting. J Mater Cycles Waste Manage
14:38–46
Xu Q, Jain P, Smith J, Tolaymat T, Townsend T (2014) Steady-state saturated zone equations for
liquids addition devices at landfills. Unpublished manuscript
Chapter 11
Leachate Control, Storage, and Treatment
Integrated liner systems and LCRS are fundamental design features of modern engi-
neered landfills, and are installed with a primary objective of minimizing deleteri-
ous impacts of leachate on the environment. Leachate management considerations
do not end with the construction of these features, however. Leachate removed from
the landfill must be handled and disposed properly. Leachate migration from the
landfill via other possible routes, namely the exposed surfaces of the landfill, must
be guarded against. While the proper management of leachate is an important ele-
ment at all landfills, it is even more critical at facilities where liquids are added or
recirculated as part of implementing sustainable landfill technologies.
Leachate forms predominantly as a result of rainfall and the manner that storm-
water is controlled plays a major role in the overall water balance of a landfill.
At sites where the moisture content is elevated, such as those where liquids addition
is practiced, contamination of stormwater as a result of exposure to leachate out-
breaks or seeps on the landfill surfaces is more likely, and thus such outcomes must
be planned for as part of both design and operation. Occurrence of leachate seeps at
wet landfills such as bioreactors is relatively common, meriting a discussion of
causes, remedial steps, and preventative measures as part of this chapter.
Although some landfill sites might be equipped with a force main that allows direct
discharge of leachate to an off-site treatment facility, leachate that is pumped or grav-
ity-drained from the LCRS is normally stored on site in a tank, pond, or similar device,
whether as a part of, or prior to, leachate treatment. The degree to which leachate is
treated on site depends greatly on facility-specific constraints and objectives, and
ranges from simple storage with only minimal changes in leachate chemical concen-
trations to comprehensive treatment plants designed to produce an effluent of sufficient
quality for on-site discharge. As leachate recirculation represents one of the practices
implemented to promote rapid waste stabilization, the need for external leachate treat-
ment might differ for these sites in comparison to traditional landfills. Alternatively,
specific treatment objectives that compliment leachate recirculation might be a desired
approach. The later part of this chapter summarizes standard practices for leachate
storage and treatment, and focuses on issues related to leachate management at sites
implementing technologies such as liquids addition and leachate recirculation.
Chapter 6 outlined a general approach for assessing moisture needs for landfill
operators attempting to maximize waste stabilization through liquids addition.
Because of the relatively large target volume of liquids that may be required to
achieve operational goals, additional sources of moisture beyond that resulting from
leachate generation may be required. The amount of leachate produced at a landfill
site is strongly dictated by site stormwater management practices and the nature of
the incoming waste material. Therefore, an important topic of discussion with
respect to leachate is the management of stormwater run-on and runoff.
Rainfall intercepted by the landfill surface, typically referred to as stormwater, is
normally considered leachate when it comes into contact with solid waste or other
leachate. While much of the production of landfill leachate results from moisture
infiltrating through the surface of the landfill into the waste, the flow of water over
the surface of the landfill as a result of rainfall runoff can also add to leachate vol-
umes. The majority of excess leachate observed at a landfill immediately after a
large storm event is, in fact, produced as a result of stormwater that flows directly to
the LCRS at the interface of the compacted waste and the liner system, or other
direct channels to the LCRS (i.e., highly permeable cover layers). Run-on control
describes the steps an operator takes to route non-impacted stormwater intercepted
by the soil or vegetation on the landfill surface away from areas of exposed waste,
thus preventing leachate formation. Similarly, runoff control describes the process
of routing clean stormwater to a designated discharge location and stormwater
impacted by leachate to an appropriate collection area.
The area of a landfill where waste is exposed to the environment after deposition
and compaction and prior to the placement of daily cover (or cover with additional
waste)—typically referred to as the working or active face—presents a source for
potential leachate generation. The working face should be maintained in a manner
to divert clean stormwater from covered areas away from the exposed waste to as
11.3 Managing Leachate Seeps 245
Seeps, which are also referred to as weeps, springs, or breakouts, result when leach-
ate migrates laterally to the side slope of the landfill instead of downward to the
LCRS. Seeps may be observed at all exposed landfill surfaces, but are most com-
monly observed on side slopes or at the base (toe) of the landfill. The seep may be
evident because of a discoloration, malodor, or the presence of insects (see
Figs. 11.1, 11.2, and 11.3). A primary concern with seeps is the potential for leach-
ate migration beyond the landfill’s footprint. Accordingly, seeps pose problems for
Fig. 11.3 Leachate seep flowing at base of landfill slope next to access road
landfill operators because they can contaminate stormwater and lead to prohibited
discharges as well as cause odors, attract insects, pose slope stability concerns, and
interfere with operations. Given the importance of addressing seeps at landfills with
high moisture levels and those practicing liquids addition, this section focuses on
seep causes, prevention, and mitigation issues. The information here follows that
presented by Xu et al. (2013).
11.3 Managing Leachate Seeps 247
Fig. 11.6 Schematic of leachate seepage at landfills caused by decreasing hydraulic conductivity
with depth
with depth in landfills coupled with compacted waste’s anisotropy and subsequent
tendency for preferential lateral flow, lead to a wider lateral spread of moisture dis-
tribution with depth, which in turn may cause leachate seepage at the side slope.
Since volumetric waste moisture contents at deeper sections of the landfill are gen-
erally higher (since porosity is reduced), leachate seeps are often most common at
the base of large landfills with elevated liquids levels.
In some cases, seeps on the side slope or base of a landfill may be the result of
highly permeable cover materials used on side slopes. As illustrated in Fig. 11.7,
when stormwater flows through permeable cover material along an exterior landfill
slope, the water contacts waste and creates leachate, which can then seep out at the
bottom of the landfill. In cases where the cover material is high in organic matter or
iron, the stormwater can become discolored and have an appearance similar to
leachate. Such occurrences are common when ground wood or yard trash is used as
a cover material on steep slopes.
11.3 Managing Leachate Seeps 249
Fig. 11.7 Schematic of leachate seepage caused by high-permeability cover at the side slope
Fig. 11.8 Schematic of leachate seepage at a landfill caused by high injection pressure
The addition of liquids to the landfilled waste has the potential to contribute to
leachate seeps. As described in Chaps. 8 and 9, landfill operators at large facilities
often rely on adding liquids under pressure to distribute liquids within the landfill.
Without the addition of pressure, it may be difficult to distribute liquids within the
landfill in the desired time period. Leachate seepage occurs when the saturated zone
reaches the side slope surface as conceptually illustrated in Fig. 11.8. The size of the
saturated zone formed by the added liquids depends to a large extent upon the liq-
uids injection pressure. The higher the injection pressure, the larger the saturated
zone, which in turn results in a larger lateral spread. The flow modeling techniques
described in earlier chapters can be used to estimate the degree to which pressurized
liquids might migrate to the landfill side slope.
250 11 Leachate Control, Storage, and Treatment
Landfill engineers and operators can employ several strategies to prevent, minimize
or otherwise control seeps at landfills. Such measures may be implemented during
the design stage of a liquids addition system, as well as during construction and
operation the system. The landfill operator also has a great ability to impact the
potential for seeps by practicing certain fundamental waste and cover soil placement
practices; examples of these practices are highlighted in the following sections.
The selection and placement of cover material requires special attention during
landfill operations and planning. An ideal cover soil would be one where the perme-
ability of the cover soil is similar to the disposed waste—however, numerous other
factors come into play when determining the optimal type and amount of cover
material (e.g., availability, cost, meeting minimum regulatory requirements for vec-
tor reduction, litter prevention, and other considerations). The use of alternative
cover materials such as foams or tarps that result in less heterogeneity in landfilled
materials serve to reduce stratification and resulting pathways for liquids short-
circuiting. When site conditions dictate that potentially problematic cover materials
(from a seepage perspective) be used, operators often attempt to scrape as much of
the soil from the underlying waste surface as possible before placing the next waste
lift. This practice of soil scraping has positive and negative economic impacts on
operations: scraping provides additional airspace for new waste disposal, but the
time and equipment requirement to scrape the soil might be more costly. Even if
only a fraction of the soil layer is removed, a reduction in heterogeneous stratified
layers will minimize lateral movement of liquid to the landfill slope. Landfill opera-
tors with low permeability cover soils sometimes dig “windows” in the cover to
promote liquids to drain from one lift to the next.
Some sites receive large amounts of low-permeability granular wastes such as
ash, and they are often allowed by permit to use these materials as daily cover.
Lateral movement of leachate along compacted ash layers has been found at several
sites to result in seep problems. The economic benefit of such practices (i.e., receiv-
ing an approved soil-like material for daily cover for free, a reduced cost, or even
with a tipping fee) must be weighed against longer term operational and mainte-
nance issues related to seep control. One strategy includes using ash as cover on
interior landfill surfaces and avoiding its use in areas near side slopes. Ash disposed
toward the edge of the landfill should be mixed with other wastes to avoid the for-
mation of distinct layers.
One strategy to avoid seeps involves grading waste lifts adjacent to the side slope
(and, by extension, grading the layers where cover material is placed) towards the
interior of the landfill. As shown in Fig. 11.9, the cover soil layer is sloped near
the edge of the landfill to drain inward and thus preclude added liquids from reach-
ing the landfill surface and emerging as a seep. This operational practice would also
need to consider the impact on stormwater management, as the development of an
inward gradient could cause more stormwater accumulation near working areas if
not otherwise diverted.
11.3 Managing Leachate Seeps 251
Fig. 11.9 Cover removal and lift grading strategies to minimize seeps
For sites with GCCS, efficient operation of the GCCS can aid in seep control.
Preferential flow paths out the side of the landfill caused by landfill gas migration
can promote the flow of leachate through these same paths. A well-operated GCCS
will induce a negative pressure within the landfill and direct gas movement towards
gas collection points, and by extension reduce the preferential flow paths that leach-
ate can travel. When directing gas movement (and, to a degree, liquids movement)
toward gas collection points, the accumulation of liquids can occur at gas collection
points. This can be remedied by installing temporary or permanent pumps to remove
liquid build-up (discussed in Chap. 13).
The extent to which seeps are likely to occur will be largely influenced by the
moisture content of the waste itself and the degree to which liquids addition occurs.
As discussed in Chap. 6, facilities that specify large liquids addition volumes will
require aggressive approaches to meet the design targets in the form of more liquids
addition devices and/or greater liquids addition pressures. When such a strategy is
pursued, the engineer and operator must expect that seeps will occur and must there-
fore take steps both in system design and in development of the operations plan to
address these events. Alternatively, the objective at some sites might be to apply a
more conservative approach and add less liquid with a goal of preventing seeps from
occurring (e.g., designing liquids addition devices to be far away from side slopes
and specifying low injection pressures). These two approaches are conceptually
illustrated in Fig. 11.10 for a landfill employing horizontal liquids addition. In one
approach, injection lines are added throughout the landfill to penetrate as much of
the waste mass as possible (Fig. 11.10a). In the conservative approach, the number
of liquids addition devices is deliberately minimized and maintained at a conserva-
tive distance away from the side to avoid seeps (Fig. 11.10b).
Designers and operators can take some steps when constructing liquids addition
devices to minimize seeps. Liquids addition systems that employ horizontal trenches
often place injection lines on consecutive or alternating waste lifts, with each line
penetrating the side of the landfill and connecting to an external distribution manifold
252 11 Leachate Control, Storage, and Treatment
Fig. 11.10 General approaches for seepage control at a bioreactor landfill (a) operate conserva-
tively to avoid seeps, (b) operate to manage seeps
(see Chap. 9). Pipe perforations begin after a specified setback distance away from
the side slope. In practice, seep collars can be constructed to reduce the preferential
flow of leachate along the pipe and towards the side slope. For example, clay can be
placed in a horizontal trench between the side slope and the start of perforations and
compacted in place. An additional construction technique for horizontal trenches
includes sloping the pipes toward the interior of the landfill to encourage gravity
drainage into the landfill, rather than toward the side slope.
Adjusting the number of penetrations into the landfill caused by liquids addition
devices represents another approach to minimizing leachate seeps; fewer side slope
penetrations will result in fewer seepage issues. Adjusting the location where
injection device perforations begin may also prove a useful seep prevention strategy,
particularly since injection pressure is greatest at the point of the first penetration of
the injection device. Starting perforations more toward the interior of the landfill
and then branching toward exterior areas should reduce seep potential; this strategy
is illustrated in Fig. 11.11. In Fig. 11.11a, each injection device has a penetration
into the landfill, which allows for greater control of injection parameters into a
device, but results in a large amount of penetrations; pressure is greatest at the point
11.3 Managing Leachate Seeps 253
Fig. 11.11 Schematic of approaches for the penetration of horizontal leachate injection line (a)
approach with multiple penetrations, (b) approach with limited penetrations
where the penetrations first start, nearer the landfill slope. In Fig. 11.11b, multiple
injection lines are tied together by a manifold in the interior of the landfill and a
single side slope penetration is constructed—this reduces liquids addition control
for individual devices but reduces the number of side slope penetrations.
The use of a distribution blanket instead of buried trenches may have a similar
result as the case shown in Fig. 11.11b. In the case of a buried distribution blanket
or a design with minimal side slope penetrations, the design and construction must
be performed to mitigate the potential for failure of the line, since failure would
preclude the ability to distribute liquids to a large area. Possible failure modes to be
planned for include crushing, clogging, or pipe disconnection.
When specifying the distance that a liquids addition device must be kept from the
landfill side slope to avoid seepage, factors such as waste properties, device configu-
ration, and planned operation should be considered. The techniques described in
Chaps. 8 and 9 for predicting the extent of liquids migration from a device (a func-
tion of flow rate, pressure, waste characteristics, and operation time) can be used to
estimate the location of the saturated zone with respect to the slope and thus to
determine a necessary setback distance. Xu et al. (2013) examined appropriate set-
back distances to avoid seeps and found that for typical design and operating condi-
tions the influence zone from vertical wells placed adjacent to or on a side slope
should not intercept the landfill surface and the appropriate setback distance (XD)
from a horizontal trench to the edge of the landfill could be approximated as:
Q
XD =
3K z
when KX/KZ > 20 and where Q is the liquids flow rate per length of trench and KZ
and KX are the vertical and horizontal hydraulic conductivity values, respectively.
As described earlier, however, the presence of waste or soil layers of high or low
permeability could result in liquids migration over much large distances.
254 11 Leachate Control, Storage, and Treatment
Regardless of the steps taken to prevent seep formation, the operator should rou-
tinely inspect for the presence of seeps and have contingencies in place to address
seeps when they are observed. The designer should include a seep management plan
as part of the site’s operations plan. Once leachate seepage occurs, the seep must be
promptly addressed to avoid further environmental issues. Operators may use sev-
eral methods to address seepage issues, but regardless of the approaches employed,
the appropriate procedures and supplies must be readily available to quickly address
observed problems.
With respect to the seep management plan, all necessary landfill personnel should
be trained to identify seep-related situations such as wet areas on the surface of the
landfill, surface cracks, and erosion. This can be accomplished as part of a routine
visual inspection of the landfill site. Inspection components most often include: (a)
examination of the landfill surface and side slopes for signs of seeps (depending on
the size, configuration, and cover type, this may be performed by walking the site or
observing from an on- or off-road vehicle), (b) examination of exposed liquids pip-
ing for signs of leakage, (c) examination of liquid, air, and gas pipes and hoses for
signs of breakage or wear, (d) visual inspection of the storm water management
system, and (e) visual inspection of the valve positions. These daily inspections
should be conducted by one or more trained individuals and recorded on an inspec-
tion sheet or using a form loaded onto a laptop or tablet computer.
The conventional practices for addressing seeps often involve placing additional
cover soil (usually a low-permeability clayey soil) on top of the area of concern fol-
lowed by compaction of the material. Depending on the magnitude of the seep
source, this may only provide a temporary solution, as liquids may migrate around
the compacted soil to reemerge somewhere else on the slope. Figure 11.12 illus-
trates this conventional soil-compaction approach along with expanded strategies
that allow the moisture to be redirected away from the slope. A more detailed
approach includes excavating the area of the seep, adding stone (or some similar
drainage media), and providing a drain or chimney that permits leachate to be
directed back into the landfill or the LCRS.
For sites with a high likelihood of seeps, such as those employing a more aggres-
sive strategy for liquids addition, the design of a robust toe drain at the base of the
landfill represents a useful control strategy. The toe drain should be designed and
constructed to allow the operator to connect slope drains that are constructed as seep
locations are identified and remediated. Since a prime location for leachate seeps is
at the base of landfill slope, benches, or access roads, the design of seep drains as
part of this infrastructure can be incorporated into the landfill liner system design
and into the long-term operations and closure plan for the landfill.
Future planning for long-term landfill cover and closure is an integral part of the
planning process for leachate seeps at landfills practicing aggressive liquids addition
strategies. Some operators pursue a “close-as-you-go” approach where components
of a final closure system are constructed on outer landfill slopes as they reach final
11.4 Leachate Storage 255
Fig. 11.12 Strategies for addressing a side slope seep: (a) seepage occurring; (b) excavation; (c)
filling and compaction; (d) placing a surface drainage system
grade (this approach is discussed in greater detail in Chap. 17). Such a practice enables
the engineer to link a toe drain system for seep control to future vertical phases of
landfill construction over time. In scenarios when accelerated slope closure is not
feasible, alternative side slope cover strategies may integrate well with seep control.
For example, placing exposed geomembrane caps or geosynthetic rain covers on the
side slopes as a means of shedding rainfall and minimizing erosion can reduce seep-
age problems. Such systems could tolerate a greater degree of leachate seepage com-
pared to traditional cover systems, though the integration of slope and toe drains
would need to be implemented. Creative management of the interface of the bottom
liner system with the side slopes and eventual cover system—especially systems that
integrate stormwater runoff, seep control, and gas recovery—should be considered.
An integral component at nearly all landfill facilities is the leachate storage system. At
facilities practicing on-site treatment, leachate storage might be integrated into treat-
ment operation. At facilities where off-site leachate disposal is practiced (unless the
discharge point is located very close to the landfill, or where leachate is recirculated
to the landfill), storage will be necessary to provide necessary equalization of flow
256 11 Leachate Control, Storage, and Treatment
Fig. 11.13 HDPE lined leachate storage pond equipped with floating aerators
and to provide holding capacity when off-site discharge or recirculation is not possi-
ble. The selection of a storage system depends on considerations such as the type of
waste disposed, volume of leachate expected, available space, and cost. Common
storage mechanisms include ponds, lagoons, and tanks.
Lined leachate ponds or lagoons are commonly used at landfill sites. The liner
system is similar to the one used for the barrier layer at the bottom of the landfill. In
some cases, a double liner with a leak detection system might be employed. A com-
mon practice includes placing floating aerators in the pond to provide initial leach-
ate treatment and reduce odor emissions (Fig. 11.13); the aeration system provides
oxygen and promotes mixing. Solids accumulated on the bottom of a storage pond
may need occasional removal. This can be accomplished by draining the pond and
removing residues by hand or using a vacuum truck. Care must be taken to avoid
damaging the liner system.
If the leachate storage unit is not covered, the volume of leachate requiring man-
agement will be influenced by local climate conditions. In dry areas, a net loss in
leachate may result because of evaporation. In wet climates, however, a net addition
of water as a result of rainfall will add to the total volume of leachate. A solution to
this problem for ponds in wet climates is to place a floating geomembrane cover on
top of the leachate (Fig. 11.14). Rainwater that accumulates on top of the geomem-
brane can be periodically removed using a pump.
Some facilities employ leachate storage tanks or structural basins, with primary
construction materials including steel, fiberglass, and concrete (Figs. 11.15 and 11.16).
For some storage systems, the top of the tanks remain open to the atmosphere, and
often include manifold diffusers for air addition. In other systems the tanks are closed
11.4 Leachate Storage 257
Fig. 11.14 HDPE lined leachate storage pond equipped with surface rain cover
Fig. 11.15 Fiberglass leachate storage tanks surrounded by secondary containment system
to the atmosphere, with either a floating cover or a fixed cover that vents to the atmo-
sphere. Just as with leachate ponds and lagoons, the sediment that builds up in the
bottom of the tanks must be occasionally removed. Secondary containment for tank
systems is provided using an outer concrete vault or wall, or by placing the entire tank
in a lined unit.
258 11 Leachate Control, Storage, and Treatment
Fig. 11.16 Glass-lined steel leachate storage tanks surrounded by secondary containment system
One of the most common and desirable methods for managing leachate is through
discharge to an existing wastewater treatment facility not associated with the land-
fill. Leachate transport is typically accomplished using a force main (a direct pipe
connection) or tanker trucks. Since leachate quantities should be small compared to
influent volumes for a domestic wastewater treatment facility, this method of man-
agement offers the advantage of diluting some of the more difficult-to-treat con-
stituents. Municipal landfills are often subject to pretreatment standards prior to
discharge to a domestic wastewater treatment facility, and if these pretreatment
standards are not met, the wastewater treatment facility may either stop accepting
the leachate or charge a higher price to receive the leachate. If pretreatment stan-
dards pose a substantial challenge, construction and operation of infrastructure for
pretreatment on-site may be necessary. Alternatively, a more distant wastewater
treatment plant may be willing to accept the leachate for treatment, but hauling
costs would increase. In addition to municipal facilities, private wastewater treat-
ment facilities may be available, though costs and travel distances may be greater.
When off-site treatment options become too limited, construction and operation
of an on-site treatment facility may be necessary. A variety of methods are avail-
able for treating leachate on-site, and the method selected depends on the fate of
the leachate and the regulatory requirements associated with the discharge loca-
tion. Given the nature and variability (with respect to quantity and quality) of
leachate, the treatment system design should allow flexibility to modify capacity
and treatment processes required. Earlier in Chap. 2, general categories of leachate
chemical constituents were outlined and the type of treatment technology most
effective for each constituent category differs. Table 11.2 summarizes leachate
treatment options associated with these categories. The rest of the chapter summa-
rizes some of the more common treatment technologies used for landfill leachate as
well as treatment issues to be considered at facilities implementing sustainable
landfill practices.
260 11 Leachate Control, Storage, and Treatment
Fig. 11.17 On-site leachate treatment plant consisting of multiple biological and physical treat-
ment operations
Fig. 11.18 A bank of spiral wound reverse osmosis membranes used as a polishing step in an
integrated leachate treatment operation
Fig. 11.19 Wetlands leachate treatment system located adjacent to an operating landfill
water source (Kadlec and Knight 1996; Kadlec 1999; Bulc et al. 1997; DeBusk 1999).
Wetlands treatment may function better in warmer climates that allow the vegetation to
flourish for a greater portion of the year. A variety of wetland configurations have been
employed, including natural wetlands, aquatic plant systems, constructed subsurface
flow wetlands, and constructed surface flow wetlands (Fig. 11.19). Much of the perfor-
mance data on wetlands treatment has been collected for systems designed to remove
pollutants from municipal and industrial wastewaters. However, the pollutant-removal
mechanisms that have been identified in wetlands receiving domestic or industrial
wastewater should operate in a similar manner for the treatment of landfill leachates.
Wetlands treatment provides both physical and biological treatment mechanisms to
remove pollutants such as nitrogen, phosphorus, metals, and organic compounds.
11.5.2 L
eachate Treatment Considerations for Sustainable
Landfill Operations
Fig. 11.20 Leachate treatment strategies for landfill practicing leachate recirculation
off-site treatment plant to handle excess leachate, and where on-site treatment is
desired or necessary, to implement technologies that complement a landfill with
liquids addition. Several leachate treatment approaches have potential to integrate
with leachate recirculation activities and goals. For instance, leachate evaporation
can be used as a means of reducing leachate volumes and, if the landfill is still
active, residual solids can simply be disposed in the disposal operation.
One approach used in conjunction with leachate recirculation to produce an
effluent of quality to be discharged on site while recirculating the rest of the leachate
is reverse osmosis (RO) or similar membrane systems (Fig. 11.20). In this process,
leachate is separated into dilute (permeate) and concentrated streams by placing the
leachate under pressure in contact with a RO membrane. RO membranes have been
found to be successful in rejecting most pollutants, including salts, dissolved organic
matter, and heavy metals (Linde et al. 1995; Ahn et al. 2002; Ushikoshi et al. 2002).
These systems do not require the degree of permeate production as RO systems
designed for desalination of drinking water, and thus can be performed under much
less pressure. This technique can be used in combination with leachate recirculation
to remove net moisture from the landfill.
Another strategy that lends itself to those facilities practicing leachate recircula-
tion is the use of oxidizing chemicals to transform some of the recalcitrant organic
matter in the leachate to a form that can be biologically consumed (and turned to
biogas) within the landfill. For example, Fenton’s reagent has been extensively inves-
tigated and applied to treat landfill leachate (Gau and Chang 1996; Bae and Kim
References 265
1997; Yoon and Cho 1998; Kang and Hwang 2000; Lin and Chang 2000; Zhang et al.
2006), and this chemical process has been specifically evaluated as a technique that
could be used with landfills practicing leachate recirculation (Batarseh et al. 2007).
The reaction involves H2O2 and a ferrous iron catalyst; the decomposition of H2O2 is
enhanced by the ferrous iron acting as a catalyst, resulting in the generation of
hydroxyl radicals that can oxidize the refractory organics. Lopez et al. (2004) reported
that approximately 60 % of COD was removed by Fenton’s reagent pretreatment and
the BOD5/COD ratio was increased from 0.2 to 0.5
At some point, active recirculation of leachate or other liquids will cease and
then leachate management will primarily consist of removal of leachate collected in
the LCRS and appropriate treatment and disposal. Leachate volumes should dimin-
ish over time as free liquids migrate out of the landfill under gravity. The long-term
rate of leachate generation will depend on the effectiveness of the closure system for
diverting rainwater from infiltrate into the waste. Final determinations of how land-
fill leachate will be managed depend on regulatory post-closure care requirements
designed to protect human health and the environment. These issues are discussed
for the landfill as a whole in Chap. 17.
References
Ahn W, Kang M et al (2002) Advanced Landfill Leachate Treatment Using an Integrated Membrane
Process. Desalination, 149(1-3), 109–114
Aktas O, Cecen F (2001) Nitrification inhibition in landfill leachate treatment and impact of acti-
vated carbon addition. Biotechnol Lett 23(19):1607
Aluko O, Sridhar M (2005) Application of constructed wetlands to the treatment of leachates from
a municipal solid waste landfill in Ibadan, Nigeria. J Environ Health 67(10):58–62
Amokrane A, Comel C, Veron J (1997) Landfill leachates pretreatment by coagulation-flocculation.
Water Res 31:2775–2782
Bae J, Kim S et al (1997) Treatment of landfill leachates: ammonia removal via nitrification and
denitrification and further COD reduction via Fenton’s treatment followed by activated sludge.
Water Sci Technol 36(12):341–348
Batarseh EB, Reinhart DR, Daly L (2007) Liquid sodium ferrate and Fenton’s reagent for treat-
ment of mature landfill leachate. ASCE J Environ Eng 133(11):1042–1050
Baumgarten G, Seyfried C (1996) Experiences and new developments in biological pretreatment
and physical post-treatment of landfill leachate. Water Sci Technol 34(7–8):445–453
Berge ND, Reinhart DR, Townsend TG (2005) The fate of nitrogen in bioreactor landfills. Crit Rev
Environ Sci Technol 35:365–399
Bila D, Montalvao A et al (2005) Ozonation of a landfill leachate: evaluation of toxicity removal
and biodegradability improvement. J Hazard Mater 117(2–3):235–242
Borghi A, Binaghi L et al (2003) Combinted treatment of leachate from sanitary landfill and
municipal wastewater by activated sludge. Chem Biochem Eng Q 17(4):277–283
Bulc T, Vrhovsek D et al (1997) The use of constructed wetland for landfill leachate treatment.
Water Sci Technol 35(5):301–306
Cecen F, Aktas O (2001) Effect of PAC addition in combined treatment of landfill leachate and
domestic wastewater in semi-continuously fed batch and continuous-flow reactors. Water SA
27(2):177
Comstock SHE, Boyer TH, Graf KC, Townsend TG (2010) Effect of landfill characteristics on
leachate organic matter properties and coagulation treatability. Chemosphere 81:976–983
266 11 Leachate Control, Storage, and Treatment
DeBusk W (1999) Evaluation of a constructed wetland for treatment of leachate. In: Mulamoottil
G, McBean EA, Rovers F (eds) Constructed wetlands for the treatment of landfill leachates.
Lewis, Boca Raton, p 175
de Morais JL, Zamora PP (2005) Use of advanced oxidation processes to improve the biodegrad-
ability of mature landfill leachates. J Hazard Mater 123(1–3):181
Diamadopoulos E, Samaras P et al (1997) Combined treatment of landfill leachate and domestic
sewage in a sequencing batch reactor. Water Sci Technol 36(2–3):61–68
Fang H, Lau I et al (2005) Anaerobic treatment of Hong Kong leachate followed by chemical
oxidation. Water Sci Technol 52(10–11):41–49
Gau SH, Chang S (1996) Improved Fenton method to remove recalcitrant organics in landfill
leachate. Water Sci Technol 34(7–8):455–462
Grady LCP Jr, Daigger GT, Love NG, Filipe CDM (2011) Biological wastewater treatment, 3rd
edn. CRC, Boca Raton, p 17
Gulsen H, Turan M (2004) Treatment of sanitary landfill leachate using a combined anaerobic
fluidized bed reactor and Fenton’s oxidation. Environ Eng Sci 21(5):627–636
Inanc B, CalIl B et al (2000) Characterization and anaerobic treatment of the sanitary landfill
leachate in Istanbul. Water Sci Technol 41(3):223–230
Kadlec R (1999) Constructed wetlands for treating landfill leachate. In: Mulamoottil G, McBean EA,
Rovers F (eds) Constructed wetlands for the treatment of landfill leachates. Lewis, Boca Raton
Kadlec R, Knight R (1996) Treatment wetlands. CRC, Boca Raton
Kang YW, Hwang KY (2000) Effects of reaction conditions on the oxidation efficiency in the
Fenton process. Water Res 34(10):2786–2790
Kurnianwan TA, Lo WH, Chan GYS (2006) Physico-chemical treatments for removal of recalci-
trant contaminates from landfill leachate. J Hazard Mater 129:80–100
Lin S, Chang C (2000) Treatment of landfill leachate by combined electro-Fenton oxidation and
sequencing batch reactor method. Water Res 34(17):4243–4249
Linde K, Jonsson A et al (1995) Treatment of three types of landfill leachate with reverse osmosis.
Desalination 101:21–30
Lopez A, Pagano M et al (2004) Fenton’s pre-treatment of mature landfill leachate. Chemosphere
54(7):1005–1010
McCreanor PT, Reinhart DR (2000) Mathematical modeling of leachate routing in a leachate recir-
culating landfill. Water Res 34(4):1285–1295
Metcalf & Eddy, Tchobanoglous G, Stensel HD, Tsuchihashi R, Burton F (2013) Wastewater engi-
neering: treatment and resource recovery, 5th edn. McGraw-Hill, New York
Powrie W, Beaven RP (1999) Hydraulic properties of household waste and implications for land-
fills. Proc ICE Geotech Eng 137:235–247
Soh I, Hettiarachchi H (2009) Potential lateral migration of leachate in flushing bioreactor landfills
during aggressive leachate recirculation. Pract Period Hazard Toxi Radioact Waste Manag
13(3):174–178
Steensen M (1997) Chemical oxidation for the treatment of leachate—process comparison and
results from full-scale plants. Water Sci Technol 35(4):249–256
Tatsi AA, Zouboulis AI, Matis KA, Samaras P (2003) Coagulation-flocculation pretreatment of
sanitary landfill leachates. Chemosphere 53:737–744
Timur H, Ozturk I (1997) Anaerobic treatment of leachate using sequencing batch reactor and
hybrid Bed filter. Water Sci Technol 36(6–7):501–508
Ushikoshi K, Kobayashi T et al (2002) Leachate treatment by the reverse osmosis system.
Desalination 150(2):121–129
Uygur A, Kargi F (2004) Biological nutrient removal from Pre-treated landfill leachate in a
sequencing batch reactor. J Environ Manage 71(1):9
Wiszniowski J, Robert D et al (2006) Landfill leachate treatment methods: a review. Environ
Chem Lett 4:51–61
Xu Q, Powell J, Tolaymat T, Townsend T (2013) Seepage control strategies at bioreactor landfills.
J Hazard Toxic Radioact Waste 17:342–350
Yoon J, Cho S (1998) The characteristics of coagulation of Fenton reaction in the removal of land-
fill leachate organics. Water Sci Technol 38(2):209–214
Zhang H, Zhang DB et al (2006) Removal of COD from landfill leachate by electro-Fenton
method. J Hazard Mater 135(1–3):106–111
Chapter 12
Slope Stability
Assessing the stability of a landfill side slope is a primary element in the design
process for all landfills. The engineer evaluates the stability of cover material com-
ponents on the slopes of the landfill and the internal stability of the waste mass
(waste plus soil) itself. A thorough assessment of slope stability is especially impor-
tant at landfills where liquids addition is practiced, as elevated pore-water pressures
in a landfill resulting from added liquids and generated gases can lead to a decrease
in the effective stress placed on the waste and on waste-soil or waste-geosynthetic
interfaces. In addition, changes in waste properties due to biological decomposition
of the waste can result in strength changes of the landfilled material.
Several landfill side slope failures have been attributed, at least in part, to ele-
vated liquid levels within the landfill (Blight 2008; Hendron et al. 1999; Stark et al.
2000; Thiel and Christie 2005) and the consequences of these failures have been
severe, including multiple human deaths in some cases (see Fig. 12.1 for an exam-
ple catastrophic failure in the Philippines). Koerner and Soong (2000) discussed the
influence of leachate in landfills on slope stability. The role of leachate under several
different scenarios was described, including perched zones of leachate within the
landfill, leachate head on the liner, and added pore pressures resulting from liquids
Fig. 12.1 Slope failure at the Payatas landfill, Philippines (Photo courtesy of Scott Merry)
addition. The potential role of landfill gas pressure was highlighted as an issue by
Merry et al. (2006). In a review of six landfill slope failures from sites around the
world, Landva and Dickinson (2012) found that the properties of decomposed waste
played a key role in the observed failures.
Slope stability assessment is an extensive engineering topic unto itself, and entire
design texts and software packages are devoted to such analyses (Abramson et al.
2002; Das 2005; Duncan and Wright 2005; GeoSlope International Ltd 2007; Krahn
2007). This chapter is devoted to slope stability because of the great importance of
slope stability considerations at landfills that add liquids to the waste mass (GeoSlope
International Ltd 2007).
The factor of safety (FOS), defined as the ratio of the shear strength (s) of the media
(or the interface between different media) to the shear stress (τ) required to maintain
equilibrium, is a term commonly used to quantify the ability of a slope to prevent or
resist movement compared to forces that would cause slope movement. The FOS is
represented as follows:
Shear strength ( s )
FOS = (12.1)
Shear stress (t )
12.2 Slope Stability Fundamentals 269
c + (s V - u ) tanf
FOS = (12.4)
t
A slope failure is expected to occur when the shear stress exceeds the shear strength
(i.e., FOS < 1.0). The plane where the failure occurs is referred to as the slip surface.
Typical engineering practice is to design for FOS of 1.2–1.5. Shear strength as
described above refers to an internal quality related to a single medium (e.g., soil,
waste), but similar concepts apply to the interface between two media (Koerner 2005).
When two types of media are involved, the interface friction angle between the two
media (typically denoted as δ) and the adhesion (typically denoted as A) must be
measured or otherwise estimated. Throughout the remainder of this chapter, slope
stability concepts are discussed and illustrated by describing a given medium with an
internal angle of friction and cohesion; the role of interfaces between different materi-
als is discussed, but not quantitatively examined. The design engineer must recognize
appropriate interfaces between different media in the system and design accordingly.
A common engineering practice for assessing landfill slope stability is to use a
computer-based slope stability model with site-specific inputs. Examples of slope
stability model input parameters include landfill configuration (e.g., height, slope),
waste characteristics (e.g., friction angle, cohesion), and characteristics of the inter-
face between the waste and other landfill components (e.g., soil, geosynthetics). As
discussed previously, a slope failure can result from several different factors. An
increase in pore-water pressure (as might occur when liquids are added or not
appropriately drained) reduces the effective stress and the resulting shear strength.
A reduction in media properties such as friction angle and cohesion (as might occur
when waste decomposes) can likewise reduce shear strength. Changes in configura-
tion (such as a slope change or the removal of a soil at the base of a slope) can result
in a decrease in those forces restraining movement.
A slope failure has the potential to occur during several different phases of land-
fill construction and operation, including liner construction, waste placement, and
after landfill closure (Abramson et al. 2002). Figure 12.2 illustrates common slope
270 12 Slope Stability
Fig. 12.2 Basic slope failure modes at landfills: (a) a circular failure; (b) a block failure; and (c)
a veneer failure
With a circular failure, often referred to as a rotational failure, the failure occurs
within the waste mass (Abramson et al. 2002). The slip surface is often illustrated
and modeled as a circular arc to simplify the calculation process. In general, circular
failures are more common in slopes composed of homogeneous material. A block
failure occurs along a weak failure plane within the waste mass or at the interface
between the waste mass and the surrounding infrastructure (e.g., the landfill liner
system and the interface between the waste, soil, and geosynthetic layers resting
above or as part of the liner). A veneer failure, also referred to as a cover failure,
may be more likely to occur during the construction of a landfill cover system, and
usually occurs along weak interfaces between the waste and geosynthetics on the
landfill slope. Water seepage and loads applied by large construction equipment in
the cover system are typical contributors to veneer failure (Abramson et al. 2002).
One of the basic methods to evaluate the FOS during slope stability analysis is the
ordinary method of slices, which was developed in the 1920s. As illustrated in
Fig. 12.3, the method of slices examines slope stability by assuming a circular fail-
ure plane. In this method, a trial circular slip surface is drawn in the cross section of
the slope, and the slip surface is divided into several vertical slices of equal width
(ΔL). The weight of each slice can be resolved into two components: one normal to
the base of the slice (Wn) and one parallel to the base (Wp). It is the parallel portion
(Wp) that tends to cause sliding. Waste cohesion and higher internal friction angles
can increase resistance to failure. The cohesion is equal to the product of waste
Fig. 12.3 Typical slice and forces for the ordinary method of slices, where W is the slice weight;
Wn is equal to Wcosα; the normal force is on the bottom of the slice; Wp is equal to Wsinα; the
sliding force is on the bottom of the slice; and ΔL is the length of each vertical slice
272 12 Slope Stability
cohesion (c) and the slice width (ΔL), and the internal friction is equal to normal
force, Wn, multiplied by the friction coefficient (tan ϕ). The FOS can then be calcu-
lated as follows (Liu and Evett 2001):
c × DL + åW cosa × tanf
FOS = (12.5)
åW sina
The ordinary method of slices provides a technique to calculate FOS directly and
is convenient for hand calculations. However, it is less accurate because it ignores
inter-slice forces among the vertical slices. Several similar methods have been
developed for FOS analysis that considers inter-slice forces and moment equilib-
rium. Table 12.1 summarizes the most commonly used methods for slope stability
analysis. Refer to other seminal texts for detailed calculation procedures related to
these methods (Abramson et al. 2002; Duncan and Wright 2005; Gunaratne 2006;
Liu and Evett 2001).
As described in Table 12.1, for more sophisticated methods (Spencer and
Morgenstern-Price), an iterative, trial-and-error calculation procedure is needed to
satisfy moment and force equilibrium for each slice, which makes hand calculation
impractical (Abramson et al. 2002; Krahn 2007). Many computerized programs
have been developed for slope stability analysis, and most computer programs can
handle a wide variety of slope geometries, shear strengths, pore-water pressures,
and external loads, and have the capabilities for automatically searching for the
most critical slip surface with the lowest FOS (Duncan and Wright 2005).
Commonly-employed computer programs include SLOPE/W, SLIDE, UTEXAS4,
XSTABL, and WINSTABL. Each of these programs can investigate different slope
failure modes (e.g., circular, block, veneer, combination) that result from variations
in site-specific conditions and landfill design (Pockoski and Duncan 2000). When
assessing slope stability at a landfill site, the design engineer will use design speci-
fications for site configuration, dimensions, and material characterization. In many
cases internal and interface friction angles (as well as cohesion and adhesion) will
be determined from laboratory tests on site-specific materials or material combina-
tions. It would be rare, however, for project-specific ϕ and C data for MSW to be
collected. The engineer would in most cases use data (or a range of data) gathered
from the literature; Chap. 5 presented a review ϕ and C data reported in the litera-
ture for MSW.
As pore-water pressures play such a crucial role with respect to slope stability
at landfills, the designer should incorporate pore pressures into the simulations for
projects where liquids are added or exist present in large amounts. One approach
that some designers use involves simulating an elevated liquid level originating at
the base of the landfill. This would be representative of typical slope stability
analysis in earthen embankments or dams where water seeps as a result of differ-
ent water levels on each side. For modern landfills with a functioning LCRS, how-
ever, such elevated pore pressures above the base of the landfill would not be
expected (Koerner and Soong 2000). Pressures would more likely be elevated
within the waste mass surrounding liquids addition devices. Many slope stability
12.3 Methods for Assessing Slope Stability 273
Simplified n
software packages can be integrated with moisture seepage (using the principles
described in Chap. 5), allowing the engineer to simulate the interconnected role of
pressurized liquids addition and slope stability. In the following section, examples
of such an analysis are presented.
274 12 Slope Stability
Xu et al. (2012) examined the factors affecting slope stability at landfills practicing
pressurized liquids addition by coupling a porous media fluid flow model (SEEP/W)
with a slope stability model (SLOPE/W). Pressurized liquids addition using a bur-
ied horizontal trench (similar to that presented in Chap. 9) was simulated in the
analysis. Slope stability was evaluated under a variety of operating scenarios,
including simulations representing clogging of the LCRS, use of different types of
cover soil, varying operating conditions to limit leachate seepage, and varying
injection pressures.
Figure 12.4 presents the results of a base simulation scenario representative of a
landfill where subsurface liquids addition beneath a side slope is practiced. A hori-
zontal injection trench (1 m × 1 m) is located 30 m above the base of the liner in a
50 m deep landfill beneath the side slope (with a configuration of three horizontal to
one vertical (3H:1V)) of a lined landfill. In the base scenario, liquids are added
under a constant injection pressure of 5 m water column (wc). As discussed in Chap.
9, this amount of injection pressure would fall on the high side of those typically
used in liquids addition operations.
A conceptual zone of elevated moisture surrounding the trench is illustrated in
Fig. 12.4, along with the FOS that develops over time. The bottom of the landfill is
treated as free-draining and represents a well-designed and operated LCRS; the
moisture profile therefore is more sharply delineated at a defined distance from the
trench. As discussed in Chaps. 5 and 9, the extent of moisture distribution and resul-
tant pore-water pressures is dictated by several factors including the operating con-
ditions (e.g., pressure, flow rate, operating regime), waste properties (e.g., hydraulic
conductivity, anisotropy), and trench design (e.g., size, bedding media). In this case,
Fig. 12.5 The effect of the LCRS on slope stability, comparing a functioning LCRS (LCRS
works) to a poorly functioning LCRS (LCRS fails)
where the LCRS is functioning properly, an FOS of greater than 1.6 is maintained
throughout the simulation period. The slip surface associated with the lowest FOS
is presented in Fig. 12.4. The simulation results suggest that for the conditions mod-
eled, slope integrity would be maintained even when liquids were added under pres-
sure directly beneath the side slope.
The impact of LCRS performance was evaluated by modifying the base scenario
and treating the bottom surface as an impermeable layer (rather than as a freely-
draining surface); this simulates a scenario where the LCRS is insufficiently
designed, improperly operated, or otherwise clogged or compromised. Consequently,
liquids mound on the bottom liner, which results in the development of increased
pore-water pressures at the base of the landfill. Figure 12.5 illustrates the conceptual
zone of elevated moisture surrounding the trench (with time), along with the FOS
that develops over time. The resultant flow rate in this simulation did not differ
dramatically from the base scenario, indicating that the flow impedance at the base
of the landfill did not have a major impact on the ability to add liquids. However,
the FOS decreased to less than 1.5 after about four years; by year 8 the FOS was less
than 1.2. The slip surface predicted at the sixth year of simulation (the surface cor-
responding to the lower FOS) occurs at the base of the landfill slope. While a FOS
less than 1 is not reached during the simulation period, the results clearly demon-
strate the critical nature of an adequately designed and properly functioning LCRS.
Landfill operators place cover soil (or alternative materials that perform equiva-
lently) on exposed waste throughout the landfill’s operating life to comply with
regulations. Typically, regulations stipulate the function of the cover material speci-
fied (e.g., it must reduce the presence of disease vectors and reduce odors), but the
specific material properties are not identified in regulations. Thus, a variety of mate-
rials have been used by landfill operators and, in some cases, the material consists
276 12 Slope Stability
Fig. 12.6 The effect of low-permeability intermediate cover soil on slope stability, compared to
cover soil with the same permeability as waste
where a saturated zone from a liquids addition device intercepted the surface, and
one where no such layer was present. A modest decrease FOS in the simulation with
the low permeability liner on the side slope was observed. The effect of increased
injection pressure was evaluated; even though the FOS did decrease with increased
injection pressure, even at very high pressures, the FOS did not decrease a dramatic
amount for the scenario simulated.
The impact of two contrasting liquids addition strategies was also evaluated. In
one strategy, liquids are added to the landfill at a lower flow rate (and thus pressure)
continuously. In the other strategy, the same total volume of liquids is added, but in
distinct intermittent “pulses” at higher flow rates (and pressures) over short periods
of time. The resulting simulations found that while the FOS decreased under the
“pulsed” strategy, the decrease was not dramatic. As a comparison, simulations of
the elevated “pulsed” injection pressures on a continuous basis resulted in a dra-
matically reduced FOS. These results support that although high pressure liquids
addition reduces FOS, when practiced intermittently with appropriate recovery time
between liquids addition events, high pressures may be safely utilized, even beneath
the side slope.
Several additional factors should be considered by the design engineer when
performing slope stability analyses or interpreting the results of such analyses. Most
commercial models allow the designer to integrate water flow (and resulting pore
pressures) and slope stability analysis. In reality, fluid flow in a landfill will consist
of multiple phases and gas pressures have the potential to contribute to pore pres-
sures (Merry et al. 2006). Most commercial modeling software will not consider gas
contributions. As described in Chap. 5, in addition to landfilled MSW being aniso-
tropic with respect to hydraulic conductivity, the hydraulic properties across the
depth of a landfill are not constant (i.e., the hydraulic conductivity decreases as the
depth within the waste increases). Xu et al. (2012) examined the potential impact of
this occurrence by simulating slope stability with pressured liquids addition beneath
the slope under conditions where the waste had decreasing hydraulic conductivity
with increasing waste depth. The results showed that when hydraulic conductivity
reduction with depth was greatest, the FOS decreased to a larger extent.
Multiple conclusions can be drawn from the modeling exercises described above.
First, pressurized liquids addition, even when performed under the side slope at high
pressures, does not necessarily result in a slope stability concern. The key design and
operational challenge to minimize potential slope concerns is to avoid the excessive
buildup of pore pressure. From an operational perspective, this can be accomplished
by maintaining and monitoring the LCRS, avoiding the creation of low permeability
zones within the landfill where leachate can become perched, and allowing appropri-
ate time in between large pressure liquids addition events. Appropriate design of the
LCRS is crucial (Chap. 10), as this is perhaps the most critical element in the landfill
that must function to avoid a slope failure. The designer should also consider com-
plicating factors in the design and simulation process such as elevated pore pressures
due to gas (a properly functioning GCCS will help reduce this potential concern) and
the decrease in waste hydraulic conductivity with depth.
278 12 Slope Stability
The previous sections described methods for performing slope stability assessments
for specific operational conditions and the factors that should be considered most
important. One of the lessons learned is that liquids addition may in many cases be
accomplished without a major slope stability concern even when the device is
located close to or under the side slope. To provide guidance to the designer and the
regulating engineer as to an appropriate setback distance (defined as the distance
from the side slope to a liquids addition device), Xu et al. (2014) created a series of
design charts that indicate minimum setback distance for liquids addition devices
and injection pressures. Three different subsurface liquids addition methods were
evaluated: horizontal injection trench, horizontal blanket, and vertical well.
Figures 12.7, 12.8, and 12.9 provide setback distance design charts for horizontal
trenches, horizontal blankets, and vertical wells, respectively. In each case, the land-
fill was modeled as having 3H:1V side slopes with a waste friction angle of 25°, a
cohesion of 5 kPa, a unit weight of 7.8 kN/m3, a lateral hydraulic conductivity of
10−5 cm/s, and a vertical hydraulic conductivity of 10−6 cm/s (Xu et al. 2012 can be
used to assess the potential difference in FOS for other conditions). The modeled
horizontal devices (trench and blanket) were located at an elevation of 30 m above
the base of the liner in a 50 m deep landfill; setback distances were modeled as the
distance from the side slope to the device (see Figs. 12.7 and 12.8). Allowable injec-
tion pressure is presented as a function of required setback distance needed to main-
tain the selected FOS (1.0, 1.2, 1.5). The modeled vertical well was located on the
side slope and was provided with a screen length of 10 m. The allowable depth of
liquid above the top of the screened well section is presented as a function of
required setback distance needed to maintain the selected FOS (1.0, 1.2, 1.5).
Fig. 12.7 Injection pressure as a function of setback distance for a horizontal trench
12.5 Design Recommendations for Slope Setback Distance 279
Fig. 12.8 Injection pressure as a function of setback distance for a horizontal blanket
Fig. 12.9 Liquid level above the top of vertical well screen as a function of setback distance
The results presented in Figs. 12.7, 12.8, and 12.9, as well as the example simu-
lations earlier in the chapter, provide the designer a sense of the results that can be
obtained with slope stability modeling coupled with pressurized liquids addition. A
site-specific slope stability analysis should be included as part of any landfill design
involving liquids addition; as new techniques are developed to account for factors
not evaluated here, such as the influence of gas flow, the designer should consider
these. The results presented in this chapter suggest that liquids addition under pres-
sure can occur without compromising slope integrity, but other site conditions such
as LCRS operation, perched liquid levels due to low permeability layers, and
280 12 Slope Stability
References
Abramson LW, Lee TS, Sharma S, Boyce GM (2002) Slope stability and stabilization methods.
Wiley, New York
Blight G (2008) Slope failures in municipal solid waste dumps and landfills: a review. Waste
Manag Res 26:448–463
Das B (2005) Fundamentals of geotechnical engineering, 2nd edn. Thomson Canada, Toronto
Duncan JM, Wright SG (2005) Soil strength and slope stability. Wiley, New York
GeoSlope International Ltd (2007) GeoStudio, version 7, build 4840. Geo-Slope International,
Calgary
Gunaratne M (2006) The foundation engineering handbook. Taylor & Francis, Boca Raton
Hendron DM, Fernandez, G, Prommer PJ, Giroud JP, Orozco LF (1999) Investigation of the cause
of the 27 September 1997 slope failure at the Dona Juana landfill. In: Proceedings of the
Sardinia’99 seventh international waste management and landfill symposium, Cagliari,
p 545–554
Koerner R (2005) Designing with geosynthetics. Pearson Education, Upper Saddle River
Koerner RM, Soong TY (2000) Leachate in landfills: the stability issues. Geotext Geomembr
18(5):293–309
Krahn J (2007) Stability modeling with SLOPE/W 2007: an engineering methodology, 2nd edn.
GEO-SLOPE International, Calgary
Landva A, Dickinson S (2012) Landslides in landfills. Int Soc Soil Mech Geotech Eng Bull
6(1):10–18
Liu C, Evett JB (2001) Soils and foundations. Prentice Hall, Upper Saddle River
McCreanor PT, Reinhart DR (1999) Hydrodynamic modeling of leachate recirculating landfills.
Waste Manag Res 17:465–469
Merry SM, Fritz WU, Budhu M, Jesionek K (2006) Effect of gas on pore pressures in wet landfills.
J Geotech Geoenviron Eng 132(5):553–561
Pockoski M, Duncan JM (2000) Comparison of computer programs for analysis of reinforced
slopes. Center for Geotechnical Practice and Research, Virginia Tech, Blacksburg
Stark TD, Eid HT, Evans WD, Sherry PE (2000) Municipal solid waste slope failure. II. Stability
analyses. J Geotech Geoenviron Eng 126(5):408–419
Thiel RS, Christie M (2005) Leachate recirculation and potential concerns on landfill stability. In:
Proceedings of the NAGS 2005/ GRI 19 conference, Geosynthetica.net, Jupiter
Xu Q, Tolaymat T, Townsend T (2012) Impact of pressurized liquids addition on landfill slope
stability. J Geotech Geoenviron Eng ASCE 138(4):1–9
Xu Q, Jain P, Tolaymat T, Townsend T (2014) Setback distance for landfill liquid addition devices:
Slope stability considerations. Unpublished manuscript
Chapter 13
Landfill Gas
Fig. 13.2 LFG production curves for a traditional facilities and a facility operated to promote
rapid waste stabilization
As discussed in Chap. 2, the methane (CH4) and carbon dioxide (CO2) gases
produced as a result of biological waste decomposition, along with nitrogen and
oxygen from entrapped air and trace gases emitted from the waste, are collectively
referred to as LFG or biogas. GCCS are required for a variety of reasons: safety
concerns (e.g., preventing formation of explosive mixtures in the atmosphere), envi-
ronmental protection (e.g., reduction of the emission of toxic constituents and
greenhouse gases), reduction of nuisances (e.g., odors), and regulatory require-
ments (discussed later in this chapter).
Active LFG collection systems employ a series of collection devices or wells
(vertical or horizontal) connected through one or more common header pipes.
Vertical wells are commonly installed in areas of the landfill that have reached a
desired waste depth (typically 30 ft or more); whereas horizontal LFG wells may
be installed as early as the first lift of waste. Vertical and horizontal wells each
have advantages and disadvantages in terms of cost, ease of installation, and
performance.
Vertical wells for gas collection are installed using a large-diameter drill rig
(e.g., an auger drill rig) that bores through the waste, creating holes that typically
have a 0.6–1 m (2–3 ft) diameter (Fig. 13.3). Perforated or slotted piping, primarily
polyvinyl chloride (PVC) or high-density polyethylene (HDPE), is placed in the
center of the borehole and surrounded by a backfill of permeable material such as
rock (Figs. 13.4 and 13.5). The perforated or slotted pipe transitions to a solid pipe
near the landfill surface as a means of reducing the introduction of air into the well
or waste once the well construction is finished and a vacuum is applied to the well.
284 13 Landfill Gas
Fig. 13.3 Bucket auger rig for excavating borehole for LFG well
Fig. 13.4 Placement of slotted well pipe into excavated borehole (Photo courtesy of Jones
Edmunds)
Fig. 13.5 LFG well under construction, including protective grate and well pipe; the contractor is
measuring the depth of the granular fill surrounding the pipe (Photo courtesy of Jones Edmunds)
For vertical and horizontal systems, the individual collectors are connected to a
common header pipe (and manifold) which routes the collected gas to a control sys-
tem where the LFG is combusted or otherwise processed.
The primary driving force causing LFG to exit a MSW landfill is pressure, and
while sufficient gas pressures develop within waste to cause gas to migrate to a well,
engineered strategies are normally needed which include applying a vacuum to the
well to increase extraction efficiency (referred to as an active GCCS). In the absence
of applied vacuum (a passive GCCS), a larger fraction of the gas will migrate to
the landfill surface or side slopes and escape to the atmosphere. The vacuum in
active GCCS is created using mechanical blowers at one or more control locations;
piping that joins individual LFG collection devices, thus allowing vacuum avail-
ability at desired points, is routed to the blower system and is often referred to as
header piping. Gas wells are connected to the header piping via a flexible hose that
can accommodate settlement of the landfilled waste and pipe movement, though
sometimes intermediate piping (sometimes referred to as lateral piping) can be used
to connect the individual well to the header piping. Control of the applied vacuum
and resulting gas extraction is achieved by using a well head that includes a control
valve and devices that provide for the measurement of flow, pressure, and tempera-
ture. Figure 2.7 provided a photo of a gas well head and pertinent features; Fig. 13.6
illustrates a typical cross section of a LFG well including the wellhead and the con-
nection to the gas collection header. Figure 13.7 shows the construction of a gas
collection header.
286 13 Landfill Gas
LFG is typically saturated with water vapor and is produced at a temperature that
is usually warmer than ambient conditions. As a result, water condenses in the col-
lection piping. This gas condensate must be removed from the collection pipes at
points of low elevation to avoid blocking the LFG system with liquid. The design of
typical GCCS includes minimizing the number of low points in the header pipe.
Depending on the site configuration and design, condensate removal points or
knock-outs will be placed at various locations within the well field and at points
outside of the landfill footprint and at the blower extraction area.
LFG collection systems require routine operation, maintenance, and perfor-
mance optimization. This is normally completed by a trained operator equipped
with a portable meter capable of measuring gas pressure, composition, and flow
(Chaps. 15 and 16). The operator must adjust valve settings on the well heads at
various extraction points to maintain sufficient vacuum to provide gas collection
without creating conditions where air is pulled into the landfill or piping system
(a potential cause of fires or explosions). In cases where LFG extraction is con-
trolled by automated systems with set points that adjust vacuum to individual gas
collectors, an operator still has a role in operating and maintaining these automated
systems.
13.2 LFG Generation, Control, and Design fundamentals 287
Chapter 5 introduced techniques for modeling gas flow in landfills; the most com-
mon gas flow modeling performed as part of the standard design process is a predic-
tion of the rate of gas generated (this is the amount that would be emitted from the
landfill surface or surrounding soils in the absence of any control measures). The
typical approach to modeling LFG production is to approximate waste decomposi-
tion as a first-order decay reaction, with gas produced in a volume proportional to
the mass of waste decomposed. The relationship for the rate of gas production from
a unit mass of waste as a function of time may be expressed as:
G ( t ) = 2 L o kM o e - kt (13.1)
where, G(t) = LFG production (m3 year−1) at time t (year); Lo = the CH4 generation
potential (m3 CH4 Mg−1 solid waste); k = the CH4 generation rate constant (year−1);
288 13 Landfill Gas
Fig. 13.8 Illustration of first-order LFG production model for five batches of waste
and Mo = the mass of solid waste in the batch (Mg). The factor of 2 is based on the
assumption that the landfill comprises 50 % CH4 and 50 % CO2; this can be changed
if a different biogas composition is anticipated.
Estimated gas volumes produced from individual batches of waste deposited in a
landfill over time are normally summed to estimate the composite LFG production
rate for the entire landfill using a relationship such as follows:
n 1
æM ö - kti , j
QCH4 = å å kLo ç o ÷e (13.2)
i =1 j = 0.1 è 10 ø
where, Q CH4 = CH 4 generation at time t (m3 year−1), i is a 1-year time increment,
j is a 0.1-year time increment, and the other values remain the same as previously
defined. The designer can use estimates of annual waste disposal amounts over the
predicted life of a landfill to predict the amount of gas produced, both during the
operational years of a landfill, as well as in the years following landfill closure.
Figure 13.8 illustrates this technique, predicting LFG generation for landfill dispos-
ing waste for 5 years at a rate of 100,000 Mg per year (k = 0.08 year−1, Lo = 100 m3
CH4/Mg). The cumulative LFG production curve is shown, as well as the individual
waste batches used to produce this curve (one batch for each year of waste).
13.3 Design and Operation Challenges 289
The model described above is the same as that provided in the US federal
r egulations for LFG emissions, the US AP-42 guidelines for gas generation (US
EPA 1998), and the US EPA’s LFG Emissions Model (US EPA 2005). While they
may differ with respect to complexity and number of parameters included, most
major models of LFG production are based on the first-order decay concept (e.g., the
Intergovernmental Panel on Climate Change’s Waste Model and several region- and
country-specific LFG models [Central America, China, Columbia, Ecuador, Mexico,
Philippines, Thailand, and Ukraine) developed by the US EPA’s Landfill Methane
Outreach Program (LMOP 2014)].
Adding liquids in a controlled manner to a MSW landfill, whether during the land-
fill’s operation or after the landfill has been filled, can pose several challenges to the
design and operation of the LFG collection system. Table 13.1 presents a summary
of the major design considerations related to GCCS design for landfills where
liquids are added.
A major impact of liquids addition at landfills is the increased LFG generation
that affects several components of the GCCS design. For instance, the modeling of
LFG production differs from a conventional landfill since the rate of LFG genera-
tion will be increased and the amount of liquid added will impact the projected LFG
generation rate. The designer of a GCCS system at a landfill that adds liquids must
also consider the impact the increased LFG generation rate has on the timing of the
GCCS construction. Ideally, the GCCS elements in areas where liquids are added
will be in place before liquids addition commences, since liquids addition increases
the LFG generation rate. Early installation of GCCS can capture the must of addi-
tional LFG generated, thus reducing emissions and enhancing the viability of a
beneficial use project to take advantage of the increased LFG quantity. The remain-
der of this section examines issues with accelerated gas production and increased
liquid levels in the waste.
The standard LFG modeling approach is appropriate for a landfill operated conven-
tionally, but the methane generation rate constant, k, will increase at sites where
liquids addition is practiced. To appropriately model gas production at wet landfills,
the designer can utilize the standard LFG modeling approach, but should adjust as
necessary to account for the increase in LFG generation rate, the fact that some parts
of the landfill may be wetted while others may not, and the potential that liquids
addition may not commence until several years after gas production has already
begun.
Table 13.1 Landfill design elements impacted by increased LFG generation at landfills that add liquids
Design category Example Specific considerations for landfills that add liquids
LFG collection LFG production modeling Increase in LFG generation rate
system Accounting for wetted fraction of waste
Collection system header and lateral Collection pipes must accommodate higher gas flows during and after liquids addition
piping operations commence
Collection system wells Selection of devices compatible with additional liquids in the landfill and increased
settlement of the landfill surface. May be required to incorporate individual well
pumping system design
Consider designs to incorporate LFG collection at the landfill surface beneath an EGC
Placing LFG collectors in a manner consistent with planned recirculation devices
Collection system control devices Selection of blower/extraction system(s) consistent with anticipated LFG generation
and collection rates
Collection system operation Timing of collection system construction
Timing of collection system operation
Recordkeeping for performance assessment and regulatory compliance
Collection system monitoring Frequency of measuring LFG flows
Frequency of measuring surface emissions
Design for additional monitoring devices (e.g., thermocouples placed in situ)
Condensate management system design Design condensate management system to account for higher condensate generation
LFG beneficial use system design Account for timing of sustainable landfilling operation and increased gas generation
and collection
Account for anticipated gas flows when selecting end use for collected LFG
Leachate Layout of pipes, cleanout lines Integrate LFG collection into leachate collection design
collection system Operations monitoring to assess quality of gas from leachate collection system
Liquids addition Specification of device location and Selection of liquids addition devices
system spacing Selection of function of devices (liquids addition only, use for LFG collection then
liquids addition, etc.)
Cover system Placement of daily and intermediate cover Selection of materials compatible with liquids addition and LFG collection
Timing and placement of covers
13.3 Design and Operation Challenges 291
Fig. 13.9 Gas production for a landfill with 5 years (1 year per batch) of waste placed subject to a
change in k at year after the last batch of waste has been placed
The standard LFG production modeling approach uses one k value for a given
waste batch. At some sites, however, liquids addition may not be initiated for several
years; during this time gas production occurs at a lower rate initially and increases
later. The standard first-order modeling approach can be modified to simulate waste
prior to (k1) and after (k2) liquids addition. The term tc is defined as the time when
the rate constant would change from initial (conventional) conditions to accelerated
conditions.
For the time period that k1 is in effect (0 < t < tc), the gas production relationship
is as follows:
G ( t ) = 2 L o k1 M o e - k1t (13.3)
when t > tc, the following relationship applies.
( )
G ( t ) = 2 L o k 2 M o e - k1t c e
- k2 ( t - tc )
(13.4)
Fig. 13.10 Example LFG production model for 10 years of disposal for (i) landfill with k = 0.04,
(ii) landfill with k = 0.3 year−1, and (iii) landfill with k = 0.04 year−1 for the first 5 years of each batch
and k = 0.3 year−1 for times after first 5 years for each batch
landfill over a 5-year period, but at year 5, the k value transition from 0.04 to
0.3 year−1. The bottom part of Fig. 13.9 presents the individual batches while the top
part presents the total gas production. This example could be used to forecast gas
production from a landfill where gas if added to a landfill for 5 years under standard
(dry) conditions, and then at the end of waste placement (corresponding to the
end of year 5), liquids addition is commenced and gas production is enhanced.
The designer could use this technique to transition different areas of the landfill to
wetted conditions at different times.
Figure 13.10 further illustrates this approach for a facility where waste is
disposed of for 10 years (each year of waste is modeled as one batch). At the end of
5 years of waste placement, the k for the entire facility (each batch of waste) changes
from 0.04 to 0.3 year−1. This process simulates the scenario where liquids addition
commences in a given area of the landfill after a specified time period, with the
entire landfill being operated in this after for the remainder of landfill operation.
Shown for comparison purposes are the gas production curves that result when a k
of 0.04 year−1 is used throughout and a k of 0.3 year−1 is used throughout.
Another factor to consider when modeling gas production at landfill practicing
liquids addition is that not all of the waste will be wetted; moisture distribution may
only be limited to certain parts of the landfill. Assuming the fraction of landfilled
waste in a batch that is wetted as w, the following equations can be used to estimate
the gas production from a batch of solid waste that is exposed to liquids at time tc.
For time 0 < t < tc:
G ( t ) = 2 L o k1 M o e - k1t (13.5)
294 13 Landfill Gas
( )
G ( t ) = 2 L o k1 M o e - k1t (1 - w ) + 2 L o k 2 M o e - k1t c e
- k2 ( t - tc )
w
(13.6)
Using this approach, the designer can estimate landfill gas production for facilities
where only part of the waste is wetted, and where liquids addition in the wetted
areas occurs after a specified period of time.
In regulatory jurisdictions with well-developed rules for landfill design and opera-
tion, regulations for LFG collection and control are typically included. At the most
basic level, LFG control is required to prevent the off-site migration of LFG through
13.4 LFG Regulations for Bioreactor Landfills 295
Fig. 13.11 Ballooning of geomembrane at the landfill surface as a result of gas pressure
Fig. 13.12 Illustration of perched liquids impacts on leachate levels in vertical LFG wells
296 13 Landfill Gas
soils and the formation of explosive gas conditions in adjacent structures and
beyond the property boundary. Some rule programs may also be structured around
reducing atmospheric emissions, both for air quality in the site vicinity (e.g., odors,
harmful chemicals) and to address regional-scale concerns (ozone precursors,
global warming). Chapter 3 provided a review of basic landfill regulatory require-
ments, including aspects related to planning of sustainable landfill practices.
Some regulatory requirements have been developed to address the potentially
greater amounts of LFG produced at landfill sites that add liquids. In the US, LFG
collection and control are addressed as part of the National Emissions Standards for
Hazardous Air Pollutants (NESHAP): MSW landfills (Code of Federal Regulations
2003). Under the US NESHAP rules, a “bioreactor” landfill is defined as an MSW
landfill or a portion of a MSW landfill where any liquid other than leachate (includ-
ing LFG condensate) is added in a controlled fashion into the waste mass (often
in combination with recirculating leachate) to reach a minimum average moisture
content of at least 40 % by weight to accelerate or enhance anaerobic biodegrada-
tion of the waste. Attaining a 40 % moisture content triggers the LFG collection
and control system requirements, not simply the addition of liquids other than
leachate or gas condensate. Table 13.3 presents rules specific to bioreactor landfills
in the US.
Internationally, specific regulations defining sustainable landfills are somewhat
limited, although cases have been reported whereby sustainable landfilling tech-
nologies are used to meet specific regulatory requirements. For example, Woelders
et al. (2007) discussed a bioreactor test cell that was operated to meet the EU
Landfill Directive’s definition of an inert waste landfill through examination of
leachate chemical quality following recirculation. Additional international experi-
ences are discussed in Chap. 4.
GCCS bioreactors constructed after November 7, 2000 (NMOC) threshold for conventional landfills does not
apply in the case of bioreactor landfills
Collection and control requirements must be in
accordance with the specifications provided in 40 CFR
60.759 or an alternate collection and control design
plan submitted for approval
Operation of GCCS Within 180 days after the landfill initiates Moisture content is evaluated based on in-place waste,
liquids addition or 180 days after the bioreactor initial moisture content, and amount of moisture added
has reached 40 % moisture content (by weight), The calculation of moisture content must be conducted
whichever is later in accordance with the procedures listed in 40 CFR
63.1980 (g) and (h) section 63.1980 (g) and (h) to
determine when the 40 % moisture content is reached
297
298 13 Landfill Gas
The techniques in the previous section provide the engineer with the ability to
predict the enhanced gas generation rate at landfills practicing accelerated waste
stabilization techniques. Flow measurement devices at wellheads may need to be
larger (larger openings in orifice plates, large flow straightener pipe diameters with
pitot tubes) to accommodate greater than normal flow rates. Pipes that route the
collected LFG from individual collection devices (the manifolds and headers)
to control points must be sized to handle the maximum LFG generation rate. The
design and sizing of the control devices (i.e., blower systems and destruction
devices) must consider a potentially greater maximum LFG generation rate so that
the collection system can handle the maximum quantity of gas expected.
The designer should incorporate infrastructure to handle the added volume of
gases and liquids likely to be encountered in the GCCS to allow the GCCS to effi-
ciently collect the gas that is produced. In addition to more condensate as a result of
enhanced gas flow, the gas pressures in some gas collection devices will be suffi-
ciently large that leachate is expelled from the landfill into the collection gas collec-
tion header; this liquid may have a much greater suspended solids content compared
to condensate. Slopes and pipe sizes should be sufficient to handle the added GCCS
liquid volumes; the piping system must be properly designed to ensure that liquids
do not cause a blockage in the GCCS piping. Additional condensate knockout and
drainage locations may be required to handle the added liquid volumes. Additional
drainage features such as toe drains may be required, which can be integrated into
the GCCS (they also serve to assist in seepage management, see Chap. 11).
As discussed previously, a borehole must be drilled into the waste to install vertical
wells. This procedure usually entails placing a PVC well in the borehole, backfilling
the annulus with a permeable material, and then sealing it near the surface. By vir-
tue of the installation method, liquids within the landfill (e.g., condensation result-
ing from collecting the LFG, leachate within the waste matrix) tend to migrate to the
boreholes during and after well installation. Consequently, the well’s perforated or
slotted portion can become filled with liquid, thus blocking some or all of the
screened portion intended for LFG collection. Pumps that are specifically designed
to remove liquids may be used (see Fig. 13.13), which necessitates increased opera-
tion and maintenance and overall costs. Again, this is often an issue at conventional
landfills and the problem is exacerbated at sustainable landfills.
Experience with vertical wells used for liquids recirculation has shown that it is
unlikely that these wells could be used as the primary LFG collectors. Normally,
recirculated liquids saturate the zone surrounding the well and may also fill up part
or all of the screened area. Often, this standing liquid slowly decreases over time
13.5 Design and Operation Strategies 299
and only a limited amount of screened interval is available later for LFG collection.
Thus, facilities that employ vertical wells for leachate recirculation must have addi-
tional devices that will be used to collect LFG.
As with vertical wells, the challenge with collecting gas from horizontal trenches at
landfills practicing liquids addition or with inherently high liquids levels is the high
moisture content of the surrounding waste. As detailed in Chap. 9, a horizontal col-
lection is constructed as waste is deposited in the landfill as trenches are excavated
in the waste. These trenches are filled with highly permeable backfill material and a
perforated or slotted pipe is embedded within the fill material. For horizontal sys-
tems, the pipe can either act as a LFG collection device or as a device to deliver
liquids to the permeable filler material, which then distributes the liquids to the sur-
rounding waste. If the purpose is gas extraction, the presence of liquids can cover
perforations of the LFG collector, thus impeding the ability to capture gas, just as
with vertical wells. This can ultimately result in additional and potentially costly
operation and maintenance.
Many attempts to collect gas from horizontal trenches used for liquids addition
have not been successful; once sufficient liquids have been added to the device, gas
collection becomes problematic (Townsend et al. 1994). This results from the satu-
rated conditions surrounding the trench; although gas generation is enhanced and
gas pressures are often quite large, gas migration through the saturated waste to the
trench for extraction is not the path of least resistance, and thus gas migrates away
300 13 Landfill Gas
from the device to other parts of the landfill. Alternatively, operation of horizontal
devices first as LFG collectors followed by liquids addition devices may allow a
given device to serve two purposes.
Figure 13.14 shows an operational scenario that involves installing a series of
horizontal trenches in each lift of waste. The initial horizontal devices that are
depicted in Fig. 13.14a are part of the LCRS (the gravity drain lines; gas collection
with the LCRS is discussed in greater detail in an upcoming section). Throughout
the operating life of the landfill, the LCRS trenches collect both leachate and LFG;
gas collection begins when sufficient waste has been placed above the LCRS. Once
the landfill is active and the first lift of waste is completed, a set of horizontal
trenches will be constructed, as shown in Fig. 13.14b (following same techniques as
described in Chap. 9). Initially, the horizontal trenches will not be used for gas
extraction, or for leachate recirculation. Once a layer of waste has been placed and
compacted on top of the trenches above the first lift of waste (to prevent air intru-
sion), the pipes within these horizontal trenches will extract LFG.
Upon completion of the second lift of waste (Fig. 13.14c) another series of hori-
zontal trenches will be constructed. When the horizontal trenches above the second
lift of waste are completed and covered with a layer of compacted waste (Fig. 13.14d)
the function of the trenches in the first lift of waste will transition to adding liquids,
while the trenches in the second lift of waste will collect gas. Ultimately, the gas
collection will always take place in the upper-most horizontal trenches to decrease
the issues posed by excess liquids in the landfill. Accordingly, as indicated in
Fig. 13.14e, when the third lift of waste is filled and the horizontal trenches are
complete (and sufficiently covered with compacted waste), they will function as gas
collectors, while the trenches within the second and first lifts of waste will serve as
liquids addition devices.
Once the horizontal trenches are used for liquids addition, they are typically no
longer effective for LFG collection. As depicted in Fig. 13.14, as a landfill is built,
newly-installed pipes near the surface are used for LFG extraction while previously-
installed trenches are used for liquids addition. Eventually, LFG is collected only
from the LCRS and the uppermost horizontal trenches. The operator may still desire
to occasionally draw gas from wetted trenches, but they should be prepared for
greater than usual liquids extraction and possibly very limited gas removal.
Gas will migrate from the landfill under pressure. If a well or a trench is present,
especially if it has been placed under vacuum, gas will migrate to these locations.
One region of the landfill where some gas will typically always migrate to is the
LCRS. Plumbing the LCRS for LFG collection is critical at landfills where liquids
are added because of the aforementioned difficulty in collecting LFG with tradi-
tional GCCS devices in deeper sections of waste that are dense and wet. The nature
of the LCRS (which is intended to quickly drain liquids across the entire landfill
13.5 Design and Operation Strategies 301
Fig. 13.14 Illustration of staged use of horizontal trenches for liquids addition and gas collection
(a) Initial landfill LCRS; (b) after first lift placement, collection of gas from LCRS (continues
throughout) and construction of horizontal trench layer 1; (c) liquids addition into layer 1, con-
struction of layer 2; (d) extraction of gas from layer 2; (e) liquids addition into layers 1 and 2, gas
collection from layer 3
302 13 Landfill Gas
Fig. 13.15 Example of gas escape from LCRS through manholes or pump stations
Fig. 13.16 Manhole of LCRS sealed with HDPE geomembrane and plumbed for gas collection
footprint via gravity) lends itself well to LFG collection, especially if dedicated
p iping is designed with the intent of LFG collection from the beginning (Townsend
and Miller 1997).
Figure 13.16 illustrates the use of a conventional external leachate collection
pumping system for LFG collection. It is common for landfill operators to notice
gas buildup in the pumping stations and manholes. As portrayed in Fig. 13.15, gas
migrates through the primary leachate drain into the manhole. If the manhole is
appropriately designed from the beginning for gas collection (e.g., gas-tight covers,
13.5 Design and Operation Strategies 303
Fig. 13.17 Maintaining leachate levels in external manholes to promote gas collection from
extraction points within the landfill
gas extraction vents), each manhole can serve as a gas extraction point. It is quite
common; however, that such systems are not designed from the beginning with such
practice in mind, and the challenges in retrofitting these locations to be gas-tight
have prompted some owners and operators to find other solutions. For example,
Fig. 13.16 shows a manhole that was completely encased in a geomembrane to
facilitate gas collection. Retrofitting manholes for gas extraction can be accom-
plished, but consideration for this function should be explored in the initial LCRS
design.
Figure 13.17 illustrates an alternative approach. The liquid levels in an external
manhole (or pumpstation) are raised so that the migration pathway for LFG into the
manhole is cut off. Pump stations will normally be equipped with on/off switches
that are triggered by the depth of water in the manhole, and these can be adjusted
to effectively isolate the headspace of the manhole from the interior of the landfill.
A gas vent at some location connected to the LCRS inside the landfill must be pro-
vided as an extraction point for the gas. Retrofitting LCRS cleanout is a common
practice, though again, design this function from the beginning would be most
efficient.
Chapter 11 discussed the use of toe drains to help control seeps at the bottom of
the exterior slopes of above-grade landfills, and described how these devices can
also provide an effective extraction point for the collection of LFG. Toe drains will
typically be connected to the LCRS to provide necessary drainage. Design and con-
struction of a toe drain as an element in the GCCS provides another promising
method for LFG collection; once sufficient waste is placed, a low vacuum can be
created on the toe drain to collect gas that accumulates at the landfill perimeter and
that which may be exiting the landfill from the LCRS. It is also possible for the
toe drain GCCS element to be integrated into the overall LCRS gas collection
infrastructure. Figure 13.18 illustrates gas collection from a landfill toe drain that is
integrated with the facility’s LCRS.
304 13 Landfill Gas
Fig. 13.18 Gas extraction from a leachate toe drain connected to the LCRS
Fig. 13.19 Horizontal gas collection trench (perforated pipe bedded in stone) installed on top of
the LCRS drainage layer (Photo courtesy of Jones Edmunds)
Even without considering gas collection, most LCRS will still facilitate gas
c ollection. For many sites, a better approach, especially for wet landfills, is to integrate
gas collection into the original design of the LCRS as a separate component. This
can be accomplished in several ways. Figure 13.19 shows a site where a horizontal
gas collector was installed on top of the LCRS sand drainage blanket. A perforated
HDPE pipe was installed in a shallow trench in the LCRS sand; permeable stone
13.5 Design and Operation Strategies 305
Fig. 13.20 Horizontal gas collector (perforated pipe blanketed in a geocomposite) placed on top
of the LCRS drainage blanket
was then mounded around the pipe and covered with more sand. These gas collection
pipes were then routed to the side of the landfill and connected to wellheads for later
gas extraction. Figure 13.20 shows another approach, where perforated pipes were
wrapped in a geocomposite placed on top of the sand drainage blanket of the land-
fill’s LCRS. Similarly, these pipes were routed to the side of the landfill for eventual
connection to the site’s GCCS.
Fig. 13.21 Construction of surface trenches for gas collection to be placed under an exposed
geomembrane cap
vacuum to be induced on devices within the waste mass. The EGC can also assist in
managing liquids on side slopes by mitigating liquid or leachate seeps that occur
and allowing the seeps to be routed to a toe drain system (see Chap. 11). Chapter 17
discusses the pros and cons of using an EGC as a sustainable landfill practice.
Care during construction, operation, and maintenance is required with this type
of system (just as with a traditional LFG collection system). Several specific design
and construction issues must be considered at landfills that employ a temporary
geomembrane. The temporary EGC should be constructed over all areas with LFG
collection and subject to quality control procedures to check integrity of seams for
signs of protrusions through the cap. Wellheads connected to extraction points
should include seals or gaskets that reduce air intrusion; in fact, wellhead connec-
tions should be minimized. Checks for tears in the cap or at seams must be rigor-
ously conducted.
As described earlier, a major operational issue with gas collection systems (at land-
fills in general and potentially more so at landfills implementing sustainable
practices) involves additional liquids impeding LFG collection; LFG collection
devices can become flooded, significantly reducing the efficiency of the GCCS.
13.5 Design and Operation Strategies 307
Another problem with modern LFG collection is that the waste surface settles after
a landfill is filled, which can cause the GCCS wells to shift, ultimately resulting in
surface seeps and breaks in the GCCS. This can cause the vacuum system to intake
ambient air, and these problems require frequent maintenance by landfill operators.
An alternative LFG collection system may be used in which extraction pipes are
installed before the waste is placed and plumbed in a way that LFG collection may
commence during the early stages of landfill development and so that LFG can be
collected concurrent to liquids removal. As the landfill is filled up, the pipes that
were installed originally will be continually extended upward until the landfill has
reached its final grade or elevation. A similar technique has been practiced at some
landfills with traditional GCCS, where the wells start after the first lift of waste has
been placed (not connected to the LCRS). Figure 13.22 shows the progress of con-
struction of a downward draining gas collection system. In Fig. 13.22a, combination
gas collectors/leachate drains are installed with construction of the LCRS; these
devices are plumbed so that leachate drains to the LCRS while gas can ultimately be
extracted as part of the GCCS. Figures 13.23b–d illustrate the progress of extending
the collectors/drains upward as the landfill is filled. If a gas-tight seal is maintained
at the top, gas collection can occur even as waste operation continues. Upon final
waste filling, the devices can possibly be buried beneath the final cover. This
approach thus has the potential to provide a less operationally intensive GCCS after
completion; the engineer should consider differential waste settlement around the
devices as part of the design and operations plan development.
This method should significantly reduce the issue of LFG collection wells filling
with liquid since both gas and liquids will be collected and extracted through the bottom
of the landfill. Furthermore, as a landfill is filled, the gas collection wells can be extended
vertically to continue collecting gas from new layers of waste. The engineer would need
to provide a design that allows continued extension of the collectors/drains while mini-
mizing interaction with the atmosphere once sealed (preventing future air intrusion).
This approach also offers strong promise to alleviate excess pore water pressures that
can lead to slope stability issues in landfills with added liquids (see Chap. 12).
Figure 13.23 shows a landfill site employing vertical gas collection wells
that begin in the LCRS and that are raised as waste filling progresses; these wells,
however, are equipped with collection devices at the top of the well. A primary
drawback of where wells are constructed as the waste is deposited is the compatibil-
ity with landfilling operations—the site operators would need to ensure LFG wells
are protected from incoming waste trucks and landfill equipment. Furthermore,
operational difficulty may be experienced through the progressive enabling and
disabling of gas collectors as vertical extensions are installed.
Delaying liquids addition is an approach that allows the landfill cell to be filled first
before the addition of liquids. The landfill would then be covered by a low-permeability
cap (EGC, traditional multi-layered cap, or a temporary earthen cover such as clay).
308 13 Landfill Gas
Fig. 13.22 Construction sequence for a downward draining GCCS. (a) Combination gas collec-
tors and leachate drains are installed as part of liner and LCRS construction; (b) waste filling com-
mences; (c) devices are raised as needed; (d) waste filling continues; (e) the devices are buried
under the cover systems and gas collection occurs through the LCRS
13.5 Design and Operation Strategies 309
Fig. 13.23 Example of vertical gas well constructed from the LCRS and extended upward as
waste depth increases
The cap construction would allow for more effective gas collection as well as assist in
liquids management. The primary disadvantage for delaying liquids addition is that
the operator misses out on the LFG generation/recovery from the early years of opera-
tion. Also, as discussed in Chap. 6, since the total volume of liquids targeted for addi-
tion may be large, and since the greatest volume of liquids available for addition
occurs during the operational years of the landfill, delay of liquids addition may have
some other operational consequences. The gas forecasting methods described earlier
in this chapter can be used to estimate the overall fraction of gas that will be collected
under different liquids addition and start time scenarios.
At some point in a landfill’s life, the volume of gas may not be sufficient to warrant
collection for energy recovery, and flaring may require an additional source of gas.
At this point, several other approaches may be required to most effectively reduce
CH4 emissions to the atmosphere. Landfill aeration for older landfills has been uti-
lized frequently in Europe as a means of reducing CH4 emissions; aerobic conversion
of remaining organic matter, along with oxidation of remaining methane, results in
CO2 being the primary exit gas. This process is discussed in detail in Chap. 14.
Another approach allows the remaining gas to vent through an adequately
thick and vegetated cover soil layer to promote biological methane oxidation. Such
activity has been widely documented at landfill sites (Visvanathan et al. 1999;
Christophersen et al. 2001; Barlaz et al. 2004). Design of landfill covers to maximize
310 13 Landfill Gas
their methane oxidation potential has been proposed as a strategy for mitigation of
CH4 emissions, particularly at older landfill sites (He et al. 2007; Rachor et al. 2011).
CH4 oxidation has been reported to be influenced by soil properties such as par-
ticle size distribution, moisture content, soil texture, mineralogy, and porosity, as
well as environmental factors such as barometric pressure, temperature, the pressure
gradient, oxygen availability, microbial population, and vegetation (Visvanathan
et al. 1999; Streese and Stegmann 2003; Borjesson et al. 2004; Barlaz et al. 2004;
Spokas and Bogner 2011). He et al. (2007) reported significant increase in methano-
trophic bacteria population over time in a methane-rich environment and proposed
use of a soil with previous exposure to a methane-rich environment such as soil
reclaimed from old landfill for soil cap. Rachor et al. (2011) used a column study to
evaluate the CH4 oxidation capacity of soils available to site owner for landfill cover
construction. In addition to the soil properties, the type of vegetative cover has also
been reported to influence the methane oxidation rate (Reichenauera et al. 2011).
Bohna et al. (2011) attributed increase in methane oxidation with vegetation cover
to factors such as improved oxygen diffusivity in soils via roots and enrichment
of soils with plant cover. When assessing methane emissions to the environment,
some modeling techniques include a specific term or factor to account for methane
oxidation (IPCC 2006).
References
Barlaz MA, Green RB, Chanton JP, Goldsmith CD, Hater AR (2004) Evaluation of a biologically
active cover for mitigation of LFG emissions. Environ Sci Technol 38:4891–4899
Bohna S, Brunkea P, Gebertb J, Jagera J (2011) Improving the aeration of critical fine-grained
landfill top cover material by vegetation to increase the microbial methane oxidation efficiency.
Waste Manag 31(5):854–863
Borjesson G, Sundh I, Svensson B (2004) Microbial oxidation of CH4 at different temperatures in
landfill cover soils. FEMS Microbiol Ecol 48:305–312
Christophersen M, Holst H, Kjeldsen P, Chanton JP (2001) Lateral gas transport in a soil adjacent
to an old landfill: factors governing emission and methane oxidation. Waste Manag Res 19:
126–143
Code of Federal Regulations (2003) National Emission Standards for hazardous air pollutants:
municipal solid waste landfills. 40 CFR 63 Subpart AAAA, 16
He R, Ruan A, Shen DS (2007) Effects of methane on the microbial populations and oxidation
rates in different landfill cover soil columns. J Environ Sci Health A 42:785–793
IPCC (2006) Waste. In: Guidelines for national greenhouse gas inventories: Intergovernmental
Panel on Climate Change, vol 5. IGES, Japan
Kim H, Townsend T (2012) Wet landfill decomposition rate determination using methane yield
results for excavated waste samples. Waste Manag 32(7):1427–1433
LMOP (2014) International landfill gas models. http://1.usa.gov/1kMhcp3. Accessed 10 Mar 2014
Owens JM, Chynoweth DP (1993) Biochemical methane potential of MSW components. Water
Sci Technol 27:1–14
Rachor I, Gebert J, Grongroft A, Pfeiffer E (2011) Assessment of the methane oxidation capacity
of compacted soils intended for use as landfill cover materials. Waste Manag 31(5):833–842
Reichenauera T, Watzingera A, Riesinga J, Gerzabekb M (2011) Impact of different plants on the
gas profile of a landfill cover. Waste Manag 31(5):843–853
References 311
Reinhart DR, Faour AA, You H (2005) First-order kinetic gas generation model parameters for wet
landfills. Report no. EPA-600/R-05/072. US EPA
Spokas K, Bogner J (2011) Limits and dynamics of methane oxidation in landfill cover soils.
Waste Manag 31(5):823–832
Staley B, Barlaz M (2009) Composition of municipal solid waste in the United States and implica-
tions for carbon sequestration and methane yield. J Environ Eng 135(10):901–909
Streese J, Stegmann R (2003) Microbial oxidation of methane from old landfills in biofilters.
Waste Manag 23:573–580
Tolaymat T, Green R, Hater G, Barlaz M, Black P, Bronson D, Powell J (2010) Evaluation of land-
fill gas decay constant for municipal solid waste landfills operated as bioreactors. J Air Waste
Manag Assoc 60(1):91–97
Townsend T, Miller W (1997) LFG extraction from leachate collection systems. J Solid Waste
Technol Manag 24(3):131–136
Townsend TG, Miller WL, Bishop R, Carter J (1994) Combining systems for leachate recirculation
and LFG collection. Solid Waste Technol 8(4):18–24
US EPA (1996) Turning a liability into an asset: a LFG-to-energy project development handbook.
EPA 430-B-96-0004
US EPA (1998) Section 2.4: municipal solid waste landfills. Chapter 2: solid waste disposal. AP 42,
5th edn, vol 1. US EPA
US EPA (2005) LFG emissions model (LandGEM) version 3.02 user’s guide. US EPA-600/R-05/047
US EPA (2006) Landfill bioreactor performance. Second interim report: outer loop recycling and
disposal facility. EPA/600/R-07/060
Visvanathan C, Pokhrel D, Cheimchaisri W, Hettiaratchi JPA, Wu JS (1999) Methanothropic
activities in tropical landfill cover soils: effects of temperature, moisture content and methane
concentration. Waste Manag Res 17:313–323
Woelders H, Hermkes H, Oonk H, Luning L (2007) From a landfill bioreactor to a sustainable
storage. http://bit.ly/1ga5pJ2. Accessed 10 Mar 2014
Yazdani R, Kieffer J, Sananikone K, Augenstein D (2006) Full scale bioreactor landfill for carbon
sequestration and greenhouse emission control. Final technical progress report. US Department
of Energy (DOE)
Chapter 14
Landfill Air Addition
In some aerobic landfill applications, the operator introduces air only during targeted
periods of a landfill’s operation as a means to meet specific objectives. In other
applications, the operator attempts to maintain aerobic conditions throughout the
Table 14.1 Comparison of aerobic and anaerobic biological conditions
Feature Comparison
Biological reaction Biology in aerobic operations is thought to be more diverse, including bacteria and fungi. Anaerobic biology consists
of a more defined set of microbial groups that are dependent on one another. The aerobic reaction pathway is
generally viewed as more rapid and robust. Anaerobic pathway is sensitive to environmental conditions and may take
time to become established
Energy released and temperature Both reactions are exothermic, but the aerobic reaction releases more energy, due to the more favorable
thermodynamics of micro-organism use of O2 as an electron acceptor. Landfilled waste often reaches temperatures of
60 °C, while temperatures as high 70 °C or greater are often reached in aerobic composting processes. Greater
temperatures result in more rapid reaction rates and pathogen destruction. High aerobic temperatures can contribute
14.2 Achieving Benefits from Air Addition
landfill’s operational life. As a result of more rapid reaction rates and the ability
to more completely transform some chemical constituents, landfill operators can
utilize controlled air addition to meet a number of desired sustainable operation
targets; see Table 14.2 for several of these potential applications.
Some landfill researchers and operators have attempted landfill air addition to
utilize aerobic biological activity as the dominant waste stabilization mechanism,
replacing the anaerobic pathway. Instead of CH4 and CO2 being the dominant gas-
phase decomposition products, an aerobic landfill would have a gas composition
consisting primarily of N2, CO2, and possibly O2. Leachate quality differs in the rate
at which organic strength (BOD, COD) is reduced, as well as other differences such
as pH, nitrogen, and heavy metals.
Several researchers have compared performance and outputs of aerobic and anaer-
obic landfill operation in the laboratory and at pilot scale. Stessel and Murphy (1992)
demonstrated in a set of laboratory lysimeter experiments that recirculating leachate
through simulated landfilled waste while simultaneously adding air resulted in reduced
leachate concentrations of organic compounds and more rapid waste degradation
rates, measured by means of waste settlement. Optimal degradation (maximum waste
settlement) was observed under the minimum moisture content, moisture addition
rate, and air addition rates of 75 %, 0.09 m3/m2-day, and 40,000 m3/m3 water applied,
respectively (Stessel and Murphy 1992). Similarly, Matsufuji et al. (2004) compared
solid waste stabilization in semi-aerobic (often referred to as the “Fukuoka method”;
discussed more later in this chapter) and anaerobic landfill cells at the laboratory
scale, and found that leachate BOD concentrations decreased much faster in the simu-
lated aerobic landfill cells, along with decreased BOD to COD ratios (<0.05 after
3 years of experimentation) and low NH3–N levels as compared to anaerobic landfill
cells. Using data gathered from large scale lysimeters where semi-aerobic, recircula-
tory semi-aerobic, and aerobic conditions were tested, Matsufuji et al. (1993) reported
that aerobic landfill conditions metabolized 72.4 % of the organic waste mass within
10 years as compared with 56.7 % under anaerobic landfill conditions.
Bilgili et al. (2007) utilized four laboratory-scale systems to investigate the effect
of leachate recirculation on aerobic and anaerobic waste degradation and leachate
quality, and observed that conductivity, TDS, and chloride concentrations were
greater under aerobic conditions due to the higher pH values; pH in the aerobic
treatment remained between 8 and 9 after study day 100, in contrast to anaerobic
cells where pH rose steadily from roughly 6 at day 100 to 7.5 on day 500. Air addi-
tion effectively reduced organic matter and ammonia leachate content (Bilgili et al.
2007). In laboratory columns containing a waste stream designed to represent the
composition of fresh MSW, Sartaj et al. (2010) found that aerobic conditions were
effective in reducing the concentration of heavy metals, attributing this to the
adsorption of metals on waste materials and precipitation of metal oxides due to the
increased pH. Kim et al. (2011) operated four waste-packed laboratory columns,
two each under aerobic and anaerobic conditions for a period of 1,650 days, and
observed differences in leachate heavy metal concentrations; some elements were
greater in concentration under the aerobic environment, while others were greater
under anaerobic conditions. Cr(VI) accounted for approximately 45 % of the Cr in
Table 14.2 Potential beneficial applications of air addition to landfills
Application Description
Improving leachate quality Introduction of air into the bottom layers of a landfill (the pipes and drainage stone of a LCRS) stimulates a leachate
treatment zone, particularly for the organic matter
14.2 Achieving Benefits from Air Addition
Primary waste treatment Similar to an aerobic waste composting operation, adding sufficient air to landfilled waste will result in the bulk of waste
decomposition to occur via the aerobic biological pathway
Waste conditioning A brief period of air addition to newly-deposited waste may better prepare waste for subsequent anaerobic decomposition
in that waste temperatures are increased (particularly beneficial in colder climates) and rapidly degradable organic matter
can be stabilized to avoid uncontrolled acid production by anaerobic decomposition conditions
Waste curing The addition of air following anaerobic waste treatment serves to stabilize the residual waste
Nutrient management Air addition is used to biologically transform recalcitrant ammonia nitrogen to nitrate, which can be further denitrified in
other anoxic areas of the landfill
317
318 14 Landfill Air Addition
Fig. 14.1 Differences in pH and COD in landfill leachate from simulated bioreactor landfills (Kim
et al. 2011). One pair was operated aerobically and the other was operated anaerobically
aerobic lysimeter leachate while chromium in the anaerobic lysimeter leachate was
below the detection limit. Kim et al. (2011) found that metal leachate concentrations
decreased significantly in leachate from the aerobic lysimeters as waste stabilized,
while concentrations in the anaerobic columns remained stable. Figure 14.1 pro-
vides the pH and COD for this experiment and illustrates the difference between
these two environmental extremes.
While most biodegradable organic matter can be equally treated through aerobic
and anaerobic pathways (although reaction rates may differ), for some chemical
constituents, aerobic treatment offers treatment capabilities not possible with anaer-
obic systems. For example, the dominant form of N in anaerobic landfill leachate is
ammonia nitrogen (as discussed in Chaps. 2 and 11), and this constituent tends to be
conserved in the landfill over time, and thus increases in concentration, presenting a
treatment challenge (Berge et al. 2005). Using aerobic treatment, ammonia can be
nitrified to nitrate, which denitrifies to N2 gas in a subsequent anoxic step, thereby
removing it from the system (Berge et al. 2006, 2007). This approach has been
examined in several different configurations as illustrated in Fig. 14.2.
Some landfill operators practice external nitrification in tanks, and then recircu-
late the nitrified leachate back into the landfill to promote the anaerobic conversion
of nitrate to nitrogen gas. This approach has been described by some as a hybrid
bioreactor landfill. Other researchers have investigated the potential for adding air
to specific regions within a landfill so that the nitrification step can occur within the
landfill itself (i.e., in situ). Leachate treatment (including ammonia transformation)
14.2 Achieving Benefits from Air Addition 319
Fig. 14.2 Alternative
strategies for promoting
leachate nitrogen removal
by using air addition (a)
external aeration (b) air
addition into the waste mass
(c) air addition into the LCRS
Some landfill operators practice the addition of air early in the active life of the
landfill for a limited period, allowing the bulk of biological treatment to occur
through anaerobic conversion (Rich et al. 2008). Early air addition has been utilized
as a method for increasing the temperature of the waste (a particularly valuable
function of aerobic operation in colder climates), thereby conditioning the waste for
subsequent conversion to an anaerobic environment, and as a means to provide
treatment of readily degradable organic compounds that otherwise might result in
rapid acid formation in anaerobic environments. An additional early-phase air addi-
tion strategy has included the induction of air into surficial regions of landfill (spe-
cifically, recently-added waste lifts) as a means to control CH4 emissions to the
atmosphere prior to LFG collection device installation (Hansen et al. 2002; Jung
et al. 2011). In this system, LFG is extracted into a horizontal collection layer at
the base of the targeted waste lift with the goal of inducing air from the surface of
the landfill into the waste, thus minimizing anaerobic CH4 production. Later, when
additional waste is placed on top of this area, the devices are repurposed as hori-
zontal collectors for anaerobic biogas; air addition piping can also serve a later
purpose as liquids introduction devices for bioreactor landfills.
A common practice, especially in Europe, has been the addition of air to landfills
toward the end of their active life as a method of promoting near-complete stabiliza-
tion of waste that has already undergone anaerobic decomposition. In some cases,
infrastructure for LFG extraction is reconfigured so airflow into the landfill can be
induced. In other cases, wells are added to older landfills for the specific purpose of
air addition. Low-pressure aeration projects have been undertaken extensively in
Germany. The Milmersdorf landfill represents one such case where >90 % of biode-
gradable organic carbon was stabilized via oxidation with active aeration and active
off-gas extraction through wells installed in the waste (i.e., the AEROflott® tech-
nique) (Ritzkowski and Stegmann 2012).
The design of an air addition system includes estimating the volume of air that must
be added (or extracted in an induced system) to meet design objectives, selection of
the type of air addition system (vertical wells and/or horizontal pipes), detailed
specifications on sizing and configuration of the air addition devices, setting spacing
between individual devices, and selection of materials for air piping. Finally, the
design should include specification of other control and monitoring devices such as
pressure and temperature measurement gauges and automated controls (e.g., emer-
gency shut-off valves that engage at a high pressure threshold) as desired. This sec-
tion reviews design objectives, methods for estimating air addition volume
requirements and rates, and air addition system infrastructure.
14.3 Air Addition System Configuration and Design 321
The engineer will consider multiple objectives in the design of a landfill air addition
system. A primary objective will be the conveyance of air to the targeted area of the
landfill. Infrastructure will be required to actively deliver (an active system) or pas-
sively encourage (a passive system) air to the landfill region of interest. In the case
of active systems, mechanical blowers, fans, or compressors must be connected to a
piping network capable of accommodating the desired flow rates to the targeted
addition points. In the case of passive systems, infrastructure (e.g., vents, drains)
must be located and appropriately sized to promote air entry into the landfill based
on temperature gradients.
Integral to the design of the air conveyance system will be the identification of
the target air volume and addition rate so that the infrastructure can be sized appro-
priately. This determination will depend on overall project objectives such as the
purpose of air addition (e.g., primary waste treatment versus targeted waste heating
or curing) and needed performance requirements. In addition to air volumes and
flow rates, appropriate air addition pressures that promote necessary distribution of
air into the waste mass must be considered.
As a result of concerns such as explosive gases and excessive waste heating, it is
critical that the engineer maintain the objective of designing a system that can be
monitored and appropriately controlled during operation. Important monitoring
parameters include gas composition, gas temperature, and waste temperature.
Coupled with monitoring must be a plan and equipment specification for addressing
concerns that may be revealed as an outcome of monitoring. For example, if elevated
temperatures create excessive waste temperatures, the monitoring and operations
plan must include contingency procedures to slow or mitigate the high temperature
conditions.
In a similar manner as the CH4 potential (Lo) for waste undergoing anaerobic
decomposition (see Chap. 13), an O2 consumption potential for waste undergoing
aerobic decomposition can be estimated. This may be measured in the laboratory
or estimated using assumptions regarding waste composition and the fraction of
waste potentially subject to aerobic decay. The following equation is commonly
cited in design texts for solid waste and represents the O2 demand as a function of a
generic stoichiometric representation of waste’s chemical composition (Haug 1993).
æ 4 a + b - 2c ö æbö
Ca H bOc + ç ÷ O2 ® aCO2 + ç 2 ÷ H 2O (14.1)
è 4 ø è ø
When this equation is simplified to the aerobic degradation of cellulose (C6H10O5),
we arrive at:
Fig. 14.3 Air addition requirement for complete aerobic waste stabilization as function of waste
disposal rate
Fig. 14.4 Results of aeration pump tests at a MSW landfill: backpressure as a function of added
flow rate (Jain et al. 2005)
(New River Regional Landfill, Florida, USA). Figure 14.4 shows a representative
graph of pump test results, where the flow rate was measured as a function of injec-
tion pressure in wells installed at varying depths within the waste. Greater injection
pressures resulted in greater air addition rates, and the achievable rates declined as
the well construction depth increased, which was attributed to the greater overburden
324 14 Landfill Air Addition
pressures in deeper sections, larger amount of moisture present, and increased gas
pressures present from anaerobic decomposition.
Since pressurized air addition in landfills has not been practiced to a large extent,
methodologies for the design and placement of air addition devices lag similar
efforts for liquids addition and LFG extraction. Some basic concepts from modeling
gas extraction in landfills, however, may be applied (e.g., the concept of radius of
influence). Additionally, the large body of design information available for air addi-
tion and vapor extraction for soil/groundwater remediation systems can be con-
sulted and adapted. Air injection system design (blowers, manifold, and injection
wells) methodology takes into account the necessary air volume, air flow rate, air
entry pressure for the surrounding media, constituent mass to be degraded, friction
and minor pressure losses, and a factor of safety to decrease the potential for air flow
backup due to high pressures within the media. In the case of a landfill, leachate
surrounding an injection well may cause a need for increased injection pressure
(Marley et al. 1995; Hudak 2000; Leeson et al. 2002). Air addition systems may also
be designed with the intent of pulsed or periodic air injection, possibly necessitating
a higher air addition rate while blowers are operating to achieve the overall air addi-
tion volume over a fixed time period. The unique challenge for designing these
systems for landfills is the heterogeneity of the waste material and the potential for
elevated temperatures and subsurface heat-generating reactions. When air is added
to an injection well, aerobic decomposition activity will occur to the greatest extent
in the area immediately surrounding the well. The rapid reaction rates associated
with aerobic activity may result in large amounts of heat generation, and it has been
observed that temperatures within the waste can increase beyond the upper range
where aerobic microorganisms thrive (discussed in next section) (Stone and Gupta
1970; Powell 2005).
The selection of a blower depends on the volume of air required, desired flow
rate, and anticipated pressures required to add the amount of desired air. There are
several factors that can influence the effectiveness of the air addition system. Due to
the heterogeneous nature of the solid waste placed in the landfill, a wide variation
of achievable addition rates should be expected. Another consideration is the pres-
ence of higher moisture contents in the landfill waste; moisture acts as a physical
barrier to air flow and can reduce the flow significantly (observed by Jain et al.
2005). In practice, it may be impossible to have a completely aerobic landfill,
because waste in deeper sections of the landfill is dense and well compacted, and
thus air permeability is very low (see Chap. 5). For aerobic landfills, the balance of
air and water addition is critical. Sufficient water must be available to provide a suit-
able environment for microorganisms to thrive. If sufficient water is not available,
excessive heat production can result in the combustion of the waste. However, if an
excessive amount of water is present, hydraulic limitations make it difficult to add
sufficient amounts of air evenly to the waste, resulting in short-circuiting and uneven
treatment of the waste mass. Finally, with respect to heating of the waste, sufficient
infrastructure must be in place to allow generated heat to escape, as discussed in the
following section.
14.3 Air Addition System Configuration and Design 325
Fig. 14.5 Vertical well air addition strategies (a) (modified from Ritzkowski and Stegmann 2012)
(a) addition of pressurized air (b) combined extraction-aeration system inducing low-pressure
aeration (c) aeration into the LCRS and waste mass (d) extraction system allowing air introduction
to a vent open to the atmosphere
Since the volume of air required to stabilize a unit mass of waste is greater than
the volume of LFG produced under anaerobic conditions, the sizing of system
infrastructure (blowers, pipes) will necessarily be larger. Aerobic systems may also
be operated following a pulsed period so more effective oxidation for a larger radius
of influence is achieved (Boersma et al. 1995; Bass et al. 2000; Yang et al. 2005).
14.4 Operation, Monitoring and Control 327
Because of the uncertainties related to air addition and the potentially dramatic
consequences that might result from improper operation (e.g., excessive waste heat-
ing or smoldering conditions), proper operation, monitoring and control are critical.
This section reviews these issues, including a focus on explosive gas control and fire
prevention.
14.4.1 Operation
Aerobic waste degradation results in the release of more heat than anaerobic activ-
ity, thus leading to an increase in landfill temperature relative to typical anaerobic
landfill environments. The rapid release of heat can increase the waste temperature
and result in combustion or combustion-like conditions, referred to as landfill fires,
subsurface oxidation events, subsurface exothermic reactions, or hot landfills. This
must be controlled by careful monitoring of temperature and by installing a system
to add water if needed. The explosivity range of CH4 is from 5 to 15 % (volume) in
air, thus the potential to create explosive conditions may exist when air is added.
Furthermore, landfills (particularly larger facilities) are typically well-insulated,
thus rapid heat increases within the landfill are often difficult to dissipate.
328 14 Landfill Air Addition
The primary operating constraints for an air addition system will include pressure,
air or gas flow rate, gas composition, and temperature (gas or waste). The operating
pressure (or the required injection pressure) will be based on limits or ranges estab-
lished at the design stage. The design pressures are typically calculated using literature-
reported values for waste properties (e.g., intrinsic permeability), possibly coupled
with fluid flow modeling (see Chap. 5) and may be supported through limited field
pump tests to establish site-specific constraints or conditions (see Fig. 14.4). In addi-
tion to the pressure considerations related to injecting air into the waste mass,
another factor to consider is the backpressure experienced within the piping infra-
structure—blower and compressor systems have an upper limit of backpressure that
can be experienced before mechanical shutdown. Again, in this case it is useful to
establish pressure profiles as part of pump testing prior to specification of mechani-
cal blower equipment so that under- or over-design can be avoided. Given that
pulsed or periodic air injection has been shown to be advantageous over continuous
injection for aeration (Boersma et al. 1995; Yang et al. 2005), these techniques
should be considered for landfill aeration systems and design flow rates should
account for the possibility of operation as a pulsed system. Air channels (i.e., pref-
erential airflow pathways) form within the surrounding media and pulsed operation
increases mixing of aerated pore space with landfill gas or leachate through forma-
tion and collapse of these flow paths (Johnson et al. 1993; Yang et al. 2005).
Temperature monitoring and control are among the most critical factors in the
operation of aerobic landfills. Landfills that are in a regulatory environment that
requires extraction and monitoring of LFG [e.g., US landfills that are subject to
the US EPA Emission Guidelines or New Source Performance Standards under the
Clean Air Act (Code of Federal Regulations 1996)] may be required to monitor gas
temperature. However, in aerobic environments, additional temperature measure-
ment and monitoring is often warranted for multiple reasons, including within the
waste mass itself. First, extracted gas temperatures can include the temperature of
gas produced radially outward from a given gas extraction point, thus the measured
gas temperature represents a combination of gases produced in all directions from
the given extraction well. Second, gas temperatures are often lower than actual
waste temperatures, thus the measurement of a given gas temperature may not accu-
rately reflect the temperature conditions of the waste itself, particularly near areas
where air is added. Finally, the frequency of gas temperature measurement in regu-
latory environments like those in the US EPA regulations is limited (monthly),
which does not provide the operator sufficient data to understand whether air
addition is effective or if excessive temperatures are occurring within the waste.
Gas composition is another key operating parameter that must be measured
during air addition operations. Similar to waste temperatures, measuring gas com-
position provides an opportunity to understand the degree of effectiveness of air
addition. The number of gas composition monitoring points must be balanced with
cost; ideally, a larger number of monitoring points allows for more information on
the landfill environment, but too many monitoring points (which could consist of
piping comprised of stainless steel, carbon steel, PVC, or CPVC probes drilled ver-
tically into the waste) could be cost prohibitive. Table 14.3 summarizes these key
Table 14.3 Summary of key aerobic landfill operating parameters and associated monitoring devices or approaches
Operating parameter Monitoring devices or approaches
Pressure Pressure should be measured at individual air addition devices, at the blower or compressor station, and possibly
within header pipes. Dial pressure gauges may be appropriate, with the pressure range spanning the design pressures
at a minimum. The injection pressures may be adjusted manually and monitored visually or they may be tracked and
adjusted using a supervisory control and data acquisition (SCADA) system
Flow rate Total system flow rate (at the blower station) and flow rate into individual wells should be monitored at a minimum.
Rotameters have been shown to be effective devices to track flow rate and should be specified to include the range of
anticipated design flow rates at a minimum. Similar to pressure measurement, flow rates may be adjusted manually
and monitored through visual inspection, or flow may be controlled through a SCADA system
14.4 Operation, Monitoring and Control
Temperature Temperatures may be monitored by installing thermocouples within the waste (extension-grade Type T thermocouple
wire has been successfully used in MSW landfill environments), either placed directly in waste or housed in a
sheathing such as a small-diameter PVC pipe. The density of thermocouple placement depends on the number of air
addition points and the design objectives, but a prudent approach would include providing thermocouples around air
addition points and at different depths. At sites where a large number of thermocouples is desired, routing all
thermocouple wiring to one or more central locations for data logging is preferred to reduce labor and provide a more
robust dataset. Temperature can be read from thermocouples via manual readout devices or connected to a multiplexer
and data logger for frequent, automatic measurements
Gas composition Gas composition monitoring can occur at dedicated monitoring points, at LFG extraction wells, or within LFG
extraction header pipes. A specialized meter designed to analyze LFG is recommended to reduce potential
interference. These meters typically analyze CH4, CO2, O2, and calculate a balance gas. LFG meters can provide
frequent measurements. In some cases, the monitoring of certain trace gases may be required (e.g., CO or H2),
particularly in cases where high temperatures are observed. Order-of-magnitude measurement of CO in the field can
be accomplished by collecting samples in non-reactive gas sample bags (e.g., Tedlar) and sampling the gas using a
colorimetric detector tube (Powell et al. 2006). Confirmatory laboratory sampling may be achieved by collecting gas
samples from monitoring points of interest in a Tedlar bag or a passivated stainless steel canister
329
330 14 Landfill Air Addition
A concern at all landfills is the formation of explosive gas mixtures, as CH4 is flam-
mable when mixed in a certain proportion with O2. Locations at a landfill where
LFG has the potential to mix with air, and thus CH4 to mix with O2 (such as pump
stations, valve vaults, buildings near the landfill, and GCCS infrastructure) require
periodic monitoring to assess whether potentially explosive conditions have formed
(a spark or ignition source must be present for an explosion to occur when an exp
losive gas mixture is present). Landfill operators attempt to avoid explosions by
minimizing locations where explosive gas conditions exist, and where they might
exist, avoiding potential ignition sources (e.g., explosive proof switches at pumping
stations, prohibiting smoking in or around active or closed landfill areas). Clearly,
landfill operators that purposely promote air entry into the landfill must be extra
vigilant with regard to avoiding explosive conditions.
When evaluating landfill gas for flammability, the most typically cited values are
a 5 % lower explosive limit and a 15 % upper explosive limit, by volume (ATSDR
2001). These values refer to the percentage of CH4 present in air. When the CH4
content is less than 5 %, not enough fuel is present to sustain a flame (the mixture is
too lean), whereas when the CH4 is greater than 15 %, the mixture is too rich. These
values, however, refer to the occurrence of CH4 in air. In reality, CH4 would almost
always be accompanied by another gas such as CO2, and other non-flammable gases
act to dilute the CH4. The presence of “diluent” gases therefore reduces the range
over which CH4 is flammable.
Given the impact of diluent gases, it is more helpful to describe CH4 flamma
bility in the form of a flammability chart, as opposed to a fixed set of CH4 concentra-
tions. Figure 14.7 presents a flammability chart, with O2 shown as a function
of CH4, and zones delineated that express whether the mixture is flammable or not
(following the procedure of Coward and Jones 1952). The relative concentration of
the primary diluent gases expected, N2 and CO2, will vary depending on specific site
conditions, thus the chart presents the flammability zone with N2 treated as the dilu-
ent gas, as it provides a larger (more conservative) range.
14.4 Operation, Monitoring and Control 331
Heating events within the waste mass, which are also referred to as subsurface
fires, subsurface oxidation events, subsurface exothermic events, or hot landfills,
among other terms, are a concern at all landfills. Landfill fires on the surface are
fairly common in the US and internationally, and the causes can vary widely (FEMA
2002). Generally, heating events can be caused by external factors (such as hot or
smoldering materials delivered to the landfill) or caused by reactions within the
waste itself (such as intrusion of atmospheric air that results in aerobic reactions).
In anaerobic systems, temperatures as high as 55–60 °C are sometimes reached in
the landfill interior, and this temperature becomes self-regulating since higher
temperatures will limit the activity of the anaerobic organisms. With aerobic
systems, however, temperatures can reach 70 °C or more; Powell (2005) reported
waste temperatures increasing approximately 20 °C to more than 70 °C within 1 week
of initiating air addition at an MSW landfill in the US. While the aerobic process
may be self-regulating to a degree, the well-insulated conditions within a landfill
may prevent the heat produced from aerobic reactions from exiting the waste.
For example, waste temperatures following cessation of air addition as reported by
Powell (2005) showed very slow temperature declines, which is in contrast to the
rapid temperature increases brought about by aerobic operation. At this point, heat-
ing reactions may create smoldering or pyrolysis-like conditions within the waste
(with temperatures ranging from 80 to 100 °C or more), which is supported by work
reported by Moqbel (2009).
332 14 Landfill Air Addition
Fig. 14.8 Temperature control chart used as part of the NRRL Aerobic Bioreactor Research
(Ko et al. 2013)
Given the complexities inherent with landfilled solid waste (and accompanying
cover material), adding sufficient air to a full-scale landfill operation at a rate that
meets air addition objectives but does not promote excessive waste heating, com-
bustion, or pyrolysis conditions may be challenging. Landfill operators who add air
must have monitoring points to measure in-situ temperature of the waste to under-
stand subsurface conditions and to regulate air addition rates; methods for monitor-
ing temperatures within landfills are summarized in Chap. 16. The engineer who
develops a site’s operations plan must establish monitoring equipment, methods, freq
uencies, and operating thresholds to maximize control over the system. Figure 14.8
presents the temperature threshold regime utilized as the New River Regional
Landfill described in Chap. 4 (air addition was practical at this site as summarized
by Ko et al. (2013).
When monitored temperatures reach a level of concern, the typical first course of
action is to reduce or stop air addition. Given the insulating environment present
within landfills, cessation or reduction of air addition may slow or stop the increase
in temperatures within the waste, but that may not always occur, at which point other
measures such as addition of liquids in the area of concern may be needed. The
amount of liquid added must be balanced with the goal of future air addition, since
hydraulic limitations to air addition will occur with excessive liquids addition.
14.5 Air Addition Experience 333
CH4 and other gas-phase compounds produced in anaerobic landfills necessitate the
installation of recovery and treatment systems, both to meet regulatory and environ-
mental considerations, and for energy recovery. As stated earlier, one of the cited
goals of some practitioners of air addition to landfills is the suppression of CH4
generation. This raises the fundamental question of whether aerobic landfill exhaust
requires collection and treatment. Even if a landfill were designed, constructed,
and operated to be completely aerobic, because of hydraulic and other constraints
already discussed, it is likely that CH4 generation could not be completely sup-
pressed. Thus, in a regulatory environment it is not likely that avoidance of active
LFG collection would be possible. In this case, the addition of air would need to be
balanced with the need to actively collect LFG produced anaerobically in the land-
fill. This leads to a complex situation where the goals of operating a landfill aero
bically would need to be consistent with the requirements typical of active LFG
collection systems. For example, US EPA Clean Air Act requirements for active
LFG collection systems place a limit on the amount of O2 (5 % by volume) or N2
(20 % by volume) that may be present at LFG collection wells or devices. The obvi-
ous conflict can be seen when considering the composition of air that would be
introduced into a landfill during aerobic operation. These regulatory considerations
must be examined at the design stage and the approach to aerobic operation would
need to be discussed with the appropriate regulatory officials to ensure the aerobic
operation would be consistent with existing regulatory operating constraints.
In recent years, aerobic bioreactor landfill technology has received increased atten-
tion due to the cited potential benefits (Matsufuji et al. 1993; Rich et al. 2008;
Ritzkowski and Stegmann 2010). The concept of the aerobic bioreactor landfill has
been applied—although with varying practices and techniques—in several coun-
tries, including Japan, Germany, and the US. These experiences and approaches are
summarized in the following sections.
14.5.1 Asia
Fig. 14.9 Configuration of large diameter LCRS drain for the semi-aerobic landfill
Fig. 14.10 LCRS of semi-aerobic landfill after construction and before waste placement; con-
nected rock drains are shown (Photo courtesy of Yasushi Matsufuji)
in several regions, particularly in Asia (Chong et al. 2005). The core fundamental of
the Fukuoka method is to create as much of an aerobic zone as possible within the
landfill by building an air introduction system in a manner that promotes natural
ventilation into the waste. The method does not require the use of mechanical extrac-
tion systems (e.g., air pumps or blowers) and allows for locally-available and less
expensive materials to be used.
Air entry into the semi-aerobic landfill is achieved through two means. First, a
large leachate collection pipe, typically at least 0.45 m diameter and as large as
0.6 m, serves as the primary leachate drainage port for the landfill and extends out-
ward to the point of discharge and open to the environment (Figs. 14.9 and 14.10).
This pipe should be bedded in drainage rock and at least two-thirds of the pipe
diameter should remain open to provide for passive air inflow to the bottom of the
landfill. Deep aeration was observed in lysimeter experiments to provide the quick-
est degradation of organic carbon as well as enhanced nitrification compared to
injection of air at shallower waste depths (Wu et al. 2014).
14.5 Air Addition Experience 335
Fig. 14.11 Illustration of the semi-aerobic landfill concept (a) LCRS vents, (b) LCRS and vertical
well vents, and (c) LCRS, vertical well, and horizontal vents
Figure 14.11a illustrates air entry into the semi-aerobic bioreactor from the
LCRS. The second means of promoting air entry is the connection of vertical pipes
to the leachate drainage pipes (Fig. 14.11b). The Fukuoka method recommends a
spacing of the vertical pipes of 20–40 m, with closer spacing recommended for
deeper landfills. These pipes (sometimes referred to as vents) serve as a means for
heated vapor within the landfill to rise to the surface and thus draw air into the waste
from the bottom. The vents can be constructed in a similar manner as LFG collec-
tion wells placed during waste filling (see Chap. 13), but the method encourages
innovative use of construction techniques and less expensive construction materials
(Matsufuji et al. 1993, 2004; Chong et al. 2005). Figure 14.12 shows a vent con-
structed for a semi-aerobic landfill in Thailand. The Fukuoka Method developers
336 14 Landfill Air Addition
Fig. 14.12 Vertical vent of a semi aerobic landfill after construction and before waste filling
describe the ability of the vents to draw air into the landfill as critical to the success
of the technology, and if site-specific reasons preclude close spacing of vents, addi-
tional horizontal vents exiting the side of the landfill should be constructed
(Matsufuji et al. 2004). The horizontal vents should be connected to the vertical
risers and should slope downward toward the vertical wells to promote gravity
drainage of liquids (Fig. 14.10c, Matsufuji et al. 2004).
14.5.2 Europe
Fig. 14.13 Inlet point for air addition and gas extraction at a closed landfill undergoing aerobic
treatment (Photo courtesy of Marco Ritzkowski)
Air addition into landfills has received limited application in North America. In the
1960s, air addition was explored at a large landfill in California, where Merz and
Stone (1966) added air through an access well in the center of a 20-ft deep landfill
338 14 Landfill Air Addition
Fig. 14.14 Blower housing and exhaust gas treatment system at a closed landfill (Photo courtesy
of Marco Ritzkowski)
Fig. 14.15 Wind-powered
air vent at a closed landfill
undergoing aerobic treatment
(Photo courtesy of Marco
Ritzkowski)
14.5 Air Addition Experience 339
Fig. 14.16 Cross section illustrating construction of the VSA biostabilization technique
test cell using a mechanical blower. Aerobic conditions dominated within the test
cell (as characterized by exhaust gas composition) and waste settlement in the first
year was four times greater than a corresponding anaerobic control cell. Waste tem-
peratures as high as 190 °F were measured, and at times the exhaust gas exhibited
smoke and signs of fire, although the issue was reportedly remedied by blower shut
down for a period of 50 days (Merz and Stone 1966).
At some large landfills where leachate recirculation is practiced, air is first added
to the horizontal leachate addition lines as a means of increasing temperature and
stimulating biological activity, especially in colder climates. For example, at the
Outer Loop landfill (Kentucky, USA), air was added to a horizontal piping network
[4-in. pipes spaced 60 ft apart (10 cm diameter pipes spaced 18.3 m apart)] approxi-
mately 30 days after one lift of waste was placed over the pipes to accelerate decom-
position. Air addition, via compressed air injection, proceeded for periods of 30–90
days (at a flow rate of 57 m3/min), until one of three set points were reached: (1)
waste temperature reaches 71 °C, (2) temperature change of 6.7 °C (12 °F) in a 24-h
period, or (3) air addition duration of 90 days.
At the Sullivan County Landfill in Monticello, New York, a technique described
as vacuum-induced semi-aerobic (VSA) biostabilization was explored (Hansen
et al. 2002). In this process, horizontal trenches containing 30-cm perforated con-
duits were placed on the landfill surface (Figs. 14.16 and 14.7). After wetting the
waste with leachate and placing a synthetic daily cover, a vacuum was placed on the
pipes using the site’s existing LFG collection system, causing atmospheric air to
be drawn through the surface of the landfill. The objective was to provide rapid
340 14 Landfill Air Addition
Fig. 14.17 VSA trench under construction (Photo courtesy of David Hansen)
Fig. 14.18 Distribution manifold and air addition well for an aerobic landfill in the Southeast US
be very challenging, primarily due to the inability of many wells to accept air. At the
Yolo County Landfill in California (also discussed in Chap. 4 for more details), a
vacuum was placed on horizontal gas collection pipes (1–15 cm) placed in shredded
tire-filled trenches to draw air through the permeable surface of the landfill (Yazdani
et al. 2010). This study reported challenges with respect to suppressing anaerobic
activity and maintaining an aerobic state. Even in areas with substantial air injec-
tion, anaerobic pockets still persisted, and the presence of anaerobic pockets was
more prevalent in areas where moisture content was greatest (Yazdani et al. 2010).
References
ATSDR (2001) Landfill Gas Primer: an overview for environmental health professionals, Nov 2001
Bass DH, Hastings NA, Brown RA (2000) Performance of air sparging systems: a review of case
studies. J Hazard Mater 72(2–3):101–119
Berge ND, Reinhart DR, Townsend TG (2005) The fate of nitrogen in bioreactor landfills. Crit Rev
Environ Sci Technol 35:365–399
Berge ND, Reinhart DR, Dietz J, Townsend T (2006) In situ ammonia removal in bioreactor land-
fill leachate. Waste Manag 26:334–343
Berge ND, Reinhart DR, Dietz JD, Townsend T (2007) The impact of temperature and gas-phase
oxygen on kinetics of in situ ammonia removal in bioreactor landfill leachate. Water Res 41:
1907–1914
Bilgili MS, Demir A, Özkaya B (2007) Influence of leachate recirculation on aerobic and anaero-
bic decomposition of solid wastes. J Hazard Mater 143:177–183
Boersma PM, Diontek KR, Newman PAB (1995) Sparging effectiveness for groundwater restora-
tion. In: Hinchee RE, Miller RN, Johnson PC (eds) In situ aeration: air sparging, bioventing,
and related processes. Batelle, Columbus, pp 39–46
Bolton N (1995) The handbook of landfill operations. Blue Ridge Services, Atascadero
342 14 Landfill Air Addition
Powell J (2005) Trace gas quality, temperature control and extent of influence from air addition at
a bioreactor landfill. University of Florida, Gainesville
Powell J, Jain P, Kim H, Townsend T, Reinhart D (2006) Changes in landfill gas quality as a result
of controlled air injection. Environ Sci Technol 40:1029–1034
Read AD, Hudgins M, Phillips P (2001) Perpetual landfilling through aeration of the waste mass;
lessons from test cells in Georgia (USA). Waste Manag 21(7):617–629
Rich C, Gronow J, Voulvoulis N (2008) The potential for aeration of MSW landfills to accelerate
completion. Waste Manag 28:1039–1048
Ritzkowski M, Stegmann R (2007) Controlling greenhouse gas emissions through landfill in situ
aeration. Int J Green Gas Control 1:281–288
Ritzkowski M, Stegmann R (2010) Generating CO2-credits through landfill in situ aeration. Waste
Manag 30:702–706
Ritzkowski M, Stegmann R (2012) Landfill aeration worldwide: concepts, indications and find-
ings. Waste Manag 32:1411–1419
Ritzkowski M, Heyer K-U, Stegmann R (2006) Fundamental processes and implications during in
situ aeration of old landfills. Waste Manag 26:356–372
Sartaj M, Ahmadifar M, Jashni K (2010) Assessment of in-situ aerobic treatment of municipal
landfill leachate at laboratory scale. Iranian Journal of Science and Technology, Transaction B,
Engineering, Vol 34, B1, 107–116
Stessel RI, Murphy RJ (1992) A lysimeter study of the aerobic landfill concept. Waste Manag Res
10:485–503
Stone R, Gupta R (1970) Aerobic and anaerobic landfill stabilization process. J San Eng Div,
Proceedings of the American Society of Civil Engineers, 96(SA 6):1399–1414
Wu C, Shimaoka T, Nakayama H, Komiya T, Chai X, Hao Y (2014) Influence of aeration modes
on leachate characteristic of landfills that adopt the aerobic-anaerobic landfill method. Waste
Manag 34:101–111
Yang X, Beckmann D, Fiorenza S, Niedermeier C (2005) Field study of pulsed air sparging
for remediation of petroleum hydrocarbon contaminated soil and groundwater. Environ Sci
Technol 39(18):7279–7286
Yazdani R, Mostafid ME, Han B, Imhoff PT, Chiu P, Augenstein D, Kayhanian M, Tchobanoglous
G (2010) Quantifying factors limiting aerobic degradation during aerobic bioreactor landfill-
ing. Environ Sci Technol 44:6215–6220
Chapter 15
Operations
Abstract This chapter builds on concepts that were previously presented in the
more design-oriented chapters by highlighting the importance of establishing a
crosswalk between design and operation. The duties of landfill operation staff for
sustainable landfills are presented, including a comparison with typical duties of
operations staff at traditional landfills. Elements that a landfill operations plans
should contain to accommodate sustainable landfilling procedures are presented;
important operator responsibilities include monitoring and data collection (e.g.,
tracking the liquids balance), infrastructure inspections and record keeping. Recom
mendations for using effective system performance monitoring metrics are presented
at the end of the chapter.
Although much of this book focuses on technical information regarding the science
and engineering of processes related to landfills undergoing rapid waste stabiliza-
tion to promote long-term environmental protection, we cannot escape the fact that
even the best planned and designed system can fail without careful and dedicated
operation. For example, an engineer may design a LCRS that provides necessary
drainage for leachate from the landfill, but if the removal pumps are not properly
maintained, inspected and operated, liquid levels can build up within the landfill and
result in consequences ranging from poor gas collection to side slope failure. Even
if the designer provides robust plans for a pumping and piping system that promotes
even distribution of liquids throughout the landfill, successful installation of the
infrastructure depends on coordinating construction with routine waste disposal
operations and appropriate recordkeeping.
While many facets of other chapters relate to operational issues, the role of the
operator is so critical with regard to environmental safety and successful outcomes
of sustainable landfill processes, a separate chapter highlighting the role of operation
is warranted. In addition, this chapter, coupled with the monitoring technologies in
the chapter that follows, should provide operators of advanced landfill systems a
strong foundation for the efforts required beyond traditional landfill o peration.
This chapter, in concert with the other more science and engineering-oriented
chapters of this book, should provide the design engineer helpful insight on how
best to prepare a system design and operations plan that maximize the operator’s
chances for success.
The landfill operator must comply with all requirements of the governing regu
lations and the facility’s permit. Beyond this, the landfill operator should be respon-
sible for operating and maintaining the facility in a manner that is protective of the
facility’s other employees, site visitors, nearby residents, and the environment in
general; such expectations apply not only to the current operational time frame, but
to every extent possible, the future. These duties and expectations, along with the
fiscal responsibility of operating a facility within the constraints of the provided
operating budget, pose a challenge for the operator of any landfill.
Best practices for operation of sanitary landfills are described in several different
documents developed by the professional and regulating community (Bolton 1995;
IRL EPA 1997; EuropeAid 2010), and some resources include information of oper-
ation of landfills operated as bioreactors (Reinhart and Townsend 1997; ITRC
2005). Routine duties of landfill operators include weighing and inspecting waste
loads and directing vehicles to designated unloading areas, moving waste to appro-
priate disposal areas after unloading, compacting the waste, and placement of
required cover material at the necessary frequency and amount (see the general
discussion of landfill operation in Chap. 2). Standard site maintenance activities
include mowing grass, maintaining roads, and repairing erosion damage. A common
regulatory-required operator duty is the examination of incoming and deposited
waste loads to identify and remove prohibited material (waste screening).
Some operational duties may be performed with facility staff, but the operator
may elect for an outside contractor to provide such services. Examples include
operation and maintenance of the leachate and gas management systems, monitor-
ing groundwater and soil vapor, conducting topographic surveys to track landfill
elevation and topography, and vehicle maintenance. Record keeping and reporting
are also major operator responsibilities, and likewise these duties may be handled
with facility personnel, with outside consultants, or some combination of the two.
At facilities employing sustainable landfill technologies, operators will be
charged with additional responsibilities beyond routine landfill operation. Some of
the responsibilities result from added operational requirements (e.g., installing
additional infrastructure, adding liquids and air), while others come about because
of an increased degree of required monitoring (e.g., liquid levels, degree of waste
stabilization). Similar to many of the routine landfill operator tasks, some facilities
perform these responsibilities in-house while others contract with outside parties.
15.4 Construction, Oversight, and Recordkeeping 347
The construction of infrastructure for liquids and air addition to landfilled waste and
the installation of monitoring equipment and instrumentation fall outside the typical
duties of routine landfill operation, and the operator must determine to what extent
landfill personnel will perform these duties. Some operators take an active role in
the construction of pipes and trenches for delivering liquids to the landfill, while
others rely solely on outside contractors for such installations. Regardless of the
level of involvement the operator plays in actual construction, it is important that
348
Table 15.1 Typical elements of a landfill operation plan and additional considerations for sustainable landfills
Landfill operation plan element Broad definition and additional considerations for landfills using sustainable technologies
Waste screening and inspection A plan for inspecting and removing prohibited wastes. Materials that promote or inhibit waste stabilization may
be targeted for placement in designated areas
Filling sequence and waste A plan for waste filling includes site drawings indicating the placement location and dimensions of waste cells
placement and lifts within the landfill unit, providing directions to the operator from beginning of waste placement to final
landfill build-out. Infrastructure for liquids and air addition, gas extraction, liquids drainage, and monitoring
instrumentation, can be included as part of fill sequencing
Soil cover application Directions describing placement areas and/or characteristics of daily, intermediate and final cover soil (or other
waste covering activities). Landfills with liquids addition may specify additional requirements for as cover
removal and stricter controls on the types of cover materials to be used. The plan may also designate alternative
cover materials that are more compatible with liquids addition operations
Leachate management A description of techniques and protocols to manage the expected quantity and characteristics of leachate from
the cell or group of cells at the site. Additional leachate pumping and distribution to the waste may be required,
along with expanded duties with respect to construction, operation, maintenance, and monitoring of leachate
systems
Storm water management A plan to route and manage water from rainfall that does not contact solid waste. In sustainable landfills,
stormwater may be managed differently to promote liquids retention in the landfill. Stormwater management
infrastructure may be integrated into seep and gas control systems
Gas management A plan to manage an active GCCS—this plan may be part of the larger operations plan or a stand-alone document.
GCCS operation and monitoring may be practiced earlier and at an enhanced level at sustainable landfills. The
presence of more liquids in the landfill may result in more maintenance and monitoring of the GCCS
Record keeping and reporting Directions outlining the type and frequency of data collected as part of regulatory requirements, permit
requirements, or simply for good landfill management practice. Additional site records and reporting are typically
required at landfills using sustainable technologies
15 Operations
15.4 Construction, Oversight, and Recordkeeping 349
Fig. 15.1 Operator inspecting and observing construction of a liquids addition device
Some components of the constructed system may not be connected until a later time
(e.g., a buried injection trench that will be routed to a distribution manifold), and very
likely will not be operated immediately, so it is very important for the locations and
elevations of all buried pipes, trenches, or related devices be surveyed and recorded.
This should be a requirement of the permitted operations plan.
The operator must coordinate construction activities with routine landfill opera-
tions. Considerations include the location of the construction area with respect to the
area of active waste disposal; the location of access roads, storm water drainage
features, and structures; and the location of buried pipes for gas and liquids mana
gement. In addition to the devices themselves, sufficient area will be required for
storing scraped soil, excavated waste, bedding materials, and pipe. For some designs,
excavated waste will be pushed back over the construction area and compacted in
place, but in other cases, the excavated waste will require loading and transport to
the active disposal area (see Chaps. 9 and 10). Some landfill operators construct the
piping on their own using thermal polyethylene (PE) or chemical (PVC) welding.
The site’s liquids addition operations plan, which is a component of the overall site
operations plan or a separate document, provides a framework under which liquids
operations should proceed. The liquids addition operator (or operators) carries out
the tasks in the operations plan and uses judgment based on knowledge of the sys-
tem’s specifications, system response, and other relevant training to ensure effective
operations. The operation of a liquids addition system has the potential to impact
other facets of typical sanitary landfill operation. As such, the operator must be
aware of other permit-related and operational requirements that may be impacted
and coordinate closely with other site personnel responsible for such duties.
Operational tasks or performance metrics that are likely the responsibility of the liq-
uids addition system operator include those necessary to achieve and monitor liquids
addition rate, inclusive of flow rate (overall system flow rate and/or flow rate to spe-
cific devices) and cumulative volume added. Liquids addition monitoring is one of the
critical elements of an operations plan since system performance can be closely tied to
the liquids addition data and observations. Similar to the liquids addition rate, the
liquids addition pressure is also a critical component. Design pressures are established
using empirical data and engineering assumptions to avoid the creation of excessive
pore pressures within the waste mass (Chaps. 8 and 9), which could in turn impact
waste mass stability (see Chap. 12). Liquids addition pressure should be checked by
the operator routinely (either manually or through data-logged components).
15.5 Liquids Addition Operation and Monitoring 351
The liquids addition system should be designed so that the pumping, piping,
metering, and control system supports liquids addition rates, volumes and pressures
required for operation. In some cases, the engineer might include automated con-
trols (e.g., through a supervisory control and data acquisition (SCADA) system)
that maintains flow rates and pressures in the desired range and opens and closes
pumps and switches valves at designated intervals; SCADA systems could be con-
trolled via devices connected to the internet or a mobile network. Other systems
may require predominantly manual involvement of the operator as described in
Chaps. 7–9; while liquids addition have the potential to be practiced over extended
periods, it is most common to limit addition to times when a trained operator of the
system is on site. Table 15.2 presents an example operational sequence for a liquids
addition system associated with intensive manual operation.
For some types of liquids addition systems, maintaining the liquid level below
the surface of the landfill can be challenging. This normally requires extensive oper-
ator inspection and adjustment to make sure that specified liquid levels are not
exceeded. Given that achievable liquid addition rates will normally decrease with
time in a given area, operator interaction and evaluation is critical. While technolo-
gies exist that can automate such liquids addition (water level sensors and piezom-
eters controlled by actuated valves), the expense associated with such techniques
may limit widespread application. The viability of enhanced control systems should
be evaluated at the design stage and periodically after commencing operation to
assess new technological capabilities or changing cost conditions.
Operators of buried systems are often less concerned with maintaining liquid lev-
els below a specified level as these systems have the flexibility of being operated
under pressure. Normal operational routine for buried horizontal trenches or blankets
352 15 Operations
includes operating the pumping system, adjusting valves to accommodate the desired
operation strategy [i.e., which trenches will be utilized in a given liquids addition
cycle (Fig. 15.2)], monitoring flow rate and pressure, and inspecting for seeps. In the
operations plan, the operating pressure guidelines must be provided so that appropri-
ate system constraints are clearly identified for the operator. Incidences of high
pressure in a liquids addition line may be the result of a pipe or trench failure (e.g.,
crushing or buckling of the pipe), so when high pressures are observed, the system
should normally be shut down (either manually or automatically) in these areas to
allow for further exploration of the problem. The operations plan should include
troubleshooting guidelines to determine whether an injection line can be salvaged or
should be abandoned. Pressures that can still be achieved in functioning systems, but
that are greater than desirable slope stability thresholds, should be identified by the
operator along with appropriate responsive measures.
Among the more important sets of data the operator must collect, interpret and
maintain are the different components of the liquids balance for the landfill. This
includes the elements needed for tracking the landfill’s liquids budget, but also
distinct volumes and depths of liquids at different points within the landfill and its
associated infrastructure (ITRC 2005). An accurately documented liquids balance
requires measurements and estimates of major inputs (infiltration from rainfall,
added liquids) and major outputs (leachate removal, evapotranspiration from the
landfill surface). Infiltration from rainfall will be the difference between rainfall
15.5 Liquids Addition Operation and Monitoring 353
intercepted by the landfill surface and that running off as storm water. Rainfall is
simple to measure and track using rain gauges or weather stations, and indeed many
landfills are required to track weather conditions under their permit conditions.
Storm water runoff, however, cannot be directly measured, and thus can only be
estimated using engineering techniques that factor in slope, soil type, and other site-
specific features. Hydrologic models such as HELP (Chap. 5), and the associated
engineering methods that serve as the basis for this model, can be used to estimate
water balance components such as infiltration and evapotranspiration.
A major responsibility of the operator includes tracking liquid volumes added to
the system. Such measurements normally utilize flow meters associated with the
pumping systems (Fig. 15.3). Available meter output typically include the flow rate
and the cumulative volume passing through the meter. In some cases, the meters are
equipped with data logging equipment and the operator’s responsibility is to peri-
odically compile recorded data, evaluate results, and organize the information for
proper recordkeeping and possibly regulatory submission. In other cases, the opera-
tor is required to manually record meter readings. Flow rate readings provide an
assessment of system performance and changes over time. Cumulative readings
provide volumes added over specific time intervals. Daily measurements of liquids
addition (rate and volume) are typical. Where possible, flow meter readings should
be collected from as distinct of a landfill area as possible (i.e., knowing flow rates
from multiple different collection points or landfill cell is more useful to under-
standing system performance than a single combined measurement).
354 15 Operations
Liquids removed from the landfill in the form of leachate is normally measured
using flow meters attached to pumping systems. As a backup, pump run times can
be recorded to provide an estimate of pumped volumes. In cases where all leachate
removal is through gravity drainage, alternative meters (e.g., weirs with water level
detectors) can be utilized. The volume of leachate discharged off site, even if in
batches, should be tracked, and together with liquids addition volumes, should be
compared to leachate removed from the landfill cell. All of the water budget data
can be used to calculate changes in moisture storage with the landfill system, and
if an initial estimate of moisture content is known, average moisture content for
the landfill cell can be estimated. Under some regulatory jurisdictions, tracking the
landfill’s water balance is critical to determining when regulating thresholds become
active (see discussion of US EPA bioreactor rules for landfill gas in Chap. 13).
15.5.3 Inspection
Chapters 13 and 14 describe the roles of landfill gas collection and control, as well
as air addition, in sustainable landfill operation. Thorough and careful operations
are critical to successful implementation of both these elements. A key component
of assessing the performance of landfills operated to enhance waste stabilization is
evaluation of landfill gas quantity and quality. Techniques and frequencies of land-
fill gas measurement at such facilities are not altogether different when compared
to a conventionally operated landfill with a GCCS. Objectives of a landfill gas-
monitoring program may vary from conventional landfilling particularly as tracking
gas production is an invaluable tool for monitoring waste stabilization and overall
system performance (Fig. 15.5).
Landfill GCCS operators must be provided with appropriate training; several pro-
fessional training courses have been developed by different organizations and GCCS
operation instructional documents are available (e.g., ISWA WG-Landfill 2005).
The GCCS operator must coordinate with the design engineers regarding future
planning of liquids addition devices and gas collection system components to assess
the required vertical and horizontal offsets to help avoid issues with watering out of
gas collection components. The operator must evaluate gas well liquid level mea-
surements (for vertical well systems) to assess potential operational changes to the
liquids addition system that may be warranted. Alternatively, gas collection perfor-
mance data from individual wells may be used to evaluate whether watering out is
occurring—in cases where frequent watering out occurs, remedial actions such as
installation of a dedicated pump may be warranted.
Addition of air requires elements similar to both the liquids addition system
operation and GCCS operation. As outlined in Chap. 14, the motivation for air
addition may differ among sites, and it is important for the operator to possess a
clear understanding of site-specific objectives and designed outcomes. An overrid-
ing objective in operating a landfill aeration system would be to provide sufficient
oversight and monitoring to avoid formation of fires or explosive gas conditions.
Liquid pressure Pressure (psi, in. w.c.) The pressure of added liquids may be limited to avoid concerns with seeps and
slope stability. Operator will need to monitor pressure and adjust or cease
operation if thresholds are exceeded
Liquid depth Depth (in., m) The depth of liquid may be limited, such as depth of leachate on liner system or
in a vertical well. The operator will need to monitor depth and adjust or cease
operation if thresholds are exceeded
Leachate composition Concentration (mg/L) Leachate samples will be periodically analyzed. In the short-term, some changes
may indicate that operations require adjustment (e.g., rapid decrease in specific
conductance may indicate too much stormwater is entering leachate collection
system; sudden decrease in pH and increase in BOD may indicate portions of
system are stuck in acid-forming phase). In long-term, leachate composition can
be used to help assess the progression of landfill stability
Air and gas flow rate Volume per time (cfm, lpm) Air flow rates added to or extracted from the landfill will be periodically
measured for individual devices. For air addition, flow rate limits will be specific
in the operation plan. For gas extraction, for wells with large flow rates
(especially at small vacuums) may suggest that additional extraction points are
warranted. Flow rate can be directly measured or calculated (e.g., based on
differential pressure across an orifice plate)
357
(continued)
358
Table 15.3 (continued)
Monitoring parameter Typical units Description
Gas pressure Pressure (psi, in. H2O) Gas pressures at well heads, points in the GCCS network, or points within
landfill are measured
Gas composition Concentration (percent, Portable or fixed meters will be used to determine composition of major gas
part per million) components to assess performance of gas extraction and air addition systems.
Portable sampling containers may be used to analyze major or trace gases
Temperature Degrees (°C, °F) Measurement of internal landfill temperature provides an assessment of waste
biological activity. Temperature is a critical parameter for monitoring landfill fire
occurrence. Temperature of landfill gas may be measured using a portable meter
(often the same meter used to measure composition)
Moisture content % Wet weight Internal moisture sensors may be used to assess the efficiency of moisture
distribution systems
15 Operations
References 359
References
Bolton N (1995) The handbook of landfill operations. Blue Ridge Services, Atascadero
EuropeAid (2010) Waste governance—ENPI east: landfill operations guidance manual. European
neighborhood partnership instrument east region, Brussels, Belgium
IRL EPA (1997) Landfill manuals: landfill operational practices. Environmental Protection Agency
Ireland, Ardcavan
ISWA WG-Landfill (International Solid Waste Association Working Group for Sanitary Landfills)
(2005) Field procedures handbook for the operation of landfill biogas systems. Austria, Vienna
ITRC (Interstate Technology & Regulatory Council) (2005) Characterization, design, construc-
tion, and monitoring of bioreactor landfills. ALT-3 Interstate Technology & Regulatory
Council, Alternative Landfill Technologies Team, Washington
Reinhart DR, Townsend TG (1997) Landfill bioreactor design and operation, 1st edn. CRC,
New York
Chapter 16
Tools and Techniques for Landfill Monitoring
Abstract Specific tools and techniques that can be used for landfill monitoring,
with particular focus on methods that enhance monitoring and maintenance proce-
dures at landfills practicing sustainable technologies. Key monitoring parameters
and methodologies covered include liquids (e.g., volume, depth, pressure, chemical
composition), landfill gas (e.g., flow rate, pressure chemical composition), and
properties of the waste solids themselves (e.g., moisture content, methane poten-
tial). The use of instrumentation placed within the landfill to measure temperature,
moisture content, and pressures are described.
Fig. 16.2 Pressure transducer used for measuring the location of liquids surface with a landfill well
larger physical pressures on the device (not necessarily the pressure being measured)
relative to those sensors only used for measurement of liquid depth. For devices
buried in the landfill, it is important to consider that the pressure reading measured
represents a combination of both liquid and gas pressure (Kadambala et al. 2011);
for some applications, this could confound the results. The application of in-situ
pressure sensors will be discussed in greater detail in Sect. 16.8 of this chapter.
Organic strength Organic chemicals are created from biological decay of the waste and BOD
measurement leaching from waste components themselves. Some organic matter (OM) Chemical Oxygen demand (COD)
parameters represent biodegradable OM, while others characterize total Total organic carbon (TOC)
OM. The concentration and type of organic matter determines treatment
requirements and provides an indication of the waste stabilization
environment inside the landfill
Inorganic strength Depending on waste composition, leachate contains substantial amounts TDS
measurements of dissolved inorganic ions. There may be measured in bulk (TDS) or Anions (chloride, potassium)
individually (anions, cations) Cations (sodium, calcium, magnesium)
Nutrients Several nitrogen and phosphorous chemicals are present in leachate, though Ammonia
nitrogen is more prominent. Ammonia-nitrogen content often strongly Total Kjeldahl Nitrogen
controls treatment options, although dissolved organic nitrogen can be Nitrate/nitrite
limiting when a treatment facility discharges to nutrient-limited water bodies
Total phosphorus
Trace constituents A variety of trace constituents, both organic and inorganic, leach from Heavy metals and metalloids
waste components in the landfill Organic pollutants
365
366 16 Tools and Techniques for Landfill Monitoring
standard water and wastewater analytical method compendia (Rice et al. 2012; US
EPA 2013). Leachate quality can vary tremendously from site to site (and within a
single site) as a function of waste type, age, climate and operating conditions.
Numerous publications describe leachate quality (Chu et al. 1994; Kjeldsen et al.
2002); example ranges of several major constituent concentrations are presented for
landfills practicing sustainable technologies to provide likely magnitudes and trends.
Leachate samples can be collected from multiple locations, including wells or simi-
lar boreholes within the landfill, leachate sumps or pumping stations, pressurized
pipes, and external storage areas (tanks, ponds). Since leachate originates from
multiple locations within a landfill unit or from different landfill cells are often com-
bined as part of the collection and conveyance system, the sample collection location
should be appropriately noted and considered when interpreting results. In some
cases, leachate samples can be obtained directly from a sampling port or accessible
leachate surface, but certain locations will require sampling pumps or manual bail-
ers. Sample agitation may impact analytical results. Exposure to air can alter some
water quality parameters (e.g., dissolved oxygen, oxidation reduction potential, vola-
tile organic compound concentrations) and excessive stirring of sediments from sam-
pling locations may result in elevated suspended solids content (which can in turn
increase the concentration of other parameters if included in the measurement).
The pH of a leachate sample is a measurement of the hydrogen ion (H+) concen-
tration in the leachate and describes how “acidic” or “basic” the solution is. The pH
is reported as a numerical value in the range of 0–14. Acid solutions have a low pH,
while basic solutions have a high pH; a pH value in the range of 6–8 is considered
neutral. Most MSW leachates are relatively neutral, though as discussed in Chap. 2,
a pH outside of the neutral range may occur, which would be reflective of a distinct
stage of waste decomposition in the landfill. Figure 16.3 provides pH data for two
landfills over a 20–25 year period; both landfills practiced technologies to enhance
waste stabilization (described in Chap. 4). The vast majority of all pH data for these
two sites fall in the 6–8 range.
The specific conductance (also referred to as electrical conductance or conduc-
tivity) provides a measure of the ionic strength of a solution by measuring the
degree that a sample conducts an electrical current. Both positively-charged dis-
solved ions (cations) and negatively-charged ions (anions) contribute to the overall
ionic strength. A greater concentration of dissolved ions in a liquid sample results in
a larger specific conductance. All leachates contain dissolved ions, but landfills co-
disposing ash will typically have higher conductivity because of the greater mass of
inorganic ions leaching from the ash. Conductivity provides a quick, simple means
of estimating the total dissolved solids (TDS) content of leachate, and measure-
ments are typically reported in units of μmho/cm or mS/cm; example data are pre-
sented in Fig. 16.3.
16.3 Leachate Chemical Composition 367
Fig. 16.3 pH and specific conductance at two landfills practicing liquids addition
As described in Chap. 2, the type of OM present in landfill leachate varies with the
dominant landfill environment and stabilization stage. Looking at several different
organic strength measurements thus provides useful information. Because so many
different kinds of organic chemicals may be present in landfill leachate, it is not
practical to measure them individually. However, since organic chemicals have the
potential to be oxidized, laboratory measurements of oxygen demand provides a use-
ful means of measuring organic strength. BOD consists of biologically degradable
dissolved organics in the leachate, while COD is a measure of chemically oxidizable
components and reflects the combined oxygen demand represented by BOD and
368 16 Tools and Techniques for Landfill Monitoring
Fig. 16.4 BOD, COD and BOD:COD at two landfills practicing liquids addition
16.3 Leachate Chemical Composition 369
Fig. 16.5 TDS and chloride concentrations at two landfills practicing liquids addition
While many chemicals may be considered a nutrient, in the context discussed here,
the term nutrient refers to nitrogen and phosphorous compounds in a wastewater
such as leachate. Ammonia nitrogen is the most abundant nutrient in landfill leach-
ate, and as nitrogen is released into leachate as a result of biological decay of waste
components, ammonia nitrogen concentration increases. The form of ammonia
nitrogen, either NH4+ (ammonium) or NH3 (dissolved or ammonia gas) depends on
pH; under neutral and acidic conditions, the majority will exist as NH4+. Ammonia
is conserved in the anaerobic environment of a landfill and thus it builds up in leach-
ate over time similar to ions such as chloride and sodium. Figure 16.6 present the
total ammonia leachate concentrations for the DSWA and ACSWL facilities;
the concentration trends are similar to that of chloride (for similar reasons). Total
Kjeldahl Nitrogen (TKN) is also a commonly used nitrogen parameter measured in
wastewater; it represents the sum of the ammonia nitrogen and organic nitrogen
species (the majority in leachate will be ammonium N). Nitrate (NO3−) and nitrite
(NO2−) are other inorganic nitrogen species, and while the concentration of these
ions should be relatively small compared to the total nitrogen content (because of
the anaerobic nature of most landfills), these ions may be important in systems
where air addition is employed. Phosphorous occurs at low concentrations com-
pared to nitrogen. Phosphorous analyses frequently performed include the inorganic
form as well as total phosphorous (TP).
16.4 Gas Volume, Pressure, and Flux 371
The bulk organic and inorganic strength of leachate (along with ammonia-N, which
will principally be present as one of the major ions) dominate treatment consider-
ations. The trace pollutants, however, which occur in much lower concentration, often
dictate regulatory concerns because of their potentially adverse health effects. These
parameters are necessary measurements when determining how a leachate may be
managed outside of the landfill. Examples of trace heavy metals include arsenic,
cadmium, mercury, lead, and zinc, while examples of trace organic compounds
include benzene, vinyl chloride, acetone, and anthracene. While the concentrations
of these chemicals are relatively low compared to the other leachate parameters
discussed, their presence may be important when assessing treatment options and
long-term leachate management options, and when evaluating potential ground
water impacts.
In general, one would expect trace chemical constituents to decrease with time.
Since most of these chemicals are not routinely detected, however, they may not
exhibit trends in the pronounced manner that the bulk constituents do. The fate of
organic trace chemicals will be highly dependent on the specific chemical com-
pound and properties such as volatility, absorption potential, and biodegradability
(Reinhart and Pohland 1991). Some trace metals will be bound within the waste, but
concentrations will be highly dependent on species, pH, and ORP (Kim et al. 2011).
Fig. 16.7 Measurement of landfill gas flow rate as a wellhead using a portable meter
flow measurement device. These systems will normally include instrumentation for
gas composition measurement. The station utilizes a programmable logic controller
that allows establishment of set points to adjust applied vacuum, typically in
response to low CH4 and/or elevated O2 concentrations. These measurements will
typically be data logged. At blower/flare stations or gas-to-energy systems (where
present), combined gas flow rate from the entire collection system is measured
using mass flow meters, with flow rates and cumulative volume continuously
recorded at specified intervals. Figure 16.8 provides an example of such data, show-
ing both total volume of methane collected over time and the percentage of methane
for a landfill practicing liquids.
Several methods are available for monitoring gases at the landfill surface. Some regu-
latory programs require surface CH4 emissions monitoring on a routine basis (typi-
cally four times per year) in areas where gas is being actively extracted. The instrument
used for this monitoring normally consists of an flame ionization detector (FID) or a
photoionization detector PID and the concentrations of interest are much lower than
that produced within the landfill (e.g., 500 ppm is the US-specified surface concentra-
tion limit). This monitoring approach can provide insight regarding areas where high
gas production rates are occurring and/or poor GCCS performance.
374 16 Tools and Techniques for Landfill Monitoring
Fig. 16.8 Gas flow rate and percentage methane measured over time at the gas collection station
of a landfill practicing liquids addition
Fig. 16.9 Basic setup of flux chamber use for measuring landfill surface emissions
Flux chambers (sometimes referred to as flux boxes) have been used at several
landfills to assess surface emissions (Reinhart et al. 1992); these devices measure
the flux of gas emitted through the surface area into the chamber (Fig. 16.9). These
instruments are typically used for research purposes and can provide more robust
data in a specific area when compared to surface monitoring using an FID or a PID;
16.5 Chemical Composition of Gas 375
Fig. 16.10 Illustration of open path technique for measuring surface emissions at a landfill
however, high spatial variability across the landfill surface has been observed when
using flux chambers (Borjesson et al. 2000; Spokas et al. 2003).
Other techniques have been developed to measure the surface emissions flux
from large landfill surfaces. The open-path FTIR technique involves sending a
series of energy waves over the surface of landfill, reflecting them back, and mea-
suring a resulting change in the wave that corresponds to the amount of a particular
gas in the air above the landfill (Fig. 16.10; US EPA 2006b; Thoma et al. 2010).
These measurements can be converted to an emission rate or flux. This technique is
somewhat expensive, but may provide a better estimate of emissions over a large
area compared to single-point flux box measurements. This technique has been
most commonly applied for measuring landfill CH4 surface emissions.
These field devices typically are also equipped with sensors to measure pressure,
flow, and/or temperature at GCCS well heads. N2 concentration is not directly mea-
sured in the field, but is often assumed as comprising the “balance” after subtracting
the concentration of CH4, CO2, and O2, which are normally measured directly.
Samples for laboratory analysis are collected in non-reactive sampling bags (e.g.,
Tedlar) or passivated stainless steel canisters. Both CH4 and CO2 are typically ana-
lyzed in the lab using a gas chromatograph (GC) equipped with a (FID), while O2 is
typically measured with a device equipped with an electrochemical sensor.
A number of trace chemicals are present in landfill gas and Table 16.4 provides a
summary of many trace chemical classes. Trace gases may be of concern for a vari-
ety of reasons. Hydrogen sulfide (H2S) is a problematic gas because of strong odor
and public health issues when emitted to the atmosphere, and when collected high
levels of H2S can create problems with energy production equipment and other
mechanical gas moving devices because the gas can transform to sulfuric acid and
prematurely wear these components. Siloxanes are a group of chemicals that are of
concern at landfills with energy production equipment, as these chemicals can build
up on gas moving equipment and their oxidation product, silicate, can cause prema-
ture wear, similar to H2S.
Another group of chemicals that may be measured is non-methane organic com-
pounds (NMOCs). This is a group of compounds that have the potential to cause a
variety of human health and environmental impacts. These compounds cause the
formation of acid rain, contribute to global warming, and lead to other adverse
effects. The amount of NMOC emissions is one of the factors that dictate whether
an MSW landfill is required to collect actively LFG in the US. Another trace con-
stituent that can be measured is carbon monoxide (CO). The presence of CO can
suggest that subsurface oxidation or smoldering is occurring; this is a concern of
landfills with active GCCS (see Chap. 13) and landfills where air addition is prac-
ticed (see Chap. 14). Although researchers have not established CO levels that
Table 16.4 Examples of trace LFG constituents commonly measured and associated measurement
techniques
LFG chemical or
chemical group Measurement option(s)
CO Colorimetric detector tube (field), electrochemical cell (field), also
laboratory analysis
H2S Colorimetric detector tube, electrochemical cell attachment to LFG
analyzer (field), or laboratory technique (e.g., thermal conductivity
detector)
NMOCs Laboratory analysis (sample collection in passivated stainless steel
canister)
Siloxanes Laboratory analysis (different sample collection methods include collection
in passivated stainless steel canister and the use of in-line midget
impingers)
16.6 Landfill Volume, Density, and Topography 377
Fig. 16.11 Use of GPS technology to measure surface elevation and location
378 16 Tools and Techniques for Landfill Monitoring
Density relates the mass of a media to the volume it occupies; specific weight relates
the weight of a medium to volume (see Chap. 5 for a description of the relationship
between the terms). Specific weight is an important parameter to track at landfills as
it reflects the efficiency of airspace utilization for a landfill unit. Most commonly, the
specific weight is estimated by measuring the weight of incoming waste loads depos-
ited in the landfill and estimating the volume of utilized airspace capacity in that
same time frame based upon surface topography data. This type of measurement,
however, is not the true value for the landfilled waste materials as it does not include
the weight of the cover soil (which is not normally measured in routine landfilling
operations). Another complicating factor is that waste volume changes (settles)
through both physical and biological mechanisms (see Chap. 5). As described in
Chap. 5, it is common to track the apparent density (or specific weight) at a landfill
site—this represents the mass (or weight) of disposed waste per volume of landfill
space (waste plus soil) and is commonly used in landfill capacity projections.
Specific weight or density can also be calculated by excavating or augering
material from a landfill, weighing the removed material, and applying a measured
or estimated volume of the excavation (Zekkos et al. 2006). Borehole measure-
ments using this technique have found, as expected (see Chap. 5), that waste mass
(waste plus soil) densities increase with depth within the landfill and increase with
decomposition (as the heavier soil becomes more prominent; Jang 2013).
Fig. 16.12 Illustration of settlement plates used for measuring elevation changes of points within
a landfill
well for closed or inactive landfills that have reached a point in operation where
surface conditions no longer change, or change slowly. At a site where surface
topography continues to evolve through waste filling or other changes (e.g., soil
placement), routine survey techniques may have limited utility for tracking tempo-
ral settlement unless these additional material surcharges are specifically tracked
and subtracted from previous measurement points.
Several approaches have been utilized for tracking the elevation of locations
buried within a landfill even as waste placement continues. Settlement plates have
been used at some sites (e.g., Jang 2013), and consist of flat plates connected to
vertical rods (settlement bars) that are placed on the desired location at a point on
the landfill surface. Prior to placing waste on top of the plates and around the rods
(which must be performed very carefully to avoid damage), a solid pipe (casing
tube) is installed over the rod so that the in the future, settlement measurements at
the top of the rod correspond directly to that occurring at the top of the plate
(Fig. 16.12). When several settlement plates are installed at different depths in the
landfill, measurements of elevation at the top of each rod can be used to estimate the
settlement of different layers.
Electronic instruments can also be used to measure elevation changes at specific
points within a landfill. Some vendors manufacturer transducers designed to be
placed in subsurface environments to measure elevation changes. In some cases the
device is permanently located in a buried pipe or similarly protected location, while
in other cases the devices are periodically inserted into and moved through a length
380 16 Tools and Techniques for Landfill Monitoring
Fig. 16.13 Illustration of buried conduit application for measuring internal elevation changes
of buried pipe. Figure 16.13 illustrates how elevation changes in a buried pipe could
be used to measure the settlement of the underlying media (waste and/or soil).
One type of instrument that could be used, either as a permanent device or one
that is moved through a buried conduit, is a pressure transducer with a measurement
tip connected to an external reservoir of liquid maintained at a known or measured
elevation outside the landfill (the reservoir would require refilling as needed). The
transducer provides a measurement of the difference in elevation between the trans-
ducer location and the reservoir surface (Fig. 16.14a). An inclinometer is another
type of instrument useful for measuring elevation changes when passed through a
conduit. As the inclinometer is moved through the conduit, the angle of the instru-
ment is recorded as a function of location within the conduit (Fig. 16.14b). When
the resulting elevation profile is compared to a previous profile from an earlier time,
the settlement occurring over that time increment can be calculated.
Slopes are routinely measured as part of surface topography surveying. Other slope
measurements might also be used to assess the slopes of pipes that are constructed
to provide gravity drainage and to monitor side slopes for potential movement. As
described in Chap. 10, both the base grade of a landfill liner and the collection pipes/
trenches are sloped to provide gravity drainage of leachate to low points in the land-
fill (for removal). LCRS pipes are often inspected and cleaned. The slopes of theses
pipes can be assessed using instruments such as inclinometers or settlement cells as
described in the previous section. In Chap. 12, the importance of slope stability was
16.7 Excavated Solids Properties 381
Fig. 16.14 Methods for measuring elevation (and settlement) using a buried conduit. (a) Pressure
transducer connected to an external fluid reservoir and (b) slope inclinometer
A direct method that can be used to examine the relative state or degree of decom-
position of landfilled waste involves the collection and analysis of physical samples.
This technique has been utilized at a number of facilities practicing sustainable
landfilling (Townsend et al. 1996; Mehta et al. 2002; Kelly et al. 2006; Kim and
Townsend 2012). The difficulty with this approach is that waste sample collection
is an intrusive, expensive, and time-intensive operation. Most methods for collect-
ing a waste sample involve auguring into the landfill with a mechanical drill rig and
retrieving samples from a specific depth; multiple samples from different depths can
382 16 Tools and Techniques for Landfill Monitoring
be collected from the same borehole. Considerations for developing a solids sam-
pling plan include the number of samples desired, the location of the samples, the
mass of samples needed for analysis, and cost of the collection procedure. A key
issue to consider when developing a sampling plan is determining how best to
obtain samples representative of the landfill location targeted and obtaining a statis-
tically significant number so that observations or conclusions drawn from the solids
analysis are valid.
Fig. 16.16 Methane yield results using BMP assay for excavated landfill samples from two facilities
a techniques for assessing landfill waste stabilization. The ratio of cellulose and
hemicellulose (C + H) to lignin (L) decreases with time. Analysis of these constitu-
ents usually involves contacting the sample with an acid solution to dissolve C + H
(Barlaz 2006). The VS content of the remaining solids (once plastics are removed)
is considered lignin, and the acid reaction is analyzed for component sugars of
C + H. Figure 16.17 presents data from several studies where both BMP and cellu-
lose and/or lignin analyses were performed. Figure 16.17a shows BMP versus cel-
lulose the Yolo County (Mehta et al. 2002) and Outer Loop landfills (US EPA 2006a)
(see Chap. 4) and over a dozen landfills studied by Virginia Tech University (Bricker
2009). Figure 16.17b compares BMP versus (C + H)/L for samples from the Yolo
County and Outer Loop landfill sites.
The microbial processes responsible for waste decomposition are exothermic; tem-
peratures within a landfill may thus be elevated relative to ambient temperatures,
especially when biologically activity is at a maximum. As described in Chaps. 13
and 14, aerobic microbial respiration releases more heat than anaerobic systems,
and thus temperature measurement is of critical importance during or at landfills
with active gas extraction to prevent subsurface oxidation or fires from forming.
Internal landfill temperatures have been reported in a multitude of studies where it
can be seen that they can reach as high as 170 °F or more (Townsend et al. 1996;
Powell 2005; Yazdani et al. 2010; Hanson et al. 2010).
386 16 Tools and Techniques for Landfill Monitoring
Fig. 16.17 (a) BMP methane yield results as a function of cellulose, (b) BMP methane yield
results as a function of the ratio of cellulose plus hemicellulose to lignin ((C+H)/L)
Fig. 16.18 Resistivity-based moisture sensor utilized by Gawande et al. (2003) and Kumar et al. (2009)
Yuen et al. (2000) examined the use of neutron probes at a landfill in Australia prac-
ticing leachate recirculation. Seven aluminum access tubes were installed and a
neutron probe was used to measure surrounding moisture content using a calibra-
tion curve produced using sand. While the technique was found successful at assess-
ing relative moisture levels, the technique did not provide a measurement of actual
waste moisture content.
Gawande et al. (2003) reported on an electrical resistance moisture sensor for use
in landfills. A stainless steel rod embedded in a granular matrix surrounded by stain-
less steel mesh was used; electrical resistance across the granular media decreased
as moisture content in the media increased (Fig. 16.18); a thermocouple wire was
included with the sensor for temperature measurement. Calibration curves (a func-
tion of temperature and solution ionic strength) were developed to relate resistance
to surrounding moisture content. Kumar et al. (2009) reported the results of the
field-scale application of these sensors after 6 years of operation. The sensors pro-
vided a reasonable estimate of local moisture content when appropriately calibrated,
but did not provide representative estimates of the landfill moisture content as a
whole. This was concluded to be a result of preferential channeling of liquids to
sensor location, likely a result of the boreholes used to install the sensors.
Time domain reflectometry (TDR) sensors work on the principal that the bulk
dielectric permittivity of a medium is related to its moisture content. The dielectric
permittivity of water is much greater than MSW; when an appropriate calibration
curve is developed, TDR probes installed within landfilled waste can be used to
estimate moisture content (Masbruch and Ferre 2003; Li and Li 2011). Li and Zeiss
16.8 In Situ Moisture, Temperature, and Pressure 389
(2001) used waste from loads of residential garbage to pack columns that included
TDR probes; calibration curves were developed by adding incremental volumes of
water to increase moisture content. They evaluated the effects of waste properties
and leachate ionic strength and concluded TDR to be a viable method for measuring
in-situ moisture content in MSW. Jonnalagadda et al. (2010) compared in-situ
resistivity and TDR sensors at an operating landfill where liquids addition was prac-
ticed. While both technologies were observed to measure transient moisture changes
in the landfill, magnitudes of moisture content measured were higher than those
predicted using mass balance. The resistivity sensors were found to be less expen-
sive, easier to install, and more reliable.
Several researchers have examined the use of geophysical techniques such as
electrical resistivity tomography (ERT) for assessing the presence of moisture in
landfills (Gueerin et al. 2004; Clément et al. 2010; Hossain et al. 2011; DeCarlo
et al. 2013). In ERT, electrical resistance between electrodes at different spatial
locations is measured with results providing information regarding the media in
between the electrodes, including the presence of moisture. While ERT can utilize
boreholes, where it has been particularly attractive for landfills is as a surface geo-
physical technique. Surface ERT can provide an image of subsurface conditions and
locations of elevated moisture.
The partitioning gas tracer test (PGTT) provides an estimate of moisture content
in a region of landfill between two points used for gas tracer injection and gas
extraction (Imhoff et al. 2003; Han et al. 2006). In using PGTT, conditions in the
landfill are created in which the addition and recovery of two different gas tracers
are measured and compared (for example, between two different wells). The differ-
ence in travel time of each tracer can yield an estimation of the degree of saturation,
and with an estimate of waste density and porosity, can be used to estimate moisture
content. Knowledge of temperature in the area of the test is necessary, and while
ionic strength of the pore water affects results, this impact have been found to be
small for some tracers.
With the exception of geophysical techniques such as surface ERT, a primary
challenge in the use of buried moisture sensors is installation in the landfill without
damaging the sensors and the associated cables, and without creating conditions
around the sensor that would encourage short-circuiting of liquids to the sensor. The
two options for installation include placement as the waste is deposited and com-
pacted in the landfill and installation after waste placement by excavating (or drill-
ing) into the waste.
Placement during landfill operation involves excavating a small area of the sur-
face of waste lift, placing the sensor in the excavation, and then backfilling around
the waste with an appropriate protective material, and then covering with waste;
waste filling would proceed and eventually new waste would be placed on top of the
instrumented area. The wires are routed to an appropriate terminal point, normally
in trenches excavated on the surface, possibly in protective conduit. Efforts should
be considered to minimize any preferential fluid flow along the wires and/or con-
duits back to the sensor. This approach does not require any specialized equipment
for installation and minimizes the potential for preferential channels that might oth-
erwise short-circuit moisture to the instrument. Because this approach occurs
390 16 Tools and Techniques for Landfill Monitoring
throughout the landfill operational period, however, it has a greater potential for
interference with routine operations and damage from equipment and vehicles, and
thus demands careful planning and coordination.
The second approach involves drilling into the landfill and placing the sensor, or
a series of sensors in the borehole. In this case, the instrument cables will be run up
to the top of the landfill and then connected to an appropriate monitoring point. This
approach minimizes interference with normal landfill operations as it takes place
after waste placement has been largely completed. An outside contractor will nor-
mally be required as drilling into the landfill will be necessary. One of the biggest
challenges of this approach is the natural channel for moisture short-circuiting
resulting from the borehole. It is imperative that areas of the borehole not occupied
by the sensor be backfilled with a low permeability material such as clay or grout.
Internal pressures, either pore pressure (the combined liquid and gas pressure in the
pore space) or the pressures exerted by the weight of the landfilled mass, have been
measured at several landfills. This type of measurement normally utilizes electronic
pressure transducers buried within the landfill. These transducers are connected via
a cable to an external power source and an output measurement or recording device.
Pore pressure readings from buried transducers can be used to assess moisture
(Kadambala et al. 2011) and gas (Ko et al. 2013; Larson et al. 2012) movement and
magnitude and thus help assess effectiveness of liquid or air addition and gas extrac-
tion systems. Important to recognize in the interpretation of such data is that the
measured pressure represents a combination of both liquid and gas pressure. For
example, pressure transducers have been proposed as a measurement tool for liquid
head on the liner; the resulting pressure, however, constitutes the depth of liquid on
the liner plus the gas pressure above the leachate.
Pressure transducers designed to handle burial under applied loads should be
specified. Pressure transducers designed for submersion in water may not be able to
withstand the forces exerted by the overlying waste mass. Installation can be diffi-
cult, both in terms of appropriate burial to prevent damage, securing cables in a
manner to avoid damage, and preventing preferential paths for fluid flow that might
influence the results. The challenge of short circuiting of liquid and gas flow is simi-
lar to that discussed for the installation of buried moisture sensors.
Larson et al. (2012) installed pressure transducers within a leachate recirculation
trench and at multiple points away from the trench. Prior to liquids addition, changes
in internal gas pressure in response to barometric pressure fluctuations were used to
estimate waste permeability. Later, these transducers were used to examine pressure
changes as a result of liquids addition. In this case, the transducers were installed
during waste filling. The trenches were first excavated, then sensors encased in a
sand-filled cloth bag were placed in the excavation and wires were placed in electri-
cal conduits day-lighted out the side of the landfill (see Fig. 16.19). Waste was carefully
16.8 In Situ Moisture, Temperature, and Pressure 391
Fig. 16.19 Pressure transducers for measuring internal landfill pore pressures installed by burying
in sand-filled bags within the waste as filled
392 16 Tools and Techniques for Landfill Monitoring
Fig. 16.20 Pressure transducers for measuring internal landfill pore pressures installed by place-
ment in grouted borehole
placed on top of the excavation until enough overburden material was present to
avoid crushing of the transducer by heavy equipment.
Kadambala et al. (2011) placed pressure transducers at different elevations radi-
ally surrounding a vertical liquids addition well. Results were used to examine
moisture and pressure distribution, and provided insight on waste anisotropy (i.e.,
the degree of directional dependence in terms of permeability). Transducers were
placed in existing waste by excavating a borehole and inserting the instrument
(Fig. 16.20). In this case, the transducers were attached to the side of a pipe and
lowered into an augured hole with the wires exiting the hole through the center of
the pipe. To prevent short-circuiting, the hole was then filled with a bentonite-
cement slurry.
Pressure transducers can also be configured to measure the total weight resulting
from overlying landfill material (overburden pressure). Total earth pressure cells
(TEPC) consist of pressure transducers connected to round, flat plates containing a
hydraulic fluid. The greater the overlying weight of the material above the plates,
the more pressure is exerted on the transducer. Timmons et al. (2012) examined
changes in landfill weight (overburden pressure) with time by installing TEPC in
the LCRS of a lined landfill. Changes in overburden pressure were observed with
the placement of waste lifts.
References 393
References
Barlaz M (2006) Forest products decomposition in municipal solid waste landfills. Waste Manag
26:321–333
Borjesson G, Danielsson A, Svensson B (2000) Methane fluxes from a Swedish landfill deter-
mined by geostatistical treatment of static chamber measurements. Environ Sci Technol
34(18):4044–4050
Bricker G. (2009). Analytical methods of testing solid waste and leachate to determine landfill
stability and landfill biodegradation enhancement. Master of Science thesis, Virginia
Polytechnic Institute and State University, Blacksburg
Chu L, Cheung K, Wong M (1994) Variations in the chemical properties of landfill leachate.
Environ Manag 18:105–114
Clément R, Descloitres M, Günther T, Oxarango L, Morra C, Laurent JP, Gource JP (2010)
Improvement of electrical resistivity tomography for leachate injection monitoring. Waste
Manag 30:452–464
DeCarlo L, Perri MT, Caputo MC, Deiana T, Vurro M, Cassiani (2013) Characterization of a dis-
missed landfill via electrical resistivity tomography and mise- à-la-masse method. Journal of
Applied Geophysics 98:1-10
Gawande NA, Reinhart DR, Thomas PA, McCreanor PT, Townsend TG (2003) Municipal solid
waste in situ moisture content measurement using an electrical resistance sensor. Waste Manag
23:667–674
Gueerin R, Munoz ML, Aran C, Laperrelle C, Hidra M, Drouart E (2004) Leachate recirculation:
moisture content assessment by means of a geophysical technique. Waste Manag 24:785–794
Han B, Jafarpour B, Gallagher VN, Imhoff PT, Chiu PC, Fluman DA (2006) Measuring seasonal
variations of moisture in a landfill with the partition gas tracer test. Waste Manag 26:344–355
Hanson J, Yesiller N, Oettle N (2010) Spatial and temporal temperature distributions in municipal
solid waste landfills. J Environ Eng ASCE 136(8):804–814
Hossain S, Kemler V, Dugger D, Manzur SR, Penmethsa KK (2011) Monitoring moisture move-
ment within municipal solid waste in enhanced leachate recirculation landfill using resistivity
imaging. Sustain Environ Res 21(4):253–258
Imhoff PT, Jakubowitch A, Briening ML, Chiu PC (2003) Partitioning gas tracer tests for measure-
ment of water in municipal solid waste. J Air Waste Manag Assoc 53:1–10
Imhoff PT, Reinhart DR, Englund M, Guérin R, Gawande N, Han B, Jonnalagadda S, Townsend
TG, Yazdani R (2007) Review of state of the art methods for measuring water in landfills.
Waste Manag 27:729–745
ITRC (Interstate Technology & Regulatory Council) (2005) Characterization, design, construc-
tion, and monitoring of bioreactor landfills. Technology & Regulatory Council, Alternative
Landfill Technologies Team ALT-3, Washington
Jang Y (2013) Field-monitored settlement and other behavior of a multi-stage municipal waste
landfill, Korea. Environ Earth Sci 69:987–997
Jonnalagadda S, Kumar D, Jain P, Gawande N, Townsend TG, Reinhart DR (2010) Comparison of
resistivity and time domain reflectometry sensors for assessing moisture content in bioreactor
landfills. ASTM Geotech Test J 33:183–191
Kadambala R, Townsend TG, Jain P, Singh K (2011) Temporal and spatial pore water pressure
distribution surrounding a vertical landfill leachate recirculation well. Int J Environ Res Public
Health 8:1692–1706
Kelly R, Shearer B, Jongmin K, Goldsmith CD, Hater G, Novak J (2006) Relationships between
analytics methods utilized as tools in the evaluation of landfill waste stability. Waste Manag
26:1349–1356
Kim H, Townsend TG (2012) Wet landfill decomposition rate determination using methane yield
results for excavated waste samples. Waste Manag 32:1427–1433
Kim H, Jang Y-C, Townsend T (2011) The behavior and long-term fate of metals in simulated
landfill bioreactors under aerobic and anaerobic conditions. J Hazard Mater 194:369–377
394 16 Tools and Techniques for Landfill Monitoring
Kjeldsen P, Barlaz MA, Rooker AP, Baun A, Ledin A, Christensen TH (2002) Present and long-
term composition of MSW landfill leachate: a review. Crit Rev Environ Sci Technol 32(4):
297–336
Ko JH, Powell J, Jain P, Kim H, Townsend TG, Reinhart D (2013) Case study of controlled air
addition into landfilled municipal solid waste: design, operation, and control. J Hazard Toxic
Radioac Waste 17:351–359
Kumar D, Jonnalagadda S, Jain P, Gawande NA, Townsend TG, Reinhart DR (2009) Field evalu-
ation of resistivity sensors for in situ moisture measurement in a bioreactor landfill. Waste
Manag 29:1547–1557
Larson J, Kumar S, Gale SA, Jain P, Townsend T (2012) A field study to estimate the vertical gas
diffusivity and permeability of compacted MSW using a barometric pumping analytical model.
Waste Manag Res 30:276–284
Li R, Li L (2011) A review of TDER applications for in situ monitoring of bioreactor landfills.
Sustain Environ Res 21(4):211–218
Li RS, Zeiss C (2001) In-situ moisture content measurement in MSW landfills with TDR. Environ
Eng Sci 18(1):53–66
Masbruch K, Ferre TPA (2003) A time domain transmission method for determining the depen-
dence of the dielectric permittivity on volumetric water content: an application to municipal
landfills. Vadose Zone J 2:186–192
Mehta R, Barlaz M, Yazdani R, Augenstein D, Bryars M, Sinderson L (2002) Refuse decomposi-
tion in the presence and absence of leachate recirculation. J Environ Eng ASCE 128(3):
228–236
Owen WF, Stuckey DC, Healy JB Jr, Young LY, McCarty PL (1979) Bioassay for monitoring
biochemical methane potential and anaerobic toxicity. Water Res 13:485–492
Owens JM, Chynoweth DP (1993) Biochemical methane potential of municipal solid-waste
(MSW) components. Water Sci Technol 27:1–14
Pohland F, Cross W, Gloud J, Reinhart D (1992) Behavior and assimilation of organic and inor-
ganic priority pollutants codisposed with municipal refuse. EPA-600/R-93-137a. US EPA,
Cincinnati, Ohio, USA.
Powell, J (2005) Trace Gas Quality, Temperature Control and Extent of Influence from Air
Addition at a Bioreactor Landfill. UF Thesis, Gainesville, Florida
Reinhart D, Pohland F (1991) The fate of selected organic pollutants codisposed with municipal
refuse. Water Pollut Control Federation Res J 63(4):780
Reinhart D, Cooper D, Walker B (1992) Flux chamber design and operation for the measurement
of municipal solid waste landfill gas emission rates. Air Waste Manag Assoc J 42(8):1067
Rice EW, Baird RB, Eaton AD, Clesceri LS (2012) Standard methods for the examination of water
and wastewater, 22nd edn. American Public Health Association, American Water Works
Association, Water Environment Federation, Washington
Spokas K, Graff C, Morcet M, Aran C (2003) Implications of the spatial variability of landfill
emission rates on geospatial analyses. Waste Manag 23:599–607
Thoma E, Green R, Hater G, Goldsmith D, Swan N, Chase M, Hashmonay R (2010) Development
of EPA OTM 10 for landfill applications. J Environ Eng 136(8):769–776
Timmons J, Cho YM, Townsend T, Berge N, Reinhart D (2012) Total earth pressure cells for mea-
suring loads in a municipal solid waste landfill. Geotech Geol Eng 30:95–105
Tolaymat T, Kremer F, Carson D, Davis-Hoover W, (2004) Monitoring approaches for landfill
bioreactors. National Risk Management Research Laboratory Office of Research and
Development U.S. Environmental Protection Agency EPA/600/R-04/301 Cincinnati, Ohio
Townsend TG, Miller WL, Lee HJ (1996) Acceleration of landfill stabilization using leachate
recycle. J Environ Eng 122:263–268
US EPA (2003) Landfills as bioreactors: research at the outer loop landfill, Louisville, Kentucky,
first interim report. United States Environmental Protection Agency, Cincinnati, Ohio, USA
US EPA (2006a) Landfill bioreactor performance: second interim report, outer loop recycling and
disposal facility, Louisville, Kentucky. EPA/600/R-07/060 U_S_E_P_A_ Cincinnati, Ohio, USA
References 395
US EPA (2006b) Optical remove sensing for emission characterization from non-point sources.
EPA Test Method OTM 10. US EPA Office of Air Quality Planning and Standards, Durham,
NC
US EPA (2013) Test methods for evaluating solid waste, physical/chemical methods. SW-846.
United States Environmental Protection Agency, Cincinnati, Ohio, USA
Wang Y, Byrd C, Barlaz M (1994) Anaerobic biodegradability of cellulose and hemicellulose in
excavated refuse samples using a biochemical methane potential assay. J Ind Microbiol
13:147–153
Yazdani R, Mostafid ME, Han B, Imhoff PT, Chiu P, Augenstein D, Kayhanian M, Tchobanoglous
G (2010) Quantifying factors limiting aerobic degradation during aerobic bioreactor landfill-
ing.Environ Sci Technol 44:6215–6220
Yuen ST, McMahon TA, Styles JR (2000) Monitoring in situ moisture content of municipal solid
waste landfills. J Environ Eng 126(12):1088–1095
Zekkos D, Bray J, Kavazanjian E, Matasovic N, Rathje E, Reimaer M, Stokoe K (2006) Unit
weight of municipal solid waste. J Geotech Geoenviron Eng ASCE 132(10):1250–1261
Chapter 17
Final Landfill Disposition
When the disposal capacity of a landfill site, or a specific operational area of a land-
fill, is reached, several decisions regarding how to manage these areas must be eval-
uated. The term closure designates the process of finalizing waste surface
configuration and installing infrastructure designed as the final containment and
control system for this area of waste. Post-closure care (PCC) refers to activities
performed to operate and maintain closed areas so that desired performance and
environmental protection are accomplished.
Landfill operators at all facilities must plan for and implement closure and post-
closure care, but when implementing sustainable landfill technologies, owners and
operators have additional options to consider. This chapter reviews such consider-
ations, starting with a description of typical closure procedures followed by a discus-
sion of specific concerns and opportunities for landfill operators practicing liquids
addition and rapid waste stabilization. One major issue is the determination of when
landfill operation and activity are considered complete, such that post-closure main-
tenance and operation can be ceased, or at least reduced in frequency. As described
in Chap. 3, a primary objective for some facility owners attempting accelerated waste
stabilization is early completion of PCC activities. An additional topic is whether a
landfill practicing sustainable technologies might not need to be closed in the same
fashion as traditional disposal facility, but perhaps instead stabilized materials can be
reclaimed at the end of operation and reused either at the landfill or off site.
While many existing landfilled elements will be integrated into the design of a land-
fill closure system (e.g., gas collection, leachate management, stormwater control),
a substantial new feature is the final landfill cover, often referred to as a cap. The
primary objective of a landfill cap is to minimize rainwater entry into the landfill as
a means to reduce future leachate production. Another major function is to aid in the
control of landfill gas. At some facilities, final cover systems are only installed after
an entire landfill unit has reached its ultimate configuration and surface elevation. In
other cases, distinct areas of the landfill unit are closed while operation in other
areas continues. Considerations in determining which approach to pursue are dis-
cussed later in this chapter.
In general, two types of final cover systems are designed and constructed: the
barrier layer approach and the capillary layer approach. A barrier layer cover system
relies on a low-permeability material (e.g., geomembrane, compacted soil) to pre-
vent liquids from entering the landfill. The cover must be designed so the intercepted
rainwater is routed off of the surface of the landfill via overland flow, channelized
flow, or subsurface flow above the barrier layer. This type of design is common in
modern landfill operations and is often the technique specifically required by regula-
tion. Barrier systems integrate well with a GCCS, as collected gases under the geo-
membranes or compacted soil can be extracted at designated exit points. The intent
of a capillary layer system is to promote evaporation and transpiration of infiltrating
rainwater. Under certain climatic conditions, if an appropriate thickness and grada-
tion of soil is selected, along with selected vegetation, infiltrating moisture can be
retained within the cover soil layer until it is removed through evapotranspiration.
Figure 17.1 provides typical cross-sections of both cover system types.
Components of a cover system include layers of different soils (and possibly
geosynthetics), each selected to serve a desired function. The final soil layer above
the waste is applied and graded to smooth out uneven spots and to provide needed
slope for moisture drainage. This layer is important to protect other cover system
components from damage by underlying waste; excavated on-site soils are com-
monly used. Since gas will accumulate beneath a barrier layer (resulting in higher
gas pressures and thus potentially high gas flux through the cap), a gas venting layer
is constructed using coarse sand or a similar material that will not harm overlying
geomembranes and promotes gas movement to designated extraction or exit points.
Materials used for barrier layer construction are similar to those used for landfill
17.2 Elements of the Closure and Post-closure Process 399
Fig. 17.1 Components of a typical landfill final cover system (a) barrier layer approach (b) capil-
lary layer approach
When a barrier layer is included in the cover system, a drainage layer must be
provided on top of the barrier to promote removal of infiltrating rainwater. If exces-
sive water builds up above the barrier layer, leakage may result and the mechanical
stability of the cover may be reduced. The top cover soil layer (infiltration layer)
consists of soils that promote plant growth and allow for water retention and even-
tual evapotranspiration. Appropriate vegetation includes shallow-rooted plants con-
sisting primarily of grasses that can help control erosion. A well-vegetated landfill
surface is important to promote overland flow of water to stormwater collection
points and to minimize soil loss. For a capillary layer system, the barrier layer is
substituted with an infiltration layer designed with a sufficient depth and moisture
retention capacity to promote necessary evapotranspiration.
may result in failure to direct stormwater properly off the site. In instances where
erosion problems are noted or drainage control structures need to be repaired, proper
maintenance procedures should be implemented immediately to prevent further
damage. Failure to maintain the physical integrity of the landfill cover will promote
additional infiltration into the landfill and eventually cause generation of larger
leachate quantities. This will also exacerbate problems associated with leachate col-
lection and disposal.
Record keeping requirements include site inspections and summary reports at
some specified frequency during the years following closure. For instance, quanti-
ties of leachate removed and transported must be recorded, and monitoring of gas,
groundwater, surface water, and leachate are commonly required. As described in
Chap. 16, monitoring landfill gas and leachate provides valuable information about
the landfill’s conditions, and as discussed in a following section, will be instrumen-
tal in determining whether or not PCC criteria are met.
The LCRS and GCCS will continue to be operated after closure and therefore
will require attention during PCC. Both systems must be maintained to ensure
effective operation. LCRS maintenance includes periodic leachate collection pipe
cleaning, collection tank cleaning, and pump preventative maintenance and repairs.
Collected leachate must be treated or disposed of in an appropriate manner, and the
quantity of leachate treated or removed should be recorded. GCCS maintenance
will consist of regular maintenance of pipes, hoses, wellheads, blowers, pumps, and
other infrastructure. Withdrawal pipes and collection lines may require condensate
removal and repairs if damage from differential settlement occurs.
The point of transition from an active, operating landfill to a closed facility depends
on site-specific conditions, operating objectives, and regulatory requirements.
Operators have pursued several different approaches with respect to implementing
the initiation of closure. One approach is to delay closure construction as long as
possible; waste filling continues, expanding laterally in new disposal areas as neces-
sary, with a final cover system constructed over a very large areas, often the entire
landfill unit. Another approach involves bringing distinct sections of the landfill to
final topographic conditions as soon as possible and closing these areas as part of
individual construction projects.
The first such approach is illustrated in Fig. 17.3. A landfill with the capacity to
dispose of 15 years of waste is filled to a specified waste height that is short of the
permitted final topography. Waste filling progresses laterally until the specified
waste height is reached, and then the entire landfill is filled to the permitted waste
height. A closure system is then installed for the entire landfill. This approach has
17.3 Closure Considerations for Sustainable Landfills 403
Fig. 17.3 Illustration of landfill final cover system installation after entire landfill reaches final
permitted elevation
Fig. 17.4 Illustration of landfill final cover system installation throughout operation
Fig. 17.5 Exposed geomembranes cap used as final cover (Photo courtesy of Jones Edmunds)
EGCs and similar systems have the potential to be used at any landfill as a
replacement for traditional final cover systems; this would require regulatory
approval, however, as this approach differs from those prescribed in most regula-
tions. With respect to sustainable landfilling, EGCs might serve as temporary cover
systems prior to later waste filling, reclamation, or placement of a final soil layer.
Since EGCs have successfully been used as temporary covers at many landfills, the
major determining factor when considering EGC deployment is cost. If an EGC
must later be removed to install a traditional final cover system, EGC benefits are
likely outweighed by the added cost. However, if the EGC can serve as a replace-
ment for all or part of the required closure system, such an approach might be
feasible.
With the placement of the final cover system, the volume of leachate produced
should decrease. Continuation of leachate recirculation or liquids addition will cer-
tainly affect post closure leachate production, but once all major moisture inputs are
stopped, if the final cover system is well designed, constructed, and maintained,
leachate production should decrease to a relatively small constant rate. Leachate
collection volumes from well-maintained cover systems should not be subject to
major fluctuations in response to wet weather, and should decline or remain rela-
tively constant. If such variations are encountered, the integrity of the cap should be
investigated to determine continuing sources of moisture intrusion and these prob-
lems addressed.
The PCC plan will outline steps necessary for operating, maintaining and moni-
toring the performance of the LCRS. The ultimate goal will be to reduce or elimi-
nate LCRS operation; steps that would need to be considered are described in the
following section. Such decisions would be made based on information on both the
amount of leachate produced and the chemical quality of the leachate. Chapter 2
illustrates leachate chemistry changes with time; after biological consumption of
the readily biodegradable organic matter in the waste and leachate, dominant leach-
ate constituents included refractory organic matter (large molecular weight humic
and fulvic compounds), inorganic ions (chloride, sodium) and ammonia-nitrogen.
As described in Chap. 11, conventional biological wastewater treatment is largely
ineffective for reducing chemical constituents in mature leachate (other than possi-
bly ammonia), and more effective treatment strategies include dilution (addition to
a domestic POTW or discharge to water bodies), physical-chemical treatment pro-
cesses (coagulation/precipitation or carbon absorption for organic matter), and con-
centration (evaporation or membrane processes). Chapter 11 also describes several
leachate treatment technologies that have the potential to work well when coupled
with leachate recirculation.
In a similar manner as the LCRS, the GCCS must be operated until requirements
for the PCC permit are met. Landfills where enhanced waste stabilization is practi-
cal may reach this point much sooner than a traditional landfill (see Chap. 13). As
gas production decreases with time, the required vacuum will decrease and neces-
sitate adjustment at the individual wellheads and the blower station. At some point,
designated wells will be removed from the collection network when they are shown
to be unproductive. An ultimate goal is to switch the gas system operation from
active to passive; the process for making this decision is outlined in the following
section.
Once passive control is instituted, remaining gas emissions could potentially be
addressed by installing passive wells (these can be equipped with solar sparking
devices that combust built-up gases with or without an external fuel) or wind-driven
extractors. Chapter 13 discussed the potential for biocovers to act as a polishing step
to mitigate methane and other gas emissions; such options should be considered as
part of the GCCS and final cover design. Additionally, as described in Chap. 14, a
GCCS may be retrofitted to serve as a system that aerates the landfill to further
reduce potential methane emissions. Allowing passive aeration via the LCRS at the
same time might encourage additional in-situ leachate treatment.
408 17 Final Landfill Disposition
Major PCC activities include maintaining the integrity and effectiveness of the final
cover, maintaining and operating the LCRS, maintaining and operating the GCCS,
and monitoring the groundwater quality. All of these processes are necessary to
ensure the objective of environmental safety is met, but they do come at an expense
to the owner, and ideally would cease or be reduced to a point when risk is suffi-
ciently reduced. A critical element in the PCC process is thus defining the length of
time that the landfill owner and operator must comply with the PCC plan and con-
tinue PCC activities. Because many regulatory programs require financial assurance
(see Chap. 2) to guarantee availability of resources for PCC, the PCC period has a
major impact on landfill economics. The US federal regulations specify a post-
closure care period of 30 years after site closure, although less than 30 years is
allowed if the landfill owner can demonstrate that the reduced period is sufficient to
protect human health and the environment (US Government 2012). Similar to the
US regulatory framework for PCC, the European Landfill Directive specifies that
landfill monitoring and maintenance during PCC should be conducted for as long as
the facility poses a hazard (European Council 1999).
Defining when a landfill poses an acceptable risk to human health and the envi-
ronment is a challenge as this term is subjective and often not well defined in regula-
tory programs. Government agencies provide guidance to assessing risk from closed
or abandoned waste sites in more general terms (e.g., US EPA 1989, 1996, 1998).
This process typically involves assessing the risk posed by current and future emis-
sions from a waste site, and considers risk pathways such as contaminant release to
water supplies, soil and air, and may include an evaluation of risk to ecosystems or
specific ecological receptors. The question that must be addressed is whether or not
the facility, in its current and future state, will result in unacceptable risk if PCC
activities are altered.
Various approaches have been proposed to assess the risk to human health and
the environment that ultimately can be used as a part of the demonstration needed to
establish an appropriate and technically sound post-closure care period (Barlaz
et al. 2002; ITRC 2006; Morris and Barlaz 2011). These approaches suggest a
framework where landfill emissions that may pose a risk to human health and the
environment are monitored as part of PCC and compared to accepted risk levels to
determine when a change in PCC is warranted. In some cases these landfill emis-
sions are measurements of chemical concentration (e.g., gas, leachate), or estimates
of contaminant mass release rate (a combination of the flow rate and contaminant
concentration). Figure 17.7 illustrates such an approach (ITRC 2006), where PCC
monitoring data are collected and evaluated as part of a modular assessment of four
primary landfill components (leachate collection and control system, landfill gas
collection and control system, groundwater monitoring system, and cap system).
In this approach, data are gathered and evaluated, and when the results suggest that
a change in PCC activity is justified (e.g., a less-frequent maintenance or monitoring
schedule), the change is implemented (Fig. 17.8). For example, a portion of the GCCS
might be converted from an active collection system to a passive system; methane
17.4 Determination of End of Post-closure Care 409
surface emissions in the period following the change would continue to be monitored.
Under this approach, confirmatory and surveillance monitoring would allow the oper-
ators to ensure that reduced maintenance and monitoring do not result in an unaccept-
able risk to human health and environment. The modular approach allows for a
phased reduction of PCC activities for only those components of the landfill system
that warrant them. A likely outcome of this approach is not that the operator com-
pletely stops PCC, but rather the facility evolves into routine and potentially reduced
long-term care.
Landfill operators implementing sustainable practices are offered the potential for
reaching a custodial care phase more rapidly than conventional landfills. A landfill
operated to enhance waste stabilization during its operating life should at some point
have lower leachate and landfill gas contaminant release rates relative to sites where
this activity was not practiced. Evidence at operating sites has demonstrated the
impacts of sustainable landfill operations on landfill gas production. Evidence is less
readily available with regard to leachate as the amount of liquids added during opera-
tion may be much greater, and thus even when concentrations are lower due to accel-
erated waste stabilization, a greater mass flow rate may be present until the landfill has
had sufficient time to drain. Additionally, traditional landfills where liquids addition
was not practiced may have misleadingly low leachate mass release rates as much of
the landfilled waste was never exposed to added moisture and thus leachate pollutant
release rates could increase in the future if final cover system integrity is ever compro-
mised. Landfill operators practicing enhanced efforts to stabilize landfilled waste can
410 17 Final Landfill Disposition
Fig. 17.8 Potential approach of the post-closure care performance evaluation process (adapted
from ITRC 2006)
MSW landfill reclamation (also referred to as landfill mining) refers to the pro-
cess of excavating previously-disposed materials from a landfill, in many cases
processing it, and then re-disposal (or reuse) of materials in another location.
17.5 Landfill Reclamation and Reuse 411
Table 17.2 Primary factors motivating consideration and implementation of landfill reclamation
projects
Factor Description
Environmental Older landfills without a bottom liner system or with poorly functioning
protection systems are often continuous sources of environmental pollution as a result of
leachate and landfill gas releases. Removal of this waste provides an
alternative to expensive ongoing remediation that may have limited
effectiveness
Create new At many landfill sites, older disposal areas were not efficiently used (small
disposal slopes, large amounts of cover soil). Reclamation of these areas may allow
capacity for the construction of more efficient new landfill units and thus allow a given
site to expand its operational life
Reduce closure When landfill reclamation reduces the overall footprint associated with
costs permanent waste disposal (area), both closure costs and PCC costs are reduced
Material The reclamation process allows recovery of potentially valuable materials
recovery such as steel and aluminum. Soil and degraded waste can be reclaimed as
cover soil in existing disposal operations and potentially off-site use
Landfill mining has been practiced to a limited extent around the world. In some
cases, landfilled materials are simply excavated from an unlined disposal area and
deposited in a lined landfill unit without any processing or recovery of materials;
this practice is commonly referred to as waste relocation. In other cases, the exca-
vated material is processed to reduce the magnitude of materials that must be dis-
posed again. The primary factors that have motivated landfill operators to consider
and implement landfill reclamation are presented in Table 17.2 and these motivating
factors are discussed in more detail below. Landfill reclamation merits attention
since the process has potential to reduce the environmental impact of existing land-
fill sites and can be integrated into purposeful material recovery operation at new
facilities (described in Chap. 19).
Waste deposited in landfills operated prior to regulatory liner requirements has
been documented as the source of groundwater contamination at many landfill sites
(Reinhard et al. 1984). This contamination results from both leachate discharge into
the groundwater and landfill gas migration. When addressing groundwater contami-
nation problems, the preferred option (when feasible) is to remove the source of
contamination; thus reclamation may be considered. However, the cost of landfill
mining must be weighed against the cost and effectiveness of other techniques used
to address groundwater contamination; an advantage of waste removal is that long-
term liabilities are significantly reduced.
Siting new landfills has become more difficult and more costly in recent years due
to increased land value in many areas, public opposition, and stricter environmental
regulations. Consequently, more facilities are examining how to utilize effectively
and efficiently all available airspace at existing facilities. Landfill reclamation pro-
vides opportunities for recovery of existing landfill airspace and allows for the cre-
ation of new disposal areas that use airspace more efficiently. Some facility owners
have undertaken reclamation at unlined cells for future use in constructing a new
lined landfill unit, while others have utilized landfill mining as a means of reducing
412 17 Final Landfill Disposition
the size of the landfill unit prior to closure to reduce costs. The recovery and sale of
recyclable materials reclaimed from a landfill (particularly metals) may present an
added source of revenue for landfill reclamation projects. Recovered soil (along with
degraded organic matter) can be used as a substitute for excavated soils in ongoing
landfill operation, and possibly used off-site. Landfill reclamation has also been con-
sidered to recover refuse-derived fuel from landfill sites for combustion and energy
recovery. The financial costs and benefits of landfill mining are addressed in more
detail in Chap. 18.
bulky items and hazardous wastes from the mined material are important steps.
Depending on the waste processing methods and equipment used, larger-sized
pieces (e.g., appliances) may also need to be sorted out before processing the mined
material using a mechanical screen. A front-end loader working with the excavator
can be used for this purpose.
The primary purpose of screening the mined material is to separate the soil or
fine fraction from the larger components. The fine fraction, while being composed
primarily of soil used as a daily cover and intermediate cover, will also include
degraded organic materials (e.g., biostabilized paper, food waste) and small pieces
of other waste components (e.g., glass). The two types of mechanical screening
equipment most often used for screening fines from larger materials in excavated
waste are trommel screens and shaker or vibratory screens. Figure 17.12 shows a
landfill mining project where a shaker screen was employed. Screening the soil
fraction may be difficult in landfills where waste is frozen.
The screen opening size used depends on the quality and final use of the recov-
ered soil. If the recovered material is to be used as a daily cover at a landfill, a larger
sized screen can be used. However, if a better-quality soil (for off-site application)
is desired, a smaller screen size should be used. Screened materials (soil and waste)
must be transported to the place of disposal and/or the location of the approved final
use. Depending on the location (on-site or off-site) of the final use or disposal and
the condition of the roads, dump trucks or off-road trucks can be used for hauling
the processed material. In some cases conveyor are used to transport screened soil
from the mining area to the stockpile (Fig. 17.13).
17.5 Landfill Reclamation and Reuse 415
Fig. 17.12 Processing mined landfill material utilizing a shaker screen and overhead drum
magnet
Fig. 17.13 Landfill mining processing equipment including conveyor system, trommel screen,
and an excavator
416 17 Final Landfill Disposition
Table 17.3 Summary of documented experience and lessons learned from several landfill
reclamation projects
Case study site Description
Naples Landfill The site contained a 33-acre unlined cell that contained 15-year-old highly
(Collier stabilized waste with minimal landfill gas issues. Recovering recycled
County, FL) material proved too expensive to process and was unsuccessful. Landfill soil
(Murphy and recovery was successful accounting for 40–60 % mass of total excavated
Stessel 1991) material. The project resulted in a potential gain of $1.00 per ton of reclaimed
soil for Collier County and the US EPA (1997) reported 10 acres of land
reclaimed
Town of A 1-acre demonstration project, part of a 5-acre municipally owned unlined
Edinburg (NY) landfill. This landfill received waste from 1969 to November 1991.
(NYSERDA Excavation equipment included track excavator (2.4 yd3 bucket), 2–3 wheel
1992) loaders (2.5 and 4.0 yd3 buckets), and 1–2 20-ton dump trucks. Vibratory
screens and a trommel screen were used to sort the excavated material. The
project resulted in approximately 14,930 yd3 of excavated waste in the first
two phases of the project; and additional 1.6 acres yielded 31,000 yd3 in the
third phase. The contractor cost was $3.00/yd3 for this phase
Frey Farm The excavated material was screened using a 1-in. trommel screen. 41 % of
(Lancaster recovered material was soil, 56 % was used as fuel at a municipal waste
County, PA) combustor, and 3 % was incombustible and reburied at the site. The
reclaimed material had an estimated energy value of 3,080 BTU/lb. By 1996,
the project resulted in 300,000–400,000 yd3 of excavated and processed
waste at a rate of 2,650 tons/week. Extensive air monitoring was conducted
during the project. The project resulted in a net revenue of $13.3/ton for
Lancaster County
Wyandot The project site was a sanitary landfill of 188 acres consisting of lined and
County unlined cells. Only waste relocation (i.e., no processing) was done via an
(Carey, OH) excavator to excavate waste, an off-road truck hauled the material to an
on-site lined unit. The overall rate of waste relocation was 300,000 yd3 per
year and as of 2006, 30 acres of land had been reclaimed. The total amount of
waste excavated is approximately 1.4 million yd3 and the cost was estimated
at $4 per yd3. Since the project began an improvement in groundwater quality
was observed
Shawano The site consisted of a combination of lined and unlined cells. A waste
County (WI) relocation project was initiated to decrease the cost of treating the leachate
collected from a perimeter toe drain. The mining process consisted of
excavation of waste from the unlined cell (using two excavators) and hauling
it to the on-site lined cell. No screening was done. Bulk soil was separated
into “clean,” “mildly contaminated” and “contaminated soil.” 12 acres were
reclaimed by relocating 0.3–0.4 million yd3 of waste from unlined to lined
cells, and approximately 2 ft of underlying soil was scraped and stockpiled on
the clay lined area. The project cost was approximately $3/yd3
(continued)
17.5 Landfill Reclamation and Reuse 417
Table 17.3 (continued)
Case study site Description
Central The landfill had a lined cell and a 10-acre unlined cell using well-
disposal decomposed mined waste as daily cover. The excavation consisted of using
systems (Lake one backhoe with a 5-yd3 bucket and hauling to an on-site lined unit via four
Mills, Iowa) trucks. Explosivity of landfill gas became a concern on some occasions.
1,000–1,500 yd3 of waste relocation occurred per day resulting in 10 acres
reclaimed and an overall relocation of 250,000 yd3 of waste as of 2006
Pike Sanitation The site contained 40 acres of unlined cells and a 125-acre lined cell
(Waverly, permitted in 1996. One to two backhoes were used for excavation in
Ohio) conjunction with four to six off-road trucks. No materials were processed.
The asbestos containing materials, if encountered, were sprayed with water to
minimize movement to air. All waste moved to lined cells amounted to
700,000–800,000 yd3 of waste at a rate of 40,000 yd3 per month
La Crosse The site consisted of an unlined cell approximately 25 acres with 1.2 million
County (WI) yd3 of waste and a lined cell. Excavation used 2 backhoes with 4 yd3 buckets
and material was hauled to a lined cell via 12 off-road trucks with 12 yd3
buckets each. Soil from the cap was recovered and used for future landfill
operation while larger waste (i.e., furniture) was placed in a lined cell; WTE
ash was placed in an ash monofill. 25 acres were reclaimed by relocating
approximately 500,000 yd3 of waste in the first phase
Dean Forest Site layout consisted of 4 quadrants (three lined and one partially lined). The
(Savannah, site contained some MSW, construction and demolition (C&D) debris, and
GA) sludge. The excavation operation consisted of excavation using two
excavators and waste hauling to lined cells via 6–7 off-road trucks. Waste
was not processed and was relocated at a rate of 7,000 yd3 per day, resulting
in 130 acres reclaimed and 650,000 yd3 of waste relocated. Waste was not
processed because of space and time constraints
Clovis (CA) The site consists of unlined and clay-lined units made of a synthetic
composite liner system used since 1998. Excavation consisted of a dozer
scraping and pushing waste to an excavator. The waste was screened using a
trommel (with a 2-in. screen) and loaded to 40 yd3 open-top dump trucks and
hauled to a lined cell. Soil, which amounted to 60 % of the total material
excavated, was transported, collected, and consolidated into a soil stockpile.
Waste was mined at 1,100 yd3 per day, 190 days a year (75 % of the total
working days), totaling to 2.1 million yd3, and costing $4.84/yd3
Winnebago The site was composed of lined and unlined cells. Waste was relocated from
County (WI) unlined cells to a closed lined cell to fill depression on this cell. The relocated
waste was spread with a dozer and an electromagnet was used to collect
ferrous metals. No other processing technique was used. Approximately
3–4 acres of land were reclaimed during the project
Phoenix Rio The site comprised of more than 600 acres spanning Phoenix North Central
Salado landfill, Del Rio Landfill and various others. Only the waste that was within
(Phoenix and the project construction zone was mined. Waste was screened at two sites, of
Tempe, AZ) which 150,000 tons of waste was removed and segregated while 100,000 tons
were screened with a trommel and grizzly screen for re-use as clean soil. As
of 2005, more than 380,000 yd3 of C&D debris, 20,250 yd3 of MSW, and 600
tons of tires were mined. Approximately 80 % of the mined materials were
re-use/recycled
418 17 Final Landfill Disposition
17.5.3 D
esign, Permitting, and Operation of Reclamation
Projects
Odor issues may be less pronounced when the waste material is well decomposed.
A landfill can also aggressively operate the gas extraction system, if present, to
minimize gas emissions and odor problems during mining. Chemical perfumes or
masking agents may provide temporary relief from odor.
Depending on the age of the waste and degree of decomposition, gas monitoring
(e.g., explosivity, toxic gases) during operations may be prudent. Methane can form
explosive mixtures when mixed in certain proportions with air (Chap. 14). Elements
specific to landfill gas monitoring should be included in mining operation plans and
should discuss monitoring devices and frequencies, establish action levels, and
specify remedial procedures if action levels are met or exceeded.
Waste excavation and screening can potentially cause the generation of dust and
windblown litter. While many previous landfill reclamation projects have reported
minimal dust issues because the excavated waste was moist, in the event that dust is
a concern, a tanker truck to spray water around the excavation and processing area
may be required. Litter control devices such as portable fences or other suitable
devices may also be used.
Regulations for operating landfills require application of daily cover at the end of
each day to minimize adverse impacts such blowing litter, odors, disease vectors, or
fires; this might also be required for exposed waste from a reclamation operation.
Landfill mining project plans or health and safety plans should address the issue of
cover and should establish a protocol that is consistent with local regulations, which
may include identifying the source of cover material, and the amount and frequency
of application.
Mining of landfilled waste will result in a change of existing grades at the site;
reclamation projects should implement a stormwater management plan to minimize
the contact of stormwater with stockpiled or exposed waste. Stormwater that con-
tacts solid waste is considered to be leachate and must be managed as such. Leachate
may also be generated from excavated waste that is wet. As with routine landfill
operation, stormwater can be controlled using diversion berms, by grading the sur-
face adjacent to the waste to direct stormwater from the working face, or excavating
waste in a given direction to minimize leachate generation.
When material is reclaimed from a landfill and processed, the two major resultant
components are a fine fraction and a larger fraction consisting of waste. The fines
result from a combination of the soil originally used as cover in landfill operation,
degraded waste, and small pieces of disposed waste. Table 17.5 summarizes the
reported composition of reclaimed material from several landfill mining studies.
The fines fraction has been reported to constitute approximately 50–85 % of the
recovered material (weight basis).
Potential reuse options for recovered soil include daily and intermediate landfill
cover (uses inside the landfill) and construction fill (uses outside the landfill)
420 17 Final Landfill Disposition
Table 17.5 Reported bulk composition of material extracted during landfill mining projects
Study Fines Identifiable bulk waste materials
Murphy and Stessel 50 % 10 % paper, 7 % plastic, 5 % wood, 2 % aluminum,
(1991) (0.5-in. screen) 5 % metal/stone, 5 % glass/ceramic, 18 % misc
NYSERDA (1992) 84.50 % 3 % paper, 2.80 % plastic, 0.70 % wood, (2.5/1.4)
(0.5-in. screen) metal/stone, 1.30 % glass/ceramic, 3.80 % misc
US EPA (1993) 59.1 % 3 % paper, 4.3 % plastic, 2 % yard waste, 5.2 %
(29 samples) (1-in. screen) wood, 0.9 % textile, 0.6 % rubber/leather, 2.4 %
metal, 2.1 % glass/ceramic, 20.5 % misc
Kilmer and Tustin 75 % Not reported
(1999) (1-in. screen)
Earle et al. (1999) 75–87 % Not reported
(1/4-in. screen)
Zornberg et al. (1999) >56 % Not reported
(80 samples)
Jain et al. (2005) 58 % 12 % paper, 13 % plastic, 3 % yard waste, 3 %
(78 samples) (1/4-in. screen) textile, 6 % metal/stone, 5 % glass/ceramic
McKnight (2005) 49 % 18 % paper, 7 % plastic, 12 % yard waste, 5 %
(19 samples) (1/4-in. screen) textile, 7 % metal/stone, 2 % glass/ceramic
Quaghebeur et al. 40.1–67.8 % 1.9–11 % plastic, 0.5–11.6 % wood, 0.6–2.3 %
(2013) (23 samples) (0.8-in. (20-mm) textile, 0.5–14.5 % rubber/leather, 0.1–0.2 %
screen) metal, 18.5–28.3 % stone, 0.4–0.8 % glass/ceramic
Kurian et al. (2003) 40.1–67.8 % 0.5–13.9 % paper, 8.2–9.5 % plastic, 1.1–1.3 %
(58 samples) (0.8-in. (20-mm) yard waste, 2.9–5.4 % textile, 4.2–5.7 % metal/
screen) stone, 0.2–0.5 % glass/ceramic; 4.8–11.5 % misc
Hull et al. (2005) 51–55 % Not reported
(1-in. screen)
(US EPA 1997). Other end uses will be dictated by available markets, the quality of
the material, and the regulatory framework for reuse. The issue that would most
likely limit the reuse of mined landfill fines outside of the landfill environment
would be the presence of trace chemicals. Given that a large variety of household,
commercial, and industrial waste containing chemicals are disposed in MSW land-
fills, the potential impact of these chemicals on the environment if the mined resi-
dues were reused must be considered. When evaluating likely chemicals of concern,
it should be noted that most organic chemicals should eventually be degraded in the
biogeochemical environment of a landfill (Field et al. 1995; Reinhart and Townsend
1997). Non-degradable chemicals such as heavy metals, however, will remain in the
waste unless leached out. Several investigations indicate that heavy metals would be
retained in the landfill (Belevi and Baccini 1989; Finnveden 1996; Bozkurt et al.
1999). The concentrations of these chemicals in the mined material would likely
dictate the degree to which mined residue can be reused outside of the landfill
environment.
While most regulatory jurisdictions will not have regulated limits specific to
materials reclaimed from landfills, they will often have risk-based thresholds for
contaminated soil or water that may be applicable. Typically, the concentration of
17.6 Final Site Use and Configuration 421
chemicals in the fines would have to be characterized to assess (a) the risk to human
health from direct exposure of the material if it is reused outside of a landfill envi-
ronment and (b) the risk to groundwater or surface water. The process used in many
states is to compare a concentration that is statistically representative of the material
proposed to be reused (e.g., the 95 % upper confidence limit (UCL) of the mean
constituent concentration) to a health-based risk level. For use in commercial or
industrial settings, some assurance would need to be provided that the property
where the material was being reused remained commercial/industrial (known as
institutional controls). To evaluate the potential risk to groundwater or surface
water, the reused material may be tested for leachability and compared to the appro-
priate water quality risk thresholds.
Once a landfill site has been successfully closed, the owner then decides whether to
isolate the site from the general public or open the site for some useful purpose,
usually one focused on community activities (common for municipally-owned
facilities). This decision is often made at the planning stages well in advance of the
closure date. Closed landfill sites have been successfully used for parks and recre-
ation, botanical gardens, ski slopes, toboggan runs, coasting hills, ball fields, amphi-
theaters, playgrounds, and parking areas. The use of a closed sanitary landfill as a
green area (a community park) or open space is very common and presents rela-
tively fewer challenges compared to a use that incorporate buildings and similar
structures. The most commonly used vegetation is grass, though shrubs and small
trees may be added where funds are available and if this type of vegetation is com-
patible with the end use and final cover design. Another use of closed landfills
includes redevelopment into a golf course (see Fig. 17.14). As discussed in Chap. 19,
landfills are growing in popularity as sites for placement of solar panels and wind
turbines for energy production.
Closed landfills are typically not well-suited for construction of buildings,
because of mechanical and geotechnical concerns, as well as potential issues associ-
ated with landfill gas accumulation and formation of explosive conditions. Small,
light buildings such as concession stands, sanitary facilities, and equipment storage
sheds are often required at recreational use areas. A geotechnical engineer should be
consulted if plans call for structures to be built on or near a completed sanitary land-
fill. The cost of designing, constructing, and maintaining buildings is often consider-
ably higher than it is for those erected on a well-compacted earth fill or on undisturbed
soil. Roads, parking lots, sidewalks, and other paved areas should be constructed of
a flexible and easily repairable material such as gravel or concrete pavers.
Buildings or other structures may be designed and built to accommodate for
potential settlement and to minimize gas problems that might result in explosive or
toxic conditions in any enclosed spaces. The GCCS and LCRS will normally still
be operational, and associated infrastructure should be appropriately isolated,
422 17 Final Landfill Disposition
Fig. 17.14 Golf course constructed on a closed MSW landfill (Photo courtesy of CDM-Smith)
p rotected, and labeled with precautionary signage. All construction activities should
incorporate appropriate protection and repair of the final cover system, particularly
any geomembranes or compacted soil barrier layers. Other issues that should be
addressed at closed landfill sites include ponding, cracking, and erosion of cover
material. Periodic maintenance includes regrading, reseeding, and replenishing the
cover material; maintenance work is required to keep the fill surface from being
eroded by wind and water.
References
Barlaz M, Rooker A, Kjeldsen M, Gabr M, Borden R (2002) Critical evaluation of factors required
to terminate the post-closure monitoring period at solid waste landfills. Environ Sci Technol
36(16):3457–3463
Belevi H, Baccini P (1989) Long-term behaviour of municipal solid waste landfills. Waste Manag
Res 7:43–56
Bozkurt S, Moreno L, Neretnieks I (1999) Long term fate of organics in waste deposits and its
effect on metal release. Sci Total Environ 228:135–152
Council E (1999) Directive 1999/31/EC on the landfill of waste. Off J Eur Union L182:1–19
Earle CDA, Rhue RD, Earle JFK (1999) Mercury in a municipal solid waste landfill. Waste Manag
Res 17:305–312
Finnveden G (1996) Valuation methods within the framework of life-cycle assessment. IVL Report
No B1231. IVL, Stockholm
Hull R, Krogmann U, Strom P (2005) Composition and characteristics of excavated materials from
a New Jersey landfill. J Environ Eng, ASCE 131(3):478–490
References 423
ITRC (2006) Evaluating, optimizing, or ending post-closure care at municipal solid waste landfills
based on site-specific data evaluations. Prepared by the Interstate Technical & Regulatory
Council, Alternative Landfill Technologies Team, Sept 2006, Washington, DC
Jain P, Kim H, Townsend T (2005) Characterization of heavy metals in soil reclaimed from a
municipal solid waste landfill. Waste Manag 25:25–35
Jain P, Townsend TG, Johnson P (2013) Case study of landfill reclamation at a Florida landfill site.
Waste Manag 33(1):109–116
Jain P, Powell J, Smith J, Townsend T, Tolaymat T (2014) Life-cycle inventory and impact evalu-
ation of mining municipal solid waste landfills. Environ Sci Technol 48(5):2920–2927.
doi:10.1021/es404382s
Kilmer KS, Tustin J (1999) Rapid landfill stabilization and improvements in leachate quality by
leachate recirculation. Proceedings of SWANA’s fourth annual landfill symposium, Denver,
28–30 June 1999
Kurian J, Esakku S, Palanivelu K, Selvam A (2003) Studies on landfill mining at solid waste dump-
sites in India. Proceedings Sardinia ’03, Ninth International Landfill Symposium, Cagliari,
Italy, 248–255
McKnight T (2005) Engineering properties and cone penetration testing of municipal solid waste
to predict landfill settlement. Master’s Thesis, University of Florida, Gainesville
Morris J, Barlaz M (2011) A performance-based system for the long-term management of munici-
pal waste landfills. Waste Manag 31(2011):649–662
Murphy RJ, Stessel RI (1991) Optimization of landfill mining: interim final report-phase I. Center
for Solid and Hazardous Waste Management, Gainesville
NYSERDA (1992) Town of Edinburg landfill reclamation demonstration project. Final report pre-
pared for the New York State Research and Development Authority
Quaghebeur M, Laenen B, Geysen D, Nielsen P, Ponikes Y, Van Gerven T, Spooren J (2013)
Characterization of landfilled materials: screening of the enhanced landfill mining potential.
J Cleaner Prod 55:72–83
Reinhart DR, Townsend TG (1997) Landfill bioreactor design and operation. Lewis, Boca Raton
Reinhard M, Goodman NL, Barker JF (1984) Occurrence and distribution of organic chemicals in
two landfill plumes. Environ Sci Technol 18:953–961
US EPA (1989) Risk Assessment guidance for superfund: volume I human health evaluation manual.
Office of Solid Waste and Emergency Response 9355EPA/540/1-89/002, Dec, Washington, DC
US EPA (1993) United State environmental protection agency. Evaluation of the Collier County,
Florida Landfill Mining Demonstration, EPA/600/R-93/163, Risk Reduction Engineering
Laboratory, Office of Research and Development, U.S. Environmental Protection Agency,
Cincinnati
US EPA (1996) Soil screening guidance: user’s guide. Office of Solid Waste and Emergency
Response 9355.4-23, Washington, DC, July
US EPA (1997) Landfill reclamation. Office of Solid Waste and Emergency Response (5305W)
EPA530-F-97-001, Washington, DC, July
US EPA (1998) Solid waste disposal facility criteria—technical manual. EPA/5-3/R-93/017
US Government (2012) US code of federal regulations: Title 40 Part 258. Criteria for Municipal
Solid Waste Landfills
Zornberg JG, Jernigan BL, Sanglerat T, Cooley B (1999) Retention of free liquids in landfills
undergoing vertical expansion. J Geotech Geoenviron Eng, ASCE 125(7):583–594
Chapter 18
Economics
Abstract Cost considerations are one of the most important features in landfill
planning and management. This chapter presents a series of examples of how tradi-
tional landfill costs and benefits can be impacted by sustainable landfilling opera-
tions. In particular, a discussion including the types of costs (e.g., leachate
management, gas recovery, and reuse) and likely or potential magnitude of costs
and benefits of sustainable landfilling is presented. The reader is given multiple
tools to guide the site-specific decision-making process associated with implement-
ing sustainable landfilling of which include liquids management, gas management,
and airspace recovery. A conceptual discussion of social costs is provided, in addi-
tion to economic considerations after the landfill closes, including how post-closure
care plans (and timing) and landfill reclamation can affect life-cycle costs.
18.1 Overview
An important part of planning for solid waste disposal in a landfill is the estimation of
construction, operating, and maintenance costs and anticipated revenues or benefits. As
with any decision that would impact construction and operation at a landfill, the short-
term and long-term costs and benefits of sustainable landfilling should be considered
carefully and ideally compared to some base scenario so that owners and operators can
make informed decisions to proceed with sustainable landfilling. These economic fore-
casts could be conducted prior to a landfill facility being built (i.e., a brand-new facil-
ity) or at an operating landfill that is planning future disposal areas or cells.
Short-term costs include landfill design and permitting, land acquisition (for new
sites or significant expansions), site preparation, cell construction, and financial
assurance. Long-term costs and benefits include operation and maintenance (O&M)
over the life of the landfill, closure, gas collection and beneficial use, and post-
closure care (PCC); the long-term costs also include social or external costs and
benefits such as loss or gain of local amenities, pollutant emissions (including
greenhouse gases), nuisances, and fossil fuel offsets, which require an intergenera-
tional comparison of cost for future landfill effects.
Substantial time and resources are normally required to develop a landfill project
prior to any construction of the landfill unit itself. In addition to land acquisition, the
site must be designed and permitted with the appropriate regulatory agency or agen-
cies (see Chap. 3). If a permit is granted, the site must be appropriately developed,
including construction of access roads and installation of required utility connections.
Approximate up-front (pre-construction) costs for a landfill may range from USD
0.75 million to more than USD 1 million (Duffy 2005a; KDEP 2012). Support struc-
tures (e.g., roads, buildings) will be needed along with appropriate materials resources
(e.g., borrow pit for cover soil). Table 18.1 provides a list of cost elements associated
with the development and a construction of a landfill project outside of the disposal
unit itself. Factors influencing the magnitude of the costs elements and related factors
pertaining to sustainable landfill practice implementation are also presented.
18.2 Fundamentals of Landfill Economics 427
Table 18.1 Typical costs elements for associated with landfill site development beyond those
involved with the landfill unit
Sustainable landfilling
Cost element Factors differential considerations
Site evaluation, Planned site size, past and current land use, None expected
planning, design hydrogeological conditions, surrounding
land use, geotechnical/soil conditions
Permitting State and local regulations, inclusive of Lack of state regulations
environmental/solid waste regulations and allowing liquid addition,
related regulations (e.g., management of greater potential for permit
surface water, wetlands, etc.), land-use appeal, potential increased
regulations (e.g., zoning, conditional use permitting costs to address
permitting, certificates of need, and regulatory questions and
impacts to infrastructure) comments
Borrow source Availability and quantity of borrow Low permeability soils
material, type and characteristics of borrow must be addressed in
material operations, could result in
greater incurred cost
Land acquisition Area required (buffer zone, landfill Potentially reduced area
capacity, site geometry, support facilities), for waste disposal,
land costs increased buffer needs
Site fencing and Cost of fencing, perimeter distance Potentially reduced area
access control requirements
Site buildings/ Cost per area, types of buildings (offices, None expected
structures gatehouse, gas management, maintenance/
storage, public drop-off centers for
recyclables and hazardous wastes)
Weigh scales Scale cost, number of scales None expected
Site utilities Connections to electric grid, sanitary None expected
sewer, natural gas, potable water
Access Roads Road construction and upgrade unit costs, Potentially reduced area
length of roads requirements
Landscaping Unit cost of landscaping, area to be Reduced area requirements
landscaped
Financial assurance Cost of maintenance of the financial Potentially reduced
assurance bond or other instrument, length duration of long-term care
of operating life, local regulations, length
of post-closure care period
Table 18.2 provides costs elements associated with the construction of the lined
landfill unit itself. Cost elements associated with the construction include earthwork
for sub-grade preparation, compacted clay liner or geosynthetic clay liner construc-
tion, geomembrane liner installation, and leachate collection system construction.
The total cost for these components may range from USD 150,000 to USD 450,000
per constructed acre, and does not include earthwork needed for ground improve-
ment or grade preparation for the constructed components (which may vary signifi-
cantly from site to site). As landfill units are constructed on a frequent basis,
(particularly at large sites where lined cells are constructed to provide several years
428 18 Economics
Table 18.2 Cost elements associated with the construction of the lined landfill unit
Sustainable landfill
Cost element Factors differential considerations
Site clearing and Unit cost of clearing and excavation, Area requirement
excavation soil hauling, area of construction,
depth below grade
Site berms Unit cost of construction and soil, Area requirement
berm design
Liner systems Unit cost of liners, soil, drainage Additional liner requirements
material, compaction, and liner
installation; berm height; volume of
soil; total area of liner; thickness of
liner elements
Leachate pumping Tank, pump, piping cost Greater capacity for storage
and storage and pumping for recirculation
Leachate collection Unit cost of purchase and installation Additional permeability in
and recirculation of drainage material, piping; length of drainage layers; additional
system pipe; volume of leachate collection piping for recirculation,
and recirculation pipe trenches; length injection facilities; toe drains
of horizontal trenches and/or vertical or similar infrastructure for
wells for recirculation managing seeps.
Gas extraction Cost of piping procurement and Increased gas generation
installation, length of pipes, number rates, earlier installation of
and cost of wellheads gas extraction system
Monitoring wells Number of wells and well installation, Reduced area requirements,
(groundwater and gas) perimeter distance, well depth increased monitoring
requirements
of capacity while not being so large as to be cost prohibitive), lined unit construc-
tion costs will be incurred throughout much of the life of the facility.
While the construction of the GCCS may lag several years behind the construc-
tion of the liner system and LCRS, it is another major capital expense associated
with the landfill unit. GCCS construction may be required earlier at sites practicing
sustainable landfill technologies. The capital costs associated with a GCCS include
installation of extraction wells/trenches, piping network, a blower/flare system with
associated controls, and a condensate management system. GCCS construction cost
(exclusive of blower and LFG destruction devices) may range from USD 24,000 to
USD 35,000 per acre (Duffy 2005b; US EPA 2015). Wells for monitoring for
groundwater and landfill gas will require installation on the perimeter of any new
lined landfill unit.
Operation and maintenance (O&M) costs of landfills consist of waste handling,
cover use, litter control, training, utilities, permitting, financial assurance, sampling
and compliance monitoring, leachate treatment, and transportation. Table 18.3
presents a summary of cost elements associated with landfill operation. O&M
accounts for a substantial portion of a landfill’s overall cost, normally comprising
18.2 Fundamentals of Landfill Economics 429
Table 18.3 Cost elements associated with landfill operation and maintenance
Sustainable landfill differential
Cost element Factors considerations
Daily Costs of equipment procurement and Reduced leachate treatment
operations maintenance, personnel, utilities, leachate volume, additional monitoring
treatment—labor costs, waste receipt rate, requirements, additional
leachate generation rate, utility costs short-term gas generation
Daily cover Area of daily waste placement, unit cost Increased permeability or
(or revenue) of soil or alternative cover material removal of daily cover
Monitoring Area of landfill, specific permit Leachate recirculation
costs conditions, number of monitoring points infrastructure monitoring,
(e.g., groundwater monitoring wells, leachate additional LFG collection
sumps, gas collectors), prevailing labor rates, infrastructure (if present),
use of third party contractors potential settlement monitoring
more than 50 % of a landfill’s overall cost. Reported O&M cost for a GCCS is
approximately USD 4,100 per acre per year (US EPA 2015), although this figure
may vary substantially depending on the number of gas collection wells, system
configuration, and monitoring frequency required. O&M of automated GCCS that
have set control points adjusted by a computerized control system may be expected
to have reduced O&M costs compared to a site that has GCCS components adjusted
manually by an operator, but these systems would carry a greater construction cost.
Monitoring of landfill gas, groundwater, and other site features associated with
the landfill (e.g., stormwater, waste elevation) will be normally required throughout
the operational life of the landfill (and after closure). As described in Chap. 16, an
expense that might be incurred at sites implementing sustainable practices relates to
additional monitoring. This could take the form of more labor in terms of collection,
added laboratory analytical expense, additional fill materials to address areas of dif-
ferential settlement, and additional monitoring equipment.
At the end of the landfill’s operating life, the landfill must be closed according to
the site’s permit and applicable regulatory requirements. At some facilities, closure
is implemented only when waste acceptance activities at the site have reached com-
pletion, while other facilities practice closure of smaller areas at greater frequency
(see Chap. 17 for a discussion on these strategies). Closure involves the construction
of final cover system (a cap). Approximate closure cap construction cost (excluding
GCCS) ranges from USD 150,000 to more than USD 300,000 per acre (Duffy
2005b; KDEP 2012; MDE no date).
After the site has been closed, the landfill must continue to be cared for to ensure
environmental protection and compliance with the site’s permit. This will include
such activities as removing and managing leachate, collecting gas, maintaining the
cover system, and continued monitoring. Annual post-closure care costs (which
includes site security, cap maintenance, environmental monitoring) range from
USD 2,000 to USD 3,000 per acre. Additional costs (and possibly revenue) might
be incurred depending on the final end use of the facility (Table 18.4).
430 18 Economics
Table 18.4 Cost elements associated with closure, post-closure, and final site use
Cost Sustainable landfill differential
element Factors considerations
Final Unit cost of procuring, delivering, and Closure timing, potential for recovery
cover installing materials (vegetation support, of air space
geotextiles, low and high permeability soil,
geomembrane, landscaping), thickness of
layers, area of closure
Post- Annual cost of final cover maintenance and Reduced length of post-closure care
closure replacement, well monitoring, operation of period, reduced long-term gas and
care leachate and gas collection systems; number leachate generation
of years of post-closure care
Final May include electrical service, buildings, Similar considerations as post-closure
site use surface preparation, significant fill material care. Potential could exist for more
(e.g., golf course), miscellaneous flexibility in final site uses if
infrastructure—the inclusion and extent of site-specific data show reduced LFG
these factors depends strongly on the final production, slowed settlement, and
site use that is planned improved leachate quality relative to
conditions from a traditional landfill
Municipalities and private companies that own and operate landfills derive their
primary revenue source from the fees charged to dispose of wastes in the landfill.
These tipping fees are most often based on the weight of waste disposed, as mea-
sured by scales placed near the entrance of the facility, although facility owners may
also charge fees on a per-truckload or volume basis. Van Haaren et al. (2010), based
on a nationwide survey, reported that the statewide average tipping fees in the US
ranged from USD 15 to USD 96 per ton. The tipping fees are used to pay for the
construction and operation of the landfill unit and the costs associated with closure
and post-closure care. They may also include revenue for other government func-
tions in the case of municipally-owned facilities and will include profit in the case
of private operations.
Another potential source of revenue at landfills sites is the sale of electricity or
processed landfill gas. Technical information regarding the use of landfill gas as an
energy source is provided in Chap. 19. Since the beneficial use of landfill gas is one
logical outcome of sustainable landfill practices, additional economic information
on gas-to-energy is provided in subsequent sections. The electricity or processed
landfill gas sale prices are highly contingent upon the electricity and natural gas
prices and incentives or other governmental incentives for renewable power. Other
opportunities for energy recovery at landfill sites include solar power and wind
power. As described in Chap. 19, if planned for appropriately, landfills may serve as
the hub of material recovery operations, thus providing another potential source of
revenue for the site owner.
18.2 Fundamentals of Landfill Economics 431
Given the known costs associated with properly closing a landfill and maintaining it
during the post closure care period, regulatory agencies require that landfill owners
and operators demonstrate that funds will be available to close the landfill upon
completion regardless of the revenues collected by the facility. Cost elements that
must be accounted for in the financial assurance demonstration include closure of
the landfill unit, care of the landfill unit during the designated post-closure care
period (including environmental monitoring), and possible corrective action to
address environmental releases. Landfill owners must demonstrate how the funds
necessary for financial assurance will be provided during the active life of the land-
fill. Estimates of closure and PCC costs require the following: (a) knowledge of the
point at which operations make closure most expensive; (b) an assumption that a
third party will perform closure activities; (c) awareness that estimates are to be
revised as conditions change; (d) an understanding that PCC includes periodic and
annual costs; and (e) the ability to revise the estimated costs for inflation.
The owner/operator may be required by regulations to have a detailed written
estimate of the cost of hiring a third party to close the largest area of the landfill that
will ever require final cover during the active life and place that cost estimate into the
operating budget. If changes to the closure plan increase or decrease the maximum
cost of closure, the cost estimate must be changed accordingly. Similarly, regulations
may require the owner/operator to follow the same criteria for PCC and dictate the
same criteria for corrective action. Owner and operators may be required to demon-
strate that funds are available to meet the cost of closure, PCC, and corrective action.
Several mechanisms to meet these monetary obligations are provided in Table 18.5.
Table 18.5 Examples of instruments that can be used to demonstrate the satisfaction of financial
assurance requirements for landfills
Financial
instrument Description
Trust fund Asset set aside to pay for closure, PCC, and contingencies, typically held by a
third party (e.g., bank). Funds typically established through fees collected during
operations
Surety bond Site owner pays a premium to a surety company that guarantees to pay the
“penalty sum” of the bond to the designated agency should the owner/operator
fail to perform the agreed-upon closure and PCC. These are typically used in two
forms: financial guarantee bond and performance bond
Letter(s) Commitments from third parties, typically commercial banks, to provide monies
of credit if and when needed in accordance with credit agreement signed with the bank
Insurance Contractual agreement whereby the insurer agrees to compensate the
policyholder for losses
Financial Also known as “self insurance”, consists of a series of tests (e.g., a government
tests may have financial, recordkeeping, and public notice requirements, while a
corporation may have size, assets, and financial soundness as part of its test)
Guarantees A guarantee can demonstrate that the required costs (all or a portion) can be paid
for and that the guarantor can fulfill the financial obligations if the owner/
operator fairs to perform
432 18 Economics
In general terms, economy of scale involves reducing a unit cost by realizing opera-
tional efficiencies. In a landfill context, many O&M costs (e.g., equipment, mainte-
nance, fuel, equipment operators, technicians, administrative staff, and other
administrative costs) are required regardless of the quantity of waste accepted at the
landfill. Thus, increasing the amount of waste accepted will reduce the landfill’s unit
operational cost since the amount of revenue will increase. Accepting larger quanti-
ties of waste can also allow for waste to be placed in larger daily working areas or
cells, which can allow the operator to use relatively less cover soil, which may also
be a cost reduction.
The additional costs and benefits of sustainable landfilling were briefly discussed in
the previous section. Additional capital and O&M costs are borne for liquid injec-
tion and leachate recirculation, air injection (if used), additional monitoring, side
slope seep control, and early construction and operation of GCCS. Potential eco-
nomic benefits of sustainable landfills include an extension of the active life of the
landfill with more efficient airspace utilization, reduced leachate treatment/disposal
costs, deferred new cell and final cover construction, earlier beneficial reuse of land,
post-closure care savings from fewer monitoring and financial assurance require-
ments and reduced maintenance, and larger gas production which represents an
opportunity to generate additional revenue when converted to energy.
Berge et al. (2009) conducted economic modeling of traditional and sustainable
landfills, and information from this work is referenced in the following discussion.
The sustainable landfill scenarios included as-built (initially designed and con-
structed as a sustainable landfill), retro-fit (converted into a sustainable landfill at
closure), and aerobic. One clear result was that without advantages associated with
reduced PCC, retrofit sustainable landfill and traditional landfills carried similar
present worth (PW) costs. Increased O&M costs appeared to offset advantages asso-
ciated with leachate treatment and air space recovery in a retrofit sustainable land-
fill. As-built sustainable landfills have lower costs than traditional and retrofit
sustainable landfills, mainly because of utilization of the recovered air space and
reduction in leachate treatment and management cost. The cost of aerobic landfills
is greater than anaerobic where gas recovery and use is possible; the difference
reduces when no gas recovery is planned or where leachate treatment costs are high.
Figure 18.1 provides a breakdown of the major cost elements (construction,
O&M, leachate treatment, and post-closure care) as a function of total PW (since
some of the costs are incurred in the future, costs were discounted in order to com-
pare their present worth); assumptions for the cost calculations were provided in
Berge et al. (2009). The magnitude of each cost element greatly depends on many
18.3 Costs and Benefits of Sustainable Landfill Practices 433
60
Post-Closure Care
50 Leachate Treatment
Operation & Maintenance
40 Construction
30
20
10
0
Traditional Landfill Sustainable Landfill
local factors including size of the landfill, cost of land, local/state regulatory envi-
ronment, availability of materials and utilities, and other economic conditions.
Construction costs represent the largest fraction of PW for both sustainable landfills
and traditional landfills and may be greater for sustainable landfills than traditional
landfills. Berge et al. (2009) assumed PCC to be of equal duration and is approxi-
mately 15–25 % of PW costs. PCC costs represent a greater fraction of PW for
sustainable landfills because of the continued intensive operation of the landfill after
closure to recirculate leachate and monitor the landfill, although this impact could
vary depending on when leachate recirculation is initiated and terminated. Thus, for
sustainable landfills, the potential to reduce the length of PCC can result in signifi-
cant cost savings. Additional results from this analysis are presented as part of the
following sections, which focus on issues related to liquids management, gas recov-
ery, airspace gain, and external costs.
The cost associated with the installation and maintenance of the liquids addition
system is one of driving factors that influences the selection of a liquids addition
approach (see Chap. 6–9). Some techniques are costlier than others (e.g., some sur-
face techniques can be implemented with little cost and effort whereas installation
of subsurface systems require more extensive resources). The availability of in-
house resources (e.g., equipment, operators) has a significant impact on total project
costs. If an operator has an excavator available on site, the construction of horizon-
tal liquids introduction trenches might be a method that can be accomplished with
existing landfill staff.
434 18 Economics
Table 18.6 The influence of leachate recirculation and treatment volumes on total
PW costs (Berge et al. 2009)
Percent of leachate Percent of traditional Percent of traditional
recirculated landfill leachate treated landfill treatment cost
50 50 60
75 25 35
100 0 0
As described in Chap. 13, the operation of sustainable landfills (without air addi-
tion) can result in greater production rates of LFG, which can have multiple eco-
nomic impacts. A major factor specific to costs and benefits related to LFG includes
accelerated gas production during landfill operations and the difference in gas pro-
duction rates after the landfill closes. Therefore, if the additional gas produced early
on in the landfill’s life can be captured, the difference represents additional energy
production that can occur. Inefficient gas collection during the landfill’s active
phase, however, reduces the benefit of enhanced gas production and could add bur-
dens such as increased gas emissions and potentially odors.
Since the potential to produce gas in a given mass of waste is fixed, accelerating
gas production during active operations means that gas production after closure will
18.3 Costs and Benefits of Sustainable Landfill Practices 435
Table 18.7 Considerations when accounting for airspace gain in sustainable landfilling economic
evaluations
Factor Discussion
Permitting Accounting for future filling activities for recovered airspace must be
and planning included as part of the planning and permitting process in advance.
Failure to do so may eliminate airspace recovery as one of the
economic benefits of sustainable landfilling operations
Recirculation If recirculation is initiated during active filling of a given cell (or cells),
initiation during a potentially large amount of the airspace that could be recovered
active filling would occur during active filling or shortly thereafter, which should be
accounted for in the economic analysis
Recirculation Economic evaluation should examine what degree of the airspace gain
initiation after could be practically recovered, with a focus on landfill dimensions.
completion of Note that some airspace gain (e.g., on lower portions of some side
active filling slopes) may not be practically recovered or represent an airspace gain
that is too small to justify re-filling with new waste
Degree of settlement The degree of settlement (and thus airspace gain) should account for
the anticipated operations time of the sustainable landfill technologies,
operating conditions of the sustainable landfill technologies (e.g.,
leachate recirculation rates), waste composition, and waste dimensions
Landfill infrastructure The disturbance of landfill infrastructure as part of re-filling gained
disturbance airspace (e.g., temporary covers like EGCs, GCCS infrastructure) must
be accounted for of re-filling of airspace is to occur after the landfill
initially reaches capacity
Social or external costs of landfilling include (1) amenity and land use impacts, (2)
pollutant emissions, and (3) damages due to greenhouse gas emissions. Social costs
are generally more difficult to quantify, but must be considered in estimating the
value of sustainable landfills compared to traditional landfills.
Amenity losses may include odors, noise, visual intrusion, reduced property val-
ues, attraction of animals, traffic, and social stigma. These impacts are immediate,
affecting the generation that created the disposed waste. However, as local impacts,
their effects may be felt disproportionally by the immediate landfill neighbors rather
than the true population that it serves. In many cases, those affected by lost ameni-
ties are compensated by landfill owners through profit sharing or community
resource building, but this does not ensure sustainability of landfilling. The value of
amenity losses is relatively low and has been estimated at approximately 1 USD/ton
or 2 % of a landfill’s PW (Mery and Bayer 2005).
Communities operating sustainable landfills will experience similar amenity
losses to traditional landfills, although the impact of odors could be greater due to
greater LFG production rates if the LFG is not efficiently collected. Conversely, a
sustainable landfill could be touted as being “green”, thereby reducing the social
stigma of hosting a landfill. The value of a “green” landfill warrants further research.
Pollutant emissions could occur as a result of uncontrolled leachate (either
breaches in liners or uncollected leachate after the end of PCC). Although leachate
during active operation of a landfill is largely controlled in well designed and con-
structed landfills, estimating future pollutant emissions requires making assump-
tions about the integrity of liners over many years (perhaps centuries). Long-term
risks would be expectantly lower than traditional landfills due to faster waste stabi-
lization. Estimating the cost of pollutant emissions is challenging, but again these
external costs should be considerably lower than other cost components.
As discussed earlier, the release of GHG emissions is an important factor when
considering sustainable landfills compared to traditional landfills. Absent efficient
LFG collection systems, the use of sustainable landfill technologies represents a
greater potential for GHG emissions in the early stages of a landfill’s life compared
to a traditional landfill. However, if the LFG is controlled efficiently, substantial
reductions in lifetime GHG emissions can be realized as was reported by Amini and
Reinhart (2012). The external cost of damages due to GHG releases are estimated
to be 21 USD/ton of CO2 (Handley 2010). These costs could be offset by financial
benefits (both external and private) due to sale of landfill gas or electricity/heat gen-
erated by the gas. External benefits would only occur if the sales offset the use of
fossil fuel. Thus, for example, these benefits would be more pronounced if the
energy offsets that which would be produced through coal combustion, but would
be less pronounced if the energy replaces that derived from nuclear sources. This is
a fundamental consideration when examining life-cycle impacts of sustainable
landfilling (Chap. 3).
438 18 Economics
After closure, the landfill enters a PCC period that involves maintaining and moni-
toring the site for a regulatory- or permit-defined period (and possibly for perpetu-
ity). As described many times already in this book, one of the major motivating
forces behind the implementation of sustainable landfill technologies is the rapid
stabilization of waste in the landfill so that the potential for deleterious environmen-
tal emissions is greatly reduced. Given that a primary driver in PCC is to make sure
that the landfill does not produce such harmful emissions, rapid waste stabilization
should, in theory, result in a landfill that can exit PCC requirements much sooner
than traditional landfills, or at least the degree of monitoring can be reduced sooner.
Chapter 17 outlined a potential approach for transition a landfill from PCC to a
custodial care phase where only minimal maintenance and operation are required
(Barlaz et al. 2002; ITRC 2006). The proposed framework involves monitoring
landfill emissions that may pose a risk to the human health and the environment as
part of PCC and comparing resulting data and trends to accepted risk levels to deter-
mine when a change in PCC is warranted. The framework involves collecting and
evaluating data as part of a modular assessment of four primary landfill components
(leachate collection and control system, landfill gas collection and control system,
groundwater monitoring system, and cap system).
Leachate quality should reach stable conditions more rapidly for facilities prac-
ticing sustainable technologies, although the volume of leachate may be greater for
a period of time. As long as the final cover system has been adequately maintained,
this should result in reduced long-term risk to groundwater resources. When waste
is rapidly stabilized, potential maintenance issues with the final cover system can be
addressed earlier, thus mitigating long-term performance issues. Gas production
will decrease more rapidly when liquids addition is practiced, thus reducing the
longer-term risks posed by methane escape to the atmosphere.
18.4 Costs and Benefits After Landfill Closure 439
Chapter 17 described the process of landfill reclamation, where landfill waste and
cover soil are excavated (mined) and processed. Proposed targets of landfill reclama-
tion include facilities where the waste reaches a point of adequate stabilization such
that the landfilled material and additional disposal capacity can be recovered. The soil
fraction, when segregated from waste, can be beneficially used to replace new soil
used for cover material, and possibly beneficially used outside of the landfill. Potential
economic benefits include the avoided cost for new cover soil, the sale of recyclable
materials reclaimed from a landfill, and airspace gain for new waste disposal. Such
practices can greatly extend the operating life of the landfill and allow the receipt of
additional tipping fees with much reduced expenditure of capital construction costs,
as well as closure and PCC costs. The overall waste disposal footprint becomes less
and thus carries with it the other benefits associated with decreased land use.
The reclamation process, of course, comes with an added expense in terms of
equipment and labor. Table 18.8 presents the reclamation costs associated with a
number of projects conducted in the US. Cost elements associated with the reclama-
tion include excavation equipment and operation, processing costs (screens,
magnets), labor cost, hazardous or problematic waste screening, and management
of bulky items. Material transportation costs represent another major project
expense, and depend on the number of size fractions the excavated waste is sepa-
rated into, the production rate of each fraction, haul distance, and route condition
and traffic. Other costs not listed above may be associated with the execution of a
mining project such as design, permitting, mobilization/demobilization, other envi-
ronmental considerations, and contingencies.
The cost effectiveness of landfill reclamation will usually depend on the amount
of material that can be separated for beneficial use. Some fraction of the excavated
material will require re-disposal in the landfill, and if this fraction is large, the pro-
cessing costs will outweigh any savings associated with additional airspace recov-
ery. While some of the airspace recovery costs will be associated with reclaimed
materials for recycling, most will derive from the use of previously placed cover soil
and degraded organic waste to replace cover soil in the active landfill operation (or
to use for application outside the landfill unit). An economic feasibility analysis for
landfill reclamation should include field investigations (e.g., auger borings, test pits)
to estimate the relative amount of material that can serve this purpose.
440 18 Economics
Table 18.8 Summary of excavation volumes and costs for landfill mining projects in the US
(IWCS 2009)
Waste
Years of volume mined Mining
Project City, State operation (million m3) cost ($/m3)
Phoenix Rio Salado Project City of Phoenix, AZ 1999–2005 0.31
Clovis Landfill City of Clovis, CA 1998–2008 1.61 $6.33
Naples Landfill East Naples, FL 1986–ongoing $2.94a
Dean Forest Landfill, Statham, GA 1997–2006 0.50
City of Savanna
Central Disposal Systems Lake Mills, IA 2000–ongoing 0.19
Town of Edinburg Town of Edinburg, 1990–1992 0.04 $6.54
NY
Pike Sanitation Landfill Waverly, OH 1996–2000 0.54–0.61
Wyandot County Carey, OH 1999–ongoing 1.07 $5.23b
Environmental Sanitary
Landfill
Frey Farm Landfill Lancaster, PA 1990–1996 0.23–0.31 $11.51
La Crosse County WI 2005–2007 0.31
Shawano County WI 2001–2002 0.23–0.31 $3.92a
Perdido Landfill FL 2009–2011 0.38 $8.37
a
Per ton basis
b
No waste processing. Waste was relocated from unlined cell to a lined cell
References
Amini H, Reinhart D, Mackie K (2012) Determination of first-order landfill gas modeling param-
eters and uncertainties. Waste Manag 32(2):305–316
Barlaz M, Rooker A, Kjeldsen M, Gabr M, Borden R (2002) Critical evaluation of factors required
to terminate the post-closure monitoring period at solid waste landfills. Environ Sci Technol
36(16):3457–3463
Berge N, Reinhart D, Batarseh E (2009) An assessment of bioreactor landfill costs and benefits.
Waste Manag 29(5):1558–1567
Duffy D (2005a) Landfill economics part I: siting. MSW Management, May–June 2005
Duffy D (2005b) Landfill economics part III: closing up shop. MSW Management, Sept–Oct 2005
Handley J (2010) Government panel estimates cost of CO2 pollution: $21/t and rising. www.car-
bontax.org/blogarchives/2010/06/03/govt-panel-estimates-cost-of-c02-20t-and-rising/.
Accessed Feb 2011
ITRC (2006) Evaluating, optimizing, or ending post-closure care at municipal solid waste landfills
based on site-specific data evaluations. The Interstate Technical & Regulatory Council,
Alternative Landfill Technologies Team, Washington, DC. www.itrcweb.org
IWCS (2009) Landfill reclamation demonstration project. A report prepared by Innovative Waste
Consulting Service s, LLC and submitted to Florida Department of Environmental Protection
and Escambia County Division of Solid Management. http://bit.ly/1lvjqWZ. Accessed 11
Mar 2014
KDEP (2012) Landfill permitting overview. The Kentucky Department for Environmental Protection
MDE (no date) Estimated costs of landfill closure fact sheet. Maryland Department of the
Environment, Baltimore
References 441
Mery J, Bayer S (2005) Comparison of external costs between dry tomb and bioreactor landfills:
taking intergenerational effects seriously. Waste Manag Res 23(6):514–526. doi:10.1177/0734
242X05060857
Powell J, Townsend T (2004) An economic analysis of aerobic versus anaerobic bioreactor land-
fills. In: 19th international conference on solid waste technology and management, Philadelphia,
PA, Mar 2004
US EPA (2015) LFG energy project development handbook. Landfill methane outreach program.
http://www.epa.gov/lmop/publications-tools/handbook.html
van Haaren R, Themelis N, Goldstein N (2010) The state of garbage in America. In: 17th nation-
wide survey of MSW management in the US. Biocycle, Oct 2010
Chapter 19
The Role of Landfills in Integrated Materials
and Energy Recovery Facilities
Abstract The book concludes with a discussion of sustainable practices for landfill
design and operation within the broader context of integrated waste management
and associated facilities. With the rising importance of preserving material resources
and avoiding wasted potential energy, facilities implementing sustainable landfill-
ing techniques can serve as a companion to other processes in an integrated fashion
to create a more sustainable materials management system overall. The employ-
ment of technologies to extract energy from landfill gas and to use the landfill itself
as an energy production center through solar cells or wind power, in addition to
using landfills as waste treatment and materials recovery cells, are just a few exam-
ples of the opportunities explored.
This book began by discussing the evolution of solid waste management technologies
over the past half-century. In many countries, waste management has transitioned from
open dumps to sanitary landfills, while some have gone beyond sanitary landfilling
into more advanced waste treatment and recovery processes. This book provides guid-
ance on how landfill owners and operators can plan, design, and operate landfills to be
more sustainable, which is of critical importance given continued reliance on landfill-
ing worldwide as a method to manage waste. Although landfills are still the least pre-
ferred option in most waste management hierarchies, landfilling is the method used to
handle approximately 70 % of the world’s wastes. The landfills we create today will
remain, and based on projections (World Bank 2012), we will continue building and
using landfills for many years to come. In light of this trend, it is of critical importance
to view landfills as opportunities to innovate and to improve the manner in which land-
fills are designed and operated to optimize resource use and energy recovery.
Thus far this book has described both fundamental and practical aspects of sus-
tainable strategies for landfill design and operation. Topics have largely focused on
methods to design and operate so deleterious impacts to human health and the
Given that landfills will still be a necessary piece of an overall waste management
system, the landfill site can host of a multi-process facility where recyclables and
other materials of value can be extracted from the waste stream, potentially con-
verted (through physical, biological, and/or chemical processes) and shipped off
site for reuse or beneficially used on site. An example of on-site resource recovery
is the operation of a co-located materials recovery facility for the recycling of waste
constituents such as aluminum and plastic containers, followed by consolidation
and export from the site to a manufacturing facility, also possibly located adjacent
to the landfill. In this case, the landfill acts as a central facility where waste materials
are delivered and a separate processing area on site is used to provide for the extrac-
tion of targeted recyclables. Any discards from the recovery operation can be easily
transported to the landfill.
An example of on-site reuse is the segregation of vegetative waste materials,
subsequent size reduction, and ultimate use as a cover material at the landfill. In this
case, the reuse is beneficial because it provides a productive use for a waste material
and avoids the extraction and transport of virgin materials (e.g., nearby soils) that
may have otherwise been used as a cover material. Some of the vegetative material
would convert to gas for possible recovery as part of a LFG-to-energy system.
Table 19.1 provides several more examples of waste materials that may be benefi-
cially used at the landfill itself.
As mentioned throughout the book, despite the presence of other technologies
that can be considered more sustainable than landfills (e.g., composting facilities,
energy-from-waste facilities), the landfill still plays a critical role in the event that
waste production outstrips available capacity at these other facilities, if the facilities
experience downtime due to equipment maintenance or failure, for disposal of WTE
and MRF residuals, or if the processed materials do not meet specifications. Thus,
the co-location of other waste processing technologies or facilities at the landfill site
can provide economic and environmental benefits such as reduced transportation
costs, utilization of energy produced and harnessed at the landfills to power these
facilities, and the potential to save costs on labor by personnel cross-training to
perform multiple tasks or functions at different co-located facilities. These facilities
can serve as community centers that encourage and promote sustainable materials
management (Fig. 19.1).
19.2 The Role of Landfills in Integrated Waste Management 445
Table 19.1 Examples of waste materials that may be beneficially used at landfills
Waste material Beneficial use at a landfill
Ash from This material may be used as an alternative daily cover material
waste-to-energy
Asphalt shingles Un-processed shingles can be used on interior landfill roads to improve
access and reduce dust generation
Glass Crushed glass can be used as a permeable medium provided it meets
required specifications. Examples may include permeable media
surrounding liquids addition devices or gas extraction devices
Tires Size-reduced tires may be a permeable medium used in liquids addition
systems or gas collection systems
Yard waste/ Size-reduced yard waste can be used as a cover material at the landfill’s
vegetative waste working face or potentially in other areas that require an intermediate cover
(e.g., areas where, based on the filling sequence, waste will not be placed
for several months or longer). Depending on the chemical quality and local
restrictions, mulched yard waste may be used on site for landscaping or
marketed as a product for businesses or individuals in the community
As discussed in Chap. 17, landfills can serve as a repository for materials that
may not currently have sufficient value to warrant extraction at the time of delivery
to the landfill, but may have enough value in the future to justify excavation, pro-
cessing, and resource and energy recovery. Jain et al. (2014) reported that signifi-
cant environmental benefits can be realized with recovery of resources deposited in
MSW landfills. When a landfill is included as a part of a larger integrated waste
management facility, the tracking and planning that can go into landfill mining is
446 19 The Role of Landfills in Integrated Materials and Energy Recovery Facilities
greatly facilitated relative to a case where waste handling facilities are scattered
throughout a community. As an example, the co-location of a landfill with a waste
to energy (combustion) facility would allow the landfill to act as temporary storage
in cases where the waste acceptance rate at the facility exceeds that which can be
combusted—in this case, the waste is placed in the landfill and extracted at a later
time when capacity becomes available.
LFG extracted using a GCCS can be described as a medium energy value gas
because of the relatively high CO2 content that is present. In this form, the gas can
be utilized for energy recovery with a variety of technologies. Alternatively, the
LFG can be first processed or cleaned to produce a high energy value gas that opens
up other energy recovery opportunities. The specific market need for the gas dic-
tates the type and level of processing and treatment required to deliver the gas in a
form that meets the necessary specification or energy project objectives.
As discussed in Chap. 13, the kinetics of LFG production are altered when oper-
ating a landfill with liquids addition. If necessary GCCS components are in place
and are designed to accommodate the liquids addition system components (and the
greater amounts of liquids), tremendous opportunities exist to enhance the viability
and effectiveness of a LFG beneficial use project. While the ultimate volume of gas
that can be produced from the waste remains the same, the period of production is
compressed into a smaller timeframe that may enhance the economic viability of a
LFG-to-energy project. These higher gas production rates should be planned at the
design stage of the GCCS, particularly since some LFG beneficial use technologies
and markets may be more sensitive to variation in LFG collection rates than others.
The following sections outline the major types of LFG beneficial use options
available for energy conversion. These include conversion to energy, medium
energy content (medium BTU) application, and high energy content (high BTU)
applications. Where appropriate, we provide specific commentary on how each
technology’s use may be impacted by sustainable landfilling operations.
Electricity generation is one of the most common techniques to harness the energy
content of LFG. Some of the benefits of electricity generation include many years of
demonstrated success at hundreds of landfill sites, operating parameter flexibility,
and the ability to expand or contract the system in response to increasing or decreas-
ing LFG collection rates. When electricity is produced, it can be used for on-site
power needs or sent to the power grid. A variety of technologies are available to
generate electricity from collected LFG, many of which are summarized in Table 19.2.
19.3 Beneficial Use of LFG 447
Table 19.2 Technologies that may be used to convert collected LFG into electricity
Technology Description
Cogeneration Generate thermal energy and electricity from steam or heated water.
(Combined heat Can be installed to recapture heat losses from turbines and engines to
and power, CHP) produce steam, thus increasing the overall efficiency to as much as
85 % (US EPA 2010; ACEEE 2009)
Combined This system utilizes both gas and steam turbines. The gas turbine
CycleEngine provides the heat needed to generate steam that is then fed to the
steam turbine. Combined cycles are utilized for scales larger than
most internal combustion projects (US EPA 2010) Efficiencies for
combined cycles range from 54.5 % to 60 % (MNSU 2014)
Gas Turbine Can operate at lower CH4 concentrations; gas turbines typically
require larger volumes of gas for economic feasibility resulting
in effciencies as large as 60 % (US Department of Energy 2014;
US EPA 2010). More resistant to wear and damage than
other systems
Internal A common type of electricity generation technology, efficiencies
Combustion Engine typically range from 25 % to 35 % (US EPA 2010). CHP can
be implemented with internal combustion engines as well
to further enhance overall system efficiency. The LFG may
need to be pretreated for removal of contaminants such as
siloxanes and H2S
Microturbine Used for smaller-scale power generation operations; units with rated
capacity as low as 35 kW are commercially available. Typically
employed in areas with lower gas flow rates. Pretreatment of LFG to
remove moisture is necessary in addition to the usage of activated
carbon to remove other impurities. Microturbines can operate at low
CH4 concentrations. Efficiencies for this system ranges from 20 % to
30 % (US EPA 2010)
Boiler/ LFG is directly used by combusting it in a large boiler to generate
Steam Turbine steam that is fed to a steam turbine. Generating electricity in this
manner is fairly uncommon (US EPA 2010)
Stirling Engine An external combustion engine that mixes air and fuel within the
cylinder of the unit to facilitate combustion. Pretreatment of LFG is
not needed because of the engine’s high tolerance for siloxanes and
other such impurities. An average efficiency obtained is 30 %
(US EPA 2010)
Internal combustion engines are one of the more common electrical generation
technologies for LFG (Fig. 19.2). In electricity generation applications, a key design
consideration involves the examination of actual and projected LFG collection rates
so that the engine(s) can be economically phased in and out of operation. Ideally,
LFG-to-energy projects are sized based on actual historical collection rates rather
than desktop projections. This is especially true in the case of sustainable landfills,
as field data demonstrating greater LFG production and collection rates would serve
as necessary justification to implement greater electric generating capacity.
448 19 The Role of Landfills in Integrated Materials and Energy Recovery Facilities
Fig. 19.2 Internal combustion engine for converting landfill gas to electricity
Medium energy applications are often referred to as direct use in that the LFG is
utilized without elaborate LFG processing, often in an industrial process such as a
boiler or a kiln. Another direct use application of LFG includes leachate evapora-
tion. The combustion equipment at the receiving facility often requires minimal
modification to accept the LFG, and the benefit to the receiving facility can range
from partial or complete replacement of other fuels such as natural gas. End users
typically must have some baseline or steady fuel demand for the benefits of LFG
utilization to be maximized, although other industrial plants that may only need fuel
on a periodic basis (e.g., a batch asphalt plant) could still be a viable option. The
distance from the landfill to the end user must be evaluated as part of the planning
and permitting process. Although gas treatment is normally not required, removal of
condensate, and possibly corrosive gases and particulate matter, is typically done.
The benefits of enhanced gas production from sustainable landfill operations are
only truly realized if the additional gas is captured and the LFG-to-energy system’s
demand is large enough to accommodate additional energy potential.
While the CH4 content of LFG is sufficiently high for combustion in many types of
energy recovery units, several additional beneficial use options become available
when the gas is cleaned to a level similar to natural gas. Such applications are
19.3 Beneficial Use of LFG 449
referred to as high energy content (or high BTU content) because the LFG con-
stituents (e.g., CO2) that do not have sufficient energy content that can be effi-
ciently harnessed are removed. Several technologies are available to remove major
and trace LFG constituents as summarized in Table 19.3. These include various
chemical wash technologies, membrane separation, and pressure swing adsorption
(Fig. 19.3).
Once major problematic trace gas components are removed from LFG, the resul-
tant gas can be used in several high BTU applications. Like other LFG beneficial
use technologies, the selection of a given energy use or conversion technology
depends on numerous factors including availability and quality of LFG, capital and
operating cost of the technology, demonstrated use of the technology, and availabil-
ity of end uses or users for the final product following conversion of the
LFG. Table 19.4 presents some of the technologies that can be employed in high
BTU applications.
Table 19.3 Summary of LFG cleanup technologies for high BTU applications
LFG cleanup
technology Technology description
Selexol process Uses a solvent derived from dimethyl ether and polyethylene glycol for
removing NMOCs, CO2, H2S, and water vapor. The solvent is regenerated at
the end of the process and recycled. This process does not remove N2 and O2
Kryosol process Uses methanol to physically absorb water, CO2, and other trace constituents
such as heavy hydrocarbon and H2S in a stepwise fashion. Methanol is
regenerated and reused in the process. The recovered CO2 can be used to
produce food-grade quality liquid CO2 that can also be sold
CO2 wash Gas is treated to remove H2S and water vapor before it enters a CO2 wash
process column where the gas is cooled to liquefy and accumulate on the top tray of
the column. A portion of this liquid CO2 is sent down to adsorb LFG
contaminants (mainly volatile organic compounds). The exit gas constituents
are CH4 (75 %), CO2 (25 %), and any O2 and N2 present in the inlet LFG;
this process cannot remove O2 and N2
Membrane Raw LFG is introduced into a vessel filled with separation polymers
technology (typically consisting of a bundle of hollow fibers), separating CO2 from CH4,
taking advantage of the fact that different LFG constituents flow through
polymeric membranes at different rates. Provides limited removal of O2 but
N2 is not removed
Pressure swing Separates CO2 from CH4 by selective adsorption of CO2 on the surface of
adsorption special porous solid absorbents. The adsorption occurs at an elevated
pressure, and when the pressure is reduced, the adsorbed CO2 desorbs.
Because of a cyclic, continuous change in pressure, this technology is
referred to as pressure swing adsorption. Two types of adsorbents used for
cleanup of LFG are molecular sieve and activated carbon. A molecular sieve
is a packed bed of granular material, typically aluminosilicate minerals
called zeolites. These materials are porous and have a high internal surface
area that can adsorb CO2. The raw LFG must be pre-treated to remove
sulfides and water vapor for an effective adsorption of CO2. A molecular
sieve can be configured to remove N2. The process does not remove O2
450 19 The Role of Landfills in Integrated Materials and Energy Recovery Facilities
Fig. 19.3 Pressure swing adsorption and membrane equipment for cleaning up landfill gas to high
energy content
The space occupied by landfills, including the disposal area and associated buffer
space and support facilities, represents an additional opportunity to recover energy
beyond the conversion of collected LFG. Landfill sites offer unique advantages for
hosting renewable energy technologies. For example, landfill sites have infrastruc-
ture that supports electricity generation and, because of their commonly rural loca-
tions, may not be a viable host for technologies that clean-up the gas to natural gas
quality because of a limited base of potential users. New energy technologies pro-
vide additional revenue to site owners and job opportunities in rural areas.
Renewable energy projects can offset the environmental impacts of fossil fuel-
based options and of the landfill itself. In considering the use of a landfill as an
19.4 Additional Energy Opportunities 451
energy park, several factors have to be considered including available area, climate,
geographic location and compatibility with the site’s closure plan. Two examples
discussed in this section are solar panels and wind turbines. Detailed design, opera-
tion, permitting, and cost discussion lies beyond the scope of this book, but the fol-
lowing discussion presents fundamental considerations associated with these energy
options and how these considerations tie in with conditions at sustainable landfills.
Landfills can provide favorable opportunities for solar power generation once they
are closed; landfills typically have large exposed areas where solar photovoltaic
panels can be placed. Harnessing solar power in this manner may represent the larg-
est energy generation opportunity for closed landfills (Millbrandt et al. 2013).
Table 19.5 details the reported characteristics that influence the suitability of solar
infrastructure development at landfills (US EPA 2013).
Once a landfill is determined to be a locational fit for a solar project develop-
ment, guidance related to matching up appropriate photovoltaic technology to the
landfill site should be consulted. Integration of solar panels with power generation
infrastructure is an integral step in the solar energy implementation. Several factors
may facilitate this process (Messics 2009a), including close proximity of power
lines (e.g., three-phase power may be needed for large installations), the local util-
ity’s need for renewable energy, and the presence of a LFG to energy plant on-site.
As LFG production declines, energy production capacity can be supplemented with
solar power. The use of generated power on-site is generally more financially ben-
eficial (since it replaces retail-rate power) than wholesale of generated power to the
grid, but the benefits of sending some or all of the produced power to the grid should
be considered at the feasibility analysis step of the solar project.
Table 19.5 Fundamental considerations for the utilization of solar power at landfills
Fundamental
consideration Description
Meteorological For economic feasibility based on current panel/flexible panel
conditions installation cost, a minimum of 3.5 kWh/m2/day of solar radiation is
generally advised, in addition to at least 6 h of sufficient sunlight on the
winter solstice (lowest yearly sunlight exposure) as a baseline. Optimal
topography includes flat or gently sloping grades (US EPA 2013)
On-site energy needs A solar photovoltaic system can be capable of meeting 100 % to 120 %
of a landfill’s on-site energy requirements (US EPA 2013)
Grants or incentives Economic incentives increase the overall value of energy produced
that are in place from photovoltaic cells. The Database of State Incentives for
(tax breaks) Renewable Energy (DSIRE) is a guide that provides information
regarding grants, incentives and policies on federal, state, local, and
utility levels in the US (Messics 2009a; US EPA 2013)
452 19 The Role of Landfills in Integrated Materials and Energy Recovery Facilities
power system at the site complements an on-site LFG-to-energy plant. The panels
were chemically adhered to the geomembrane on the south-facing side slope. The
electrical conduits were installed in the anchor trenches for the geomembrane.
Approximately 7,000 flexible solar panels were installed over approximately 10-acre
area of the 45-acre landfill cell at Hickory Ridge Landfill (Atlanta, Georgia) in 2011
(Fig. 19.5). The panels were installed on top the thermoplastic exposed geomem-
brane cap. The total cost for this 1-MW system was reported to be USD 5 million.
From a sustainable landfill perspective, one unique opportunity lies in utilizing
an exposed geomembrane cap (EGC) installed early (e.g., when a side slope meets
its design grade but well before a final cap is to be installed) to capture LFG during
initial stages of gas production and coupling solar panels with the EGC. In this case,
the landfill operator would realize the benefits of additional gas production while
also harnessing solar power through the use of EGC-mounted panels. As with solar
energy projects with closed landfills, the feasibility scenario described above would
need to be examined on a site-specific basis and account for aforementioned poten-
tial challenges such as panel placement, potential settlement, and other factors.
Wind energy projects involve placement of wind turbines on large towers. Turbine
blades rotate in response to passing wind movement and this movement turns the
shaft of a generator to produce electrical power. A transformer at the base of
the tower steps up the voltage to the necessary level for the accompanying power
19.5 Wind Power at Sustainable Landfills 455
distribution system. Turbines are typically spaced apart 14 or more times the blade
length. Industrial size blade lengths range from 100 to 150 fit in length.
In regions where sustained winds support economically viable wind power,
closed landfills offer potential locations for wind energy projects (Fig. 19.6).
Millbrandt et al. (2013) reports that wind power ranks second (behind solar) in
opportunity for renewable energy resource development on landfills in the US
(including closed landfill sites), with a total potential of 2,700 TWh (energy effi-
ciency of 30–65 %, based on class II turbines with an installed energy generation
density of 5 MW/m2). Wind speeds of approximately 16 km/h (10 mph) are nomi-
nally required; wind speeds increase at greater distances from the ground. The wind
speed affects the type of wind turbine that would be selected for a site, class II, the
most common type of turbine is typically for sites up to 8.5 m/s average wind speed.
A landfill site may be ideally suited from a location perspective as it is often
higher in elevation than surrounding land, provides a large area of tree- and building-
free land, and is often already located a sufficient distance away from homes and
businesses. The presence of the LFG to energy system at a site, as with solar proj-
ects, increases feasibility of wind projects at a landfill site due to in-place power
transmission infrastructure (although the complications of numerous piping sys-
tems infrastructure may pose a technical challenge; Millbrandt et al. 2013).
Table 19.7 details relevant resources and considerations for siting of wind turbines,
sites should have adequate, sustained wind speeds.
US EPA’s Wind Decision Tree is one resource available for wind-turbine siting.
Computer-based geographic information systems (GIS) can also aid in consider-
ation of many siting factors at once (economic as well as environmental and eco-
logical impacts). While Millbrandt et al. (2013) did not consider wind power at
landfill sites, given the abundance of other marginal lands and the relatively small
areas of landfills, collocation may be feasible if standard wind power constraints are
456 19 The Role of Landfills in Integrated Materials and Energy Recovery Facilities
Table 19.7 Siting and land usage considerations for wind turbines at landfills
Resources and
considerations Description
Wind resource maps These maps show locations where strong, sustained winds are
expected based on historical data on wind speeds and area
elevation above sea level (US EPA 2012)
Topography Landfills greater than 80 m above sea level and with wind speeds
below 5.5 m/s are not appropriate for wind power (US EPA 2012)
Land use considerations Generally, sites like landfills are preferred when possible for
siting wind turbines so that green space can remain undisturbed,
sometimes referred to as “marginal lands” and estimated at
roughly 11 % of total US land in the contiguous 48 states
(Millbrandt et al. 2013). These marginal lands include landfills,
brownfields, abandoned crop land, other barren lands
Landfill-specific At sites with a LFG to energy system, power generated via wind
considerations turbines can be “piggybacked” onto existing power infrastructure
(either on the waste
footprint or within site
boundaries)
Exclusions Some criteria that would preclude a site from wind turbine
installation (Millbrandt et al. 2013):
• Slopes >20
• High-value lands
• Urban areas
met. van Haaren and Fthenakis (2011) reported use of GIS for a state-wide assess-
ment of potential wind-farm sites in the state of New York; infeasible sites were first
excluded, then economic assessment of remaining sites was performed, and impact
on birds were considered. Landfills have the benefit of having no land clearing
requirement, which can be a substantial cost (68–84 % of total project cost) (van
Haaren and Fthenakis 2011).
A challenge for installing wind turbines on landfills is the design of a foundation
that provides necessary support. Foundation types include spread footings, deep
anchors, and tensionless pier foundations. In addition to utility-scale wind power
projects, the use of a small number of turbines or chimneys possibly with storage
capacity to provide small, site-scale power to provide energy for day-to-day landfill
functions, such as sump pump, gas collection system, air blower (for air circulation
to waste in aerobic systems) operation has been suggested (Stormont et al. 1998).
Hickman et al. (2014) conducted a study evaluating the potential for closed
Florida landfill sites to be used as energy parks. A screening tool was created that
utilized broad criteria such as landfill location, size, and site conditions to select
landfill sites that might be suitable for three alternative energy technologies, landfill
gas to energy, solar power, and wind power. These criteria were based on readily
available data, such as atmospheric and weather conditions (e.g., historic wind
speeds, cloud cover, and precipitation), landfilled tonnage, area availability, and
surface irradiation. Landfills that were potentially suitable for the technologies were
further evaluated using site-specific variables. Technologies were evaluated with
19.6 Landfills as Waste Treatment and Materials Recovery Operations 457
19.6 L
andfills as Waste Treatment and Materials
Recovery Operations
Landfills by their nature are intended to be the final resting place for discarded solid
waste. Throughout this book, practices to enhance the stabilization of landfills were
presented and techniques to extract energy at or from landfills were described.
Additionally, the concept of landfill reclamation (introduced in Chap. 17) raises
possibilities for perhaps the most sustainable manner in which a landfill might be
operated: a treatment operation where the landfill cell serves as a temporary treat-
ment unit designed to be emptied and later refilled.
Figures 19.7, 19.8, 19.9 and 19.10 illustrate this concept. The first landfill unit
would be constructed, filled with waste, and operated using practices such as liquids
addition and LFG collection and beneficial use (Fig. 19.7). Unit 2 would be built as
Unit 1 is filled. After reaching capacity, Unit 1 would be closed using technologies
such as an EGC (possibly equipped with solar cells) that would be less permanent
(and less costly) than a traditional final cover system. Unit 1 would continue to be
operated to stabilize the waste and harvest LFG while Unit 2 was filled (Fig. 19.8).
Unit 3 would come on line as Unit 2 reached capacity (Fig. 19.9). During this
time, Unit 1 would be at the point where the waste is largely stabilized and thus
prepared for reclamation. While Unit 4 operates, Unit 1 would be mined and made
ready for acceptance of new waste upon closure of Unit 4 (Fig. 19.10). In this con-
ceptual model, it is expected that some residual materials will be left over. As
described in detail in Chap. 17, the mining process could involve varying degrees of
material screening during the excavation process. The ultimate volume that would
be reclaimed in this process would depend on the degree of stabilization that the
waste achieved during sustainable landfilling operations, the nature of the waste, and
other factors. But this concept illustrates an idealized version of what sustainable
landfilling can be when planned from the beginning and cells are built, sequenced,
operated, and harvested with the primary concept of preparing the waste to be
treated, treating the waste, and utilizing the stabilized residuals. In light of society’s
anticipated continued reliance on landfilling as a means of managing discarded
458 19 The Role of Landfills in Integrated Materials and Energy Recovery Facilities
Fig. 19.7 Conceptual sustainable landfill operation. Cell 1 constructed and operated
Fig. 19.8 Conceptual sustainable landfill operation. Cell 1 closed, treated, and gas harvested. Cell
2 constructed and operated
19.6 Landfills as Waste Treatment and Materials Recovery Operations 459
Fig. 19.9 Conceptual sustainable landfill operation. Cell 1 operated and decommissioned; Cell 2
closed, treated, and gas harvested for energy; Cell 3 constructed and operated
Fig. 19.10 Conceptual sustainable landfill operation. Cell 1 reclaimed; Cell 2 operated and decom-
missioned; Cell 3 closed, treated, and gas harvested for energy; Cell 4 constructed and operated
460 19 The Role of Landfills in Integrated Materials and Energy Recovery Facilities
References
ACEEE (American Council for an Energy-Efficient Economy) (2009) Combined heater and power
and clean distributed energy policies. Washington, DC. http://www.aceee.org/files/pdf/fact-
sheet/chp_policyposition0809.pdf
EPIA (2012) Solar photovoltaic technology. http://www.epia.org/about-us/about-photovoltaics/
solar-photovoltaic-technology/
Hickman N, Reinhart D, Toth M (2014) A multi-level decision tool for converting landfill Sites
into sustainable energy parks. Submitted to the Hinkley Center for Solid and Hazardous Waste
Management, Gainesville
Jain P, Powell J, Smith J, Townsend T, Tolaymat T (2014) Life-cycle inventory and impact evalu-
ation of mining municipal solid waste landfills. Environ Sci Tech 48(5):2920–2927
Messics MC (2009a) Site considerations—what makes a site desirable for a solar project? renew-
able energy at closed landfills workshop, Mansfield/Foxboro Holiday Inn, MA, 17 June 2009
Messics MC (2009b) Case study—Pennsauken landfill solar project, Pennsauken. In: Renewable
energy at closed landfills workshop, Mansfield/Foxboro Holiday Inn, MA, 17 June 2009
Millbrandt AR, Heimiller DM, Perry AD, Field CB (2013) Renewable energy potential on mar-
ginal lands in the United States. Renew Sustain Energy Rev 29:473–481
MNSU (Minnesota State University) (2014) Combined cycle plant. Mankato. http://cset.mnsu.
edu/engagethermo/systems_combinedcycle.html
Sampson G (2009) Solar power installations on closed landfills: technical and regulatory consid-
erations. US Environmental Protection Agency, Office of Solid Waste and Emergency
Response, Office of Superfund Remediation and Technology Innovation, Washington, DC
SRA International (2008) Technical assistance: solar power analysis and design specifications.
U.S. Environmental Protection Agency’s Brownfields Program
Stormont JC, Ankeny MD, Kelsey JA (1998) Airflow as monitoring technique for landfill liners.
J Environ Eng—ASCE 124(6):539–544
Tansel B, Varala PK, Londono V (2013) Solar energy harvesting at closed landfills: energy yield
and wind loads on solar panels on top and side slopes. Sustain Cities Soc 8:42–47
US Department of Energy (2014) How gas turbine power plants work. Washington, DC. http://
energy.gov/fe/how-gas-turbine-power-plants-work
US EPA (2010) Landfill methane outreach program project development handbook. United States
Environmental Protection Agency, Washington, DC
US EPA, OSWER, OCPA, RE-Powering America’s Land Initiative (2012) Wind decision tree
US EPA (2013) Best practices for siting solar photovoltaics on municipal solid waste landfills.
Technical report: NREL/TP-7A30-52615
van Haaren R, Fthenakis V (2011) GIS-based wind farm site selection using spatial multi-criteria
analysis (SMCA): evaluating the case for New York state. Renew Sustain Energy Rev
15:3332–3340
World Bank (2012) What a waste: a global review of solid waste management. Urban development
series
Index
D EGC
Darcy velocity, 104 advantages, 405
Darcy-Weisbach Equation, 147 construction, 404–405
Delaware Solid Waste Authority (DSWA), operational and maintenance issues, 405
55–57 vs. traditional final cover system, 406
Denitrification, 135, 261, 319 final site use and configuration, 421–422
Dewsbury Landfill, UK, 85–86 leachate and gas management, 406–407
Direct push technology (DPT), 169 PCC
Drainable porosity, 99, 182, 216 components, 408, 409
Drip irrigation system, 42 modular approach, 409
ACSWL, 57–59 potential approach, 407, 409
liquid surface systems, 155–156 risk assessment, 408
Dry tomb landfills, 9, 32 reclamation and reuse
DSWA. See Delaware Solid Waste Authority excavation technique, 412, 413
(DSWA) fundamentals, 410–412
health and safety requirements, 418
mining processing equipment, 414, 415
E process flow chart, 412, 413
Economics. See Landfill economics reclaimed material composition,
Effective porosity. See Drainable porosity 419–421
EGC. See Exposed geomembrane cap (EGC) reclamation projects, 416–417
Electrical-resistance technology, 68 scraping technique, 412, 414
Electrical resistivity tomography (ERT), 389 screening, 414, 415
EU Landfill Directive of 1999, 39 waste excavation, 419
EU’s Waste Framework Directive of 2008, 39 waste filling, 402–404
Evapotranspiration, 24–25, 112, 398 Flammability chart, 330–331
Exposed geomembrane cap (EGC) Food waste, 125–126
construction, 404–405 FOS. See Factor of safety (FOS)
LFG collection, 305–306 Fukuoka method, 316, 333–336
LLDPE textured, installation of, 68
operational and maintenance issues, 405
vs. traditional final cover system, 406 G
Extraction points, 23–24 Gas collection and control system (GCCS), 281
bucket auger rig, 283, 284
collection header, 285, 287
F condensate, management of, 24
Factor of safety (FOS) downward collection systems, 306–308
analysis, 271–273 extraction points, 23–24
computer programs, 272 flaring process, 287
conceptual liquids addition, 274, 275 gas well head and pertinent features,
effective stress, 269 285, 286
setback distance, 278 horizontal wells, 284, 285
shear strength, 269 impacts, 298
slip surface, 269 landfill design elements, 289, 290
Fermentable organic compounds, 138 LCRS integration
Field capacity, 98, 102, 103, 140, 224 downward collection systems, 307, 309
Final cover system, 24–25, 61. See also Final horizontal gas collector, 304, 305
landfill disposition manholes and pumping stations, 302, 303
Final landfill disposition plumbing systems, 301, 302
closure and post-closure process toe drain, 303, 304
barrier layer cover system, 398–400 leachate seeps, 251
capillary layer cover system, 398–400 operation and monitoring, 355–356
definitions, 397 slotted piping, 283–285
planning process, 400–402 surface trenches, 305–306
Index 463