PB-239 357
AN EVALUATION OF LANDFILL GAS MIGRATION
AND A PROTOTYPE GAS MIGRATION BARRIER
W ins ton-Salem Department of Public Works
Prepared for:
Environmental Protection Agency
1975
DISTRIBUTED BY:
National Technical Information Service
U. S. DEPARTMENT OF COMMERCE
-------
I. Title anJ Silt-title
BIULIOCRAPHIC DATA
SNCCT
I. S «,'on N'u.
r.PA/530/SW-79d
PB 239 357
An Evaluation of Landfill Cos Migr.it.ion and
A Prototype Cas Migration Barrier
S. Krpufi lljiie
197*
f. Auihatd)
Cttv of '.'Ingtnn-salca. 'i.C. anil Enviro-F.ngtncers. Inc.
B. I'rilnnuiiift Ufpaniz^nun Kept.
No.
'. I'eifoiiiunji OIIEJIII/AIIUII N.imc anJ AUtlir«s
Dcpartnont of Public Works, City of Uinston-Salcm,
North Carolina, and Enviro-Engineors, Inc./Kngincering-
Science, Inc., Atlanta, corgia
10. I>i0|ret/lasli link Unit No.
ni No.
S-801519
12. Sponsotmg Orf.ani/.itu \loiJfc and Document Anal)si*. 17o. Uesciiptuik
Landfill, Gas Migration, gas Barriers
17k. Idrntificfs/Opcn-Kndcd Tcims
Gas
17c. COSATI Field T.ioup
»!OS SUBJECT n CHANGE
IB. Availability Statement
19. Srcutiiy C'.Uis (Ihis
Repim)
tl.Vf I.ASSIHI.P
2O. Sri in ii / (
I'jfir
UNf I
(I lii»
21. No. of
22. I'nce
l«tv. IO-TII kNUUICSKI) HV ANSI ANI> UNKS.
11II& HIRM MAY HI. HI l-KIIUUt.kl>
u*co
kltC •»•»-•'»•
-------
NOTICE
THIS DOCUMENT HAS BEEN REPRODUCED FROM THE
BEST COPY FURNISHED US BY THE SPONSORING
AGENCY. ALTHOUGH IT IS RECOGNIZED THAT CER-
TAIN PORTIONS ARE ILLEGIBLE, IT IS BEING RE-
LEASED IN THE INTEREST OF MAKING AVAILABLE
AS MUCH INFORMATION AS POSSIBLE.
-------
AN EVALUATION OF LANDFILL GAS MIGRATION
AND A PROTOTYPE GAS MIGRATION CARRIER
final report (SU-?9d) describes uork performed
for fie Federal solid isa.de nznasenent programs
dencistration Grant project S-3D2&3P ixta written cv
DEPARTMENT OF PUBLIC WORKS, CITY OF 1JI1JSTON-SALEH, NORTH CAROLINA
and EHVI'RO-EHGIMEERS IHC. and, uith nitior exceptions,
is reproduced ae received fran the grantee
U.S. ENVIRONMENTAL PROTECTION AGENCY
1975
-------
This report as submitted by the grantee or contractor has not been
technically reviewed by the U.S. Environmental Protection Agency (EPA).
Publication does not signify that the contents necessarily reflect the
views and policies of EPA, nor does mention of comerclal products
constitute endorsement or recommendation for use by the U.S. Government.
An environmental protection publication (SW-79d) In the solid waste
management series.
11
-------
CONTENTS
Chapters Page
I Sunatuiry 1
II Introduction 3
III An Accoi-nt of tlic Evplosio.i in the. Winston-Sali-m 10
National Guard Armory
IV Rrvii-i: of G.-is> Ciiu-r.itIon a-»d Kovvr.ient From Solid Waste 23
Ln:.d fills
V Star i—of-thc-Art ir. Gas Control Technology 30
VI Selection, Design ind Ir.st.il lation of I'rotoLypc- Gas AO
Barrier
VII bcscripti!" of Prototype Has K.-irrier 75
V11I P=rfonnajice Evaluation of Prototype Gas Barrier 87
IX Tocluiiqiu-s for r.vjluntin;; Solid Waste Landfills \.'ill>
Rer,'ir
-------
FIGURES (continued)
Mo. Page
1 Final Topofiraphy of Che Lint: Road landfill Site, 1972 14
4 Cross-sccHons Through Completed Link Road Landfill, 1973 16
5 Initial Cas .VonUorlns Stations, 1970-1971 19
6 Combustible Cas Concentration Distribution in Link Road 45
Disposal Kite 1973
7 Soil-Air Prt-ssutc Gradient Toward Amory Prior to Barrier 46
Operation
8 Gen. rail *eJ Corimst iMe Gas ConrentraLlons at Several 47
Stations Near Arnory
9 Soil Pcrrk-»!»j]liy S.-iqdcs anil Ass oc In ted l'»obc Locations fit
in Vicinity of Proposed Barrier
10 His torlf.il Variations of Coniust 5blc Cas Concent ration nt 70
Station 7
11 Historical and Projectt-J Concentration of CoiAx. :tlMe 71
G;JS Migraine From Disposal Area at Station 7
JZ Location of Carrier fells. Header, Pu/.y and Vent 76
13 Typical Cas Micratlon Barrier Veil 77
14 Sampling Locations for OeiceLlon anil Alarn SystcD 80
IS Probe Locations Tor Cvis Barrier Evaluation 90
16 Prossuro Pro'ui; and Manometer Detail 92
17 Average Concentration of Corbustlblc- Cas on Kiilier Side 96
of Car. Ilii;rdttoii Darrirr Trier to and During Operational
Hodos
Iv
-------
FICUKr.S (continued)
No. rape
18 Percentage of Lovi-r Explosive Llnlt al lndividu.il 98
Probes in and !.Var Buildinc*: 1'rior to and During
Barrier Opcration.il Modes
19 Variations of Baronotric Presr.urc During tlic Barrier 102
Test Period
20 Drawi'o ,-n Curves Between Barrier Wells 5 and 6 During 10-i
Barrier Test Period
21 Relative Pressure at Two D> pths Midpoint Between 105
Barrier Wells 5 and 6 During Barrier Test Period
22 Relative Pressure at Various Distances Fron Barrier 106
During the Test Period
23 Fre ;ucncy Analysis of t.'oisc Produced by Vai nun Pututt 109
24 Freqi. ncy Analyst of Background Noise Measured Three 110
Feet from Vermin Purp Gearbox
25 Comparison of Vin.uuin Pur.ip Noise Levels Measured at 10 111
Feel to the Background Noise Levels Measured During
Normal Traffic Conditions
List of Tables
No.
1 Estimated Costs of Alternative Gas Barrier Systems 50
2 Relative Effectiveness of Alternative Cas Barrier Systems 52
3 Relative Maintainability of Alternative Gas Barrier Systems 52
4 Relative Controllability of Alternative Gas Barrier Systems 53
5 Relative Environmental Impact of Alternative Cas Barrier 55
Systems
6 Relative Disruption During Construction of Alternative 55
Cas Barrier Systems
-------
TABLES (continued)
No.. Page
7 Relative Novelty of Alternative Gas Barrier Systems 56
8 Summary of Relative Scores for Alternative Gas Barrier 58
Systems
9 Particle-Size Distribution and Permeability of Soils in 65
Vicinity of Proposed Barrier
10 Cas Chromatographic Analysis of Barrier Pump Discharge Gases 99
11 Cas Chromatographic Analysis of Samples from Selected Probe 100
Locations
12 Statistical Data on Comparative Effectiveness of Soil-Air 107
Pressure Reduction at Various Distances from Barrier
Under Two Operational Modes
13 Power Requirements and Costs for Gas Migration Barrier 114
System
14 Comparison of Actual and Estimated Construction Costs of 115
Gas Barrier System
vl
-------
ACKNOWLEDGMENTS
The completion of this project In essential fulfillment of most of
tbe objectives originally conceived, despite the limitations which ensued
(namely, the impractlcallty of obtaining definitive data adequate for
comparative analysis at the sites available and the unforeseen modifica-
tion of the Link Road landfill) Is due In no small measure to the close
liaison and assistance provided by EPA project representatives Keith
Hartley and Truett DeCeare. They redirec. d the project to changing con-
ditions and barrier design aspects to permit the development, design,
installation and testing of a prototype gas barrier system.
The Winston-Salem project staff was most diligent in fulfilling its
responsibilities, ana strong support and invaluable assistance were pro-
vided by Joe H. Berrlcr, Director of Public Works, Pat Svann, Assistant
Director of Public Works, and Robert H. Davis, Assistant Superintendent
of Sanitation Division. A special note of conmendation is due Frank
Styers, whose intimate knowledge and understanding of gas-barrier prin-
ciples and his familiarity with the Link Road, site resulted in high-
quality construction and performance, of the gas barrier-.
Acknowledgment Is also due representatives of the North Carolina
National Guard for their close cooperation, particularly Hajor James N.
Stoneman.
Key personnel from b'nviro-Engineers, Inc., an affiliate of Engineering-
Science, Inc., who participated In this project were M. DeVon Bogue. Pro-
gram Manager, and Dr. Rahman Sheikh. The overall program was conducted
under the technical direction of Dr. Donald L. Fcuerstein.
vll
-------
CHAPTER I
SUMMARY
The Department of Public Works, City of Uinston-Salea, North
Carolina, assisted by Enviro-Engineers, Inc., cordueted a study of gas
hazards associated with the Link Road landfill and nearby structures,
particularly the North Carolina National Guard Armory, f«J» August 1972
through June 1974. The study Included a literature review of landfill
gas generation and movement and of the s:ate-of-tho-art of landfill gas
control technology; a review of the explosion attributed to ignition of
nigrat.ng landfill gas in the Armory on 27 September 1969; the design.
Installation and evaluation of a unique type of active gas barrier between
the landfill and the Armory; and the development of guideline considera-
tions for gas barrier design and detection of landfill gas migration:
hazards.
The review of literature- on landfill f.-\. generation and movement
Indicated that information on this subject in sparse. The state-of-the-
art of landfill gas control technology is limited In scope and applica-
tion, consisting primarily of high-capacity active systems, installed
after landfills were convicted and developed without specific considera-
tion of gas flow principles or conditions at the site of barrier instal-
lation. The review of the Armory explosion did not delineate any likely
causes other than migration! accumulation and Ignition in the Armory
rupp]y room of gases from the Link Road landfill.
The gas barrier system, designed, installed and evaluated at the
Lluk Road site, was developed on a rational basis, considering the
anticipated generation of gases In the landfill, gas permeability of
-------
the adjacent soils, gas flow principles and sinilar factors. The system
Incorporates a« series of gravel-filled veils, each containing a perforated
pipe and valve assembly. A header connects all veils to a snail-capacity
vacuua pump and vent assembly. The system is monitored by a gas analyzer
console in the Armory, connected to a aeries of gas probes in the soil
between the barrier and the Armor/ and in interior spaces, along with
audible and visual alarms to indicate gas concentrations exceeding 25
percent of the lower explosive Unit.
Testing of the installed system Included measurements of gas concen-
trations on both sides of the barrier and the measurement of pressure
profiles along and transverse to a section of the barrier. Coob us tittle
gas concentrations were reduced on the Armory side of the barrier to
sero or near-zero levels wlttiin several veeks after pumping began, and
have remained at these levels through June 197A at minimal pumping rates.
Ubiquitous negative pressure levels were established in the soJl along
and near the barrier shortly after pumping be Ran and have been maintained
throughout the test period. It would appear that the efficacy of the
barrier has been demonstrated.
Although the program did not permit refined or detailed guidelines
for the detection of gas hazards by simplistic procedures perforoed by
landfill management personnel, general comments were derived based on
project experience which may be helpful.
-------
CHAPTER II
INTRODUCTION
PROJECT RACKCROt'IJD
The degradation of burled solid wastes under anaerobic conditions
results In the generation of nethone gas which nay rlgrate through the
soil Into areas peripheral to the landfill and accumulate In explosive
concentrations. Methane is odorless and cannot be rc.idLly detected with-
out special instruments; consequently, neons are needed not only to de-
termine the node and extent of migration of landfill gas but also to
provide systems to protect buildings and other structures fron gas entry
and accumulation. This report provides the results of a project to In-
vestigate an explosion attributed to accumulation . nd ignition of land-
fill gas at the National Guard Armory in Winston-Salon, North Carolina;
to investigate gas migration characteristics; to design, install and
evaluate a prototype gas barrier; and to provide criteria and <*tsign
guidelines for gas barriers. The following sections describe background
clencnts and project objectives.
Armory Lxplosi on
tt» 27 September 1969, at about 8:40 a.m., a flash fire or explosion
occurred in the nupply room of the North Carolina National Guard Armory
in Uinaton-Salcm, North Carolina. Twenty-five guardsmen were injured
and three died as a result of the accident.
The nearby Link Road solid waste landfill was considered to he the
source of explosive- gases, which were thought to have migrated through
the Boll and into the building. An account of the explosion, including
a review of possible contributing factors, is presented in Chapter III.
3
-------
Early Investigations
Following the explosion, a number of investigations were Bade in
an effort to determine the cause of the incident, the potential for a
recurrence* whether explosive gases were migrating from the landfill
into neighboring areas through the coil and to evaluate the factors de-
termining the migration and accumulation of gases from the landfill.
Those investigations relating to the explosion are detailed in Chapter
111.
The Winston-Salem Department of Public Iforks then sought the assist-
ance of Enviro-Engineers, inc. (EE), who had performed investigations of
migration of feas from landfills trv the L«* tofceUs,, California area, to
define and Initiate a gas monitoring prog ran at the Link Road landfill
for the purpose of determining combustible gas concentrations and migra-
tion potentials.
Twenty gaa sampling probe locations, with probes located at l-» 10-,
25-, and 40-foot depths, were established around the northwest periphery
of the landfill and around the Armory. The probes, patterned after
those used by EE in the Los Angeles studies, were attached to plastic
tubes leading to ground level where they could be connected tP portable
gaa analyzers Cor direct reading of combustible gas in percent of the
lower explosive limit or in percent by volume. Gas samples were also taken
for gas chcomatographlc analysis. Gas concentrations were determined at
those probes nearest the Armory once daily and at the remaining probes
once weekly. (Tlicso Initial monitoring probe locations ace shown in
Figure 5, Chapter III. In later investigations, additional probes ware
installed. Some were destroyed and others were not used in some of the
Investigations.)
Combustible gas ton cert rat ions in airi near the landfill were at high
levels throughout the surveillance period, and relatively high levels of
combustible gas were noted in those probes located on the side of the
Armory nearest the main noay of the landlill. The probes nearest the
supply room showed consistently high levels of combustible gas at all
-------
depths, and probes beneach the building slabs in the supply and assembly
rooms showed combustible gas concentrations well above the lower explosive
limit most of the time.
It was generally concluded that conditions existed wherein combus-
tible gas (primarily methane) was being generated in the landfill in
considerable quantity, that it was migrating through the soil to and
beneath the Armory and that concentrations above the lower explosive
limit could accumulate beneath or in the building, posing a likely hazard
of recurrence of explosions if the Armory were reoccupled. (The Armory
was vacated shortly after the explosion.) It was further concluded that
safe beneficial occupancy and use of the Armory could not be resumed un-
less combustible gas concentrations near and beneath the building could
be brought, and maintained, well below the lower explosive limit. The
extent and results of early investigations are included in Chapter III.
Project Development and Funding
The Winbton-Salen Department of Public Works, after discussions
with U.S. Environmental Protection Agency (EPA) Solid Waste Program
representatives, submitted, in January 1970, an application for a demon-
stration grant to conduct a study and demonstration related to the
Armory explosion and to design and evaluate a gas barrier at the Link
Road landfill.
After several revisions of the original proposal, which resulted in
a considerable expansion of the total project, a grant award was made on
26 June 1972, and the project was initiated by EE to provide engineering
consultation, design and evaluation functions. The project is more fully
described in the following section and project findings are embodied in
the succeeding chapters.
PROJECT OBJECTIVES
The project, as finally developed and funded, included four main
objectives: (1) investigations pertaining to a review and report of
the Armory explosion; (2) a review of the state-of-the-art of gas migra-
tion and gas barriers, a review of gas barrier technology and designs,
5
-------
the design of a gas barrier for installation at the Link Road site and
its evaluation; (3) investigations of factors Influencing gas generation
and migration from landfills and development of procedures for detecting
and evaluating gas migration in terms of management guidelines for land
operators; and (4) establishment of design criteria for gas barriers.
A number of factors, including placement of new soil cover on the
Link Road landfill, resulted in only partial fulfillment of a number of
the tasks and in their termination, with EPA approval. It was not pos-
sible to accurately define the relationship of settlement, cell structure
and cover thickness, climate and other factors with gas generation and
migration; to evaluate effects of gas on vegetation as an indicator of
gas potentials; or to establish a meaningful "checklist" technology for
landfill operators. It appears that achievement of these objectives
could best be fulfilled at a landfill where the various parameters could
be more accurately measured rather than at a completed landfill. The
barrier design and evaluation was reoriented, with EPA approval, toward
the design and testing of a low-flow barrier system rather than the com-
bined high-flow and gravel-trench system originally conceived. This re-
vision has resulted in the development, evaluation and demonstration of
new technology in gas barriers.
The project objectives listed below are those pursued in the final
project to completion or to a point where meaningful information or
observations were possible.
Investigation and Report of Armory Explosion
This objective included the accumulation and assembly of all known
data, facts, accounts and reports on the occurrence and aftermath of
the explosion; assembly and review of all investigations and reports of
conditions at the site before and after the explosion; determination and
description of the development and operation of the landfill; a review
of similar incidents elsewhere; an investigation of the landfill site
and the Armory to document features which might have contributed to the
-------
migration of gases and their accumulation and ignition in the Armory; a
determination, if possible, of the cause of the explosion; and prepara-
tion of a full account of the explosion, investigations and conclusions,
if any. Chapter III presents a complete account of the fulfillment of
this objective.
Evaluation of Factors Related to Gas Generation and Migration
This objective included a review of the technical literature on gas
generation and movement to establish those parameters which should be
investigated and which would be most applicable within a checklist meth-
odology for landfill operators in assessing gas migration and hazards.
Other activities included the selection and characterization of appro-
priate test landfills; the design ard installation of monitoring systems
for combustible gas concentrations, settlement, vegetation effects and
climatological factors; collection and analysis of data; development of
gas monitoring techniques; and development of a checklist and rating
system. These efforts were only partially successful, and effort in
achieving this objective was reduced with EPA approval. Consequently,
the gas monitoring techniques and the checklist procedure are of a very
preliminary nature. Findings are presented in Chapter IV, and recom-
mended techniques are given in Chapter IX.
Review of State-of-the-Art in Gas Barrier Design
The objective included a review and summary of the technical litera-
ture and analysis of the design aspects, functioning and application of
existing or new technology to the conditions of the Link Road landfill.
The review included both active and passive systems and systems incor-
porating features of both, plus assessment or development of new tech-
nology, including high-flow and low-flow systems. A summary and analysis
of the state of the art is provided in Chapter V.
Data Accumulation and Investigations Related to Barrier Design
The task required collection of data for design purposes, including
gas production potentials, soil characteristics and gas permeability.
-------
area to be protected, cost, technological and construction considerations,
and monitoring systems. Findings are presented in Chapter VI.
Selection and Design of Cas Barrier
The objective required selection of the most appropriate gas barrier and
preparation of detailed design drawings and specifications, including
cost estimates and bid document information. The details of fulfillment
of this objective are presented in Chapter VI.
Cas Barrier Installation
The objective included selection of a contractor, equipment pro-
curement and construction supervision. Details are included in Chapters
VI and VII.
Monitoring Program
Throughout the project, gas concentrations were monitored at the Link
Road landfill and at two other landfills (not reported herein) in accor-
dance with the needs of the various objectives. Monitoring included the
installation of gas probes, the periodic determination of gas concentra-
tions using a portable combustible gas detector and analysis of samples
by gas chromatography. In addition, records were collated on rainfall and
other climatologieal data. Monitoring was done on a relatively continuous
basis at the Link Road landfill and was revised (following selection of
the barrier design) to provide base-line data and performance-evaluation
data for the barrier. Monitoring programs arc described in Chapter III
and in Chapters VI through IX.
Cas Barrier Evaluation
Prior to installation of the gas barrier, gas probe layout and in-
stallation was devised, along with a detailed program for determining
barrier performance. The evaluation included monitoring of gas flows In
various parts of the system, determination of pressures in the soils
adjacent to the barrier and determination of gas concentrations on both
aides of the barrier and inside the Armory by means of portable gas
analyzers, gas chromatography and the installed automatic monitoring
systems. Evaluation of gas barrier performance Is given in Chapter VIII.
8
-------
MANAGEMENT OF THE STUDY
A work plan was developed wherein the accomplishment of the various
tasks required to fulfill each objective was subdivided into subtasks,
and responsibilities were assigned to the Vinston-Salem project staff
and to EE. For the most part, data collection was perfonrcd by the
Winston-Salem project staff in accordance with EE recommendations.
Data analysis was performed primarily by EE. Technology reviews, barrier
selection and design were an EE responsibility, along with development
of barrier evaluation programs. Barrier installation and developnent of
barrier performance data were Winston-Salem responsibilities, whereas
analysis of performance data and development of design guidelines were
done by EE. The account of the Armory explosion was largely a Vinston-
Salem staff effort with EE's major input being directed toward interpre-
tation of documented investigations and specific further investigations
to complete a technical review of the factors possibly related to the
incident. EE had the major responsibility for developing and preparing
the technical report.
-------
CHAPTER III
AN ACCOUNT OF THE EXPLOSION IN THE WINSTON-SALEM
NATIONAL GUARD ARMORY
ACCOUNT
On 27 Septenber 1969. at approximately 8:40 a.m., a flash fire or
explosion occurred in the supply room of the North Carolina National
Guard Armory at 2000 Silas Creek Parkway (formerly Link Road) in Kinston-
Salem, North Carolina. Of approximately 25 Guardsoen injured in and
near the Supply Room, 1Z were seriously injured, seven of vhom are now
partially or totally disabled. Three Guardsmen died as a result of burns
or subsequent complications.
The National Guard Armory was built in 1962 adjacent to the Citv's
operating solid waste landfill, whose continued operation was a condition
of the deed. In addition to the Armory, the National Guard utilized ad-
joining landfilled property for parking and servicing a large fleet of
supply vehicles, including large-capacity fuel tankers. Other National
Guard structures included a garage/repair building, a small shed and an
underground fuel-storage tank in the parking area, and a prefabricated
steel storage building constructed on a slab placed over a landfilled area
to the rear of the Armory. The \rmory was built on a six-inch reinforced
concrete slab and, except for the central drill-hall area, is of one-story
conerote-block construction. A two-story addition to the building,
»>uilt by the City in 1965, housed Winston-Salem's Police Academy. The
slab was poured over four inches of crushed stone. The site was prepared
by leveling existing topography. No portion of the building extends over
10
-------
a landflllcd area and closest proximity to the landfill (about 30 feet)
is at the rear of the building. Prior to placing foundations, it wan
found that soil conditions at the rear of the buildings did not provide
adequate bearing capacity; consequently, spread footings were used to a
depth of six feet along the rear of the building. No nechanized ventila-
tion system was provided for the main building.
Other structures nearby on land of or adjacent to the Link Road land-
fill Include an abandoned abattoir (now used for fire training purposes),
a fire training tower and a peripheral road constructed around the fire
training ai°a, developed by the City for Fire Department use. The -fire
training tower is built on a landfilled area as is the peripheral road.
Fire hydrants are supplied by a looped eight-inch water main crossing
beneath the vehicle storage area from Silas Creek Parkway.
Utility structures include water and electrical services to the
Armory, a sanitary sewer serving the garage and Armory, building sever
and drains and a 72-inch storm drain constructed through the landfill
and discharging to Salem Creek. The stona drain was emplaced during land-
fill construction. Street drains from the fire training area, the en-
trance ro£.d and paved areas enter the storm drain.
Figure 1 provides general details on the Armory, the nearby landfill
and associated facilities. Figure 2 shows the boundaries of the landfill
and structure and Figure 3 shows final landfill topography (prior to place-
ment of additional cover in July 1973).
The Link Road landfill was initiated in 1949 when the City's onsite
incinerator was closed and was near completion in 1969 when the Armory
explosion occurred. The present vehicle storage area is placed over the
area formerly used to landfill incinerator residue. Dead animals and
wastes from the abattoir were disposed in trenches close by. No other
segregation of wastes was practiced (except surface composting of leaves
and vegetative trimmings) and wastes deposited throughout the remainder
of the fill include intermixed domestic, industrial and commercial wastes.
In general, wastes were placed in an organized fashion, compacted to the
11
-------
SANITARY SEWER
V
SILAS
CREEK
PARKWAY
STORM DRAIN
SANITARY
SEWER -
PAVED
AREA
\
\Y
.. i_)
VEHICLE
METAL PARKING AREA
0SHEO (GRAVEL)
M J* H M M M M_M<
a - v»
TOWER/ ( ^~ d&i
i ^ V.*.>-O=-/J>
ARMORY
SUPPLY
ROOM—»T
^PAVING J
/^^ar^llT^
11\ rr^A
VEHICLE
PARKING
(GRAVEL)/
SANITARY
SEWER .
N
*
72"DRAIN
---- ~'*
CONCRETE SLAB
SEWER MANHOLE
VENT ASSEMBLY
VEGETATION
LANDFILL
A
LANDFILL
VEGETATION
\
TO SALEM CREEK
\
LEGEND
— APPROXIMATE EDGE
OF LANDFILL
PONDED WATER
AREA OF MAJOR
SUBSIDENCE
Figure I: National Guard Armory and associated structures, 1972
-------
SILAS CREEK PARKWAY
PARKING
oSHEO
LOCATION OF \
FORMER I
SEWAGE /
TREATMENT J
PLANT
x
„. FORMER
CONCRETE
SLAB
INCINERATOR
LINK
ROAD
LANDFILL
EXISTING STRUCTURES
APPROXIMATE LANDFILL
BOUNDARY
Figurt 2: Link Rood landfill and associated structures
-------
APPROXIMATE LIMIT-
OF LAND FILLED
WASTES
nr,7/111 "?•'•'•
LEGEND
STRUCTURE
PAVING
GRAVEL SURFACE
0 100 200 400
SCALE
Figure 3' Final topography of tht Link Road landfill sift, 1972
-------
best of ability of equipment provided and covered with compacted soil
cover, If not daily, at frequent Intervals. Operation of the site,
while perhaps not In strict accordance with high standards set for sani-
tary landfills today, was equivalent to or better than operation of its
contemporaries. Depth of the landfill ranges from 10 to nearly 40 feet
in places. Inmedlately back of the Arnory. wastes were packed along a
steep natural bank (extreme settlement occurred and Is continuing to
occur at this location). The landfill is reported to have incorporated
a high ratio of soil cover to wastes deposited. Most of the bulk of the
landfill Is saturated from groundwater and surface water. Total quantity
of wastes deposited is estimated at 1.4 oillion cubic yards. Figure 4
shows typical cross sections through the landfill.
In the summer of 1965, a welder installing part of the storm drainage
system near the Armory received minor burns in a flesh fire. In November
of the sane year, a fireman working near one of the street drains In the
fire training area dropped a lighted match into a manhole and received
minor burns in the resulting minor explosion. In December 1965, the Guard's
Executive Officer reported a flash fire while welding downspouting on the
Armory's roof drain system wnlch he was connecting, to an underground
drainage system extending into the filled area back of the Armory. The
Fire Department determined presence of combustible r.as in the abattoir
but none in the Armory. In 1966, representatives of the Federal Solid
Waste Program investigated conditions in the fire training area and con-
firmed methane in the storm drains. In July 1966, following complaints
of strong odors froei the landfill, a blower was installed in a brick
building, surmounted by a tall vent stack, over a large manhole in the
stora drain to the rear of the Armory. This ventinp, system was operated
intermittently thereafter and was apparently affective in dispersal of gas
and odors. During these early episodes, investigations centered around
accumulation of combustible gases in sewers, a condition not uncommon to
sewers, though more common to sanitary sewers than to storm drains.
No detailed investigations were made to determine occurrence or extent of
gas accumulation in the Arnory.
IS
-------
1200
Figure 4: Cross-sections through completed Link Rood
landfill, 1972
16
-------
Only a week or so before the explosion occurred, three to four feet
of additional cover had been compacted over the landfill Just oack of the
Armory, where extensive settlement had occurred. The Guard subsequently
dismantled and removed the prefabricated building nearby, due to extensive
deformation of the building slab caused by settlement of the landfill in
this area.
On 26 September 1969, the day before the tragedy, City representatives
not with Guard represpntatives to investigate the occurrence of "odors
or gas" from the arms storage vault off the Armory supply room. No source
of gas, defective systems or other causes were noted. Arrangements were
cade to have the Fire Department check out the vault (after being closed
several days) with portable gas detection equipment the following week.
Since the vault would be opened during drills, the Guard proceeded with
scheduled drills. The explosion occurred the following morning.
Immediately following the explosion, it was conjectured by some that
the "blue flame" explosion or fire was caused by ignition of gas migrating
from the landfill and accumulating in the vault or supply room. Investiga-
tions did not reveal presence of petroleum products or flammable materials
in the supply room or vault, nor was an ignition source definitely estab-
lished. It is presumed that one of the Guardsmen lit a match, despite
posted "No Suoking" regulations.
The explosion lifted the roof of the supply room at least-six inches
at the southeast corner. Windows were broken, but it was not known vhether
they were broken by the occupants or from the force of the explosion. The
rear wall and celling were blackened and scorched, but most of the stored
materials (such as blankets) were undamaged. A one-half inch horizontal
separation between the rear wall and the floor slab was noted following
the explosion.
Combustible gas sampling in the supply room, drill hall and pistol
range shortly after the explosion, using a portable combustible gas indi-
cator, revealed combustible gas concentrations near or above the lower
17
-------
explosive limit (LEL) only at cracks in the slab. In soils outside the
Armory* readings were above the LEL. It was thought that soil saturation
by recent heavy rains prevented landfill gas from venting through the
landfill cover, thereby increasing lateral gas migration while lack of
ventilation in the building may have contributed to gas accumulation in
the supply room. Subsequent readings on 6-9 October 1969 were inconclusive,
but seemed to indicate a significantly higher concentration in the vault
than in the supply room (but no concentrations were noted above the LEL).
Gas sampling was done in nearby sanitary sewers and storm drains on
9-10 October and 21 November 1969. whereas high combustible gas concentra-
tions were noted in manholes near and behind the Armory in the early tests
and petroleum vapor was reported in two of the manholes, there was no
conclusive evidence of gasoline or other petroleum vapors entering the
Armory from the sewers, and all combustible gas concentrations had dimin-
ished by late November 1969.
A network of 20 holes was drilled 30- to 40-feet deep around the
Armory in 1970, and gas sampling probes were inserted at 1-, 10-, 25-
and 40-foot depths. The probes, whose locations are shown on Figure 5,
were connected to plastic tubes leading to the surface from which com-
bustible gas concentrations were determined periodically with a portable
combustible gas detector. Subslab sampling points were established by
drilling holes through the floor slab in the drill hall and supply room.
Combustible gas concentrations in the range of 30 to 40 percent by volume
were recorded in the exterior probes over a period of many months. Con-
centrations in excess of the LEL were consistently recorded below the
building slab. This sampling program verified that combustible gas from
the Link Road landfill was migrating to and beneath the Armory, presenting
a continuing hazard of gas accumulation and explosion. Concern about
this situation prompted interest in developing and demonstrating a gas
barrier between the landfill and the Armory.
*For methane, the LEL is four to five percent by volume in air.
18
-------
H M M M I
'16
15 J
il >ft.M * K i
18 .17
' H °lf M H M MJ»<
r~z
.
•10
LEGEND
• SAMPLING STATION
I, 2 PROBE NUMBER
Figure 5 • Initial gas monitoring stations, 1970-1971
-------
The Armory was vacated shortly after the explosion and since has
remained vacant. The landfill was completed in March 1970 but continued
use has been made of the fire training center and the Guard's vehicle
storage and maintenance area. Wild vegetation has grown over most of the
landfill surface. Except for periodic monitoring of combustible gas coi -
centrations, little activity was conducted on site until August 1972,
when Investigations began related to studies of gas migration and develop-
ment of a gas barrier. A portion of these investigations was directed
toward a review of the Armory explosion and to possible causes.
FINDINGS
During the placement of probes in the landfill for combustible gas
concentration evaluations in 1973, it was found that high moisture con-
ditions occur throughout the landfill. Combustible gas measurements
indicate that the gas migrates tnrough the soil a considerable distance
from the landfill boundary and occurs beneath the Armory in explosive con-
centrations. Investigations indicated that the soil cover, though variable
in consistency and thickness, is relatively porous, but that saturation
of the surface layers by rainfall may increase lateral migration. The
landfill is continuing to settle unevenly, particularly in a large area
immediately back of the Armory and in an area on the east side of the
fire training area (see Figure 1). In the summer of 1973, three to four
feet of new soil cover was placed over the upper portion of the Armory.
This new soil cover places additional weight on this portion (which con-
tinues to subside) and may further reduce venting of gas through the
cover or intensify local gas migration through adjoining soils.
Analysis of soils along the line of the barrier indicated considerable
variation in the soil's gas permeability along the landfill periphery,
with the greatest permeability being noted back of the Armory. In early
tests of the gas barrier in 1974, rapid gas flow was noted into three
of ehe barrier's suctirn wells, located about midway in the vehicle storage
area. Indicating local areas of high gas permeability in the nearby soils.
Porous incinerator residue underlies this area.
20
-------
Gas flow toward the Armory appears to be greatest immediately to the
rear of the Armory, where the land::! 11 periphery is closest and where major
settlement is continuing. Significant combustible gas concentrations also
occur on the vest side cf the Armory, indicative either of relatively long-
range but attenuated gas migration from the main body of the landfill oz
of moderate gas flow from wastes landfilled in the incinerator-residue area.
In 1973, a ditch was dug around the southeast corner of the Armory to
a depth of about eight feet in order to determine if direct channels for
gas migration existed between the landfill and the Armory in this area.
Examination indicated that several types of relatively coarse, sandy soils
were intermixed, with discrete lenses of sandy-micaceous material ocurring
throughout. No direct conduit or obvious avenue for ready migration of
gases was noted. Tests on the soil did not indicate presence of petro-
leum products. At the same time, field test? were made which indicated
that building drains were functioning properly. It was noted that tile
drains, formerly connected to roof-gutter downspouts, had openings near
the building which would permit venting of any gases entering the drains
before they reached the building. There was no connection between roof
drains and the Armory interior.
Inquiries to the National Guard regarding possible spillage of gaso-
line into sewers on nearby soils, with likely entry into the Armory, did
not substantiate this possibility.
In summary, the exact cause of explosive gas accumulation in the
supply room or vault and its subsequent Ignition has not been determined.
It does not appear that utility gas or petroleum-vapor sources were likely.
No explosive gas, other than combustible gas migrating from the landfill,
vas discerned and the Armory was reasonably veil-protected from accumula-
tion of gases from normal sources. Combustible gas was likely migrating
from the landfill through adjoining soil to a-.id beneath the Amory prior
to the explosion, but its presence or hazard potential had not been con-
firmed. Early assumptions that combustible gas concentrations were con-
fined to sewers were reasonably valid, in the light of extant knowledge
and-practice. Although operation and construction of the Link Road land-
fill resulted in conditions contributing to extensive gas production and
21
-------
migration, Che nature and potential hazards of gas migration were not
then well known. The landfill was operated in a manner common to landfill
practice prevailing at that time.
The Armory explosion has highlighted (unfortunately with tragic
aftermath) the extreme importance of recognizing the hazardous explosion
potentials from landfill gas migration possibly resulting from operation
of solid waste land disposal sites. Measures to -minimize gas production
and migration potentials during and after landfill operation are in order.
along with the protection of structures nearby from gas accumulation by
structural features and by active or passive gas barriers. Such measures
are included in modern guidelines for sanitary landfill design and opera-
tion.
Litigation following the Armory explosion has apparently been settled.
Tests on the gas barrier have thus far indicated favorable results. Gas
concentrations between the barrier and the Armory have been reduced to
zero or near-zero levels by operation of the gas barrier. If, after a
suitable period of operation, the gas barrier remains as effective as
the preliminary results have indicated, and the barrier system is propetu
operated, regularly monitored and constantly maintained at speci'fied *
operating levels, normal use and occupancy of the Armory should be
possible.
22
-------
CHAPTER IV
REVIEW OF GAS GENERATION AND MOVEMENT FROM SOLID WASTE LANDFILLS
LITERATURE REVIEW
The paucity of information on gas generation and movement from
solid waste landfills is attested to by only brief mention In a principal
solid waste disposal reference (Reference 1) and the listing of only 48
articles in the most recent bibliography on the subject (Reference 2).
Many of the articles are of a nontechnical nature and, in many cases.
discussions of gas generation and movement are only a small part of the
article listed. Only a few are based on scientific measurements over a
significant period of time, and several of those containing specific data
are based on observation of simulated landfills, on test refuse cells or
on laboratory-scale experiments. The most informative of the early
articles, based on field studies, was that reporting studies related to
groundwater and to carbon dioxide production from landfills, conducted by
or for the California Department of Water Resources, the California
Water Pollution Control Board and the California State Water Quality Con-
trol Board. Studies of gas explosion episodes were limited in thorough-
ness or scientific approach; the most extensive report being of an
explosion in an abandoned power substation in Atlanta, Georgia, which was
being converted to a recreation facility adjacent to a landfill. An ex-
tensive study by Los Angeles County included a review of gas production
and movement at 10 landfills in the Los Angelas area and extensive lab-
oratory studies to determine diffusion coefficients for landfill gas
in a variety of soils (Reference 3). The study also presented design and
construction criteria for buildings on or near solid waste landfills.
23
-------
Studii-s of pl'ot landfills or refuse cells in southern California
(References 4, 5, 6 and 7) have contributed considerable information on
gas production and migration. Most of the reported work on gas genera-
tion and movement has been undertaken in recent years (i.e., since
1954); the earliest article was published in 1940.
The literature review was supplemented by field inspection and re-
view of information on existing solid waste landfills in the Los Angeles,
California area where a number of gas migration studies or observations
have been made and where gas barrier systems have been installed.
CAS GENERATION FROM SOLID WASTE LANDFILLS
In ttie early stages of solid waste decomposition in a landfill,
decomposition takes place through processes which utilize the initial
supply of oxygen entrained in the wastes. In the first six months to a
year, carbon dloxidn predominates. Principal gases produced are carbon
dioxide, hydrogen, nitrogen and methane. Carbon dioxide and methane
mak£ up over 90 percent of the total gas produced (Reference 8). Hydro-
gen sulfide iuay be produced in areas where salt water is in contact with
the refuse. As the entrained oxygen supply diminishes, anaerobic decom-
position ensues and an increasing amount of methane is produced while
other gas production diminishes. Significant methane production may
occur within six months to a year, and production may continue over long
periods of time. The production of gases depends on a number of factors,
including the amount of oxygen available, the organic content of the
solid wastes, particle size and degree of compaction and the amount of
moisture available. In general, high organic content and moisture will
increase gas production. Smaller particle size (by exposing more of the
refuse to bacterial action) may have a similar effect. There are some
Indications that densely compacted refuse will decompose at a slower rate
than loosely compacted refuse and that gas production may be prolonged
In densely packed landfills (Reference 7). Several studies of gas produc-
tion have indicated that from very moist or saturated "noraal" domestic
re'Juse, carbor, dioxide may be produced in quantities of about 0.2 cu ft/lb,
whereas from the same type of refuse which is relatively dry, the
24
-------
production is only about 0.04 cu fc/lb (Reference 9). In a sealed test
chamber of mixed refuse, 2,027 cu ft of decomposition gases were pro-
duced from 73 cu yd of refuse over 906 days, or about 28 cu ft/cu yd
(Reference 7). In a similar test, over a 200-day period, gas production
was approximately 28 cu ft/cu yd of refuse, or about 0.04 cu ft/lb
(Reference 7). Theoretically, a pound of refuse can produce 2.7 cu ft
of carbon dioxide and 3.9 cu ft of methane (Reference 10). How much of
this potential production will be realized in a particular landfill over
a given period of time will depend on a number of factors which aid or
binder gas production, and the variations in production will be largely
dependent on variations in these factors.
Anaerobic conditions may occur within a month after wastes are buried
(Reference 11). Dry landfills apparently produce much less gas than those
which are initially or subsequently saturated with moisture. Moisture
additions to a landfill can reinitiate or increase carbon dioxide and
methane production. Under such conditions, once initiated, methane pro-
duction may continue f«r a long period of time.
GAS MIGRATION FROM LANDFILLS
In studies of migration of decomposition gases from landfills
(References 4, 8, 12, 13 and 14), it has been noted that carbon dioxide
and methane migrate upward and downward as well as laterally through
surrounding soils. Most of the gas produced will migrate upward through
soil cover. Field observations have reported explosive concentrations
consistently as far as 600 feet from a landfill (Reference 15).
Field observations in a study of a landfill in southern California
(Reference 12) Indicated that carbon dioxide moved vertically through
the soil cover at more than 10 times the rate it moved downward or Later-
ally through adjacent soils. Other studies indicate that 18 to 24 times
the quantity of carbon dioxide and methane, respectively, pass through
the cover than into the surrounding soil (Reference 13). Carbon dioxide
movement rates ranged between 0.22 and 0.8 ft/day vertically and betvesn
0.24 ft/day and 1.4 ft/day horizontally (in undisturbed alluvial soils)
25
-------
in one study (Reference 9) to vertical and horizontal velocities of 0.22
ft/day to 0.24 ft/day in another (Reference 14).
Little is known about gas migration within a landfill. In landfills
where well-compacted waste and cover does not prevail or where settlement,
groundwater intrusion or fires have caused destruction of cell integrity,
gas will likely migrate through voids or channels throughout the nass,
perhaps towards a section of the landfill's periphery where localized
(rather than general) gas nigration may occur. Numerous instances have
been noted where methane or other gases issue from surface cracks in
landfills in large quantity and at considerable velocity. Maintenance
of a gas-permeable cover or other means of venting, would appear to be
most important in insuring that lateral landfill gas nigration is mini-
nized. Impermeable scaling of a landfill surface at any time will likely
cause a prompt increase In lateral gas migration (Reference 16). Even
If a landfill has been in place for many years and decomposition of
early-deposited refuse is in an advanced stage (with theoretically de-
clining gas production rates), gas migration controls should consider
characteristics of the more- recently deposited wastes, wherein gas pro-
duction and migration potentials may be highest. Critical areas of a
landfill in terms of gas migration would likely be those where the mois-
ture content is highest, where highly organic wastes were deposited
(these are often concentrated in one area of the landfill, commonly with
minimal compaction or cover) and areas where the landfill surface nay
be most impermeable.
In addition to the general gas migration characteristics of a land-
fill, it is apparent that local variations in gas production and migra-
tion occur, perhaps related to reduced permeability of soil covet due to
precipitation, settlement or other phenomena. Ho long-range correlation
of gas production and migration of landfill gases with precipitation has
beor. documented, but field observations have indicated a general increase
It. gas migration following extensive precipitation. Monitoring of gas
concentrations related to gas control barriers has indicated daily and
periodic fluctuations or "excursions" of gas concentrations both above
26
-------
and below the average. Consequently, gas migration is likely to be
extremely variable and control systems should be based on theoretical
maximum potentials for gas production and migration.
HAZARDS DUE TO LANDFILL GAS MIGRATION
The principal hazard from migrating landfill gas is likely to be
from methane. It has a relatively high production rate in comparison
with other gases, it has a .cndency to migrate laterally with greater
facility than other gases when venting is impeded and it has a long-term
production potential from landfills. Its explosive hazard and the in-
ability to detect the presence or concentration of methane without in-
struments further emphasize its prime importance among the migrating
landfill gases. The tendency of carbon dioxide to migrate downward to
groundwatcr and to acidify it increases the corrosivity of the water.
Methane, by contrast, is relatively insoluble in water. Maximum produc-
tion of carbon dioxide from a landfill is relatively short-lived, in com-
parison with methane production. Hydrogen sulfide, because of its noxious
odor, is relatively c&rily detected and is produced primarily in those
relatively few landfills in contact with sulphate-containing sea water.
Relatively few occasions of explosions or other tragic consequences
from migration of landfill gases have been reported, and only a couple
have been investigated closely. A survey of 2CS counties in 1965 Indi-
cated that only 21 percent had built on landfills. In these cases, aany
of the structures were plagued with settlement and gas problems (Reference
17). Fires in San Francisco's waterfront landfill were attributed to land-
fill gas (Reference 18). A fire in a manhole of a sewer built through a
landfill near Knoxville, Tennessee occurred in 1947, within two weeks
after the landfill was sealed with soil cover (Reference 16). At a large
landfill near Seattle, Washington, pipes inserted through the soil cover
were used to vent and burn off escaping gases and to partially eliminate
odors (References 19 and 20). In the New York City area, gas accumulations
In a sewer near a landfill were noted, and an explosion occurred in a small
building situated on another landfill. More recent explosions include
27
-------
one at a warming hut at an above-ground sanitary landfill converted to a
ski slope, an explosion in a former power substation being converted to
a recreational facility next to an active landfill in Atlanta, Georgia
(Reference 21). and the tragic explosion in the KMnstcn-Salem, North
Carolina National Guard Armory, built close to a landfill just being
completed. The latter two episodes resulted in five deaths and in in-
Jury to 12 building occupants. In the Atlanta incident, methane gas
accumulated in the building from landfilled refuse placed against and
under the building and was ignited hy a workman installing natural-gas
heating facilities. Ignition of leaking natural gas from the gas
facilities was ruled out as a cause of the explosion. In the Hinston-
Salem incident, early flash fires in sewer lines near the landfill had
been attributed to gas migration from th_ landfill, but gas migration
beneath the Armory had not been, suspected. Following the explosion,
presence of methane in the soil next to the building and beneath the
building's floor slab was detected in explosive concentration. These and
later investigations, reported in Chapter III of this report, failed to
Indicate a source of explosive gas other than combustible ga& from the
landfill. The influence of a large storm drain through the landfill,
either as a channel for combustible gases to the Armory or as a con-
tributor of water to the landfill, was not demonstrated, Monitoring of
gas probes around the site in recent yars indicates tne continued genera-
tion of combustible gas and its migration to the vicinity of the Armory
in explosive concentrations.
Host of the explosions have been attributed to accumulation and
ignition of methane. Perhaps as a result of these other hazardous inci-
dents, construction of buildings on or near landfills has apparently
dwindled or such development is done with appropriate precautions.
OBSERVATIONS OF GAS MIGRATION AT VINSTON-SALEM
Early investigators attributed the accumulation of combustible gas
in she Armory just prior to the explosion to an antecedent heavy preciplta-
tltitij which reduced venting through the landfill cover with & resultant
increased lateral migration* and to a lack of building ventilation. Gas
28
-------
permeability •• asurements of soil samples, taken relative to the gas
barrier recently constructed, indicate that the peripheral soil is gen-
erally permeable to gas flow with highest permeability in the soils near
the supply room, where the explosion occurred. Other possible influences
include extensive settlement near the landfill edge nearest the Armory
and subsequent additional placement of cover soil (shortly before the
explosion).
It was not possible to make field measurement of gas generation rates
or the relative effects of cover type, thickness or settlement on gas
flow. It was determined, however, that high moisture occurs throughout
the landfill and that it was actively producing combustible gas, which
migrates readily through the surrounding permeable soils.
A more detailed account of observations and findings is presented
in Chapter VIII.
29
-------
CHAPTER V
STATE-OF-THE-ART IN GAS CONTROL TECHNOLOGY
INTRODUCTION
Relatively little Investigative work has been performed and reported
in the literature concerning the generation and movement of decomposition
gases from sanitary landfills. Nevertheless, uncontrolled migration of
decoiaposltion gases from a landfill is recognized to be capable of posing
a great threat to its environs due to pollution of groundwaters, damage
to vegetation or the hazards of gas explosions. Only in recent years
have active measures been undertaken to investigate and provide means
for controlling the movement of decomposition gases from landfills. The
limited literature on the subject of gas control measures makes reference
to procedures which should or could be utilized to mitigate the problems
of gas migration, but xt contains very little reference to the actual
means for achieving the desired results. This is probably due in large
part to the experimental nature of the art of gas control technology.
Testing and evaluation of the various control methods are still in progress
in different parts of the country, and definitive conclusions are yet to
be formulated. In general, detailed information on design and operation
jf gas control systems is lacking in the literature, mainly because this
area is a recent addition in the field of landfill design and utilization.
GAS CONTROL SYSTEMS
This section contains information on gas control systems known to have
been installed, mainly in the western United States, either for experimental
purposes or in attempts to eliminate existing problems at particular sites.
Gas control systems can be categorized into two major subsystems—pre-
construction systems and post-construction systems. The pre-constmctlon
30
-------
systems consist of measures that arc incorporated into the design of the
landfill to mitigate the movement of gases from the landfill. Integrated
design measures taken to prevent the entry of gases into structures
constructed on or adjacent to landfills are also included under the heading
of pre-construction systems. Post-construction systems include all of
those facilities which are designed and constructed after completion of
the landfill or construction of structures or. or near the landfill.
Post-construction systems are developed as a solution for the problems
created by uncontrolled landfills. Host gas control systems in existence
today are of the latter type.
Pre-Construction Systems
Pre-construction systems are relatively recent innovations and their
development has come about as a result of the increased awareness and
knowledge on the hazards of landfill-generated gases. Various gas con-
trol methods nay be utilized in the pre-construction stage. These methods
are classified as "permeable methods" and "impermeable methods." Some
permeable methods include the use of gravel vents or gravel-filled trenches
to provide a more permeable path for the escape of- gases from the land-
fill to the atmosphere. In this fashion, decomposition gases are Inter-
cepted and prevented from migrating beyond the Units of the fill. Other
permeable methods consist of a system of gravel-filled trenches and/or
wells, with vent pipes and collecting laterals inserted in the trenches
and the wells for conveyance of gases out and away from the fill and ad-
jacent structures.
Impermeable methods are based on the use of natural or synthetic bar-
rier materials to prevent gases from migrating beyond the fill limits. A
venting system is usually utilized along with the physical barrier system
in order to prevent the buildup of positive gas pressure within the con-
fines of the fill. Such venting would particularly be required if the
surface of the landfill is later sealed by paving or by the use of some
other impermeable cover materi*?.. The barrier system may consist of clay.
synthetic membrane, concrete, asphalt or other bituminous material.
31
-------
It has.beep suggested (Reference 22} that the most conaon, and pos-
sibly the most practical, impermeable barrier method involves the use of
compacted clay, either as a liner in the bottom and sides of the fill or
as a curtain wall to block underground gas movement. It was concluded from
laboratory experiments (Reference 23), which were conducted to,determine
toe thane diffusion coefficients for various soils with differing particle
size under varying conditions of noisturr and pressures, that soils with
a high percentage of clay can act as a gas barrier under favorable soil
moisture conditions. However, the impermeable soil barrier method still
remains relatively untested. Some synthetic membranes have been utilized
in landfill construction, but these have been used primarily for the pur-
pose of controlling the flow of leachates.
The literature contains very little reference to pre-construction
gas control systems constructed in conjunction with sanitary landfills.
It is probable, however, that much of the work in this area is still in
progress and has not been publicized or reported. One example of such a
system (Reference 24) is a sanitary landfill located in Kansas City,
Kansas, which was initiated by the Kid-American Regional Council, Kansas
City, Missouri. The project is supported in part by a demonstration grant
from the Office of Solid Haste Management Programs, U.S. Environmental
Protection Agency. The completed site is to be utilized for a regional
park, which .will include several public buildings and recreational facili-
ties. Several homes are also located adjacent to the landfill site.
Becognizing the potential dangers to these structures from landfill gen-
erated combustible gas, a combination permeable and impermeable system
for gas venting is being constructed as part of the landfill. A soil
blanket under the fill, laid primarily to control downward flow of
leachates, also will provide a relatively impermeable barrier to the
downward movement of gases. To allow ventilation of gases from the fill,
a portion of the one-foot thick daily soil cover of each lift of the fill
at the site perimeter is being removed to allow gases to pass readily up-
ward through each lift. A trench two to three feec deep will t«» ex-
cavated and backfilled with gravel at tne top of the fill in the final
32
-------
earth cover around the site edge. A permeable channel is thereby pro-
vided through which landfill gases can travel readily and vent to the
atmosphere. The effectiveness of this method of gas control has not
been established.
More common pre-construction systems are those that are designed and
constructed for protection of structures which are erected on or adjacent
to a completed landfill. Numerous examples of these gas control systems
are in existence, although reported cases in the literature are rather
limited (Reference 25) and include examples of techniques used to protect
various structures such as school buildings, a market building and a
residential area. In most cases the control systems involve vent pipes
around the perimeter of the structure or within the building itself,
gravel trenches, impermeable barriers around the foundations and under
the floor slabs, or a combination of these.
Observations of some recent gas barrier installations in southern
California were obtained from interviews with designers, developers and
public agency personnel. Information from these interviews is presented
below.
For protection of a public market located on natural ground adjacent
to a landfill, a series of trenches were excavated in the subgrade under
the building and backfilled with gravel. The trenches were spaced approxi-
mately 12 to 15 feet on center and were 12 inches deep and 12 inches
wide. These trenches extend the full width of the building, terminating
at the foundation walls. Two-inch diameter vertical vent pipes were
placed in the walls of the building and connected to each gravel trench
at tne perimeter of the building. As an additional precaution, the entire
subgrade beneath the floor slab was covered with a protective impermeable
10-ttll polyethylene membrane which was laid on a two-inc* sand bed.
Venting of the trenches and subgrade area occurs by natural ventilation
only, and no provisions were made for forced ventilation. Periodic
measurements of gas concentrat. >is have not detected the presence of
33
-------
combustible gas within the confines of the building, indicating that the
precautions taken are satisfactorily venting the combustible gases from
beneath the building.
At the site of a new apartment building complex located adjacent to
an old 70 to 130 feet deep landfill which was formerly a dlatomaceous earth
mine, a gas interceptor trench was employed as a means to channel land-
fill generated gases away from the building complex. The trench extends
the full width of the development and was installed in natural material
between the fill and the buildings. The trench is approximately 18 feet
deep and 24 inches wide, and is filled with gravel. To vent the trench,
which will eventually be sealed over with pavement, four-inch diameter
pipes were Installed vertically at intervals of 30 to 40 feet.
The public agency that constructed this fill also utilized a similar
gravel filled cut-off trench in a different location to intercept the
gases that might flow from the fill to adjacent private property and
home sites. In this case, the trench was 30 inches wide, 20 feet deep
and 1,050 feet long. This trench was left open to the atmosphere for
venting purposes and has successfully arrested all gas migration to the
endangered properties. Recent measurements of gas concentrations on both
sides of the venting crench indicate that the trench is still acting
as a satisfactory deterrent to migration of the gases.
A more sophisticated gas control system will be used for a public
service building complex to be constructed adjacent to a former landfill
where monitoring has Indicated the presence of combustible gas. The entire
construction area is being filled with earth cover for a depth of approxi-
mately SO feet to bring the level of the completed surface to that of the
adjacent developments. Prior to filling, a trench was excavated in the
original ground in which a six-inch diameter perforated gas collector
pipe was placed. The trench is seven feet deep and two feet wide and is
backfilled with gravel and sealed at the top with a plastic membrane.
The total control system extends for a distance of approximately 850 feet
-------
and consists of two separate segments. The first segment, which is 450
feet long, consists of a gravel trench, perforated pipe and four vertical
gravel-filled wells, which extend upward from the trench to the surface
of the new fill. A four-inch diameter perforated vent pipe was installed
in each 24-inch diameter vertical well and connected to the perforated
collector pipe. These pipes extend 10 feet above the finished surface.
Each extended vent pipe, which is nonperforated, is capped with a
gas burn-off device. The second segment, which extends for approximately
400 feet, consists of the gravel trench, six-inch diameter perforated
collector pipe and five 24-inch diameter gravel-filled wells. These wells
are approximately SO feet deep and contain a four-inch diameter perforated
pipe connected to a six-inch diameter collector pipe. This segment con-
tains only one four-inch diameter vent pipe which is connected to the
collector pipe and extends upward through the new fill, terminating 10
feet above the finished surface. In the middle of the control system,
each collector pipe is connected to a six-inch diameter nonperforated
riser pipe, which extends vertically to the surface and is capped. If
needed these two pipes can be utilized for forced ventilation of gases
by installing a vacuum pump on the system. A series of gas monitoring
probes are also being installed as part of this system for measurement
of gas concentration at various depths and for continual monitoring and
evaluation of the system.
Other examples, which are mostly of similar design (but for which
little definitive information is available), exist and consist of gravel-
filled wells, gravel-filled trenches or both. Some control systems employ
forced ventilation in conjunction with gravel-filled wells and trenches,
whereas others rely entirely upon natural ventilation. The permeable
method for gas control appears to be more commonly employed because such
systems arc known to have performed satisfactorily and are relatively
inexpensive and simple to install.
35
-------
Examples of pre-construction systems utilized elsewhere in the nation
are few. One of these systems is located near Boston, Massachusetts at
the site of the new university of Massachusetts Columbia Point Campus in
Boston Harbor (Reference 26). The seven-building University complex is
being constructed on a completed dump site where substantial quantities
of methane gas and some hydrogen sulfide have been detected. To assure
safe beneficial occupancy of the buildings after completion, an elaborate
gas detection and control system has been designed and installed for each
building at this site. Each system consists of a network of perforated
pipes laid in a gravel blanket beneath the floor of the building. The
perforated pipes are connected to collector pipes which convey the gases
to a vertical venting system existing above roof level. The gravel blanket
Is overlain with an impermeable gas barrier (40-rail polyvinyl chloride
membrane) placed over a concrete mudsill, and all subterranean utility
entries to the building are sealed. Each piping network is of varying
size, predicated upon quantities of gas flow and head losses in the system.
Automatic gas analyzers and control panels in each system provide con-
tinual printout readings of combustible gas concentrations beneath the
buildings and from occupiable interior spaces. The control panels activate
the gas-evacuation pumps and air-inlet valves when combustible gas con-
centrations reach a pre-set level. All such systems vill be monitored
at a central console and may be separately monitored and controlled.
One building at a lower grade is protected with a membrane-enclosed
rock blanket filled with nitrogen under pressure, and a sub-ramp space
is constantly purged by means of a blower and pipe network. Performance
checks of the University of Massachusetts systems are underway, and all
systems appear to be performing satisfactorily.
Post-Construction Systems
As mentioned previously, the majority of the gas control systems
reported in the literature are of the post-construction type and have
been constructed to eliminate an existing and potentially dangerous con-
dition. Many of these systems have been experimental in nature and are
the forerunners of many of the pre-construction systems discussed above.
36
-------
Three expi-riiaental post-construction gas control systeas have been
developed and tested in Los Angeles County as part of a three-year
research study (Reference 3).
The system that has been most extensively tested and evaluated con-
sists of five wells which are spaced 40 feet apa^t and are constructed In
natural ground adjacent to the landfill Interface. Each veil Is 30 inches
in diameter and 60 feet deep. The wells are divided into three separate
levels. A six-inch diameter perforated pipe is installed to the bottom
wf each level of the gravel-filled well. Each level Is topped with a
layer of concrete to eliminate short-circuiting -within the well during
pumping fro.ii various levels. The three perforated pipes from each
well arc connected to an vij>ht-inch dianeter header pipe that is connected
to a 25-horsepover exhaust blower.
An extensive testing and monitoring program was conducted to ex-
amine the relative effectiveness of this gas control system in reducing
or preventing gas migration beyond the vertical plane of the well system
to the adjacent properties. The basic conclusions drawn fron the program
were that natural ventilation of the wells did not provide an effective
barrier to gas migration; that some forced ventilation across the plane
of the wells is necessary to arrest gas migration; and that with con-
tinuous pumping at the rate of 80 scfn per well, well spacinge of 150
to 200 feet nay be sufficient for the control of migrating gases. It
should be noted, however, that these conclusions arc for the particular
test installation described and reported upon, and nay not be exactly
applicable to other areas with different soil and other environmental
conditions.
An extensive post-construction system has been employed by the City
of Los Angeles for controlling the migration of gases from an active land-
fill Into an adjacent residential area and a school building complex
(Reference 27). Several test systems were constructed and evaluated prior
to installation of the principal permanent system. The basic permanent
system consists of 17 ventilation welle 25 feet deep, spaced on 150-foot
centers with a design withdrawal rate of 200 scfm from each well. It
37
-------
had been determined from earlier test prograas conducted on three of these
wells that the spacing of the wells, and thus the zone of influence of
each well, was a function of the withdrawal rate froo each veil, and that
the depth of the well was not a significant factor in tlie overall effec-
tiveness of the control system. These criteria were used in designing
the main system. The City chose a 25-foot depth, which was considered
to be sufficient. The wells were placed along two sides of the 125 to
ISO feet deep fill only (opposite the residential area on one side and
the school area on the other side). The other two sides of the former
gravel pit had previously been lined with a clay barrier, placed as the
fill progressed, to prevent water intrusion from adjacent spreading grounds.
It was expected that the clay liner would also serve as a gas barrier,
and therefore no forced ventilation system was installed on those two
sides. In recent (1974) observations, it is apparent that gas is
migrating beneath the wells. Installation of additional wells, 100 feet
deep, is planned.
The wells in the above system are two feet in diameter and are drilled
in natural ground. A six-inch diameter perforated polyvinyl chloride
pipe was placed in each well, and the wells were backfilled with gravel.
Each well pipe is connected to a vitrified clay header pipe that runs
three-fourths of a mile to a 15-horsepower blower. The blower feeds an
induced draft vapor rume incine.-ator that uses natural gas as a supple-
mentary fuel. To date, tests have indicated that gas migration has been
controlled by this ventilation system and that odors can be controlled
by the fume coL-_ustion chamber. However, further tests are being con-
ducted to determine optimum flow rates, effective areas of influence of
each well, the total amount of gas produced and the rate of gas produc-
tion with respect to time.
A system similar to that being used by the City of Los Angeles (as
described above) was utilized by the County Sanitation Districts of Los
Angeles County to control the movement of gases from the completed por-
tion of one of their active sanitary landfills. The fill Is adjoined
-------
would not migrate Co these areas, a series of 18 24-inch wells, varying
in depth from 30 to 40 feet and spaced 100 feet apart, have been drilled
to date (1974). A six-inch diameter polyvinyl chloride perforated pipe
was placed in each well, after which the wells were backfilled with gravel
to within eight feet of the surface. Each well was capped with a layer
of driller's mud and three feet of concrete, and was then backfilled with
earth to the finished surface. Each well pipe was connected to a header
pipe, which was connected to a S-horsepower suction blower. Two locally-
fabricated burners, each receiving 400- to 625-scfm gas flow, are in-
stalled on piping assemblies from the blower exhaust and operated in
parallel. Although gas flow is less than the design value, preliminary
results indicate that gas movement into the adjacent properties are
being arrested by this control system. It became necessary, however, to
install seven additional wells to the original 11-well system. The
burners, equipped with automatic pilot lights, operate continuously.
A gas control system utilizing natural ventilation was installed by
a manufacturing company to protect its building coirplex. The system was
installed approximately 150 feet from the edge of an old landfill in con-
junction with the construction of an addition to an existing building and
extended the full length of the complex bordering upon an old landfill
site. The installation consists of a series of gravel-filled wells spaced
20 feet on centers and a gravel-filled trench five feet in depth. Each
veil is two feet in diameter and 40 feet In depth. A four-inch diameter
perforated pipe was placed in the trench to collect the gases from the
veils and the trench area. Vertical risers, four inches in diameter and
connected to the trench pipe at a spacing of approximately 100 feet,
extend up the wall of the building for emission of gases above the top
of the buildings. The effectiveness of this control system is presently
unknown.
39
-------
CHAPTER VI
SELECTION. DESI3J AND INSTALLATION OF PROTOTYPE CAS BARRIER
HISTORY OF SITE DEVELOPMENT
Development of the Link Road landfill, ..he Armory ano other facili-
ties on or near the landfill Is described in Chapter II. Primary features
of Interest with regard to design and installation of the gas barrier
included an escinatc of the gas production potential of the landfill's
contf-its, the permeability of the adjacent soils, the history of com-
bustible gas concentrations and migration, the proximity to the landfill
of buildings and facilities to be protected and the location of under-
ground facilities.
In 1973, the Link Road landfill was 24 years old, the last refuse
having been placed near the Armory In September 1969. Extensive settle-
neat had occurred just back of the Armory and the fire training area.
Settlement was continuing, the landfill surface was covered with weeds
and landfill-generated gases were venting from the soil cover through
extensive cracks in a number of areas.
In July 1973. over 100,000 cubic yards, or approximately 162,000
Sons, of new soil cover were placed over the old cover throughout most
of tne undeveloped upper portions of she landfill to restore final grade
and to fill settlement areas. This represents an additional load on the
upper surface of the landfill of over 500 pounds per square foot. Depth
jf nev cover placement back of tr.e Armory was three to four feet.
-------
SITE DESCRIPTION
Physical Features
Topography- -
The landfill occupies approximately 40 acres of a 60-acre tract
in the southeast portion of Winston-Salem. The perimeter of the landfill
extends about 5,600 feet.
Originally, the site consisted of a relatively steep hillside In
the vicinity of the present Armory, with a ravine running easterly along
the present line of the 72-inch storm drain. A small stream ran through
this area and in the flat portion of the site, near Salem Creek, the
stream bed contained fairly extensive swampy areas. The terrain slopes
generally southeast from the highest elevations near the eastern boundary
of the landfill site towards Salem Creek, which is about 70 feet lower.
Final topography and typical cross-sections are shown in Figures 3 and
4. As landfilling progressed, wastes were Initially deposited in the
areas near Salem Creek. Drainage ditches were used to dewater the flat
areas and work progressed up slope, finally terminating near original
grade in the vicinity of the Armory. During operation of the abattoir,
highly organic slaughterhouse wastes and dead animals were burled nearby.
A small amount of refuse was also compacted over a limited section of the
incinerator residue area and covered prior to conversion of this area
into a parking and maintenance area.
The present fire training area is grassed and relatively flat,
sloping gently to the center of the main landfilled area. Some settlement
areas still exist in which water stands. Surface cracks are evident. The
peripheral road drains through a storm-sewer system to the 72-inch main
storm drain. Most of the parking and maintenance area is surfaced with
a 12- to 18-inch thick well-packed layer of crushed stone. A small por-
tion is paved with asphalt, and street drains divert runoff to the main
atorm drain. The entry road and the immediate area around the Armory are
paved with asphalt and drain to the main storm drain. Relatively shallow
sanitary sewers cross the parking area from Forsyth Technical Institute
41
-------
and connect with sewers entering the property from Silas Creek Partway
and passing southeastward by the Armory. Sewers and drains fron
the Armory connect to this sewer. A looped eight-inch hater main passes
through the parking area to serve the hydrants in the fire training area.
The upper surface of the completed landfill is relatively flat,
but beyond the fire training area and to the east of the Armory, it begins
sloping rapidly towards Salem Creek. Surface drainage is to the south-
east, and no severe surface drainage problems have been noted.
Soils—
The soils at the base of the landfill are primarily sandy, with a
relatively low clay content. Soil information obtained during preliminary
testing for foundation design in the vicinity of the Armory indicated
a rather sandy soil, with unsuitable bearing capacity for building con-
struction in some areas. The cover material obtained on site was generally
of a sandy nature with relatively low clay content. Varying amounts
of soil cover material were brought in from time to time from a number
of sources during the development of the landfill, and there is no con-
sistent pattern of soil cover by type.
Climate—
The Wins ton-Salem area has a relatively moderate climate character-
istic of inland areas of the Piedmont. Average annual precipitation
is about 42 inches with a major amount of precipitation occurring in
the early summer and early winter months. Temperatures in summer rarely
exceed the mid-90's. Winter weather is generally moderate with snow
and subfreezing weather of relatively light intensity and duration.
Hydrology—
Boring logs of subsurface explorations, placement of probes and
excavation for the prototype gas barrier indicated that the groundwater
table is between 25 and 30 feet below the surface along the line of the
barrier, i.e., outside the land ."ill at the higher elevations. Similar
exploration in the landfill, nowev^r, indicated water levels within three
-------
to five feet of the surface toward the center of the fill. These findings
indicate occurrence of saturated areas throughout the landfill due either
to water penetrating the soil cover or emanating from subterranean springs,
with most occurrences in settlement areas. It is apparent that most of
the mass of the landfill has reached field saturation, a condition con-
ducive to maximum gas production.
Deposited Refust
Field checks indicated that wastes buried in the landfill had an
average density of about 26 pounds per cubic foot. The average depth of
fill was considered, for design purposes, to be 35 feet. As reported by
Winston-Salem officials, the refuse to cover ratio was 2 to 1.
Other Features —
The principal areas of concern with regard to gas migration and
potential hazards at the Link "oad site are along the boundaries between
the landfill and its most elevated periphery. Here the major construc-
tion of facilities has taken place, and the facilities are in the upgrade
zone of gas migration. In the fire training area, deposition of highly
organic wastes and lack of compaction may be resulting in high gas genera-
tion and migration toward the parking area and the Armory. Gas was not
found above the periphery road in early investigations, however.
The placement of roads, buildings, crushed stone, sidewalks and
paved parking and access areas in the general vicinity of the Armory
has reduced vertical venting of gases from underlying refuse and has
directed migration of gases toward the Armory. The surface seal afforded
by the paving and crushed stone will, however, serve a useful function
In the operation of the gas barrier since short-circuiting of air
through the soil into the active barrier system will be reduced.
COMBUSTIBLE CAS CONCENTRATION'S AT LINK ROAD DISPOSAL SITE
Areal Distribution
The concentration of explosive gases in soil voids within and ad-
jacent to the landfill area is a positive Indication of sources of gas
43
-------
generation and general direction of long-tern movement. Because of the
vide fluctuations observed in gas concentrations from day to day, average
annual gas concentrations from the deep (25 feet) sampling probes were used
to obtain the combustible gas concentration contour map shown on Figure 6.
High concentrations were observed throughout the landfill, with rela-
tively abrupt declines at the peripheries. Near the Armory gas concen-
trations persist for a considerable distance from the landfill edge, which
is an indication of the relative ease of gas passage through the media
lying between the fill and the building. Gas contour gradients between
the fill and the truck parking area and the garage are somewhat larger,
indicating lower gas permeability in the soils. Relatively high concen-
trations persist at the outermost probe at the landfill edge nearest the
abattoir.
The observation of a relatively gentle gas concentration gradient,
i.e., low resistance to flow, in the direction of the Armory in 1973 is
coupled with the observation of a soil-air relative-pressure gradient in
the same direction. Manometers connected to pressure probes installed in
the peripheral area of the landfill in a line extending toward the Armory
(Figure 15) consistently recorded decreasing pressures in the direction
of the building over a one-month period of measurements in early 1974*
as shown in Figure 7. Gauge pressure readings within the landfill reached
as high as 200-mra water column, whereas those nearest the building vas-
cillated near zero. Thus, the pressure and the concentration gradients
along this particular direction induced both mass flow and the diffusion
of combustible gas toward the building prior to the construction of the
barrier.
Temporal Variations
Monitoring of combustible gas concentrations from a large number
of probes in the Armory was conducted during the period 1970 to early
1974. The large quantity of data obtained were plotted, general'.zed
and a number of representative or critical stations were selected for
presentation in Figure 8.
44
-------
GARAGE
•S5
CONTOUR INTERVAL: 10 PERCENT CH* BY VOLUME
Rgure 6: Combustible 909 concentration distribution in Link Rood disposal site,
1973
-------
§
0
••
o
9
£
E
UJ
-------
V)
S
Ul
CO
o
u
u
o
u
c
111
• BARRIER EVALUATION INITIATED
•GAS MIGRATION BARRIER SYSTEM CONSTRUCTED
•W*cu yd ADDITIONAL SOIL COVER PLACED OVER LANDFILL
• TRENCH ALONG S AND E SIDES OF ARMORY BUILDING
EXCAVATED AND NEW ANALYZER DEPLOYED
IU NtW ANALT£e.n Utr-LUItU N
I I I I I I I ( I I I I I I I I I I M I I I ill I I I I I 1
§ »-
1972
Figure 8: Generalized combustible gas concentrations at several stations near
Armory
-------
Rigorous attempts at correlating combustible gas concentrations with
rainfall and temperature variations failed to demonstrate any regular
patterns. However, it can be observed from Figure 8 that a steadv decline
of combustible gas concentration had been occurring over the period of
monitoring, before barrier construction, at all stations in the vicinity
of the Armory (Stations 7, 8, 9, 10 and 11, Figure 5).
The sudden rise in concentration of explosive gases noted in late
1972 coincides with the excavation of an L-shaped trench along the south
and east sides of the Armory. It is believed that backfilling of this
trench with loose material caused migrating combustible gas from areas of
higher concentrations to move more readily toward the probe stations than
was possible previously. At about the same time that the trench was
dug, a new gas analyzer was purchased and deployed by the monitoring
personnel. It was a few months later that the new gas analyzer was found
Co be reading five to 10 percentage points higher than the previously-
used analyzor. The COL ji nation of these two events tend to explain the
sudovn rise in gas concentrations that occurred in late 1972. Soon after
the ,udden rise, a gradual decline was observed with a rate somewhat
higher than that observed in the early part of the monitor ing period.
Altogether, trends shown in Figures 6, 7 and 8 confirm the specific
conclusions reached earlier about gradual gas concentration reductions
over time, the response of the concentration at a given peripheral point
to permeability of wdia between that point and the landfill, and the ex-
istence of an overwhelming mass flow component in the migration of gases
away from the landfill.
EVALUATION OF ALTERNATIVE GAS BARRIER SYSTEMS
Three alternative gas barrier systems were evaluated for the purpose
cf selecting a suitable method of gas control for the Link Road disposal
site. The alternatives considered were forced ventilation, natural
ventilation and impervious membrane. Three modes of operation of the
forced ventilstion method were considered: high-flow 0-100 scfm/well)
ventilation, low-flow (MO scfm/well) ventilation corfeined with the use
48
-------
of a surface seal and lew-flow (ilO scfin/well) ventilation without the
use of surface seal.
The objectives and functions of high-flow, forced-ventilation and
low-flow, forced-ventilation systems are quite different. The objective
of the high-flow, forced-ventilation system is to create a relatively
large bulk flow of soil-air gases, which Includes a substantial amount
of overlying atmospheric air that is pulled through the surface of the
landfill, from the area of influence of each well. Therefore, in addi-
tion to providing an effective barrier against migration of landfill-
generated gases, this system effects a dilution of combustible gases in
the soil-air on the landfill side in the vicinity of the barrier. There-
fore, in this sytsm, no attempt is made to exclude atmospheric air, ex-
cept in the immediate vicinity of each well. The objective of the low-
flow, forced-ventilation system is to create a vertical plane, or barrlert
of ubiquitous negative pressure between wells which will intercept land-
fill-generated gases naturally migrating toward the barrier and cause
them to flow to a well due to the induced negative pressure gradient.
In this system, a minimum amount of landfill-generated gases, I.e., only
that necessary to create the barrier of negative pressure, is extracted
from the area of influence of each well. To be effective, this system
requires that the inflow of overlying atmospheric air into the landfill
in the vicinity of the barrier be kept to an absolute minimum.
As described in Chapter V, the state-of-the-art of gas control tech-
nology is by and large experimental at this point in time. The experi-
mental and demonstration projects involving gas migration barriers have
generally failed to provide universal design criteria. However, existing
systems of gas migration control have demonstrated the feasibility of
controlling the flow of gases from landfills. Furthermore, the limited
experience has provided some field data on the mode of operation and
effectiveness of several gas ventilation methods. These data are of
value for developing new design criteria for conditions other than those
for which direct experience is available.
-------
Several criteria were considered in evaluating the various methods
of gas migration control. The methods considered were ranked under each
criterion and given an appropriate score in accordance with the quantita-
tive or subjective information presently available. The judgement on
the final ranking of the systems may be more readily Justifiable in some
cases than in others where performance data are scarce or inapplicable.
Nevertheless, it is believed tnat a fair and impartial evaluation of
each system was obtained with the scoring system used.
Cost
Total annual costs were obtained from estimates of capital cost
and operation and maintenance cost for each alternative, and are shown
In Table 1. All capital costs were amortized over a 10-year period at
an interest rate of aix percent. As indicated in Table 1, the highest
rank or score of five was assigned to the system with the least total
annual cost and a score of one to the most expensive. Detailed informa-
tion on capital cost requirements for forced ventilation systems,
natural ventilation systems and impermeable barrier systems is pre-
sented in Appendix A.
Table 1. ESTIMATED COSTS OF ALTERNATIVE GAS BARRIER SYSTEMS
(dollars)
Total Annual Total Rank/
Alternative system capital cose 04M cost annual cose Score
Forced ventilation
a. High flow
b. Low flow with seal
c. Low flow without seal
Natural ventilation
Impermeable barrier
52,000
73.600
37,600
10,000
24,500
3,000
1.200
850
0
500
10,000
11,500
6,100
1.400
4.000
2
1
3
5
4
50
-------
Effectiveness
Because effectiveness is the primary factor of concern in the selec-
tion of a gas control system, this parameter was assigned a weight of
three. Thus in a one-to-five relative ranking of the alternatives (with
the highest rank assigned to the most effective method), the score for
effectiveness of an alternative system vould range from three to IS.
Experience with existing systems Indicates that forced-ventilation systems,
as a group, are far more effective than either the natural ventilation or
the impermeable barrier systems. Natural ventilation systems have gener-
ally been least effective in preventing gas migration from a disposal site.
because high-flow, forced-ventilation barrier systems have proven to be
effective, this alternative was assigned the highest score.
Low-flow, forced-ventilation barrier systems have not been tested to
date as a means of preventing gas migration. Therefore, their relative
ranking is based upon theoretical fluid flow behavior which clearly
indicates that the low-flow, forced-ventilation system would be far more
effective than natural ventilation or Impermeable barrier methods. By
the same token, it is surmised that the effectiveness of low-flow, forced-
ventilation methods may be somewhat lower than high-flow, forced-ventila-
tion systems. Theoretical considerations also imply that the use o£
a surface seal, as discussed in the sections below, would Increase the
effectiveness of low-flow, forced-ventilation systems in intercepting
migrating gas.
The assigned rank and score of alternative systems from the viewpoint
of effectiveness is presented in Table 2.
Maintainability
Maintainability is important because It bears directly upon the
reliability of and continuous safety'provided by a given barrier system.
Generally, the simpler the system the more maintainable it will be. Con-
versely, the more complicated and larger the barrier appurtenances, the
greater are chances for breakdown and the need for maintenance. There-
fore, the natural ventilation and the impermeable barriers would be more
maintainable than those requiring pipelines, pumps and control structures.
51
-------
Table 2. RELATIVE EFFECTIVENESS OF ALTERNATIVE GAS BARRIER SYSTEMS
Alternative system
Forced ventilation
a. High flow
b. Low flow with seal
c. Low flow without seal
Natural ventilation
Impermeable barrier
Rank
5
4
3
1
2
Score
IS
12
9
3
6
The rank and score for the five alternative systems with respect to
maintainability is presented in Table 3.
Table 3. RELATIVE MAINTAINABILITY OF ALTERNATIVE CAS BARRIER SYSTEMS
Possible naintenance Rank/
Alternative systems problems Score
Forced ventilation
a. High flow Pump failure, control mechanisms 1
maintenance, short-circuiting
of air flow
b. Low flow wich seal Pump failure, control mechanisms, 2
seal breakage
c. Low flow without seal Pump failure, control mechanisms, 3
short-circuiting
Natural ventilation Plugging of porous material by 4
surface drainage
Impermeable barrier Seal breakage 5
52
-------
Controllability
This criterion is a measure of the ability of operators to adjust
the operation of a system in accordance with the feedback provided by the
monitoring system. Thus, in a highly controllable system, working with
100 percent effectiveness, operating costs can be lowered by reducing
flow rates or decreasing purapage periods, or both. The energy input to
the same system nay be increased to intercept any gases upon the incep-
tion or Increase in gas migration. The rank and score of the five alter-
native systems with respect to controllability is shown in Table 4.
Table 4. RELATIVE CONTROLLABILITY OF ALTERNATIVE CAS BARRIER SYSTEMS
Alternative system Control feature Rank/Score
Forced ventilation
a. High flow Flow rate (decreasable), 3
well spacing
b. Low flow with seal Flow rate (wide range of varia- 5
tion), well spacing, strip
width
c. Lou flow without seal Flow rate (wide range of varia- 4
tion), well spacing
Natural ventilation Hone 1
Impermeable barrier More 1
Environmental Effects
The most noticeable environmental effects of disposal site gas migra-
tion barrier systems are noise, odor, options for final use of the site
end the aesthetic appearance of system appurtenances. Noise and odor are
primarily caused by the forced-ventilation systems involving use of motors
and the discharge of pumped refuse gases into the atmosphere from a con-
centrated point source. The high-flew, forced-ventilation systems, by
za'.ure. pose the greatest noif level and the highest discharges of
malodorous gases. The low-flow, forced-ventilation systems, on the other
53
-------
hand, are relatively quiet and contribute much smaller volumes of landfill
gases to the surrounding areas. The existence or absence of surface seal
does not appear to contribute significantly and in a direct manner to
change in either noise level or odor dispersal. Indirectly, however, it
may reduce total punpage flow requirement, which would lower noise and
rates of malodorous gas extraction. Both natural ventilation and imper-
meable barrier systems are completely noise free. The natural ventila-
tion systems conceivably can contribute more naladorous gases along the
barrier by providing freer venting access than can the impermeable barrier
system.
' From the point of view of options for final use of the disposal
site, the forced-ventilation systems are favored over the others due
to the fact that such systems would more readily enable beneficial uses
of areas on and adjacent to the completed disposal sites due to their
greater effectiveness. Such uses include public parks and recreation
sites and other facilities which nay be developed upon the completed
fill areas.
rj Aesthetically, the impermeable barrier system is completely innocuous
after construction is completed, whereas the natural ventilation system
will have some protruding vent pipes. The forced-ventilation systems.
on the other hand, pose structural unsightliness in proportion to the
magnitude of pumpage and surface sealing. To the extent that these
features detract from the natural open-space setting, the forced
ventilation systems would be aesthetically more objectionable. The rating
scores given each alternative barrier system with respect to environmental
effects are presented in Table 5.
Disturbance Due to Construction
In the course of constructing new gas migration barrier systems, a
certain amount of earth excavation, well drilling, pipeline installation
and other activities are certain to disturb the site for the duration of
const'.-uction work. Placement of either a natural ventilation system or
en impermeable membrane will require a large amount of excavation and
b&ckfilling work with the attendant stockpiling of fill material as well as
54
-------
disposal of excavated soil. These disruptions are monumental relative to
well drilling and surface seal application required in the forced-ventila-
tion systems. The assigned scores for disruption due to construction of
the alternative systems are presented in Table 6.
Table 5. RELATIVE ENVIRONMENTAL I Iff ACT OF ALTERNATIVE GAS BARRIER SYSTE11S
Rank/Score
Alternative system
Forced ventilation
a. High flow
b. Low flow with seal
c. Low flow without seal
Natural ventilation
Impermeable barrier
Noise
1
3
2
5
5
Odor
1
3
2
4
5
Final use
options
3
5
£
2
1
Aesthetics
2
2
3
4
5
Table 6. RELATIVE DISRUPTION DURING CONSTRUCTION
OF ALTERNATIVE GAS BARRIER SYSTEMS
Alternative system
Rank/Score
Forced ventilation
a. High flow
b. Low flow with seal
c. Low flow without seal
Natural vcncilation
Impermeable barrier
5
4
5
1
1
55
-------
Experimentation and Innovation
Because of the paucity of information presently available on the
entire problem of gas migration control, it is deemed important to
emphasize innovation and experimentation in this field. Furthermore, a
system with built-in flexibility, allowing variation of the independent
parameters (pumpagc rate, strip width, etc.), is considered to be a
great advantage. The advantages are twofold. First it allows the system
to be "tuned" to the particular peculiarities of the site, and secondly
it provides data for use in design and operation of similar gas migration
barriers at other locations. The innovative character of each system
and its flexibility are judged on the basis of past experience or lack
thereof with the respective systems. The impermeable barrier system is
one of the oldest concepts, followed closely by the natural ventilation
alternative. On the contrary, application of a surface seal and a low-
flow, forced-ventilation barrier is the most novel concept in gas migra-
tion control. High-flow, forced-ventilation systems have been in use in
several locations for some time and are thus ranked intermediate among the
other alternative systems insofar as experimentation and innovation are
concerned. The ranking and scoring of the alternative systems from this
respect are summarized in Table 7.
Table 7. RELATIVE NOVELTY OF ALTERNATIVE GAS BARRIER SYSTEMS
Alternative system Novelty feature Rank/Score
Forced ventilation
a. High flow None 3
b. Low flow with seal Low flow, surface seal 5
c. Low flow, without seal Low flow 4
Natural ventilation None 2
Impermeable barrier None 1
56
-------
SELECTION OF THE GAS BARRIER SYSTEM
A summary of the scores assigned each alternative gas control system
is presented in Table 8. The low-flow, forced-ventilation system with
a surface seal received the highest cumulative score, followed by the
low-flow sysLem without the surface seal. The high-flow, forced-venti-
lation system has the next highest score, followed by the impermeable
barrier system. The natural ventilation system has the lowest total
score.
On the basis of these scores, the low-flow, forced-ventilation
system with surface seal was selected for design to provide a barrier
against migrating gases from the Link Road landfill.
THEORETICAL CONSIDERATIONS
Laminar flow of fluids through porous oedia is governed by Darcy's
law. In order to be able to use analytic tools to describe the movement
of landfill gases toward outlying areas, it is necessary to make certain
justifiable simplifying assumptions concerning the fluids (air-methane
mixture), the porous media (refuse, fill and undisturbed soils adjacent
to the landfill) and the flow process. These assumptions and the justi-
fications therefor are given, in some length, before a presentation of
the theoretical considerations tor the design of a low-flow, forced-venti-
lation system with surface seal.
Assumptions
Laminar Flow—
Darcy's law has been experimentally found valid for all flows in
the laminar range, i.e., with Reynold's numbers below unity. Because
the kinematic visocity of air is about 10 times greater than that of
uatev, the law allows an air flow 10 times that of water within the
applicability of Darcy's law. Therefore, inasmuch as Darcy's law is
universally used successfully to describe groundwater movement under
most conditions, laminar flow should prevail for all gas movement pro-
blems encountered in barrier system design and related analyses.
57
-------
Table 8. SUMMARY OF RELATIVE SCORES FOR ALTERNATIVE GAS BARRIER SYSTEMS
wn
oo
Total
Alternative annual
systems cost
Forced ventilation
a. High flow
b. Low flow
with seal
c. Low flow
without seal
Natural ventilation
Impenieable barrier
2
1
3
5
4
Effec-
tiveness
15
12
9
3
6
Maintain-
ability
1
2
3
4
5
Control-
lability
3
5
4
1
1
Con
struction
disturbance
5
4
5
1
1
Environ-
mental
effects
7
13
11
15
16
Experimen-
tation &
innovation
3
5
4
2
1
Total
score
3fi
4?
39
3)
34
-------
Isotropic Media—
Mechanics of fluit1 flow in porous media almost universally assume
isotropicity of the medium. Flows occurring. In a well-type barrier system
are in fact three-dimensional, even though the principal direction of
movement of gas is perpendicular to the barrier. The only anisotropicity
of significance may be due to the various horizons developed during the
soil-forming process. As long as flow along the vertical dimension is
minimal and differentiation of soil horizons is not highly exaggerated,
the assumption of isotropicity is not expected to limit the applicability
of Darcy's law co flow situations under consideration. Generally, the
effects of anisottopic media upon flow conditions are only of theoretical
significance. In practical applications these effects are subordinated
by the scale of the flow parameters.
Homogeneous Media—
The assumption of homogeneity of soils along and in the vicinity
of the gas migration barrier led to the use of a single representative
permeability factor for the entire length of the gas barrier. The super-
ficial fallacy of this assumption is evident In the range of perraeabilities
measured for the air-dried samples taken from various points In the vicin-
ity of the barrier. Soil permeability variations are an common within
relatively short distances that unless a great number of samples are
obtained, It is not safe to assign the results of one test to the zone
of Influence of a number of wells. Thus a single representative per-
aeability value was judiciously selected for use in the design of the
entire barrier systea. Natural soil variations will generally be averaged
out for the entire system, as well as for each well, resulting in flows
close Lo theoretical derivations. The problem of lack of homogeneity
Is dimilarly handled in groundwater flow analyses successfully.
Steady-State Flow—
At the begi.i.iing of thi. operation of the gas migration barrier sys-
tem, a transient flow condition will exist wish complex flow and pressure
distribution patterns in the vicinity of the barrier. This time-variant
59
-------
condition will last until streamlines and the pressure distribution net-
work finally become stabilized La a steady-state pattern closely resen-
bling those predicted by the equations. During periods of wetting and
drying of the soil surfaces (such as at the onset of the wet and the
dry seasons) nonsready-state gas flour will prevail due to changing soil
permeability in those periods. A Ions-terra decrease in gas generation
rate In the disposal area vil.1. impose an additional transitory character
to the gas flow parameters.
The effects that the nonsteady character of flow axe liable to exert
upon flow conditions arc all such that design based upon a steady-state
assumption is biased toward providing a conservative system.
Constant Gas Density—
In the range of negative pressures produced by the pumping.equinment
proposed for the barrier system (less than 0.01 atra), pas densities should
not vary by no re than one percent. Snail variations will not affect the
accuracy of predicted flows to a significant or measurable degree.
Analytical Relationships
For the condition under consideration, namely the presence of a
lower confinement to gaseous flew in the forn of the groundwater table
which is relatively impermeable cocpdred to the overlying unsaturated
soil and an upper confinement in th? form of paving or other appropriate
seal, the flow of gas tovard any well in the barrier is given by the
Thietn equation for radial flow of fluids in a confined aquifer (an
adaptation of Darcy's Law); viz.,
-2 ir h k AP
q u In (re/rw> (1)
where: q ** flow to each well (cfs)
h » depth La groundwater table (ft)
k " Intrinsic soil permeability (ft2)
&p •* imposed pressure differential between well and Unit
of influence, re (psf)
li " average gas viscosity (lb-sec/ft2)
re/rv B ratio of radius of influence of well to radius of well
60
-------
In order to intercept all gases moving toward the protected area.
wells should be spaced along the barrier so that pumpage from all wells
in the barrier equals gas movement toward the barrier. Flow from each
side of the barrier may be calculated by using Darcy's equation directly,
«-*»oS (2)
where: Q = total gas flow front each side to the gas barrier (cfs)
L « length of barrier (ft)
&? » average pressure differential between the barrier ant
an orthogonal distance AX away (psf)
AX - distance over which AP acts (ft)
Accurate estimation of the pressure gradient, AP/AX, is the key to the
usefulness of liquation 2. Because the negative pressure along the barrier,
indicated by Ap, will vary from a maxima at the wells to a minima at mid-
points between the wells, the effective negative pressure at the barrier
is assumed to be about 0.5 Ap. Therefore,
nP - 0.5 fip (3)
The total system flow is equal to 2Q and the number of wells along
the barrier, n, may be determined by dividing the total flow from both
sides of the barrier by pumpage rate from each well; viz.,
• -V m
Spacing bntwecn wells, S, is obtained by dividing the length of the
barrier system by the nuiaber of required wells; i.e.,
S " n (5)
Spacing between wells can also be determined independently cf most: system
variables by substituting Equations 1, 2, 3 an.i 4 into Equation 5; viz.,
, 2» AX
In <.-r) (6)
61
-------
To assure confined aquifer conditions, the width of seal along the
barrier, W, must be only sufficient to provide the pressure gradient
AP/AX. Therefore,
W = 2 AX (7)
Cost Minimization
The economic trade off; between the width of the sealed barrier and
the number of wells along the barrier can be evaluated using the relation:
C - CQ + Cx + C2 (8)
where: C = total cost of the barrier
CQ = costs independent of number of wells and spacing
C^ " costs associated with the number of wells in the
system «* Kj n
KI = cost of each well and appurtenances thereto
C2 - costs associated with the sealed strip = 2 K2 L AX
K2 » cost of paving a unit width along the entire length of
the strip
A minimum total cost solution can be obtained if C Is related to
AX by substituting Equations 1, 2, 3 and 4 into Equation 8; i.e.,
L In (rc/rw) Kj
C ' C0 + —TVAX + 2 K2 L AX (9)
Differentiating C with respect to AX gives:
2 K, L
_dC L ln (re/rw> Kl
2 TI (AX)2
Setting dC equal tc zero to obtain the minimum cost solution gives:
d(AX)
AX =
The first multiplicand is a physical constant for a given system and the
second can be determined from prevailing unit costs of the various
elements of the barrier system.
-------
CAS BARRIER DESIGN
Widtn of Surface Seal
The nost economical AX is determined from Equation 10. Tor a well
radius, r , of one foot and an assumed radtus of influence, r , of 100 feet:
AX - \/ln (100/1) /An
- 0.605X^/1(2 (11)
Choice of an appropriate numerical value for r depends, to a certain
extent, upon the expected final value of the width, W, of the sealed
strip along the barrier. The width of seal will determine the approxi-
mate radius of Influence of the well by imposing a lower limit of AX
feet upon the path that the air stream will traverse horizontally before
arriving at the intake perforations of the wells. Thl; assignment,
arbitrary as it may appear, is not too critical in the solution of Equa-
tion 10, because of the logarithmic and square root function In which
r appears. (A 50 or 100 percent error in estimation of r will result
in only a 4.6 or 8.4 percent error, respectively, in the value of AX.)
The unit cost of wells, K., is comprised of the costs of drilling,
piping, gravel placencnt and monitoring and control devices for each
well. These costs, under the existing conditions, and for a 30-feet well
depth are:
Drilling 0 $20/ft - $600/well
Piping @ 3/ft = 90/well
Gravel @ 2/ft - 60/well
Monitoring (orifice meter) - 350/well
Control (butterfly valve) - 150/well
TOTAL RX - $1,250/we 11
The unit cost of sealing the barrier strip, K_, is the unit cost of
placing sealant on each square foot of the barrier. Cost of sealant is
estimated to be S0.10/ft".
63
-------
Replacing K^ and KZ in Equation 11 gives:
flX - 0.605 >/i."250/0. IB = 50 ft
Tills value of AX is Chen used as a basis for Che computation of flow
Coward wells, spacing of wells and actual strip width.
The resulting width of the paved strip alone, the line of well is:
H - 2 AX - 100 ft
Flow Rote and Capacity
An important parameter in the computation of flow toward the barrier
and inco tlio wells is soil permeability, k. Thirteen soil samples from
eight different locations along the proposed barrier and at two depths
In five locations were analyzed in the laboratory. A nonste,-\
-------
Table 9. PARTICLE-SIZE DISTRIBUTION AND PERMEABILITY OF SOILS IN VICINITY OF PROPOSED BARRIER
Probe*
1
2
3
8
25
33
54
55
g
Location
Southwest of Armory supply
Sample
depth,
ft
room 6
South of Armory supply room 5
Southeast of Armory supply
South of Armory building0
West of Armory building0
East of Armory building6
South of truck park area
South of metal shed
room 6
4 to 6
14 to 16
4 to 6
14 to 16
4 to 6
14 to 16
4 to 6
14 to 16
4 to 6
17 to 19
Sieve analysis of soils.
percentage passing sieve
No. 4
4760u
86.5
100.0
99.3
98.4
99.4
99.6
97.1
98.7
99.4
97.0
97.9
85.3
98.4
No. 10
2000u
59.5
100.0
94.2
95.6
97.8
95.8
88.6
95.2
97.5
93.0
93.6
76.9
95.1
No. 40
417u
21.7
62.4
54.6
63.1
68.8
66.8
60.6
73.2
69.0
69.2
68.9
53.7
80.6
No. 200
74u
5.3
15.9
9.8
22.5
28.6
19.5
18.5
37.4
35.1
33.1
35.4
24.3
35.0
Permeability,
io-12 ft2
425
37.8
160
126
61.2
33.4
43.2
95.5
110
76.5
54.0
2.16
81.0
Figure 9 shows location of probes in the waste disposal site and vicinity.
b Samples were collected from the bottom of a trench dug along the southern wall of the supply room.
c Core samples were obtained from the given depths using standard sampling techniques (Reference 28).
-------
SILAS
CREEK
PARKWAY
SANITARY-v >
SEWER \
^^__- • '-
i
i
Y/.i
K
— • «
X
V
r~r~
GARAGE , ! X
^_- — S
SHED
®55 54
f ""
«.^FIRE LAB
/DRILL \
1 E3 1
VTOWERy
1
i
i
i
1
J
/
f
'
jP%^%?3
*! ^^<^^^ ARMORY
^^^ O ' \ v *^^^
CONCRETE
SLAB
YW^\
NOTE:
NUMBERS REFER TO PROBE
LOCATIONS SHOWN ON
TABLE 9
/- SANITARY
# SEWER
^
jsigo
••E^^MHBES
0 50
200
Figure 9: Soil permeability samples and associated probe locations in
vicinity of proposed barrier
-------
The choice of a viscosity value tor gases flowing In the vicinity
of the proposed barrier was made with full cognizance of pressure and
temperature Influences, as well as metnane concentration, upon the gas
mixture. The pressure dependence of the gas-mixture viscosity, in the
pressure ranges encountered, is negligible. The value chosen for vis-
cosity of the gas mixture allows for the most conservative computation
of flows in the temperature and methane concentration ranges expected
to be experienced under various operating circumstances.
Cas viscosity decreases with decreasing temperature and with increas-
ing methane concentration. The lowest probable soil-air temperature is
about 10°C at the depths where gas flow under forced ventilation will
occur. The estimated ncthane concentration of the f»as being generated
in the landfill is expected to be about 29 percent in 1973, the time
that the barrier system will become operational (see Figure 11). Linear
Interpolation of reported values for viscosity of air and of methane at
standard temperatures results in a design value for the methane-air nix-
Lure of 157 micropoises.
Using fquation 1 to calculate flow toward each well, and using the
following data:
h • 30 ft, as measured in the field from ground surface
to the groundwater table along the proposed barrier
— 12 2
k = 69.9 x 10 ft , selected representative intrinsic
permeability based upon laboratory tests
Ap = 0.01 atmosphere (.21.2 psf), selected vacuum to be imposed
by pump
p - 157 micropoises (3.28 10~7 lb-scc/ft2)
r = 100 ft, assumed radius of influence of a well
r • 1 ft, a design parameter
the flow to each well will be:
_i i
2 ir
3.28*10~7(4.61)
67
-------
Using Equation 2 to compute the total flow from each orthogonad
direction toward the gas migration barrier, and the additional appropriate
data:
L - 1.000 ft, the length of the main legs of the barrier
AP • 0.5&P - Q.OOS atm (10.6 psf)
AX « 50 ft, as determined from cost analysis
the total flow from one side of the barrier will be:
Q . -6.99.10-11(30>
-------
It is logical to presume that combustible gas generation rate is
in some direct fashion related to combustible gas concentration in the
surrounding areas to the disposal site. Such a relation would exist
whether gas transport were by diffusion or advectlon. Although com-
bustible gas generation data are not available, combustible gas concen-
tration values have been obtained at the Link Road disposal site over
the past three years at several sampling locations. Individual sampling
data were plotted and smoothed to eliminate the day-to-day variations
in the concentration of the coafcustlble gases. A plot of combustible
gas concentrations for one typical sampling station is presented in
Figure 10. To obtain a reasonable estimate of the pseudo-equilibrium
gas generation rate suggested by Figure 10, a few logical assumptions
are required regarding the quantitative nature of the events and con-
Li
ditJlons at the Link Road disposal site.
It has been reported (Reference 10) that a pound of refuse will
ultimately produce 3.9 ft of methane gas upon complete biochemical
decomposition. For conservative design purposes, this value of refuse
conversion to methane is assumed to apply at the Link Road disposal site.
The density of conpactcd refuse at the Link Road disposal site is
estimated to average about 26 Ib/ft .
The volume of the entire disposal site was estimated from cross-
sectional profiles of the filled area determined by field surveys. A
total volume of abcut 1.4 x 10 yd has been filled with refuse and cover
material. The refuse:cover ratio was estimated at 4:1 to be conservative
in the design (the City has indicated a ratio of 2:1). The perimeter of
the disposal site was measured at about 5,600 ft and the average fill
depth was determined from actual borings to be about 35 ft.
A semi-logarithmic plot of combustible gas concentrations at
Station 7 (see Figure 5 for location) versus time, which is shown in
Figure 11, indicates the following relationship:
69
-------
I 70
o
O
o
-------
IOO
50
J3
5
!O
o
<§
o
in
o
u I
00
00
u
0.5
O.I
1970
-0.288t
1975 I98O 1985
YEAR
1990
1995
Figure IT- Historical and projected concentration of
combustible gas migrating from disposal area at station 7
71
-------
-0.288 t
C - C0C
where C- and C are combustible gas concentrations initially (t » 0) and
at time, t, respectively. Figure 11 was based on onlv one sampling
."Cation because it provided the largest number of data and, of all
sampling stations, it is located in the closest proximity of the Armory
supply room where the flaeh fire occurred.
Assuming that gas generation rate is proportional to gas concentra-
tion at the periphery of the landfill during tht decomposition period,
Equation 12 can be rewritten for gas generation rates; viz.,
r _„ _ -0.288 t
where F_ and F are gas generation rates during the initial year and at
some later year, t, after completion of the disposal operation. The
value of F can be computed by integrating gas generation rate over the
life of the landfill and equating it to total gas production expected
from a pound of refuse; i.e.,
F dt - 3.9 ft3/lb
This procedure yields FQ = 1.082 ft CH./lb in the initial year after
disposal operations have been completed. Thus* Equation 13 becomes:
In laid- 1973, when the barrier presumably will become operational, the
annual gas generation will amount to 0.222 ft Cll./lb.
On this basis, the average CH^ flow rate during 1973 across a 1,000
feet long vertical cross-section of the landfill periphery will he:
x l.AxlO6 yd3 x 26 x 27 x 0.222 x 5.256 x 105
5600 9"* cfln
72
-------
The various simplifying assumptions made to arrive at this value are of
such a nature as to make it a conservatively high estimate, particularly
since no allowance was given to vertical venting through the surface of
the disposal site, which may only be approached during periods when the
-disposal site surface is made nearly impermeable to upward gas movement
due to precipitation, confining it to lateral migration. The 59 cfn
maximum gas migration rate computed above is far below the design value
of 163 cfm for the main branch (1,000 feet) of the low-flow, forced-
ventilation system, which provides for a reasonable factor of safety.
Pipe Sizes
For a maximum allowable total dynamic head (friction) loss in the
system of 10 Inches of water column and a design total flow rate of 196
ficfm, the suction header pipe should be four inches in diameter. A con-
stant diameter suction header is provided throughout the system to simplify
procurement and installation.
For the reasons discussed in che next section, the underground
piping, consisting of the wells and suction header, was sized to accom-
modate a high-flow system, should it be required. In this case, for the
sane maximum allowable Lotal dynamic head loss in the system of 10 inches
of water column and a design total flow rate of about 1,800 scfm, the
suction header pipe should be eight inches in diameter throughout.
Because the flow in each short-section of well pipe will be relatively
small and the concomitant head loss negligible, the well pipes were
sized at four inches in diameter to allow for greater ease of perfora-
tion in the field.
CONSTRUCTION STRATEGY
Due to the relatively high estimated cost of the surface seal
(about 45 percent of the total system cost) and the possibility that
the existing surface may provide an adequate seal, a step-wise con-
struction strategy was employed.
73
-------
Because any of the three forced-ventilation systems are considered
to be acceptable for controlling gas migration from the Link Road dis-
posal site, the two low-flow, forced-ventilation systems can be designed
to provide a backup capability by sizing the pipelines for maximum flows
required In a high-flow, forced-ventilation system.
Preferred System; Low-Flow. Forced-Ventilation Without Surface Seal
Low-flow operation without surface seal was rated next-to-highest
in total score in Table 8. However, because it is the least expensive
of the forced-ventilation alternatives, it was chosen as the first system
to be tested, and to be used if proven effective. Should the testing
program indicate that satisfactory gas migration control is achieved
with the low-flow system without a surface seal, no further construction—
and additional expenses—would be necessary.
Contingency Plan 1; Low-Flow. Forced-Ventilation With Surface Seal
If the low-flow system without a seal does not prove effective during
the test period, a surface seal would be applied along the barrier,
with width and other parameters specified in later sections. The pro-
vision of surface seal should assure gas migration control with the Low
flows predicated in the theoretical considerations.
Contingency Plan lit High-Flow. Forced-Ventilation
In the unllkelv event that the testing program reveals gas migration
beyond the barrier system while the low-flow, forced-ventilation system
Is in operation, flows can be increased by installation of much larger
pumping capacity so that adequate protection is provided.
The initial design of the pipelines and pumpage facilities for the
proposed barrier are carried out with the necessary provisions for con-
version to high-flow puim>agc rates of the magnitude already proved effec-
tive in other ateas close to solid waste disposal sites. Thus, if the
low flows prove not to be wholly effective (as judged by the testing
program), pumpage rates can be increased without incurring excessive
pipeline head losses.
74
-------
CHAPTER VII
DESCRIPTION OF PROTOTYPE CAS BARRIER
PLANS AND SPECIFICATIONS
Detailed plans and specifications were prepared prior to construc-
tion of the gas barrier system. A brief description of the various
components of the barrier system is presented herein in order to maintain
adequate reference for the evaluation of the system performance, given
in Chapter VIII.
The gas migration barrier is essentially a curtain of negative
pressures, in relation to the atmosphere, maintained indefinitely by
pumpage of gases from a series of wells along the curtain. Continuous
maintenance of negative pressure at every point along the barrier—placed
between the refuse disposal area and the Armory—is the key to provision
of protection to the occupants. The line of wells, in relation to the
landfill and the building, is presented in Figure 12.
BARRIER WELLS
Barrier wells, which provide the means for creation of negative
relative pressure, are all similar to the one shown on Figure 13.
The wells arc all interconnected through an eight-inch diameter polyvinyl
chloride header pipe, laid underground along the entire length of the
barrier. A butterfly valve, installed between each well head and the
header, controls flow from each well. This provision is important for
balancing system flow and equalizing pressure drop along the barrier.
Veil heads are also equipped with quick-disconnect assemblies for field
measurement of vacuum created in each well during operation. The valve
75
-------
15 14 13 12
FIRE LAB
PUMP
NEW FILL
NEW
FILL
N
A
LEGEND
BARRIER WELL
HEADER
SCAU - r«r
0 SO 100 ISO
2S 7S 123
Figure 12: Location of barrier wells, header, pump and vent
-------
QUICK DISCONNECT
VACUUM GAUGE
GROUND SURFACE
MANHOLE
HEADER
SEAL
15 FEET OF
COMPACTED
FILL OR
CONCRETE
SLOTTED
PIPE (PVC)
CONDENSATE
DRAIN AT
LOW POINTS
BUTTERFLY
VALVE
ROCK FILL
2- FOOT DIAMETER HOLE
~30-FEET DEEP
CAPPED
Figure 13: Typical gas migration barrier well
77
-------
and pressure gauge connection are located in accessible manholes. The
wells consist of slotted four-inch diameter polyvinyl chloride pipes
surrounded by coarse gravel. A 15-foot length of the unslotted section
is lamed lately below the ground surface and at least 10 feet of the
slotted section is above the groundwater table.
VACUUM PUMP
In order to produce .1 continuous relative negative pressure along
the barrier, a positive-displacement, lobe-type vacuum pump, capable
of pumping 250 scfro (the nearest available to design capacity of 196 ecfra)
of air-methane mixture at a vacuum of 10 inches of water column was
provided. The pump was installed on the northwest corner of the Aitnory,
and its suction pipe was connected to the northern leg of the harrier,
as shown on Figure 12. A variable-speed drive system, consisting of
a variable-type sheave, companion pulley and belt, was provided in order
to afford a capability for testing system performance undar different
flow conditions.
CAS MONITORING AND SIGNALLING SYSTEM
The purpose of the monitoring and signalling system is twofold:
(1) to provide the occupants of the National Guard Armory with an early
warning of intrusion of refuse-generated combustible gas into the space
between the Armory and the barrier and into the building and <2) to
determine the long-term effectiveness of the gas migration barrier.
Because failure in a coob-jstible gas barrier system could result
in a threat to human life and property, it is Imperative that the gas
control system be provided with a fail-safe monitoring and signalling
system designed to detect dangerous gas build-up in the building and
to warn the occupants. Data obtained from the long-term monitoring
program will supplement intensive test data which were obtained at the
beginning of the operation of the barrier system. Valuable additional
datat pertaining to the effectiveness of the selected system, can thus
bo acquired while at the same time ensuring the safety of the building's
occupants.
78
-------
Selected System Design Criteria and Characteristics
Components of the monitoring and signalling system must meet
certain functional requirements to satisfy the system objectives. These
components include the sampling network, the detection system and the
signalling system.
Sampling Network--
Sampling point locations were selected to ensure that gas will not
accumulate unnoticed in the Armory. The network consists of probes
located within the building and in the ground between the building
and the gas migration barrier. Probes within the Armory provide a direct
measurement of the parameter of concern, namely the presence of minute
amounts of combustible gas in the Armory. Probes located in the ground
between the Armory and the barrier provide a warning of the presence
of gas passing the barrier toward the Armory. Such an event will
allow Implementation of remedial measures prior to detection of any
explosive gas within the Armory.
Sampling points between the Armory and the gas migration barrier
are located in such a manner as to intercept the most probable pathways
for gas movement, namely midway between the gas extraction wells. The
sampling points within the ground are located adequately remote from
the Armory so that early warning of gas migration toward the Armory will
allow remedial action, if deemed necessary. Gas samples are withdrawn
at a depth which will provide an indication of maximum prevalent com-
bustible gas concentrations.
Cas sampling points within the Arnory arc installed in locations
where landfill gas might be expected to infiltrate or accumulate in the
building. Possible danger points are deformation cracks in the below-
grade structure of the building and unventilated high points in rooms
or confined spaces.
The selected sampling network consist:, of four interior sampling
points and six exteri^i sampling wells. Locations of sampling points
are shown in Figure 14. Choice of locations of sample probes was based
79
-------
NORTH-
LEG
ee
o
-PUMP
i
? ,'
9 '
WNT
NORTH
^ ANALYZER PANEL AND
S AUDIBLE ALARM
1 .
CAROLINA ^^
NATIONAL GUARD ^-^
ARMORY
0
BUILDING Q
^x
WEST
LEG
CONTINUES
EAST
LEG
CONTINUES
^
*
GAS
*
^
NOTE: PROBE NUMBERS DO NOT CONFORM TO
THE NUMBERING SYSTEM USED FOR
PROBES IN PRIOR SYSTEMS.
Figure 14: Sampling locations for detection and alarm system
-------
on the results of the sampling program conducted during four years
(1970 to 1974; and the predicted Impact of the control measures on gas
migration.
Interior gas sarpling points are located near the ceiling in the
supply room, vault, wool room and at the ground floor of the Police
Academy in the southwest portion of the Armory. The exterior sampling
veils will provide gas concentration data from 10 feet deep probes
located approximately h ilfway between the gas control barrier and the
Armory and spaced about midway between the gas extraction wells. This
configuration and sampling depth will ensure best interception of the
most probable pathways of gas migration, should any gases cross the
barrier and provide adequate warning of increasing p.as concentrations
before hazardous levels arc reached near any potential entry point.
Detection System--
The combustible gas analyzer selected for use In the pas monitoring
system is relatively simple and robust and of a type common to mining and
other gas detection applications. Extreme accuracy in analysis is
of less importance than dependability of service. The selected system
Is designed to function reliably with minimal attendance, because techni-
cal staff is not expected to be available for frequent maintenance or
testing of the equipment after the initial evaluation program.
Host of the available combustible gas detection devices have been
developed for the mining industry. They consist of an electrically heated
coil (sensor), the resistance of which varie: in proportion to the heat
of combustion generated by the burning of a gas mixture ignited by the
coll, and a device for measuring the coil resistance. Where several
locations are to be sampled and analyzed, it Is both economical and
convenient from the point of view of operation to centralize measurement
and recording. In the selected system, samples arc pumped from the
various probe sources to a single sensor located at a central monitor-
ing station.
81
-------
The detection system, includes a MSA Model F gas analyzer, a 10-
point recorder, sample pump and visual and audible alarms, mounted in a
console located within the Armory (Figure 14). A continuously pumped
sampling system with a centrally located gas sensor was considered pre-
ferential to individual sensors at every sampling veil because pimping
of the sample assures positive displacement of gas across the sensor.
Each exterior sampling well is provided with an access point for COD-
bustiblc gas injection 10 test the function of the system, whereby known
concentrations of test gas may be introduced into the well and monitored
at the gas monitoring console.
Samples are withdrawn continuously from each interior and exterior
campling point and monitored sequentially in a step-wise manner. Gas
from each interior sampling point is passed sequentially across the
probe for a three-ninutc period; thus the interval between routine
measurements of combustible gas from a particular point within the
Armory is 12 minutes. This sequence is interrupted every two hours
when gases withdrawn from the exterior sampling points are passed
sequentially across the probe [or a three-minute period. This lapse of
time between exterior samplings Is deemed adequate in light of the
results of the evaluation program presented in ChfDter VIII. Thus,
monitoring of interior samples is interrupted for a period of 18 minutes
while exterior sanples are being measured. Analyzed gas samples and
surplus by-pass gas are vented to the atmosphere above the roof of the
Armory.
Measurements of percent of lower explosive limit of combustible
gas are made and recorded on a 10-point recorder located in the monitoring
console.
Signalling Sygtem--
The signalling system was designed to warn immediately all occupants
of the Armory should explosive gas concentrations be approached within
the confines of the building. An alarm, audible in all parts of the
Armory, is provided, together with visual signals at the central monitoring
82
-------
station. In addition, visual signals arc provided at the central
measuring station if combustible gas is detected at any exterior sam-
pling probe.
The detection system is also provided with audible and visual alarms
to warn Anaory occupants of electrical or mechanical failure of any of
the system components.
The visual and audible alarms are activated at a pre-set combustible
gas concentration of 25 percent of the lower explosive limit (LEL). This
level Is well within the analyzer's capability for accurate analysis and
sufficiently below the LEL to rllow necessary action long before the LEL
is approached.
OPLRATLNC PROCEDURES
Based upon preliminary results of the evaluation program, a set
of operating procedures has been developed for use by field personnel
during the active life of the landfill and the beneficial occupancy of the
Armory. These procedures are divided into routine procedures, system tests
and maintenance. It is crucial to the safety of the Armory premises that
these procedures be followed regularly and precisely, and that any irregu-
larities be promptly and properly corrected.
Routine Procedures
The pump is to be operated continuously at the lowest speed practi-
cable (about 600 rpm) with the drive system provided. This speed of
rotation should provide the necessary protection with minimal equipment
wear. The regular operation of the system will, in the long tern, re-
sult In breakdowns and failure of parts. Replacement parts and repair
facilities must be readily available to minimize down time and prevent
the possibility of gas movement into the vicinity of the Armory during
these periods.
The recorder charts must be carefully reviewed by operating person-
nel at the time that they are changed. Any time that readings above zero
percent lower explosive limit are obtained, time, date and concentration
should be permanently recorded in a special book. This will allow ready
83
-------
reference and appropriate decision on the necessity for changing opera-
tional node of the system. Although pressure readings need not be ob-
tained as.part of the routine procedures, it is important that access to
and operation of the pressure probes (Ml through M10, Figure 15) be Main-
tained and their destruction be avoided. Should future testing become
necessary, existence of these probes will be of vital importance.
System Tests
Periodic tests of the integrity of the various system components
are necessary to minimize repair and down time as veil as to maximize
reliability of the monitoring and signalling systcn. For major equip-
ment, such as puop, motor, analyzer, recorder, etc., the instructions
specified by the manufacturers should be followed carefully and strictly.
Once every six months, test gas with a known concentration of
methane shouldtbe injected into the exterior sampling wells through the
access points provided, while the system is placed on manual control for
testingrsamples from .a particular well. The volume of injected gas
should be sufficient to fill the well casing and the volume immediately
adjacent to it, or at least enough to effec* a response at the monitoring
panel. .This quantity will be ascertained *-. field personnel through
experience. Response of the analyzer to these tests should be noted
in :the special monitoring book kept near the analyzer panel.
-After.every shutdown (for repair, replacements, etc.), the record
of the analyzer should be critically viewed for possible gas Intrusion.
System Maintenance
All manufacturers' recommended maintenance procedures must be
followed for pump, motor, analyzer, recorder and other equipment used
for .the Link Road landfill gas barrier system. Breakdowns and failures
oust be imraediajtely attended and repaired.
SAFETY FEATURES
There are a number of safety features Incorporated in the design,
operational procedures and contingency provisions of the gas migration
84
-------
barrier system. The design of the barrier involves gas-well spacing and
flow specification which together Include provisions amounting to a very
hlj»;> factor of safety. While this margin may appear excessive, it should
be noted that gas migration barrier design is still evolving and that
human life may be jeopardized without provision of excess capability.
The pump as well as the analyzer sampling units are designed to be ex-
plosion-proof, and excess combustible gases are vented to the atmosphere
at a level above the roof of the Armory.
Operational procedures, described above, also include ample safetv
features which, if followed faithfully, should prevent any possibility of
failure or danger to occupants.
Any mechanical system upon which safety of humans is based needs
back-up contingency provisions. The gas barrier Is no exception. There-
fore, a series of back-up safety provisions are available for implementa-
tion upon indication of need. These provisions—none of which will
probably ever be needed—are:
(1) If at any time the lowest speed of pump operation fails to
intercept gas movement across the barrier, pump speed can be increased
gradually until gas concentrations at the exterior probes (Probes 5
through 10, Figure 14) drop to zero.
(2) If provision (1) does not provide the necessary protection, all
butterfly valves on Wells 9 through 15 (western leg of the barrier, shown
on Figure 12} should be shut off. This action will almost double the flow
from the wells immediately surrounding the building and should inter-
cept any gas movement across the barrier.
(3) If both provisions (1) and (2) fail to provide protection, a
surface seal (asbestos-cement, concrete, asphalt, bentonite, plastic
sheeting, etc. as may be determined when and if the need arises) should be
placed to a width of 100 feet along the barrier. This provision will
prevent movement of atmospheric air along the barrier Into the wells.
The net effect will be the creation of a substantially greater vacuum
midway between wells and hence a far greater assurance of interception
of gases moving toward the building.
85
-------
(4) Ultimately, if all the above provisions fail, the vacuum pump
should be replaced with a larger pump with higher capacity. The veils
and headers are sized such that the pumpage rates can be increased to
as high as 10 to 20 times the rate which Is expected to be satisfactory.
Thus, there is no conceivable possibility that any provisions beyond
this step need be implemented.
86
-------
CHAPTER VIII
PERFORMANCE EVALUATION OF PROTOTYPE GAS BARRIER
INTRODUCTION
Protection of tmcan lite, property and the environment in and around
the facilities near che Link Road waste disposal site has been the prin-
cipal objective of constructing the gas migration barrier system. An
indisputable assurance of its reliable and safe operation—over extended
periods of rime—is necessary befoie there can be safe occupancy and
regular use of the buildings. A carefully planned, systematic and
rigorous testing program has been conducted, and the results thoroughly
examined and evaluated. These results were favorable, and indicate
that, if the system continues to operate at the tested level, safe
occupancy of the Armory peraises can be anticipated in the relatively
near future.
The primary objective of the testing and evaluation program was
to aid in the selection oE the most cost-effective mode of operation
for the barrier system. The test sequence was planned to permit evalua-
tion of each mode of operation independently and under similar conditions,
insofar as prevailing constraints allowed. The program was designed
to tfst and evaluate least-cost methods at the outset and to progress
to mare costly modes of operation only if the lower-cost methods failed
to provide the necessary protection.
A secondary objective was to accumulate data and experience for
better understanding of gas-flow ohenomena near solid waste land disposal
sites under the influence of the forced-ventilation gas barrier system.
87
-------
The information will aid in designing other gas barrier systems in
similar circumstances.
Measuring the zone of influence of individual wells and of the
whole barrier under actual operating conditions was a key element in
the program. This parameter helps determine the maximum well spacing
which assures pressure reduction along barriers in future systems. Other
parameters determined include noise levels at the pumps and flow dis-
tribution among the wells under different pumpage rates.
TEST EQUIPMENT AND INSTRUMENTATION
Some of the equipment used in the test program was similar to or
identical with equipment Installed for the continuous monitoring and
signalling system. In addition, gas concentrations were determined
using portable combustible gas analyzers. Soil-air pressures were
determined from probes placed near the barrier. Atmospheric pressure
was obtained from the meteorological station at the Greensboro, North
Carolina, Airport approximately 35 miles from the Armory.
Flow Meter
A cumulative flow meter was installed on the discharge manifold of
the vacuum pump. Cumulative flow over a known period of time was read
and recorded, including the date and time at which readings were
obtained.
Gas Probes
The ultimate purpose of the gas barrier is to reduce gas concentra-
tions in the immediate vicinity of the protected facilities sufficiently
below the lower explosive limit to assure their safe use. This para-
meter is a primary criterion for the barrier system's success under each
mode of operation. Measurements of explosive gas concentration can be
correlated with such independent variables as flew rate, duration of
flow and weather conditions for the purposes of future design considera-
tions .
88
-------
It is not sufficient to measure gas concentrations in terns of
percentage of the lower explosive limit. Concentrations may exceed LEL
at the beginning of barrier operation or when test flow rates are too
low to reduce all concentrations below the explosive limit. A portable
gas analyzer unit (J-W Gas Kit, Model HPK) was used to determine explosive
gas concentrations, in percent by volume. When gas concentrations in
probes not connected to the barrier monitoring and signalling system
fell below the lower explosive limit, readings in terms of percent lover
explosive limit were taken with the portable unit. This allowed more
sensitivity to variations of concentration at the lower ranges of con-
centration. The installed gas monitoring and signalling system (con-
sisting of a sampling system, combustible gas analyzer, programmer,
control and signal devices and a 10-point recorder) were used to deter-
mine gas concentrations (in terms of LEL) from the probe locations
serving the system. Locations of probes serving the iLonitoring and
signalling system, as well as those for supplemental evaluations, are
shown in Figure 15. Subterranean Probes 5 through 10 (see Figure 14)
consist of standard wo11 points driven 10 feet below grade, connected
permanently by copper sample lines to the gas analyzer. Access is pro-
vided through water-meter-type enclosures.
Pressure Probes
Maintenance of a ubiquitous negative pressure in the vicinity of
the barrier is an essential requirement; consequently, it is important
to verify the ubiquity of the negative pressure and to measure its varia-
tions. Relations between pressure distribution and pressure variation
in the soil and with system flow rate and atmospheric conditions help
describe the effects of soil properties and gas movement at the Link
Road site. Probes were installed to permit pressure determinations at
10- and 20-foot depths at locations shown in Figure 15.
Specially constructed U-tube manometers were used to measure sub-
terranean soil-air pressures at each pressure probe. A schematic drawing
of a typical pressure probe installation and manometer is shown in
89
-------
ANALYZER PANEL
GARAGE
FIRE LAB
JL GA$ EXTRACTION WELL
• GAS ANALYZER PROBE, EXTERIOR
0 GAS ANALYZER PROBE, INTERIC <
O GAS CONCENTRATION PROBE
A GAS PRESSURE PROBE
N
.29
IZS
Figure 15: Probe locations for gas barrier evaluation
-------
Figure 16. In this instance, manometer assemblies were fabricated locally
and permanently mounted on plyboard. The manometer assembly was mounted
on a cart for transport to each sampling point.
Vacuum Gauges and Taps
Each well is equipped with a vacuum gauge tap so that flow among
the barrier wells may be equalized (uniform vacuum) by use of butterfly
throttling valves. Consistent recording of vacuum readings and of valve
settings at each well are required to provide information for operational
and test regimes.
TEST PROCEDURES
Ease levels of gas concentration and pressures in the soils were
established before any pumping (even for equipment testing) took place.
These tests were perfonuod regularly for approximately one month prior
to the first activation and operation of the pump. The on-going gas
monitoring program continued uninterrupted and readings were obtained
on a regular, daily (5 day/wk) schedule, coinciding with gas concentra-
tion readings throughout the test period.
Mode 7; Low-Flow Pumpage without Surface Seal
The vacuum pump was operated at the lowest flow rate of 160 scfm
(corresponding to lowest available speed setting of 585 rpm) for about
two weeks. This period was not enough to produce an equilibrium pres-
sure distribution in the soils along the barrier; however, most para-
meters were approaching steady-state, as shown later under discussion
of results. During this period, gas concentration, pressure, flow rate
and cumulative volume of gases removed were recorded.
Modes II. Ill and IV; Varying Flow Pumpage without Surface Seal
After it was ascertained that the system was satisfactorily effective
at the minimum pumpage rate, pump speeds were increased for a similar
length of time. This sequence could be repeated, if necessary, by step
increases In the pump spec*' (745, 905 and 1,065 rpm) until the pump's
ultimate capacity Is reached. System flow rates at these pump (settings
91
-------
1/2 PLYBOARO
1/8" POLYPROPYLENE
TUBING
GLASS TUBING
VULVES
1/8" POLYPROPYLENE
TUBING
1/2 STONE
PERFORATED PROBE
1/2 STONE-
METER STICK
GLASS WOOL
-GLASS TUBING
COLORED WATER
(p= 100 g/ee)
MANOMETER
SOIL SURFACE
7?3ZV~
CEMENT SEAL
—CEMENT SEAL
PRESSURE
PROBE
\*—+\— 6" TO 8" DIAMETER
DRILLED HOLE
Figure I6: Pressure probe and manometer detail
92
-------
were 168, 210 and 250 scfm, respectively. Throughout each test sequence,
all parameters were measured and recorded.
Maintenance of a ubiquitous negative pressure and no passage of
landfill gas Co the protected side of the barrier (when the barrier is
operating ac low flow with no sealants) was used as a positive criterion
of successful performance.
Mode V; Contingency Tests for Surface Seal with Varying Flow
An ultimate back-up provision, the necessity for which appears
highly unlikely, is application of the designed surface seal along the
barrier. Application of the surface seal will make it possible to effect
confined flow conditions, with no flow of atmospheric gases into or
between wells along the barrier. Standard asphalt paving is considered
nost appropriate as a seal, bul application of a seal was not required
or tested in the evaluations conducted and reported herein.
The test progr.in for such a contingency condition was designed to
give the maximum amount of data for future design purposes as well as
to effectively evaluate this unique feature, not connon to existing
barriers. While the seal is being installed, pumping should continue
at the lowest design rate and complete data collection should be con-
ducted. A chronicle ot Che progress of seal applications should be kept
to correlate the time of seal placement with the reductions in gas con-
centrations or increases in negative pressure.
The seal can be placed in successively increasing widths and lengths
(e.g., exclude western leg), allowing sufficient tine between dimensional
Increments to permit equilibration of the flow and pressures. Incremental
seal application and concurrent lapses of Lime will be dependent on
available test results, equipment availability and logistics.
In the highly unlikely event that the low-flow and surface-seal
combination system proves unsatisfactory (under some severe combination
of conditions, certainly not expected at the Link Road site), high-flow
pumps may be installed and their performance tested in much the same
general pattern t's prescribed above for the low-flow ays ten. The success
93
-------
of high-flow ventilation barriers, has already been demonstrated else-
where (Reference-27).
NoiseSurvey
Noise levels resulting from operation of the vacuum pump equipment
Installed for gas evacuation were measured by R. J. Reynolds Tobacco
Company personnel on 15 Hay 1974 for the City of Wins ton-Salem. The
objectives of the survey were to determine if the vacuum pump produced
noise levels which would be annoying and generate objectionable complaints
and to determine if the noise levels during operation were hazardous to
human hearing. On chat basis, the procedure used was to measure the
noise levels produced by the vacuum pump and to measure the prevailing
daytime background noise levels with the vacuum pump off.
The following instruments were used during the noise survey: General
RaJlo Octave-Band Noise Analyzer, Type 1558-BP, Serial No. 2155; and
General Radio Piezoelectric Microphone Assembly, Type 1560-P6, Serial-
No. 1614.
The A-weighted sound level and octave-band sound pressure levels
were measured at two positions. The pump speed was 1,065 rpm (motor
speed 1,750 rpm). All measurements were taken at a height of four feet
above the ground. The first measurement was taken close to the pump.
The microphone was located three feet north of the vacuum pump gearbox.
The second measurement was taken 10 feet north of the vacuum punp gear-
box at the sidewalk. Kith the vacuum pump off, background noise levels
were alea measured three feet north of the vacuum pun? gearbox during
normal and peak traffic conditions on Silas Creek Parkway.
The A-weighted sound levels obtained were used to evaluate noise in-
duced hearing hazard. The A-weighting network has a frequency character-
istic similar to that of the human ear, i.e., both arc more sensitive
to middle frequencies than very high or very low frequencies. The. octave-
band frequency analysis, i.e.., the distribution of sound pressure level
in the nine preferred octave bands with center frequencies from 31.5 to
6(000 Hz provides additional information concerning the character of
94
-------
the noise and for estimating subjective annoyance effects such as
speech interference.
RESULTS
A large volume of data on gas concentrations and pressures was
generated during the course of system testing and evaluation, under
two modes of operation, lasting about three months. These data
firmly corroborate the theoretical computations which were used as
a basis for the conceptual design of the gas migration barrier.
Because most data are confirmatory, only representative samples of
data showing trends over time and space are presented In this section.
The remainder of the data are not graphlcallv presented in order to
maintain simplicity and prevent repetition and confusion of the pre-
sentation.
Records obtained from flow rate meters, gas concentration probes
and pressure probes, and those obtained less frequently from other instru-
ments were analyzed and evaluated promptly. Because many decisions re-
garding additional construction and operation of the barrier system had
to be based upon these results, prompt and proper analysis and careful
evaluation of data were conducted as the data were collected and as system
operation continued during the test period. Overall evaluation criteria
on each of the pararicturs are outlined below.
Combustible Gas Concentrations
The most Important parameter, from the point of view of human safety
from explosion hazard.1;, is concentration of explosive gases on the protected
side of the barrier. Early results, after Initiation of pumping, Indica-
ted a definitive trend, with gas concentrations in the protected areas
rapidly falling with Line. The trends continued ai all probe locations
with a firm trend toward zero readings throughout the monitoring period.
Trends of average combustible go., concentrations are shown on Figure 17
for shallow (10 feet) probes and deep (15 feet) probes on both sides of the
barrier. In order to maintain comparability of the two sets of data, only
95
-------
to
<
O
UJ
_l
CD
t-
«n
CD
2
O
O
u.
O
100
90
80
70
DAYS AFTER START OF PUMPING
-25 -20 -15 -10 -50 5 10 15 20 25 30 38 40 45
~ LEGEND
PROTECTED JIDE OF BARRIER
DEEP L° 115 ft° " """""
LANDFILL SIDE OF BARRIER
ond 17
prob«l2,l4,l5ondl8
j
z
UJ
O
O
u
UJ
cr
UJ
L11
20 25 90
MARCH
10 15 20 25 90
APRIL
DATE - 1974
10 15 20 25 30
MAY
Figure
17: Average concentration of combustible gas on either side of gas
migration barrier prior to ond during operational modes
-------
those data obtained from Probes 11, 13, 16 and 17 (Armory side) and 12, 14.
15 and 18 (landfill side) were utilized to obtain the averages plotted on
Figure 17. Other data present a similar picture. From probes located
inside the Buildings, readings of percent of the lower explosive limit
(i.e., percentage of five percent methane by volume in air), recorded
with automatic equipment, verc obtained and plotted as shown on Figure
18. Percent lower explosive limit (LED is a very sensitive and hip.hly
refined measure of safety or lack, of it in an occupied space. (For example,
25 percent of LEL, at which point alarm systems are activated, corresponds
to a methane concentration in air of only 1.25 percent by volume.)
Gas chromatographic analyses of combustible and other gases sampled
at the pump discharge and at selected probes provided additionil data
durlnc the evaluation period. Concentrations of oxygen, nitrogen, methane
and carbon d oxide in these samples were measured. Table 10 shows the
chroaatographic analysis for the pump discharge and Table 11 for other
selected probes from the deep (15 feet) access points. As expected, the
gas mixture pumped from the volume of soil on both sides of the barrier
was rich in methane and carbon dioxide at the beginning of the operation.
As pumping continued, evacuation of combustible gas from the protected
side of the barrier and admission of atmospheric- air from both sides caused
a gradual decline of CH. and CO. concentrations with a simultaneous in-
crease in 02 and N^ concentrations. A similar trend from t?as chromatograph
concentrations at the selected probes was not as evident during the brief
tine and limited extent of gas chromatographic analyses, in the early
period of barrier operation, but pronounced reduction in concentrations
at probes near tha barrier were verified by gas analyzer readings as
pumping continued. Concentrations at most probes had declined to or near
zero levels, with lowered gas concentrations persisting at a few deep
probes, primarily on the landfill side.
Flow from Individual Wells
A primary parameter in the operation of the barrier system at the
Link Road site is rate of flow of the gas mixture. It is imperative
that flow from Individual wells be maintained uniform along the entire
-------
100
390
$5
.80
O
5 TO
a:
S60
o
I50
v>
g40
330
0
oa
O 10
o
DAYS AFTER START OF PUMPING
•tO -5 0510620 30354045
r'5 '*P .'?. .P. m5T '° g. 2Q. 2S. ff 35. -40. i45
j\ I I I I I I njll M I I I I I I! II H I I I I I I H I I I I t II I I I I I I M t I I
PERCENT n
S 5,6,7,8,9,10 J\ j j
> ^ n. . _ t I • •
'IT Mf1
_/
[ PERCENT
^-P
PROBES 5
(EXTERIOR)
BACKGROUND
KI
\ i
-\-MOK I-
•MODE M-
NOTE - PROBE NUMBERS
REFER TO LOCATIONS
SHOWN ON FIGURE 14
I
\
'. n
H I
/-PROBES I,Z,3,<» I \[
f (INTERIOR) If
iJniilu ili. iiL^-Ji
\-fc-PROBE 5
PROBE 9
j^ PROBE 6
PROBES 1,2,3,4,
3,9,10
*
-ALL PROBES
• L,
nil
20 25 30
MARCH
25 30
10 IS 20
APRIL
DATE-1974
10 15 20 25 30
MAY
Figure 18: Percentage of lower explosive limit at individual probes in and
near buildings prior to and during barrier operational modes.
-------
Table 10. CAS CHROMATOCKAPHIC ANALYSIS
OF BARRIER PUMP DISCHARGE CASES
DATE
16
18
19
22
23
24
25
26
29
30
1
2
3
6
8
10
13
IS
17
Apr
Apr
Apr
Apr
Apr
Apr
Apr
Apr
Apr
Apr
May
May
May
May
May
May
May
May
May
74
74
74
74
74
74
74
74
74
74
74
74
74
74
74
74
74
74
74
Cos coiqiosition , Z by
°2 N2 CH4
7.
3.
7.
9.
7.
5.
5.
5.
3.
IS
12
6
4
4
0
6
1
5
8
1
.2
.5
10.7
21
20
18
19
8
9
13
.4
.4
.4
.6
.0
.5
44
43
47
48
51
54
54
50
61
63
62
73
78
76
76
75
68
67
74
.7
.4
.4
.5
.2
.3
.5
.0
.0
.0
.9
.6
24.
31.
20.
20.
19.
20.
19.
22.
14.
11.
7.
8.
4
1
3
6
8
6
8
8
4
2
5
1
Trace
.8
1.
2.
2.
11.
11.
7.
2
6
6
8
3
0
volume
C02
23
24
24
29
20
2C
20
22
20
8
17
10
.0
.7
.6
.5
.7
.0
.8
.2
.6
.8
.0
.0
Trace
1
2
2
11
11
6
.0
.4
.2
.1
.9
.7
99
-------
Table 11. GAS CHROMATOGRAPHIC ANALYSIS
OF SAMPLES FROM SELECTED PROBE LOCATIONS*
Gas composition, % by volume
DATE 0 N CB C0
Probe 11D
24
30
2
Apr
Apr
May
74
74
74
9.
2.
6.
2
8
3
30.
18.
35.
2
7
2
37
45
30
.0
.1
.5
22
33
28
.8
.0
.0
.Probe 12D
3 May 74
-6 May 74
4.1
5.1
10.6
20
53.1
43.8
31.3
30.7
Probe 24D
18 Apr 74
19 Apr 74
22 Apr 74
23 Apr 74
25 Apr 74
26 Apr 74
1 May 74
4.0
-5.2
.1.2
8.6
3.0
4.5
3.8
13.9
9.1
10.1
35.0
20.4
26.8
33.1
53.8
55.0
62.8
35.7
52.9
46.5
36.4
25.6
30.4
26.0
20..2
-24.7
71.6
26.1
*Probe .locations 12 and .24 are on the landfill side of Che barrier.
Probe .11 is in the gravel-surfaced parking lot, inside the barrier
and west of the .Armory. See Figure 15.
100
-------
barrier for the duration of its operation. An unusually high Clow In
a well usually indicates leakage of air or a relatively high-permeability
medium surrounding the well. In such cases, the valve at the well head
was partially or totally throttled so that the well would not interfere
with the operation of other wells. An unusually low flow may Indicate
a low permeability medium around the well. Such an observation was not
recorded during the conduct of the test procedure.
Direct flow measurement on each well head was not practical. How-
ever, an indirect, though accurate, measure of flrw was obtained from
relative vacuum readings. A portable vacuum gauge was used on quick-con-
nect nipples provided on each well head. Excessively low vacuum readings
were encountered on three wells (Wells 11, 12 and 13, Figure 12) located
in an area where higher soil permeabilities were suspected from earlier
soil tests. The rushing sound of air moving through these particular
well heads emphatically corroborated that excessive flows were taking
place from those wells. Therefore, in order to maintain as high a vacuum
as possible and to preserve the uniformity of flow from other wells, the
valves on the three wells were closed. While this modification may reduce
effectiveness of the barrier in the vicinity of the wells, flow at the
remaining wells is increased thereby. No potential for hazardous gas
accumulation existed nearby so further modification, such as application
of a surface seal, was not attempted.
Flow rates from the entire system were 160 and 250 scfn during op-
erational Modes I and II, respectively. These flows correspond to aver-
ag?. flows per well of 11 and 17 scfm, respectively. It should be noted
that the flow rates found satisfactory in the tests are far lover than
those reported for other existing forced-ventilation operating systems.
Pressure Distribution
All pressure measurements were made relative to the atmosphere.
It was found that atmospheric pressure variations were experienced almost
immediately at the depths manometer probes were Installed. In view of
the wide atmospheric pressure variations encountered, as shown In Figure
19, and because flow, in & given area, occurs in direct response to
101
-------
770
760
E
E
. 750
UJ
I
UJ
-------
gradients of relative pressure, use of absolute pressures was avoided.
Typical draw-down curves between wells were obtained before and during Che
test period as pumping progressed. Pressure distributions at 20-foot depth
between Wells 5 and 6 at selected points In time are shown on Figure 20.
It Is noteworthy that a ubiquitous negative pressure at all points along
the barrier was obtained immediately after r—>lng began at the 160-scfm
rate and that the negative pressures at all points between wells, partic-
ularly midway, continued to increase significantly, particularly at the
250-scfm rate.
The increases in negative pressure with time at the deep probe level
at the H6 through M10 stations at a point midway between wells and at
increasing distances from the line of wells are illustrated in Figures
21 and 22, respectively. The significance of the drop in pressure over
time is dramatically exhibited in a statistical test of the differences
between average pressure before pumping and that during each mode of
operation of the barrier system. The range, average, standard deviation
and statistical significance test of the differences for each mode of
operation throughout its duration are shown on Table 12. It is noted
that the influence of the barrier significantly extends to a distance
of at least 18 meters (about 60 feet) from the barrier in both operating
nodes (160 and 250 scfn).
Noise Levels
The Federal Occupational Safety and Health Act of 1970 (OSHA),
using the Walsh-Healy Public Contracts Act, paragraph 50.204.10, Safety
and Health Standards for Federal Supply Contracts, specifier the maximum
number of hours per day an employee may be exposed as a function of the
A-veighted sound level when measured with the A-network of a standard
sound level meter at slow response.
According to the regulation, the noise level limit is 90 dB(A) for
an eight hour per day exposure over a working lifetime. For noise levels
lower than 90 dB(A), longer exposure times are permitted. The converse
IB also true.
103
-------
MANOMETER STATION
M2 M3 M4 MS
WELL 6 MI
DAr —
MINUS IS
DAY ~~
MINUS 5 —
DAY I —
IO 5 0 5 10
DISTANCE FROM MIDPOINT BETWEEN WELLS, m
Figure 20: Drawdown curves between barrier wells
5 and 6 during barrier test period
104
-------
o
*
E
E
(E
O
CO
UJ
a:
a.
I
10
0
10
20
30
40
50
60
70
80
O
to
-90
-100
-25 -20 -
DAYS AFTER START OF PUMPING
15 -10 -5 0 5 10 15 20 25
MODE I
(iBOscfm)
30 3S 40 45
i
-MODE JT-
(25Oscfm)
x-SHALLOW PROBE (10 FEET)
ill
„ DEEP PROBE (20 FEET)
i
20 25 30
MARCH
10 15 20 25 30 5
APRIL
10 15 20 25 30
MAY
DATE - 1974
Figure 21: Relative pressure at two depths midpoint between barrier
wells 5 and 6 during barrier test period
-------
DAYS AFTER START OF PUMPING
25 30 36 40 45
20 25 30
MARCH
6 20
APRIL
30
DATE- 1974
NOTE: PRESSURE READINGS WERE OBTAINED FROM MANOMETER PROBE
LOCATIONS M6 THROUGH MIO (FIGURE 15), ESTABLISHED AT 6-
METER INTERVALS, IN A LINE PERPENDICULAR TO THE LINE OF
THE BARRIER AND EXTENDING INTO THE LANDFILL FROM A
POINT MIDWAY BETWEEN WELLS 5 AND 6
Figure 22'- Relative pressure at various distances
from barrier during the test period
106
-------
Table 12. STATISTICAL DATA ON COMPARATIVE EFFECTIVENESS OF SOIL-AIR 1'RESSURE
REDUCTION AT VARIOUS DISTANCES FROM BARRIER UNDER TWO OPERATIONAL MODES.'
M
S
static-.
Ml
MS
M7
MS
X9
SIO
Distance fros
barrier.
•s ft
0
6
12
IS
24
30
0
19.7
39. i
59.0
78.7
98.4
Frinsurc raise, — 1'2°
«.-.rt,-n.n'
+7 to -22
0 to -8
+12 to 0
+30 to 8
+214 to -18
+116 to -44
v«!e I
-2 to -72
-14 tc -47
0 to -25
0 to -3
+157 to -155
+33 to -17
"efe II '
-71 to -124
-43 to -97
-i: to -54
-9 to -47
+156 to -13
+105 to +7
"c.ir r-
lll^-OU-
-2
.4
5
6
95
27
e^jre. rr- )!20
1 vo-lc I
-42
-29
-3
-3
43
13
•Vue II
-97
-73
-2?
-25
+39
+40
St
"tee-«r«-.J
9
3
9
10
87
41
"o-'c I
12
8
b
3
79
10
*"ode II
17
15
13
12
50
23
-«:*:
-10.6*
•11. I1
-5.2ia
-8.40*
+l.T9b
+1.42b
"e'e II
•2C.95a
-16.S9"
-9.33*
-10.52*
-1.07b
-1.29b
Str-t!tca-.tly difCerc-t fron baekprojrd coriiltloiR at the °9 percent level of centlder.cc.
Net •Igntflcnntly different fron bnc*jrc-jrd conditions at tbc 99 pcrcc-.t level of confidence.
-------
By way of comparison, the A-weighted sound levele measured at three
and 10 feet from the vacuum pump gearbox operating at maximum speed (1,065
(1,065 rpra) were 86 dB(A) and 75 dB(A), respectively. Results of the
noise analysis are presented in Figures 23 and 24. Thus, operation of
the vacuum pump does not present a serious hazard to human hearing.
Since the pump operator is not required to spend a considerable length
of tide near the unit, personal ear protective devices, such as ear plup.s
and ear muffs, are not required.
A further evaluation of the pump noise levels at the sidewalk vas
made due to the possibility of annoyance when the vacuum pump noise
Interferes with speech intelligibility. The Preferred Speech Inter-
ference Level (PS1L) is a good guide to the interfering effect of an
intruding noise on speech.
The extent co which a steady continuous noise interferes with
speech communication depends upon the distance between the speaker and
listener at various levels of voice effort. The PSIL is the arithmetic
average of the sound pressure levels in the three octave bands, with
center frequencies at 500, 1,000 and 2,000 Hz. The interference levels
are for average male voices with speaker and listener facing eachn
other, using unexpected word material. It is assumed there are no
nearby reflecting surfaces that aid the speech sounds.
The PSIL of the noise produced by the vacuum pump at the sidewalk
is 70 dB. The average person would need to shout to Bake himself" under-
stood at a distance of six feet with the vacuum pump in operation. A
comparison of the PSIL produced by the vacuum pump (70 dB) and the PSIL
of the prevailing background noise in the area during normal traffic con-
ditions (54 dB) Indicates that the average person would be understood
at a distance up to about six feet using a normal conversational voice
level with the vacuum pump off, as shown in Figure 25.
108
-------
100
CJ
E
>
z
m
i
o
X
CD
•o
o:
a
o
o
o
I
CD
UJ
O
O
60
AT IO-FEET\
50
40
30
20
IOOOO
Figure 23! Frequency analysis of noise produced by
vacuum pump
109
-------
100
(M
E
-^
•z.
10
g
X
CJ
ffi
X3
UJ
cr
or
CL
o
I
o
o
100
• IOOO
FREQUENCY, Hr
10000 dB(A)
Figure 24: Frequency analysis of background noise measured
three feet from vacuum pump gearbox
no
-------
IOO
CJ
E
10
ID
•o
UJ
-I
OT
V)
UJ
8
10
CD
1
zo
10000 PSIL
Frequency, Hz
Figure 25' Comporison of vacuum pump noise levels measured at
10 feet to the background noise levels measured during
normal traffic conditions
111
-------
EVALUATION OF RESULTS
Safety for Occupancy
The combustible gas concentration profiles, presented ii. Figures 17
and 18. show conclusively that the system is providing and maintaining gas
concentrations at or near zero inside the barrier and In occupied spaces
under the least Intensive operating mode. If careful monitoring confirms
that these conditions continue to obtain over a period of months, the
Armory should then be free from explosion hazards and suitable for normal
occupancy and use. That adequate protection is apparently provided
Is best evidenced by the plot of percent lower explosive Unit at all
inside monitoring probe locations before and after the start of the
system, as shown on Figure 18. Probes loceted inside the Armory (Probes
1, 2, 3 and 4) consistently recorded zero LEL readings throughout the
test period. Whereas combustible gas concentrations were uniformly above
100 percent LEL before the start of the system at all probes outside the
Armory, they dropped precipitously upon start of the pumping process.
The sharp, but ephemeral, resurgence of gas concemrations at two outside
probes (Probes 5 and 9) after the initial drop, is Attributed to the
arrival of gases that had previously moved under and beyond'the Aratory.
It is expected that after all the explosive gas on the protected side
is extracted there will be no more source of supply of gas. north of the
barrier, to give rise to any further such peaks. With the pumpage rates
of 160 to 250 scfia used, it Is estimated that complete removal was accom-
plished in the duration of the test period.
A comparison of average combustible gas concentrations on both sides
of the barrier at two depths, as shown in Figure 17, is revealing. Whfle
combustible gas concentrations at both depths (10 and 15 feet from the sur-
face) declined to zero on the protected side of the barrier, those on
the landfill side declined to values above zero. This trend is wholly
expected due to the continuous generation of combustible gas in the land-
fill. The decline in concentration is due to admixture of progressively
greater anouni.s of air entering from the surface. This phenomenon also
explains the faster rate of decline of combustible gas concentration from
the shallower probes than from the deeper ones on both sides of the
112
-------
barrier. Afcer several months, when equilibriun conditions are fully
established, the average concentration of cocibustiblc gas froa the probes
on the landfill side of the barrier—at any given punpage rate—nay be
used as an index to the approximate gas generation rates in the fill.
Barrier Effectivcnesa
While gas concentration data help establish the effective operation of
the system, pressure distributions reveal the degree of success of the
barrier, its extent of influence and the need, if any, for its improvement.
The "draw-down" curves in Figure 20 show the gradual lowering of pressure
(increasing negative pressures) at all points between wells. While a
minimal pressure drop of a few millimeters of water at the midpoint be-
tween wells would be sufficient Lo ascertain protection, i.egative pressures
of above 100 nllllmetcrr. water obtained under Mode II operation of the
barrier provide a very high factor of safety. The pressure decline o'er
the test period at the critical midpoint between wells is illustrated
on Figure 21. The high negative pressures obtained virtually guarantee
existence of ubiquitous negative pressures throughout the length and dcoth
of the construction barrier.
Extent of Influence of Barrier
A very significant question in the design of new barrier systems
is the radius of Influence oz wells along the barrier. Thij parameter
helps establish maximum spacing of wells, flow rates to be employed and.
In some cases, the need for a surface seal along the barrier. The extent
of influence of the barrier can be readily visualized from Figure 22 in
which declines in pressure are illustrated as a function of time from
Che start of operations and distance from the barrier. Although only
the pressure readings at the deeper (20 feet) probe locations are shown,
the trends arc the same for the shallow probes, albeit, more pronounced.
It is evident from Figure 22 and from statistical analysis presented in
Table 12 that the influence of the barrier extends to at least Station
M8, approximately 60 feet from the barrier. Thus, it may be rurmised
thac under existing flow conditions, a well spacing greater than the
113
-------
existing 67-foot interval nay be permissible. Conversely, it is evident
that lover flow rates can be effective in providing the necessary pro-
tection vith the existing spacing.
Noise Levels
The pump motor does not present a hearing hazard and does not con-
tribute any additional neighborhood noise at the property line, located
more than 50 iset from the pump. There is some noise at the sidewalk
directly in front of the pump. However, this should not present a public
hazard because people are not expected to be working for Ions periods
In that area. The only person who would be clc e to the pump at any time
1 the operator. That person is not required to spend a considerable
length of time near the equipment during normal operation.
Fover Requireman c and Cos t
The major component of the barrier system requiring power is the
motor driving the vacuum pump via the variable-speed sheave and pulley
system. Power requirement for the pump was computed directly, using
actual voltage and current measurements during operation at the two extremes
of the pinup speed range. Power requirement for the gas analyzer sampling
pump is relatively minor. It was calculated using a similar procedure.
Unit power cost at the time of these computations was $0.0251/kw-Jhr.
The power required and its cost at the two speed extremes are presented
in Table 13.
Table 13. POWER REQUIREMENTS AND COSTS FOR GAS MIGRATION BARRIER SYSTEM
Unit
Hot or
Analyzer
Total
700
Power,
kw
3.229
0.918
4.147
Pump spee-J,
Cost,
S/vr
697
198
895
rpm
1050
Power,
kw
4. 835
0.918
5.753
Cost,
S/vr
1,044
198
1,242
114
-------
Barrier System Cost
The total construction cost for the barrier system was $88,100,
which consisted of $64,400 for the barrier and $23,700 for the monitoring
equipment.
The itemized costs for various elements of the gas barrier system,
as constructed, are presented in Table 14, which also indicates the
corresponding estimated costs used in the evaluation of alternatives.
Table 14. COMPARISON OF ACTUAL AND ESTIMATED CONSTRUCTION COSTS OF
GAS BARRIER SYSTEM
Element
Gas wells
Manholes
Suction header, 8 in.
Vacuum pump
Monitoring system
Total barrier
Cost,
Actual
19,800
7,380
24,970
12,250
(23,700)
64,400
dollars
Estimated
17,000
6,100
8,900
8,000
-
40,000
The greatest discrepancy between actual and estimated costs is the
much higher (more than twofold) actual cost of the eight-inch dia-
meter polyvinyl chloride suction header. The higher actual costs in
all other elements of the system can be attributed to the high rate
of inflation that occurred between the time of estimation and the time
of procurement and construction.
115
-------
CHAPTER IX
TECHNIQUES FOR EVALUATING SOLID WASTE LANDFILLS
WITH REGARD TO CAS MIGRATION
MINIMIZING ADVERSE EFFECTS OF GAS PRODUCTION AND MIGRATION
Decomposition gases are a normal, expected consequence of land dis-
posal of solid wastes. Sanitary landfill procedures will necessarily
establish anaerobic conditions; consequently, provisions should be made
to control the migration of the gases, particularly methane. Certain
precautions or provisions can be taken to minimize adverse effects by
proper site selection and by the design, construction and operation of
the sanitary landfill. In general, open burning dumps pose lesser pro-
blems from gas production and migration then do sanitary landfills since
dumps usually afford ample ventilation and organic content is reduced by
burning. However, any buried refuse whether in A dump or landfill nay
produce gases which may migrate to and accumulate in structures.
Site Selection
To minimize gas production from landfilled wastes, they should be
kept as dry as possible. Consequently, the site for a sanitary Landfill
should be selected so as to minimize entry of surface water or grounrtwater.
Locations with high elevation, low water table and adequate natural sur-
face drainage features are preferable. Sites should be selected so as
to be least affected by nearby development which would increase surface
runoff onto the landfill. To minimize effects of migrating gases, sites
should be chosen which are remote from nearby existing structures or froa
likely future development on nearby land. The determination of proper
116
-------
separation between landfill sites and nearby development should be based
on engineering evaluations, including gas permeability of soils. Sites
in areas designated for open-space, recreational or similar nonconstructlon
uses are generally perferable, although adequate barriers can be constructed
to permit on-site or nearby development in some cases. Sites should be
selected so that the soil cover obtained on site will be sufficiently
permeable to allow venting of gases through the cover, yet not so permeable
that gases will easily migrate from the landfill periphery. Sandy soils
and those with voids and rocks, etc. are likely to have high permeability
for gas. Clay and similar soils would have low permeability. In general,
sites should be selected which require a minimal amount of construction
(such as surface drainage, barriers for gas controls, etc.) in order to
minimize costs.
Landfill Design
A number of design features are important for the control of gases.
If natural surface-drainage features are minimal, the landfill design
should include a permanent surface drainage system to divert surface water
away from the landfill. If groundwater levels are high, a membrane or
other Impermeable barrier should be required to exclude groundwater,
or wastes should be deposited above ground. The landfill's final grade
should be at least one percent to drain surface water without erosion
of landfill soil cover. Landfill-edge slopes should also be protected
against erosion. The operational plan should require adequate daily
compaction and covering of incoming wastes so that voids are reduced and
water entry during deposition is minimized. The area of solid wastes
exposed to precipitation should be minimized by limiting the size of the
working face. The design may also include installation of gravel-filled
trenches, membranes or other permeable barriers or areas of undisturbed
soil between waste deposits as natural barriers or vents (depending on
soil permeability). The design should also stipulate the ultimate antic-
ipated use of the completed landfill. If construction is to be pemitted
on site or nearby, developers should follow construction precautions to
minimize potential for gas accumulation. Undisturbed areas of soil nay
117
-------
be left for future building sites, utility conduits and roads. If soil
permeability permits extensive lateral gas migration, it may be necessary
to leave undisturbed an extensive peripheral area around each area where
wastes are buried, within which building will be prohibited or restricted.
Landfill Construction and Operation
It Is important that design and operational plans be closely follow.-4
during operation r; the landfill and that any deviations therefrom be
undertaken with caution and be recorded. In developing the landfill, it
is important to Dininize the proportion of the site which will be subject
to receiving precipitation runoff and to conduct operations so that proper
grade and cover will be restored as soon as possible. Natural cover and
beneficial runoff characteristics should be retained as long as possible.
Every effort should be nade to insure prompt, maximum contraction and
covering of the wastes within a minimal space. The highest and dryest
portion of the site should be reserved for the placement of liquid and
highly organic wastes, and they should be disposed so that surface drainage
will be away from the main body of the deposited wastes. Absorption
of liquids by spreading such wastes over the main body of drier wastes
will serve only to accelerate gas production therein, over a wider area.
Records should be kept on the location and character of special-waste
deposition areas such as animal carcasses, sludges, etc., so that unusual
gas-migration suspect areas will be demarcated. Such special-waste areas
should be located in those parts of the landfill where on-slte or nearby
construction is least likely to occur. Such areas may be enclosed by
impermeable or other gas migration barriers. Use of water for dust con-
trol should be kept to a minimum, and water should not be added to the
wastes to aid in compaction. Pre-construct!on of trenches and stock-
piling of cover materials should be minimized in order to reduce ponding
of water and interference with proper drainage. Portable pumps should
be used to drain ponded water rapidly, and finished areas of the landfill
should be regraded and additional cover applied as necessary to eliminate
ponding due to settlement and to restore cover integrity, finished areas
118
-------
of the landfill should be revegetated as soon as possible in order to
restore proper runoff, retain cover and stabilize slopes.
Ultimate Use
Development of the completed site should follow the designed plan.
The finished site should be recorded in property records with a proper
description of Its conversion from prior condition to finished landfill,
its intended use, zoning and use restrictions. Until ultimate-use plans
are implemented, the finished site should be maintained so as to minimize
water entry. Such maintenance will possibly Include regrading, filling
or revegetation of the site and the maintenance of surface or subsurface
drainage.
GAS MONITORING PRACTICE
It has not been common practice to design or install systems at
solid waste land disposal sites to detect or to measure the concentration
of decomposition gases escaping from the surface of the landfill or
migrating into adjacent areas. It has been shown that for landfills where
initial or subsequent moisture in the wastes is limited, production and
migration of gases will be relatively low and the need for monitoring
gas migration will be minimal. Many landfills, particularly those con-
structed wholly or partially above original ground level, will have one
or more edges, plus the landfill surface, exposed to the atmosphere,
and monitoring will be required only where migration through ad.lacent
soils will likely occur.
NEED FOR GAS MONITORING
For landfills where peripheral venting systems or barriers (active
or passive) have been installed, monitoring needs will be minimal, and
consist primarily of checking the effectiveness of the venting system
or barriers. The principal need for monitoring will be at those landfills
where:
(1) field saturation of the landfilled wastes occurs or is expected
to occur at any time,
119
-------
(2) subterranean utilities arc installed nearby,
(3) buildings or other structures are constructed on or near the
landfill (dependent on gas permeability of the soil),
(4) permit requirements or regulations require monitoring for
migrating gases, or
(5) the landfill surface cover permeability is reduced by paving
or other sealants, either during operation or following land-
fill closure.
Monitoring would not generally be required for landfills where
ultimate use will be for open space, agriculture or other uses where
utilities or enclosed structures will not be required. Suitable pre-
cautions should be taken, however, when and if utilities or structures
are installed and in any instance where site-use plans are modified to
uses which require on-site or nearby construction.
Tho&e landfills whose cover materials are of high permeability will
likely have high permeability soils adjacent, If the cover material was
obtained on site. Nevertheless those landfills where adjacent soils are
highly permeable to gas flow are most in need of monitoring. Even
though the landfill will Likely vent gases rapidly through the permeable
cover, surface water may easily infiltrate the cover to expedite p.as
production. Periodic saturation of the soil cover will at least period-
ically reduce cover permeability for brief periods, possibly resulting
In brief subsequent episodes of lateral gas migration. Consequently, Ras
monitoring will likely be most important following heavy precipitation
episodes.
In those instances where landfills have highly permeable soil cover
but relatively impermeable surrounding soils cr impermeable barriers,
there will be minimal need for gas monitoring.
Xt has been noted that landfills will produce decomposition gases,
particularly methane, for 30 to 40 years. Also if conditions change,
such as introduction of groundwater or surface water, so that moisture
increases occur in a formerly dry landfill, gas production may be
120
-------
reinitiated. Consequently, it cannot be safely assumed that a vintage
landfill will no longer produce gas, and some form of monitoring or sur-
veillance may be required indefinitely.
PHYSICAL OBSERVATIONS
The operator or manager of a solid waste land disposal facility, or
those examining a site with regard to ultimate use or nearby construction
can, through observation of some physical features, sometimes gauge the
potential for gas migration or detect signs of gas migration occurring.
Recording of such observations during the active lifetine of a sanitarv
landfill will provide some guidance to future development of gas migra-
tion problems. Physical observations, performed without dependence en
expensive instruments, may serve as an early warning of potential problems
from gas migration which can be followed up with instrumented surveil-
lance. Early warning signs of potential problems include observation
of unusual settlement, effects on vegetation, occurrence of gas in sewers
or other subterranean structures (which may require Instrumented sur-
veillance), odors or gas bubbles in water accumulations. Observations
on these phenomena at the Link Koad facility, as related to gas production
and migration, are noted in the following sections.
Settlement
Whereas the surface of the Link Road landfill is generally flat,
a number of random undulations occur throughout, indicating that local
settlement has occurred. Over most of the landfill, the settlement
observed was similar to that occurring in most landfills of its type (i.e.,
moderate but not extensive). Two specific areas of extensive and dramatic
settlement were in the area just back of the Armory where wastes were
deposited along a steep natural bank and in the fire training area. The
peripheral road is undulating, indicative of settlement throughout. Just
above the lower peripheral road, large depressions were overgrown with
cattails (a water-loving plant) and after rains, all depressions held
water for long periods of time. Similar cattail and water-filled areas
were noted just north of the concrete slab back of the Arnory. Extensive
121
-------
cracking of the soil cover was noted in the fire training area and in
a line paralleling the bank crest back of the Armory, indicative of
BetClement and local infiltration of surface water.
After nev cover was applied in the summer of 1973, it was noted that
rapid settlement was continuing back of the Armory and that a crack was
re-established along the bank crest.
Settlement plates placed throughout the fill were lost when the new
cover was placed, and only a minimal amount of settlement was noted for
the few months they were in place. The landfill is settling slowly in
general, but rapidly in a few locations. The settlement areas are gener-
ally characterized by relatively high comb.-stible gas concentrations.
The rate of settlement immediately prior to the Armory explosion and the
effect of settlement in increasing gas migration at that time could not
be established. Obviously, the rate of settlement in the fire training
area and in the area back of the Armory is much faster than nearby areas
of the landfill.
As at Link Road, rapid-settlement areas in most landfills are often
those where the highest gas concentrations occur. Where settlement is
the result of rapid decomposition of wastes in the immediate acea, rapid
gas evolution is to be expected and its migration through nearby soils
results both froia internal pressures caused by shifting oC the mass and
from increasing gas pressure. Additionally, channels may be created
through disruption of cell integrity, soil-cover displacement and erosion
processes by surface water entry through which gas escapes readily. These
processes arc likely all combined in the Link Road landfill's fire train-
ing area.
In the settlement area back of the Armory, the deposited wastes are
not likely to produce gas at as high a rate as those in the fire training
area because the wastes placed near ihc Armory were less putrescible.
Consequently, the settlement is likely due to physical causes, primarily
the sliding of the wastes down the natural slope, hastened by the entry
of water from ponding and cracking and the weight of additional cover
material. From these observations, it can be confirmed that settlement
122
-------
areas in landfills arc likely locations where gas is rapidly being gen-
erated or gas migration Is being hastened and abetted by physical pro-
cesses. In cither situation, observations of settlement are a means of
discerning locations of high gas migration potential for further observa-
tion.
Visual features associated with settlement areas include, in addition
to the differences in elevation, which may not be apparent for moderate
settlement, specific types of vegetation and surface cracking. Water-
loving vegetation such as cattails nay grow in settlement areas which
frequently hold water. In grassy areas, grass docs not grow in the
Immediate vicinity of the crack so that cracks in the soil cover are
readily visible when the grass is short and dormant but are less visible
when the grass is long or growing vigorously.
Vegetation
A preliminary assessment was made at the Link Road landfill of che
potential for visual effects of gas damage to vegetation serving as a
monitoring aid for detection of gas migration. A report of this assess-
ment is included in Appendix B.
Vegetative growth at the Link Road landfill was heterogeneous but
had no apparent correlation with areas of high or low gas concentration
(neither was there a discemable pattern of gas concentration at the Land-
fill surface indicating consistently hif.h and low j>as concentration
areas). Vegetation at other landfills in the area was either too sparse
or too recently established to permit meaningful observations.
Areas peripheral to the Link Road landfill where gas data were avail-
able were not vegetated, being covered with either asphalt or crushed stone.
The one area where visible effects on vegetation were noted was in
the fire training area. It appears that in the immediate vicinity (one
to two inches) of a crack from which pas is ernerRing, vegetation does
noi grow. This effect may be due to the drying effect of the gases In
the grass1 root zone. On either side of the bare area along the cracks,
the grass appeared to be growing well, with no gradation in color or
123
-------
growth rate being observable. Grass (fescue) at the Link Road landfill
is apparently not a sensitive vegetative indicator of gas migration.
Nevertheless, plantings of grass on the perimeter of a landfill and into
adjoining areas might aid in increasing the visibility of any cracks or
openings in the surface from which gas may be emerging.
In some areas, it has been reported that certain ornamental plants
have been killed by gas migrating from landfills and that it is difficult
at some landfills to maintain grass cover or plantings because of gas.
The potential for gross observation of effects on vegetation to serve
as an indicator of gas migration remains undefined.
Other
It has been noted in the literature that at some landfills, bubbles
of emerging gas can be noted in surface-ponded areas after a rain. This
phenomenon was noted by one observer at the Link Road landfill in the
ponded water filling settlement depressions in the fire training area.
In general, occurrence of a local strong concentration of odor might
be taken as an indicator of gas migration, but it would likely be difficult
to distinguish the precise location or to gauge the concentration. Methane
is odorless and could not be detected in this manner. Other gas
-------
INSTRUMENTED SURVEILLANCE
Gas concentrations at the Link Road landfill were determined through
Instrumented surveillance, taking gas samples fron installed probes.
A review of these procedures and findings follows.
Gas Probe Layouts
A series of probes were installed at the Link Road landfill and at
two other landfills in order to determine gas concentrations. The probes
were of the Los Angeles type (Reference 23), consisting of a short length
of perforated plastic tubing attached to snail-diameter plastic tubing.
The probes were placed at various depths (1, 10 and 25 feet, normally) in
six-Inch diameter holes at various sampling locations, and the holes were
back-filled with soil 'between probe placements) to the ground surface.
Each probe tubing lead was marked according to probe depth. For the
earlier studies, probes were established around the Arnorv. Later, addi-
tional sampling locations vere installed on a grid patter, extending
several hundred feet outward in the landfill. In many of these locations,
water uas encountered within two to three feet of the landfill surface.
Upon installation of the gas barrier, additional sampling locations were
established on either side of the barrier. In addition, well-point-type
probes were established inside the barrier and connected to the gas mon-
itor. Four occupied-space interior probes were connected to the monitor
installed inside the Armory. The various probes and their locations are
described in Figures 5, 6, 14 and 15.
In order to determine pressure differentials in the soils along the
barrier and between wells, 10 sampling locations (Figure 15) were estab-
lished along a portion of the barrier back of the Armory supply room.
Probes were established at two depths at each location, and pressure
readings were taken by use of a mobile manometer (Figure It).
Instrument Use
Two portable conbustibl gas detectors (J-W Gas Kit, Model HPK) were
used to determine gas concentrations fron the probes at each sampling
125
-------
point. This Instrument Is calibrated to show combustible gas concen-
trations In percent by volume in air. In use, the operator Inserts
the instrument's metal probe into the probe tubing at the sampling point
and permits the Instrument's vacuum pump to evacuate the tubing until
a proper sample Is obtained. The steady-state concentration reading is
recorded for each probe and sampling station. During the studies., concen-
trations were .determined daily in sone instances, and at weekly intervals
on others. The instrument was calibrated against standard gases and was
checked for "zero" (acmospheric) readings frequently.
In addition to the gas concentration data obtained by use of the
portable gas detector, samples were also analyzed by gas chronatography.
Samples were obtained from the probes by using evacuated flasks connected
to the probe tubing, following evacuation of the tubing with the portable
instrument's vacuum pump.
SIGNIFICANCE OF FINDINGS
Sampling Pattern
The distance between sampling stations ranged from about 10 to
over 100 feet. The original aeries was dispersed in accordance with
the need to establish gas concentration profiles around the Armory, with
closest spacing nearest the Armory supply room. Outlying sampling
points were established to determine concentrations in presumed high gas
areas (such as those in the fire training area) or to determine the
boundaries or -extent of gas migration potential. Shallow probes of one-
foot depth were established to determine if gas was emerging through the
soil surface, whereas the deep probes were established at the approximate
level equivalent to the base of the landfill.
In general, the concentrations at the lowest probe of a sample loca-
tion were highest while Chose at mid-depth were somewhat lower. Concen-
trations at the one-foot probes were erratic and are considered unreliable.
Gas Migration
Typical combustible gas concentration profiles arc shown in Figures
6. 17 and 18.
126
-------
Cas concentrations in the landfill were generally only slightly
higher than those in probes between the landfill and the Armory,
indicative of the relative ease with which gas migrates through the
peripheral soil. Concentrations above the LEL were noted in the parking
lot northeast o£ the Armory at considerable distance from the landfill;
but north of the Armory, concentrations vere at zero levels.
The findings indicate that gas migrates readily to the Armory am!
at considerable distance from the main landfill across the former incin-
erator-residue area. Where the gas apparently encounters undisturbed
soil, as at the northern boundary of the fire training area, the undis-
turbed soils near the front of the Ar..iory and the parking area southeast
of the Anaory, it is more rapidly attenuated. These areas, however, are
at considerable distance from the landfill boundary, and attenuation nay
be attributed as much to distance as to decreasing soil permeability.
The gas samples consisted primarily of methane, as the gas chromato-
graphic analyses attests (Tables 10 and .'!).
Cas production and migration from the Link Road landfill continues
year-round at a singlficant rate, though it is steadily declining. At
the present rate, significant gas production may continue 10 years or
more. Seasonal variations in gas production and migration have been
noted, with peaks occurring in February or March and declining through the
Bumaer months to lows in mid-winter. Combustible gas concentrations In
the 40-percent range are still occurring near the landfill boundary, and
explosive-concentration accumulations arc a potential for years to cone.
There appears to be a short-range increase in gas migration occurring
after rainfall episodes; consequently, there is a need to be concerned
with gas migration problems at this site on a relatively continuous basis.
The Link Road landfill seems to have the following characteristics,
with regard to gas migration and explosive hazards:
127
-------
(1) the landfill is largely saturated and is producing methane gas
steadily with both seasonal peaks and periodic increases
following precipitation episodes;
(2) gas migrates readily through the soil to nearby structures,
due to high gas permeability of the intervening soil;
(3) major structures are in close proximity to the landfill or
are built on site and until recently have had no means to
prevent gas entry;
(4) major settleaent is continuing to occur in areas of the
JLandfill closest to stuctures and permeable soil strata; and
(5) -gas migration appears to be attenuated below the LEL within
200 feet of the landfill boundary when passing through un-
disturbed soil.
It appears likely that despite the high production of gas at this
site and the relatively permeable nature of the soil, normal building
construction could have been followed for structures built on natural
.ground 100 to 200 feet from the landfill boundaries (though not on the
incinerator-residue portion).
GAS DETECTION
Observations in Subterranean Structures
In retrospect, the occurrence of the flash fires in sewer drains
•and roof gutters could have served as adv.need warnings of gas migration.
In fact, the fires in the sewers of the fixe training area were assumed
to be associated with the entry of landfill gas. Occurrence of gas in
sanitary sewers, as noted in a later investigation, is less definitive.
Rapid settlement in both heavy-settlement areas was noted, but its possible
relation to lateral gas migration was not suspected. There were no ob-
served effects on vegetation indicative of gas migration or of gas bubbles
In ponded water.
128
-------
Physical Phenomena
At this site, physical phenomena related to gas migration were
either not of sufficient magnitude, could not be properly observed or
were associated with factors net related to gas migration to the Armory.
Instrumented Surveillance at Shallow Depths
It has been suggested that sampling for gas with a portable instru-
ment at shallow (1 foot) depths in the soil adjacent to a landfill boundary
could serve as an early warning of gas migration. The erratic nature
of data fron the one-foot probes in the initial gas sampling network
indicates that this is an unreliable procedure per se. It was observed
Chat any sampling from an unclosed system wherein air may be entrained
with the gas sample gives erratic results. Any such system should be
based on the sampling of permanently installed probes installed at least
three feet below the surface.
Sampling Patterns
The placement of sampling stations to monitor gas migration should
be based on some knowledge of the soil permeability. If gas pemeability
tests of peripheral soils indicate low permeability, it would probably
suffice to place sampling stations at 100-foot intervals along and 25
to 50 feet from the landfill edge. At landfills adjoined by highly per-
meable soils, stations should be at 50-foot intervals or less with a second
line of stations 100 feet from the landfill edge. The nature and extent
of the pattern will depend in large measure on the gas permeability of
the soil and proximity of the facilities to be protected. Intervals
suggested herein are admittedly arbitrary. At landfills where peripheral
soils are relatively impermeable, it might only be necessary to install
monitoring probe networks in the near vicinity of areas of the landfill
where high gas production is anticipated (i.e., wet-waste disposal, sludpe
disposal, animal carcass disposal, etc.). In general, where the likelihood
of gas migration is greatest and the resultant hazards are most significant,
the gas monitoring network should be of greater sophistication.
129
-------
ROUTINE OBSERVATIONS
Relatively simple observations should be required of landfill oper-
ators- to provide early warning of potential gas migration. These include
recording:
(1): occurrence of rapid or extensive settlement;
(2) occurrence of extensive cracking in landfill cover or nearby
soils;
(3) occurrence of vegetation browning or die-off in peripheral
areas;.
(4) concentration of odors;
(5).- occurrence of. flash fires or minor flares on or off site; and
(60- occurrence of bubbles in ponded water on'landfill periphery.
INSTRUMENTED OBSERVATIONS AND RECORDS
If- any of the crude observations or a combination thereof indicates
that gas migration may be occurring, landfill supervisors should institute
surveillance based*on use of portable gas detectors and relatively shallow,
but' per&anent&y installed, probes. This should include:
(1) installation of a.suitable probe network in the suspect areas,
probes'being established at least three feet deep; and
(2) sampling and recording combustible gas, using a portable gas
detector, according to the following schedule:
(a)S weekly, during spring and early summer;
(b)) daily follovlr .teavy precipitation; and
(c)3 at regular two-week or monthly intervals during winter
and fall months.
FOLLOH^-UP GAS, MONITORING
If concentrations are found near the explosive range, additional
probes, placed at 10- to 15-foot depths will aid in determining the extent
130
-------
and range of gas migration. The range of probe placement should be ex-
tended if gas concentrations continue to occur at the outer edge, in order
to determine extent of migration. If levels continue or are in the high
range, it will be necessary to consider barrier Installation or prohibi-
tion of construction within the near vicinity of the landfill.
GAS MONITORING INSTRUMENTS
Relatively simple, low-cost portable gas detectors are available
which indicate presence of combustible gas and are calibrated to read
gas concentrations in percent of the lower explosive limit (LEL). Gas
readings near or in excess of the LEL (52 by volume of methane in air)
would indicate the possible necessity of follow-up monitoring, using a
nore expensive portable Instrument calibrated to read higher combustible
gas concentrations directly in percent by volume.
Instruments of this type are commonly used by utility companies, pub-
lic works departments and fire departments. Arrangements could probably
be made with such agencies for routine nonitoring and reporting of pas
concentrations in subterranean facilities near landfills.
Gas samples can be taken from inexpensive subsurface probes in
evacuated flasks for gas chronatographlc analysis in commercial labora-
tories. Many water and sewer department laboratories can perform such
analyses at minima] cost.
SUMMA
In summary, careful selection, design and operation of sanitary land-
fills will usually minimize gas production and migration. Gas production
and emergence through soil cover is to be expected in a sanitary landfill,
but on-site or nearby structures should be protected from gas migration
or accumulation by structural features, distance or by passive or active
gas barriers.
During landfill operation, careful observation and recording of
physical features such as settlement, damage to nearby vegetation, flash
fires, odor occurrences, gas in subterranean facilities, etc. will provide
131
-------
early warning of potential gas migration. Use of portable gas detectors
to sample combustible gas concentrations fron permanently installed probes
can verify gas migration problems, and follow-up monitoring with more
sensitive instruments will aid in the decision to control gas migration
or restrict use of the area of gas migration.
Observations at the Link Road landfill verified the utility of settle-
ment and subterranean-facility gas determinations as indicators of gas
problems but did not afford opportunity to verify utility of vegetation
effects as a gas migration indicator. Shallow sampling of soils with
a portable gas detector is unreliable. Probes three co 10 feet deep
will provide, reliable indication of gas movement. Gas production and
migration at the Link Road site peaked in the spring months, declined
through the summer and increased through the fall months.
Monitoring of landfill gas migration should be based on soil permeabil-
ity, proxitaity o£ nearby structures and knowledge of the location and
extent of high gas production areas. In some landfills, monitoring may
be restricted to known high gas production areas.
132
-------
CHAPTER X
REFERENCES
1. Municipal Refuse Disposal. American Public \Iorks Association, 2nd ed.,
Chicago, Public Administration Service, p. 128-132, 134, 135 (1*»66).
2. "Landfill Decomposition Gases - An Annotated Bibliography," Ceyer,
Janes A., National Environmental Research Center, Environmental
Protection Agency, SW-72-1-1, 28 pp. (June 1972).
3. Development of construction and use criteria for sanitary landfills,;
final report on a solid waste management denonstration grant.
Department of County Engineer, County of Los \npeles, and Engineering-
Science, Inc. Environmental Protection Publication SW-19d. U.S.
Environmental Protection Agency, 1973. [511 p.] (Distributed
by National Technical Information Service, Springfield, Va.
as Pn-218 672.)
4. Merz. R. C., and R. Stone. Gas production in a sanitary landfill.
Public Works. 95(2):84-87, 174-175, Feb. 1964.
5. "Quantitative Study of Gas Produced by Decomposing Refuse," Merz, R.C.
and Stone, R., Public Works. 99(ll):86-87 (November 1968).
6. "Sanitary Landfill Behavior in an Aerobic Environment," Merz, R.C.
and Stone, R., Public Works. 97(1):67-70 (January 1966).
7. Herz, R. C., and R. Stone. Special studies of a sanitary landfill.
Environmental Protection Publication SU-8rg. Rockville, lid.,
U.S. Department of Health, Education, and Welfare, 1970. (Distri-
buted by National Technical Information Service, Springfield, Va.,
as PB-196 148. [222 p.])
8. In-situ Investigation of Movements of Cases Produced from Decomposing
Refuse. California State Water Quality Control Board, Final Report-,
Publication No. 35, Sacramento (1967).
9. "Effect of Solid Waste Disposal on Ground Water Quality," Coe, J.J.,
Journal of the American Water Works Association. 62:776-783 (Dec-
enber 1970).
10. "Gas Generation and Movement in Landfills," Anderson, D.R. and Callinan,
J.P., In Proceedings; National Industrial Solid Wastes Management
Conference. Houston, (March 1970).
133
-------
11. California State Water Pollution Control Board. Report on
the investigation of leaching of a sanitary landfill. Publication
No. 10. Sacramento, 1954. [92 p]
12. "Water Pollution Hazards from Refuse-Produced Carbon Dioxid««,"
Bishop, U.D., Carter, R.C., and Ludwig, H.F., Journal of Water
Pollution Control Federation. 38(3):328-329 (March 1966).
13. In-situ Investigation of Movements of Cases Produced fron Pcconposintt
Refuse. California State Mater Quality Control Board, Publicatio.i
No. 31,. Sacramento (1965).
14. "Gas Movement in Landfilled Rubbish," Bishop, W.D., Carter, R.C.,
and Ludwig, H.F., Public Works. 96(11):64-68 (November 1965).
IS. Eliassen, R. , F. N. O'llara, and E. C. Monahan. Sanitary landfill.
gas control. American City. 72(12):115-117, December 1957.
16. "Gas Fire in Sewer Manhole Traced to Sanitary Fill Operation,"
Engineering News-Record. 139(26):51 (25 December 1947).
17. How to use your completed landfills. American City. 80(R):91-94,
August 1965.
18. "Gas Fires in a Sanitary Fill," Engineering News-Record. 140(2):86-87
(8 January 1948).
19. "Landfill Gas Burned for Odor Control," Dunn, W.L., Civil Engineering.
27(11):60-61 (November 1957).
20. "Storm Drainage and Gas Burning at Refuse Disposal Sites," Dunn, W.L.,
Civil Engineering. 30(8):68-69 (August 1960).
21. "Methane Gas Explosions Delay Building on a Landfill," Solid V'astes
Management - Refuse Removal Jou-nal, 12(7):20 (July 1969).
22. Brunner, D. R., and D. J. Keller. Sanitary landfill design and
operation. Environmental Protection Publication SW-65ts.
Washington, U.S. Government Printing Office, 1972. 59 p.
23. Development of construction and use criteria for sanitarv landfills:
an interim report. County of Los Angeles, Denartment of County
Engineers, Los Angeles, and. Engineering-Science. Inc. Cincinnati,
U.S. Department of Health, Education, and Welfare, 1969. [267 p.]
134
-------
24. "Model Sanitary Landfill in Kansas City. Kansas." Gerald A. Neely
and Nicholas S. Artz, Civil Engineering (October 1972).
25. "Gas Explosion Hazards in Sanitary Landfills. I.e. MacFarlane, Public
Works (May 1970).
26. "A Gas Monitoring and Control System," Public Works. 101 (12), p. 39
(December 1970).
27. The Desl&n and Installation of A Cas Migration Control System
for A Sanitary Landfill. C.P. Callal.an and R.H. Curske, Bureau
of Sanitation. City of Los Angeles (May 1971).
28. Methods of Soil Analysis - Part i. American Society of Agronomy,
Madison, pp. 520-528 (1965).
135
-------
APPENDIX A
COST ESTIMATES FOR FORCED VENTILATION, NATURAL
VENTfLATION AND IMPERMEABLE MEMBRANE GAS CONTROL SYSTEMS
136
-------
PRELIMINARY COST ESTIMATES FOR FORCED
VENTILATION CAS CONTROL SYSTEMS
(dollars)
I leu
18 Gas wells3
Construction
Butterfly valves
Fittings
Precast boxes
Suction header
Pipe, fittings, etc.
High-flow
system
13.500
2,500
1,000
6,100
8,900
Low-flow
system with-
out seal
13.500
2,500
1,000
6,100
2,500
Low-flow
system with
seal
13.500
2.500
1,000
6,100
2,500
Vacuum pump & fittings
Pump, valves, electrical
panels, fldme traps,
etc. 16,000 8,000 8.000
Surface seal — — 36,000
Contingency A.OOP 4.000 A.OOP
TOTAL 52,000 37,600 73,600
Includes 15 wells in the main branch and three wells in the small
branch of the gas control system.
137
-------
COST CALCULATIONS FOR NATURAL VENTILATION AND
*
IMPERMEABLE MEMBRANE CAS CONTROL SYSTEMS
ASSUMPTIONS
The following assumptions were made for calculating capital costs
of the natural ventilation and impermeable membrane gas control systems:
(1) A dragline can excavate 81.6 yd /hr at the rate of $45/hr.
(2) A 20 feet deep trench 2 feet wide can be excavated for the
natural ventilation system.
(3) Gravel will cost about- $2.25/ton delivered on site.
(4) A side slope of 1.5 vertical to 1 horizontal will be re-
quired for impervious membrane trench when clay is used
as' the barrier material.
(5) The impervious membrane will consist of a dry core 2' feet
vide and 20 feet deep.
(6) Clay will cost $4.25/ton delivered on site.
3
(7) Clay tamping will cost about $0.80/yd .
(8) Clay and gravel have a density of 2.20 ton/yd .
COST BREAKDOWN
Natural Ventilation System
Excavation cost • $ 800
Gravel materials - 7,400
Other costs • 1.800
Total costs- $10,000
*Costs will, vary somewhat, by location
138
-------
Impermeable Membrane
Excavation cost • $ 6,200
Clay materials - 14,000
Clay tamping " 1,150
Other costs - 3.200
Total costs $24,550
VOLUME CALCULATIONS
Natural Ventilation
1,000-foot trench x 20 feet deep x 2 feet wide = 40,000 ft
Trench volume • 1,480 yd
Impermeable Membrane
1,000-foot trench x ?0 feet depth x 15.32 feet average width
for side slope of 1.5 vertical to 1 horizontal
Total volume - 306,400 ft3 - 11,400 yd3
Clay volume - 40,000 ft3 • 1,480 yd
139
-------
APPENDIX B
EVALUATION OP EFFECTS ON VEGETATION
REGARDfNG SANITARY LANDFILLS AT WINSTON-SALEM. NORTH CAROLINA
140
-------
Report
Co Engineering-Science, Inc.
on the
Detailed Work Plan
for
Evaluation Effects on Vegetation
Regarding
Sanitary Landfills
at
Winston-Salcm, North Carolina
19 April 1973
Caythcr 1.. Plunmu-r, Ph.D.
995 Timothy Road
Athens, Georgia
141
-------
The purpose of the work plan was to "Obtain ana Review Data on
Effects cf Gas on Vegetation."
A general ^crusal of the work plan gives the initial impression
that the project is well conceived, well organized, having clearly de-
fined objectives, good control characteristics, and the problem to be
solved is simple, straight forward with potentially clear-cut results.
The problem looks challenging partly because the nature of the objec-
tives have an intriguing origin and partly because the concept of
identifying sentinel plants to indicate an impact of mankind on the
environment is relatively underfined.
The work plan calls for the derivation tl diversity indices to
relate to gas potentials aiiH to effects on sentinel plants. This part
of the work plan seemed to have some inherent difficulties built into
it having been designed without first-hand knowledge of the way di-
versity indices are interpreted bared upon field conditions. To
devise the indices would be no problem, but to relate the meaning of
that information to the objectives of t'.ie project are obstacles that
appeared to be difficult to achieve.
A trip to Winston-Salcm, North Carolina, and to the landfills
was undertaken in conjunction witli a student in the Botany Department
at the University of Georgia to determine the feasibility of completing
the work plan to the satisfaction of Engineering-Science, Inc. We
vent on the weekend of 6-8 April 1973. A day and a half were spent
in the field. Both the Link Road site and the Overdale sites were
visited under the guidance of Mr. Robert Davis.
Communications with Engineering-Science, Inc. advised me that
methane gas was a principal component of those emitted from the old
city dump. A very brief search of the literature revealed also that
carbon dioxide, hydrogen, sulfur dioxide, and hydrogen sulfide were
among the possible kinds. The effects of gas on vegetation would be
complicated, therefore, by mixtures of gases of unknown composition
and of unknown concentrations. At L'IO site it was easy to learn that
age of the fill and time difference.-, in decomposition rates were added
complications.
142
-------
The soil was cracked in placed, sometimes around subsidence pits,
so gases Immcrgcd from place to place according to the distribution
and kinds of subterranean materials.
Vegetation on the old city dump also varied from place to place.
Some was of the onc-ycar-old abandoned field type, that is crabgrass
and ragweed steins. Others were of the six- to eight-year-old field
type--Coldcn rods, Asters, etc. Yet others, wetland Cattails, and
dryland Hroomedgc grass, grew side by side--an unnatural situation
brought aljut by physical changes in the level of the land. Diversity
indices related to site quality arc not meaningful indicators of effects
of gas on vegetation.
Gas oozed from pools of water in the pits as out body weights
shifted from place to place. It became conclusive, therefore, that
changes in atmospheric pressure of one inch could change the load
factor on gas emission by approximately two tons in a pit covering about
300 square feet. This means that cloudy ueathcr and a pressure drop
would probably cause gases to accumulate internally whereas sunny skies
and high pressures would cause these gases to be forced out.
Soils differed greatly within one square meter. Some places the
clayey soils were tough enough to exclude all root penetration; in
other places the sandy soils allowed good plant growth. Chalky alkali
sites evidently excluded some species. High nitrogen places occurred,
as did nitrogen-deficient sites.
Within the chain-linked fence at the Link Koad site, the fescue
grass was growing poorly in some places and excellently in others. It
seemes to have been planted uniformly. It also seemed to have been
affected by poor growth in a small area where odors of certain gases
were almost suffocating to us. Bermuda grass, however, grew very well
where fescue did not. The ground was cracked in placed, odors came
from the cracks, and all grasses within four to five inches on each
side of the cracks were dead. This small situation (and there «rore
several lines of such features) was the only place in which vegetation
was obviously affected by something that came apparently from the cracks.
143
-------
Otherwise, there was absolutely no doubt that soil textures and fertility
were responsible for most other variations in the vegetation. The cer-
tainty of harmful effects of gas along these cracks cannot be demon-
strated clearly without further study.
One plant, the Chickwcek, was yellowed in certain situations but
not in others. Careful examination showed that the lower leaves were
in fact dead. Other leaves had yellow spots on them similar to-such de-
fects as those caused by sulfur dioxide in other plants. This sign of
yellowness suggests the possibility of a sentinel plant. Because the
manifestations were very localized and not widely spread raises some
doubts about the real indicator values o£ such "effects." There is no
doubt in my mind that other plants can be affected similarly, and
probably arc. But the Interpretation and meaning of these facts must
be tested under experimental conditions before any useful indicator
values of sentinel plants can be suggested more particularly. To com-
pare these effects with concentrations of gases in the field would be
nearly impossible, especially without knowledge of the way in which a
single kind of gas injures a plant species. The hopeful thing about
all this is that detrimental effects do occur, but they do not seem
to be widely spread.
The assumption all along seems to be that detrimental effects are
the chief manifestations of gas concentrations. Perhaps this stems from
Crocker's bood Growth of Plants in which the affects of gases on plants
arc treated as extensively as was knowledge about the subject until
1946. On the other hand, beneficial effects could be possible because
methane fumes arc known to stimulate the growth of potato buds (see
Biol. Ahst. 1972, item 33720). Marsh gas, of which methane if the prime
constituent, is commonly associated with plants. No doubt many plants
have become quite tolerant to methane. Hydrogen sulfide is not really
very toxic to plants, but it is to animals. A very good possibility
exists that the vegetation on a sanitary landfill, or any old dump,
could be enhanced in places by the totality of the local environment.
144
-------
The Overdale situ is a marvelous example of what landfills ought
to be. The fescue plantings added acceptability of the site. That
grass did have several unusually green patches growing hc>*e and there—
probably from dung piles dropped by stray dogs, perhaps urine spots,
etc. In general, the grass looked nitrogen deficient, but uniformly
so. Obviously, the site was more homogeneous than the old dump. No
detrimental conditions were outstanding. The question was raised:
What sentinel plants would indicate gas at different concentrations
here? The assumption was cither that gases would ooze from the ground
uniformly, or that they would move laterally and come out near some
pre-existing harrier.
A search for such a plant is underway. Perhaps one can be found,
such as Chickvecd, but it is more likely to be an annual rather than
a perennial; it would probably be a temporary rather than permanent
resident on the site; and it would be seasonal in its responsive indi-
cator qualities. When that plant,
------- |