Ipr eduction and Management of Leachate from
Municipal Landfills: Summary and Assessment
iCalscience Research, Inc., Huntington Beach, CA
Prepared for
Municipal Environmental Research Lab,
Cincinnati, OH
May 84
PB84-187913
U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
NIIS
-------
EPA-600/2-84-092
May 1934
PRODUCTION AND MANAGEMENT OF
LEAC3ATE FROM MUNICIPAL LANDFILLS:
SUMMARY AND ASSESSMENT
by
James C. 5. Lu
Bert Sichenberger
Robert J. Stearns
Calsciencs Research, Inc.
Huntington Beach, California 92647
Contract No. 68-03-2361
Project Officer
Wendy J. Davis-Hoover
Solid and Hazardous Waste Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45263
-------
TECHNICAL REPORT DATA
(Pfcase read Inuruct ons o,, the reverse be/ore cornr’1errn )
1.REPO T NO. .
EPA-600/2-84-092
4. TITLE N0 SL8TITLE
3. ECIPtENTS ACCESSION NO.
PS 4 i 791 ;
S. PE O T DATE
PRODUCTION AND MANAGEMENT OF LEACHATE FROM
MUNICIPAL LANDFILLS: SUMMARY AND ASSESSMENT
6. s GORGANIzArON CODE
7 .AUTHOR(S I
James C.S. Lu, Bert Eichenberger
8. PER OAMINQ ORGANIZATION EPO T NO.
Robert J. Stearns, Ihor Melnyk
.
LPERFORMING ORGANIZATION NAME AND ADDRESS
10. PROGRAM ELEMENT NO.
Cal Science Research Inc.
Huntington Beach, CA 92647
ll.cONTR GApANrNo.
68-03-2861
. SPONSORING AGENCY NAME AND ADDRESS
Municipal Environmental Research Laboratory——Cm., OH
Office of Research and Development
13. TYPE OF REPORT AND PERIOD COVERED
Final 9/79 — 3/82
1A.SPONSORINGAGENCYCOOE
U. S. Environmental Protection Agency
Cincinnati , Ohio 45268
EPA/600/14
15 SUPPLEMENTARY NOTES
(Stephen C. James - Proaram
Wendy Davis-Hoover, Project Officer 513/684—7871
Coordinator 513/684-7871
. A STMACT
An assessment was made to evaluate production and management of leachate
from municipal landfills for purposes of identifying practical information and
techniques which may be useful to design engineers and site operators. Also
assessed were: advantages, limitations, and comparative costs of various
approaches for the estimation and mitigation of environmental and public
health impacts, management options, and additional research needs on the
generation, control, and monitoring of landfill leachates. Numerous
mathematical models have been proposed for estimating leachate generation and
are usually based on the water balance method. Several models have been
proposed which are fairly successful in simulating the change in leachate
strength with increasing landfill age or cumulative leachate volume A zone
of saturation monitoring program is established to give a prompt indication of
groundwater contamination.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IOENTIFIEPSIOPEN ENDED TERMS
C. COSATI Field/Cioup
1 LOISTRISUTION STATEMENT
PFI FASF Tfl P1!Rt IC
19. SECURITY CLASS (TIns Repo iI
UNCLASSIFIED
—
21. NO. OF PAGES
474
20. SECURITY CLASS (711.: Q j
UNCLASSIFIED
22.
PA F..m 2220 —I (Rev. 4._77) R viOUs EDITION IS OSSOLETI
1
-------
NCTICE
TEIS DOCUMENT EAS BEEN REPRODUCED
FROM THE BEST COPY FURNISHED US BY
TEE SPONSORING AGENCY. ALTHOUGH IT
IS RECOGNIZED TEAT CERTAIN PORTIONS
ARE ILLEGIBLE, IT IS BEING RELEASED
ZN THE INTEREST OF MAKING AVAILABLE
AS MUCH INFORMATION AS POSSIBLE.
-------
DISCLAIMER
The information in this document has been funded
wholly or in part by the United States Environmental
Protection Agency under Contract tb. 68—03—2861 to
Caiscience Research, Inc. It has been subject to the
Agency’s peer and administrative review, and it has
been approved for publication as an EPA document. Men-
tion of trade names or commercial products does not
constitute endorsement or recommendation for use.
ii
-------
FOREWORD
The U.S. Environmental Protection Agency was created because
of increasing public and government concern about the dangers of
pollution to the health and welfare of the American people. Nox-
ious air, foul water, and spoiled land are tragic testimonies to
the deterioration of our natural environment. The complexity of
that environment and the interplay of its components require a
concentrated and integrated attack on the problem.
Research and development is that necessary first step in
problem solution; it involves defining the problem, measuring its
impact, and searching for solutions. The Municipal Environmental
Research Laboratory develops new and improved technology and sys-
tems to prevent, treat, and manage wastewater and solid and
hazardous waste pollutant discharges from municipal and couunity
sources, to preserve and treat public drinking water supplies,
and to minimize the adverse economic, social, health, and aes-
thetic effects of pollution. This publication is one of the
products of that research and provides a most vital communica-
tions link between the researcher and the user community.
This report evaluates the production and management of
leachates from municipal landfills for purposes of describing
current practices, available control technology, comparative
costs of alternae ve measures, management options, and mitigation
of environmental impacts.
Francis T. Mayo, Director
Municipal Environmental Research
Laboratory
iii
-------
ABS TRACT
This report evaluates production and management of leachate
from municipal landfills for purposes of identifying practical
information and techniques which may be useful to design engi-
neers and site operators. Also assessed are: advantages,
limitations, and comparative costs of various approaches, for
the estimation and mitigation of environmental and public health
impacts, management options, and additional research needs on
the generatiOfl control, and monitoring of landfill leachates.
The quantity of leachata produced from a municipal landfill
may vary considerablY with management or operating practices.
Uumerous mathematical models have been proposed for estimating
leachate generation and are usually based on the water balance
method. More than one hundred different approaches are available
for water balance calculations; however, few comparisons have
been made in the past to identify which approach can achieve
better results, or which approach is suitable for what types of
landfill conditions. The applicability, accuracy, and sensitivity
of water’bal3.flce techniques are, therefore, greatly Lacking.
The composition of municipal landfill leachate is highly
variable from site to site, and major changes in leachate
compositiOfl occur during the life oi a. landfill. Leachate
strength is influenced by factors which include refuse composi.-
tion, refuse density, rate of water infiltration, landfill depth,
and landfil 1 temperature. Several models have been proposed
which are fairly successful in simulating the change in leachate
strength with increasing landfill age or cumulative leachate
vol e. At present, however, the models have only been used in
interpreting experimental data, and their usefulness as ore—
dictive tools in field-scale situations is either limited
or yet to be tested.
The attenuation and migration of contaminants in the soil /
water system are influenced by physical, chemical and biological
mechanisms. Quantitative data relating to the mobility and
attenuatiofl.0f individual constituents by specific mechanisms
are not available; consequently, considerable effort has been
placed on the development of sophisticated conceptual-mathematical
leachats migration models. !owever, due to the highly complex
leachate/ 501 1 environments, no leachate migration model exists
that can simulate all of the physical, chemical, and biological
iv
-------
processes occurring in a typical landfill system.
Controlling leachates involves preventing their formation,
preventing their movementonce formed, and controlling their
composition. Surface water control measures such as contour
grading and su.rf ace water diversion, surface sealing, and re—
vegetation are used to minimize the quantity of water entering
the landfill for purpose of reducing leachate generation. Plume
management procedures involve actively altering the course of
leachate movement by either adding or removing water from around
the landfill. In general, a long term reduction of the strength
and contaminant flu.x of leachates can be obtained by high water
application rates, small refuse particle size, and shallow Land-
fill depths.
The successful treatment of high strength leachates will
probably require different combinations of unit treatment
techniques, the appropriate combinations dictated by the
character of the leachate. For a high strength leachate
containing high concentrations of both organic and inorganic
contaminants, a combination of biological and physical/chemical
processes will be needed. Regardless of the system selected,
leachate recirculation can be beneficial in reducing the organic
strength and volume of leachate, as well as reducing treatment
costs.
A zone of saturation monitoring program is established to
give a prompt indication of ground-water contamination. The
size of the landfill, hydrogeologic environment, and budgetary
constraints a.re factors which will dictate the actual number of
wells. Non-sampling methods provide for the determination of
water content and water movement in the vadose zone. Incorpora-
tion of vadose zone devices can provide an early warning of
potential ground—water pollution. Additionally, an effective
vadose zone monitoring network could reduce or largely preclude,
the requirements for ground—water monitoring.
This report was submitted in fulfillment of Contract No.
68-03-2881 by Calecience Research, Inc. under the sponsorship
of the i3.S. Enviroz mental Protection Agency. This report covers
the period September, 1979 to May, 1982, and work was completed
as of May, 1982.
V
-------
CONTENTS
Abstract..................................,., . ...._..,... iv ’
?igi res..................................................
Tables...... ••••••• ..S 4444 .4 .• ••• .... . .. ..xiV
1. Introduction. . . . . . . . . . . . • • • • • . . . . . . . . . . . . . . . . . . . . .
2. Conc lusionsandRecoznmendatjons . .................. 3
3. Leachate Generati.on. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Ix trod.uct2.on. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Factors Affecting Leachate Generation........ 7
Precipitation. . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Surface RLin—onandRtmoff.. .,..,........ 14
Ground-Water tntrusion and Irrigation... 14
Refuse Decomposition. ... . . . ... .... ‘• • 16
Codisposal of Municipal and Industrial
Sludges....................,.,........ 17
Evapotransp .rat .on... ... .. .. .. .. ........ 21
Infi1t.rat .ort. ...... .... . •....... .. .. .. . . 25
Moisture Retention. ....... .. ... ......... 27
Perco1at .on.. .4.•••SS • .,. . . .. . ••• •• •4$Ø 27
Concepts and Techniques Describing Leachate
Generation. . . . . . . . . . . . . . . . . . . . . 31
Deterxui.n .ng Prec ipitation............... 35
Determini gsurfaceRunoff.,............ 36
Determining Infiltration................ 51
Determining Evapotranspiration.... ...... 56
Determining Water Storage Capacity...... 66
Determining Time of First Appearance
of Leachate. . . . . . . . . . . . . . . . . . . •... • • e • 69
Determining Ground-Water Intrustion..... 73
Examples of Leachate Generation Models....... 76
Quantification of Leachate Generation
Factors. • •• • • • • • • • 76
Water Balance Calcu lation............... 80
HSSt ibS Model... •....... . . •4 44444 ..... 91
Discussion................................... 95
Sensitiv .ty Analysis.................... 95
Applicability and Accuracy.............. 98
4 • Leachate Composition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . a 1.08
Introduction • • . . • • • •4 • . .. . . 108
Chemical Cornpositiort.........................108
vi
-------
4icrobio1ogical Coiuposition . 123
3acteria. • • . . . . . . . . . . . . . • . 123
7irm es 125
Fi.ir gi. . . . . . . . . . . . . . . • . . • . . . • . . • . . . . . . . . . .127
Parasites . . . . . • . . • 127
Principal Mechanisms Leading to Transfer
of Refuse Mass To The Percolating Water...,.128
pH. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Redox Potential. •....... . . . . . •........ . .130
Adsorption and Complexation 130
Temperature. . . . . . . . . . . . . . . • • • • • . • • . . • . . . .13].
Biological Mechanisms............... ... . .13].
S ary. . . . . . .. • ...S..• •• . .132
Factors Affecting Laa.chate Composition..... .. .133
Refuse Composition. .. . . . ,..... • •••••.••• .133
Refuse Processi.ng . ... 136
Landfill Age . . 144
Rate of Water App licatiori................152
DepthofteaChed3ed 161
Landfill Temperature.....................162
163
Effects of Adding Sludges to Municipal
Landf ills. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
dd ti .on of ‘, WTP Sludges... .. .. .. .. .. ... .166
Addition of Industrial Sludges..... ... . . .167
Leachate Composition Models. ... . . . . . .. ...... . .169
Curve Fitting Approaches..... .. ...... ... .170
Process Modeling Approaches..............173
S .=ar f.. . . . . . . . .. 182
S. Leachate Migration . . . . . . . . . . . . . . . . . . . . . 188
Int’oductjo...,. • •........ ••••• •.•• ....... .. .188
Leachate/EflViroxUZtefltal Interactions.... . ... . . .189
Soil Properties. . . . . . . . . • • • • . . . . . . . . . . . . . 289
Migration Mechanisms.....................190
Migration Trends of Contaminants........ .200
Types of Leach.ate Migration Models........ . . ..214
Descriptive Models.......................214
Physical Models. •. . . . . . . . . . . . . . . . . . . . . . . . 214
Analog Models............................2 ] .6
Mathematical Models......................216
Model. Distinctions...... . . . . . . . . . • . . • .216
pirical or Conceptual Models. ...... . . . .222
Deterministic or Stochastic Models.....,.222
Static or Dynamic Models................222
Spatial Dimensionality of the Model......223
Conceptual-Mathematical Models....... . ........ 223
Saturated—tJnsaturated Transport Models... 224
Saturated-Only Transport Models. . .. . 224
nsaturated—Only Transport Models. . , 224
Analtyical Models..... 224
Mathematical Solutions For Conceptual Models..229
vii
-------
Analytical Methods • 229
Nt erical Methods.......... . 229
Suxmna..ry.. . . . . 231
6. Available Control Technology 233
Introduction 233
LeachateVOltm eC0ntrdl ..... 234
Ground—Water Control Measures ....... 234
Surface Water Control... .. . . ......... .... 253
Leachate CompOsitiofl Control .. 259
Landfill Construction and Operation
F eat’.ires. • • . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
LeachateReC .rCUlati.On............ . ” 26 S
Addition of Municipal and
Industrial Sludges ...... .... 267
Addition of Selected Sorbents........ .... 269
LeachateCO1leCti0nSYStemS... .t. . 5 ” 27 °
Shallow Drains . . . . . . . . . . . . . . . . . 270
Well Systems . . . 27].
Leachate Treatment. .... 272
Basic Leachate Treatment Methods ..272
Leachate Treatment Systems..
Treatment Costs. . . . . •. . . . . . . . . . . . 283
Suary. . •. . •5t• •. • • • • . . . •.•• •• • • •• 288
7. En n.roflmental Monitoring .............. • .... 290
Introduction...... . . . . . . . . . .. . . • • . . . . . . . .. . . . . 290
Vadose Zone...... . . . • •• St...... .. . .•.•••• . . . . .290
Monitoring Approaches. .. 292
Available Monitoring Equipment
and Costs. . . . . . . . . . . . . . . . . . . . . 292
Zone of Saturation. . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
Monitoring ....311
Available Monitoring Equipment
and Costs.... ... .. ...• .. . . ...313
Monitoring Approaches and Considerations.. ....328
Selection of Sampling Locations...... . .. .328
Monitoring E requency.. ...... ......... . .. .335
Mon itoringParameters....
Sample Collection Methods, Contain-
ers and Sample Preservation, and
Analytical . ...337
Analytical Procedures and Costs...... ....342
....342
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .344
Appendices. .......s... . •..... . . . . . . .........t• . . . . . . . . . .1S .383
A. Thornthwaite Tables. . . . . • • . . , • . . , . . . . . . . . .. . . . . . . . . , .383
B. Case Study Sites Used for Evaluation of Leachate
Generation Models........................ .418
C. Municipal Landfill Leachate Composition. ..........
D. Additional Research Needs . .... .. . . . . . . .433
viii
-------
FIGURES
Nuxn er
1 Factors affecting leacha.te volume generation. . 9
2 Mean annual rainfall in the United States based
on4Oyearsofrecord(18991938) 12
3 Mean annual snowfall in the United States. 13
4 Mean annual effective precipitation
(calculated from Figi .res 2 and 3)................... 15
5 Maximum non—leach. .ng sludge to refuse ratio........... 18
6 Leachate production resulting from 1üdge compaction.. 19
7 Leachate production resulting from sludge compaction
and decomposition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Mean annual pan evaporation........................... 22
9 Mean annual net effective precipitation (prepared
from. Figure s 4 and 8) . . . . . 2 3
10 Relationship of rainfall, runoff, evapotranspiration,
and infiltration during a rainstorm event . 26
11 Potential infiltration for landfills in various
geographical areas of the United States............. 28
12 Moisture content as a function of time for a
draining soil or refuse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
13 Example showing the effect of moisture content on
pe.rmeabi lityf ora claYS oil...... . 32
14 Municipllandfill waterbalance............. . ...... 34
15 Estimation of direct runoff amounts from storm
rainfall . . . . . . . . . . . . . . . . . 46
16 Chart for estimating 50-year—frequency peak rates
of flow •I. ...•• 48
ix
-------
Number Page
17 DistributiOn o rainfall fact 5 used with the
modified Cook’ s method 49
18 TemperatUre versus saturated Vapor pressure 62
19 TemperatUre versus A for use in the Penman
equation (Equation (28) ) . . • • . • • • • • • • • • • • • • • . • • . . . . 63
20 Water storage capacitY of t S A soils. 67
21 Refuse field capacitY versus refuse dry density..... 68
22 Time needed for one inch infiltration per month
to bring soil from wilting to field capacity 72
23 Time needed for percolate to pass through the
• • .•.. • ••....... . ... 74
24 Estimating rate of landfill tderflow 73
25 Flow chart of water balance calculation. ............ 77
26 Generalized flowchart for the hydrologic
sjilationmodelMS S....... ........... 93
27 Leachate constituent concentration ranges........... 110
28 RelatiOnship between refuse permeability and
refuse density. . . • . . . . . . . . . . . . • . . • . . . . . . . . • • • • • • • • 138
• 29 Mass removal of COD versus cumulative leachate
volt IZte. • • • • 145
30 RelatiOnShip between landfill age and leachate
composition...... .•. . ... . . .......•.•••• . .•• .. • . •.. 146
31 Volatile acids found in leachates during early
stages of stabilization... .. . . . • . . .•.............. 154
32 Trends in the identified fractions of leachate
TOC Vs. landfill age. . . . . . . . . . . . . . . . . . . . • . . . . . • . . . 155
33 MoistUre addition and leachate response......... .... 158
34 Leachat total solid s concentration from experi-
mental landfills subject to different moisture
application rates. •••.... •.• . ...• ..... •.........• . 160
35 Leachat COncentration history curves for simulated
landfil l sof different ...... 163
x
-------
______ Pace
36 Simulated and experimentally derived concen-
tratior. histories for COD and chloride
from Wigli (1979) . . . . . . . . . . . . . . . 172
37 Single well—mixed reactor for model of leachate
inorga.nics. . . . . . . . . . . . . . •. . . . . . . . . . . . . . . . . . . 176
38 Measured and simulated results of leachate
total solids concentration history from
Strau,b (1979)...................................... 178
39 Single well-mixed reactor for model of leachate
organics 179
40 Measured and simulated results of leachate COD
concentration history from single reactor
model from Straub and Lynch (1982) 180
41 Vertically cascaded reactor model of landfill. 181
42 Measured and simulated results of leachate COD
concentration history for vertically cascaded
reactor model from Straub and Lynch (1982) 183
43 USDA soil textural classification 191
44 Solubilities of major phosphate solids ...... 208
45 Total soluble phosphate levels in leachates as
affectedbypHandalkalinity..... ............209
46 Design specification for surface and interceptor
ditches . 255
47 Landfill design and operational features promoting
rapid landfill stabilization. .. .. .. ..... . .. ........ 263
48 Landfill design and operational features
promoting gradual landfill stabilization........... 264
49 COD removal efficiencies for the treatment of
leachate. •. • ........ •,•, •, .. . , ..... . ... .... ..... . .. 277
50 Percent removal of NH 2 , Fe, Zn, Ni, Cd, and Cu by
various treatment a cternat .ves ... ... 278
51 Schematic of complete leachate treatment plant....... 280
52 Effects of added moisture on leachate COD 284
‘ ci
-------
Number ? ace
53 Hydrogeological cycle.. . 291
54 Generalized cost of coring in unconsolidated
and consolidated formation, December, 1981 296
55 Pressure/vacuum lysiineter.. . ..... .......... ......... 297
56 Cost of pressure/vacuum lysimeter for various
depths and sampling densities, December, 1981..... 299
57 Schematic representation of tensiometer and section
through the ceramic cup. . . . . . . . 300
58 Cross section of tensiometer-pressUre transducer
assembly. ....•...• ...... ..........•• 301
59. Cost of tensiozneter and electrical meter for
various depths and sampling densities, December,
. . 303
60 Soil psychrometer . . . . . . . . . . . . . . . . . . . . . . . 304
61 Cost of psychrometer and electrical meter for
various depths and sampling densities,
December, 1981.. •..........••• •.. ..............• • 305
62 Gypsum electrical resistance block 306
63 Cost of multiple electrical resistance blocks and
soil moisture meter for various depths and
sampling densities, December, 1981 309
64 Equipment and principles of r.eutron moisture
loggings..... ... .......••• . . ... .....•• .... . •. •... . 310
65 Cost of neutron logging for various depths and
sample space densities, December, l981............ 312
66 Determination of approximate ground-water flow
direction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
67 Costs of well drilling in unconsolidated
formations, December, 1981........................317
68 Costs of well drilling consolidated formations,
December, 1981. . . . . . . . . . . . . . . . * . . . . . . • . . . . . 318
69 Costs of PVC casing, December, 1981...... .... 320
xii
-------
______ Page
70 Total costs o.f 4—, 5—, and 6—inch diameter wells
in consolidated formations, Decem.be:, 1981......... 322
71 Total costs of 4—, 5—, and 6—inch diameter wells
in unconsolidated formations, December, 1981....... 323
72 Single screened ground-water monitoring well......... 325
73 Costs of 2—, 3—, 4—, 5—, and 6—inch diameter
monitoring wells, December, 1981 326
74 Well clusters . . . , . . . . . . 327
75 Single (A) and multiple (B) installation
configurations for air—lift samplers 329
76 Costs of 1¼—, 1½—, and 2—inch diameter air
lift samplers. . . . . . . . . . . . . . . . . . . . 330
77 aypothetical hydrologic conditions and possible
Location of vadose monitoring devices.... 332
78 Cross section of Genasco Landfill and underlying
materials showing a portion of the monitoring
well—lysimeter network.. ...... .. . . .. . .. .. ... ....... 334
xiii
-------
TABLES
Number Page
1 Approximate Seasonal Transpiration of Selected
vegetation. . . . . . . . . . . 24
2 Gen.eralized Water Balance Equation at a
Municipal Landfill Site. . . . . . . . . . . . . . . . . . 33
3 Runoff Coefficients for Storms of 5— to 10—Year
Frequex .cies . • 38
4 Runoff Coefficients for Drainage Areas with
Different Topograph.y, Soil, and Cover Conditions 39
5 Runoff Coefficients as Affected by Cover
Material and Slope...... .....•• ... .. ......•• ... •..... 40
6 Runoff Coefficients Used by Salvato, etal., for
Leac .ate Estimation . . .. . . .. .. . . .. •....... . . . .. 40
7 Rainfall Limits for Estimating Antecedent Moisture
Conditions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . 41
8 Hydrologic Soil Groups Used by Soil Conservation
Se,rv’ice. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . 42
9 Runoff Curve Numbers for Hydrologic
goil—CoverComplexas(forAMCII).................... 43
10 Runoff Curve Number (CN) Conversions and
Constants. . . . •. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
11 Incremental W Values for Use in Cook’s Method.......... 47
12 Frequency Factors for Use With Cook’s Method........... 50
13 Typicalf 1 ValuesforBa.reS oils...... ............ 52
14 Vegetation Cover Factor for Estimating Infiltration
Capacity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
15 Values of the Constant K for Degree—Day Equation 54
xiv
-------
L’Iumber Page
16 Eva otrans iration Equations ......... 58
17 Seasonal Consumpt±ve— se Coefficients k in 1aney—
Griddle and, Elaney-Morin Equations, for
Irrigated Crops in Western United States ... 59
18 Daytime—Hour Percentages, or lOOp, in Blaney-Criddle
and Blaney—Mori.n Equations........ ...... ............. 60
19 Mean Possible Duration Expressed in Units of 30
Days of 12 Hr Each, or the Adjusting Factor
for Potential Evapotra.nspi:ation Computed by
the Thorrithwaite Equation . 61
20 Midmonthly Intensity of Solar Radiation on a
Horizontal Surface, in Millimeters of Water
Evaporated. per Day, or the Value R in the
Penxna.n Equation . 64
21 Values or B .n the Penman Equati.on...................... 65
22 Water Absorption Ranges for Solid. Waste Components..... 70
23 Predicted Range of Absorptive Capacity of Municipal
Refuse as Received at Oceanside Landfiil............. 71
24 Potential Evapotranspiration Computed by Thornthwaite
Method. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31.
25 Moisture Routing Through a Sanitary Landfill Soil
Cover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 3
26 Moisture Routing Through Underlying Compacted
Re fuse (First Year) 85
27 Mositure Routing Through a Landfill Soil Cover
After One Year of Emplacement........................ 87
28 Moisture Routing Through Underlying Compacted
Refuse (Second Year) . . . . . . . . . . . . . . . 99
29 Moisture Routing Through Underlying Compacted
Refuse (Third Year) •...... .... •• ......... •.•• • •• . ... 89
30 Water Balance Calci4ation for a Landfill in
Ohio..................................... 90
31 Water Balance Calculation for a Landfill in
LOS Angeles, California . 92
xv
-------
Number Page
32 Leachate Production at Blue Valley Sanitary
Landfill Under Different Assurtt tjons, in
Inches of Water . . . 96
33 Summary of SensLt .v .ty Study Results. ............ 97
34 Comparisons Among Leachate Estimation Methods......... 100
35 Methods Used for the Evaluation of Leachate
Generation. . . . . . . . . . . . . . . . . . . . • • • • • • • , • • • • • • • • • • • , • • 106
35 ComDariSoXtS of Leachate Generation Between
Measured and Calculated ResuLts........ ..... 107
37 Organic Compounds or Classes Identified in Landfill
I e achates. 122
38 Bacteria and Bacteriologic Indicators Identified
in Municipal Leachates... .. .. . . . . ..... 124
39 Cumulative Mass Removal of Refuse Pollutants
by Laboratory Columns and Mass Removal by
Extraction Testing. . . . . . . . . . . . . . . . . . . . . . . . . . 135
40 Cumulative Masses of Leachate Parameters per kg Dry
Refuse Ofl a 3asis of Equal Leachate Volume..... 140
41 Sustained Maximum Concentrations of Leachate
Contaminants (from Streng, 1976) .... .... ... . .. 168
42 Total Available Mass Removal (from Wi h, 1979)........ 173
43 Summary of Leachate Compos .ti.on Models. ... ............ 184
44 Soil Organ.1.sms. . . . . . . . . . . . . . . . . . . . . . . . . . . . •. . 192
45 Important Solubility Products (Ksp) of Trace Metals... 195
46 TheoretiCal. Redox Conditions for Redox Couples........ 202
47 Major Phosphate Controlling Solids.................... 207
48 Factors that Influence the Movement of Viruses in
Soils. • . . . . • • • • • • • • • • • • • • • • • . . . . . . . . . . . . . . . • . . . • • 213
49 Example Models and Their Classification into
Different Groups. .1.1.:..... . . .. I... • I •. • .. . 215
xvi
-------
Number Pace
50 Partial List of Available Transport Models for
Application to Ground—Water Quality Problems. 217
51 Explanation of Symbols used in the Mass Transport
and Flow Equations . . . . . . . . . . . . . . . . . . . . . • • . . . 227
52 Partial List of Equations Gsed to Describe
AdsorptionReactions ... 229
53 Properti.esorCotonJ.y sedLiners 236
54 Typical Characteristics of Synthetic Liners 237
55 Selection C: teria for Synthetic Liners 239
56 Synthetic Layers for Asphalt, Clay, and 3entonita.. . -. 242
57 Costs of Flexible Polymeric Membrane, Plastic,
and Rubber Liners . . . . . 243
58 Cost Estimations of Soil, Admix Materials, and
Asphalt Membrane Liners .. 244
59 Costs of Ground— 7ater Control Measures 245
60 RelativeCostsofGrout ............... .249
61 Leachate Plume Control 251
62 SurfaceWaterControl . 256
63 Casts of Sur:ace Seals...... . . . . . . . . . . . 258
64 Grasses and Leg aes Coonly used for Vegetation 260
65 teachate Recjrculation and Collection Technologies.... 266
66 Zinetic Parameters for Aerobic Treatment of Leachate
(High Strength) and Domestic waseewater 274
67 Summary of System 1 Qperat on Data.................... 279
68 A Summary of Cost Estimates for Leachate Treatment.... 286
69 Estimated Comparative Azinuai. teachate Treatment
Costs for Recirculated and Equivalent Non-
Recirculated Landfills . . . . . . . . . . . . . . . . . . . . . . . /
xvii
-------
‘Tumber Page
70 Compa.rative Savings of Recic ilated Case over
on—Recirc 1L4ted Case. . . . . . • . . . • . . • . • • . • • • • s • . • • e . 288
7 . Basic ?arts and Costs of a Vii eyer Soil Sample..... 295
72 Advantages and Disadvantages of Zone of
Sat rati n Monitoring Oevices................,. . . ., ,
73 Reeon endations for Sampling and Preservation of
Samples. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
74 Costs of 7ater Quality Analyses, Dece n bex,
343
xviii
-------
AC OWLZZGMENTS
Caiscience Research, Inc. would i.iice to thank our Project
Officers, Mrs. Wendy J. Davis-Ecover and Mrs. Laura A.
Ringenbach, for their assistance throughout this project.
We are particularly grateful. for the support and valuable
guidance given s by Mrs. Wendy J. Davis-ifoover.
xix
-------
SECTION 1
INTRODUCTION
In 1990, a projected 295 to 341 million metric tons of mu-
nicipal refuse will be produced annually in the ztited States
(Doggett, et al., 1980). The disposition of this huge välume of
material ECianitary landfill conveys with it an inherent
Potential for pollution of indigenous water resources. With the
increasing demands on our land and water resources, the protec-
tion of these resources from the impacts posed by municipal solid
waste disposal is imperative.
A major environmental impact, resulting from the disposal of
municipal solid waste into a sanitary landfill, occurs when water
passing through the refuse accumulates various contaminants.
This percolate or leachate may enter underlying ground waters and
seriously degrade the water quality of the aquifer.
Potential pollution from a refuse disposal area must be con-
sidered in its design and location. understanding of the factors
and conditions involved in the production of leachate is neces-
sary for its elimination and control. A major goal in planning,
designing, and operating a sanitary landfill is to minimize all
Sources of water reaching the refuse so that leachate productton
is avoided. Leachate product on be minimized or nearly elim-
inated by preventing water contact with the refuse by the use of
surface and subsurface drainage and properly selected cover ma-
terial that is graded and seeded with a high evapotranspiration
crop.
In some cases it will not b economically feasible to elimi-
nate ].eachete production from these refuse areas and treatment
of the leachate will be necessary. The prOcesses to treat this
material effectively have not been studied in detail, but either
chemical or biological treatment should be effective. It is also
possible that a combined treatment system in which the major
treatment is done by chemical and physical means, with the
Polishing being done in lagoons or other biological processes,
would be the most economical. In all cases, however, minimizing
the quantity. of leachate production, as well as protection of the
ground and surface waters, should be the major goal.
An extensive body of literature is available that describes
1
-------
sanitary landfill leachate production, composition, migration,
control, and monitoring. Out of these works have evolved design
and operational criteria which are primarily empirical in nature
and which may or may not have a relationship to environmental
conditions. Various studies concerning sanitary landfill behav-
ior have been conducted in recent years to better understand them
and to delineate and define significant management options.
However, a review of existing literature has revealed a number
of areas requiring additional research for technological, develop-
ment.
The objectives of this study are: (1) to clarify the under-
standing of leachate production and management options through a
critical review and analysis of existing information, (2) to
identify practical information and techniques which can be used
by design engineers and site operators, (3) to delineate weak-
nesses in the available data base, and (4) to identify techniques
which may be useful in the estimation and mitigation of environ-
mental and public health impacts caused by leachate generation.
Specific objectives are to prov’ide documentation of current meth-
odologies, advantages and limitations of various approaches,
comparative costs, and additional research needs on the genera-
tion, significances, and associated costs of controlling and
monitoring leachate from municipal landfills. This information
is presented in terms of the following subject areas:
• Leachate generation;
• Leachate composition;
• Leachate migration;
• Available control technology; and
• Environmental monitoring.
The presentation of each of these subjects is based upon
a review of the available literaturi and site histories.
2
-------
SECTION 2
CONCLUSIONS AND RZCOb ENDATIONS
CONCLUSIONS
A critical evaluation of production and management of leach-
ate from municipal landfills was performed to identify practical
information and techniques that might be useful for landfill de-
sign and. operational purposes. This information, including
economic evaluation, is important in efforts to estimate, reduce,
and control the risks of enviroz mental and public health impacts
caused by leachate production. Existing literature revealed a
significant volume of information on the generation, composition,
significance and costs of controlling leachates, as well as an
absence of data necessary to the development of leachate predic-
tive techniques. Report conclusions re presented in the order
of their discussion in the text.
.The quantity of leachate produced from a municipal landfill
may vary considerably with management or operating practices,
depending on whether leachate is viewed as a short or long—term
problem. operational factors which affect leachate quantity may
include: cover material handling, watering prior to compaction,
watering following compaction. daily variation in cell construc-
tion, and variation in waste composition (e.g., municipal refuse
alone or codisposal of municipal refuse and sewage sludge or
industrial wastes; milled or ummilled refuse).
Numerous mathematical methods have been proposed for a quan-
titative estimation of the volume of leachate generated from
landfills and are usually based on a mass balance approach
(i.e., water balance method). Components of these models are
relatively easy to obtain but other model variables such as the
surface runoff coefficients, runoff curve number, and evapotran—
spiration from the landfill surface are more difficult to develop.
Limited field data exists for verification for many of the leach-
ate generation models. Therefore, the applicability, accuracy,
and sensitivity of leachate generation models, is largely unknown.
The composition of leachate produced from a municipal land-
fill is also highly variable, depending on factors such as refuse
composition, refuse processing, landfill age, the rate of jnf ii—
tration, landfill depth, and landfill temperature. The quality
3
-------
of leachate can be controlled to a large degree. Shredding and
baling of refuse, landfill depth, and the rate of water applica-
tion to the landfill surface can influence the rate at which
contaminants are released from refuse, and can determine whether
leachates have a long or short-term pollution potential. Based
on empirical data garnered from a number of studies addressing
landfill and leaching behavior, several models have been developed
which describe leachate quality as a function of time or cumula-
tive leachate generation. The most promising of these models
attempts to simulate the physical/chemical and biological pro-
cesses which occur during leaching. Presently, however, the
leachate composition models are usefui. only in the interpretation
of experimental results, rather than finding application to
field—scale problems.
teachate migration models are based upon constituent mass
transport and water flow equations. Numerous models are avail-
able for predicting chemical and physical migration: biological
models, however, are generally lacking. While a large number of
conceptual-mathematical models exist, none are universally
applicable for the simulation of all of the physical, chemical,
and biological processes that are operative in a typical waste
disposal system. The complexity of these processes, which operate
in a simultaneous and interactive ma .ner, are probably such, that
the development of a. generic model would be impractical because
the resulting program would undoubtedly become so large and
complex that the .cose of operating it woul4 be exorbitant.
Leachate control teclmoi.ogy is a relatively well-developed
methodology for the management of landfill leachates. Various
ground-water and surface water control approaches are available
for the control of leachate into or out of the fill. Approaches
in which dewatering or counter pumping (injection) is practiced
appear to be effective control measures, although the long term
energy costs for pumping are restrictive. Accelerated stabili-
zation processes such as leachate recycling and nutrient addition
are promising, but require additional study. Extensive liner
technology is available, however, field verification studies have,
in general, been inconclusive regarding anticipated liner life-
time.
?actors which affect leachate composition and are amenable
to control are: (1) physical characteristics of refuse, (2)
rate of water application or leachate recycle, and (3) land.f ill
depth. In general, shredded refuse, higher water/leachate
application rates, and shallower landfill depths axe conducive to
increased refuse stabilization and, ultimately, a dilute leachate.
The successful treatment of municipal landfill leachates has
been demonstrated. The selection of an appropriate treatment
depends on the character of the leachate and the proximity to
a wastewater treatment facility. Whereas conventional wastewater
4
-------
treatment poses a legitimate treatment option for landfills close
to a treatment plant, leachates of landfills distant to a plant
may be treated by aerated lagoon, land application, or by con-
ventional biological and/or physical/chemical treatment
techniques. Regardless of the system selected, leachate
recireulation can be beneficial in reducing the organic strength
and volume of leachate, as well as reducing treatment costs.
Vadese zone monitoring has received little attention in
contrast to the advanced te hnology for monitoring in the zone
of saturation. This discrepancy reflects, to a large extent,
the greater complexity of flow in the vadose zone, compared to
saturated flow, and the related problem of obtaining a repre-
sentative sample for analyses. Incorporaton of vadose zone
devices can provide an early warning of poten€ial ground-water
pollution. If remedial measures are implemented prior to the
onset of extensive ground-water contamination, the associated
renovation costs could be reduced significantly. Additionaly,
an effect±ve vadose zone monitoring network could reduce or
largely preclude, the requirements for ground-water monitoring.
The savings in costs for construction of ground-water wells
could be significant, particularly, in western regions where
water tables are often hundreds of feet deep.
RECO 4ENDATIO TS
Review of the literature and case study materials revealed
a ni. ber of areas that warrant further study. These are saim
marized in the following discussion.
Leachate generation cannot be accurately quantified due to
the inadequacy of much of the data concerning the availability
of water, landfill face conditions, and refuse conditions.
Therefore, additional research should be conducted regarding
such variables as water contribution from refuse degradation,
refuse permeability, evapotranspiration, surface runoff co-
efficients, and runoff curve numbers. The acquisition of
reliable data may be applied to existing models for improved
predictive capabilities or may be used to derive a modified
technique for prediction of leachate generation.
With regard to leachate composition, the persistence and
viability of enteric bacteria, viruses, and parasites in land-
fills and 3.eachates require further study. Because the leaching
behavior of heavy metal contaminants is highly variable, add-
itional study is required on the roles of complexaton, adsorption,
and precipitation in influencing heavy metal mobility. Also,
leachate composition modeling efforts clearly need a broader
empirical base in the areas of solid waste hydraulic properties,
contaminant leaching patterns, and microbial activity.
5
-------
Additional research should be directed towards the develop-.
nient of simulation technology for leachate migration models.
A better understanding of the different mechanisms operative in
waste disposal systems is required in order to construct more
efficient and reliable.simulation models. In particular,
modeling efforts should be concerned with single—ion, key-
parameter models, rather than complex approaches which are both
inwieldly and impractical. Further research should be directed
towards the development of biodegradation models which would be
of great value in predicting the stabilization rate of landfill
refuse. Attention should also be given to the development of a
simplified chemical/physical model which incorporates the ef-
fects of soil chemical and physical properties.
Leachate control techniques which require. additional study
include the effects of nutrient and alkaline industrial waste
in ecticn into a landfill for the purpose of accelerating the
stabilization process. Additional work should also include the
effectiveness of natural and synthetic sorbent materials in
removing contaminants from leachate. Leachate recycling should
also receive attention since it appears to offer both increased
stabilization and an effective leachate treatment scheme. The
effectiveness of a leachate withdrawl and injection well system
should also be investigated.
Vadose zone monitoring requires additional study to better
define both the dominant phenomena occurring in this zone as
well as optimum monitoring equipment. The integration of pas-
sive or non-sampling vadose zone devices should be studied to
determine their possible use. Sample retrieval devices for
both the vadose and zone of saturation should be examined to
determine their effects on obtaining a representative sample.
Attention should also be given to potential environmental
impacts of surface run—off from sanitary landfills.
6
-------
SECTION 3
LEACUATE GENERATION
fl TRO DUCT ION
A quantitative description of leachate volume generation
from sanitary landfills is essential for determining effects on
water quality and the value of control methods. Methods of
estimating leachate generation have received considera.ble atten-
tion. Field tests, physical and mathematical models, and
monitoring of actual installations have all been performed with
the intention of increasing the accuracy of these predictive
methods. Because of the numerous factors involved in leachate
generation; however, estimates are only accurate within an order
of magnitude (Duvel, et al., 1979). This lack of accuracy is
due primarily to the ina tlity to measure the surface runoff
coefficients and evapotranspiration of the landfill site and
adjacent area. Other problems include estimation of the in situ
permeabilities (i.e., hydraulic conductivities) of wastes, and
of underlying and cover soils. All of these factors contribute
to the complex problem of estimating leachate volume.
In this section, the various factors affecting leachate
generation will be discussed and pertinent generation concepts
and measurement techniques will, be evaluated. Examples of
selected leachate generation models are provided. Calculations
are compared to the actual field data where this is feasible.
The advantages and limitations of existing models are discussed.
FACTORS AFFECTING LE.ACHATE GENERATION
Leachate volume generation at a landfill site is dependent
on many factors; in general, it is determined by the following
four conditions (Merz, 1954: Remson, et al., 1968; Qasint, at
al., 1970; Tunga.roli, 1971; Rover, etTl ’ 1973: Caffrey,
IL, 1974; Fenn, at al.., 1975: Dass7e al., 1979; Funqarofl
! • 1979; and Lutton, at al., 1979):
• Availability of water;
• Landfill surface conditions;
• Refuse conditions; and
7
-------
• Underlying soil conditions.
The major factors which affect the above conditions are summa-
rized in Figure 1.
Factors affecting the availability of water include direct
precipitation surface run-on, ground-water intrusion, irr iga-
tion, refuse decom ositiofl, and codisposal of liquid waste or
sludge with refuse. Of these water sources, the primary contrib-
utor is direct precipitation (Caffrey, et al., 1974; Fenn, et
al., 1975; Dass, et al., 1977). Precipi a Ion in adjacent areas
can be channeled onto the landfill site (i.e., surface run-on).
If refuse is placed in or just above the ground-water table, a
ground-water mound may saturate the refuse and result in consid-
erable quantities of leachate. Irrigation applied to the
landfill surface may also contribute to leachate generation.
This is especially true if the completed landfill is used as a
golf course or for similar recreational purposes (Fenn, et al . ,
1977). Water from refuse decomposition is another contriEutor
(Remson, et al., 1968; Rovers, et al., 1973; Dilaj, et al.,
1975). Co—disposal of liquid wastes or sludge in municipal
landfills may also contribute to leachate generation (Stone,
1974; Parkhurst, 1978). ALl of the above mentioned leachate
sources will be discussed in detail later in this section.
Water reaching the landfill surface by precipitation,
surface run-on, or irrigation may either evaporate or transpire,
infiltrate through the landfill surface, or leave the site as
surface runoff. Pathways available to the water and the distri-
bution of water among them will depend principally upon the
landfill surface conditions. Surface conditions which may affect
leachate generation include vegetation, cover material (type,
dimension, compaction, permeability, moisture content, etc.),
surface topography, temperatures humidity, and wind speed above
the landfill (Chow, 1964; Remson, et al., 1968; Rovers, et al.,
l 73; Caffrey, et al., 1974; Dass, et al., 1977; and. tutton,
s ! ;‘ 1979).
Once field capacity of the surface cover material is
attained, leachate will percolate through the refuse. Retention
and transmission characteristics of the refuse control the
percolation rate. Leachate may be channelled through the refuse
and dispersed by refuse intermediate cover layers, or may seep
through the pores of the refuse. If channelling does not occur,
leachate will not be produced in a landfill until at least a
portion of the refuse reaches field capacity. Any additional
moisture will then cause leachate movement. Underlying soil
*
Field capacity is defined as the maximum moisture content which
a soil or a solid material can retain in a gravitational field
without producing continuous downward percolation.
8
-------
______ ____ — ‘
C. LI p —- . ‘p __ ti . .
___ _______ ____ L —— - ptm .’
4 ‘ —- -i — —
I _ —I’ t s
Pig e 1. Tactors affecting leachate volume generation.
9 rRep,oduced from
Jbest availabi. copy
S.
-------
conditions can modify both the rate and amount of leachate
generation when soils underlying and. surrounding the site
have lower permeabilities than the cover soils and refuse.
Details of the leachate percolation through the refuse and
the effects of underlying soils will be discussed later in
this section.
The quantity of leachate produced may also differ consider-
ably with management or operating practices (Duvel, et al.,
1979). Operational factors may include cover material handling,
watering prior to compaction, daily variation in compaction and
cell construction, and variation in waste composition (e.g.,
municipal refuse alone or municipal refuse plus sludge or
industrial wastes; milled or ui milled refuse; etc.). Basically,
these factors, may affect the conditions of cover material and
refuse fill, and; therefore, affect the leachate generation.
Specific information describing the effects of management and
operating practices on leachate generation; however, is not
available.
Precipitation
In most cases, precipitation will be the principal source
of leachate, provided the contribution by irrigation, ground-
water intrusion, or liquid wastes codisposal, is minimal (Fenn,
et al., 1977). Precipitation includes rainfall and snowfall.
Rai !all can affect leachate generation more directly and rapidly
than snowfall (Caffrey, etal., 1974; Fenn, etal., 1975; and
Dass, et al., 1977).
Tour rainfall characteristics which affect 1eac ate gene:-
ation are amount, intensity, frequency, and duration (Chow,
1964). Rainfall is usually used to represent the total quantity
of rain water reaching the ground surface during a certain time
period (e.g., a month or a year) for a given location. This
rainfall amount may result from a single storm or multiple
storms. A niiber of rainfall analyses in terms of amount are
generally available in the forms of mean-annual-, mean-seasonal-,
mean-monthly—, and mean-weekly-rainfall data (Chow, 1964). Many
gross estimations of leachate generation are on the monthly basis
(e.g., Remson, et al., 1968; Fenn, et al., 1975; and Dass, et al.
Amount: Rainfall amount is the quantity of rain reaching the
ground surface in terms of depth of water (e.g., mm. or in.).
Intensity: Rainfall intensity is the amount of rain per unit
time period (e.g., mm/hr or in./hr).
Frequency: Rainfall frequency is the repeated occurrence of a
certain rainfall characteristic at a given interval.
Duration: Rainfall duration is the duration of a storm (e.g.,
mm or hr).
10
-------
1977). Such estimation methods for leachata generation usually
neglected the effects of rainfall intensity, frequency, or dura-
tion. Rainfall intensity was found to influence the impact of
raindrops on the surface soil particles. Such an effect could
change the infiltration rates (Chow, 1964) and thus change the
leachata quantity produced. Rainfall frequency and duration also
could affect leachate generation through their influence on
infiltration and surface runoff. For example, an annual rainfall
of 10 in. in the southwestern United States probably will not
produce any significant leachate if such a rainfall amount occur-
red during numerous storms spread over the entire year period.
But leachate could very easily be generated if such a rainfall
amount is caused by a few deluge—type rainstorms lasting hours or
days. The quantitative information on the effects of such rain-
fall characteristics on leachate generation; however, is greatly
lacking.
Rainfall varies with both the physiographical setting and
the season and is well—documented in most hydrologic literature.
Regional effects, such as latitude, mountains, large lakes, and
oceans, influence rainfall patterns. Dry s .mimers on the West
Coast, the stability of monthly precipitation in New England, the
winter and spring maxima from the Central Gulf to the Ohio Valley
states, and the sudden transition from June drought to July
thunderstorms in Arizona, are notable examples (Chow, 1964; and
Viessman, et. al., 1977). Leachate quantities from rainfall are
likely to 5 iss in the western states than in the east (refer
to Figure 2).
Significant variations in rainfall may occur in certain
localized areas, especially in mountainous regions, requiring
caution in using generalized rainfall data or hydrologic maps.
An abnormally wet or dry year may occur which also cannot be
shown by generalized maps or data. In using rainfall data for
leachate generation calculations, it is advisable to seek data
specific to the landfill site. Methods of determining rainfall
are discussed later in this section.
Snowfall on the landfill may also yield water for subsequent
leaching. Snowfall varies with season and location. Figure 3
depicts the mean annual snowfall in the U. S. In estimating the
availability of snowfall as a leaching source, the amount of
snowfall, the subi.imation of snow, and the melting rate of snow
must be considered. The determination of the portion of the
snowpack lost by sublimation is very difficult. The sublimation
is probably small during the active snowmelt period when dew
points* ar above 0°C (32°F) (Chow, 1964) • In many mountainous
Dew point: Dew point is the temperature at which an air mass
just becomes saturated if such air mass is cooled at constant
pressure and without moisture addition or removal.
U.
-------
lean
annual
rainfal
LEGEHD
Unit In
Inches
>60
56—
60
51 —
55
36 —
50
26 -
16 -
<‘5
35
25
7! E 1
t I1
• Figure 2. Mean annual rainfall in the United States
based on 40 years of record (1899-1938)
(prepared after Chow, 1964)
0
a.
C
-.n
—m
0 -
‘0
I
-------
U
96—12 1 mi
32—96
16—32
aD
1 igure . Mean annual snowfall In the United States (Prepared after Chow, 1964).
I-
(A)
Leyenda
Bnowfall (In.)
a128
-------
areas during the winter, dew points are low and the total evapo-
ration loss from the snow pack in areas of low snow accumulation
may be significant (Chow, 1964). In the winter and early spring,
before sn wmelt runoff begins, sublimation losses can be assumed
to be about 13 r (0 .5 in) of water equivalent per month (Chow,
1964). The melting rate of snow can be affected by temperature,
wind speed, snowpack characteristics (the ripening of snow), site
conditions, antecedent snowfall conditions, and rainfall. Melt-
ing usually occurs at rates considerably below the infiltration
capacity of unpacked soils (Chow, 1964). Therefore, surface
runoff from snowmelt may be rare. aowever, when the landfill
surface is steep and compacted tightly which reduces infiltration
sudden changes in temperature may result in surface run—off from
snowmelt. In general, it is suggested that 250 m (10 in) of
snow is equivalent to 25 (3. in) of rainfall (Linsley, et al.,
1972). A mean annual effective precipitation* in the U. S. is
calculated on this basis and is shown in Figure 4. This figure
is a modification of Figures 2 and 3 as presented previously,
and can be used to indicate the potential relative source of
leachate in various geographic areas in the United States.
Surface Run-On or Runoff
Major land surface conditions affectinc’ surface run—on or
runoff include surface topography, covermaterial, vegetation,
soil permeability, antecedent soil moisture, and artificial
drainage (Chow, 1964; Caffrey, et al., 1974: and Dass, et al.,
1977). Surface topography (size, shape, slope, orientation,
elevation, and surface cor.figu.ration) control flow at the surf ace
xnong these topographic factors, slope may be most significant.
Surface soil material, soil, permeability, and antecedent soil
moisture affect infiltration rates which in turn can affect
surface run—on or runoff. Details of these effects are dis-
cussed later under “infiltration”. Landfill surface vegetation
also has significant effects on surface run—on or runoff. The
significance of this influence depends on the species compositiont
age, and density of the vegetation, as well as the season.
Quantitative data showing the effects of slope, cover material,
and vegetation, on the surface runoff are discussed in the next
subsection entitled “Concepts and Techniques Describing Leachate
Generation.”
Ground-water Intrusion and Irrigation
Landfill leachate from ground-water intrusion will occur
if the base of the landfill is below the ground-water table.
The vo1 e of saturated refuse, contact time, and flow direction,
will affect leachate production. Precise measurement of under-
flow is not feasible. Detection of underfiow and reasonable
* Effective precipitation = rainfall + 1/10 snowfall.
14
-------
Figure 4. Mean annual effective precipitation (calculated from Figures 2
and 3).
Le9ends
Effective
PrecipititiOn lie.)
>96
48-96 1111
24-48
16-24
0-16 LI
*
Effective Precipitation — rainfall + 1/10 anowfall
-------
approximation of flow rates are possible; however, with a hydro-
logical investigation and calculations using Darcy ’s Law (Chow,
1964; DeWiest, 1967; and Fenn, et al., 1977). Quantification
techniques for the ground-water intrusion estimation are shown
later entitled “Concepts and Techniques Describing Leachate
Generation 1 ’.
Total amounts of irrigation water applied to a landfill
can be measured with a flow meter. Because the water loss
caused by surface runoff during, irrigation is negligible, the
leachate produced is the difference between irrigation input
and evapotranspiration.
Refuse Decozn ositiofl
Microbial decomposition of the biodegradable organics in
refuse can generate water, which will contribute to the refuse
moisture for leachate production. Both aerobic and anaerobic
reactions may generate biochemical water (California State
Water Pollution Control Board, 1961; and aoluche, 1968):
(Organic Matter) + (Oxygen) Aerobes (Water)
+ (Carbon Dioxide) + (Acids) —
+ (Inorganic Minerals) (e.g., PO 4 SO 4 ,
NC ;, or
(Organic Matter) Macrobes (Water) + (Methane)
+ (Decomposed Organics) + (Carbon Dioxide)
+ (Axmuonia) + (Inorganic Minerals).....(2)
The extent and rate of microbial activity and consequent
water qeneration depend mainly on the amount and pH of inter-
stitial moisture, temperature, presence of oxygen, the
composition and particle size of the refuse, the type of organ-
isms present, and the degree of refuse mixing (Remson, et al.,
1968; Rovers, et al., 1973; Caffrey, et al., 1974; and
et al.,, 1975). In general, decomposition is faster under
i r Sic than anaerobic conditions. California State Water
Pollution Control Board (CSWPCB, 1961) reported that under
aerobic conditions, decomposition may reduce volatile matter and
carbon concentrations by 50 percent in a year. For anaerobic
conditions, although precise data is not available, the decom-
position rate is known to be substantially lower. Qualitative1y
the no na1 series of events would begin with an initial rapid
aerobic decay of the refuse. The refuse would gradually become
anaerobic as easily digested substances are depleted and slower
heat production reduces convectional transport of oxygen. After
16
-------
a period of anaerobiosis, which may last several years, decay
slows, and the aerobic processes may be supported by the oxygen
diffusing into the fill (SCWPC3, 1961). The rate at which water
is produced by decomposition would vary as these events occur.
Microbial phenomena occurring in a landfill can be describ-
ed qualitatively. However, the present information is not
adequate to calculate rates of refuse degradation and water
generation (Remson, at al., 1968; Rovers, al., 1973; and
Dilaj, et al., 1975). Research conducted by the California
State W erPollution Control Board (1961) measured decomposition
water production in a municipal landfill operated by the City of
Riverside. This study found that decomposition water may reach
42 /m (0.5 in./ft) of rófuse. No other data for biochemical
water production are available. Based on the above measured
decomposition water, it can be concluded that leachate generated
from refuse decomposition is very minor compared to other
sources of leachate.
Codjsoosal. of Municipal and Industrial Sludges
- Codisposal of municipal and industrial sludges in municipal
landfills may contribute significant amounts of leaching water.
The volume of water available for leaching may be affected by
the type and amount of sludge, moisture content, moisture
holding capacity, and the effect of compaction and decomposition
on the release of water from sludge (Charlie, at al., 1979;
and P arkhur zt, 1978).
Experimental works performed by the Los Angeles County
Sanitation District (Parkhurst, 1978) provides information on
the generation of codisposal leachates as function of sludge/
refuse ratio, sludge moisture content, landfill depth, and
effects of coxn action and decomposition (Figures 5 to 7). Figure
5 shows the maxim non-leaching sludge to refuse (wet weight)
ratio as a function of sludge solids content at various landfill
depths. The data demonstrates a large dependence of the max-
imt a non-leaching sludge to refuse ratios on the depth of the
disposal mixture. For example, at 25 percent solids, a sludge
cake can be placed at a 0.93 to 3. ratio at the 7.5 meter depth
without leachate generation. However, at the 100 meter depth,
the non—leaching ratio was reduced to 1 part sludge to 9 parts
refuse. This phenomenon is mainly due to the pressure effects
resulting from the surcharge provided by successive lifts
(Parkhurst, 1978). Figure 6 illustrates the leachate release
from the codisposal landfill caused by surcharge loads when the
non—leaching ratio is exceeded. Figure 7 further depicts the
ultimate leachate production resulting from both surcharge loads
and decomposition.
17
-------
4
4J
0)
0
- ‘ .4
i
1.4
(0
I-’ ::j
0)
0
4.)
a)
‘ c i
‘.1
3
I
S Cake o1ids
0
0 $0 20 30 40 50 60 70 80 90
$00
Figure 5. Maximum non-leaching sludge to refuse ratio (Parkhurst, 1978).
-------
% Cake Solids
Leachate production resulting from sludge compaction
(Pa khurst. l978 ’.
-
- -
— —
— - —NON LEACHING CAXE
SOLIDS AT VARIOUS
LANDFILL OEP1HS
—— —
— —
‘
‘
LANDFILL DEPTH I
,— lOOM (330 F ,.)
/
—ISMI5OFP.)
,—7.5M(ZSF?.)
\
\
l0000
9000
8000
7000
6000
5000
4000
3000
2000
1000
0 .
N
II
10 20 30 40 50
GO 70 80 90 100
Figure 6.
-------
z
0
C
C
C
0
0
4J
I
Figure 7.
Leac1 ate production resulting from sludge compaction
a d decomposition (Parkhurst, 1978).
% Cake Solids
20
-------
Ev ape tr an so iratio n
Evapotranspiration is the si. of water loss by evaporation
and transpiration (plant water consumption). Factors control-
ling evaporation and transpiration are well recognized, but
quantitative evaluation is difficult because of their inter-
dependent effects (Chow, 1964). For a free—water surface, the
rate of evaporation depends on the vapor pressure of the body
of water and that of the air. Major factors affecting these
vapor pressures include temperature (both water and air), wind,
him idity, atmospheric pressure / quality of the water, and the
nature and shape of the surface. With the exception of the
last two, the factors vary seasonally and. geographically.
Average annual pan evaporation presented in Figure 8 illustrates
geographical variations 1 From this figure, the mean annual net
effective precipitation, which is the difference between “effect-
ive precipitation” (Figure 4) and “potential evaporation (Figure
8), for the t7. S. can be prepared (Figure 9). Teqative values
of net effective precipitation represent potential water loss
from the land surface to the atmosphere (i.e., evaporation >
effective precipitation). Landfills located, in areas with
negative values of net effective precipitation are expected to
have a negligthle source of leaching water from precipitation.
Evaporation is proportional to the vapor pressure difference
between water in the ccii. and the air, so temperature increase
may not prpportionately affect the evaporation rate and thus, a
high correlation between the air temperature and evaporation
cannot be ex ected. Wind removes water molecules from the
region near the surface, thus also increasing evaporation rates.
Hi idity and atmospheric pressure are inversely related to
evaporation. Evaporation rates decrease as dissolved solids
concentrations increase. The evaporation rate decreases about
1 percent for each 1. percent increase in specific gravity
(Zeen, et al., 1926; Fisher, 1927; Penman, 1948; V’eibmeyer,
et al. ,r955, C1 ow, 1962).
Evaporation of water from soil, surfaces is controlled by
the same factors that evaporate water from free-water surface.
A difference is that soil, particles tend to bind water molecules
against evaporation (Penman, 1948). This attractive force is a
function of the moisture content of the soil and the character-
istics of the soil. unsaturated wet soils maintain a nearly
constant rate of evaporation over a range of moisture content
(Keen, et at., 1926; Fisher, 1927). Evaporation is reduced,
according to veibmeyer, et j ., (3.953), at or below the permanent
wilting point*o.f the soil ’ Evaporation from the soil surface
* Permanent wilting point, or permanent Wilting percentage, is
the moisture content of soil at which soil cannot supply water
at a sufficient rate to maintain tiargor, resulting in wilting
of vegetation.
21
-------
Evaporation (in.)
44 M 64-80
a28—144 fi 48—64
112—128
96—112
80-96
Legend:
0-32 Li
32—48
Figure 8. Mean annual pan evaporation
(after Chow, 1964).
-------
Hut Ettective Precipitation tfective Precipitation — EveI ratlOn.
(Notez neqetive valuea represent yearly potential water loss trois the land
surface to the atausphuru)
Figure 9. Mean annual net effective precipitaLion (prepared from FiguroB 4 and B).
‘U
( .4
Legend:
Net Effective
(In. ,yr)
D
Oto-lO
-------
will continue as long as the shallow surface layer (about 100 itim
(4 j ) for clays and about 200 mm (8 in.) for sands) contains
moisture above the permanent wilting point. Evaporation from
deeper soil, is insignificant (Chow, 1964).
Factors af fecting transpiration are physiological and envir-
onmental. Physiological factors include the type and density
of vegetation, leaf structure, and plant condition and age.
Table 1 provides examples of seasonal transpiration ranges for
several types of vegetation (Urqi.thart, 1959). The data shows
transpiration has the potential to remove significant fractions
of water from the soil.
TABLE 1. APPROXIMATE SEASONAL TRA SPIRATIO 7 OF SELECTED VEGETA-
TIO T(URQURART, 1959)
Vegetation
Transpiration (iz Tqrow ing seaso ri)
Alfalfa and Clover
>2.5
Coniferous Trees
4—9
Deciduous Trees
7—10
Lucern Grass
26—65
Meadow Grass
22-60
Oats
28—40
Rye
>18
Wheat
20—22
Among the environmental factors influencing transpiration
are season, temperature, solar radiation, relative humidity,
wind speed, and soil, moisture when the permanent wilting point
is reached (Caffrey, et a].., 1974; Molz, et al., 1974). The
season of the year and the associated sunlTght conditions will
affect the temperature of the leaf and thug, transpiration.
Solar radiation is particularly important because it stimulates
the “guard ce11s1 * to open the leaf pores. About 95 percent of
daily transpiration occurs between sunrise and sunset (Molz,
et a].., 1974). The removal of water vapor next to the leaf
sur! ce by wind also can increase transpiration. With wind
speeds of eight to twenty—four kilometers per hour (5 to iS
mph), transpiration rates may be twenty and fifty percent higher
than at zero wind speed (C1 ow, 1964; Molz, etal., 1974).
Transpiration is also affected by the soil moisture content when
it is reduced to the permanent wilting point (Chow, 1964).
Evaporation and transpiration are difficult to measure
separately. Consequently, water lost to the atmosphere is most
* Pores on the leaf surfaces that are surrounded by special ceil
are called “guard cells.”
24
-------
often reported as total evapotranspiratiorl. Methods of calculat-
ing’ evapotranspiration are presented in the next subsection.
Infiltration
Infiltration is the flow of water through the soil surface.
It is only one of the possible fates for water applied to the
landfill surface, the other possible fates being runoff and
evapotranspiration. Factors affecting the rates of surface
ra .m tf and evapotranspiration. .will, therefore, indirectly affect
infiltration. The characteristics of.the cover material (per-
meability, moisture content, porosity, organic content, degree
of comvaction, thickness of surface permeable layers, etc.),
surface topography (slope, surface, storage capacity, etc.),
vegetation, and underdrainage condition were found to influence
the infiltration directly (Chow, 19647 Remson, et al., 1968;
Gray, . , 1963; Caf fray, j j ., 1974; Viessinan, at al.,
1977; and Lutton, et al., 1979)
The most important factor affecting infiltration is usually
the moisture content of the landfill cover material. Infiltra-
tion rates are reduced with an increase in the moisture content
in the top soil (Chow, 19647 Gray, et al., 1968). This relation-
Ship can be illustrated by the infi ration rate changes after
a rainstorm, as shown in Figure 10. The maximum infiltration
rate occurs at the beginning of the rainstorm when the soil is
relatively dry. The rate decreases first rapidly and then
slowly approaching a stable minimum, the minimum infiltration
rate usually approaches the percolation rate (to be discussed
later) of the soil. The changes of infiltration rates versus
time during a rainstorm can be approximated by the following
eq ’ .iation (Viessman, j, ., 1977):
-kt
f • fc + f Ø fo) e e_ — — — e —eee — — — —a __ __ __ — _ __ ( 3)
where:
f — infiltration rate at some time t;
k — a constant representing the rats of decrease in f;
• a final or equilibrium rate; and
— the initial infiltration rate.
Infiltration into a landfill surface is also determined by
the structure of the surface layer of soil. Important character-
istics of soil tending to retard infiltration axe high non-
capillary porosity and high organic content of the subsoil
* Non—capillary porosity is the fraction of air—filled voids
in a porous medium at field capacity.
25
-------
I
z •i-I
o
H
14 >4
v E 4
[ - 4 ( 1 )
1 -1z
H14
11414
zz
HH
Figure 10. Relationship of rainfall, runoff, and infiltration during a rainstorm
INTENSITY
to = the initial infiltration rate
fc = the final stable infiltration rate
to
TIME (lirs)
event.
-------
(Chow, 1964). Raindrops falling on bare soil tend to compact
the surface layer and wash fines into the large pores, also
causing a sharp reduction in infiltration.
The Eype and density of landfill vegetation are other impor-
tant factors affecting infiltration. Vegetation protects the
soil surface from the impact of rainfall. It also retards sur-
face flow velocity so that water is retained on the landfill
surface for a longer time.
Water can enter the soil no more rapidly than it flows down-
ward within the soil. Underdrainage, therefore, affects
infiltration. Where underlying layers are unsaturated, the
infiltration rate is regulated by the permeability of the upper
layers. Where underlying layers are saturated, the layer with
least permeability will control infiltration (Chow, 1964).
Figure ii. is constructed to compare the potential infiltra-
tion for landfills located in different goegraphical areas of
the United States. This figure is derived from the mean annual
net effective precipitation (Figure 9) and the potential landfill
surface runoff (0.25 is assumed for the runoff coefficient).
This figure .s not to be misconstrued as 5howing the actual
amount of water that infiltrates, but rather provides a compari-
son of potential landfill infitration rates. Such potential
exists for the northwestern, northeastern, and mid-eastern
United States. Rowever, this does not mean that leachate will
not be a problem from precipitation in other areas. Leachate
could very easily be generated during a deluge-type rainstorm
lasting hours or a prolonged rainfall over a period of days.
Moisture Retention
Figure 12 depicts the moisture retention properties of an
unsaturated soil or refuse (Bauer, 1956). Commencing with
saturation, the water initially drains rapidly and the rate
progressively decreases. Field capacity remains essentially
unchanged unless the vegetation root zone is in the layer of
consideration. A soil or refuse located within the root zone
can be dried to the permanent wilting point as shown in Figure
12. Approximate field capacities and wilting points for soil
and refuse are discussed in the next subsection.
Percolation
Landfill percolate is generally defined as the quantity of
water which exceeds the refuse field capacity. Theoretically,
the water movement through a compacted refuse cell will not
occur until the refuse field capacity is exceeded. Practically,
because of the heterogeneous nature of the refuse, some channel-
ing of water may occur causing some percolation to appear prior
to attainment of field capacity. In field situations, the
27
-------
PotentIal Infiltration
n/acre/vr million qailacre/vi
>0.815
tflfll
15
to
30
0.401
to
0.815
10
to
15
0.212
to
0.407
5
to
10
0.136
to
0.272
L 1
0
to
5
0
to
0.136
<0
‘0
LI.J
Figure 11. Potential infiltration for landfills in various geographical areas of the
United States (See text for explanation).
4J
-------
PE ANENT
WiLTING POINT
_ — — — — — —
SOIL ISflJRE CONTENT CHANGES
WITh PL TS GRQ 4ING CN SOIL
figure 12. Moisture content as a function of time for a drain-
ing soil or refuse (Bauer, 1956).
SATURAT I ON PC I NT OF SO I L
— — — a — — — — — a — — — — — — —
it
— — — — — — 1 FIE CAPAC In —
SOIL OISTURE CONTENT C’ ANGES
WIThOUT PLANTS
a — — a — a
0
TINE
29
-------
significance of such a channeling effect is difficult to estj—
mate. Due to the fact that the refuse seldom achieves satura-
tion (e.g., Remson, et ., 1968; Rover, 1973; and
Lutton, et al., 1979), and the high water absorbing capability
(to be disci4ssed later) of most refuse, the channeling amount
may be small. Fenn, j . (1975) suggested that the channeling
amount should be small and certainly not a continuous flow, and
was ass zned negligible in their percolation estimations.
As water percolates vertically through a landfill soil.
cover, and into the refuse, layers with different structures and
properties are encountered. If water is moved in a saturated
layer of the refuse (which may Occur for top layers of the refuse
right after a deluge—type or prolonger rainstorm), the gravity
and pressure forces, which form the hydraulic gradient, will
influence the water movement. The velocity and flow rate of
such percolates can be expressed by Darcy’s Law as follows:
V. Ki (4)
and Q KAi (5)
where:
V a flow velocity;
Q a flow rate;
K refuse permeability;
A a cross-sectional area perpendicular .to the flow
direction; and
j hydraulic gradient.
The refuse permeability as shown above is one of the most i.mpor—
tant parameters for determining percolation rates in the satura- .
tion condition. it is generally concluded that the permeabi1it .
of refuse is affected by refuse particle size and density.
ifowever, the quantitative relationship is greatly lacking. A
recent study conducted by Funga.roli, et al. (1979) could not
establish a Significant relationship between saturated perme-
ability, density and milled refuse size. Nevertheless, the
ranges of refuse permeability was available and reported to be
i0’ cm/sec for the low density (1e s than 500 pounds per cubic
yard) large particle refuse, to l0 cm/sec for the high densjt
(greater than 500 pounds per cubic yard) large and small
refuse (Fungaroli, et al., 1979).
Water movement in an unsaturated layer of the refuse fill
j affected by thermo—osmosis and capillary forces (Chow, 1964;
.. Wiest, 1967, in addition to gravity and pressure forces,
thermO-osmosis forces can move water by diffusion of water vapo
in the soil or refuse pores. This phenomenon may be signif1can
especially when warm temperatures exist resulting in water
vaporization (De Wiest, 1967). In an active landfill, the
theriflophilic condition could result in such an effect. However,
30
-------
a study or. the thermo—osmOsis effect for water movement in the
unsaturated landfill is lacking. Chow (1964) stated that the
capillary poter.tial gradients are large and usually control the
water movement throughout most of the moisture content ranges
of the unsaturated layer. The modified Darcy’s L aw for un-
saturated flow was thus derived, mainly based on gravity and.
capillary potential (Chow, 1964; and Philip, 1969). Zn this
derivation, the permeability, K, in Equations (4) and (5)
becomes a nonlinear function of moisture content, G. Figure
13 is an example of such a nonlinear relationship.
Darcy’ s Law for unsaturated flow indicates that any mois-
ture movement must occur in the direction of decreasing gravity
and capillary potential (Chow, 1964). The gravity potential is
constant in a landfill. In order for a wetting front to move
downward from a fine medium into a coarse medium in an unsat-
urated refuse, the fine medium must become nearly saturated to
crea.te sufficient capillary potential (Colman, et al., 1944).
In the case of flow from a coarse medium into an un r1ying
finer stratum, the rate of percolation is controlled exclusively
by the fine particles because of the relatively low permeability
of the fine particles. Where water is applied more rapidly than
the fine—grained percolation capacity, either the upper layer
will become saturated, or the water will flow Laterally.
After the percolate reaches the landfill/underlying soil
boundary, leachate will form. Percolate is identicai. with
leachate generated if ;round-water intrusion, changes in mois-
ture retention, and biochemical contributions are negligible
(Fenn, et al., 1977; Lutton, et al., 1979). Percolation can be
estimated by direct measurement or calculation. Percolation
is measured with a sub—surface water trap. Water traps and
their application are discussed in the literature (Fenn, et
1977). The calculation methods for percolation are discussed
in the next subsection.
CONCEPTS AND TECHNIQUES DESCRIBING LEAC1 ATE GENERATION
Numerous mathematical methods have been used for quantita-
tive estimation of the volume of leachate generated from land-
fills (Thornthwaite, et al., 1957; Remson, et al., 1968; Fenn,
etal., 1975; Dass, eTaI, 1979; Duvel, etal7 1979; and
Lutton, et al., 1979). Although the approaches employed in
these metho vary, they are all basically derived from the
water balance or water budget principle. The water balance
principle is based on: (1) a one—dimensional flow model and
conservation of mass relationships among various components
of the leachate sources, and (2) the retention and transmission
characteristics of the refuse and cover soil. The generalized
water balance relationships are shown in Table 2 and are
illustrated in Figure 14.
31
-------
12
C
6
>4
4
2
VOLUMETRIC MOISTURE CONTENT
Figure 13. Example showing the effect of moisture content on
permeability for a clay soil (a f tar Philip, 1969).
10
8
0
32
-------
TABLE 2. GENERALIZED WATER BALANCE EQUATION AT A Mt ICI?AL
LANDFILL SITE
wp+WSR+WIB aI+R (6)
where:
Input water from precipitation
W Input water from surrounding surface runoff
= Input water from irrigation
I Infiltration
R a Surface runoff
PER 5 a — E — Ass
PEBR I - E — As 5 + W — AS (8)
+ - ASa
where:
PER 5 and PERR a Percolation in soil and refuse respectively
— Water contributed by solid waste decom—
position
AS 5 Change in moisture storage in soil
SR = Change in moisutre storage in refuse
£ a EvapotranspiratiOn
a.x d I. a PER. + ( 10
where:
L Leachate generation
— Input water from underf low
33
-------
p
VIRGIN GROUND
W 2k R
PRECIPITATION +
I RRIGATI ON
INFILTRATION
EVATRANSPI RAT ION
COVER SOIL
I 4 EACIIATE
Fijure 14. Municipal landfill water balance.
-------
Equation (6) in Table 2 illustrates the water balance
re1ationshi of the landfill surface. Water contributed by
precipitation (We), surface runoff (WSR) , or irrigation (W )
will either become surface runoff CR), or infiltrate into tne
cover soil (I). A portion of the infiltrated water leaves by
evapotranspiration (E) and part will recharge the cover soil.
Once the field capacity of the cover soil is attained, vertical
percolation (?ER ) will occur. The quantity of percolation
through the cover material can be calculated by using equation
(7). At first the percolate from the cover material will be
absorbed by the refuse. As the landfill refuse reaches field
capacity, refuse percolate (PERu) is generated in quantities
described by equation (8) or (9). The refuse percolate will
eventually evolve as refuse leachate (L). If ground—water
intrustion (W ) occurs, leachate generation estimates can be
modified usin ’ equation (10) (Table 2).
Because leachate generation mechanisms are site specific,
any estimation method requires knowledge of site topography,
geology, hydrology, climatology, and meteorology. The many
poorly known factors in the water balance equations often
require empirical, rational, or experimental methods. These
different approaches reflect the wide variety of concepts
describing leachate generation. In this report, only those
techniques which are appropriate for municipal landfills are
discussed.
Determinin Precibitation
The amount of precipitation is expressed as the amount of
water (in inches or millimeters) which would accumulate on a
sealed level surface. For the purposes of leachate quantIty
estimation, existing precipitation records or gauge measurements
may be used to determine precipitation quantities.
Existing Precipitation Records-—
The U.S. Weather Bureau collects precipitation data for
the U.S. The methods of handling the data and its publication
have varied greatly since the inception of the Bureau in 1891.
Records are presently stored at the National Climatic Center at
Asheville, North Caro1ina Local, regional or national pre-
cipitation data can be obtained from this agency.
Several states and local agencies publish data from rain
gauges maintained by them. Precipitation data may also be
obtained from military installations, local airports, harbors,
fire stations, and universities. Independent data maintained by
these organizations are sometimes very useful for specific
* National Climatic Center, National Oceanic and Atmospheric Ad-
m .njstratjon, Asheville, North Carolina 28801, (704) 258-2850.
35
-------
landfill site locations.
Precipitation Gauges-—
Precipitation gauges can be installed on or near a landfill
for extended leachate generation investigations. Precipitation
gauges may be of the nonrecordirtg or recording type. The
standard U.S. Weather Bureau nonrecording gauge is a circular
cylinder 20 cm (8 in. in diameter), which serves as an overflow
can. A funnel shaped receiver of the same diameter is connected.
to a measuring tube with a cross section equal to one—tenth that
of the receiver. When 2 .5 cm (1 in.) of precipitation falls
into the receiver, the measuring tube is filled to a depth of
25 cm (10 in.). Precipitation is measured with a graduated rod
to the nearest hundredth of a centimeter or hundredth of an
inch. Many co mnercially available gauges exist, and the selec-
tion of one should be made with the particular purpose of the
monitoring program in mind.
Recording gauges are also available. The weighing gauge
and the tipping—bucket gauge are two cocmon types. The weighing.
gauge has a bucket supported by a spring or lever balance. Move. .,.
merit of the bucket is transmitted to a pen which traces a record.
of the increasing weight on a clock-driven chart or punched
paper tape. The tipping-bucket gauge consists of a pair of
buckets pivoted under a funnel in such a way that when one
bucket receives 0.01 cm (or 0.01 in. depending on equipment use
of precipitation, it tips and discharges its contents into a
reservoir. The other bucket is then positioned under the funne’
A recording mechanism indicates the time of occurrence of each
tip. The tipping—bucket gauge is well adapted to the measure-
ment of rainfall intensity for short periods. The weighing
gauge is more rugged and can record snowfall.
Determining Surface Runoff
Surface runoff on or adjacent to a landfill site can be
determined by surface measurement, empirical formulas, or
graphic methods. Exact measurement is a difficult task because
of the non-point discharge characteristics of the overland
flow and the large area involved. Nevertheless, actual runoff
measurements may be obtained by carefully fencing a test plot
of the landfill or a test section of the proposed cover system.
The rational equation, one of the empirical formulas, is widely
used by hydrologists and sanitary engineering designers (Chow,
1964; and Mterican Society of Civil Engineers, 1960). Although
this formula is based on a number of assumptions which cannot
be readily satisfied under actual circumstances, its simplicity
makes it a practical tool. Many other empirical formulas or
graphical solutions for surface runoff have been suggested by
researchers. Hundreds have become almost obsolete in design
practice because of their complexity or because the values of
coefficients in the formulas are not known in terms of specific
36
-------
conditions (Chow, 1964). The following section will discuss
only those methods suitable for landfill situations.
Su.rface Measurement-—
Landfill surface runoff caused by precipitation or irriga-
tion can be directly measured either in the field or in la.bora-
tory pilot studies. For field measurements, fencing a test plot
of the landfill will collect the overland flow. If tests are
performed on certain sections of the landfill, representative
vegetation, slopes, soil types, and other landfill surf ace con-
ditions should be considered. A precipitation gauge should be
located nearby to record precipitation daily or even more fre-
quently.
Rational Method-—
The rational method is usually expressed in terms of the
following equation (Chow, 1964):
R a C . P . — (11)
where R and P are surface runoff peak discharge and uniform rate
f rainfall intensity, respectively. A is the landfill area.
The runoff coefficient, C, indicates the fraction of total
precipitation that flows off the suxfac . If the units for P
and A are inches per hour and acres, then the unit or a is in
cfs. Equation All) can also be written:
C 3. 2)
where W is the input precipitation in inches or mm. In this
case, t e units for R become in. or ttun depending on the units
of W 9 .
The key to the successful estimation of surface runoff by
the rational method is the correct choice of the runoff co-
efficient. The runoff coefficient for a particular area depends
on surface characteristics, type and extent of vegetation, sur-
face slope, and other less important factors as described
previously. Considerable work has been done developing more or
less standard runoff coefficients for various agricultural
and engineering situations. Rowever, runoff coefficients for
municipal landfills have received less attention. Some values
of the runoff coefficient were reported by a joint conittee
of the American Society of Civil Engineers and the Water Pollu-
tion Control Federation (Table 3) (American Society of Civil
Engineers, 1960). These values axe for storms of 5— to 10-year
frequencies. Values shown in Table 4 are based on individual
effects of topography, soil type and cover conditions (Bernard,
1932). The “Engineering Manual” edited by Perry, suggests a
sat of runoff coefficients based on cover material and slope
(Table 5) (Perry, 1976). Salvato, et al. (1971) in their land-
fill leachate study also suggested a set of runoff coefficients
37
-------
(Table 6). Because there are so many physiographic and physical
variables that can affect the runoff coefficient, the results
obtained are not consistent. Additional investigation of the
runoff coefficients for landfill sites is required if rational
estimation methods are to be used. Significant infor iation
omitted in the above tables include: the quantitati ie relation-
ships between runoff and precipitation duration, precipitation
frequency, antecedent soil moisture content, and the degree of
compaction of the cover material.
TABLE 3. RUNOFF COEFFICiENTS FOR STORMS OF 5- TO 10-YEAR
FREQUENCIES (American Society of Civil Engineers, 1960).
Surface Condition Runoff Coefficient, C
Grass Cover:
Sandy soil, flat, 2% 0.05 — 0.10
Sandy soil, average, 2—7% 0.10 - 0.15
Sandy soil, steep, 7% 0.15 - 0.20
Heavy soil, flat, 2% 0.13 — 0.17
Heavy soil, average, 2—7% 0.18 - 0.22
Heavy soil, steep, 7% 0.25 — 0.35
Suburban Area 0.25 — 0.40
Pi.ayground 0.20 0.35
Unimproved Area 0.10 - 0.30
Drives and Walks 0.75 — 0.85
Curve Number Method-—
The curve ni.unber method is suggested by the U.S. Department
of Agriculture, Soil Conservation Service (SCS) for predicting
direct surface runoff on agricultural land (SCS, 1972). Esti-
mation of runoff is based on the amount of rainfall, soil type,
land use, land cover, and antecedent moisture condition.
Calculation procedures are as follows:
(1) Determine the antecedent moisture condition-—
The antecedent moisture conditions of cover
soils are classified into three groups, based
en 5—day total antecedent rainfall: AMC I,
38
-------
TABLE 4. RTYNOF ’ COE TICIE S FOR DR .INAGE AREAS WITH DIFFERENT
TOPOGRAPHY, SO IL , AND COVER COND iTIONS (Chow, 1964;
and Bernard, 1932)
Site Description
Value, v’
To ograohy:
E 1at land, with average slope of
1 to 3 ft/m4.1 .e
0.30
Rolling land, with average slope
of 15 to 20 ft/mile
0.20
Hilly land, with average slope
of 150 t; 250 ft/mile
0.10
.
Soi1
Tight impervious clay
0.10
Medium combinations of clay and
loam
0.20
Open sandy loam
0.40
Cover:
Cultivated lands
0.10
Pasture
0.15
Woodland
0.20
* Add values v’ for topography, soil and cover, and subtract
from unity to obtain runoff coefficient.
39
-------
TABLE 5. RUNOFF COEFFICIENTS AS AFFECTED BY COVER MATERIAL
? ND SLOPE (Perry, 1976)
Runoff Coefficient, c
Type of area Flat: slope aolling: slope Eilly: slopi
<2% 2—10% >10%
Grassed Areas 0.25 0.30 0.30
Earth Areas 0.60 0.65 0.70
Meadows and Pasture 0.25 0.30 0.35
Lands
Cultivated Land:
Zmpermea 1e (clay) 0.50 0.55 0.60
Per ea’ole (loam) 0.25 0.30 0.35
TABLE 6. RUNOFF CoEFFICIENTS USED BY SALVATO, ET AL. FOR
LEACHATE ESTINATION (Salvato, et al., 1971).
Percent Percent Surface Runoff
Surface Condition Slope Sandy Clay or
Loam Silt Loam • Clay
Pasture or Meadow 0—5 (flat) 10 30 40
(Surface with 510 16 36 55
cover crop) (rolling)
10—30 (hilly) 22 42 60
No Vegetation 0—5 (flat) 30 50 60
(Raw Soil Surface) 5— 0 40 60 70
(rolling)
10—30 (hilly) 52 72
40
-------
AMC II, and AMC I . Table 7 provides criteria
for these three classifications.
(2) Determine the hydrologic soil group——
The hydrologic properties of a soil or a group
of soils can be classified according to their
properties as described in Table 8.
(3) Determine the land use or cover and the hydrologic
conditions
Land cover is any material (but usually vegeta-
tion) covering the soil and providing protection
from the impact of rainfall. Detailed information
about the land cover is seldom available. It is
therefore necessary to rely on land use as an
index of cover conditions. The types of land use
or cover are listed in Table 9. The hydrologic
condition of the vegetation should also be
determined for runoff estimation. Good, fair,
and poor conditions are designated for indicating
vegetation density. Good conditions are represent-
ed by a high proportion of alfalfa or other close—
seeded legumes or grasses that will improve tilth
and increase infiltration. Poor conditions occur
for raw crops, small grains, and fallow land in
various combinations. A vegetated sanitary
landfill cover will be similar to pasture,
range, or meadow. The classifications of hydro-
logic conditions can be seen in Table 9.
TABLE 7. RAINFALL LIMITS FOR ESTIMATING ANTECEDENT MOISTURE
CONDITIONS (Soil Conservation Service, 1972)
—
A.ntecedent
Condition
Moisture
Class
*
5-day Total Antecedent Rainfall, in.
Dormant Season
Growing Season
I
Less than 0.3
Less than 1.4
II
0.5 to 1.1
1.4 to 2.1
III.
Over 1.1
Over 2.1
* i in. — 2.54 cm.
(4) Determine the value of curve number (CN) -—
After the antecedent moisture, hydrologic soil and
cover conditions are determined, the value of the
curve number (CN) can be determined from the SCS
curve number table. Table 9 provides runoff curve
numbers for various combinations of soils and
41
-------
TABLE 8. EYDROLOGIC SOIL GROUPS tJSED BY T SOIL CONSERVATION
SERvIcE (Soil Conservation Service, 1972)
aydrologic Soil Group Soils Included*
A (Low runoff potential) Soils having
high infiltration rates even when
thoroughly wetted, consisting chiefly
of sands or gravel that are deep and
well to excessively drained. These
soils have a high rate of water trans-
mission.
3 Soils having moderate infiltration
rates when thoroughly wetted, chiefly
moderately deep to deep, moderately
well to well drained, with moderately
fine to moderately coarse textures.
These soils have a moderate rate of
water transmission.
C Soils having slow infiltration rates
when thoroughly wetted, chiefly with
a layer that impedes the downward
movement of water or of moderately
fine to fine texture and a slow
infiltration rate. These soils have
a slow rate of water transmission.
D (High runoff potential) Soils having
very slow infiltration rates when
thoroughly wetted, chiefly clay
soils with a high swelling potential;
soils with a high permanent water
table; soils with a clay pan or
clay layer at or near the surface;
and shallow soils over nearly imper-
vious materials. These soils have a
very slow rate of water transmission.
*
Soils are classed in the next lowest category when a h gh
percentage of stones is present.
42
-------
TASU 9. PWIQFF Q V2 M%$SUS FOR IIYCRCt.QGZ SOZL-COYER C IPt.ZXZ3
(F OR . ‘!C tt)’
Rofiquacs; Soil Coes.rw.tioo S .rvf ci (1975)
Closa-frill.d or roadcut
Z.icIudlnq rf35t-ol.,iys
Ciflcacfos, of aseive astw. or r*nqe:
M dre1 qf Cdltion
Tr ia . it
Und Us. or Cower or Prectici
4ydrOIoqic
Condizboo
N$relcqic
Soil
Gsouo
A
3
C
0
Pillow Str*iqS t row
Aow cros Sb e1qht row
Stniq t row
COntOured
COntoured
Cootoored &
tarrtcad
C toured &
torraced
I 1 3r*ifl S r ,iq it row
Stralgftt row
COntoured
Contoured
Citoured I
arricid
C t ured £
earracad
.
Poor
Poor
GOOd
Poor
Gøa d
Poor
Good
Pe or
lood
Poor
G Ood
Poor
Good
77
72
87
70
55
64
62
68
63
43
41
51
59
36
31
78
79
75
74
11
78
5
74
73
72
70
91
88
33
34
35
30
73
34
33
32
31
79
73
94
91
39
95
35
42
31
33
57
85
34
32
31
Clcse-s.eded 1.ç eo* ftralqsit row
ol ritation iaadGw Strai9nc row
Catcvr ,d
Contoured
Contoured I..
tarr*cad
Contoured I
iarr*c*d
Po o,
Good
Poor
Good
Poor
Good
66
33
54
55
53
51
77
75
75
59
73
57
33
31
33
73
80
76
s
33
43
33
*3
.
50
Pu re or rancid
Contoured
Contoured
Contoured
Poor
Fair
Sood
Poor
Fair
(3aod
6*
.49
39
47
23
5
79
49
61
57
89
35
*6
79
74
31
75
70
39
54
SO
83
33
79
idow (p,.nons,t)
docdlat,ds (fine wood1ot )
Good
Poor
Fair
GOod
30e
45
34
23
55
66
60
SI
71
77
13
70
7*
53
79
77
F.re ,ta.d ,
Roids, lard.surfac,.
g
‘z
74
7j
sz
34
32
v
80
5
n
35
Poor
fair
Good
Viostitiva Condition
4s avily 9r% d no euldi, or havinq p4an cover
on bone thaw aedur 50 circent Of the area.
Moderatuly 9rued; bs ...n aoogc SO and 755
Of th ares with plant cover.
Ltqnt!y 9raa,4; nore u an Oout 751 of thi area
vito plant cover.
43 Jipreduced from ____
[ t available copy .
-------
covers at AMC II. The CN values of other AMC’s
can be found by using a conversion table shown
in Table 10.
(5) Estimate direct runoff——
Direct runoff can be estimated by a graph (Figure
15). Figure 15 is appropriate for use with storms
of 1-day duration or less. The time of day during
which the runoff occurred must be found in detail-
ed rainfall records. When a storm occurs over
several days, the runoff estimate may be made
on a daily basis, changing the antecedent mositure
condition accordingly. The total runoff for a
certain period also can be estimated directly,
using the value of CN applicable for the first
day of the storm, and the total storm rainfall
in the period. In some cases such an estimate
will, differ slightly from the one made on a day-
by—day basis. The direct runoff estimation also
can be calculated by the following equation:
[ w — 0.2( 0 (13)
+ o. ( 1000 — 10)
where R, WP are direct runoff and rainfall respect-
ively (unit in inches).
Cook’s Method——
Cook’s method uses an empirical relationship between drain-
age area and peak flow with modifications for climate, relief,
infiltration, vegetal. cover, and surface storage (Chow, 1964).
The s mmtaticn of applicable values of W (Table 11) reflects the
hydrologic conditions of the watershed. Figure 16 illustrates
the relationship between drainage area and peak flow for various
hydrologic condtions. Values obtained (Figure 16) are modified
for climatic influence (Figure 17). Values for other frequencies
can also be obtained (Table 12). Examples showin the applica-
tion of the Cook’ s Method are presen€ed in the next subsection.
Other Empirical Formulas-—
In the past, engineers developed empirical formulas to
determine design discharges for storm drains. Many such formulas
took the general form:
Q — C A P ( — -—— ) X ( 14
where Q is the peak discharge in cfs, C is a coefficient, depen-
dent on the physiographic and climatic conditions of the water-
shed, A is the drainage area in acres, P is the average rainfall
44
-------
Th$1.Z 10. UM0F c1 VE M 1 1U (ca). NV5R5Z0? zio
ca
Cend
For
Itlon
LI
C I
L
?orM4C
LZ
t ,
Miusi.
(In.)
Cljr.e’
Starts
a ( In.)
100 100 100 0.000 0.00
9 5 94 99 0.304
96 39 99 0.417
94 53 98 0.538 0.13
92 31 97 0.370
71 96 1.11
38 79 99 .3 5 0.37
56 72 94 1.53
34 68 93 1.90 3.33
32 56 92 2.30 Q .44
30 63 91. 2.30 0.30
73 60 90 2.32 056.
75 53 39 3.16 0.63
74 55 83 3.31 0.70
72 93 36 3.59 0.73
70 51 35 4.28 0.36
68 41 84 4.70 0.94
56 45 32 5.15 1.03
64 44 61 5.52 1.12
62 42 73 6.13 1.23
90 40 73 6.67 1.33
53 3$ 76 7.24 1.46
56 35 75 7.36 1.57
54 31 73 3.52 1.70
52 32 71 9.23 1.85
so 31 70 10.0 z.co
43 29 58 10.8 2.15
44 27 56 11.7 2.34
44 25 63 t2.7 z.s
42 24 62 13.3 2.76
40 22 30 15.0 3.00
33 21 53 16.3 3.29
34 19 56 17.3 3.36
34 15 54 19.4 3.38
32 16 52 V.2 4.24
30 15 90 23.3 4.56
25 t2 43 30.0 5.00
20 9 37 40.0 3.00
15 6 30 56.7 11.34
10 4 22 90.0 13.00
S 2 13 190.0 31.00
6 0 0 taflnity tnf in lty
‘ SfI?IfKE SOIl Cons.rvatlon ServIce (7973)
For C II In eli.n 1. £ It the ot.ntIal Inlflt *titn In ncI ss end We It the rain
fall In InClu.
Reproduced from
basi availabk copy.
-------
z
Figure 15.
Estimation of direct run-off amounts from storm
rainfall (Soil Conservation Service, 1972).
from
4 S 5 7 3 9 10 11 12
n;=s (We)
46
-------
TABLE 11. INCREMENTAL W VALUES FOR USE IN COOK’S METHOD (Chow,
1964)
Watershed
Characteristic Extent or Degree w
Relief • Steep rugged terrain with average 40
slopes generally above 30%
• Hilly, with average slopes 10 to 30
30%
• Rolling, with average slopes S to 20
10%
• Relatively flat land, slopes 0 to 10
5%
Infiltration C:) a No affective cover; either rock or 20
thin soil tnantle of negligible in-
filtration capacity
• Slow to take up water; clay or other 15
soil of low infiltration capacity
• Deep loans with infiltration about 10
that of typical prairie soils
• Deep sand or other soil that takes S
up water readily and rapidly
Vegetal cover (C) • No effective plant cover or equiv— 20
alent
• Poor to fair cover; clean cultivated 15
crops or poor natural cover; less
than 10% of the watershed in good
cover
• About 50% of watershed in good cover 10
• About 90% of watershed in good 5
cover, such as grass, woodlands,
or equivalent
Surface storage • Negligible; few surface depressions 20
• Well-defined system of small drain— 15
age
• Considerable depression storage with 10
not more than 2% in lakes, swamps,
or ponds
• Surface—depression storage high; 5
drainage system poorly defined;
large number of lakes, swamps or
ponds.
47
-------
1000
c1
C -,
C -,
z
C
I ’ ,
Fjgure 16.
DRAINAGE AREA, ACRES
Chart for esti nating 5c_year freqUeflCY peak rates of
flow (Chow. 1964 .
48
-------
Figure •t7.
Distribution of rainfall factors used with the wodified Cooks Method (Chow 1964).
-------
TABLE 12. FREQUENCY FACTORS FOR USE WZTH COCK’ S METHOD (CHOW
1964)
(1+ C)
Average
annual precipitation, in.
10
20
30 40
60
80
;
Ratio:
25-year/SO-year
5
0.31
0.38
0.41 0.44
0.48
0.51
10
0.41
0.50
0.53 0.58
0.63
0.66
15
0.50
0.59
0.64 0.68
0.73
0.77
20
0.55
0.65
0.71 0.76
0.82
0.87
25
0.60
0.71
0.78 0.83
0.90
0.92
30
0.64
0.76
0.83 0.89
0.92
0.92
35
0.67
0.81
0.89 0.92
0.92
0.92
40
0.71
0.85
0.92 0.92
0.92
0.92
Ratio:
10-year/50-year
5
0.05
0.08
0.10 0.12
0.15
0.17
10
0.10
0.16
0.21 0.24
0.30
0.34
15
0.16
0.25
0.31 0.37
0.45
0.51
20
0.21
0.33
0.42 0.49
0.60
0.68
25
0.26
0.41
0.52 0.6].
0.75
0.80
30
0.31
0.49
0.62 0.74
0.80
0.80
35
0.36
0.58
0.73 0.80
080
0.80
40
0.42
0.66
0.80 0.80
0.80
0.80
From Table 11.
*
50
-------
intensity in inJ r, S is the slope of the drainage basin in ft
oer 1,000 ft, and X is an exponent. The following equations are
t e examples of the empirical formula (Chow, 1964).
Burkli-Ziegl.er Formula:
Q C A P ( 0.25 (15)
McMath Formula:
Q a c A P C ) (16)
The C value of both formulas varies from 0.2 for previous rural
areas to 0.75 for highly impervious metropolitan areas. Because
C values for specific sites are not available, the use of these
emoirical formutilas is not recommended.
terInin1flg Infiltration
Numerous methods have been proposed for determining infil-
tration capacity and infiltration rate. Theoretical methods
are not recommended because of their highly simplified nature
and the requirement for determining special watershed proper-
ties. Th.ree general approaches are available for the determjn
ation of landfill infiltration: actual measurement, estimation
from surface rui off, and empirical calculation.
thfiltrouteter Measurement——
Infiltrorneters are preferred and are often used for deter-
mining infiltration rates of all watersheds, or experimental
or sample areas within large watersheds. Two general types of
jnfiltrOnteters may be used: (1) rainfall simulators, with the
water applied in the form and at a rate comparable with natural
rainfall, and 2) flooding infiltrometers, with the water ap-
plied in a thin sheet upon the enclosed area.
There are many modifications for rainfall simulators. The
inost common types in use are the F and PA types (Chow, 1964).
They generally utilize plots of 1.8 in (6 ft) wide and 3.6 in
(12 ft) long, respectively. Special nozzles are used to
simulate the rainfall intensity and impact of raindrops. More
detailed information can be found from hydrology literature.
The simplest type of flooding infiltrometer is a metal tube
forced into the soil through a test section of the solid waste
cover. Careful measurement of the amount of water added to
hold a constant reservoir depth produces data for the infiltra-
tion curve. A more advanced type of flooding infiltrometer
includes tvo concentric metal rings. c ithin the outer concert-
tric ring, forms a second reservoir. This outside reservoir
produces a buffer to minimize the lateral spreading of water
below the inner tube. The tubes are usi l.1y 23 cm (9 in.) and
51
-------
35 cm (14 in.) in diameter. Details of the flooding infiltro—
meters also can be found from hydrology literature.
Estimation From Su.rface Runoff-—
Infiltration can be calculated as the difference between
precipitation and surface runoff. Results for surface runoff,
obtained by the methods described in the previous section, can
be used in Equation (6) (in Table 2) to calculate the infiltra-
tion.
ASCE Method--
Mast field infiltration approaches a steady-state minimum
rate after one or two hours of precipitation.. Relative minirrn t
infiltration capacities for three broad soil groups (Table 13)
are the basis of the ASCE em irjcal method (Jens, 1949).
TABLE 13. TYPICAL f 1 VAL ZS FOR BARE SOILS (Jens, 1949).
Soil Class
f 1
(in.Ib.r)
Eiqh (sandy
soil)
0.50
—
1.00
Intermediate
(loam,
clay,
silt)
0.10
—
0.50
.
Low (clay, c
lay loan)
0.10
—
0.10
*
is the infiltration capacity after 1 hr of continuous
rainfall (where 1 in./hr 2.54 cm/br).
“Riqh” Besides sand, the open—structured soils of
other textures, particularly the most friable
silt loans are also included in this group.
“Intermetiata ” In this group, soils are associated with the
barns. These loans contain considerable clay
and much silt but are friable at ordinary
water contents.
“Low” Soils included are not only most clay and clay
barns, but also other soils that are dense in
structure. “Low” soils are commonly deeply
cracked, and the initial infiltration rate may
be high; however, soon after wetting, the
cracks are closed and f 1 decreases rapidly to a
more characteristic low rate.
52
-------
vegetal cover has considerable influence on the infiltra-
tion capacity of bare soil. nflltration capacity can be
increased 3 to 7.5 times by good permanent forest or grass
cover, but little or no increase wit’h poor row crops (Jens,
1949). A vegetation cover factor N (Table 14) is provided for
the infiltration capacity of vegetated soil,
(17)
where f is the infiltration capacity after one hour of contin-
uous ra nf all.
Soil infiltration capacity is also affected by antecedent
precipitation. However, this factor is not considered by this
empirical method.
TABLE 14. VEGETATION COVER FACTOR FOR ESTIMATtNG NPILTR.ATON
CAPACITY (Fenn, 1949)
*
Vegetation
Cover
Cover Factor, N
Permanent forest and
Good
3.0
—
7.5
grass
Medium
Poor
2.0
1.2
-
—
3.0
1.4
se-groWing crops
Good
Medium
Poor
2.5
1.2
1.1
—
—
—
3.0
2.0
1.3
Row crops
Good
Medium
Poor
1.3
1.1
1.0
—
—
1.5
1.3
1.1
* “Good’ Dense vegetal cover of high-quality grass having
extensive root systems, properly managed in grass
for several years.
“Medium” 30 - 30 percent of “Good” vegetal. quality and
density, well managed in grass for at least 2
years.
“poor” Less than 30 percent of “Good” vegetal. density,
low-quality grass and poor management.
53
-------
Snowmelt Infiltration-—
As discussed previously, it is generally assumed that suz—
face runoff from snownelt is rare in municipal landfills. In
general, the water contained in snowfall on landfills either
su.bliinates or melts to become landfill infiltrate. Two methods
for estimating snowmelt are suggested: the degree day equations
and the U.S. Army Corps of Engineers Equations.
The degree—day method provides a means of estimating the
potential sncwmelt on a daily basis. Local or transposed
temperature and snowfall data are necessary. In its usual forn,
this equation is (Soil Conservation Service, 1972):
M 1< (T — 32°) (18)
where M is the potential daily snowxnelt in inches of water
equivalent, K is a constant representing the watershed conditjo
(Table 15), and T is the average daily air temperature in
The difference, T—32°, is the n ttber of degree-days per day.
For example, with an average daily temperature of 47°F, the
ni.unber of degree—days for that day is 15. Because M is the
potential snowmelt for the day, the actual snowmelt cannot be
larger than the water equivalent available for the day. ew
snow may have a water equivalent of from 5 to 25 percent of
snow depth. A value of 10 percent, or 1 in. of water for 10 i
of snow depth 1 is often used for making general estimates.
TABLE 15. V? LtJES OF T CONSTANT K FOR DEGREE-DAY EQUATION
(Soil Conservation Service, 1972)
Watershed Condition
K
Low runoff potential
0.02
Average heavily forested areas;
facing slopes of open country
north—
0.04
-
0.06
Average runoff potential
0.06
South—facing slopes of forested
areas;
0.06
—
0.08
average open country
Eigh runoff potential
0.03
The U.S. Ar fly Corps of Engineers has derived empirical
equations similar to degree—day equation for the daily
*
1 in. 2.54 CTfl.
(°C x 9/5) + 32.
54
-------
springtime snowirtelt estimation (Chow, 1964). Either the mean
daily temperature, T , or the maximum daily teIm erature,
is used, depending c e temperature ranges. Following
equations are suggested:
For open sites:
H a 0.06 (Tmean — 24) (19)
H a 0.04 (Tmax —27) (20)
For forested sites:
H a 0.05 (Tmean -32) (21)
M a 0.04 (Tmax —42) (22)
For open areas (up to 10 percent of forest cover):
H a k’ (0.00508 I ) (1—a) + (1—N) (0.0212 Ta’
+ N(O.029 T 0 ’) + k (0.0084 v) (0.22 Ta’ + 0.78 Td’)
— (23)
H a snowmelt, in./day;
Ta’ difference between air temperature at 1-ft level and
snow surface temperature, °F;
T ‘ a difference between dew—point temperature at 10-ft level
d and snow surface temperature, F;
v a wind speed at 50—ft level, mph;
a observed or estimated solar radiation on horizontal
surface, (langleys);
a observed or estimated average snow surface albedo;
I c’ a basin short-wave radiation factor (between 0.9 and 1.1)
depending on average exposure of open areas to short-
wave radiation in comparison with an unshielded
horizontal surface;
F a estimated average basin forest canopy cover expressed
as a decimal fraction;
T ‘ a difference between cloud base temperature and snow
C surface temperature, F (estimated from upper-air
temperature or by lapse rates from surface station,
preferably on a snow—free site) ;
N a estimated cloud cover expressed as a decimal fract .on
I c a basin convection-condensation melt factor, depending
on relative exposure of area to wind.
55
-------
Determiniig Eva otranSPiratiCfl
Thornthwaite (1955 and 1957) has suggested a term “potentiaj
evanotranspirati0fl to define the evapotaflsPiratiofl that would
occur’ when adequate soil moisture supply is available at all
times. stActual evapotranspiration” is the actual water loss,
controlled by the amount of water actually available for plant
use. There are many methods of estimating actual and potential
evapotranspiration. but no one method can be applied generally
for all purposes. All methods, however, fall into three genera’
categories (1) theoretical approaches (based on physical pro-
cesses), (2) analytical approaches (based on energy or water
balance), and. (3) empirical approaches (based on the site and
climatic conditions). Methods which are applicable for the
landfill ccnditiOfls will be discussed.
Soil-Moisture Sampling--
Soil samples can be taken at different time intervals from
differen depths in the landfill cover for moisture content
analysis. After the moisture content of the soil samples is
analyzed, the rate of change reflecting the water loss from
evapctranspiratiofl can be calculated. Various devices exist to
measure soil moisture without the effort necessary for soil
sam ’olingt These instr flents have the further advantage of
producing i aediate results.
Lysimeter Measurements--
EvapotraflsPiratiOfl is estimated by growing the vegetation
in lysimeterS and then measuring the losses of water necessary
to maintain satisfactory growth. Unfortunately, the lysiineter
may not closely simulate the field conditions, and hence the
results may not be reliable when applied to much Larger areas.
Modification of the lysimeter results for field application is
thus necessarY.
Pan Evaporation Adjusted—-
The ratio of annual free water evaporation to annual pan
evaporation, the “pan coefficient”, averages nearly 0.7 (ranges
from 0.67 to 0.81) (Linsley, et ., 1972). The rate of trans—
piration is approximately identical to the rate of evaporation
from a free water surface provided the water availability for
* Standard procedure for moisture contents analysis: refer to
“Annual BoCk of ASTM Standards”, Part 19-Standard Method of
Laboratory Determination of Moisture Contents of Soil D2216,
p . 338—339, 1980.
Examples of instruments for in situ moisture content measure-
ments are soil moisture tensionteter, soil mesiture block
(based on electrical resistance), and soil moisture cells.
These instruments are available conunercially.
56
-------
1ants is not restricted (Chow, 1964; Linslev, et al., 1972).
stimated free water evaporation may, therefore, i icate the
potential evapot:anspi:ation from a vegetated soil surface. To
obtain the annual distribution of potential evapotranspiration,
the monthly an evaporation is multiplied by the pan coefficient
(usually 0. 7). Obviously, estimations of evapotranspjratjon
based on pan evaporation values are not sensitive to differences
between vegetation types.
£vapotransp iration Equations——
Many exzrojrical or theoretical equations have been derived
for estimating the evapotranspiratjon rates. Some typical
equations suggested by Veihxneyer (1964) are listed in Table 16.
These equations are usually based on the climate (temperature,
humidity, etc.), available heat, and mean vegetation consumntjve
data.
&edke equation——The I edke equation (Equation (24) in Table
16) estimates evapotranspiration from the sum over the growing
season of available heat. Available heat is the number of -
degree-days above the minimum growing temperature (usually
30 to 50 ’). The use of this equation may be limited because
of the lack of data for k and minimum growing temperatures.
Lowry—Johnson equation—-This equation (equation (25) in
Table 16) assumes a l near relationship between effective heat
and eva cration. The effects of types and density of vegetation
are not considered.
Blanev-Morin equation--This equation (equation (26) in -
Table 16) develops an empirical relation between evapotranspjra—
tion and percent day-time hours, mean monthly temperature and
mean monthly relative humidity. The coefficient k (in equation
(26)) varies with the type of vegetation (Table 17). The value
of p can be computed from the TJ.S. Weather Bureau’s “sunshine
tables” (Table 18).
Blaney-Criddle eguation--Thj equation (Equation (29) in
Table 16) is similar to the Blazley—Mcrjn equation, except that
the former does not consider humidity. Tables 17 and 1.3 can
also be used for equation (29) for the evapotranspiratjon cal—
ctilation.
Thornthwaite eauation--The Thornthwaite equation (equation
(27) in Table 16) is based on an exponential relationship
between mean monthly temperature and mean monthly consumption.
This method is widely used by climatologists, geologists, and
agriculture engineers. It has recently been adopted by sanitary
engineers, because of its well developed information and proce-
dures (Remson, et al., 1968; ‘enn, et al.. 1975; Dass, et al.,
1977; and Lutton, et al., 1979). The relationship is b ed
largely on experience in the central and eastern Inited States.
57
-------
1*811 I I. IVAPOIIA1ISPIIIMION 1l IMIi0NS’
Name Dais Period Unit £quai lo a
for( lorE
Uedka Equation . *930 Annual feet £ — iii
Low,-y .Jnjiason Equation . *942 Annual feet £ — 0.000 156 11 s 0.0
S
Ilaney-Ilorin EquatIon .... *942 m Inches £ • i p1( 1 14-k) (26)
months I
Thorntt .waite Equation . 1944 bbnthly Centimeters I • 1.6 3 2 (21)
a • 0.00000ti6Th(lE) - 0.0000flh(l1) $ 0.011911 • 0.49239
Penman Equation 1940 OaIly Millimeters • OP. (28)
A- ’ ,. ?,
where e • 0.35 I• ld)(I • 0.0098w 1 ) o s
U i . I i - t(l-r)(0.I8 4 O.SSS) - S(0. 6 - O.092 d )(O.l0 • 0.905)
Ilaney-Cr iddI. Equation l9 0 . Inches I ktpt • II titers I • p1 (29)
months I I
geference: after V.ibeeyer . l%4. —
•A l . of seturated.vapor-prsssurs cwvs of air at absolute temperatura in F, or dea/dI Ia i 1 19/ 1.
• • a coefficient depending on temperature (x c. text for explanation).
- saturitionvapOr pressure at mean air temperature in ma 119.
ej • iat.ration vapor pressure at mean dew point (I..., actual vapor pressure in the air) lame 1 1g. being equal to multiplied by relative
humidity in per cent.
a • daily evaporation in em.
It • annual mean relative humidity ii . per cent, in Eq. (26).
• accumulated degrma -d 1 ys above minuptun growing temperature for growing season, In Eq.(24)& or accumulated degree-days of aximum daily
temperatur, above 32 1 (or growing season, In Eq. (21) or daily heat budget at surface in ma of water, in Eq. (28).
6 • annual seasonal or monthly consumptIve-us. coefficient.
< - p • per cent of dapt imp hours of the year, occurring during the period, divided lay IOU (table Ia).
r • estimated percentage of reflecting surface.
N - mean monthly sxtrat.rrestrlal radiation in me of water evaporated par day.
° if — Ihornthwait&s teeperat rs-e(Iiclency index, being equal to the sun of I ? monthly values of heat index i — (e/S)L 4, where t is mean
monthly temperature in C.
I • mean monthly temperatur, in ‘I, in Eqs. (?6),(29) or In ‘C in Eq. (21).
I • evapotransplratloa or consumptive use for given period.
mean wind velocity at 2 m abov, the ground in miles/day, or equal to ui(log 6.6/log it), whore w 1 is measured wind velocity in miles/day
at height It in ft.
S • estimated ratio of actual duration of bright sunshine to maximu. possible duration of bright sunshine.
-------
Ecuation (27) gives only unadjusted rates of potential evapora-
tion. Factors as listed in Table 19 can be applied to Equation
(27) tO obtain the adjusted potential evaporation.
Penman ecuation——This equation (Equation (28) in Table 16)
is theoreticallY derived from the absorption of radiant energy
by the ground surface. The values of e and A can be obtained
from Figures 18 and. 19, respectively. ‘?he values of R and
can be obtained from Tables 20 and 21, respectively.
TABLE 17. SEASONAL CONStMPTIIE-rJSZ COEFFICENTS k IN BLANEY-
CRIDDLE AND BLANEY-MORIN EQUATIONS, FOR RRIGATED
CROPS IN WESTERN UNITED STATES (Veibmeyer, 1964)
Crop
Length of growing season or period
k*
Alfalfa
Between frosts
0.80
-
0.85
3eafls
3 mOnths
0.60
-
0.70
corn
4 months
0.75
-
0.85
Cotton
7 months
0.65
-
0.75
Flax
7—8 months
0.80
Grains 1 small
3 months
0.75
-
0.85
Sorghums
Orchard, citrus
4-5 months
7 months
0.70
0.50
-
0.65
cqalnuts
Between frosts
0.70
Deciduous
Between frosts
0.60
-
0.70
Pasture, grass
Between frosts
0.75
tadinc clover
Between frosts
0.80
-
0.85
Potatoes
3½ months
0.65
-
0.75
Rice
3—5 months
1.00
—
1.20
Sugar beets
Tomatoes
6 months
4 months
0.65
0.70
-
0.75
vegetableS
3 months
0.60
sntal 1
* T1ie lower values of k are for coastal areas; the higher
values, for areas with arid climate.
Thornthwa.tte Tables——
Thornthwaite and Mather (1955; and 1957) developed several
tables (see Appendix A) for calculation of evapotranspiration
rates. The method is based on the aSS U tions described pre-
viou ly for the Thornthwaite equation (Equation (27)). Three
steps are involved:
(1) A heat index is computed for each of the 12 months
and summed to produce the annual index (Appendix A,
Tables A-].);
59
-------
TABLE 18. DAYTIME—HOURS PERCENTAGES, OR lOOp, IN BLANEY-CRIDDLE AND BLANEY-
MORIN EQUATIONS (Veihmeyer, 1964) (Annual value of p 1.00)
Latitude,deg J F H A M J J A S 0 N B
North:
60
50
40
35
30
25
20
15
10
0
10
20
30
40
Q
South:
4.67
5.65
8.08
9.65
11.74
12.39
12.31
10.70
8.57
6.98
5.04
4.22
5.98
6.30
8.24
9.24
10.68
10.91
10.99
10.00
8.46
7.45
6.10
5.65
6.52
6.76
6.72
8.33
8.95
10.02
10.08
10.22
9.54
8.39
7.75
6.72
7.05
6.88
8.35
8.83
9.76
9.77
9.93
9.39
8.36
7.87
6.97
6.86
7.30
7.03
8.38
8.72
9.53
9.49
9.67
9.22
8.33
7.99
7.19
7.15
7.53
7.14
8.39
8.61
9.33
9.23
9.45
9.09
8.32
8.09
7.40
7.42
7.74
7.25
8.41
8.52
9.15
9.00
9.25
8.96
8.30
8.18
7.58
7.66
7.94
7.36
8.43
8.44
8.98
8.80
9.05
8.83
8.28
8.26
7.57
7.88
8.13
7.47
8.45
8.37
8.81.
8.60
8.86
8.71
8.25
8.34
7.91
8.10
8.50
7.66
8.49
8.21
8.50
8.22
8.50
8.49
8.21
8.50
8.22
8.50
8.86
7.87
8.53
8.09
8.18
7.86
8.14
8.27
8.17
8.62
8.53
8.88
9.24
8.00
8.57
7.94
7.85
7.43
7.76
8.03
8.13
8.76
8.87
9.33
9.70
8.33
8.62
7.73
7.45
6.96
7.31
7.76
8.07
8.97
9.24
9.86
10.27
8.63
8.67
7.49
6.97
6.37
6.76
7.41
8.02
9.21
9.71
10.49
-------
0
TABLE 19. MEAN POSSIBLE DURATION EXPRESSED IN UNITS OF 30 DAYS OF 12 HR EACH, OR
THE ADJUSTING FACTOR FOR POTENTIAL EVAPOTRANSPIRATION COMPUTED BY THE
THORNTIIWAITE EQUATION (Veihrneyer, 1964).
Latitude,deg
J
F
H
A
H
J
J
A
S
0
N
fl
0
1.04
0.94
.04
1.01
1.04
1.01
1.04
1.04
1.01
104
1.01
1.04
10
1.00
0.91
1.03
1.03
1.08
1.06
1.08
1.07
1.02
1.02
0.98
0.99
20
0.95
0.90
1.03
1.05
1.13
1.11
1.14
1.11
1.02
1.00
0.93
0.94
30
0.90
0.87
1.03
1.08
1.18
1.17
1.20
1.14
1.03
0.98
0.89
0.88
35
0.87
0.85
1.03
1.09
1.21
1.21
1.23
1.16
1.03
0.97
0.86
0.85
40
0.84
0.83
1.03
1.11
1.24
1.25
1.27
1.18
1.04
0.96
0.83
0.81
45
0.80
0.81
1.02
1.13
1.28
1.29
1.31
1.21
1.04
0.94
0.79
0.75
50
0.74
0.78
1.02
1.15
1.33
1.36
1.37
1.25
1.06
0.92
0.76
0.70
-------
°C °F
fJ)t’iO
5O 2
I
2068
10 .50
SATURATED VAPOR PRESSURE, ea, in mm of Jig.
Temperature versus iaturated vupor pressure (Veihrneyer, 1964)
loll
86
10
20
30
Ito
(30 70 80 90
100
Figure 18.
-------
.c •v
‘0
30
120
10
0
0 0.2
1.0 1.2 1.11 1.6 1.8
VALUE OP A — O II j pui •L
Figure 19. Temperature versus A for use in the Penman equation
(Equation (28), Veihmeyer, 1964).
(1
14
(.J
0.6 0.8
-------
a’
TABLE 20.
MIDMONTHLY INTENSITY
MILLIMETERS OF WATER
EQUATION (Veihmeyer,
OF SOLAR RADIATION ON A HORIZONTAL SURFACE, IN
EVAPORI1 D”PER DAY, OR TIlE VALUE 11 IN THE PENMAN
1964)
Latitude,degJ
F
H
A
14
J
J
A
S
0
N
D
North
.
90
——
——
——
7.9
14.9
18.1
16.8
11.2
2.6
——
——
——
80
——
——
1.8
7.8
14.6
17.8
16.5
10.6
4.0
“0.2
——
——
70
——
1.1
4.3
9.1
13.6
17.0
15.8
11.4
6.9
2.4
0.1
—--
60
1.3
3.5
6.8
11.1
14.6
16.5
15.7
12.7
8.5
4.7
1.9
0.9
50
3.6
5.9
9.1
12.7
15.4
16.7
16.1
13.9
10.5
7.1
4.3
3.0
40
6.0
8.3
11.0
13.9
15.9
16.7
16.3
14.8
12.2
9.3
6.7
5.5
30
8.5
10.5
12.7
14.8
16.0
16.5
16.2
15.3
13.5
11.3
9.1
7.9
20
10.8
12.3
13.9
15.2
19.7
15.8
15.7
15.3
14.4
12.9
11.2
10.3
10
12.8
13.9
14.8
15.2
15.0
14.8
14.8
15.0
14.9
14.1
13.1
12.4
0
14.5
15.0
15.2
14.7
13.9
13.4
13.5
14.2
14.9
15.0
14.6
14.3
South
10
15.8
15.7
15.1
13.8
12.4
11.6
11.9
13.0
14.4
15.3
15.7
15.8
20
16.8
16.0
14.6
12.5
10.7
9.6
10.0
11.5
13.5
15.3
16.4
16.9
30
17.3
15.8
13.6
10.8
8.7
7.4
7.11
9.6
12.1
14.8
16.7
17.6
40
17.3
15.2
12.2
8.8
6.4
5.1
3.3
7.5
10.5
13.8
16.5
17.8
50
17.1
14.1
10.5
6.6
4.1
2.8
1.2
5.2
8.5
12.5
16.0
17.8
60
16.6
12.7
8.4
4.3
1.9
0.8
——
2.9
6.2
10.7
15.2
17.5
70
1.6.5
11.2
6.1
1.9
0.1
——
——.
0.8
3.8
8.8
14.5
18.1
80
17.3
10.5
3.6
——
——
——
—
1.3
7.1
15.0
jjj .
90
17.6
10.7
1.9
——
——
——
——
——
——
7.0
15.3
19.3
-------
TABLE 21. VALUES OF B IN TI PENMAN EQUATION (Veibmeyer, 1964)
Ta.
°abs.
3,
water/day
Ta, °F
B,
water/day
270
10.73
35
11.48
275
11.51
40
11.96
280
12.40
45
12.45
285
13.20
50
12.94
290
14.26
55
13.45
295
15.30
60
13.96
300
16.34
65
14.52
305
17.46
70
15.10
310
•
18.60
75
15.65
315
19.85
80
16.25
320
21.15
85
16.85
325
22.50
95
18.10
100
18.80
* 3 c ra 4 , where c is Boltzmann constant, or 2.01 x
mm/day, and Ta is the air temperature. Heat of evaporation
was assi.m ed to be constant at 500 cal/g of water.
55
-------
(2) Daily unadjusted potential evapotranspiration is
obtained from the heat index and the tables
(Appendix A, Tables A—2 and A-3);
(3) Finally, the potential evapotranspiration is
adjusted for cnth and day lengths with correc-
tion factors provided in table (Appendix A,
Table A-4).
As indicated before, this method has been widely used by
sanitary engineers. However, the accuracy of this method for
sanitary Landfills is less studied. Examples of this rrtethod
and the accuracy of the results will be more thoroughly dis-
cussed later in this report.
Determjnjnc Water Storage CaoaCi y
Water storage capacity can be expressed as:
Storage Capacity Field Capacity — Wilting point (30)
In agricultural situations, the root zone is up to the order of
4 ft thick. At a refuse disposal site, the root zone thickness
will be limited by the landfill cover due to the inhospitality
of the refuse to vegetation root systems (Lutton, et al., 1979)
Therefore, Equation (30) is applied only to the landfUl cover.
For a refuse cell, the storage capacity is equal to the field
capacity.
The Water storage capacity (or, the “available water” for
vegetation) of the landfill cover depends on the soil type,
soil comcactiorl, and cover thickness. Figure 20 su tunarizes the
three states of soil water through the range of USDA soil types.
According to this figure, 2—ft silty loam cover will store
about 4 in. of water:
Storage Capacity = 2 ft (3.4 in./ft field capacity — 1.4
in./ft wilting point)
— 4 in.
For a refuse cell, the refuse composition, size, and den-
sity are major factors affecting field capacity (Stone, 1974;
and Fungaroli, 1971). Figure 21 is a plot of refuse field
capacity versus refuse dry density for both processed (milled)
and unprocessed refuse. Refuse field capacity increases as
refuse dry density increases. When refuse grain size decreases,
field capacity also increases significantly. Results indicate
that, for unprc essed refuse, if the dry density of refuse is
around 300 */yd , the field capacity would be roughly 2.5 in.
of water per foot of refuse depth.
66
-------
5
0
—I ,
no
0
a
a ’
3
-J
F
I
SN1 SNV( LIW4 SILT LI(iHT ClAY $ AYY ClAY
sMki Liwi SNØ( LLWI ClAY LON4 CLAY
L( H U M
Figure 20. Water storage capacity of USDA soils (Lutton, Js 1979).
-------
I I I I
t asia Study U.965) -
DT’inq .r Li Study (197
AM . z Study (1962)
RRoVr3 Study (1973)
iin Study (1970) -
.M.rz Study (1954)
A
A
Figure 21.
Refuse field capacity versus refuse dry density,
I
A
C
•Tunqax Li. Study (1
(fin. und :si s.-
•rungar 1i Study (197
(ccan!s. q und e fus
cc
2
g
a
7
)5
z
3
2
1.
0
a
S
I I
c srr
68
-------
The above data, cwever, illustrates only the apparent field
capacity of the refuse, i.e., the arnou.nt of water absorbed beyond
the original refuse moisture content. The absolute refuse field
capacity (or dry weight field caPacity) is higher than the appar-
ent value. Penn, . (1975) suggested that the absolute field
capacity of unprocessed municipal refuse varies from about 2,4
in./ft to a.bout 42 in./ft (or 20 to 35 percent of the moisture
content by volume). Several analyses performed on municipal refuse
found its original moisture Content ranging from 10 to 20 percent
by volume (Penn, et a l. , 1975). Thus, the a.p a.rent fie .d capac-
ity of municipal refuse might range from 0 to 3 in./ft of ref-
use.
Stone (1974) has conducted numerous water absorotion studies
for various refuse components. The initial moisture content and
total moisture holding capacity are reported and sir narized in
Table 22. esu.Lts are presented in units of weight of water per
weight of refuse. Stone (1974) indicated that the field capacity
of any refuse can be estimated with reasonable accuracy if the
distribution of their components is known. Results of Table 22
can be used for this estimation. An example of the field capac-
ity estimation for refuse from the City of Oceanside, California,
is presented in Table 23.
Determining Tine of First A earance of Leachate
If no channeling effect occurred in the refuse, leachate
will first appear when the soil cover and refuse reach field
ca acity. The time required for the first appearance of leachata
ca be obtained by using the moisture—routing calculation (de-
tails of the calculation procedures will be described in the
next sub—section).
The time f first appearance of leachate can also be esti-
mated using graphics prepared by Fungaroli (1971). The graphical
procedures are suzm arized as follows:
1. Moisture routing through the soil cover
a. Determine type o cover soil;
b. Determine the thickness;
c. Csing Figure 22, evaluate the time Ct ) from place-
ment when leachate will pass through .he soil cover
if the soil is placed at the wilting moisture content;
* 2.4 in./ft — 2.4 in./12 in. 20%
4.2 in.fft 4.2 in./12 in. 35%
20% absolute moisture content — 20% original moisture content
— 0% apparent moisture content — 0 jn./ft.
35% absolute moisture content - 10% original moisture content
— 25% apparent moisture content 3 in./ft.
69
-------
TABLE 22. WATER ADSORPTION RANGES FOR SOLID WASTE COMPONENTS (Stone, 1974)
Moisture content, percent dry weight _____
Water absorption capability Total moisture-holding capacity*
Component Maximum Average Minimum Maximum Average Minimum
Newsprintt 290 290
Cardboard (solid 170 17O
and corrugated) 1
Other miscella— 400 100 400
neous paper
Lawn clippings 200 60 370 140
(grass and leaves)
Shrubbery, tree 100 10 250 10
prunings
Food waste 100 0 300 0
(kitchen garbage)
Textiles (cloth 300 100 300
of all types, rope)
Wood, plastic, 0 0
glass metal (all
inorganic)
*
Calculated from water absorption plus initial moisture content in as-received
samples.
Sample variation was negligible.
F Initial moisture contents as—received were less than 6 percent in the laboratory
tests; therefore, they were considered negligible compared to the variation in
moisture absorbed.
-------
TABLE 23. PDThICIE%) RNICE OF N SO4 PFWE CAPACPI’Y OF MWICTPAL R JSE “S IU X ElSJ D AT OCE1 HSIDE,
L1 NDFIU. (Stone, 1974)
Cat onent
¶ft)Lal n isture-balding
capacity as determined
In laboratory tests
Available field absorp—
tion capacity of waste
cc*l onentst — _____
Average : : Field absorr, tive
x position capacityl
(percent)
Max1mui Mirnrnwi
Maximun Miniiiuin
Ma drnuii
Minimun
Newsprint
290
262
7.2
19
Cardboard
170
146
8.3
12
Miscellaneous paper
400 100
397
97
23.6
94 23
Leaves and grass
370 140
312
92
3.8
12 4
Prunings
250 10
207
0
6.3
13 0
Garbage (food waste)
300 0
229
0
9.2
21 0
Textiles **
300 100
284
84
2. 3
7 2
Non-absortonts
Total
0 0
0
0
39.3
100.0
0 0
178 60
*
Oven-dried s ç]es, fran Table 22, percent dry wt basis.
The absorptive capacities determined in laboratory tests reduced by the n asured noisture contents
Iran Oceanside waste saiq]es, percent dry wt basis.
Average of year’s (four quarters) oanposition of collected refuse arriving at Oceanside Municipal
Landfill. Site.
Pounds water per 100 poun:ls of average mixed refuse as received at tI landfill; derived fran product
of available absorptive capacity and average cxnposition for each o onent.
Includes tw od (absorption very sk ,) • fo n plastic (insignificant quantity), and dirt, sand, and ashes
(which entrain hit do ixt absorb).
—1
• -
**
-------
S SL F!L L 511. Ui C. Wi C
flPE CF ‘m’IE MATr!M.
Figure 22.
Tine needed for one inch infiltration per month to
bring soil from wilting to field capacity (Fungaroli,
1971).
72
r
-------
d. Calculate the time (t ) to reach field capacity,
if the placement mois ure is higher than the wilt-
ing moisture:
ti t x
where t calculated time from Figure 22;
field capacity - placement moisture content;
field capacity - wilting moisture content.
2. Moisture routing through refuse:
a. Determine field capacity of the refuse;
b. Determine the original moisture content of the
refuse;
c Calculate apparent field capacity of the refuse
(field capacity—original moisture content);
d. Determine depth of refuse;
e. tsing Figure 23 determine the tine Ct,) from
placement when leachate will appear at the
bottom of the refuse.
3. Total time for leachate appearance:
a. Add times from computations 1 and 2 (i.e.,
+ t ) to get time required for one inch
ikfilthtion per month to bring the soil
and refuse to field capacity;
b. Determine the monthly average infiltration
for a given area;
c. Total time in months (t 1 + t 2 ) x monthly
average infiltration.
Determining Ground-Water Intrustion
Determination of ground-water intrusion into a landfill
(i.e., underf] .ow) can be obtained by a hydrologic investigation
and calculations using Darcy’s Law. Because the rate and direc-
tion of the underfiow through the landfill may be different from
those in surrounding ground water, ground-water monitoring wells
should be drilled on and adjacent to a landfill site. Figure 24
illustrates a typical arrangement of monitoring wells. Wells A
and 3 are off-site wells used for determining local ground water
characteristics. These wells can also be used to determine soil
permeability. Wells 3 and C serve a similar purpose for the
actual landfill. For monitoring studies, measurement of soil and
refuse permeabilities are the most critical factors for underfiow
estimations. Permeability may be obtained by p ping tests,
73
-------
AP W OU FIiU c?I IJY (IH./rT.)
Figure 23. Time needed for percolate to pass through the refuse (Fungaroli, 1971).
1 2 3 ‘I 5 6 7 8
-------
S
Lec end :
Ve = underfiow velocity;
V = local qround-water flow velocity;
II — II
L = on-site hydraulic qradient;
L
S
off—site hydraulic gradient;
A I)rojcction area of the in-flow
boundary on a plane perpendicular
to the in-flow direction.
LAND I
SURFACE WELL D
I
WELL Pt
—I
us
V
V
S IN-FlOW DOUNDARY
H — II
a ( 1
S S
ioure 24. Estimating rate of landfill underfiow (see text for explanation).
-------
injection tests, tracer studies, or by testing of individual
soil samples removed during drilling (Chow, 1964).
After ground-water characteristics (most notably -
ground-water flow pattern and depth to ground water) and soil
permeabilities have been determined, Darcy’s Law can be
applied for identifying velocity and rate of the underf low:
V (1)
and Q = A - (2)
where V and Q are the volocity and rate of underflow, respec-
tively, K is the permeability and dh/dL is the hydraulic
gradient (Figure 24). For estimation of the leachate generatjo
rate by tmderflow, the landfill underfiow projection area (A)
peroendjcular to the ground—water flow direction, must also be
approximated. This area is ca1.culated using the projection
of the in-flow (to the landfill) boundary area onto a plan
perpendicular to the in-flow direction.
EXAMPLES OF LEACKATE GENERATION MODELS
Leachate generation calculations are presented in this su
section. The calculation (Figure 25) involves two major steps:
(1) quantification of leachate generation factors (precipita-
tion, surface runoff, infiltration, etc.), and (2) water balance
calculations.
Quantification of Leachate Generation Factors
Some of the methods (Figure 25, under “Alternative Meth-
ods”) for quantification of leachate generation factors have
been explained in the previous sub—section. In the following
pages only those methods requiring clarification will be Illus-
trated by examples.
Curve N mi .ber Method--
Exammie i-—A small landfill, having silty-clay soil (perme-
ability 10—5 cm/sac) and poor pasture cover had a rainfall o
2.5 in. The 5—day total antecedent rainfall was 0.4 in. Estj —
mate the direct runoff for this storm.
Solution:
1. Using Table 7, find the anteodent moisture condition
a AMC I.
2. Using Table 8, determine the hydrologic soil group
a
3. Using Table 9, identify ON = 89 for AMC = II.
4. Using Table 10, convert CN for AMC a that is ON
a 77
76
-------
SITS C
_I, IMA tIL
I geproduced From
best available copy.
Figure 25. Flow chart of water balance calculations.
77
-------
5. Entering Figure 13 or Equation (13) with W = 2.5 in.
and CN 77, find P. 0.74 in.
When a storm occurs over several days, the runoff estiriate
may be calculated on a daily basis and changes the ANC accord-
ingly. The total runoff can also be estiinted directly by using
the value of CN a licable for the first day of the storm and
the total storm rainfall for the calculation period. Such an
estimate will differ slightly from that made with a day-by-day
calculation.
Cooks Method--
Example 2——A 40—acre landfill located in central Ohio has
an average surface slope of 5 percent. The landfill surface
is covered with a silty-clay soil with no vegetation. The
average annual precipitation in the landfill area is 36 in.
Determine the peak rate of flow with a predicted 10-year fre-
quency.
Solution:
1. Determine the coefficient W. Using Table 11, the
component values of c4 are:
Relief 10
Infiltration
Cover 20
Surface storage 20
W=65
2. Entering Figure 16 with 40 acres at curve W = 65,
read a discharge of 150 cfs for a 50—year frequency
rate.
3. Detez zuine the rainfall factor from Figure 17 for
this location. For central Ohio 1 this is 0.75.
4. Determine the frequency factor from Table 12.
Enter the Table with the s of W for the given
conditions of infiltration and cover soil
(I + C a 15 + 20 a 35) and the annual precip-
itation of 36 in. The 10—year frequency ratio
is 0.77.
5. Compute the 1—year-frequency peak flow 150 cfs
xO.75x0.77—S7cfs.
ASCE Infiltration Method--
Example 3—-A small landfill has the conditions described j
Example 1. Find the total infiltration capacity of a 2.5 in.
storm lasting four hours.
78
-------
Solution:
1. Using TabLe 13, estimate f 1 0.3 in./hr.
2. Using Table 14, estimate N 1.3.
3. Calculate 0.39 jn./hr.
4. Total infiltration capacity 0.39 jn./hr x 4 hr. =
1.56 in.
Degree-day method--
Exarnole 4——A snow storm leaves 20 in. of snow on a munic-
ipal landfill with a low runoff potential. Find the otentia1
sncwmelt infiltration during the ten days after the storm. The
average daily air temperatures in F for these ten consecutive
days are: 30, 28, 34, 34, 36, 37, 40, 42, 44, and 45.
Solution:
1. Using Table 15, find K — 0.02.
2. Converting snowfall to water equivalent 20 in. x
1 in./l0 in. = 2 in.
3. Using Equation (18) find the total potential
snowmelt infiltration = 0.02 t(34—32) + (34—32)
+ (36—32) + (37—32) + (40—32) + (42—32) + (44—32)
+ (45—32)1 1.12 in.
4. Since 1.12 in. is smaller than 2 in. (maxim water
equavalent as calculated in Step 2 above) , so 1.12
in. is the answer.
Pan evapoxat±On adjusted-—
Exainole 5——The monthly pan evaporation rates in a municipal
landfill area are shown in the following table. Find the
potential evapotranspiratiOn from the landfill.
Solution:
Multiply pan evaporation rate by 0.7
Potential
Month Pan ‘Evaporation Evaootranspjratjon
5 anuary 0.2 in. 0.14 in.
February 0.2 in. 0.14 in.
March 0.6 in. 0.42 in.
April 1.5 in. 1.05 in.
May 2.7 in. 1.89 in.
June 4.2 in. 2.94 in.
July 4.8 in. 3.36 in.
August 4.5 in. 3.15 in.
79
-------
Potential
Month Pan Evaooratjon Eva otrans iratjon
September 3.6 in. 2.52 in. —
October 2.1 in. 1.47 in.
November 0. 9 in. 0. 63 in.
December 0.1 in. 0.07 in.
Annual Total 25.4 in. 17.78 in.
Thornthwaite Tables—-
Example 6——A municipal landfill located in Chester County,
Pennsylvania (latitude 40 N) has a mean monthly temperature as
illustrated in Table 24 (row 1). Find the monthly potential
evapotranspiratiort.
Solution:
1. Using AppendIx A, Table A-i, find the monthly heat
index (row 2 of Table 24).
2. Using Appendix A, Table A-2, find the unadjusted
potential evapotranspiration (row 3 of Table 24).
3. Using Appendix A, Table A-4, find the correction
factor for monthly duration of sunlight (row 4
of Table 24).
4. Monthly potential evaporation (row 5) daily un-
adjusted potential evapotranspiration (row 3) x
correction factor (row 4).
Water Balance Calculation
The water balance calculation can be made at several levels
of detail according to the needs of the designer or regulatory
agencies. Two types of calculation can be employed: gross esti-
mation, and moisture routing methods.
Gross Estimation——
The gross estimation method assumes that the moisture con-
tents of the entire landfill (including refuse and cover soils)
are equal or exceed field capacity during the considered period.
Leachate generation can be grossly predicted by using the water
balance equation (Table 2). The period of consideration can be
a year, a month, or some specially selected period (e.g., the
duration of a storm). The following equation calculates the
gross estimation of the annual leachate generation (Ann. L):
Ann. L = Ann. W - Ann. R - Ann. E (31)
where: p
annual leachate generation;
Ann.W — annual precipitation;
Ann.R = annual surface runoff; and
80
-------
TABLE 24. POTENTIAL EVAPOTRANSPIRATION COMPUTED BY TIIORNTHWAITE METHOD
Item Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec.
Mean monthly 0.0 1.0 5.2 11.2 17.1 21.9 244 23.5 19.8 13.7 7.4 1.1
Temperature
(°C)
Monthly heat 0.06 0.09 1.06 3.39 6.44 9.36 11.02 10.41 8.03 4.60 1.81 0.20
index
Daily Un— 0 0.1 0.5 1.4 2.5 3.5 4.0 3.8 3.0 1.8 0.8 0.1
adjusted
potential
evapotrans—
piration (mm)
Correction 25.2 24.9 30.9 33.3 37.2 37.5 38.1 35.4 31.3 28.8 24.9 24.3
factor
(Table A-4)
Monthly 0 2.5 15.5 46.6 93.0 131.3 152.4 134.5 93.6 51.8 19.9 2.4
potential
evapotrans—
piration
(mm)
-------
Ann. E = annual evapctrans iration.
Examole 7-—Assune that the 20-year average annual precipi-
tation (Ann. W ) at chippewa Falls, Wisconsin is 24.5 in. The
rimoff coefficient for this site is estinated as 0.07 and the
annual evaporation is 28 i.n. Find the annual leachate genera-
tion.
Solution:
1. Ann. R 24.5 in. x 0.07 = 1.7 in. (rational method).
2. Ann. E = 28 in. x 0.7 = 19.6 in. (pan evaporation
adjusted method).
3. using EquatiOn (31),
Ann. L 24.5 in. — 1.7 in. — 19.6 in. = 3.2 in.
Moisture routing rnethod
The moisture routing method can be used for landfills which
exceed field capacity and for those which do not. The method
begins with calculations for moisture routing through the soil
cover and uses the results to calculate moisture routing
through the underlying compacted refuse. The following example
shows the procedures for a hypothetical landfill which does’ not
exceed field capacity (Remson, et al., 1968).
ExainDle 8——A hypothetical landfill is composed of 8 ft
(2.4 in) of compacted refuse and 2 ft (0.6 m) of silt loam cover.
The field capacity and wilting point for the cover soil are 34.9
and 10 percent respectively (volume basis). The inita). moisture
content and field capacity for the refuse are 3.9 and 28.6 per-
cent respectively (volume basis). The landfill and cover are
completed on July 1 with an initial soil moisture content of
10 percent (wilting point). Find the leachate movement within
one year.
Solution:
1. Moisture routing through soil cover-—Table 25
shows the routing of moisture through the land-
fill cover for the first year according to the
Thornth aite Method. Computation is made month
by month and results are si . u arized in Table 25
and explained as follows:
a. Find the mean monthly temoerature (T) and
precipitation data (W ) for the study period
(row 1 and 2 of Table 25).
b. Find the potential evapot:anspiration (E)
(row 3, calculation method refer to Example
6, Table 24).
82
-------
YMIIA 25. *WflI*1 UrU*3 9II* U1I A 0Ml M Y IkAWIIl. )IL (UA M
I l PWà1 fl* M4 . iur.
I I ma .*L1y t uI Mt&a. 24.4 23.5 19.8
i
2 N. t.hIy 1iiIt t an. b 10* 142 100
ms ot thr)
3 r I *M1 uv tt - 1S2.4 134.5 93.4
I1I LIQfl. £
4 k utC 3 16 21 16
5 b ti u av IIab1. tt 92 121 06
Lnt1Ituii’ n. I 4’4
I - 0 -40.4 -13.6 -1.4
7 W — E , (I — L I 4’4 —40.4 —13.9 -92.5
I 5011 .I 11IL4M1S i tor a, 41) 40 60
9 a n,s 1 .wre $3 3ui4 0 0 0
10 k u 1 91 121 MS
11 Ii94 0 0 0
Oct.
LW.
I*2 .
.IN I.
LUI.
WJ4.
Nil.
M l J.1
MI1IN.
13.1
1.4
1.1
0.9
1.0
5.2
11.2
17.1 21.9
14
90
17
96
75
102
96
90 103
1132
51.9
19.9
2.4
0
2.6
15.5
46.6
93.0 131.3
143.5
21
14
12
*3
I I
15
13
1$ IS
110
43
76
45
13
64
81
12
76 U
962
11.2
56.1
62.4
73
61.5
71.5
25.4
—17.0 —43.3
218.5
-11.0 -60.1
71.2
127.3
150
150
150
150
150
133 99.1
11.2
54.1
22.1
0
0
0
0
—11 —33.3
51.9
19.9
2.4
0
2.5
15.6
46.6
93.0 I2*.
451
1)
0
39.9
73
41.5
71.5
26.4
0 0
271.1
—
0
V
-------
c. Find Runoff (R): For this example the runoff
is calculated by the Rational Method, assuming
a runoff coefficeint of 0.15 (using Table 3,
under heavy soil, and 2% slope).
d. Find the moistUre available for infiltration (I):
z — R (rowS of Table 25).
e. Find the value of I — E (row 6).
f. Find the accumulated potential water loss (PE):
PE Z negative (I - E) (row 7)
g. Find the soil moisture storage (S ): The initial
value is equal to the initial moisture content
of the cover soil (10% x 0.6 m = 60 mm). For
humid areas (defined as areas where the sum of
all the i — E values is positive, the initial
value is eaual to field capacity of landfills
older than one year (see Example 10). This
initial value of S is assigned to the last
month having a pos!tive value of i - z. In
dry areas (see Example 11), the initial and.
subsequent S values must be determined from
the appropriate table (Table A-8 to A-16 of
Appendix A) using the value of PE calculated
per row 7 above.
Other S values in row 8 of Table 25 can be found
from ap ropriate tables (Table A—S to A—16) if
I E is negative. (If I — .E-is positive, represei t
ing additions of moisture to the soil, it must
be added to the previous months S value.) No
S value can exceed the soil fiel capacity (in
tAis example, 150 mm).
h. Change in soil moisture storage (AS ): represents
the change in soil moisture from znoAth to month
(row 9).
i. Actual evapotranspiration (AZ): when I - E > o,
— E; when I — E < 0, AE — E + (I — E) ‘AS . AZ
soil moisture is depleted, the actual eva trans-
piration rate (AZ) decreases below its potential
rate (E).
j. Percolation in cover soil (PER ): when S < field
capacity, PER 0; when S > ield capa!ity, PER
S
2. Moisture routing through compacted refuse Table 26
shows the routine of moisture through the refuse
84
-------
TA&E 26. P M F(1 E IiWFTNG ‘flIIUJGII UtI ERLY1NG XI4PACt1 D REI U (FI1 F YEI R)
Depth J3elcwi
Er -of-nontb
n isture
content (am)
Lar fil1 Cover
(m)
July
Aug.
Sept.
Oct.
Nov.
Dcc.
Jan.
Feb.
Mar.
Apr.
May
June
o — 0.3
*
12
12
12
12
12
1
51.9
*
86
86
86
86
86
86
0.3 — 0.6
12
12
12
12
12
12
50.
86
86
86
86
86
0.6 — 0.9
12
12
12
12
12
12
12
38.4
86
86
86
86
1.2 — 1.5
12
12
12
12
12
12
12
12
35.9
61.3
61.3
61.3
1.5— 1.8
12
12
12
12
12
12
12
12
12
12
12
12
1.8 — 2.1
12
12
12
12
12
12
12
12
12
12
12
12
2.1 — 2.4
12
12
12
12
12
12
12
12
12
12
12
12
Leachata
Prodictlon
0
0
0
0
0
0
0
0
0
0
0
0
*
0.3 in x 3.9% 12 nm (initial nxisture content per layer).
0.3 in x 28.6% 86 nun (field capacity per Layer).
j location of wetting front (i.e., depth at which the field capacity is exc 1ed).
12 H Un + 39.9 nun (fr an Table 25) = 51.9 ma.
51.9 nun + 73 nun (fran Table 25) — 86 ama + 12 nm = 50.9 nun.
U i
-------
for the first year after placement. The landfill
is divided into eight 0.3-in layers for computational
purposes and has an initial moisture content of
12 n/per layer. The field capacity for each layer
is 86 mm (0.3 m x 28.6% 86 mm). Thus, 74 mm of
moisture are required to bring each layer of refuse
from its initial content to field capacity. Table 25
describes the moisture content in each layer at the
end of the month. The moisture surpluses are routed
from the soil cover (Table 25) into the uppermost
layer. When a layer has reached field capacity, the
excess moisture is routed. into the next layer below.
The results shown in Table 26 indicate the wetting
front is at a depth of approximately 0.9 in one year
after placement.
Exain 1e 9——Determine the time required for the first appear-
ance of leachate for the landfill described in Example 8 by the
moisture routing method.
Solution:
Moisture routing through the soil cover after the first
year of emplacement j calculated in Table 27. Calculation
methods are largely identical to those described in Example 8.
Different results are obtained for rows 7 to 11 because of the
different PE data for July, August, and September.
Table 28 shows the moisture routing through the underlying
refuse for the second year. Two years after the emplacement,
the wetting front is at a depth of 2.1 in.
Table 29 illustrates moisture routing through the underlying
refuse for the third year. Leachate is produced in January
(Table 29), that is. leachate generation occurs’31 months after
refuse placement.
Example 10—-A landfill located in Cincinnati, Ohio has a
0.6m clay-loam cover with 250 mm/in of available water. Runoff
coefficients for this landfill are 0.17 for the wet season and
0.13 for the dry season. Assuming that the field capacity of
the underlying refuse has been reached, find the yearly leachata
production.
Solution:
The calculation procedures for this example are similar to
Example 3. This landfill is located in a hi.rnid area (Table 30,
Z (I — E) >0). The initial value of potential water loss (PE)
is zero because the landfill is at field capacity at the end of
the rainy season. Negative values o.f I — E are then accumulated.
Because the refuse is at field capacity, percolation through the
86
-------
TABLE 27. I)ISPL11E WJ(ffING ¶1H )UQ1 A I )t’IIL 9ML (X)VER AF11 R (tIE YEAR OF U4PLW )4Et F
Row Paranetar
JULY AUG. SEPT. O(.1 . tOy. DF . JAN. FEU. MAR.
APR. MAY
JIJUE
*
PE = E negative (I—E) (refer to Exa*i 1e 8)
— 60.3 nun + (—60.4 nun) = —120.7 nun
— 120.7 nun + (—13.5 nun) = —134.2 lun
1 to 6
7
8
9
—I
10
11
(&nne as that listed in Table 25)
* * *
PE —120.7 —134.2 0
S 9 65.3 60 57
AS 9 —34.4 —5.30 -3.0
P*E 126.4 126.3 88.0
PER 8 0 0 0
0
0
0
0
0
0
0
—1.7
-60.3
68.2
124.3
150
150
150
150
150
133
99.7
11.2
56.1
25.7
0
0
0
0
—17
—33.3
51.8
19.9
2.4
0
2.5
15.5
46.6
93.0
121.3
0
0
36.9
73
61.5
71.5
25.4
0
0
-------
*
TABlE 28. H3ISThRE W)iJFIt 3 ¶1Il D(JGH UtI)ERLYJNG (DIPICfli) REFUSC (SFXX)ff YE1 R)
Depth Below
En1-of-rnonth
nvisture
wntent
(lila)
landfill Cover
(m)
July
Auj.
Sept
Oct.
Nov.
Dec.
Jan.
Feb.
Mar.
Apr.
May
3ur
o — 0.3
86
86
86
86
.
86
06
86
86
86
86
.
86
86
0.3 — 0.6
86
86
86
86
86
86
86
86
86
86
86
86
0.6 — 0.9
86
86
86
86
86
86
86
86
86
86
86
86
0.9— 1.2
61.3
61.3
61.3
61.3
61.3
86
86
86
86
86
86
86
1.2 — 1.5
12
12
12
12
12
24.2
86
86
86
86
86
86
1.5 — 1.8
12
12
12
12
12
12
23.2
84.7
86
86
86
86
1.8 — 2.1
12
12
12
12
12
12
12
12
82.2
86
86
86
2.1 — 2.4
12
12
12
12
12
12
12
12
12
33.6
33.6
33.6
leachate
Production
0
0
0
0
(1
0
0
0
0
0
0
0
*
Refer to
Table 26 for first year noisture rouLing data.
-------
?A1 E 29. O1 I 1 E IOUflNG ¶IWO(X11 tfl 1 LY1N Q*IWL ’LED RI JSE (‘I’IIIHD YEAR)
Depth BelciJ
Lanif ill
Co’ jer
(m)
July Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June
o — 0.3 86 86 86 86 86 86 86 86 86 86 86 86
0.3 — 0.6 86 86 86 86 06 86 86 86 86 86 86 06
0.6 — 0.9 86 86 86 86 86 86 86 86 86 86 86 86
0.9 — 1.2 86 86 86 86 86 86 86 86 86 86 86 86
1.2 — 1.5 86 86 86 86 86 86 86 86 86 86 86 86
w 1.5 — 1.8 86 86 86 86 86 86 86 86 86 86 86 86
‘0
2.1 : 2.4 33.6 33.6 33.6 33.6 33.6 70.5 6 8 :6
leachate ,
Production 0 0 0
0 0 0 57.5 61.5 7LS 25.4 0 0
*
Total leachate produced in tha thh:d year = 57.5 urn + 61.5 inn + 71.5 inn + 25.4 twa
= 215.9 tim
-------
TAI)LE 30. WATER BALA E CAIL’(JLM’IOW FDR A LAM)FILL IN CIlCINHNrI, 01110 (Fenn, et al., 1975)
*
Para ter
J1 N.
FEB.
MAR.
APR.
MAY
JU 1E
JULY
AUG.
SEPT.
(XT.
U)V.
DEC.
ANN-
UAl 4
E
0
2
1.7
50
102
134
155
138
97
51
17
3
766
W 1
80
76
89
82
100
106
97
90
73
65
83
84
1025
Runoff
0.17
0.17
0.17
0.17
0.17
0.13
0.13
0.13
0.13
0jl3
0.13
0.17
Coeff.
to
14
13
15
66
63
75
E +66
+61
+58
8
11
14
154
57
72
70
872
t6
+55
167
1106
R
14 17
14
13
12
9
I
68 83
92
84
78
64
I—
+18 —19
—42
—71
—60
—33
PE
(0) —19
—61
—132
—192
—225
S
150
150
150
150 131.
99
61.
41
33
39
94
150
0
0
0
0 —19
—32
—38
—20
—8
+6
+55
+56
lsE
0
2
17
50 102
124
122
98
72
51
17
3
658
L
+66
+6].
+57
+18 0
0
0
0
0
0
0
+11
213
*
All
L =
units in
leachate
nm (See Table
production
25
for explanation)
-------
soil cove.r will eventually become leachate. Leachate might be
forned in the month of January through April and. December for
the given landfill conditions (Table 30). The annual leachate
production will be 213 nun.
Exam ].e 11——A landfill located in Los Angeles, California
has a 0.6 in silty loam and moderately deep-rooted grass cover.
The runoff coefficient is 0.15 for only those months were
>E and surface runoff is negligible for the dry months. Find
the potential leachate generation.
Solution:
The available water for cover soil (Figure 20) is 200 rum/in.
For landfills in arid areas (Z(I - E)<0), the soil moisture at
the end of the wet season is below field Capacity. Therefore,
it is necessary to find, an initial value of PE for accumulating
negative values of I - E. Thornthwajte’s method of successive
a proxintati0flS is used (Thornthwaite, et a],., 1957). The
values (Table 31) for negative PE are und from Table A-].O of
Appendix A (soil moisture storage 200 mm/rn x 0.6 in 120 mm
at field. capacity).
BSSWDS Model
Due tO tedious procedures involved for the calculation of
leachate generation factors and moisture routing data, attempts
have been made in the past to develop Computer simulation models
for the estimation of leachate generation. Lutton al. (1979)
1979) adopted the USDAHL Model from USDA Rydrograph Laboratory
( ) and applied this model to actual landfill conditions. A
more complete model, which was designed especially for landfill
sites, is the RSSWDS (Rydrologic Simulation on Solid Waste
Disposal Sites) model, developed by Perrier, et al. (1980).
The ffsswDS Model was modified from the SCS cu e umber runoff
method and the hydrologic portion of thS USDA-SEA hydrologic
model (CREAMS) (Knisel, 1980). The model is intended to assist
the owners or operators of solid waste sites, or permit
0 fficia2.s, in arriving at a logical, well-docented decision.
The flowchart for the HSSWDS Model is shown in Figure 26.
From minimal input data, the model will simulate daily, monthly,
and annual values of runoff, infiltration, evapotranspjraejon,
and soil water storage. The model stores r any default values
f parameters estimates to be used when measured and existing
data are not available; for example, soil-water characteristics,
preCiPitati0n mean monthly temperatures, mean monthly solar
radiations and vegetative characteristics. Five years of
climatic records for a large number of stations within the United
States are on tape for easy access to be used in lieu of onsite
mea5ureme t5.
91
-------
TABLE 31. WATER BALANCE CALCULATION FOR A LANDFILL IN LOS ANGELES, CA (Fenn, et al.,
1975)
Pararneter* J F H A 14 J J A S 0 14 I) Annual
*
All units in mm (see Table 25 for explanation)
34
78
0.15
12
66
+32
36
79
0.15
12
67
+31
E
Runoff
Coeff.
R
I
0
I—E
PE
ss
AE
It
840
370
49
66
0.15
10
56
tl
-39
90
+7
49
0
59
27
0
0
27
-32
—71
70
-20
47
0
39
68
0 15
10
58
il 9
76
9
0
0
9
—67
-138
40
-30
39
0
>
94
2
0
0
2
-92
-230
19
-21
23
0
117
0
0
0
0
—117
-347
7
12
12
0
115
1
0
0
1
-114
-461
3
-4
5
0
96
5
0
0
5
-91
-552
-2
7
0
73
14
0
0
14
-59
-661
1
0
14
0
44
334
-506
52
29
0
U
29
-23
-634
1
0
29
0
52
83
+32
+31
34
36
20
+19
39 334
L leachate production
-------
ENTER CUMATOLOGICAI.
DATA; RAIN. TEMP. RAD. L.AI.
F NYDROLOGICAI. OATA
souo w*sr PARAMETERS
COMPUTE OAII.Y TEMPERATURE
RADIATION AND LEAF AREA INDEX
INmAuzE ALL
PARAMETERS
READ ONE YEARS
DAILY PREcIPITATION
I 1
COMPUTE
K ) 10WMEL.T
COMPUTE
365 RUNOFF
DAY
LOOP
COMPUTE
EVAPOTRAN IRATION
AND SOIL WATER MOVEMENT
COMPUTE
DRAINAGE
NO I 3E6?
YES
IS
I.MT YEAR
UNO
YES
CALCULATE OVERALL
STAT1 ST I
OUTPUT
TAlUS
END
?jqu.re 26. Generalized flowchart for the hydrologic sixnulation
Model HSSWD$ ( errier, et al., 1980).
93
-------
ike other leaciiate quantj a j 0 iethods, the SSt1Ds
Model also applied t Water balance principle. The techniques
ormuia used by this model are as follows:
• PrecipitatiC tJsing data from existing’ precipita-
tion records or actual field measurements;
• Surface runoff-—tJsing the SCS curve riuzziber tech-
nique;
• Snow-melt act Lmation— — Jsjng the following’ formula:
M O. 18’r (32)
w z’ M. is the snowtnei.t on day i; and T is the
temperature (CF) above freezing. This relation is
use unless M is greater than the amount of surface
snow;
• Evapotranspiration——r3 5 j the Penman equation (refer
to Table 16); and
• Infiltration and percolation——tjsing’ the soil moisture
routing’ technique for the cover soil. The cover soil
is divided into seven layers for routing. each soil
layer is subject to evaPotranspiratjon. The equation
used for calculation is shown below:
FR 9 SM 1 -Si’1 — ET + M (33)
where DR percolation (or drainage) in soil on
day it
FR infiltration on day 1;
= soil water storage on day i;
ET evapotranspiration on day i; and
H. - amount of snc melt on day i.
(Note that Equation (33) is similar to Equation (7)
as discussed previously). The moisture content
of the solid waste material is assumed at field
capacity; therefore, the volume of water entering
the solid waste by percolation through the cover
soil will immediately generate leachate.
The user of the model can request a final cover soil with a
vegetative and a barrier layer or with a uniform final cover
soil. The user can select an impermeable liner separating
the final cover soil material from the solid waste cells and
select the life expectancy of the liner. No prior experience
with computer programming is required for model usage. The
reader is referred to the original report (Perrier, et al.,
94
-------
1980) for details of the aoplication of the modal.
D:scrJSs:O
Although the water balance method has been widely used for
estimating landfill leachata, the accuracy and sensitivity of
the method for the actual landfill are less studied. More than
one hundred different approaches are available for water balance
calculation (refer to Figure 25); however, no comparisons have
been made in the past to identify which approach can achieve
better results, or which approach is suitable for what types
of landfill. ccnditions. The applicability, accuracy, and sensi-
tivity of the water balance method are discussed in the following
text.
Sensitivity Anal .vsis
In conducting water balance calculations, assumed or sub-
jective estimations of leachate generation factors or conditions
(e.g., runoff coefficient or curve nt.ber for runoff estimation;
f or N factors for ASC infiltration estimation; consumptive—
u e coefficient for evapotranspiration estimation; etc.) are
usually involved. Changes of Certain assumptions or estimations
may have little effect on the final leachate quantities calculat-
ed; however, sometimes these changes may lead to great
differences in the calculated results. Table 32 shows examples of
such effects (Dass, al., 1977). Cases land 2 of Table 32
shows that an increase of the runoff coefficient from 0.065 to
0.200 would reduce the calculated annual leachate data from 3.75
to 2.04 inches, respectively, for Blue Valley Sanitary Landfill.
Cases 1. and 3 indicate that an assumed annual evapotranspiration
of 15 in. (from bare soil in the Blue Valley Sanitary Landfill
area) instead of 20 in. (from grain crops in the same area)
could increase the calculated leachate quantity from 3.75 to
7.95 in. (about 112% difference). The same table also shows that
for Cases 3. and 4, the changes of available moisture of cover
soil. from 1.05 to 4.80 in. (more than four times difference)
only affects the leachate estimation about 20 percent (from 3.75
to 3.01 in.).
Perrier, et al. (1980) also conducted a sensitivity analysis
for the SSWDS Model for a landfill in Cincinnati, Ohjo. Overall
effects of varying Linear life, curve number, winter cover fac-
tor, depth of barrier and vegetative soils, leaf area index,
barrier soil compaction, and soil texture, on leachate generation
and other leachate generation factors, were studied. Suxttmarjza-
tion of this study is given in Table 33.
It should be noted that the sensitivity of the water balance
calculation is Site specific. Even two sites with similar refuse
and soil conditions, radically different climatological
and hydrological data may change the sensitivity results greatly.
95
-------
TABLE 32. LEACHATE PRODUCTION AT BLUE SANITARY LANDFILL UNDER DIFFERENT ASSUMPTIONS,
IN INChES OF WATER* (Dass, et al., 1977)
Case
Runoff
Coefficient
Annual
Potential
Evapotranspiration
Maximum Soil
Moisture Available
For Plant Use
Annual Leachate
1
0.065
20.0
1.05
3.75
2
0.200
20.0
1.05
2.04
3
0.065
15.0
1.05
7.94
4
0.065
20.0
4.80
3.01
* Methods for quantification of leachate generation factors:
Surface runoff—-rational method;
Evapotranspiration—-Thornthwaite Tables; and
Infiltration—-estimation from surface runoff.
(1 in. — 25.4 mm.)
-------
IMIL 33. SIII4ARY Of S(NSIIIVIIV SUIDY NL AUS (Perrier. çt ! . . i JaO)
Change Surfaçç Runoff Eva tran pIr jj % ac te Sofl Perit1j n
SeuslSIrec- Sc i1- Otrec- Suns i irec- Send- Dirac- typo *1
Paraii.ter fr j tIvl Lion Rank ii J! jj Rank j yjjy Lion yJjy Lion
lupermeeble liner S yr md. ’ M NA NA NA NA NA t I I t I Cc*puted
SCS curve .ind.nr II 95 It I tt 4 2 1 3 NA NA NA C n t nt
Winter cover factor 0.5 1.0 1 z 1$ $ i $ 4 3 NA NA NA Seasonal
Depth of barrier soil 6 in. II Ia. $1 • i I 2 1 4 3 NA 11* NA Constant
Depth of vegetative soil I? Ii. 36 Ii. I V ’4 3 1 2 i a I NA NA MA Lonstaut
t.aV area ind * Exceli 1r1 1 4 2 11 I 1 MA NA NA Seg onil
Barrier soil Couspactioo NCP C?’ I * 2 t 4 3 I 4 I MA NA MA COiSsigilt
Soil t.*ture 1
Vegetative layer-S S C t Vt I I Vi 2 6 Vs 3 MA MA NA Constant
Vegetative lsy.r-St t • p t 4 2 * Vt 3 NA 11* MA
0 Vegetative lsyar-L Stc c t i i 3 $ 4 2 NA NA 44*
Vegetitive Isyer-S&C MC? CPO 4 4 I I 4 3 4 4 I MA NA MA
Nut.: Arrow indicates direction .1 chaeges (t increase aid 4 decrease).
lank .aa ft,e change wi parineter ( related to the average annual precipitation (I largest. 3 • lowest £i$E.U4Je).
NA • Not affected and V • Variable (Arrow *nlicat.s g .ncral tendency)
• Slightly
0 Moderately
• SignIficantly
H • Highly
• teer ly
lid. — indefinite life; Irgd — hare gruund; NC? — nut cu.ipacted C? • cu.pact.d Soil leiSure; S • scud. I • lu... C - clay
a
-------
The sensitivity of the water balance calculations also varies
when different calculation approaches are used. The following
s maries are given here only to indicate some of the general
trends of the sensitivity analysis (Perrier, et al., 1980)
• Leachate generatiofl and evapotranspiration are sig-
nifiàantly affected by changes in the soil-water
storage and the available water capacity;
• The winter cover factor is seasonally dependent
and directly affects sensitivity of the evapotrans-
piratiori;
• The SCS curve number primarily affects the surface
runoff and secondarily affects both the evapotrans-
piration and the leachata generation;
• The impermeable liner only affects water that has
percolated past where there is control by evapotrans—
piratiort and surface runoff;
• The surface runoff was the most sensitive parameter
when varying the barrier sail depth;
• The effects of the leaf area index are seasonally
dependent and the parameters most sensitive to
changes in leaf area index were evapotranspiration
and leachate generation;
• The primary parameters affected by the barrier soil
comoaction were leachate generation, and surface
runoff; and
• Changes in soil texture are highly time dependent
and produce conditions where other parameters are
very sensitive.
A clicabiljty and Accura
The applicability and accuracy of the water balance
calculations can be analyzed based on the suitability of the
approaches and the closeness of the calculated results to actua’
measurements, Suitability of the approaches may be judged by
the soundness of the theory used and the completeness, reason-
ableness, and sensitivity of the proposed coefficients,
conditions, and calculation schemes. In general, the field
measurement based techniques (e.g., on—site precipitation.
surface runoff, infiltration, or evapotranspiratiorl measurenent
see Figure 2.) can achieve better accuracies for the leachate
volume estimation. HOwever, these techniques are less applicai
because of the difficulties involved for a large scale field e
measurement. Other non—field measurement based techniques are
98
-------
usually based on a set of asstmlptions or empirical data and,
therefore, their use may be site or conditions s ecific. All
non—field measurement based techniques bear the disadvantage
of unknown accuracy for the landfill conditions. This is be-
cause no statistical verifications have been conducted to
evaluate their use in actual landfill.sites. It is very diff i-
cult to judge which non-field measurement based techniques can
achieve better results for leachate estimation. Table 34 is
prepared only for the qualitative comparisons among various
leachate estimation methods based on the soundness and comulete-
ness of the approach, and easiness and wideness of the
application.
The site verification of various leachate estimation tech-
niques is largely lacking. There are no landfills designed
specificiallY for the purpose of evaluating leachate generation
models. Laboratory or field test cells have been used, however,
by various researchers to evaluate the accuracy of the models.
Wigh (1981) found that in a la ge field test cell (Cell No. 1)
the water balance calculations were reasonably accurate in
predicting the quantity of ].eachate, with only 16 percent
differences after 6.5 years. But, if average evapotranspiration
values had been used, rather than ones computed from the actual
climatic Con itiOfls experienced, the achate predicted would
have been 43 percent different than that actually collected.
In other attempts, Wigh (1979 and 1981) found that after a
orecipitation input of 2050 t to all cells the Leachate collect-
d varied from 213 tO 2347 t Ifl. The low and high values that
occurred were explained by leakage of the test cells. The veri-
fication of the model by these test results is therefore, ixa—
possible.
Other studies, such as Fungaroli, et al. (1979), and Walsh
at al. (1981) also attempted to balance the leachate generation
dInput water in refuse test cells. In general, the water
balance method is applicable and accurate for the test cell
conditi0fl5, because, in such study designs, the djffjcultto—
estimate factors such as surface runoff and evapotranspiration,
are usually not involved.
* ‘ 0 i.owing assumptions were used for calculation:
• Runoff coefficients--0.17 and 0.13 for wet and dry
seasons, respectively;
• Soil moisture storage——125 mm for final cover (2 ft)
and 90 required to bring moisture content of final
cover to field capacity;
• Refuse field capacity——330 mm/rn refuse;
• EvapotraflSPiratiOn COmPUted from the actual climatic
conditions.
99
-------
TABLE 34. COMPARISONS AMONG LEACIfATE ESTIMA’rlOtJ METHODS
0
0
Factor
Mdthods
Advantages
*
Disadvantages
Precipi-
tation
Existing
Records
.
Data are easily
obtainable
e
Data may not be suitable for
specific landfill site.
a
Field
Measurement
•
Data are more accu-
rate than that from
a nearby weather
station.
.
It requires precipitation
cjauçjes and long-term test ing
to gather necessary data.
Surface
Runoff
Surface
Measurement
•
Data are more re-
liable for specific
landfill site
conditions.
•
•
Test plots are necessary for
obtaining the data;
Field testing increases the
labor and time requirements
this method.
for
Rational
•
This method is
•
A number of assumptions cannot
Method
•
•
widely used by
hydrologists and
sanitary engineers.
It is a well eatab—
lished reliable pro’.
cedure for most
engineering cases;
It is very simple to
use;
Required inputs such
as precipitation,
soil types, surface
slope, and vegetal
cover, are available
or can be estimated.
•
•
be readily satisfied under act-
uaI. circumstances;
Runoff coefficients for various
landfill conditions are limited;
Certain important information
such as duration and frequency
of precipitation, antecedent
soil moisture content, and
compaction of landfill cover
material are not considered
in the model.
.
(Continued)
-------
TABLE 34. (Continued)
Methods
Curve
Number
Method
Advantages
• This method is widely used
by agricultural engineers.
it is a well established
reliable procedure for
predicting direct surface
runoff on agricultural
Land;
• it is easy to use but
more complicated than the
rational method;
• Required inputs such as
precipitation, soil types,
land cover, and antecedent
moisture condition are
available or can be
estimated.
Disadvantages
• It is difficult to determine
conditions of the vegetation;
• Since the method is designed
for agricultural purposes,
the curve number obtained
from the table may not be
suitable for landfills;
• For same types of land use
and soil categories, the
curve number way cover the
full range of values. This
means the estimated leachate
volume may be significantly
different depending on the
subjective selection of the
conditions.
• It is easy to use;
• Required inputs such as
slope, soil type, vegetal
cover, surface depression
can be estimated.
• Similar to the rational
method. Used for estimating
the peak flow only;
• Applicability has not been
proved for landfill condi-
Lions
Factor
Surface
Runoff
0
Cook’s
Method
Jnfi].-
Infiltrometer
•
Data
are more reliable
•
Rainfall simulators or
tration
Measurements
than
are
and
other methods which
based on assumptions
calculations only.
•
flooding infiltrometers are
required to gather the data;
Field testing increases the
labor and time requirements.
Estimation
From Surface
Runoff
•
Same
face
as the selected sur—
runoff method,
(Continued)
-------
TABLE 34. (Continued)
• Data are more reliable
than other methods for
evapotranspirat on
estimation.
• It can simulate the evapo-
transpiration for a wide
variety of landfill condi—
tions.
• Requires field sampling and
laboratory testing or field
in situ moisture measurement
equipment.
• Lysimeter may not closely
simulate the field conditions,
therefore, the results may
not be reliable when applied
to landfill areas;
• Requires lysimeter, vegeta-
tion growing time, and labor
to conduct the tests.
0
“.3
Factor
Methods
Advantages
Disadvantages
Intil-
ASCE
•
It is easy to use;
•
Antecedent precipitation is
tration
Method
•
Required inputs can be
estimated.
•
not considered in the method;
Only three groups of soil
type are used which increase
the uncertainty of assumed
infiltration capacity.
Therefore, results may be
very subjective.
Snowmelt
Infiltration
•
•
Both the degree-day or U.S.
Army Corps of Engineers’
equations are easy to
use.
In most situations, only
temperature data are
required.
•
•
No field verification has
been conducted for various
landfill conditions. The
accuracy of this method
is unknown;
In most situation, the
effects of wind, solar
radiation, etc. (refer to
Figure 1) are not considered
in ibis method.
Evapotrans-Soil.
piratlon. Moisture
Measurements
Lysimeter
Measurements
(Continued)
-------
TABLE 34. (Continued)
Factor Methods Advantages Disadvantages
Evapotrans
piration
• It is not sensitive to
differences between land
uses, vegetation types,
and soil moisture conditions;
• Landfill conditions may be
siqnificantly different: from
free water evaporation. No
field verification has been
conducted to identify the
accuracy of this method.
Pan
Evaporation
Adjusted
• It is very easy to use;
• Data are readily avail-
able.
0
Hedke
.
The equation is
very
•
Data of consumptive-use
Equation
simple and easy
to use.
coefficients and minimum
•
growing temperatures are
usually unavailable;
Factors of wind, humidity,
cover material, soil
moisture content, etc. are
not considered.
Lowry—
.
The equation is
very
•
Data of minimum growing
Johnson
simple and easy
to use.
temperatures are usually
Equation
•
unavailable;
1 ffects of types and density
of vegetation, wind, humid-
ity, cover material, soil
moisture content, etc. are
not considered.
Blaney —
•
.TL is easy
to use;
• Effects
of
wind,
and
soil
Morin
•
Required inputs can be
moisture
content
are
not
Equation
estimated.
considered.
(Continued)
-------
TABLE 34. (Concluded)
Thornthwaite
Tables
• Similar to Blaney—Morin
Eq ust ion;
• Required inputs can be
estimated
• Same as Thornthwaite
Equation.
• Effects of humidity, wind,
and soil moisture content
are not considered.
*
The accuracy of non—field measurement based methods for
has not been established. No statistical verifications
such methods under various landfill conditions.
siinulatinq field conditions
have been conducted for
Evapotrans
pirat ion
Factor Methods • Advantages Disadvantages
Blaney-
Cridd le
Equation
I- ’
a
Thornthwaite
•
It is a well established
•
The equation is based large-
Equation
•
•
•
and widely used method;
It is easy to use;
Required inputs can be
estimated;
Both potential and actual
evapotranspirations can
•
•
ly on experience in the
central and eastern United
States;
fffects of wind, and humid—
ity are not considered;
It requires tedious ca cula-
be estimated.
tions.
Penman
•
The equation is based on
•
It requires tedious
Equation
•
•
the most complete theo—
retical approach;
It is widely used in
England, Australia and
eastern United States;
Required inputs can be
estimated.
calculations.
• Same as Thornthwaite
Equation.
-------
The difficulty involved in the verification of the applica-
bility and accuracy of the water balance model for actual
landfills is the lack of accurate leachate generation data.
it is virtually impossible to collect all the leachate generated
from an existing landfill if such a landfill was not designed
for the purpose of obtaining all the leachate generated. SCS
Engineers (1976) have conducted leachate collection studies
for five existing municipal landfill sites, representing five
different geographical areas and site conditions. These sites
are selected for the evaluation of leachate generation models,
because no other data for full scale leachate collection are
available. Descriptions of the sites and actual data on leach—
ate generation 1 precipitation, and evaporatióare’ giveni± .
Appendix 3. The comparison of method accuracy based on these
five sites is informative and not to be used as guidance for
screening or choosing any method discussed previously. Zn this
evaluation, because necessary data were not available, only
twenty-five individual methods were evaluated., as shown in
Table 35. Results are shown in Table 36. Among the 125 individ-
ual cases (25 methods x S Sites) studied, 54 cases (about 43
percent) are underestimated and 71 cases are overestimated for
leachate generation. Results of 21 cases have errors less than
100 percent (see Table 36) • o one method can estimate leachate
generation with errors less than 100 percent for all the sites
studied. In general, Cases 7, 13, 4, 10, 6, 12, 23, and 11
(in the order of overall accuracy) can achieve better leachate
volt e estimation for the selected sites.
It should be noted that the above described comparisons
are based on limited field data. Part of the errors may in
fact, result from field monitoring (SCS Engineers, 1976). It
is suggested that the selection of applicable leachate volume
estimation methods be based on the specific site circumstances,
availability of data, scientific and engineering judgements,
and the experience of the designer and operator in leachate
generation estimation.
105
-------
TABLZ 35. M THOOS uS FC IH€ EVALUATION OF LEAO4ATE NE ATt0N.
CAll
, m —
3. null —__ —
3. pull ___ *
• null —
I. fill
I. 1 )1 1 1 ___ r8
7. • null _ t
4. pull —
,. 1 1, 1 1 ——
Is. ,uuls
U. full -
3. Uull__ %
1$. P P .44
I.. flu id . .Ji
1$. ll4d l
,p•l ___ j i
it. •l.Sl i
1$. lull . ......U *
34. ,lul d
fluid IIIW i
3%. Pull. —l
a.
a. fluid t
. fill
a. P1 .44 . ... $l
1.tIll•I —
ima nails lust. 33
1.II I —
t. ius ill. 3)
1.$i i -
ima us ). 31
1.II I
IAIIIS 1154. 3)
1.ti .i —
ima r..i.s 1l.4• 3)
1.V l —
3p ,.s T lu lull. 13
ul t —
i ’r’ ? .44 .i tail, I)
ts...i %Ml
i , ’1 7s 11l1 ?A I . I I
1.ll i
(p pl Thu ? 4• 1)
—
p y . l 7 . 14. 5 TIllu 4)
1.sl l
3 14ll Thill Ihiul)
1.t1 l at
(14 flaIls fl 1 4• 4)
1.$ l —
(3a4 ll CAllus 1154 1 4)
(klull Ta. •s tu.s. Ii
— ‘ — 4
(3j4U 1.54 11 ?U 14 4)
c_u. -w
c_us -w
thu. — - 4
id -u _.-4
c_ I. - w
Iitlu.tlu. P c_Ph
—P
Plu. ,fhl
Istuslc_ p-u c_Pill
? 1 u .tV
• S isl
inlu.s$S. • - u ill
‘-u p,
(3*l1iI. Pu. lPs4 i la
. - uI9
CA1uI.t*u. . c_ u,l
‘ —ft
tjsl*I P-u c_Put
3 .r Pl It
ff
CAilI.14u. ‘ c_ ’w.
‘—4,
c_ P c.
-P
3 -4$I p, 4 fil
-P
I..
-P
Iu 5u.1 •
‘-‘ft
tiI4u.$-u p-u c_Pmi
p a ..’,
st_Is.. fP —
-P
6*tIt_ • SwIll.
—p
tsit ltu. I-u c_Phi
I_ft
c_fill
5-4,
Zutt_Il P c_Pica
I _ f l
ha 44
ma
ma
ma
Ala
l,s . .b, usu lrnI. .
II. i-uutI . .
141 1 11 . 1
V . 1 1,—Ph Il
I - us t . .
l%..p.C, l 1 4he
1 - uil . .
1.uut - u Iu
it_uI_ Ill
115 11
Pu -u-ul1I
hi——.
il.,m .llrl.
c - ut,—
3 - u i , —
?iuu.e.s,,.
nt_ lu-sit.
l ull
Ii. ...-uulI . .
iI
(—it..
l)lsPhCP’All t
I - ut ,..
I - uS Ia
vt_ut—It.
till.
Ii . lit-IMu-
14-
i t-P.
i - u u1a
i3l. lP.C ’ . 1 lS
c-u’s—
TI_I_Ill
It_ I ,—
It_It- f l.
Tell.
II I I.S Itt
Mi_Ill
i 3 t _ ,— - u• I I
i _I ’ll
n—.,—c,t_i.
I—Il—
Tt_..illhIW (-uls..
tt_ it. Till.
— ., s sis.c ’ gs
106
I Reproduced from
[ e t available copy .
-------
l*6L8 36. CORPARISOIIS Of IIACIIAII 61N11A1 100 U1l&t1N MIMU8ID 880 CALCULAUD ICIULIS.
511.8
C iii
N ..
Sit. £
I. 14I(fer. .c.
c*lswlttcd
— sit. a
I. U*fftr.nCS
ikit1.UJ
sit. ç SI!Q_!____
—
I. Z4It(str*. . I. %JIfterelbc. L ZJ1 fe.’euIci I
aLc ,tditd_
I
4.39 392.1 1
22. 12 1655.SS
11.15 63.68 1.01 25*8.62 188 964 16 I I I )
6.86 444.44 1.11 2925.93 8.69 804.05 *248
2
9.16 864.2%
16.39 1200.19
184. 13
0 -100.0 0 -100.0 2.4* 229.68 2112
3
2.86 201.0$
11.14
480.95
0 -*00.0 I I -44.44 0 -100.0 161
4
0 -100.0
1.32
1316.19
1.18 14.24 4.98 1144.44 1.13 863.5% 992
5
10.01 960.0
*8.8
519.84
4.02 2*9.05 1.2$ 341.19 1.8 143.21 213
6
0.63 -fl.U
1.8*
*31 .11
0 -)00.0 0 -*110.0 0.6 -18.92 83
P
1.56 64.21
2.92
-*00.0
0 -100.0 0 -100.0 0 -100.0 *00
S
0 -100.0
0
0 -*00.0 0 -*00.0 0 -100.0 IO U
9
0 -100.0
0
4*2.10
0.88 -81.00 0 -*00.0 1.11 135.14 17$
10
2.43 155.7$
6.46
1.56 -33.04 2.94 906.09 1.13 92.10 4*6
II
2.2 131.6$
12.29
*84 -12.39 1.64 501.4) 0 -100.0 293
U
3.11 298.84
1.4
0 -100.0 0 -100.0 0 -*00.0 91
*3
0.64 • -32.63
3.15 150.0
-*00.0
0 ‘-*00.0 0 -*00.0 0 -*00.0 100
II
0 -$00.0
0
121.18
*42 -19*5 0.11 181.19 1.0$ 45.9 5 21 16
IS
4.88 392.63
10.43
9.2 35.10 8.41 3031.04 4.08 451.35 1150
16
4.8 401.26
24.92
4.61 -32.31 9.48 3403.10 2.09 290.54 1195
Il
9.2 $68.42
18.61 $381.15
0 -100.0 0 -*00.0 0 -100.0 M i S
I I
3.41 265.26
13.43
153.91
0 -*00.0 0.6 *22.22 0 -100.0 241
*9
0 -$00.8
10.16
6.11 -9.40 5.29 $859.26 4.01 450.0 9110
20
10. 11 984.2*
21.81
941.62
6.9 1.32 *4.82 5308.89 0 —*00.0 *543
2*
13.11 1280.0
*3.2
1.42 -19.19 14.21 5*62.98 0 —*00.0 1185
22
15.91 1914.14
1. 6 4
0 —100.0 4.89 1311.11 0 -100.0 406
23
1.82 91.6$
1.6
-100.0
0 -100.0 3.46 1181.4$ 0 -100.0 316
24
0 -100.0
0
3.16 -44.93 13.44 4811.30 0 -100.0 1328
25
9.8 931.5$
9.92 601.30
‘2 dSU.r . s
.usi iil • lflO2 Keiiur*d Leacho1 . Sit. & • 0.95 Sit. I • *26. SIt. C - 8.8*. SIt. 8 • 0.21. Sit. C - 0.14. ii It I i
I nch.
I- . .
0
—3
• 11.11 Ii lick.
-------
SECTION 4
LEACEATE COMPOS IT ION
INTRODUCTION
The composition of municipal landfill leachate has been the
subject of considerable research during the past two decades.
Landfill leachates are, •in many cases, highly contaminated and
can degrade surface and grourid—w ater resources. In recent years,
research efforts have shifted from merely characterizing leachate
components toward identifying the solid waste and landfill fac—
tors influencing the composition of leachate, and developing
models which describe the concentration histories of leachate.
Though the behavior of landfills and leachate generation is
generally understood, present leachate composition models are
useful only in the interpretation of experimental results, rather
than finding application to field-scale problems.
In this section, the chemical and microbiological character
of municipal landfill leachate will be reviewed, and the influ-
ence of various factors which interact to control leachate
quality will be described. In addition, the effects on leachates
of adding municipal and industrial sludges to municipal landfill 3
will be identified. Finally, leachate composition modeling
efforts will be reviewed.
C MICAL COMPOSITION
Several researchers have investigated the quality of land-
fill leachata under controlled laboratory conditions and in the
field. Merz (1954) found that ion concentrations in leachates
from percolation bins reached a maximum of 10 to 20 times the
concentration found in the applied water; ammonia nitrogen and
phosphorus were present in extremely high concentrations rel-
ative to the applied water. Qasizn and Burchinal (1970)
indicated from their lysimeter tests that leachate was high in
dissolved solids and found BOD concentrations 40 to 85 times
higher in leachate than in most raw domestic sewage sludge.
Fungaroli (1971) reported that the mean p of leachate was 5.5.
Ranges of other constituents included: zinc, 0 to 135 mg/i;
chloride, 50 to 2,400 mg/I; sodium, 100 to 4,000 mg/i; hardness,
300 to 6,000 mg/i; COD, 1,000 to over 5,000 mg/i. ughes,et al.,,
108
-------
(1971) studied four landfills in northeastern Illinois and
concluded that leachate quality was highly variable. Leachate
contained larger ollutant loads than raw sewage or industrial
waste; concentrations of heavy metals were higher than the
applicable drinking water standards.
Data from these and other studies addressing the chemical
composition of municipal landfill leachates were combined to
provide the ranges of constituent concentrations illustrated in
Figure 27. Sources of the leachate data are given by numbers
indicated within each range. The location of each number in
the range indicates the concentration reported by each study,
or, if a range of concentrations was reported, the values
indicate median values (except where noted otherwise). A legend
is presented at the end of Figure 27. Appendix C summarizes
the leachate composition data used to develop Figure 27
Tote in Figure 27 that data represent leachates collected
from actual landfill sites, field test cells, and laboratory
column tests. The sampling of leachates from actual landfill
sites was in most cases achieved through monitoring wells
placed in and beneath the fill. A few samples represent ground
water collected downstream from the landfill sites, or leachate
springs which could be sampled at the surface. Leachate sam—
pies collected above ground (data points 30,31,32,33,34) may
contain eroded soil and may have reacted with the soil to
significantly affect its quality. Many leachate studies have
involved field test cells and laboratory columns which simulate
actual landfills. A study by Wigh (1979) showed a lack of
statistical similarity between leachates of all-sca1e cells
and leachates of a field—scale cell, though the comparison was
prejudiced by possible leakage caused by faulty cell construc-
tion. In study .ng FLgU.re 27, no specific trends are evident
which link higher or lower leachate concentrations to specific
testing methods.
Figure 27 shows large variations in leachate constituent
concentrations. The variations are attributable to a myriad of
jnteractiflg factors, including refuse age, the rate of water
application, refuse moisture, landfill design and operations,
and the interaction of leachate with its envirox,ment. The
variations can also be attributed to sampling procedures, sample
preservations handling and storage, and analytical methods used
to characterize the leachates. For example, the effective pore
diameter of the material through which leachate is collected
can affect the concentration of parameters such as suspended
solids, phosphates and heavy metals since precipitation or
coatings may be formed on the material which may filter out
these materials (Chian and DeWalle, 1975). Chian and DeWalle
(1977) found rapid changes in leachate redox potential (Eh),
turbidity, suspended solids and color ixtediately after sampling.
Cook (1966) showed that leachate stored in a quart jar capped
109
-------
4
10
10
1
1
i a 2
*
denotes peak values
denotes arithmetic mean concentration
*
units in mg/i unless indicated otherwise
denotes maximum reported value (m g i 1 )
Figure 27. Leachate constittlent concent.ratiofl ranges.
COD
Legend:
/
7E0,OCO
-
Landfill
Field Test Cell
—x- Laboratory Column Study
BOD 5
/
TO C
I
57,000
\
27,700
12 .-—
— 12
06 *
10
1
1
102
-2.4—
—25 —
10
10
110
-------
Is
TDS
59,200
/
44,900
12 —
TVS
/32 ,2W \
10
13
( -16
153
1
I
102
1 .243
I
10
1 o2
10
Figure 27 (Con.tinued)
111
-------
—23—
17_
— 24
Figi.ire 27 (Continued.)
112
NH 4 4 !
/
1,106
12=—
g36
NO 3
1 3
/
1
1
1
10
1
10_i
1 o-2
27.2.
_12__€!: -
17
—24€
10
1
*
€11
12
-------
r
98
l
—Li —
16
154
i i
23: .
5,500
1
TOTQ. P
1
FO, 4 tNCR ,
1
to 4
3
-
102
10
•i.
to- i
102
7—
10
4
1.
Figure 27 (Contii ued)
1]. 3
-------
Figure 27 (Continued)
10
1 3
10
1
114
-------
i a?
1
I
1
1.
10
oZ
1
10
Figure 27 (C nti tied )
U. 5
-------
i0 4
1
I O-
10
I
10 1
1 O Z
Figure 27 (C nti e )
116
-------
10 _ I
1
•1
Cu
102
9.9
10
Figure 27 (Continued)
117
-------
Ni
PB
13.0
CD
/
0.375
—12
10
1
10 1
1
3 T
— —
10
1
io a
i 3
Figl.lrQ 27
(Continued.)
U. 8
-------
100
CR
/—\
18
— fl-.
10
10
I
•1
10
10 -
10
1
0.16
4 .
101
—3
—2
10
—3
10
Figure 27 (C r tin ed)
119
-------
4
PH
*
4.
5.000
5,000
t j .z pE rnits
SP. CCND. ’
10,000 15,000
0,000 15,000
j xnhcs/tn1
?igi re 27 (Continued)
2 3 4 5 6 7 8 9 10 11
1 .2 3 4 5 6 7 8 9 10
I I I I I I I I
20,000
960
12— 16,800
t I I .. I I I
20,000
120
-------
Where:
Reference
1—3 indicates Hughes, et al. (1971)
4—6 I ’ Pohiand (1975)
7 ‘I Ministry of Local Housing and
Goverx ment (1961)
8 Merz (1954)
9,10 ENCON A.5sociates (1974)
11,12 Chian and DeWalle (1977)
13 Fungaroli (1971)
14 Qasixn and Burchinal (1970)
15 Zenene (1975.)
16 Chun (1975)
17 Re inh art and. Ham (1973)
18,19 Johansen and Carison (1976)
20—22 Wigh (1979)
23,24 Fungaroli and Steiner (1979)
25 Fuller (1978)
26—28 “ U.S. Ar ty Engineer W.E.S. (1978)
29 1 Rovers and Farquhar (1973)
30—34 SCS Engineers (1976)
35..37 E rich and Landon (1969)
38—39 1 Apgar and Langmuir (1971)
40 I’ A.nderson and Dornbush (1967)
41 I’ Riccio and Hyde (1971)
42,43 a Meichtry (1971)
44,45 County of Los Angeles (1969)
46 Zenone (1974)
47—51 N SCS Engineers (19Th)
Figure 27 (Conc1 .ided)
121
-------
with aluminum foil and maintained at room temperature caused a
55 percent COD reduction after three weeks with most of the
decrease occurring after one week. Most studies, however, fail
to mention leachate sampling, preservation, handling, storage,
and analytical orocedures, making interpretation and cornParisort
with other studies djffjcult.
Aside from the leachate constituents presented in Figure 27
several classes of organic compounds have been identified in
municipal landfill leachates. Those compounds identified by
Burrowsr et al. (1975), Khare and Dondero (1977) , Robertson,
et al., (t 7 T , and Engers (1977) are summarized in Table 37.
Ho chain fatty acids, indicative of the primary stages of
anaerobic degradation 1 are common to all investigations. Re-
search has revealed that organic constituents in landfill
leachate are largely a factor of landfill age (Chian and De ’1a11e
1977). Further discussion of organic constituents is presented
in a later s sectiofl entitled “Factors Affecting Leachate
Composition,” (Landfill Age)
TABLE 37. O ANIC CO OUNDS OR CLASSES 2DENTIFIED IN LAN FtLL
LEACHAT S
AUTHOR COMPOUND
Burrows & Rowe (1975) Acetone
Short chain alcoho l*
Short chain acids
share & Dondero (1977) Alkanes
Ketones (Acetone,
2-butanone)
C! C1 3 I Cd 4
Aromatic solvents
(benzene, toluene,
xylene)
Short chain a1coho1
Short chain amines
Short chain acids
aebertson, Toussaint, & Jorgne Phthalate esters
(1974) Aromtaic solvents
(cresol, xylene)
Toluo ethyl-toluene
sulfonamide
Alcohols
Me thylpyri dine
Ethers
Short chain acids
Engers (1977) Short chain acids
___ (C 1 — C 7 )
122
-------
MICRCBIOL0GIC COMPOSITION
Municipal solid waste contains a large microbial population,
and may be heavily contaminated with pathogenic raicroorganisms
(Gaby, 1975). According to Caby, refuse is an excellent medi t
for supporting the survival of pathogens. Municipal solid waste
landfills often contain pet feces, animal remains, disposable
baby diapers, hospital wastes, and sometimes sewage sludges,
all of which pose a potentially significant health hazard.
,i.though the chemical properties of municipal landfill leachates
have been fairly well-characterized, the microbiological com-
position and viability of contained microbes has been less
studied. Microbiological investigations have centered on the
detection of fecalthdiCator bacteria which, if found, would
suggest the presence of microbial pathogens in municipal leach—
ates. Various bacteria, and fecal-indicator groups, viruses,
fungi, and. parasites identified in municipal landfill leachates
are described below.
Bacter
Several studies have shown that there can be a significant
bacterial copulation associated with municipal landfill leach—
ates. These studies have also indicated that this population
varies with landfill or refuse age: population densities are
high in leachates of fresh refuse but decrease with age or time
of leaching (Qasizu and Burchinal, 1970; Blannon and Peterson,
1974; Engelbrecht and zairhor, 1975; Glotzbecker and Novello,
1975). Specific bacteria and. fecal-indicator bacteria identi-
fied in municipal landfill leachates are listed in Table 38,
along with reported numbers. Note the high upper range for
the three fecal-indica.tOr bacterial groups. The presence of the
fecal streptococcal group indicates fecal contamination from
humans or chickens (Skinner and Sguesnel,19 7 8). Specific
fecal streptococcal species identified in municipal landfill
leaChate are Streptococcus faecalis biotype iii and S. e uinus
(Table 38). The presence of the fecal coliform group indicates
fecal contamination by warm blooded animals. This group is
readily identified by its ability to ferment lactose, producing
gas at 44.5°C. The third. group, total coliform, is the tra-
ditional indicator of possible pathogenic constituents in water
supplies. A water or leachate analysis resulting in a positive
indication of coliform bacteria implies a potential danger,
though the sample can contain coliform bacteria and still not
contain any pathogenic organisms.
Table 38 also shows three bacterial species raarkec by
double asterisk. These groups are judged by the U.S. Public
Health Service (1976) as Class 2, moderately infectious agents
posing a potential health hazard. According tO the Public
Health Service, those working with Class 2 agents need to take
123
-------
rdicator ro s
SC*L s sp .ococ . Lab, rc. 1.2.4,5.6.
la.MfiU . 7,3
Lab, C. 1.2,3.4.5.6,
Lanafil.L 8,9
i.a.b, FC.
Or7an.i
.OCCCC 2S $P.
F c
S. ? aecal±s ‘iax.
— L . ac .sns
aeypLcaJ. . ra.c aj.is
v iz.
S. fas aIis i0t7pe
L fascaj. .s biotype z::
L durez s
I. r c
LandfI ll.
Si lon.Ua
p ni i rian Lab,
Land fi lls
Ac ro obecter *9. Landfi LJ.
SaciJLt s sp. Landfi ll.
C1oserid1’ n fa.LL Landf I ll
Clostr . i .e3 sp. landfill,
Cor nebecterit 59. L&ndfiL l .
r c
________ r c
___ F C
________ __________ Landf Ill,
______ rrc.
____________ Landf ii. 1
_______________ L andfill
l andf ill
________ La.ndfili.
__________________ Landf i ll
Lab • Labozator3 or Lysi .t.r st; F ’C • fi.ld t.st ceLl
sin a.Laetng an act ai Landfill: Landf 11.1. • actuaL landfIll
1. 3lennon and Peterson (1974) 2. Glotzbscker and love1lo (1975)
3. !ng.l rsc’ t (1973) 4. fnqslbrec t (1974)
5. Cocper (1974a) 6. QLIjn and 8urct ina.L (1973)
7. Enqe1br*c t and ?.airhcr (1979)8. Scarpino and Oortn.lly (1979)
9. Sobsey (1978) 1.0. Cook (1967)
c3 — colony fo ing units OI Probable b.r
Aqents of crdinar’ tcts!te±a l ta:ard such as seaD )LLococci. that can auae
fissais ihsn the agent penetrates the skin (PublIc 3.alth Servics, 1975)
124 Reproduced from JP A
L sI av&Iable copy .
7A3L 8. 3ACP.I ?. A D 3Ac !P : ICxcRs E rrIF:z3
: v zc: . z.z c : s
rgaei or f.cai.-
indicator o cuo
ange of ranort.d
va l e *
‘jp. of
Stdv aaferertces
0—48 nilli.cn organ—
t /L00 n i.
<0.1—3 ni.i.licn /nl.
‘2—490.300 MPN/l00
0—1.3 ni.l lion organ—
i /l0o nL.
‘0.i.—100,000 c3V/&..
2—94.000 M1 /10Q n1 ..
(0.1—100.000 C3’ ’JIni.
‘2—1.40.000 !1/l30 L.
0—450.000 C ’ fnl
8, 1 .0
1
1
I
I
1
2 • 10
3.7
3,13
8.10
8
3.10
a
3
8
a
8
8
3.10
a
8
10
(.0
10
10
10
Aertoonas ip.
t.istsr .a ncnovtocenes
Micrococtus S
oraxsLla *9.
Serrat .a marcescerts
? .cin.tobact.r sp.
ntsrooacter cloacal
S tOaAviOcOctus aLbua
Pseudononas so.
? lavobact.rzun to.
? roojgnibact.rI *9.
-------
precautions to prevent from being innoculated by contaminated
material containing these pathogens.
A comprehensive review of studies on the survival of
bacteria in leachates was conducted by Ware (1980). In reviewing
her work, a number of trends are evident. First, increases
bacterial mortality with time of leaching or refuse age result
from the bacteriacidal effects of the leachate and landfill.
Bacterial survival is inversely proportional to landfill temper-
ature (Glotzbec1cer and Novello, 1975; Engelbrecht, 1973;
£ngelbrecht and ? mirhor, 1975). Relatively high temperatures
(60°C) achieved Ln the aerobic stage of refuse biodegradation
(FarqUhar and Rovers, 1972) can thus inhibit bacterial growth
and survival. Bacterial inactivation is more rapid at lower pH
(Engelbrecht , 1973; Engelbrecht and mirhor, 1975). Together,
temperature and pH can act synergistically to accelerate
bacterial inactivation (Engelbrecht and njrhor, 1975). Attempts
to correlate leachate chemical and biological composition impli-
cated iron and zinc cations and short chained fatty acids in
j creasing bacterial inactivation, though not all bacterial
speCiSS were similarly affected (Engelbrecht and Ainirhor, 1975).
Riley, et al. (1977) found that aliphatic acids present in raw
leaChates erted a bacterio-static action on coliforms.
Recent research on the health hazard associated with rnunic-
jDal landfill leachates by Donnelly and Scarpino (1981) has
demonstrated that leachates can contain high concentrations of
bacteria which are present in human feces, several strains being
, hogefliC for humans. Though fecal-indicator bacterial con-
entrations were usually low (below detection level) for
landfills inactive for 2 years or greater, fecal-jndjcatcr
bacteria were isolated in solid Waste from a 9 year old landfill.
p thogens found in these old landfills were Listeria monocvto—
enes, Acinetobacter Sr., Moraxella np., and Allescherja
bOVd , all of which are moderately infectious (U.S. Public
jj 1th Service, 1976). Donnelly and Scarpino also studied
bacterial population changes in leachates from small-scale
experimental lysinteters. Low coliform levels were detected
in ].eachates after 4 months. Streptococcal levels remained
jgher, but eventually dropped below detection level after
1½ years. Laboratory lysiirteters containing sewage sludges
generated leachates which contained fecal—indicator bacteria
after 2 years.
‘ iyiruS !
Because municipal solid waste may contain feca]. material
from a number of different sources, it is possible that enteric
viruses are among the pathogens entering leachate. As opposed to
bacteria, viruses are obligate parasites and cannot multiply
‘a pathogenic fungus
125
-------
outside a host organism ( 1jtchell, 1974). They may be present
in the feces of humans and animals which are subsequently die-
flosed in municipal landfills. Enteric viruses have been detected
in fecally contaminated articles in municipal solid waste (Sobsey,
1978)
In general, enteric viruses are rarely found in municipal
landfill leachates. Engelbrecht, et al. (1974) detected no
viruses in leachates produced by a large, field scale municipal
solid waste lysimeter that had been experimentally contaminated
with poliovirus type 1 during the filling peration. Peterson
(1971) etected poliovirus type 3 (150 PFtJ7100 ml) in one of 13
sam les of leachate collected from a lysimeter containing raw
municipal solid waste. The experimental lysimeter was rapidly
brought to field capacity, thus enhancing the chances of viral
detection in the leachate. Likewise, Cooper, et al. (1974a,
1974b) only occasionally detected enteric viruses in leachates
from a series of pilot scale municipal solid waste lysizneters
seeded with poliovirus. The occurrence of virus in leachate was
influenced by the rate at which the test units were brought to
field capacity: viruses were detected in leachate when they were
brought to field capacity rapidly. The investigators speculated
that the sporadic appearance of viruses in leachate may have been
the result of their irregular distribution in the fill or because
of non—uniform flow of the water applied to the fill.
Using a method which detects enteric viruses with about 40
percent efficiency in experimentally contaxninatedleachat es,
Sobsey (1978) examined 22 leachate samples from 21 different
landfills in the United States and Canada for enteric viruses.
The sites represented a broad range of conditions for solid
waste landfills and the leachate samples ranged from 10 to 18
liters in volume. Enterlc viruses were detected in only one of
the 22 leachate samples examined. Two viruses, indentified as
types 1 and 3, were found in an 11.8 liter sample obtained from
newly placed refuse at an improperly managed sanitary landfill
site. From his studies, Sobsey (1978) concluded that leachate
samples from properly operated sanitary landfills do not consti-
tute a public health hazard due to enteric viruses. This finding
was consistent with the results of previous laboratory and pilot
scale studies on the occurrence of enteric viruses in landfill
leachate (Sobsey, 1975).
Municipal leachate and landfills apparently pose a harsh
environment for the survival of viruses, though the mechanisms
of viral destruction is presently unknown. The rate of viral
inactivation in leachates is temperature—dependent and proceeds
much faster at higher temperature (e.g., 20-22°C) then at lower
temperatures (e.g., 4°C) (Engelbrecht and xnirhor, 197 ;
* Plaque forming units
126
-------
Sobsey, et al. 1975; Novello, 1974). Elevated landfill tempera-
tures can thus enhance the inactivation of viruses. It has been
speculated that adsorption of viruses onto solid waste compon-
ents and particulate matter in leachates has in part accounted
for the non-detection of viruses in viral detection studies.
Ware (1980) references various studies (i.e., Carlson, et al.,
1968; Mcore, 1975; Schaub and Sagik, 1975) whjch aV
demonstrated that viruses retain their infectious nature upon
adsorption.
Very little information is available in the literature on
the presence of fungi in landfill leachates. Cook, et al. (1967)
identified many saprophytic (or nonpathogenic) mo1dsTu as
per illuS Penicilliuztt , and ‘usarii.mi in older landfill leach—
ates. Donnelly and SCarpiflO (1981) examined leachates from
cox tercial and experimental field landfills and frequently
isolated yeast cells of basically 3 or 4 genera. Also identified
were the genera of Fusarium, Pen c .11ium, e e on uia,
£ halosPOni 1 and Allescheria boydii . Except for Allescheria
bOvdii , which can cause madura foot abscesses, all fungi are
j rophytiC. In the same study, Donnelly and Scar’pino deter-
mined the numbers of fungal cells, namely yeasts and molds, in
leachate obtained from six different experimental laboratory
lysimeters using Sabouraud dextrose agar plates. The plates were
incubated at room temperature for one week and counted for molds
and yeasts. Although each lysimeter was prepared with different
types of municipal waste, hospital waste, sewage sludge and
combinations of any t o o them, fungi cell numbers were mostly
in the range of <1.0 to 10 CFZr/ml even 23 weeks after lysimeter
construction.
parasit
A review of literature failed to show any reference to
parasites in municipal landfill leachates. Parasites, including
protozoas helminths, and nematodes are a potential constituent
of municipal leachates because of the presence of hwnan and
animal feces in landfills, and particularly when unstabilized
5 wage sludges are received by a landfill. Parasitic cysts and
ova are among the hardiest of fecal microorganisms; many landfill
conditions which inactivate bacteria and viruses are ineffective
for parasites 1 particularly nexaetodes and helminths (Hays, 1977).
Many parasites found in sewage sludge (and thus human feces) are
capable of causing disease (Lu, j ., 1982). Research to
determine the potential for parasites to survive in municipal
landfill leachates and potential health hazards associated with
parasites in leachate is needed.
* Colony forming units
127
-------
Si.tmnary
Research clearly indicates the presence of pathogenic bac-
teria in leachates of fresh refuse in a sanitary landfill.
Viruses are only occasionally detected in leachates, though their
potential presence in leachates of fresh refuse cannot be over-
looked. There does appear to be a significant decrease in
bacterial and viral populations with refuse age or time of leach-
ing. Leachate and landfill conditions appear to present major
obstacles to microbial survival in leachates. Elevated landfill
temperatures resulting from biodegradation can help to inactivate
bacteria and destroy virus. The chemical characteristics of
leachate have also been shown to contribute toward bacterial
inactivation. Little work has been done regarding the presence
and persistence of fungi and parasites in leachata.
PRINCIPAL MECHANISMS LEADING TO TRANSFER OF REFUSE MASS TO THE
PERCOt ATING WATER
Principal mechanisms by which the refuse mass is transferred to
percolating water can be divided into three major categories.
One major mechanism is the entrainment of refuse particulate and
soluble material by the percolating water. The other two major
mechanisms are the dissolution of soluble salts in refuse and
stabilization of the refuse, that is, the conversion of bio-
degradable organic material to gaseous and soluble forms. The
latter two mechanisms may be referred to as refuse solubilization
processes, as. solid refuse matter is converted to soluble form.
Since refuse solu.bilization processes can play a major role in
influencing the quality of landfill leachate, it is useful to
summarize the general sequence of refuse stabilization and to
identify those mechanisms leading to the transfer of refuse mass
to the percolating water at each stage of stabilization.
Municipal solid waste stabilization in a landfill can be
separated into two major biological stages: an aerobic decom-
position phase, and an anaerobic decomposition phase which de-
velops once the oxygen originally present in the landfill is
consumed. The transformation from the aerobic phase to the
anaerobic phase can occur rapidly after refuse placement, i.e.,
in a matter of hours (Myers et al., 1979), or may be postponed
where oxygen persists (e.g., near the landfill surface or in
air pockets*). The aerobic phase of decomposition is generally
*
Fungaroli. and St .ener (1979) showed that pockets of aerobic
and anaerobic activity can exist concurrently within the
refuse. Evidence also suggests that the attainment of
anaerobic conditions is more rapid for refuse of high moisture
(Chian and DeWalle, 1979), shredded refuse (Hentrich et al. ,
1979) and at greater landfill depth (Furigaroli and SteinF, 1979).
128
-------
short because of the high biochemical oxygen demand (BOO) f
the refuse and limited amount of oxygen present in a landfill.
Leachate produced during this phase is characterized by the
entrainment of particulate matter, the dissolution of highly
soluble salts initially present in the landfill, and the pres-
ence of relatively small amounts of organic species from aerobic
degradation. The leachate formed. during this initial phase is
likely a result of moisture squeezed out of the refuse during
compaction and cell construction.
As the initial anaerobic biodegradation processes occur,
acid fermentation will prevail, yielding a low p leachate, high
volatile acids concentrations, and considerable concentrations
of inorganic ions (e.g., C1, S04 Ca 42 , Mg 42 , Na 4 ). n this
state the p generally decreases due to volatile fatty acid
production and the high partial pressure of CO . The increase
in cation arid anion concentrations probably re ult from the
leaching of readily solubilized materials including those
originally available in the refuse mass and those made available
by biodegradation of organic matter.
Initial anaerobic biodegradation processes are carried out
by a mixed anaerobic microbial population, composed of strict
and faculta.tive anaerobes. The facultative anaerobes aid in the
breakdown of materials and reduce the redox potential so that
ethanOgefliC bacteria can grow.
The second stage of anaerobic biodegradation is character-
ized by methane fermentation via methanogenic bacteria. Methane
fermentation generally begins within one year following refuse
placement (Merz and Stone, 1968; Rovers and Farguhar, 1973;
Chian and DeWalle, 1979; Walsh and Kimnan, 1979). MethanogeniC
bacteria require the absence of oxygen, and tolerate only a
narrow pH range (6.6 to 7.4). Nutrients must be present in a
readily assimilable form. The refuse and leachate pH regulateS
the extent of retention of numerous soluble nutrients while the
redoX potential determines whether or not several inorganic
nutrient elements are in an appropriate oxidation state for
assimilation.
Leachate and gas compositions reflecting near neutral pH,
low volatile acids concentrations, low total dissolved solids
concefltrati0 , and high CE 4 contents (‘50%) in the gas are
typical of older landfill s .tes which have already undergone
substantial stabilization of the readily available organics in
the refuse. Although the rate of solubilizatioxi of most con-
stituents is significantly diminished at this point, the
stabilization processes continue to work for many years. A
nall portion of the original refuse organic content, the lignin-
type aromatic compounds, are not degraded to any extent
129
-------
anaerobically and remain in the fill material. These comunds
are important factors in adsorption and complexation mechanisms
as described below.
As mentioned above, the quality of leachate resulting from
the solubilization of the refuse mass can be influenced by a
variety of physical, chemical, and biological activities occur-
ring within the landfill. These activities not only control the
presence or absence of a constituent in leachate, but can alter
chemical species and influence the physiology of degrading micro-
bial populations. Each of the major activities and their
importance in influencing leachate quality is briefly described
below:
The pH of a refuse and leachate system will influence chem—
i al processes (precipitation dissolution, redox reactions, and
sorption) and will affect the speciation of most of the constit-
uents in the system. In general, the acid pH conditions,
characteristic of the initial anaerobic biodegradation state of
refuse stabilization will (1) increase the solu.bilization of
chemical constituents (oxide, hydroxide, and carbonate species)
(2) decrease the sorptive capacity of the refuse, and (3) in-
crease ion exchange between leachate and organic matter.
Redox Potential
The redox potential of a refuse and leachate system will
affect the oxidation state and chemical form of many constit-
uents in the system. Redox reactions are related to biological
activity in the refuse, to diffusion of oxygen from the atmos-
phere, and to addition of oxygen from rainfall. Because of
these factors and the variations in refuse Local envirox ments
(or microenviroriments), redox equi ibri .mt probably cannot be
reached. Reducing conditions corresponding to the second stage
of anaerobic biodegradation will influence the solubility of
nutrients and metallic solids, resulting in precipitation or
dissolution of these constituents. Changes in redox potential
can also alter sorption reactions because of solid species
transformations.
Adsorption and Complexation
Adsorption and complexatiort are probably the most important
processes influencing the attenuation or mobility of trace metal
constituents by the refuse mass. Under oxidizing conditions
adsorption can regulate the concentration of a constituent well
below the level controlled by preciPitatiOn effects. Ligriin-
type aromatic compounds can adsorb trace metal constituents
from the leachate, as can iron and manganese hydrous oxide oljd ,
hydrated altmiintmt oxides, and clay minerals from daily or interim
130
-------
soil cover (Jenne, 1968). In complexation, metal ions combine
with non—metallic compounds called ligands by means of
coordinate-covalent bonds. Leachates are abundantly provided
with such ligands as chloride, ammonia, phosphate and sulfate
as well as an array of organic compounds which provide conditions
under which cozuplexation must be considered in evaluating the
transport and fate of toxic metal ions (Knox and Jones, 1979;
poiüand at al., 1981). In general, complexation acts to increase
the concentration of metals to levels far in excess of their
normal solubilities (Sttn and Morgan, 1970; Snoeyink and Jenkins,
1980). Of extreme importance, however, is the impact of sulfide
..ubi1ity equilibria or the levels of trace metal complexes
whiCh can exist in the presence of sulfide (Pohiand, at al.,
1981). In essence, sulfide effectively competes with most corn—
plexing agents, and consequently many heavy metals will
precipitate as sulfides rather than remain in solution as
complexes. A comprehensive discussion of the impacts of co in—
plexatiOn of metal so1 .thility and the influence of sulfide is
presented by Pohiand j . (1981). The net effect of adsorption
and complexation mechanisms upon leachate composition is diff i—
cult to quantify because of the complex and heterogeneous nature
of refuse and landfill systems.
Temperature affects not only the solubility of many chemical
constituents and the reaction rates, but it establishes limits of
jo1ogical activity. Thus, temperature flux within a landfill
system may affect the reaction mechanisms and biodegradation
rates.
The solu.bilities of many inorganic salts increase with tern—
erature (e.g., CaPO 4 , NaC1), while a ni ther of compounds of
interest in leachate (e.g., CaCO 3 , CaSO 4 ) decrease in solu.bility
with increased temperature. Temperature represents a major
variable on biological activity. For biological systems, each
jcroorqanism will, possess an optimum growth temperature, ranging
from 0°C to as high as 80°C. An increase or decrease from the
optimt n temperature will result in decreased growth due to such
ings as enzyme destruction.
In terms of reaction kinetics, an increase in temperature
ii.i. increase the reaction rate. A 10°C increase in temperature
will roughly double reaction rates, though exceptions include
dissolutiOn and biological reactions.
3 j 0 logical Mechanisms
Throughout the life of a landfill, important biological pro-
cesses re 1ate abiotic reactions, which in turn initiate further
jo1ogical response. The effects of refuse bacteria, fungi,
131
-------
actinornycetes, and protozoa on the chemical states of various
constituents in the refuse and leachate system are numerous.
Biological activity affects oxidation and reduction, rninerali—
zation, i obilizat cn, precipitat .On , aL complexatior.. The
most important factor is the effect of biological activity or
the oxidation-reduction reactions and organic cornpouna trans-
formation.
Oxidation-reduction reactions are greatly affected by the
degradation of organic compounds. The mode of degradation not
only changes with the species of organisms, but it also may
affect the redox potential and thus the oxidation state and
chemical forms of all constituents in both solution and solid
phases of the refuse.
Through the process of mineralization, trace elements,
plant nutrients, organic chemicals, microbial tissues, and
organic—inorganic complexes may be converted to inorganic com-
pounds. Through biological assimilation, the inorganic nutrients
and trace metals may be transformed into microbial tissues, thus
biologically immobilizing these constituents. Organic complexes
accumulating in leachates from microbial synthesis and degrada-
tion have the capacity to combine strongly with trace metals
and other constituents. Through these reactions the constituents
in refuse and leachate can be mobilized, coinpiexed, precipitated,
or sorbed.
Si nmary
In stmimary, the rate of solubilization of a refuse mass j
governed by specific microbial populations, which are in turn
associated with a prescribed set of chemical and physical pro-
cesses occurring in the landfill. Among these are pM effects,
redox effects, precipitation’ ion exchange, adsorption and
complexation, biological effects, physical sorption effects, and
temperature. The interaction of these processes creates a highly
complex and dynamic system which, in turn, creates considerable
variability in the compositiofl. of leachate. Some of the basic
relationships in the refuse—leaChate system are stmmiarized
below:
1. Carbon dioxide and various organic acids formed
during refuse stabilization will reduce the pM
of leachates from relatively fresh refuse, and
consequently will influence the chemical altera-
tion process (e.g., precipitation dissolution,
sorption, etc.) of the leachate constituents
and the biological activities in the landfill.
2. Biodegradable organic compounds help to deplete
free oxygen in the landfill, dropping the leachate
redox potential (Eh), and will thus influence the
132
-------
• chemical forms and solu.bility of various constit-
uents in the leachate.
3. Clay minerals and oxides or hydroxides of iron,
manganese, aluminum and silicate, and complex
com ounds such as organic chelates or inorganic
anions in refuse can adsorb, d.esorb, or complex
the constituents in leachates and influence their
concentration levels.
4. Finely divided particulates in refuse can increase
leachate turbidity and suspended solids.
rACTORS AFFECTING LEACRATE CO OSIT ION
Leachate composition is a function of numerous factors,
jncluding those inherent in the refuse mass and landfill loca-
tion, and those created by designers and site operators. When
viewed as representing a mass of potential pollutants, the
chemical and mi robiological character of refuse is largely
uncontrollable. Similarly, ambient air temperatures and rain-
fall are unalterable characteristics of a landfill site. Con-
versely, refuse density, refuse permeability, refuse depth, and
the rate of water application to the landfill surface can be
regulated. Those factors which influence leachate composition
require identification for purposes of model development and
so that control methods and. landfill design parameters can be
dified for optimum results. This section discusses the effects
o various factors, both controllable and uncontrollable, on
leachate composition.
Refuse Composition
The mass of refuse stored in a landfill may be viewed as
epreSeflting a specified mass or quantity of pollutants. That
portion available for leaching is largely a function of the
physicochelnical character of the solid waste, rates and extent
of solid waste stabilization, and the volume of infiltration
the landfill. Research has attempted to estimate the total
amount of pollutants released from a given refuse mass and the
m jmum concentration of each species in the leachate. To do
this, experiments were developed to simulate the leaching of
solid waste in the landfill environment. One approach, callea
extraction testing, involves mixing samples of solid waste with
distilled water or other solvent (e.g., solid waste leachate),
nd allow contaminants to desorb from the waste for a period of
16—24 lIrs. Another approach involves filling laboratory columns
* Reduction processes such as incineration do represent distinct
changes in the chemical and microbiological character of
refuse.
133
-------
with solid waste, applying water to the surface, and allowing the
water to percolate t ough the waste as in an actual landfill.
Column tests are usually performed for extended periods, i.e.
greater than one year. The cumulative mass of contaminants
leached per kg of refuse for selected column studies is indicated
in Table 39, along with extraction testing results. Note that
for many of the contaminants represented in Table 39, the cum-
ulative mass removals from tests exceed the quantity identified
as available by aqueous extraction testing of the refuse. This
is particularly pronounced for COD, indicating that the biodeg-
radation of cellulose and other complex organic substates cannot
be simulated by extraction tests. Note also the large masses of
pollutants which can leach from the refuse mass. For example,
Walsh and Kinman (1981) report that 1.4 g of iron were removed
per kg dry wt solid waste over a six year period, which is equiv-
alent to approximately 1,270 g (2.8 ibs) of iron removed per ton
of refuse.
The total mass of refuse available for leaching may also
be influenced by refuse surface area and contact time between
the refuse and leaching solution. For example, solid wastes
which are relatively water impervious but which contain water
soluble constituents may exhibit leaching of these constituents
from the waste surface during initial water contact, but not
from the interior of the waste. With increasing contact time
between the waste and leaching solution, additional water
soluble waste constituents may be leached out of the waste.
However, a point of equilibrium may eventually be reached where,
although water soluble constituents are still available in the
waste, the solubility limit of the leaching solution is attained
and no further exchange can occur. Hence, as long as the
pollutant does not reach its solubility limit in the leachate,
more will be removed during a given time period. The relation-
ship between refuse surface area and leachate composition is
discussed more fully in the following section.
In some instances, research has demonstrated a pronounced
influence of specific refuse components on the character of
leachates emanating from the refuse. For example, Kemper an&
Smith (1981) found that the ct nulative mass of zinc leached from
shredded refuse was about one order of magnitude higher than
non-shredded refuse. The investigators speculated that zinc
alkaline batteries present in the refuse were split open upon
shredding allowing the zinc to be released and to be made avail-
able for leaching. In another study, Cooper et al. (1974a)
studied the effects of adding disposable diapers n the composi-
tion of leachate from refuse-filled lysimeters simulating san-
itary landfills. Fecal streptococci and coliform densities in
the leachate were extremely high, indicating widespread fecal
contamination of the refuse. These studies tend to underscore
the importance of refuse composition on leachate composition.
134
-------
TABLE 39. CL%4ULI TIVE )U.32 ME* VAL OF KOFUSE POLLUTAI1TO BY LABOI1ATUKY COLLAINS AND BAUB IIEI4OVAL BY EX7IIACTION TESTING
CumuI attva Na.s Removal tau EKtC.cted
Con.tltu.nt Walsh a,id Kjnmaz (1?I1 ) U 19h sod Bruanar JLSS1) lies et .1. 4 1978 ) W19h (1975) Yu.ujarolt anSI Steimir (*919)
46 ysar period) (S year period)T
— .11 value. lo qlkg dry vt aolid waits axcopt whero noted olhurwl .e -
COO 5% 00 1.5 44
TOC 10 (1 5%)
18 51 (I I)
1 KW 2.3 (50%) 2.1 0.3%
iI.rdne.. 10 15.0
Ca 3.10 0.55
iiq 0.55 0.56 0.56
1.4 417%) 3.5 0.0013 0.22
1.6 4.20
(Al s.ss
Cl 2.2 0.56 1.53
50 1.5 3.46
0.31 0.55
C & 0.12 0.023
Pb 0.11
Nt 0.0015 (101) 0.053
approximately 4001 am of ctu latIve InfiltratIon over * 6 year period
approxImately 4500 am at ci .su lativ. prualpltatlon over a 5 year period
4 ui,Ing synthetic 3.achet.
cumulative sac. reaüv.4 a. pvrcentaJu in .ol (4 w..tn
-------
Refuse Processi
Refuse processing refers to shredding and baling activities
which represent alternatives for refuse volume reduction for
dfillir.q,and for material and energy recovery by many utunici—
palities. In some cases, facilities have been constructed and
full-scale operations are currently underway (Savage and Trezek,
1980). The effects of refuse shredding and baling on leachate
production and composition has been included in major research
efforts sponsored by tJSEPA (Fungaroli and Steiner, 1979; Kemper
and Smith, 1981), and provide the foundation for the following
review.
Shredding--
Rather than just review the empirical evidence on the ef-
fects of refuse shredding on leachate production and composition,
it is useful to identify how, intuitively, shredding should in-
fluence leachates, and then compare the intuitive explanations
with experimental data. In this way, a more comprehensive pre-
sentation can be made to provide better insight to shredding
effects.
The physical characteristicS of refuse are changed after it
undergoes a cormnunition (shredding) process. The characteristic
particle size is reduced at least one order of magnitude as com-
pared to raw refuse. The smaller particle size of in-place
refuse can, intuitively, result in the following effects:
1. An increase in refuse surface area and therefore
greater contact of the refuse mass with the per-
colating water.
2. An increase in in-place refuse density. Decreased
particle size allows for greater compaction.
3. A decrease in the permeability of the refuse.
Smaller particles and increased density favor
a reduced flow rate through the refuse.
4. An increase in the landfill field capacity and
elapsed time before the first appearance of
leacliate. Greater in-place density should
result in greater moisture retention, and
subsequently delay the first appearance of
leachate.
5. A higher strength leachate and greater pollutant
removal. An increased contact time produced by
either reduced permeability or an increased field
capacity should increase concentrations of pollu-
tants in the leachate, and refuse ccimrtunitiofl
should allow greater pollutant removal.
136
-------
6. Accelerated refuse decomposition. The shredding
process serves to “tear” paper and textile prod-
ucts, exposing fibers that otherwise would be
more tightly interconnected.
Research has found that most of the above theoretical
effects of shredding on landfill and leachate behavior do, j
f act, occur. For example, experiments performed by Fungaroli
and Steiner (1979) using experimental cylinders (76 m diameter)
d onstrated that milling (shredding) of refuse increases greatly
the saturated field capacity of the refuse in comparison with
unmilled refuse. The increase in field capacity, however, is
0 linear and approaches a limit as particle size decreases
end density increases. The field capacities reported by
‘u garoli and. Steiner for compacted, unshredded refuse and com-
pactedi shredded refuse are 1380 mi/kg and 1819 mi/kg dry refuse
resPeCti Y. The study by System Technology corporation
cemper and Smith, 1981), using buried lysimeters loaded with
municipal solid wastes and subject to controlled water additions,
indicated that shredding had little effect on refuse moisture
retention in comparisOn with unshredded refuse, but that the
redded waste in their study appeared to have not yet reached
field capacity. This observation is supported by the fact that
measured field capacity (122 mg/kg) was somewhat lower than
jiose measured by Fungaroli and Steiner (1979) and Wigh (1979).
egardiflg the effect of shredding on the elapsed time before
the first appearance of leachate, Fungaroli and Steiner (1979)
dem DnStrated that as the milled refuse size decreased, sub-
stantial delays in first leachate appearance occurred. The
experi ntai . design consisted of a landfill with an eight (8)
foot (2.441n) refuse layer and a two (2) foot (0.61 m) soil
0 over. Infiltration was assumed to be eighteen (18) inches
(0.46 m) per year. Conversely, Kemper and Smith (1981)
observed roughly similar elapsed times before first leachate
apPearafl for shredded and unshredded refuse. Kemper and.
jth point out that dry parcels within supposedly “saturated”
5 edded refuse mass may have some influence on their results.
with regard to permeability, Fungaroli and Steiner (1979)
fQ .1fld results which generally agree with what our intuition
rediCt As indicated in Figure 28, shredding of refuse
i pearS to slightly decrease its saturated permeability relative
the larger, unshredded refuse. The figure does, however,
iLustrate the importance of refuse density on permeability:
j creasifl refuse density decreases its permeability. Eowever,
5 jgnificant relationships between saturated permeability, density,
5 d milled refuse size could not be establisheth
Regarding leachate strength and cumulative pollutant re-
moval, results from each of the aforementioned investigations are
generallY consistent with intuitive predictions. tJsing a series
of mini-lysimeters containing shredded refuse of different
137
-------
Q 0 I I I
0
200 A
—
I OA A
0
r1
> ‘ 160 A A Legend:
0 Size (ian)
• 0
0 000 •A 0.83
0 0 0 320
0
120 A AC 4.80
t •I
• 0
000 0 On 13.50
• • A OE 92.00
r 0 L b
A • 0 A A
ao c i • LA
o A _0 0 •a
00 Q
•o 0 A
.
0.
40 • 0 • 0
— • •00
. if
OA 0 JIQ
00 9 p Q A
A A 1 . 0 % o ’
150 200 300 400 500 600 700 800
Density (rounds t r Cubic Yard)
Figure 28. Relationship between refuse permeability and
refuse density (Fuiigaroli and Steiner, 1979)
-------
oarticle size, and thus refuse densities, Fungaroli and. Steiner
(1979) observed changes in leachate contaminant concentrations
over a 76-week period. Plotting these changes with respect to
refuse size and refuse density, the authors demonstrated that
pollutant concentrations in leachates increased with increasing
refuse density. A lysizneter with a refuse density of 737
pounds/yard 3 (437 kgs per meter 3 ) had the highest weekly pollu-
tant load in 90 percent of the study period. A lysimeter with
the lQwest refuse density of 522 pounds/yard 3 (310 kgs per
meter ) had the lowest concentrations of pollutant in at least
75 percent of the report period. The results show an increased
availabilitY of pollutants to the leaching solution as refuse
density increases. As long as the pollutant does not reach its
so1 bility limit in the leachate, more will be removed during
a given time period. tnterestingly, the results also showed
that except for early transient phases of leaching, particle
size does not appear to have any significant influence or
concentrations. Kexnper and Smith (1981), however, did find that
shredded refuse could. form a. more highly concentrated leachate
in comparison with unshredded refuse. But Kezuper and Smith did
not attempt to differentiate between the effects of refuse
density and. particle size, as did Funqarolj and Steiner. Thus,
refuse density may have been the ultimate cause of high constit-
uent concentrations in shredded refuse leachates observed by
emper and. Smith. Regardless of this point, each study confirms
that as a result of shredding, ].eachates can be of greater
strength than u.nshredded refuse.
Kezuper and Smith (1981) also showed that cumulative mass
removals of leachate constituents on a basis of equal leachate
volume were often greater for shredded refuse than for unshredded
refuse. The authors found that the shredded refuse cell (Cell 4,
Table 40) had the greatest mass removals of alkalinity, total
solids, TOC, COD, iron, copper, cadmium, zinc, nickel, and lead
in comparison with the remaining cell 4 s. However, the authors
point out that the variability in metals results is probably
caused by the large error possibilities associated with the
detection of very low concentrations of copper, cadmium, nickel,
chromium, and lead in the leachates. The authors attribute the
higher cumulative masses removed from the shredded cell to the
increased surf ace area of the waste and enhanced moisture inf ii-
tration, providing more sites for biological and chemical
activitY. With regard to the rate of pollutant removal,
Fungaroli and Steiner (1979) found that milling of refuse to an
effective diameter (D50) of 3.5 to 13.5 mm will significantly
increase the rate of pollutant removal. Below this value, a
decrease in rate of removal will occur. The maximum removal
rate was reported to occur at approximately D 50 — 10 mm.
Finally, in regard. to biodegradation rates, shreddinq.
appears to promote the degradation of refuse, according to Ham
and Anderson (1975). The authors noted a rapid formation of
139
-------
___ Baled, redde refuse
___ Baled, t.mshredded refuse
P 1 d , sbrer refuse
____ Sbredded refuse
Urhr ded, baled refuse ( ntrol)
In s .urmary, the following trends have been confirmed by
limited research with regard to the effects of shredding on
leachate production and composition:
• Shredding can increase the landfill field capacity,
though shredding to extremely fine sizes may increase
the chances for “dry pockets” to be formed within
the refuse mass, which delay attaixmtent of field
capacity;
• By increasing surface area and rates of refuse bio-
degradation, shredding can increase the concentration
of pollutants in leachate in comparison to unshredded
refuse leachates; and
Dartially degraded matter in lysimeters with shredded refuse,
which led to initially high pollutant loadings in leachate.
Later in the study, pollutant loadings were reduced as
highly degradable matter remained. The increased degradation
rates were attributed to the increased homogeneity of the shredd-
ed refuse and the smaller particle sizes.
.Z 40.
CF r T! P P PS K D
A 515 CF J.TAL L t VOLL 4E
JSE
l l8l)
Par3n ter
*
Cell 1
*
Cell 2
*
C fl 3
*
Cal]. 4
*
Cell 5
Alkalinity, mg/kg
Total so11 d , mg/kg
tcc, mg/kg
rr , mg/kg
D, mg/kg
5,450
10,500
6,040
241
9,230
6,960
1.5,200
9,950
411
23,200
9,420
21,200
13,300
514
36,000
17,200
40,200
21,700
712
50,000
9,280
24,200
16,500
729
34,600
Lron, mg/kg
Coprer, pg / kg
C nii. n, .ig/
Zir C, mg/kg
Tjckel, ug/kg
raaIu!t, q/kg
lead, g/kg
273
149.
60
13
539
246
598
714
152
76
20
789
304
654
1,180
490
78
10
1,060
305
564
1,940
152
112
200
1,950
234
856
540
128
84
42
759
167
577
*
Cell 1.
Cell 2.
Ce].]. 3.
Cell 4.
Call 5.
satated with water
140
-------
• Shredding can. increase both the rate of pollutant
removal and the cumulative mass of pollutants
released per. volume of leachate in comparison
to unshredded refuse.
3aliflg
Municipal landfill operations,, by nature, attempt to com-
pact solid waste into the smallest volume possible to prolong
landfill life. A loosely packed landfill is impractical and
uneconomical. For this reason, some landfill operations have
employed the preprocessing step of refuse baling. Baling has
shown to have a pronounced effect upon leachate generation and
quality.
Perhaps the most comprehensive study of the effects of
solid waste baling on leachate generation and composition was
performed by Keinper and Smith (1981). using test lysixneters,
the investigators compared the leachate produced by refuse which
was processed in five different ways before lacement in the
cells, including a cell with baled, shredded refuse (Cell 1),
baled, unshredded refuse (Cell 2), and unsh.redded, uithaled
refuse (control, Cell 5) (Table 40). The first significant
event which occurred upon water addition was that the cells
with baled refuse (Cells 1 and 2) started producing leachate
before the control cell (Cell 5). It was also noticed that the
cells with unbaled refuse retained substantially more moisture
than those containing baled refuse. The authors report that
the results “indicate that baling of refuse has an inhibitory
effect on its moisture retention capability. The tightly—packed
bales are more difficult for water to infiltrate than loose
refuse, and so absorption of water proceeds at a slower rate.”
The authors also note a study by Edens (1978) who also found
that baled waste had absorbed little moisture one year after
placement and after 8 months of leachate outflow. Analysis of
the cores of the bales indicated only a 1 percent increase in
moisture than what was originally present.
In terms of leachate composition, Kemper and Smith report
that the leachates from baled refuse cells contained less
pollutants per liter than leachates from unbaled refuse cells.
3eCaUSe of baling, substantially greater volumes of leachate
was produced. A comparison of the mass flows of pollutants from
the cells on a basis of equal leachate volume (Table 40) shows
the decreasing effect of baling on refuse mass removals in
comparison to non-baling. 3owever, Cell 3 (baled, saturated
refuse) has mass removals of most pollutants roughly equivalent
to those of Cell 5 (unprocessed). Kemper and Smith reason that
saturation of refuse in Cell 3 has obviated the inhibition of
moisture infiltration normally produced by baling, thus causing
greater leaching of pollutants than would be expected from
baled refuse. Thus, it may be that once baled refuse has
attained field capacity, the effects of baling on leachate
141
-------
production and CCm OSi iOfl are removed.
Returning to Table 40 , the metals results do not show the
inhibition of leaching by baling of the waste since Cells 1 and
2 have mass removals approximately the same as those of Cell 5
(unprocessed). An exception is zinc which is clearly reduced
mass removals of Cells 1 and 2 in comoarison with Cell 5. The
variability in metals data observed, the authors note, is pro-
bably caused by the large error potential associated with the
detection of low levels of metals in the leachates. Actual
metal removals may actually be different than that reported.
Looking again at the cumulative masses of pollutants leached
from the cells, it is interesting to compare data from Cell 1.
(shredded, baled) with Cell 2 (unshredded, baled). A greater
mass of every pollutant has leached from Cell 2 than from Cell 1,
including over twice as much COD. The effect is clearly seen
in Figure 29, showing mass removals of COD versus cumulative
leachate volume. In this regard, I emper and Smith point out
that baling of shredded waste produces bales that are even more
resistant to leaching than bales of unshredded waste. The
shredding process evidently allows a reduction in void space
within the bales, thus further retarding the movement of water
in and solids out of the bales. The leaching rate from baled
shredded refuse is reduced in comparison to baled, unprocessed
refuse.
In review, the process of baling can enhance leachate
production by decreasing the elapsed time before leaching,
reducing the moisture—retention abilitY of the waste, and by
increasing the overall volume of leachate produced. Baling of
Waste results in large volumes of relatively dilute leachate
emanating from a landfill. Cumulative mass removals of pollu-
tants from baled solid waste can be lower per unit volume of
leachate in comparison to unbaled solid waste. Eowever, mass
removals from baled waste in a saturated environment (or satu.rat..
ed baled waste) were found to be similar to mass removals of
saturated, unbaled waste.
Evidence has suggested that the processes of shredding and
baling, taken separately can produce opposite results on leach—
ate generation, compositions and mass removals. Baling of waste
results in large volume of dilute leachate with a longer period
of stabilization than required for unbaled refuse. Conversely,
shredding of waste results in the formation of smaller volumes
of highly concentrated leachate, and a shorter period of waste
stabilization. Large mass removal is favored by waste shredding,
while smaller mass removals are favored by waste baling. The
data also show that waste shredding before baling can inhibit
mass removals more than baling alone. Note that all of the
above-mentioned effects apply mainly to solid waste in an un-
saturated state. Once field capacity has been reached’for
142
-------
400
ibaied un-
hsedded
refuse, ugt—
urated with
water
CELL 2(baled. unehredded
efu u)
CELL. I (bah d , shredded
i c C use)
120 140
Cuuiative leachate voluac (tlters)x 102
160
Flqure 29.
MaBs removal of Con vereus cumulative leachate
volume (Kemper and Smith, 1981).
c 1
$4
I ,.
0
.1
320
240
160
00
CELL I (sbiutldud refuse)
20 40 60 dO 100
-------
shredded, baled, or shredded/baled waste, the effects of waste
processing on leachates tend to be less pronounced. It is
speculated that in the long-run, ci ulative mass pollutant
removal per kg of solid waste will be the same regardless of
waste processing, particularly for the so called non-biological
parameters such as heavy metals. Solid waste and leachate
constituents such as COD, TOC, volatile acids, and bacteria
may be reduced in leachates through biological activity if they
can be maintained in the landfill mass for prolonged periods.
Further analysis of this phenomenon is covered in our discussion
on leachate recycling, in Section 6, Available Control Tecnnology.
Landfill Age
Variation in leachate composition and cir ulative mass removal
of pollutants in solid waste is often attributed to age factors,
such as time since refuse placement or time since the first
appearance of leachate. Time will obviously play an importnat
role in determining leachate constituents and concentrations
due to the nature of solid waste stabilization processes,
described earlier. Mote that landfill age, in itself, does not
govern leachate compositional changes, rather it is the rates
of solid waste stabilization and the rate and volume of water
infiltration to the landfill. Age is merely a convenient means
of measuring and monitoring changes in leachate composition and.
pollutant removal from waste. The literature is replete with
data describing leachate quality as a function of time, some of
which is s umarized below.
Merz (1954) examined the quality of leachate from two per-
colation bins containing 3.1 m (10 ft) of compacted municipal
refuse. The concentration of organic components were high in
the first samples of leachate emanating from the bins and in-
creased for five weeks. Initial BOD was 33,100 mg/i and rernajn.ed
high for eight months. An 80 percent decrease in BOD occurred
after eight months; after 13 months the BOO had been reduced
to 375 mg/i. Similarly, Fungaroli and Steiner (1979) examined
leachates from an insulated lysirneter over a 6-year period.
Initially high concentrations of COD (50,000 mg/i) gave way to
lower concentrations (between 20,000 and 22,000 mg/i) after 2
years, and finally dropped below 5,000 mg/i after approximately
3.2 years. Most other leachate constituents exhibited concerl—
tration decreases over a 3 to 5 year period, including iron, zinc
phosphate, chloride, sodium, copper, organic nitrogen, total
solids, and suspended solids. The steady decreases were attri-
buted to the continued flushing of the refuse, thereby removing
the easily decornposibla and soluble materials. Operating field
scale test cells with controlled moisture additions, Encon
Associates (1974) found that the refuse cell subject to continual
flushing with water produced steadily declining concentrations
of leachate constituents over a two—year period. These
144
-------
constituents included all of the general organic indicator
groupS (e.g., COD, SOD) nutrients, major ions, fecal coliform,
and fecal streptococcal groups. Results of leaching studies
performed by pohiand (1975), using 3—foot diameter columns
00 taifliflg 10 feet of compacted refuse and 2.5 feet of cover
soil, also revealed declining concentrations of leachate con-
stituents, including SOD, COD, free volatile fatty acids,
nutrients, and various major ions over a 3—year period. However,
significant concentrations of iron (400 mg/i), and total hard-
ness (lQOmg/1Y , we ye ieachateat.the end oLthe
j ee—year tudv.
Additional studies, including those conducted by Qasim and
Burchinal (1970), Ham and Anderson (1975), Wigh (1979), and
Walsh and Kinman (1981), have demonstrated a general trend:
leachate concentration peaks for leachate constituents early in
the landfill (or simulated landfill) life, i.e., within 2—3 years
refuse placement 1 and gradual declines in ensuing years. Accord-
ing to literature, this trend generally applies to the organic
constituents and general organic indicators (e.g., SOD, COD, TOC,
etc.), nutrients, major ions, and microbiological populations
of municipal solid waste leachates. In some cases, the leaching
behavior of heavy metals has been characterized by this trend,
but more commonly, heavy metal concentrations have fluctuated
widely throughout the study periods. The inconsistency of heavy
metal contaminants may be attributable to precipitation, disso-
lution, adsorption and complexation mechari.isms which can retain
or mobilize the metals within landfill microenvironments.
Figure 30 was developed in an effort to sumnarjze the rela-
tionship between landfill age and municipal refuse leachate com-
position. Data used to develop the figure was derived from lab-
oratory column studies, field test cells, and actual landfill
sites; the data is compiled in Appendix C. For many of the
leachate constituents, upper concentration boundaries are indi-
cated. The boundaries were developed using a first order rate
equation which, in most cases, conveniently explained the behavior
of the upper limits of the experimental data. tJsing SOD as an
example, rate coefficients were calculated as follows:
dCb
-kC (first order rate equation)
C C 10 — kt ( 3 4)
where:
c., remaining (i.e., actual) concentration present at
‘ tiznet
c = maximum concentration (assume t = 0)
b 0
145
-------
• ce2. I,,.. I . 4I4t
Sl.4
:40.4..
.1
30.410
.1
I
I
20.100 •‘
. s.
I
I. I
1 I
3i.300
23.00
2 1.4 1 1
L3. 0 0 1
4 t O II 30 40 II
?tas 1 s 1 g
30 4 £4 20 4
I1. Sh 1 C2.I
Figure 30. RelationShiP between landfill age and leachate
composition. (Note: solio circles represent
actual landfills, triangles — field test cells,
and hollow circles are laboratory col ns
146
I Reproduced From
I best available copy .
I
I
-------
20 30 II
t&as. CL434tS TWU
t0
$0
“.44.
LI. 300
4. 4 00
I • 200
2.401
1.3.341
$ 0 0 0
“a.
10 0 0
I
4300
2000
LOIS
I
33 - 30 40
PLO. 103.. ca.O.. YSIII
(Contirn ed) from
Figure 30. best avajlabje copy
147
40 •
3 4.000
32.100
*t 3II
31.344
I
.3 • 230
‘4.4
a
•1 •
4
.4
.1
• t
a
.
.
a —
a
S.
a
r
• a a
I —
3 1 .4
‘ II I
4244
$200
II ’ .
32
2300
I
• II 20 I I 40
Pt.. 133O. C3a.uir, T41Ib
P lO. ILlO. 333. ITa I
-------
24
S
I
*
I
a..
‘4.
4i
Z .4
I
I La :s 34
L alS C a * Tw ’
I.
‘4
‘I ,
‘3
S.
4
‘3
33
a
21
24
S
I -—
I 2* 2* 41
?* I {La. s IT. IJ
Figure 30. (Cor tinued)
148
I
a
LI 24 20 44 30
?3. ttaS. C ..•. . YsaSni
S S
. 5
S
S •
S
S
P , 5 S
S •
I
I
S
-------
211 l
I
i I I
.
,1.
LI I
m
I
.
. 1
•
I
•.
4
.4 .
I
4
S
.
.
I
• .
. I
LOll
4S
‘SI
?2I
LIII
LIII
‘ S.
II I
II
4201 I
S
S
S
S
2
S
.4 S
I
I r
.
rk. T 4
L I a YI
?4 l&IIS Ci.aasls I?sslI
1204
I
II I
I’ll
33..
2 1 1 1
2144
L.a.
I 1410
4
‘ I I
I.
LI a oi
U. S. C34.. rTs.,e
I
I
4
I
I.
‘a.
13 54
LIII
NI
440
4 1 1
I ll
S
LI 35 a oi
?4 $L C1 . e 2.01 54
N
• LI a a 40
2311 I1 clau01 (25011)
so
Figure 30.
(Continued)
149
Reproduced rrern
best avaitab e copy
-------
U. 3
30.0
0.3
.4
0.3
S. .’
LS 20 34
?Oas C I...r.
I
4 ’
I
a
a
24.4
22.3
.3. 2
S..
4.0
LO 20 10 40
...aI, fTasts
Figure 30. (Continued)
150
U.’
LI. I
?. I
I
2.&
.
a
I
2I
34 14 44
?0 $&a 3mi . rT.ws
S.
I
3
a
S
a S S
S. S S •
S.
-------
LIII ‘T
a
‘S
‘ II
a
4
41
Ill
32
4 4S
1 4I 1 • 1 24
• I
r.
a a
4
L I I a
S
________________________________I ________________________________
1 LI I LI 20 10 40 I I
?t 1L 2. 1 (1 1 . .u—.
to
• 2 8
l OS
LI
311
13 1
304
27,
t
244 $
2 .
I
7 .4
I
I ( •
S
S
to, at?
OS I
S
4*
.
3• 5
4.
a
at £ I 3
2 ,4 25 I I 4* a. S LI 28 II 40 II
?$ 144(5 C2 (I1(S “ ?tls $2,a ( 5 C2,., . 2..,.,
Figure 30. (Continued)
I Reproduced from
151 best available copy .
-------
rate coefficient based on 10 (to be calculated)
Taking the logarithm of e uatiOfl (34) we have
log Cb = log Cb - kt (35)
setting log Cb = Y, log Cb = b, k rn, and t X, by
0
substitution Y b + mX, which is an equation of a straight line
with slope m and Y intercept b. Thus, by plotting the logaritb
of 30D 5 concentrations indicated by the boundary line shown in
Figure 30 against the time value at those concentrations
(sexnilogaritbmiC paper), the graph is a straight line with Slope
k, the rate coefficient. Because the graph is a straight line,
the rate ecuation is of the form given in equation (34).
In most cases, the rate equation is adequate for the de-
scription of observed behavior in a limited time range. There-
fore,equations derived are generally applicable for landfills
of age greater than three years and less than thirty years. For
some constituents, concentration data did not permit the formu-
lation of a rate equation and rate coefficient, due to lack of
sufficient data or the absence of an apparent trend. Trends
were generally absent for heavy metal constituents. Apparently,
the leaching of heavy metals does not necessarily follow the
patterns established by the organic indicators, nutrients and
major ions. It is speculated that the mass removal of heavy
metals from refuse does not correspond to stabilization pro-
cesses. Whereas the release of organic constituents (and
probably nutrients and major ions) is closely tied to the
various stages of refuse stabilization, heavy metals do not
this behavior. A parently considerable anounts of heavy
metal pollutants are associated with inorganic refuse constit-
uents, and their release may largely be a function of rate of
flushing of waste and of the character of the leaching
solution (i.e., a more acidic leaching solution may enhance
metals release). The absence of trends may also be attributed,
to precipitatiofl dissolution, adsorption or com 1e ?ation
mechanisms which can retain or mobilize the metals within
landfill micrcenviroflmeflts. It should be noted that consider-
ably less data was available for heavy metals in leachates
from older landfills (i.e., ‘ 20 years), and that trends may
appear as more data is gathered.
The leaching patterns of specific organic constituents and
microbiological populations as a function of time since refuse
placement require further elaboration. These topics are dis-
cussed below.
OrgaflicS
Municipal landfill leachates often show the presence of
organic compounds, which are either flushed out of the refuse
152
-------
by the leaching solution or produced during the biodegradation
of refuse. Relationships between the general organic indicators
(e.g., COD, 30D, and TOC) and landfill age ( first page of
Figure 30) indicate a substantial reduction in leachate organics
following landfill closure. Studies have shown (Lin., 1966;
Pohland, 1974) that free volatile fatty acids can represent
the bulk of the organics present in leachates formed during
the early steps of refuse stabilization, representing the acid
fermentation stage of anaerobic degradation of the refuse.
Figure 31 shows the trend of accumulation of organic acids
produced by ground refuse in a test cylinder during early stages
of stabilization (Lin, 1966). Figure 31 indicates that acetic
acid was the only significant compound found in the first few
days of leachate production, Thereafter, butyric acid pro-
duction increased rapidly and amounts exceeded all other acids
formed during the first two months of leachate production. Lin
indicated a tendency for the amount of caproic- acid to increase
after 60 days with a prolonged period of fermentation.
With increasing landfill age, humic-carbohydrate—like
compounds and fulvic—like materials become more predominate
(chian and Dec’Zalle, 1977). Chian and. DeWalle (1977) identified
the various classes of organic compounds as a percentage of TOC
and related variations to landfill age. Their results are
presented in Figure 32. The volatile acids production,
corresponding to the first stage of anaerobic degradation,
represents the major organic fraction during the early years of
landfill life. The volatile acid, carbohydrate, and protein
concentrations decrease relatively rapid.ly with increasing
landfill age. The aromatic hydroxyl compounds present in humic
and fulviclike fractions of leachate organics, show little
decline with increasing landfill age, and are a major portion
of identified TOC in older landfills.
The declining concentrations of specific organic classes
of leachate TOC with increasing landfill age are also seen for
organic degradation in natural environments. Carbohydrates
degrade rapidly with increasing time of burial or increasing
depth in sediment while a less rapid decrease is observed for
amino acids (Degens, 1964; Ishiwatari, 1971). Aromatic com-
pounds such as huinic and fulvic acids, are generally less
subject to decomposition than aliphatic compounds (Kohonova,
1961; Shapiro, 1964). Based on research findings and. available
literature (Chian, 1977) concluded that resistance of organics
to bacterial degradation follows the pattern: free volatile
fatty acids < carbohydrates, proteins, and humic—like substances
c aromatic hydroxyl, carboxyl and fulvic—like substances.
The stability of organic compounds influences the leach—
ability of trace metals. Many organic compounds containing
nitrogen, oxygen, and sulfur can form soluble complexes with
metals and thus promote leaching. Humic and fulvic acids are
153
-------
28
? CF F L (days)
Figure 31.
Volatile acids found in leac1 ata during
early stages of sta ilizatioti (Lin, 1966)
20
20
16
I:
0
0
20
30
40 50 60
154
-------
i 00
90
80
70
60
50
40
30
20
10
0
0 2 4 6
T (years)
8 10 12 14 16
Figure 32. Trends in the identified fraction of
leac ate TOC vs. landfill age (C ian and
DeWalle, 1977)
155
-------
considered strong ccrnplexing ligands (Reu.ter and Perdue, 1977)
and could be important in long term leaching of trace metals
from a municipal landfill. In studying the complexation of
cadmium by organic components of sanitary landfill leachates,
Knox and Jones (1979) found that high (>10,000) molecular weight
components contributed significantly to complexation. The
authors suggested that these leachate components might contain
phenolic hydroxyl groups having a stability constant toward
cadmium of the order of l0 . Conversely, Chian and De alle
(1977) found that the majority of metals in leachates are not
associated with organics having a molecular weight larger than
500. Knox and Jones (1979) stated that the role of cornplexation
in determining the fate of metals in sanitary landfill leachate
depends umon several factors whose importance had not yet been
adequately investigated or reported in literature.
Microorganisms-—
The effects of landfill age on the microbial populations of
leachates was alluded to earlier in this section. In review,
municipal solid waste leachates may contain significant bacteri-
al populations, but these populations show a significant
decrease in den .ty with refuse age or time of leaching. A
number of environmental factors are responsible for reducing
these populations, including high landfill temperatures, low
leachate pH, and the bacterio—statiC action of leachate organic
acids and metals. There is some indication, however, that bac-
terial populations can survive .n landfills and be present in
leachates after extended periods, i.e., up tO 9 years since
refuse placement. More research is clearly needed n this area.
Enteric viruses are rarely found in landfill leachates.
Because viruses are obligate parasites and cannot multiply out-
side host organisms, their numbers are also expected to decline
with increasing landfill age. However, evidence suggests that
viruses adsorbed to solid waste may be able to retain their
infectious nature for extended periods. Viruses can be destroy-
ed at the elevated landfill temperatures characteristic of the
initial stages of refuse decomposition. The potential health
hazard associated with viruses in leachate has yet to be
resolved.
The presence of parasites in leachates has not been studied
though sewage sludge and fecal addition to municipal landfills
provides a source for parasites to enter leachates. Parasitic
cysts and ova can remain viable for extended periods in harsh
environments, and their presence in municipal landfill leachates
should be of public health concern. With regard to fungi in
leachates,the paucity of research to date shows that the major-
ity are non—pathogenic in nature. Their number in leachate
appear substantially less than bacterial populations.
156
-------
Rate of Zater Application
The rate of water application to a landfill can influence
its behavior in a number of ways, and concomittantly will affect
j..eachate composition and pollutant removal. Assuming that the
rate of water applied to a landfill is proportional to its
infiltration to the fill material, a greater rate of water
application will accelerate the attainment of landfill field
capacity. Studies have shown that once field capacity is reach-
ed, leaching closely follows the pattern of moisture input,
j djcating that little change in internal landfill moisture
OCCUrS. Figure 33 shows the schedule of moisture addition and
the observed leachate from Fungaroli (1975). Because of this
“input-output” leaching behavior at field capacity, it is found
higher rates of water application to a landfill will produce a
more dilute leachate than lower rates of water application
(Walsh and I imnan, 1981; Wigh and Brunner, 1981). Studies have
also shown landfill microbial activity is related to moisture
levels, i.e., biodegradative processes are iminished when land-
fill moisture remains below a certain level. In light of these
facts, it is feasible that high moisture levels in a refuse mass
may result in leachate of distinctly different composition than
leachates from refuse of lower moisture content.
Several studies have addressed the effect of water applica-
tion rates on sanitary landfill leachate composition, the results
of which are somewhat conflicting. Leckie, . (1979) con-
structed five large refuse test cells and sub ected them to
varjoUs moisture conditions through the controlled application
of excess water, septic tank pumpings, and recycled leachate.
A comparison of cells with and without continuous flow-through
of water showed that continuous flushing brought about a signif-
icant trend toward lower concentrations of dissolved materials
with time than the non-flushed cell. The results exemplify the
important effects of dilution on leachate quality. Both inor—
ganiC and organic salutes were reduced at equivalent rates.
The authors also note an accelerated stabilization of refuse
materials during continuous flow-through. Using a laboratory
lysimeter 6 ft square by 13 ft deep, Fungaroli (1975) simulated
seasonal variations in rainfall through artificial moisture
applications. From his experiments, Fungaroli suggested that
leachate produced during the slow attainment of a landfill’s
field capacity will probably exhibit initial pollutant concen—
trations different than a landfill in which substantial
quantities are produced immediately. However, he adds, “once
a system transients have eliminated, both systems should produce
similar, but not necessarily identical, leachates.” In another
* using laboratory-simulated landfills, Chian and DeWalle (1979)
found that the moisture content of refuse must be above 75%
but below 100%, based on dry weight, to maximize gas production.
157
-------
0.3
0.2
z
3
a
a
0.1
Days
Figure 33. Moisture addition and leachate response (Fttngaroli,
1971).
158
-------
study using laboratory lysimeters with controlled moisture
appliCatiofli Revah and Avniinelech (1979) stated that contaminant
concentrations in leachate were independent of water application
rates. The authors demonstrated that for a given refuse type
or density, the concentrations of leachate pollutants as a
function of time were the same regardless of the volume of water
cassing through the column. These results contradict the re—
gults of virtually all other laboratory and field-scale
jnvestigatiofls to date.
Two systematic efforts to determine the affect of moisture
application on leachate quality were performed by Walsh and
i jnman (1981) and Wigh and 3runner (1981) and provide perhaps,
the best evidence of the influence of water application rates
on leachate quality. tn the Walsh and Kinman study, experimental
lab—scale columns were constructed and subjected to a different
amount of controlled moisture infiltration. The study continued
for almost six years. The results showed that though constituent
concentrations in leachates may vary with time depending on water
application rates, the cumulative mass of constituent leached per
cumulative leachate volume is generally the same, regardless of
the infiltration rate. Wigh and Brunner (1981) found similar
results. In their study, a pilot-scale cell and three Landfill
columns were operated for nearly eight years. Leachate contam-
inant concentration histories and mass removals were plotted as
a function of cumulative leachate volume, and were found similar
for varying water application rates. In comparing the similar-
ities in the experimental results of several experimental land-
fills, Brunner (1979) demonstrated that meaningful comparisons
of leachate quality can be made when results are expressed as a
function of cumulative leachate volume. Figure 34 shows the
results from Walsh and Kinman (1979) expressed in this manner.
The horizontal axIs normalizes the differences in infiltration
rates over time. Although total solids on the vertical axis
is a gross parameter which ignores the relative compositions of
specific constituents, Figure 34 presents an ordered account of
otherwise scattered leachate concentration data. This relation-
ship between leachate constituent concentrations and cumulative
leachate volume (or cumulative infiltration for landfills at
field capacity) serves as the basis for several modelling efforts
discussed later in this section.
Note that not all leachate constituents exhibit the above—
mentioned relationship between pollutant concentration and
cumulative leachate volume. Heavy metal constituents, particu-
larly iron, showed considerable variability in concentration as
a function of cumulative leachate volume. Whereas the mass
removal of other leachate constituents showed obvious decreasing
trends as leaching continued, iron release appeared somewhat
linear with time, even after 5 years (Walsh and Kinman, 1981).
Apparently 1 iron is not as readily flushed out of a landfill
as are other pollutants.
159
-------
__ 72000
• Municipal refuse leached at 813 mm/year
60000 a C l Municipal refuse leached at 406 mm/year
O ) a 0 Municipal refuse and sewage sludge leached
. 48000 )b at 406 rnm!year
A Muiiicipal refuse with 3% by weight of
0:0 CaCQ and leached at 406 mm/year
A
o uuuu
0 0O
(I) • AA0
. 0 • a .0
= 24000
o 0 ç osn
(I) & o
— Lb 0 T1O
(ii d 0 . 00
.4—a
o 12000
F— ••o.
0 I I I
0 1.0 2.0 3.0
Cumulative Leachate (Liters/Kg dry refuse)
Figure 34. Leachate total solic s concentration from experimental
landfills subject to different moisture application
rates (from Brunner, 1979)
-------
tnterestingly, organic leachate constituents a pea.r to be
flushed out of the fill as readily as most inorganic constitu-
ents. Straub and Lynch (l982 ) suggest that high moisture flow
rates can carry soluble organics and microbial cells out of the
fill, and that in such cases microbial action appears to have a
minor role in determining leachate quality. At lower flow rates,
anaerobic microbial activity is thought to be a significant
factor governing leachate organic strength, i.e., a significant
reduction in organic material can be achieved.
In terms of landfill design and operation, these relation—
ships reveal that the long term pollution potential of leachates
can be diminished with initial high rates of moisture applica-
tion. Provided that short-term leaching problems can be solved,
high moisture application rates can remove the bulk of contami-
nants early in landfill life.
p pth of Leached Bed
Refuse depth, or lift height, is a design variable which
operates in conjunction with the rate of water application to
influence leachate composition. To illustrate, consider a
solid waste landfill consisting of compacted refuse. Some
fraction of the vol e of the landfill will be void spaces
capable of holding water or gas. Water entering the top surface
of the landfill and travelling down through the refuse will move
from one void space to the next until it eventually reaches the
bottom sii.rface of the landfill. Polluting material coming in
contact with the percolating water is transferred from the solid
to the liquid phase due to a concentration difference between
them. Thus, water percolating through refuse will successively
accumulate contaminants until the solubility limit of the
leaching solution is attained and no further exchange can occur.
For a deeper landfill, the chances of leachate attaining its
solubility limit for various constituents is greater than for
a shallow landfill, because of greater contact time between
refuse and percolate.
However, much depends on the rate of water percolating
rough the refuse. Water percolating slowly through a landfill
provides greater contact time between the solid and liquid
phaSeS and increases the chances for a high strength leachate
to result. For high infiltration rates, contact tix e is dimin—
jshed between the two phases and a more dilute leachate is
expected. The concept of a residence time in the refuse for
a given volume of liquid can be used to anticipate the effect
of changes in both infiltration rates and refuse depth. That is,
for any liquid travelling the full depth of the refuse, the
rssidenCe time (and leachate strength) should increase directly
with the depth and inversely with the infiltration rate.
161
-------
One of the earlier studies confirming the importance of
refuse depth was performed by Qasim and 3urchinal (1970). These
researchers found substantiallY greater constituent concentra—
tioris in leachates from taller concrete cylinders (exceptions
were hardness and calcium). Deeper fills under similar condi-
tions of precipitation and percolation were found to require more
water to reach saturation, require a longer time for decomposi-
tion, and distrthute the bulk of extracted material over a longer
period of time. LaboratOrY column studies performed by Phelps
(n.d.), demonstrated the effects of depth on leachate composition.
Columns filled with refuse of varying depths (i.e., 38, 76, and
114 ) and su.bj acted to equal annual net infiltration showed
distinctly different concentration histories over a 560 day
period. Figure 35 illustrates the concentration histories of
300 and chloride, as well as indicate the simulated concentration
histories from the model developed in conjunction with the study.
The figure clearly shows that higher concentrations of leachate
contaminants emanated from the deeper columns. The deeper col-
umns allowed greater residence time in the refuse for a given
volume of percolate.
Given that saturation limits of leachate might be reached
for a deeper landfill, and that increased infiltration rates
allow for a greater mass of contaminants to enter the leachate,
pollutant concentrations in leachats should decrease more rapidly
in time for shorter residence time or greater infiltration rates.
In other words, since the ratio of the mass of liquid per unit
time and the mass of refuse it passes t ough is greater at
shorter residence time, polluting material is flushed out of the
refuse more quickly at greater infiltration rates.
Landfill Tem erature
Although chemical and biological processes generally are
temperature—dependent both in terms of rate and ultimate equi-
libria, at the present time it is not reasonable to speculate on
the effect of temperature on leachate composition and refuse
stabilization except in general terms. Research has shown
significant differences in landfill (experimental and field
scale) temperatures with time, depth, and mode of operation.
Major studies and results are summarized below.
Temperatures in the refuse mass are found to reach a maximum
generally within a few days of refuse placement. The high tem-
peratures correspond to the aerobic decomposition phase of refuse
stabilization (Chian, et al., 1977). Merz (1954) found maximum
temperatures in a refuse bed to range from 138° to 157°F within
8 days following refuse placement in wooden percolation bins.
Fungaroli and Steiner (1979) observed a peak temperature of 130°F
at the center of a laboratory lysimeter to occur within 10 days
of refuse placement. At depth, Wigh (1979) observed a peak tem-
perature of 110°F to occur the day of lift construction,
162
-------
Figure 35.
Legend: increasing
col in depth,
3>2>1
(Note: nu .bers rep-
resent observed values,
the lines were calcu-
lated by a leaching
znodel)
increasing co1u nn
depth, 3 > 2 > 1
I
=
C , ’
E
U,
0
0
‘ -4
‘4
0
C.,
1
.0
S0.0
240.0
320.0 400.0
Time (days)
Leachate concentration nistory curves
for simulated landfills of different
depths (from Phelps, n.d.)
163
-------
indicating a short period of active aerobic decomuosition.
Following the aerobic decomposition phase of refuse stabi-
lization temperatures at depth gradually decrease, signalling
the onset of anaerobic decomposition. Fungaroli and Steiner
(1979) found lysimeter temperatures to stabilize at approximately
80°? in 60 days following refuse placement. Evidence front sev-
eral studies, however, indicates that temperatures within the
refuse mass are not static, rather there is a tendency for refuse
temperatures to vary with ambient air temperatures, particularly
with seasonal variations. Leckie, al. (1979), for example,
studied temperature variation in compacted refuse of five field
scale experimental cells. In general, there was an apparent
thermal response of the upper several feet of landfill to the
mean ambient temperature, which varied seasonally in a range of
approximately 20°C over the annual cycle. Deeper refuse material
(e.g., 3 meters) tendad to show a much smaller thermal response.
The authors noted that the effect of the variation in mean a n—
bient air temperature was significant only to the extent of the
variation over an annual seasonal cycle. Temperatures in the
refuse beds followed the ambient temperature cycle with an
observed time lag in their response.
Other studies noted similar landfill thermal response to
ambient air temperatures. The refuse mass in test cells studied
by Wigh (1979) showed response to seasonal temperature varia-
tions, though the temperature amplitudes within the cells were
reduced with increasing depth. Temperatures at 2 m depth ranged
between 47 and 67°F over the year. Merz (1954) found refuse
temperatures in bins to fluctuate with seasonal temperature
changes, front a low of 75°? in February to a high of near 100°F
in summer months.
The thermal response of refuse beds to changes in ambient
air temperatures correspond to those presented in literature on
both the diurnal and annual cycles of heating and cooling of
soil, which also shows a typical time lag in heating and cooling
at depth compared to the annual or diurnal surfe ce cycle (Gieger,
1965; Singer and Brown, 1956). From the information presented
above, it is expected that there is a landfill depth below which
temperatures would remain essentially unchanged year round, and
this temperature is typically close to the average annual air
temperature near the ground surface. Because of active anaerobic
decompositiOfl landfill temperatures are expected to exceed soil
temperatures at a given depth (Riehl, 1972; Strahler, 1971).
Elevated landfill temperatures may exhibit significant ad-
verse effects on the survival of pathogenic agents. Englebrec
and Amirhor (1975) studied the stability of enteric bacteria to
elevated temperatures by adding cells to leachate at two dif-
ferent temperatureS (72°? and 131°F). The persistence of enteric
bacteria was found to be significantly less at 131°?. Wigh (1979)
164
-------
inoculated shredded refuse withpoliovirus type 1 and Salmonella
der and placed it in a field test cell to evaluate the micro—
ganism movement. Both pathogens were found to be inactivated
in less than 10 days. Temperature appeared to be the essential
organism inactivation factor. Estimated peak temperatures were.
ly L40 F. From these observations, elevated tern—
peratu.res corresponding to the aerobic phase of refuse decow osi-
tion appear to be unfavorable for pathogen transmission via
leaching though additional research is required for verification,
rtherassessmeflt should be conducted.
s .umua
Research on several factors which influence the composition
of municipal landfill leachates has been reviewed. Attempts
were made to identify how some of these factors can be controlled
in order to modify the character of leachate. In summary, refuse
processiflcat including shredding and baling, can significantly
affect leachate composition. Refuse shredding tends to increase
landfill field capacity, can increase the concentration of pollu-
ants in leachate in comparison to uxxshredded refuse leachates,
and leads to a increased rate of pollutant removal per volume of
leachate in comparison to unshredded refuse. Conversely, baling
promotes th production of a more dilute leachate, a delayed
attainment of landfill field capacity, and smaller !fl&SS removal
per volume of leachate compared to shredded or unbaled refuse.
Regarding the rate of water application, increased rates can
lead to a more dilute leachate and greater mass removals as a
ftmction of time than lower rates of applied water. Righ ntois-
t ire application rates can remove the bulk of refuse contaminants
early in landfill life. Increasing landfill depth will generally
promote a higher strength leachate, though at high inf ltratjon
rates the effect is diminished. Landfill temperature, a largely
incontrollab1e factor influencing leachate composition, have been
ShOWfl to 4 fluctuate with seasonal ambient temperature variation
near the landfill surface, but amplitudes are less pronounced
with increasing landfill depth.
y ECTS OF A DI G SLEDGES TO MT ICIPAL LANDFILLS
The addition of municipal wastewater treatment plant (WWTP)
5 dges or industrial sludges to municipal landfills is reported
to provide both beneficial and adverse effects on leachatas. In
uarious studies, WWTP sludge and refuse admixtures have acceler-
ated the rate of stabilization of biodegradable organic matter.
ewever , sewage sludge/refuse codisposal alternatives may pose
greater adverse environmental and public health impacts than
coiwefltioflal refuse disposal. Concern has arisen that the
additi0’ of industrial waste may result in the occurrence of
,arious toxic elements in leachates in excess of that encountered
m, nici;al refuse leachates. The addition of industrial sludges
165
-------
may also pose a deleterious impact on the biochemical stabili-
zation processes within the landfill. These effects of adding
sewage sludges and industrial sl dges to a municipal landfill
are identified in the following sections.
Addition of WWTP Sludges
A number of column and field studies have been performed to
investigate the effects of sewage sludge codisposaJ. on leachate
composition. Pohiand (1975) found that seeding municipal refuse
with primary sewage sludge in test columns increased rates of
biological stabilization of organic pollutants in refuse and
leachate in comparison with non-seeded columns. Encon Associates
(1974) reported that the addition of septic tank punipings to
municipal refuse accelerated the tnethanogenic process, as evi-
denced by increased methanogeniC activity related to the control
cell (refuse only). The study also showed that leachate bac-
terial loading did not exceed the refuse—only cell; leachate
organics and TDS were reduced, while inorganic ion cOflCefltZation.
were largely unaffected. Stone (1974) found that admixed WWTP
sludge and refuse produced an acidic leachate with a higher BOD
than the solid waste leachate, but that the chemical compositio
did. not significantly differ. Lu, et al.. (1982) determined that
codisposal leachatas posed no more of a pollution threat than
municipal refuse leachate, though cocisposal can accelerate
leachate formation through the additional moisture provided by
the sludge.
Leachates of WWTP sludge and refuse codisposal operations
appear similar to conventional refuse leachates. Reported invest..
igations differ in regard to whether sludge can enhance removal
of organic matter from refuse and leachate through accelerated
stabilization. A paucity of information exists regarding the
potential public health impacts of sewage sludge/refuse co-
disposal landfilling (Lu, et al., 1982). The degree of health
risk associated with WWTP residuals is proportional to the
degree of microbial survival from the sewage treatment processes.
An increase in pH from lime conditioning, heat treatment,
chlorination, dewatering, and incineration will significantly
diminish the survival of pathogens in sludge.
The transmission of pathogens and toxic material from w p
sludge/refuse landfill operations can occur primarily via sur-
f ace waters and ground waters. Epidemiological evidence of
disease occurring as a result of WW P sludge codisposal landfill...
ing is lacking. jtrate—flitrOgefl and enteric pathogens (i.e.,
bacteria and viruses) appear to represent the greatest potential
to ground-water contamination from leachates of sewage sludge!
refuse codisposal operations.
166
-------
Addition of Industrial Sludges
Industrial wastes vary greatly in content, physical char-
acterist±cS, and potential for environmental degradation. In
many cases, the potential for inclusion of highly toxic contain—
jnants in leachates is greater than for WWTP residuals or
municipal wastes. Limited research has been conducted concern-
ing the potential toxicity of various industrial sludges.. The
few studies on the subject, however, are su maarized as follows.
Streng, . (1976) evaluated the leaching character-
istics of several industrial residuals admixed with municipal
refuse, including: refinery sludge (PS), battery production
waste (BPW), an electroplating waste sludge, (EW), an inorganic
pigment sludge (IPW), a chlorine production brine sludge (C BS),
and a solvent based point sludge (SBPS). Notable components of
these sludges were: Cu, Fe, Eq and moisture for RS; Cd, Fe,
Cu, Pb and asbestos for SBPS: Sn, Sb and moisture for 3PW; Cr,
Fe, As, Cd., cyanide, and moisture for EW; Be, Cl, asbestos and
moisture for IPW: and Ni, Pb, Cl, asbestos, and Hg for CP3S.
Sustained maximum concentrations of leachate contaminants as
compared to refuse are presented in Table 41. Leachate contam-
mat concentrations from a sludge and refuse mixture exceed
those of solid waste, particularly for those components in
abundance in the respective sludge. Other important observa-
tions included: (1) field capacity was obtained earlier for
the majority of the sludge/refuse cells, probably due to the
added sludge moisture, (2) industrial sludge leachates exhibit
both a viricidal. and. bacteriacidal effect.
The leaching trends of chemically stabilized and unstabi-
lized industrial wastes in conjunction with municipal solid
waste (MSW) was investigated by Myers, et al. (1979). The
resu.Lts are s mmtarized as follows: —
(1) When codisposed with MSW, the release of major
metal contaminants from treated or untreated
electroplating sludge was not observed.
(2) Untreated chlorine production brine sludge
when codisposed with MSW, releases significant
quantities of Al, Cd, Cu, Cl, Hg, Na and other
dissolved solids. Chemically stabilized chlor-
ine production brine sludge when codisposed with
MSW significantly reduces the mass release of
toxic metals (e.g., Cd, Hg, and Cr).
(3) Codisposal of MSW with calcium fluoride/sewage
sludge apparently improved leachate quality
with respect to BOD, COD, TOC, alkalinity,
pH, and iron.
1.67
-------
TABLE 41. SUSTAINED MAXIMUM CONCENTPATIONS OF LEACHATE CONTAM-.
INANTS FROM STRZNG (1976)
Wmste
Ce.U
g Jb
4
RS
9
3P J 4
10
E u t
12
1P’.
13
14
Vsnadiu
ArseniCk
Se l .niu’
Anci ony 1
rink
Su1fur
Asbesos
Cyanide 1 ’
A.tuminunh
P eno1h
3erylli 1’
Leadk
ritasiush
Mercury
3erou
Nic 1 ’s.1 .
Zinc 1
<0.03
12.2
<5
<5
1 50
674
<100
<1
<0.03
4.3
<0.1
1310
cO.].
19.7
18.1.
1.360
4.3
<0.03
<3.0
(1.3)
<3
(800)
NSAn
<1.00
<1
(1.2)
((22.91)
(Q.9)
1450
cO.].
(31.2)
2.8
1.010
2.1
((0.29])
( [ 30.0])
<5
<5
(1200)
491
<100
((1.2])
(8.1)
N ?
((45.01)
((3380 ])
<0.1.
7.2
LT 3
1.319
(5.2)
<0.03
( [ 230))
c3
(3000)
<100
<1
<0.05
NA
((30.01)
(1820)
<0.1
(26.9)
(22.0)
((3505])
0.3
((0.23])
13.0
((10.01)
((17])
(2400)
(846)
((6000])
5.0
(5.4)
NA
4.5
(2140)
(0.1
(67.1)
(24.0)
(1490)
((1.1.21)
<0.03
((54.01)
<5
<5
((8000])
( [ 33621)
((3000])
< 1
((11.21)
((48.03)
((64601)
(0.6)
((3281)
((83.01)
((60201)
(8.0)
Station
4 (6) (1]
3 (8) (5]
3 (7) (3]
0 (13) (5]
0 (12) (101
igher conce rations a ’ have been detected.
b 501 4 Waste.
C•Refinery Sludge.
dBattery ?roduc: on Waste.
!1ectrop1ating Waste.
tnorganic Pigment Waste.
gc j .orjx e Production 3rine S1i dge.
) indicates .xce*dtd solid waste.
( 3 indicates ixceeded maxi.aus.
!Th. d.r3Core indicates less than solid vast..
e.Insuificient sasple available.
7ib.rs/ t
9 Not anaJ.yzed.
168
-------
utilizing residential-type solid wastes alone and in con-
junction with a metal plating sludge, Pohiand and Gould (1980)
indicated the possi i1ity for significant reductions in leachate
heavy metal concentrations and key roles of redox potential, p i
and sulfides, hydroxides and hydroxy-carbonates in sequestering
metals. Studies designed to determine the limits of heavy metal
loadings which can be sustained during codisposal without de-
leterious effects on landfill stabilization processes are
required. Studies on the reduction of these effects by addition
of appropriate sequestering agents are also needed.
There is still little known concerning the effects of in-
dustrial sludge on refuse decomposition processes and leachate
produced during decomposition. Evidence suggests, however, that
the addition of sludges, particularly those high in trace and
heavy metals, will result in elevated metal concentrations in
leachates. There is also evidence suggesting certain wastes or
landfill conditions can iobilize trace and heavy metals.
LEACEATE COMPOSITION MODELS
The development of a descriptive and predictive model of
the leaching of municipal landfills and the composition of
leachate can be advantageous for the following reasons (Phelps,
1) To assess the. potential impacts of leachate on
receiving waters;
2) To aid in the design of a leachate treatment system;
3) To help estimate the concentration of contaminants
entering an underlying ground-water system so as
to determine leachate concentration in that system;
4) To assess the possible effects of codisposal of
various liquid and/or semi-solid wastes with
municipal solid waste in landfills.
This section of the report has thus far described leaching
processes and significant factors affecting the generation and
composition of municipal landfill leachates. Though the leach-
ing process involves a ni . .ber of variables whose interactions
are poorly understood, some generalizations can be made. For
example, little leachate is collected at the bottom of a land-
fill until the field capacity of the refuse has been reached.
The field capacity of refuse has been similar among various
studies, and appears to be a function of in-p lace density
(Fuzigaroli , 1971; Revah and Avnimelech, 1929). Once the land-
fill has attained field capacity, leaching patterns closely
follow the input of moisture to the landfill, indicating that
little change in internal moisture occurs. In addition, though
169
-------
the concentration of contaminantS in leachate may vary from
study to study, there is a strong correlation between leachate
concentration and the total volume of leachate produced (Wigh,
1979; Walsh and ininan, 1981). Leachata contaminant concen-
tration reach a maximum level near the onset of leaching and
decrease as leaching continues.
Whereas models developed for estimating leachate generation
(Chapter 3) rely on observations of field capacity, water jnpu
to the landfill, and a water mass balance to route moisture
through a landfill, they make no attempt to predict leachate
quality. In this section, modeling efforts which describe and
predict the quality of leachates from municipal landfills are
reviewed. The basic leaching behavior generalizations described
above as a starting ocint, the models attempt to estimate
leachate quality as a function of time or cumulative leachate
generation.
Curve Fitting A oroaches
There are generally two approaches used when attempting
model the composition of leachates as a function of time or
cumulative Leachate volume. One approach is to quantitatively
describe the physical, chemical, and biological processes which
occur (or are assumed to occur) during leaching. Because land-
fills are dynamic environments, these types of models tend to
be complex, and make a number of simplifying assumptions. The
other approach is to avoid mathematical expression of leaching
mechanisms and to focus merely on leachate concentration his-
tories. In this approach, empirically equations are developed
which describe the shape of curves derived from experimental
data on the concentration of leachate constituents versus time
or cumulative leachate volume. Because the concentration
histories of leachate constituents are highly dependent on
variables such as refuse character, refuse density, landfill
depth, and rate of water application 1 these “curve—fitting”
models are generally effective only in modeling the experimental
data from which they were derived.
Revah and Avnimelech (1979) used the curve-fitting approach
to model the concentration history of organic carbon, volatile
acids, T , NE , NO , Fe, and Mn in leachates from refuse—filled
columns. The teep y declining concentration of these leachate
constituents was described through an exponential function:
C — ab (36)
where:
C a the concentration of leachate constituent,
t a time,
a a concentration at t = o, and
b — the exponent±al base
170
-------
EquatiOn (36) above was used by the authors to determine the
length of time (t:) needed for a leachate constituent to reach
a certain 1imitin concentration (C ), i.e., the concentrat .on
below which the leachate is not er.dlngering water quality. The
model parameter t was calculated for columns containing raw
and shredded refuse, varying refuse densities, and for coli ns
filled during different filling seasons. According to the cal-
culation, potential pollution by sanitary landfill leachates,
continues for about 3 tO 15 years after the placement of the
solid waste. The fastest decay was associated with the least
dense columns, filled during the Winter. The leaching period
is extended for sh.redded and compacted refuse. These observa-
tions are in agreement with previously reported research on
the effects of refuse shredding and COm actjort. The authors
also report a general similarity between field data and those
obtained fl this model system, indicating that exponential decay
adequately characterizes the leaching process.
Pigh (1979) also employed the curve—fitting method to model
aching patterns. Noting the relationship between leachate
concentration and the total volume of leachate produced, Wigh
proposed an equation for two consecutive first-order reactions
having the form: -k v -k v
kiCk Ce l—e2)
I c —k ———• ( 37)
21
where:
C — the concentration at any volume of leachate
and Ic rate constants
Cj unIc own concentration related to the contaminant mass
available in the refuse
v — the volume of leachate collected per unit of surface
area.
Since volume of leachate collected is some function of time, the
equation could be changed to a time dependent function.
Tra in experimentally determined leachate concentration his-
tories, a trial and error process was used to determine the best
visual fit for reaction curves and rate constants. FLgu. e 36
shows a reasonably good fit for leachate COD and chloride. While
the equation can fit well the concentration histories of various
leachate constituents in this experiment, the fit is an empirical
one, with constants having significance only for the particular
experimental landfill for which they were evaluated.
Knowing the rate constants, it was possible to compute Cj
in equation (37) and solve for the total mass of a refuse con-
stituent that might be leached. By integrating equation (37)
ough an infinite volume of leachate, the resultant total
leachable mass in mg per unit of surface area was:
171
-------
,o,000 —
300 t ,000 L, 00
tvt TtVE t2A . .r! ‘:o .u cou.zcr o eu u iVr (3f SU 1A AREA, • ‘
O. —ixp.r a.ntzL data
— quatian (37)
0 tOO 200
ctnIsLAT1V .L ClA!! vou cOulCT!0 LEE UNIt 0 ? SURfACE
Figure 36. Simulated and experimentally derived
concentration histories for COD and
chloride (from Wigh, 1979)
r ira Ifrorn
£72
O— xp.rim.ntai data
A ua or% (37)
G ,000 —
30,000 —
20.000 —
3
I-
I-
‘S
t
0
p.
2
5
a
a
2,100
1 .300
1. 1 00
1,200
900
600 ’
400 600 R0O 1.000 1,200
-------
M Cj/k (38)
The total mass of dry refuse available for leaching and percent-
age removal at 1500 nm . of leachate is presented in Table 42.
If equation (38) is an accurate representation of the leaching
process 1 extraordinary removal had occurred after only 1500 mm
of leachate.
TABLE 42. TOTAL AVAILABLE MASS REMOVAL (from Wigh, 1979)
parameter
Available Total Massa
Removal at 1500
COD
89.6
70
Sulfate
1.41
75
Chloride
2.73
81
MagnesiUm
.500
86
Iron
3.26
31
a g/ g of dry refuse
b percent removed obtained by dividing mass actually removed
at 1500 mm by total mass calculated from Equation 38.
Process Modelinc A roaches
There has also been a few attempts at developing equations
which represent a fl2ndamental description of the leaching pro-
cess which can be applied to other landfills. These models
help describe and predict moisture and contaminant movement in
sanitary landfills. The models are sensitive to such factors
as refuse placement and composition, hydraulic phenomenon and
landfill configuration. The models range in levels of sophis-
tication—the more sophisticated the model, the more responsive
the models must be to landfill construction and solid waste or
media properties.
Qasim and Burchinal (1970) assumed a simple physical pro-
cess to describe leaching from sanitary landfills. The theory
of co1 fln operation was applied to the leaching of chloride
during the vertically downward movement of water. Chloride was
used as a surrogate parameter of landfill leachate. The ap-
proach assumed that the chloride concentration may be determined
on the basis that adsorption Occurs at a rate proportional to
the concentration of solute in the liquid phase, and by the
difference between the actual and maximum possible concentration
of the solute adsorbed on the solid particle. The model is
173
-------
responsive to the physical parameters of depth of refuse, refuse
compaction, and cumulative leachate volume. Experimental and
theoretical concentrations of leachate constituents in leachate
saxvoles showed fair agreement, with maximum deviations of 30%.
Theoretical calculations tended to underestimate chloride con-
centrations as leaching continued. The duration of the leaching
experiments was only 163 days.
Another physical process model of the leaching of solid
waste has been developed and tested by Phelps (n. d.). In con-
cept, the model likens a landfill to a packed column, wherein
a solutoin percolating through the column picks up contaminants
from the solid phase due to a concentration difference across
a stagnant film of liquid on the particles in the solid phase.
The concentration on the particle surface is assumed to be
proportional to this amount of leachable material er unit
volume of refuse (voids and liquid included). The mathematical
derivation of the model is quite complex, involving a series
of partial differential equations.
The model was tested for a special case assuming a consta.z t
rate of infiltration and low initial refuse moisture content.
The analytic expression is a function of time, refuse depth and
refuse moisture content, and thus requires only three empiri-
cally derived parameters. The experiments used to test the model
cover a six-fold range in both refuse dpeth and infiltration
rate. The experiments also cover two sizes of refuse particles
and the presence or absence of septic tank sludge. Six test
results for each experiment were used: 800, total residue, cal-
cium, iron, COD, and chloride. The correlation coefficient,
expressed as a 2 , between the calculated and observed curves
for oncentration vs. tine was calculated. The resulting values
of R , generally exceeded 0.9.
The calculated curves generally agree with the data within
a few percent over the initial upper 75% to 88% of the concen-
tration values. Figure 35 shows the calculated and observed
concentration curves for SODS and. chloride for a given set of
experimental parameters. Th figure shows good agreement from
the time ].eachate first appears to about twice that time. For
longer times, the observed concentration values are consistently
somewhat above those calculated. The extent to which the later
data values exceed those calculated was found to depend on
infiltration rate and the addition of septic tank sludge.
The trends in the empirical parameters of the Phelps study
indicate the effects of the various experimental operating
conditions: infiltration rate, particle size and septic tank
sludge addition. These empir±cal parameters provide a basis
for direct comparison of leaching curves for widely varying
174
-------
0 nditiOfl5. cwever, it is questioned whether the observed
trends in the empirical parameters provide a basis for applying
the model in its present form to field situations.
Finally, a series of models have been developed by Straub
and Lynch (l 82a) which describe separately the leaching pro-
cesses of inorganic and organic constituents in landfills. These
models represent the best attempt to systematically describe and
predict leachate volume and quality from sanitary Landfills.
Like other models discussed thus far, the fundamental relation-
ips among landfill variables provide a framework for the
modeling. Unlike the previous models, however, the level of
gophistication has been greatly increased by the formulation of
specific hydraulic, physical/chemical, and microbial processes.
The types of leachate behavior simulated with the models
include:
1) Hydraulic behavior. This model simulates the moisture
flow in a sanitary landfill based on equations of unsaturated
flow in a porous media. The model is an extention of Darcy’s
quation insofar as a finite difference rt erical solution
is developed. While the water balance method of modeling
j..eachate production (Section 3) may yield reasonable predictions
of leachate production in some cases, it does not describe
internal rnoist ire flow details or contaminant movement. This
model provides a continuous description of moisture flow, which
can subsequently be used in modeling of contaminant transport.
The model applied knowledge of the h.ydraulic properties of
refuse and soil layers, as well as properties of soil layers
which affect landfill moisture flow. Typical forms of hydraulic
functions were applied to refuse and yielded good simulation
of moisture movement. However, specific empirical information
on refuse hydraulic characteristics is limited, requiring
additional research.
2) Inorganic leachate composition. The concentration
and relative strength of inorganic contaminants (represented
by total solids) is described by both a simple single reactor
modal and vertically distributed unsaturated flow model. The
gjmple reactor model considers the volume of refuse as a single
well mixed reactor (Figure 37), and that leaching can be
described as a simple exponential function of cumulative
leachate volume. The leaching pattern of inorganic constituents
is described as a function of the net infiltration rate of
jsture, the volume and field capacity of the refuse, and the
rate of generation of dissolved substances.
The vertically distributed model describes contaminant
transport in an unsaturated porous medium, and was developed
175
-------
q 4 , C 4
as:a ui
i f1ow rate, “o]./t n.
C. inflow conesntratio ,
ass/vo1
q — outflow (leac at )
C a outflow ortcertt:aeiort
C) — vo1 et:ic moisture
con tent
r — oon:amina t 5er e s:e
rate, mass/ti e.’vcL
Figure 37. Single well - Lixed reactor tnodel of a land-
fill (Straub and Lynch, 1979)
176 _________________
[ eproduced from
I best available copy .
q, C
-------
o be used with the unsaturated moisture flow model as described
j (1) above. This one-dimensional 1d.netic model allows the
simulation of the continuous transport of contaminants through-
t the vertical death of the landfill, for inputs of moisture
content, contaminant concentration, and arbitrary source terms.
The dilution, dispersion, and convection of these contaminants
by the flow of water is described. The mathematical derivation
of each of these models is extremely complex, and interested
readers are asked to consult Straub (1980) or Straub and Lynch
(1982). Figure 38 shows the observed total solids concentration
versus time for an experimental test cell as well, as results
simulated by the vertically distributed model.
In general, the simulations obtained by the vertically
distributed model, represening major hydraulic and contaminant
transport processes, are in overall agreement with leachate
data documented in various studies. However, additional experi-
mental work is required in order to identify and refine the
ScriptiO1 . of basic processes occurring in the landfill.
3) Organic leachate composition. Using the general frame-
work of the models of inorganic leachate composition, the organic
],,eachate composition models add the action of microbial digestion
of organics present in landfill moisture 4 and the movement of
oxygen in the landfill. The single well-mixed reactor model,
presented in Figure 39, depicts these and other relationships.
Separate aerobic and. anaerobic microbial populations are assumed
toutilize solubilized organic material, creating increased cell
mass and gazes of decomposition. Active cells are subject to a
death rate. Dissolved oxygen in the landfill moisture is uti-
lized by aerobic digestion, and may be replenished by the
dissolved oxygen present in influent moisture and by transfer of
oxygen from the gas within the landfill itself. The model
ass eS no direct microbial utilization of solid organics occurs;
organics must become dissolved and mixed in the moisture before
digestiofl can take place. Thus the model represents, “the
,roCeSSes of mixing and convection of several contaminants in
e landfill moisture, interactions between them imposed by
microbial action, the generation of organics into landfill
moisture, depletion of solid organics, and oxygen exchange”
(Straub, 1980). Each of these relationships is described
by mathematical expression. Figure 40 shows the measured and
simulated leachate COD for the single reactor model.
A model of four cascaded well—mixed reactors as in Figure
41. was developed as an extension of the single reactor model.
Each reactor represents one quarter of the depth of the landfill.
The material convected out of reactor i equals the material
171
-------
70.000
60.000
i -I
U ’ s
50.000
0
4J
RI
40.000
.0)
-JO
0
30 .000
20.000
I
I
I
I
I
500
600
700
000
900
Figure 38.
Measured and
history from
1
/
/
/
/
/
I
/
F
/
/
v;
/
/
/
Sisiuitt te A TS
— —
I
100
200
300
400
days
simulatod results o total solicis concentration
Straub and Lynch (1979)
-------
AEROBIC CELLS
IN INFLUENT
AERO8IC CELLS
IN EFFLUENT
I I
_________ 02 TRANSFER ___________
GAS TO LI0 O
3
T L M OISIURE
L...O2UTILIZEDBY ____
\ AE Ros’ c rY
ORGANICSN
LAN OFILL MOISTURE
.1
0RGANICS UTILIZED
J r—BY CELLS
AN AG RO B IC
CELLS
LO THOF ______
ACTIVE CELLS
REACTOR BOUNOARY
•I ‘LEVEL
SINK
- MATERIAL TRANSFER
Figure 39.
single well-tnixed reactor for
leachate or a.nics (Straub a i
model of
Lynch, 198Th)
Oi IN
INFLUENI GAS
02 IN
EFFLUENT GAS
‘1 OXYGEN IN
•—-4 LANOFILL GAS
ORGANICS TRANSFER
SOLID TO UOUIO
ORGAN ICS IN
SOLID WASTE
07 IN
INFLUENT MOISTURE
02 IN
LEAC IATE
ORGAN ICS IN
JNFUJEMT MOISTURE
ORGAN ICS N
LEACHATE
ANAEROBIC CELLS
TN INFLUENT
ANAEROBIC CELLS
IN EFFLUENT
ORGANICS UTILIZED BY
AEROBIC CELLS
ACTIVE AEROBIC
CELLS
L DEATH OF
ACTIVE CELLS’
179
-------
Figure 40. Measured and simulated results of leachate COD concen-
tration history for single reactor model from Staüb and
Lynch, 1982b)
20000
15000
0
E
i;-10000
0
0
5000
MEASURED
SIMULATED
/
— —
,
/
t--I
I
I
I
/
00
250 500 750
DAYS
1000
-------
Each Reactor Vo1 me Contains:
Solid 7aste
Constant Moisture Content =
Field Capacity
- Anaerobic Cell Concentration
- Organic Contaminant Concen—
tr at ion
Moisture Outp’ .it
Fig. re 41.
vertically cascaded reactor model of landfill
(Strai .tb, 1980).
4 /InPut Moisture
Reactor
1
Reactor
q
Reactor
- 4
1 1
-------
convected inro reactor i + l• Processes within each reactor
include: the generation of organics from solid waste to
moisture, utilization of organics by anaerobic cells, and
convection of organics and cells. This model is thought
to have greater general capability, but requires greater
empirical support in the development of hydraulic and con-
taminant dispersion processes. The model is mathematically
portrayed as a series of partial differential equations,
which are derived by Straub (1980). Figure 42 compares
the simulated leachate COD with measured COD from Pohiand
(1975).
The success with which the models described above simulate
actual leachate concentration histories demonstrates the feasi-
bility of analyzing the solid waste leaching as physical/chemicaj
and microbial rocess models. Reactor models with simplifying
assumptions of moisture content, moisture flow, and mixing have
been shown to be useful in explaining observed data. These
models provide a framework by which the dynamics of leachate
volume and quality are described. Because of their complexity,
consideration of details of specific situations may be given
with models based on the partial differential equations of u —
saturated moisture flow and contaminant transport in porous
media. L ayered effects, vertical distribution of materials
anc aiec ia properties, hydraulic phenonLenort, ana moisture anti
contaminant source/sink effects can be reflected i these
models. The systematic prediction of leachate behavior and
landfill stabilization for particular landfill configurations
and designs is provided by detailed simulation of continuous
flow of moisture and contaminants. However, as model complex-
ity increase and greater detail is simulated, empirical support
required for the formulation of the model also increases.
Straub and Lynch (1982a) claim that although the models demon-
strate promising predictive abilities, their application to
many situations will require a broadened empirical base.
Suimna.ry
Table 43 presents a general swiary of the leachate composi-
tion models previously discussed. At present, leachate composi-
tion models are appropriate primarily for research purposes,
insofar as mathematical expressions developed to describe the
leaching process or concentration histories merely interpret
experimental results. Additional experimental work is required
in order to identify and refine the description of basic pro-
cesses occurring in the landfill environment. Eventually, the
process models could be applied to field scale problems, al-
though such applications would require extensive empirical
support.
182
-------
20000
16000
0
0
C)
4000
Figure 42. Measured and simulated results of teachate COD concentration
history for vertically cascaded reactor model from Staub and
Lynch (1982b)
1000
12000
8000
0
/
I
/
/
I
/
I
I
I
SI’
MEASURED
SIMULATED
‘¼
‘ ¼
03
1
0 200
400 600 800
TIME (DAYS)
-------
IMLE 43. S1I*4MIY 4 )1 SIACIIAIL C1 lP0Sl1lDlI NilItiS
Qasi. cud
Burchinal
(1970)
Revah and
Avniielk h
(1979)
Wi9h
(1979)
IL. theory of coltnu* operation
md nass transfer Is applied
to the leaching of chloride
trun experlaental landfill
colujuins. 1 1w beacbing of
chloride over tine, among
other hectuat. constituents,
was best sbeulmted by desorp-
then breakthrough curves.
The concentration histories
of 14 other contani naitl s were
related to chLoride estinetes 1
with reasonable egreenent.
ihe d cc i in. of concentrations
of organic carbon, volatile
acids, 1KM. 11114, 1103, Fe, and
Mn over tine were described
through an exponential function;
where
C • the cuuiceutrat Ion of
the constituent
I — tine
a — concentration at L • 0
I • the exponential base
The concentration hhtory of
various ieaciute constittuciuts
Was described Iks -ough a soil-
nuipirical equation. The
equation Is based on two
consecutive first-order
reactions hiving use fain
k12M -kV • kV
C.- 1 -——— (u ‘-e 2 )
k2_Il
1.
use void space of packed refuse
is equal to susee perce.itage of
the height of the fill.
2. IL. leaching liquid is in enuilib-
luau with the solid beiis legtlied
at paints of the bed.
I. Peak concentral Ions of Teecisate
constituents occur at the onset
of leaching.
2. Leachsate constituesst concentrations
decline according to exponential
decay curves.
1. The total leachable ness of refuse
is represented by
H • Cut
where;
• an eeperiuusoustally calculated
constituent concentration
• an eaperiuse iutal ly derived
rate Cuuistasst
Model respuussive to the depth
of the ref use, refuse cuie4uact luau,
nuI tlsucusiuiattve vultause of
leachate geoir.ted by the landfill.
Model responsive to refus. density
as-id the effects of shsediling as
they effect couutneiuua,.t couscen-
trations as a (unction of tine
Since the onset leacbi*g.
Model resiIu.ssive to cujaul tiv.
volune at leachate generated
and total leachable ness of
refuse, but only for landfills
of sheila ,- rOfuse, density,
depth, etc.
Fair agresusaunt. Devi t lusts
(run experissuental results
generally ranges betweess
SI and 301 over a 163 day
testing pus-hid at one
selected depth.
Fair agreussessi.
General sinilarity between
field data and those
obtained by this ussodcl.
Good ags-euuesut. use
equation fits well the
pattern of toastaeiusmnt
concentration decsej es (or
cunulat iv . leachate vuluae
collected. ituwever, siiice
the fit Is t,m iricel in
nature, the equation dues
lEt represent a Iuuid.usscntal
description of tIe ie cbIsug
Model kesiionsivCsuess To
Agreussent of Mudel
Researcher(s) Modet Concepts Asswsptio..is landfill Cisaractesistics
WILL fxperIai eiutal O,ttg
‘-a
co
C - ats
process.
-------
lAtilt 41. (Cuutinued
Nodal Pespousiwuass to Agreonestt of Hudul
Landfill Charertesistics With txperiaesilal i) .ta
Wigli
(Coat Inued )
Phelps
( . d )
Where
C • pollutant caiicestrat loa
N • total leachable .ass per
melt surface area
V • vo*i. of laachate colLected
per unit surface area
1 1 k 2 • e.ipirically fit rate
cou stants
this model utilizes utass transfer
quatio.s based on flow through a
stagnant film on tu e refuse parti-
cles. Tb. model is built upon
both theoretically and eapirically
derived parameters. to last the
model, calculated concentration
values are lit to observed values
for six test results; BOO, total
residue. Ca, fe COO, and Cl.
the experiments used to test the
nodal cover a slit-fold range in
both refuse depth and iuliltration
rate, Iwo sizes of refuse parti—
des. and the presence or
absence of septic tank sludge.
2. live concentration histories
of ieachate contaminants can
be represented by the curves
described by consecutive first
order reactions
I. the initial moisture contestS
of the refuse is low enough
that the first leachate to
appear can be assnued to have
travelled the full depth of
the refuse.
2. iii . fluid velocity i ii live refuse
is zero before the wetting front
reaches a give. depth. and con-
stant and unifone for all points
at that depth alter the wetting
front has arrived.
3. lIve rate at which soluble material
(mass) is transferred (row the
solid phase to the liquid phase
is proportional to the difference
in concentrations between these
phases.
Pollutant concentrations in
leachates ate described as a
functio, of time, leichale
flow rate end the depth of
the refuse.
fair agremeent. Cosreiati.si
coefficients between the
calculated and observed
curves (or cuncesatra Iusi vs.
time exceeded 0.9 (It’). 11w
calculated curves generally
agree with the data (rule (lie
thee leachate lIr t ai lpeastd
to about twice that (leo.
for longer times. the uhsusv-
ed conceutrat ion values ave
consistently above those
calculated.
4. the depth of the refuse does not
change with time.
6. The laachtata can be considered a
dilute solution--changes In
leackale concenttatio* are assumed
to taut effect its density a,id hence
its velocity.
Researcher(s)
Nodal Concepts
U’
-------
lAME 41 (CuisIii w .i)
Phelps
(Continued
Two .odels are proposed which
portray the coisceetrat ion
histories of leachute inorganics
is a (unction of lime since
leachate appearance. One
model presents refuse as a
single well-.lsed reactor.
providing mass Lalances
(or both .solsture and
contaminant yields. The
other Is a vertically distributed
model Of contaminant transport
in is unsaturated porous aedit*.
Calculated concentrations are
compared with nispirical data
(ron several pest lisent studies.
6. A constant moh tuse liii I I-
tnt Ion sate is ass,assod.
br the single well-shied reactor
nodel. müdel hIasavsuters include
sisnistuic l i ii i tiitiou rate.
volume. depths, and fluid capacity
of selusO.
bus the vestlt iIy distributed—
ws gtu,at d flew *odel , isiudel
pusd.sietuis include leachable
cussijisslssesit mass, refuse moisture
co.stesst • v h tasme dopth desss I ty
and field tapacity of refuse and
.ioistuie .sppl ic tltsis rgte 5.
Very good agrecusuisi.
Simulated vs experiawsmtal
results show the lezclsliig of
Inorganic cuist ssssissaist s is
explainable i i i tusisss 01
(us4aissental trais 4murt pie-
cesses.
I. landfill moisture hums Is by
gravity drainage only, i.e..
no .solsture diflusiws and iso
cu islaislisa,st dispersionj
diUusion out of the system
occurs.
2. Concepts of hydraulic end
cosst.smlnasst transport in
soils can be applied to solid
waste.
Researct ser(s Model Concepts
Straub and
Lynch
(1982a)
Model Re 5ieuss iv0sse 5 to Agreumssent of Model
A s sta s spt l oi s s taismif Ill tlwrecteristics Vith (aperimisesslal ihmta
1-
a’
For the single wuli-mix d reactor
model;
I. Landfills can be isoriraved as
a single well-mixed scactur.
2. Contaminant concesslratlui.s OS
infiltrate are zero.
3. Refuse moisture Couatesit russ ia ins
essential I) constaist at field
capacity.
4. internal leauhsate guises-at hint is
negligible.
for Ike vertically distributed
unsaturated flow model:
(Cent issued)
-------
IMLI 41 Cunclu4ed
*csearcher(s)
Model Concepts
Model Kesponilveness Fe
landfill Cioracterist Ics
Agre mont of Iknlel
With Ixperluental Data
I- ,
— I
Ike concentration histories of
leachale orgenics are modeled.
three models were developed. based
on an unsaturated flow of moisture
contaminant generat ion and trans-
port • oxygen exchange. and micro-
biI1 activity. One ii a single
well-mixed riactor mode) similar
to that described above. Tue
second Is a vertically cascaded
weIl-aixcd reactor model. The
third incorporates ueuturated
flow and contaminant transport
modeling, described ahove, to
simulate organic Icackate
strength. Calculated con-
centrations ar. compared with
empirical data iron several
pertinent studies.
tech of the reactor models makes
simplifying asswliltions regarding
me lstui-e storage, moisture flow.
and contaminant distribution in
order to locus on contaminant
generation, microbial activity.
oxygen exchaal4Je. and dilution/
convection effects.
tack of the reactor models is
responsive to landfill volume,
void volume, moisture content
at field capacity, moisture
flow rates, concentrations of
sobstances in moisture entering
the la,idf ill, utlimate mass of
leachable constituents per unit
bulk volume of refuse, and an
assumed maximum constituent
concentration.
Very good agreement.
tack of the models
closely approximates
experimental COO
concentration histories,
bet the unsaturated flow
and contaminant transport
model is mole precise, and Is
tixnsjit to have greater
field scale potential.
Straub and
lynch
(1982 14
-------
SECTION S
LEAC AT! MIGRATION
INTRODUCTION
In this sectIon, interactions between leachate and under-
lying soils are addressed. An extensive review and analysis of
the movement of leachate contaminants through underlying soils
has been performed by Weston, Inc. (1978). This section sun—
marizes and provides supplementary information to the findings
of this study.
Soil is a dynamic physical, chemical, and biological system
which may be envisioned as material containing biodegradative
matter and chrotnatographiC properties, with numerous finite
variables that influence leachate migration. For instance, in-
organic species may be adsorbed or complexed by soil components
or leached from the surface of soil particles. Organic sub-
stances may be ultimately mineralized, yielding inorganic
substances, harmless gases, and water. Organic matter acownu—
lates as resynthesizad substances of microbial origin and as
degraded organic residue (Alexander, 1971; Fuller, 1978; McCarty
and Rjttman, 1981).
The attenuation and migration of contaminants in the soil,’
water system are influenced by the following physical, chemical
and biological mechanisms:
• Diffusion and dispersion;
• Dilution;
• Straining;
• precipitation/dissoluticn
• Adsorption/de sorption;
• Complexation;
• ton exchange;
• Redox; and
18
-------
• Microbial activity.
Although these mechanisms can be grouped as above, clear
j 5 tjnCtiOflS cannot always be made because of the simultaneous
occurrence of numerous interactive reactions. Moreover, a fac-
tor which may mobilize or attenuate a constituent in one soil,
may have no effect in another soil; a factor which may inhibit
movement of one element may have no effect on the movement of
another element. Each of the preceeding phenomena will be dis-
cussed in the following subsections in relation to the effects
on the applicable physical, chemical, and biological mechanisms.
The multitude of reactions that are operative when contami-
nants enter a soil can be, as listed above, identified with
reasonable assurance, but quantitative data relating to the
attenuation of individual constituents by specific mechanisms
are not available, consequently, considerable effort has been
placed on the development of sophisticated leachate migration
models. However, due to the highly complex leachate/soil envi-
ronments, no leachate migration model exists that can simulate
all of the physical, chemical, and biological processes occur-
ring in a typical landfill system. In this section, emphasis is
placed on a discussion of existing conceptual—mathematical
models, since these models appear the most promising, but,
unfortunately, also the most complex ones for predicting the
performances of waste disposal sites. An extensive list of
available simulation models is given, and the advantages and
jsadVantageS of these models for simulating landfill behavior
are discussed. Des .gn examples are not presented because of
the large number and complexity of simulation models. Inter-
ested designers and investigators are referred to the original
literature for detailed discussion and methodology.
,ZACHATE/ENVI RONMENTAL INTERACTIONS
cil Properties
The soil is a heterogenous, polydisperse system of solid,
liquid, and gaseous components in various proportions. The
solid component of the soil is made up of primary minerals, clay
minerals, and hydrous oxides, together with organic matter and
living organisms, forming a. polyphase system of more or less
discrete particles or aggregates. Separate particles vary wide-
ly in size and shape. The finest particles are clay——2 or less
in diameter, which constitute the major portion of the soil col-
loidal material. The organic matter, particularly humus, and
clay, affect many o the properties of soil, especially surface
pheflometta (Bear, 1955; Stuima and Morgan, 1981). In this hetero—
genous systems the soil solution acts as the medium by which
chemical reactions between members of the different phases and
of the same phase are made possible. In the USDA system, the
goil texture designation (e.g., sandy loam, silty clay), is
1 9
-------
based anountscf sand-, silt-, and clay—sized particles
in the soil. The t .angular diagram used to make this textural
classification of the soils in the USDA systam is shown in
Figure 43.
Soils are very porous bodies. About 40 percent of the vol—
ume of dry, compact sand and about 60 percent of dry, compact
highly organic soils is air. By volume, about 50 percent of the
most couunon soils (loams, silt loans, and fine sandy loans) ce —
sist of pore spaces which are of a cellular nature (Walkman,
1952; Alexander, 1971). This porosity is an important quality
that affects the movement of leachates in soils and determines
biological activities, the exchange of gases, chemical reactions,
and biological changes.
Soils are chemically sorbent bodies consisting of: (1)
inert chemical compounds, (2) difficult and easily soluble sub-
stances, (3) soluble salts and acids, and (4) conmiex insoluble
compounds assessing the property of exchanging their basic
elements for other basic elements and hydrogen. Soil also may
contain a wide variety of organisms as shown in Table 44
(Mitchell, 1974) . Investigations have discovered 320,000 to
500,000 CFTf bacteria per gram of sandy soil material, and
360,000 to 600,000 CFT3 bacteria per gram of loam. n soils that
are rich in organic matter, as many as 2,000,000 to 2OO,QOO,OQ
C?TJ bacteria per gram of material have been found (Wali an, 1952;
Alexander, 1971). In a complex physical and chemical soil medi-
um, organisms pursue their activities not as individual groups,
but as a population. A soil does not represent a homogeneous
biological entity, but ra ther a media where a series of complex
biological activities are occurring simultaneously.
The soil properties most useful in predicting mobility of
leachate contaminants are: (1) texture (clay content); (2) con-
tent of hydrous oxides (Fe, Mn, and Al); (3) type and content
of organic matter; (4) particle size distribution; (5) cation
exchange capacity; and (6) soil pH. Relative rctobilities of
leachate cont thants through soils are quite variable. Roweve ,
a qualitative prediction is possible if the important soil
physical and chemical properties are known (Korte, at al., 1975
Fuller, 1976; Fuller, 1978; Griffen and Shintp, 1978; cq ton,
1978). In the following discussion, specific examples are used
to illustrate the various reactions which influence leachate
mobility and attenuation.
Migration Mechanisms
Physical Mechanisms-—
The most important physical prOperties of a soil system
relative to leachate migration are diffusion and dispersion,
Colony forming unit
190
-------
‘9
Figure 43.. CSDA soil textural classification
(U S. Soil Conservation Service, 1971).’
I I
‘ S
silt loam
prcent sand
t
191
-------
TABLE 44. SOIL ORGANISMS (after Mitchell, 1974)
fiterotrophi c
1. Nitrogen—fixing
I a. Symbiotic
b. Nonsymbiotic
I 2. Bacteria that recuire
I combined nitrogen.
Bacteria— 1 Autotrophic
Bacteria that oxidize nitrogen,
I sulphur, iron, hydrogen, and
carbon compounds, and sulohur
L nd/0r hydrogen.
Protists 1
I Yeasts
Soil Fungi —j Molds
Organisms tinomycetes
[ 7 ie—green
A3.gae H Diatoms
s-green
Protozoa
Nematodes
Animals-I Rotifers
cqorms
Insects, etc.
192
-------
dilution, physical sorption, and straining.
Diffusion and dispersion effects——Molecular diffusion is
caused by the concentration gradient Of contam nants, that is,
the movement of a chemical component from a region of high con-
centration to one of low concentration. I the case of very low
leachate flow rates, diffusion of soluble species in the soil
solution may be a significant migration mechanism (Mang., et al.,
1978).
Hydrod.ynamic dispersion is the result of variations in pore
velocities within the soil. Within a given soil pore, the flow
rate is slower near the walls than in the center of the pores.
The soil water (or leachate) also flows faster in. the larger
pores than in small pores. Mydrodynamic dispersion is effective
in a tenuatiflg the maximum constituent concentration rather than
total qualtity of the constituent in a pulse or slug of leachate.
or large leachate inputs, hydrodynamic dispersion will not be an
effective mechanism (Ogata, 1970; Weston, 1978).
Dilution effects-—When leachate flows from the unsaturated
zone to the zone of saturation below the water table, dilution
o the contaminant concentrations occur. The rate of contaminant
dilution is proportional to the solution flux of both leachate
and ground water. The finer the soil texture the slower the
leaching solution tends to flow, and therefore, the slower the
rate of dilution. The leachate flow rate is also directly
proportional to the moisture content of the soil through which
i.eachate is being transported. If the region of soil between
the water table and the bottom of a landfill is unsaturated, the
vertical transport rate of the leachate from the disposal site
will be orders of magnitude smaller than that when soil is
saturated (Ham, 1975).
Physical sorption effects——Physical sorption is a function
of Vai der Waals forces, hydrodynamic. and. elect;okinetic proper-
ties of small soil particles. As mentioned by Fuller (1977),
OU1Y a small portion of the reactions of trace contaminants in
goji/water solutions can be defined as physical adsorption; how-
ever, physical adsorption is an important removal mechanism as
regards bacteria and viruses (Gilbert, 1976). The remaix er of-
the sorbed contaminants can be ascribed to chemical bonding or
chemical adsorption, as will be described later.
Straining effects-—Suspended solid particles in solutions
generi llY range in size from O.OOli.& to 8 . The removal of con-
stituents by soil filtering is applicable to a wide range of
particle sizes, and. not exclusively, to larger particles.
other straining effect by soils is the attenuation of nonpolar
organic compounds (such as o ..ls, grease, and. hydrocarbons of
lower molecular weight). Even a small depth of soil can remove
large quantity of such material (Mang, et al., 1978).
193
-------
Chemical Mechanisms -—
Five types of reactions occur which are basically chemical
in nature: (1) recipitation/diSSOlUti011i (2) adsorption/
desorption, (3) comple: ation, (4) ion exchange, and (5) reduc—
tion/oxidation (redox).
precipitation/dissolution reaction-—Precipitation/dissolu-
tion reactions can control concentration levels and limit the
total amount of contaminants in leachates when leaching through
soils. The migration trends of contaminants are usually toward
an equilibrium state, controlled by the solubilities of solids.
For example, when the soluble concentration of a contaminant in
leachate is higher than that of the equilibrium state, precipi-
tation occurs, thus the contaminant level of the resulting
leachate is attenuated.
precipitation/dissolution reactions are especially important
in trace metal migration in the soils (Garrels and Christ,. 1965;
Santillan—MedranO and Jurinak, 1975; Griffin, et al.,l976;
Leckie and James, 1976; Weber and Posselt, l97 7 iff in and
Shimp 1978). The important solubility products of coon
trace metal compounds are presented in Table 45. In general,
cadmium, copper, lead, nickel, and zinc can form carbonate
solids in relatively oxidizing leachate environments, and sul-
fide solids in relatively reducing environments where soluble
sulfide is present. Iron and manganese can form hydroxide or
oxide solids under oxidizing environments and sulfide solids
when soluble sulfide is present. The attenuation effects of
precipitation/dissolution of metals is more significant at
neutral, or higher than neutral pH values. At lower levels,
other effects (such as adsorption) will usually prevail. Since
precipitation/dissolution and sorption occur simultaneously i
the soil, it is difficult to separate the two processes.
Adsorptiotl/desOrptiofl reaction-—Sorption (adsorption or
desorpticn) is the most common mechanism generally associated
with trace contaminant migration in soils. Under oxidizing or
low pH conditions, adsorption can usually regulate the concen-
tration of a contaminant well below the level controlled by
precipitation. The high anion retention of soils that cannot be
attributed to anion exchange or precipitation can usually be
attributed to adsorption (Jenne, 1968). Because of the relative..
ly rapid rate of adsorption in comparison to precipitation
reactions, adsorption mechanisms may affect the concentration of
a constituent even if that constituent has a low solubility
controlling solid.
Co-precipitation is the precipitation of an otherwise
soluble substance along with an insoluble precipitate. These
two substances may precipitate simultaneously or one may fo11 w
the other (Peters, et al., 1974). The co—precipitation of a
constituent as an i urity in solid phase accumuiations and
194
-------
TAILI $ IN1O0?MI1 SOLUIILITY PItODUCTI (K .pIOV TLL&Cg e4 T1J.a. 4S 11 n s,s4 k .rtu1L 1904 . 1*71g k1n bon 1 l1J1
$tu. and Nocqan LU ))
M. toL Oa td. Mydra 4s1 . Carliosii.I. Su ltIde j rj Iha ph t blItc .tu
CdflI) 83.0 13.0
C . 41 1 ) 5.20 8.32 0.25 (C .UP0 4 ) 3.7
(cm1o1te)
8.22 20 Ci 3 (VO 4 2 )
4ar9?n1t ) . 1.14 (C .U 2 (P0 4 1 3 )
“.7
(dolo. lt.8 04 (CMIPO 4 flI 3 O ) 2 ) 585 cA-&3at
aorLl loulLu8
40.9 4Ca 4 U(P0 4 ) 3 ,
44.0 ICa 0U4PO 4 I )
120.8 (Ca 10 (V0 4 ) 1 V $
Cu(138 20.4 I I I 9.13 35.2 37.1
33.2
Cu 2 co 3 lOI) )
V . 4 1 )) 15.3 10.2 1 1. l4VuS ) 33.3 10.91
V.4118) 10 39.3 10.2 25.4
(f. 2 0 3 ) $V. 3 5 4 )
Pb($1 15.4 14.1 13.1 24.4 4.79 43.5
18.0 DIa1IPO 4 )
1.2 l.ct*v.. 49 4.ia jnu*1ta8, 444k 30.4
I I I 4bruclt.I 5.4 4ise.quehon1t b. 4$ 1 C1 3 82.8 I 4 4PO 4 fl
81.7 1H CatCO ) 3 (1 20I4). 83.2tpI 9 NN 4 PO 4 lu 2 O )o
4
5.0k ftIjUPO 4 4M 3 O ) 3
013083 1
4Co nt uod)
-------
TAbL& I IICouthu H
O Id ________ C bon t ChIor .% HIOML1hIt
bhi(ii) 13.7 9.3
iS.) Ipruolpi—
119(11 1 75 • 7 t 35.4 52.2 (HuLa— 13.4
clnnabar l
53.4 4olunabsr l
KIU —4.11 Ut
(ohoc l a aul
124
( Uaicov1 101
l -’
‘0 40.4
0% (al% 1te l
294
4N a— iiLi ot r—
I itcnltu l
15.7 (Amor housI. 10.4 25.2 4.pksl— U.?
14 gAorphOUS. a rttu).
22.0
14.5 (cryst. (wuriltul.
aged) 22.1 (pruclp*
it at ad)
vaiuas pK.p (log I when I (ionic tus pura1ure) — 0 t 4tuapursLuruI —
11aAi81 3 0 4 (.. aibitul • 71120 t — *13* t 11a * 31145(04 $ 301(1
CuA1 2 S4 2 0 5 (s, anorthitu) • •II 0 — 2A1 * 2II 4 8iO • II0II t
1 1/2 z118g 3 o 4 (.. orthoolseul • 1 1/2 *1 t 1 1/2 54 + 4 1/2 1148104 $
(CoSt tuuud)
-------
1 (Conc%udu4)
KA1 3 $1 3 0 10 d04 1) 3 4a, •uacovitu + 101120 — Mi 3 $ k 3U 0L0 • 1O0U
3 11 $0 33 AI 233 0I 307 0 10 019 2 dii 1 Na-.ont. orI11ontLe) $ 101120 — 7A1 * 111140104 * 22011 * 1141 ,
3Ca 033 Al 44 BI 333 O 30 l0lll 4 d.. C4—.oi L oc1*1onlLaI * Ull2O — L4*1 • 221148404 • 440U* C* 2 !i
CaMç CO 3 ) 2 4o1osiItii — Cal* * H j + 2CO ,
* 31120 — ZVe 3 ’ *
r*810 3 (a) $ 1140 • VS t2011 •
3 4O 3 C0$ 2 4s) — 3p4 2G $ aoii t
rbO .) $ U 0 — PII t 20UJ
PI 1 1P0 4 d 1 — PI • NPo ,
w
—1 uis *P0 4 1 ( ) — • *
HytW 4 PO 4 HU 2 0l 8 l$) * PO +
M jIIPO 4 (U 4 0l 3 .) — 119 $ UPO • 311401
N jL1 3 Il 2 0I 1*. • *
811gC1 3 10 3 0) 3 ($. c$rnall*I.) • * 3C1 t 31130 1
• 2U — 2e • 113 I t 11401
11n310 3 4.) * 114$ — 11a 2 1 $ 20 1 1 *
1190 lii * 1120 — 1192* • 201C,
I 0&O3 a) * 1120 — $ 20 1( • II0 MI
-------
crystal structures is common in nature. Since it is extremely
difficult to separate the effects of co—precipitation and adsor —
tion, both are usually included as adsorption. Some of the nos
important adsorption mechanisms in soils are those associated
with iron and manganese hydrous oxides (Jenne, 1968; Gadde and
t. aitinen, 1974; i orte, et al., 1975; Shuman, 1976; jnnjburgh
et al., 1976; Fuller, 1976; Shuinan, 1977; Fuller, 1977; Ful1e ,
T 72). Jenne (1968) has suggested that hydrous oxides are the
main controlling factor in the environmental fixation of trace
metals. This theory is supported by findings that manganese
and iron oxide contents are highly correlated with the quantity
of trace metals in soils (Taylor and McKenzie, 1966; Korte,
1976)
Clay minerals, hydrated a1 inum oxides and soil-organic
solids may also adsorb constituents from the solution phase of
soil/water system (Korte, et al., 1975; Fuller, 1976; Griffin,
et al., 1976; Kinnibu.rgh, et al., 1976; F l1er, 1977; Roulier,
1977; Fuller, 1978; Griffin ar Shimp, 1978; McSride, 1978)
As in the case of precipitation reactions, sorption processes
are strongly dependent on pE. Because of pE-dependent charge
characteristics, soils can exhibit either cation or anion
exchange characteristics, (Jenne, 1968; Frost and Griffin,
1977: Shuman, 1977; Griffin and Shim;, 1978; Stumm and Mor an,
1981)
Com ;lexationC omplexation involves the reaction of metal
ions ith inorganic anions and organic 1i ands (chelation) to
form complex inorganic ions and organorttetallic complexes (Stui
and Morgan, 1981). Complex formation may affect attenuation and
migration of constituents in two ways. In the solution phase,
it can greatly increase the concentration of constituents by
the formation of soluble complex ions. On the other hand, if
the complex formation exists between the soluble constituents
and solid surfaces, especially organic chelates, then the
levels of constituents in the resulting leachates can be
decreased (Schnitzer, 1971: Stevenson, 1976; Mattigod and
Sposito, 1977; Doner, 1978; Mang, et al., 1978). The concentra
tions of complex species in leachate are not necessarily the
result of equilibrium processes. Species which are thermo—
dynamicallY unstable but which, due to kinetic constraints,
exist in significant amounts are called nonlabile or inert
(Siegel, 1971). Because of these nonequilibrium processes,
extrapolation of laboratory determined thermodynamic stabilities
to natural conditions requires considerable caution (Ruben,
1974). Although the “apparent” stability of metal ions and
natural organics have been investigated in a few cases, the
extent of the interaction of these organic compounds with metal
ions in the real environment is not known (Malcolm, 1970;
Sch.nitzer, 1971). Since both soil and municipal refuse contaj
a wide variety of organic compounds and complexes, the net
attenuation or migration of complex species, is very difficult,
198
-------
j not impossible to evaluate with any degree of certainty.
Ion exchanqe——Most of the ion exchange effects in soil/water
yste originate from the exchange sites on layered silicate
clays and organic matter. In general, the layered aluminum-
silicate secondary minerals in soils hold a permanent negative
charge. Murr ann and Koutz (1972) pointed out that the cation
exchange property arises from the need to balance the negative
charge of clay micelles to maintain neutrality. To accomplish
this, positive ions in the soil solution become associated with
the negative charge on the exchange complex. These ions are
mobile and readily exchange with other cations in the soil solu-
tion to maintain chemical equilibrium. This process represents
cation exchange. The total capacity of soils to exchange cations
is called the cation exchange capacity (CEC); the CEC of any
particUlar soils is affected by the kind and quantity o clay
mineral and organic matter and the pH of the soil/water solution.
For the three predominant types of clay minerals, CEC values can
be ranked in the following order (Wilson, 1980): ntontznorjllonjte
, illite > kaolinite.
Soil textures greatly affect CEC values. Generally, soils
0 cntainiflg smaller particles have higher CEC values because of
the large specific surface area and correspondingly available
xchaflge sites. The organic matter fraction of a soil may con-
tribute substantially to the CEC by contributing a negative
charge to the soil exchange complex. Additions of organic
matter through sludge apulicatiort may improve the CEC of soils
with naturally low CEC (3roadbent, 1973). The cation exchange
capacity increases with increasing soil p . This effect is due
to an increased negative charge on the soil exchange complex
with rising pH as a result of increasing ionization of soil acid
gr ou pS and decreasing proton addition to the basic groups. The
0 aflge goes in the opposite direction if the pH decreases. In
mos t cases, the negative charge greatly exceeds the positive
barqe , resulting in a net negative charge ($t and Morgan,l981).
Major cations in landfill leachates (i.e., calcium, magne—
5i.um, ootassiuxtt, and sodium) are usually present in higher
concentrations than any of the trace metals. Consequently,
trace metals cannot suc essful1y compete for the cation exchange
5 ites that are dominated by the major cations (Wiklander, 1974;
Griffin’ et ! • , 1976; Fuller, 1977; Garcia.-Miragaya, and Page,
1977)• Ara result, the removal of trace metals from soil-water
interfaces by cation exchange is usually insignificant when
to other mechanisms (Murrmann and Koutz, 1972).
In nature, sorption, solid-liquid comolexation, and ion
chaflge processes are difficult to distinguish from one another.
j.though ion exchange mechanisms can be distinguised by the ini-
tial and subsequent cation/anion concentrations, these three
199
-------
mechanisms are usually grouped together as if they were one
mechanism.
Redox reaction——Redox (reduction/oxidation) reactions of
contaminants will occur when conditions of redox potential in
leachates are different than that of the soil solutions. A
discussed oreviously, redox reactions are important in the
migration of contaminants since they can affect the soluble
complexes and solid transformations, subsequently affecting the
solubilities of contaminants. In general, for iron and rnanga—
nese, reduced forms are more soluble than oxidized forms. For
other trace metals, however, a reducing environment is more
favorable for the attenuation of metals if sufficient sulfide
is present for the precipitation of metallic sulfides which are
generally very insoluble.
Microbiological Mechanisms--
Microbiological reactions affecting contaminant migration
are numerous. The following major effects can be attributed to
microbial activities (i.e., biodegradation): redox effects,
mineralization, L, obilizatiOn, precipitation/dissolution, and
complexation.
Redox reactions are greatly affected by the biodearadatjon
of organic compounds in soil. The mode of degradation not only
changes with the organism specie, but it also may affect the
redox otential and thus the oxidation state and chemical forms
of all constituents in both solution and solid phases of the
soil system.
Through mineralization, plant nutrients, organic chemicals,
microbial tissues, and organic—inorganic complexes may be con-
verted into inorganic states. Through biological assimilation,
the inorganic nutrients, and trace metals may be transformed
into microbial tissue, thus biologically immobilizing these
constituents.
Organic complexes which accumulate in soils as a result of
both microbial synthesis and degradation have a high capacity
to combine strongly with trace metals and other constituents.
Through these reactions, the constituents in soil systems can
be mobilized, com lexed, precipitated, or sorbed.
Migration Trends of Contaminants
General Parameters
The migration trends of general parameters in the soil!
water environments to be discussed are pH, redox potential (Eh),
TOC, and alkalinity.
t H——pH levels in leachates are usually acidic (refer to
Section 4, Leachate Composition). The result is due to the
200
-------
biodegradation reactions as illustrated by the following two
eneri zed ecuations:
‘C 0’ (N ) Q ( ) . - 53 302— - ,— -
- 2 ‘106 3 16. 3 4 4 bacter .a
106 HCO 3 + 53 HS + 16 NE + HP0 4 39 , r
(CE ° ‘06 ) 16 H 3 P0 4 (s) + 106 02 Aerobic
2 - 3 bacteria
106EC0 3 +16NE +HP0 +92E 4 .
Low leachate pE levels can usuai.ly be increased when leachate
oercclates through underlying soils. This is acomplished by the
dissolution of solids in the soils which contain a buffering
capacitY such as carbonate, oxide, or sulfide solids. Follow-
ing are two typical examples:
CaCO 3 Cs) + Ca 2 + HCO 3
Sn Cs) 4 rtE 2zi 4 m+ + nHS
m
(Where M 2 S represents metallic sulfides)
In general the calcium carbonate (CaCO3) content in soils is the
most jm ortant factor responsible for the leachate pH increase.
Redox otential--The redox condition of leachates from act-
jve b degradation landfills is usually reducing. Soluble
sulfides produced in the landfill are the main species which
control the redox level. Numerous redox reactions can be ex ec—
ted during leachate migration through underlying soils. Table 46
lists some important related theoretical redox reactions in the
5 oillwater syste.’n. For example, if the pM and redox potential
in the soil/water environment is 7 and -200 mV, respectively,
th from Table 46, E,S can begin to undergo transformation to
so because the Eh 8f the H 2 3-S0 4 redox couple is - 210 my.
j ice this Eh value i more negative than the soil/water Eli
(—200 WV), there will be a strong tendency for H 2 5 to be oxidized
so 4 ’ . If the original iron solid in the soil is FeCO 3 (g),
then transformation to Fe (OH) (s) also can be seen. (Eli FeCO 3 -Fe
(OH) couple - 385 mV), which tends to reduce the soluble Fe
levels. T3nder these soil/water conditions (pH — 7, Eli -200 rn’ )),
Mfl02(s) could be gradually converted to MnS(s) which would
5 sequently increase the soluble Mn levels. (Eh Mn02 (s) -
4flOOH(S) couple = + 670 mV; Eli MnOOH(s) — MnS(s) couple + 407
rn ’ !). Other redox reactions and their effects can be qualita-
tively predicted from Table 46. Redox couples not included in
Table 46, can be calculated from the thermodynamic data such as
iisted by Latimer (1952), Sillen and Martell (1964), and Sillen
and Martell (1971).
201
-------
1*011 3 •IIIOUTICAL OfilOl (1414041108$ 106 61008 (1M1 ILS (SIlica. *nd NartclI. 4984. 497% . k$a. 3 ura 1113)
*164
.83%
.130
.16 0
*680
.610
.500
‘4 , 49
.521
‘444
.430
.146
.140
-III
-166
- ISO
- lO S
- I 86
-203
•2 01
-266
.7 ”
.768
$610
$610
.670
$ 5544
.440
.429
.401
.36 1
.349
—34
-40
-198
-210
-225
-246
-241
-28 %
-321
-385
.164
.68 1
• 10
*460
.490
.430
* 3410
.3 1 .9
.287
1294
$268
-154
-160
-210
-255
-292
-3447
-300
-359
-441
-504
I.doz C.UpIS 168C1 1U11 — 4I4 1 4 -1
—
im —N 7 (,)
16.0481 (8)
- Na.103(i )
— MC0 3 ($)
1 5 ,0 0 1 1 ( i ) — 41ut0 3 (i )
o oosi(s ) —14u . 0,(s )
NnOtE( i) —44 .4(i)
1 .00 8( s) —los(s)
1 .0011(i) — 1cC0 3 (s)
S0 • —Se(s)
s0
— ias
S°(s) — N S(o)
£0719) — cii,(a)
117(9) — M 1
1 * 10 1 1) 3( 5) — 1. 5( i)
1 0( 0 1 4)3( 1) — 1 0 (03( 5)
•
3 ,5 110 .6/59’ es —1/10 N (i)’ . s/ I N 0
46407 I i) • 14 * s • 16.008 (s)
4/2 41* 1304( 1) • 3/2 co;oo21 $ 5/2 ii ’ ‘ S — 3/2 9 14(03(8) • 21420
1/2 44n0 2 ($) • 1 ,2 80 (Io ) e 3/2 N’ $ • • 1/2 I84(0 1 (s) $ 3/0 4120
14 .00 1 1( 1) • 8C0 (I0 ) • 21$’ • • • N , .C0 3 (s ) • 24170
3 4 4 .0 0 1 1(s) • I I ’ • • — 81.30411) * 21470
l i z 1 10 • 1 1 •s 1/2 140 • 1/21120
s.oo$l(i) • 85 (3 .1o ) • . • — 164 5(s) $ 24420
I/ O • 6/4 I4 • • • • 1/8 *I ‘ 120
I ,4 11ç.4 1141... 1 18rn. 1/34170
1 .0(91(8) • IIS13a10’) • 244’ $ . • 1.6(s) • 24470
1 .00441*) • t0flo ) • 24 1 • a. — 1ct0 3 ($) • 24170
UO S0 • 4/3 14 • a — 1/6 1°(s ) • 2/3 4120
118 S0 .6/414’ * a. • I/I t 1 7 $(9) $ if? 1170
I/U S0 $ 9/U 11 $ a. • 1/8 II$ • 4/? 1120
I!? O( ) • I I ’ • • • 4/2
1/8 C0 2 (pj • 4 1 • a. • I/ I l ( 1 14(9) $ 1/4 14.0
4/4 N ( ) • 4/3 1 1 . a • 4/2 l9I
1*1044)3(5) • Il5(3i 10’) * 2 1 1 4 • lo s(s) • 34170
1 5(041)3(5) • 11(0; ( 1U 2 ) • ?U $ . — 1i C0 3 (s) • 31120
Reoctjoa.s or. r.v.riIbi.
-------
TOC, BOD, and COD--The major mechanisms for the migration
of organic matter in the soi1/ ater system are bioconversion,
surface sorption, and metal-organic complexation (Mang, et a l.,
1978). Bioconversion can either decrease or increase TOC levels
in leachates through microbial synthesis and microbial degrada-
tion for both soluble and particulate organics. Organics can be
released or assimilated through surface sorption by soil parti-
cles, especially clay minerals, hydrous oxides, and non—
crystalline materials (Greenland, 1971; Fuller, 1978). Chelation
may occur in the soil/water system by specific chelates such as
microbial slimes, g .m1s, cell debris, humus, lignin polymers,
polysaccharides, proteins, and substances with low molecular
weights (aliphatic acids, amino acids, organic phosphates, and
volatile acid complexes) (Mang, et al., 1978).
Extensive data on the fate of organic matter, primarily in
terms of TOC, 300, or COD in soil/water systems, indicate that
organic matter can be substantially attenuated or unaffected
e ending on the site specific conditions. Griffin and Shimp
(1978) and Fuller (1980), demonstrated that COD and TOC were
not significantly retained by any soils tested. Griffin (1978)
concluded that in a fresh, young leachate, COD would most
likely present a high pollution hazard. Fuller (1978) reported
that the COD in soil column effluents was initially 30 to 50
times higher than the COD in the applied leachates. Fuller
(1978) concluded that soil organic matter initially contributed
to the high COD of the soil solution which was subsequently
eluted by the applied leachate. The TOC in soil column
effluents rapidly increased to the level of TOC in the applied
leachate; when the soil columns were subsequently leached with
water, essentially all of the applied TOC was eluted front the
soil. Engers (1978) found that in down-gradient ground water,
the concentration of organic matter in leachate was reduced
gjgnificantlY as a result of filtration and microbiological
conversion. In column studies using a recycled simulated land-
fill leachate, Soyupak, et al., (1978) reported that the
jnfluent COD and TOC was re ced by 90 and 94 percent, respect-
ively, over a 41-cm soil column after 70 days of operation.
The results obtained in this study can be approximated by a
developed model based on biodegradation phenomena. From limited
avadil le data (Griffin, etãl., 19? ; Fuller, 1978) th4t
icate that the sorptio o 1eachate organic matter by soil
is not extensive, Soyupak, et . (1978) concluded that the
microbial decomposition of organic matter in leachate and soil
appears to be a significant attenuating mechanism.
Alkalinity--In general, alkalinity in municipal leachates
j due primarily to soluble carbonate species. Dissolved sili-
cates, borates, aumtonia, organic bases, sulfides, and phosthates
cafl also contribute to the alkalinity. These noncarbonate con-
centrations are usually small in comparison to the carbonate
species. In the soil/water environment, it is suggested that
203
-------
the alkalinity of leachates is controlled by the dissolution!
precipitation of metallic carbonate solids in the soil, such as
calcite (CaCO 3 ).
Major Ions——
Concentrations of major cations and anions such as sodium,
potassium, calcium, magnesium, chloride,.and sulfate are high j
the soil/water system because (Mang, ai ., 1978):
• The solubilities of simple solids of major ions
are relatively high;
• The solubilitiQs of clay minerals may be low enough
to reduce the concentrations of major ions. Eowever,
the nucleation of clay minerals is extremely slow; and.
• No other mechanisms cart reduce the soluble levels
significantly.
Possible mechanisms for regulating the levels of major ions
are solubilization and ion—exchange effects. Solubilization is
significant only for calcium and magnesium. Calcite, aragonite,
dolomite, and brucite regulate concentrations through solubizi—
zation. Other complex solids, due to the kinetic constraint,
will not play important roles in precipitation/dissolution
reactions (Carrol, 1959; Griffin, at al., 1976; G if fin and
Shimp, 1978; Chart, et al. , 1979).
Ion exchange is one of the important mechanisms for control-
ling the migration of major ions. sual1y occurring on the
surface of clay minerals and colloidal organics, the capacity
of ion exchange is dependent upon the c jstalline structure of
the mineral and the chemical composition of the leaching fluids
in contact with the mineral. Ion exchange in these minerals is
a reverse chemical reaction which follows the law of mass action
and is restricted by the number of exchange sites on the mineral
and the strength of the bonding of the exchangeable ions to the
mineral surface. The ease of replacement of common soil cations
is (Rose, 1966) :
Li> Na> K> Mg> Ca> Ba> Al
In two municipal soil/leachate interaction studies, Chloride
and sodium were found to be relatively unattenuated while otas...
siwn and calcium were moderately attenuated (Griffin, al., 1976.
Rou.lier, 1977). Dilution caxi be of greater importance than
solubilizatiOn and ion exchange for the control of the major
ions, especially for sodium, potassium, and chloride ions.
Therefore, the porosity of the soil and the flow rate of the
leaching fluid may become significant factors in regulating the
migration of these ions (Nang, et al., 1978).
204
-------
Nutrieflts
Nitrogen ccnroounds-—In most natural water systems nitrogen
compounds include or an c-nitrogen, monia—nitrogen, nitrite—
jtrogen, and nitrate-nitrogen. The transformation of these
jtrcgen compounds is greatly affected by biologically mediated
reactions which, in turn, are controlled by the types and popu-
lation of microorganisms, pH, redox conditions, and dissolved
oxygen concentrations as well as the concentrations of the com-
pounds themselves. Other mechanisms contributing to the
transport of nitrogen compounds are diffusion, sorption, corn-
plexatiofli and hydrolytic reactions (Mang, et al., 1978).
All reactions carried out by living organisms indirectly
require nitrogen. t’Titrogen fixation* is a process performed
a few bacteria and blue—green algae. tf the soil/leachate
5 vstem is aerobic, nitrification (mineralization) of organic
jtrogen sources occurs readily, producing N0 3 -N as an end
product. Nitrate—nitrogen is highly mobile, moving readily
with the soil solution into the lower vadose zone and ultimately
into the ground water (Wilson, 1980). Under anaerobic condi-
tions, nitrification is inhibited and the NH4-N form predominates
(alzonificati0 ) . The incorporation of N0 3 —N or N1 -N into
jolOgiCal tissue is called imtmobjljzatiort or assimilation.
Under anaerobic conditions, nitrogen may be returned to the
atmosphere through denit.rification. Dentrification occurs quite
readily as a result of the activity of heterotrophic bacteria,
which convert N0 3 -N to gaseous N 2 or NO 2 .
Organic matter in leachates can play an important role in.
the reduction and fixation of nitrogen compounds. Most of the
nitrogen in municipal refuse is present as organic nitrogen
whjCh is only slowly released by the activity of the bictic
0 UUfljty. Thus, the rate of biodegradation of organic matter
usually influences the nitrogen fli.uc. Many species of bacteria
and fungi are capable of reducing organic nitrogen compounds,
nitrate, nitrite, and alluflonia; others can oxidize nitrogen
ompounds. All these reactions are enzyme-catalyzed and highly
pH dependent.
Due to its cationic nature, alonia—nitrogen (chiefly in
for form of aoni ) may be adsorbed onto solid materials by
jon exchange reactions with mineral or organic matter in the
5 jl material. Adsorption is considered one of the most impor-
tant mechanisms for controlling movement of axmnortia—nitrogen
ough soils and, depending upon the soil enviroz ment, might
even override biological action (Preul and Schroepfer, 1968).
riie nitrate anion is relatively mobile and not retained by ion
chaflge processes.
* Conversion of atmospheric nitrogen to soluble nitrate ion.
205
-------
Under many circumstances, nitrogen ccm ounds can be corn—
plexed with soluble metal ions (Sillen and 1arte1l, 1964; Sillen
ar.d Marteil, 1971) . There is little information on organic
nitrogen species and the thermodynamics of complex formation.
Ecwever, from data on the complexation of trace metals with
humic su.bstances and amino acids (Lu, 1976) , it can be surmised
that organic nitrogen can also complex with trace metals to a
great extent, depending on the concentration of the metals.
This complexation effect can increase nitrogen levels in leach—
ates. The importance of the resulting nitrogen complexes
depends upon the types and levels of metals and the formation
constants, activity coefficients, and levels of the leaching
solutions in contact with the soil.
Phosphorus comoounds The transport of phosphorus compou ds
in the soil/water system involves complicated physical, chemi-
cal, and biochemical interactions. The most important
mechanisms governing the phosphorus migration in the soil/water
system are solu.bilizatiofl, sorption, and biological effects
(Mang, et al., 1973)
To understand the importance of solubilization effects,
phosphate solids possible in municipal leachates have to be
assessed. The major hosphate solids which might control the
solu.bilities of phosphate in leachates are shown in Table 47.
Among these solids, Ca_OH(P ) (S) and A 1 POA(s) are most Likely
to control the soluble PhOSPh .teS in leachätes. The equilibrj
solubilities of these two solids can be seen in Figure 44.
The minimum solubility of A1PO 4 (s) is about 0.01 ppm (as P) at
pH 6. At pH 5 it increases to 0.02 ppm (as P). At a pH great
than about 6.8, the solubility controlling solid for phosphate
will shift from A1PO (s) to CaçOff(P0 4 ) 1 (S). At pH 7 and an
alkalinity of 1000 (as CaCO (s)), the solubility of the
phosphate specie is about 0.35 pm (as P). From this relation-
ship; it is likely that phosphate solu.bility is controlled by
A1PO 4 (s) when leachate pH is less than neut al. When leachate
is recirculated or buffered to 7, Ca OH(PO ) (s) may become
the predominant solid species to gov rn th oluble phosphate
levels.
Although the free soluble phosphate ion is limited by less
soluble solids(s), the complexation ions can enhance the phos-
phate solubility. For orthophosPhate, the most important
contplexatiOri ions in leachates are magnesium and calcium, which
will form CaHPO 4 (aq) and MgHPO 4 (aq) complexes (calculated by
Calscieflce Research, Inc.).
Phosphate adsorbentS also can regulate the soluble levels
of phosphate (Fox and Kamprath, 1970; Singer, 1972; Chen, et
1973j Patrick and Khalid, 1974; .St .inm and Morgan, 1981).
Hydrated iron oxide is a strong adsorbent for Dhosphate, how-
ever, it seems of little significance in leachate reduc .ng
enviro efltS because of the release of phosphate caused by the
206
-------
reduction of ferric phosphate solid to the tnore soluble ferrous
thosphate. In general, high phosphate adsorption by clays is
favored by .a lower pH. Maximum sorption of orthophosphates on
rnontmorilloflite occurs at pE 5—6. Sorption by kaolinite is
maximum at a pH near 3 (Sti.m and Morgan, 1981). It is specu-
lated that phosphate clay adsorbents will not play a i. iportant
role in landfills when pH levels are close to neutral.
TABLE 47. MAJOR PHOSPEATECONTROLLING SOLIDS (Strun m and
Morgan, 198].)
Equilibri1 . Log Equilibrium Constant
Ca 5 OH(P0 4 ) 3 (S) 5Ca 2 + 3P0 4 3 + Of -55.6
Ca 5 OH(PO 4 ) 3 ( ) + 3H 2 0 =
2(Ca 2 HPO 4 (0E 2 ) 2 ] surface 4 Ca 2
+ RPO 4 Z -8.5
CCa 2 HPO 4 (OH) 2 surface 2Ca 2 + HPO 4 + 20H -27
CaEPO 4 (5) Ca 2 + O4
FePO 4 (S) = Fe 3 + PO 4 3 -23
A1PO 4 (5) = A1 3 + P0 4 —21
Ca 2 P 2 O 7 (S) Ca 24 + P 2 0 7 2 —7.9
In view of the above, the soluble phosphate levels in the
neutral leachates are most likely controlled by Ca 0H(PO 4 ) 1 (s)
and regulated by complexation ions. Because of this pheno enon,
the calci .mt ion concentration or alkalinity level in leachates
will, be the dominant factor affecting the soluble phosphate
levels. The concentrations of the major soluble phosphate
(mainlY orthophosPhate) species were calculated and are shown
n FigUr 45.
As can be seen from FigUre45, thetotal soluble hos hate
levels in leachates may range front 0.30 to 6.5 ppm (as?) when
rances from 5.5 to 7.0 and when alkalinity ranges from 1000
to 6000 p tn (as CaCO 3 ). High alkalinity levels favor the
207
-------
2
0
3
4
5
6
7
8
Fig .re 44.
100
10
1
pif
Solu.bilities of major phosphate solids (after
Str1 mm and Morgan, 1981).
0.01
0.001
0.].
5 6 7
208
-------
6
5
4
3
E
2
1
0
5.0 5.5 6.0 6.5 7.0 7.5
pa
g/l (as CaCQ 3 )
j ire 45. Total soluble phosphate levels in leachates as affect-
ed.by pH and alkalinity.
209
-------
solu.bilization of phosthate in leachates. At the same alka—
unity, pH values between 6.0 to 6.5 appear to be least solu.bie
for hosphate , and when the pH increases or decreases from this
range, the soluble phosphate levels exhibit an increasing trend,
(see Fiqure 45)
Trace Metals-—
The movement of trace metals in the soil/water system
is extremely complex. Not only are there numerous controlling
factors, but many unknowns beyond present state-of-the—art
knowledge are involved. Thermodynamic and kinetic influences
make transport phenomena difficult to explain and predict. sox
of the major mechanisms that influence mobility of trace metals
in solid—liquid interfaces are:
• Solubilization;
• Sorption;
• Complexation; and
• Dilution.
Each of these mechanisms may exert some influence on the migra-.
tion of trace metals in soil/water systems.
The importance of solubilization effect for trace metals
has been discussed previously. Griffin, et a!., (1976) suggest...
ed that migration of Pb, Zn, Cd, and Hg in the soil/water
system could be explained by the solubilization effect. it ais
has been suggested by Mang, et a!. , (1978) and Lu (1981) that the
concentrations for most trace metals in leachates could gradua
approach the equilibrium solubility levels of predominant so1jd, 5
after long-term contact of leachate and media. Although the
solu.bility effect is an important factor for governing trace
metal migration, it is usually suggested that this mechanism j
significant only in alkaline or reducing environments (Lu. 1976;
Manq, et al., 1978; Duv l, et al. 1979; Chén, et al.; 19ao Lu
1981)
In a more acidic environment, adsorption mechanisms were
suggested to be more important for regulating the levels of
trace metals in leachates. Content of hydrous oxides (primarj,j
iron, manganese and aluminum hydrous oxides), clay minerals, p
soil and leachate organic species, and specific competing
ions (especially major cations) were concluded o be the most
significant factors affecting trace metal adsorption (Gadde an
Laitinen, 1974; Leckie and James, 1974; Korte, et al., 1975;
Griffin, et al., 1976; Fuller, 1977; Rôulier, lW77TFuller, 1980.
Griffin a SHimp, 1978). Due to a wide variety of soils and
leachates, development of a quantitative generic model to
the adsorption of trace metals in the soil/water system is
210
-------
still beyond our present knowledge.
Metallic Complex formation is another important factor con-
trolling the total concentrations of trace metals in the soil!
water system (Schnitzer, 1971; Leckie and James, 1974; Weber and
possett, 1974; Lu, 1976; Staverison, 1976; Mang, 1978; St m
and Morga 1 , 1981). In a system containing only inorganic see-
cies, the quantification of the complexation effect could usually
be estimated successfully (Lu, 1980). Because of the many un-
known organic species and lack of knowledge concerning thermo-
dynamic data in the soil/water system, a precise evaluation of
all the metallic complexes is virtually impossible.
The difference in the rates of the aforementioned reactions
and the dilution effect (controlled by the solution flux of both
leachate and ground water) also affects the transport of trace
metals in the soil/water s7stem. The leachate flow rate is
proportional to the moisture content of the soil through which
leachate is being transported. Dilution by ground water may
be important in some cases; however, the availability of ground
water for dilution may be limited by the hydrogeography of the
site. The presence of clay or sand lenses, rock fissures, etc.,
can limit drastically the water available for dilution, and can
transport leachates long distances without allowing dilution to
take place (Ham, 1975).
Chlorinated Hydrocarbons-—
It is difficult to predict the fate of chlorinated hydro-
carbons in the environment. In general, volatilization,
microbial degradation, chemical hydrolysis, oxidation, and
sorption can be involved (Boucher and Lee, 1972; Nisbet and
Sarofin, 1972; Davison, 19:78; Farmer, 19787 Griffin, 1978;
GriffJ fl and Chou, 1980). These studies suggest that sorption
may be the dominant mechanism in soil/water systems.
Previous studies have shown that chlorinated hydrocarbons
tend to be strongly sorbed by soils. The major adsorbents are
clay minerals, iron and manganese hydrated oxides, and organic
material (Davison, 1978; Griffin, 1978; Griffin and Chou, 1980).
The rate of adsorption of chlorinated hydrocarbons studies was
f nd to be rapid, with equilibrium conditions achieved within a
few hours (Griffin, 1978). The adsorption process conformed to
the Freufldlich adsorption equation (Davison, 1978; Griffin,
1978). In general, chlorinated hydrocarbons are highly resis-
tant to aqueous phase mobility through earth materials, because
of their low water solubilities and strong sorption by soils;
however, they are highly mobile in organic solvents (McCarty
and Rittman, 1981).
pesticide5
Factors influencing the fate of pesticides in soil/water
systems include: (1) adsorption/desorption, (2) microbial
211
-------
decoin osition, (3) volatilization, (4) sci]. moisture, and (3)
hysical properties of the soil (Eaily and White, 1970; Leonard,
et a].., 1976). Adsorpticn/descrptiofl is considered to be the
prime factor governing the interactions between pesticides and
soil co1loid .
Organic pesticides were classified as ionic or noniorijc by
Leonard,et a].. (1976). Ionic pesticides are subclassified as
cationic (paraquat, disquat), basic (S—triazones), and acidic
(benzoic acids, phenolds, picolinic acid). Nonionic estjcjdes
include chlorinated, hydrocarbons and organophosphates. Cationjc
pesticides are strongly retained on the exchange complex.
Changes in soil p have a complex effect on pesticides forms.
For example, Laonard, ., 1976, reported that a decrease in
pH increases the molecular form of an acidic pesticides but
increases the conjugate acid form of a basic pesticides. These
changes will modify adsorption/desorption characteristics and
the mobility of pesticides. Leonard,et a].. (1976) also reviewed
the effects of clay content and soil moisture on adsorption. it
was reported: (1) adsorption is enhanced with increased organic
content, and (2) a decrease in soil moisture may increase sur—
f ace acidity, and, thus, increase adsorption.
Microbial decomposition of pesticides depends on such
factors as microbial population, soil moisture and. temperature,
organic matter content, pH, redox potential, pesticide con-
centration, availability for degradation, and nutrient ConCSfl—
tration and availability. Leonard,et a].. (1976) presents a
detailed discussion of these factors and their effects on
microbial decomposition of pesticides.
Viruses-—
Chang and Page (1979) cautioned that i.?nmobilizaton of
viruses by soil should not be equated with virus inactivation
because many adsorbed virus particles have been demonstrated
to be infectious for significant periods of time. Viruses
immobilized by soil adsorption may also become desorbed when
the chemical composition or pH of the percolating wastewater is
changed.
Recent studies by Keawick and Gerba (1980) indicate that
virus survival within a soil relate to soil moisture content,
temperature, pM, nutrient availability, and antagonisms.
evaluation of these factors indicates that because of the
detrimental effects of aerobic soil microorganisms, evaporation
and higher temperatures inactivation of viruses will be much
more rapid near the surface than for those viruses that pene-
trate the soil more deeply. Thus, viruses which penetrate the
soil surface are expected to survive for prolonged periods of
time, as compared to those retained near the soil surface.
212
-------
Laboratory studies have shown that viruses have widely
different adsorption properties (Wellings, 1974; Green and
Oliver, 1975). This is, therefore, an oortant factor i the
movement of viruses from sources of contamination into otherwise
clean ground water. Thus, for evaluation of a site, several
ty es of viruses should be tested in order to obtain a represen-
tative picture ( eswick and Gerber, 1980).
The studies by Wellings (1974, 1975) were among the first
to demonstrate transport of viruses for long distances through
soils into ground water. Specific factors affecting the migra-
tion of viruses in soils are s minarized in Table 48.
48. FACTORS THAT INFLUENCE T MOVEMENT OF VIR JSES IN
SOILS (after Chang and Page, 1979; eswici and Gerba,
1980; Wilson, 1980)
Factor Remarks
Soil Composition Clay soils are good adsorbers of viruses
under appropriate conditions. Sandy loam
soils and organic soils are also favorable
for virus removal. Soils with a low
s ecific surface area are most effective
for virus removal.
Virus removal increases with decreasing
pH; high pM results in elution of ad-
sorbed virus.
soluble Organics 5o1 1e organic matter competes with vi-
ruses for adsorption sites on the soil
particles, resulting in decreased virus
adsorption or elution of an already
adsorbed virus. Definitive information
is not available for soil systems.
cations The presence of cations usually enhances
the retention of viruses by soil.
gairif all Viruses near the soil surface may be
eluted after a heavy rainfall.
Evapotranspiration will tend to increase
the concentration of viruses at the soil
surface.
213
-------
TYPES OF LEACEATE MIGRATION MODELS
Whenever liquid or solid wastes are de ositad on land,
leachates may be generated which can seriously degrade the
quality of underlying ground-water systems. Predicting the
potential magnitude of such ground-water pollution is a complex
technological undertaking’. The simultaneous presence of nim er —
ous physical, chemical, and biological interactive mechanisms
makes it very difficult to obtain an understanding of the
pollution potential of a given waste for a specific geohydro—
logical envirorinteflt. Consequently, many designers/investigat
have resorted to the use of simulation models to obtain infor-
mation necessary for rational landfill design and to forecast
the performance of a specific waste disposal site.
A waste disposal model represents a simplified delineation
of a real waste disposal system. As a result of simplification
different types of models exist. Leachate migration models may’
be classified according to a number of schemes. One such ar-
rangement provides the four following generic model categorj
• Descriptive models;
• Physical mcdels;
• Analog models; and
• Mathematical models.
Table 49 lists a few example models and their classificatjo
into different groupings. Each of these model categories are
described in the following text.
DescriotiVe Models
A descriptive model utilizes a qualitative judgment ap-
proach to evaluate a landfill site rather than a well defined
quantified set of steps. While the cost of this approach is a
favorable consideration, the highly subjective nature of the
method severely limits its validity because different experts
may reach different conclusions based upon this modeling ap-
proach. Because of these subjective, site-specific, and
qualitative limitations, descriptive models are not further
addressed in this report.
Physical Models
A hysical model is a scaled-down version of the actual
landfill condition. Only a few physical landfill models have
been constructed in the past (Quasizn, 1965; Fungaroli and
Steiner, 1973; Pohland, 1975). Although physical models of
214
-------
TABLE 49. EXAMPLE MODELS AND ThEIR CLASSIFICATION INTO DIFFERENT GROUPS (van
Genuchten, 1978)
MODEL DEFINITION
TYPE OF MODEL
On site inspection using
engineering judgement
Descriptive
Empirical
Deterministic
Dynamic
3D*
The Drexel University experi—
mental field landfill
(Fungaroli and Steiner, 1973)
Physical
Empirical
Deterministic
Dynamic
3D
:
Batch equilibrium study;
shaker test; solid waste
evaluation leachate test
(subsystem models)
Mathematical
Empirical
Deterministic
.
Static
00
t )
Column displacement studies; Mathematical Empirical Deterministic Dynamic ii)
thin- layer chromatography
(subsystem models)
Criteria listing; matrix
method
Mathematical
Empirical
Deterministic
Static
00
One-dimensional unsaturated
transport model of Bresler
Mathematical
Conceptual
Deterministic
Dynamic
lD
(1973) (subsystem model)
Two-dimensional saturated—
unsaturated transport model
of Duguid and Reeves (1976)
Mathematical
Conceptual
Deterministic
Dynamic
2D
Model for ground-water flow
and mass transport under
uncertainty of Tang and
Plnder (1977)
Mathematical
Conceptual
.
-_____________
Stochastic
Dynamic
20
*
Indicates spatial dimension:
3D — three-dimensional, etc.
-------
waste dis osa3. sites are generally lacking, e:( erjments can
be conducted to aidinvestigators in making more accurate pre-
dictions. Such ecperimentation may include column leaching
studies, thin—layer chromatography, or batch equilibrit n
studies. unfortunatelY, this information does not, in itself,
define a waste disposal model; therefore, it c not be used
a predictive tool. owever, experimental data such as adsorp-
tion and decay constants, dispersion coefficients, etc. may
provide necessary information for development of a mathematical
model. While it is obvious that scaled-down physical models
can provide useful information, their practical ap Lication a
predictive tool appears doubtful. They are costly to build and
also may be very time-consuming to study, especially in con-
sideration of the different chemical and biological processes
within a landfill which may be operative over a period of many
years.
Analog Models
Analog models employ a convenient transformation of a g ve
physical system into another one which behaves in a similar
manner, but which is measured more easily. Electrical analog
models have found application in ground—water flow modeling
(Jorgenson, 1975). Because of the high cost of building electri .
cal analog models for large—scale field problems, it is dOubtf 1
that many such models will be used in the near future for the
simulation of ground-water quality problems.
Mathematical Models
Mathematical models employ a set of concise mathematical
equations to describe the relationships between the various
systems parameters and their input and output variables. The
series of variables usually represent known physical and cheraj—
cal properties and relationships, which can be solved using
either analytical or numerical techniques. A compilation of
different types of mathematical models is given in Table 50.
MODEL DISTINCTIONS
Several distinctions between models can be made, depending
upon the method of analysis defined by the model and the par-
ticular approach used to solve the model. Common model
distinctions are as follows:
• pirica1 or conceptual models;
• Deterministic or stochastic models;
• Static or dynamic models; and
• Spatial dimensionalitY of the model (i.e., 1, 2, or
216
-------
•0
0
a ’
C
-. ‘.
—a’
0 ;-
a-
t’)
I -
-.4
?AULB 5 0. PAkIIAL LZ T Or AVkILA L
TMAHSI’OK? I40D E .5 VON APVI.ICATZOU ) GMOUNU-WAT R QUALITY PNO&1LEH
Nodel
No.
Nodal
MUt aCUDCUS
Guoa,utry
NutLo4
Type
3)
Type
4I
Type of
CM
aut ion.
WpIicat*on/ ja aute
A.
SATURATED-UHOATUNAItD TMMISPOK?
4()OLJ.5
20, C
LV
It
I ., An
?.4. flu
Tren.port of c 4ionucl1duu
fro a waata-4 1 .po .a) alto.
Al
Duguld an4 I .ves U5Th. 1517 1
A2
Sugot I15I , 15TH
20, 30
HV
i t
I ., An
--
—-
Al
van Genucht.n *1. fl37H
20. C
UI
I C
L, Au
M l . flu
Loacliata Novonent ito ,, a
hypothetical landfill.
*4
Sykes 115151
20, C
I1V
01
I., An
--
CoiiLaa.iisanI movo &8nL fios
a landfill.
AS
Nlzy at . 115141
20, C
0
I c
--
Ad, I e
Contaminant movumont tto*
a landfill.
AS
Pdt. a *1. 4l574
20, C
FU
i t
I.
--
Groundwater pollution from
agricultural ourcue.
*7
5I*if(.r ut . 115TH
20. C
.
vu
ft
I .
Ad, Cu, Do
Zrrlgatlon Comm flow
model, coupl ed ID,
vorticol uniiatumatud mc ,dui,
and 20, .imtu rated modul.
a.
ATSD- Ly TKANS! !
3D
lIFE
it
I., An
--
lItsiulatue ricing connate
valor tlirauqh a vertical
fault in aulti-aquituc
cystum latoady Clate flow
application only)
a j. I S7 I
(Coot inuud l
-------
0
—m
(D;-
0
C’
0
1,
I - ’
1*111.1)
6 4
Modal
ho.
$ d.l
8 .fsr*ao.a
Caom.try
at
Modal ‘
M tho4 typu
of of
8o lutton Vlow
fyp.
of 4)
8011
Typu of
Clumtca l
I teractt0n4.
App lIcatlon!t))aladl&tS
1)2
Gui *4431)18 .) 81141 Cl.ary
(1971)
31)
LV )
OL
An
A4 flu
Appliad to an uai.tlng landfill on Ioni j 1 .daml.
Coiit .imiu*nt
from a bypot1)uttonl
1)3
Pickan. and Lennoz
(1916)
20, C
81.
1.. An
Ad
Du
tranaport
landf Ill.
Ilyputbotical atudy of .u)ea .urfaoa pollution by
1)4
1)5
Bchwarta (1971)
Bradaboaft and P1n4u
21), C
20. A
HOC
HOC
81
i c
Au
An
Ad, Cu,
--
raidloactivu woatea 41975). modal ana)y ta at a
propound uautu-nnnayueunt attn (1977).
Koveluant of aalL water In confLu )ad lIulu8tonU
piudictud future co,,centraliona mild
86
(1973 )
koaLkow and Mruduho .Ct
21). A
W)C
ir
Au
Ad, Do
aqultu I
tuatod utfacta of prut øttvu pumpln j.
Uuud celibratud *udul to aviluato attucta of
.llttorunt irri.jution precticua on amliniLy
1)7
(19 74k
Roburtaon fl974)
3D, A
HOC
ir
An
Ad. (hI
c l i uu In an alluvial .traaia-aqultcr ayalem.
Tranaport of tnduiitgial and low-luvul radio-
act Iwo uautua Ililu the Uuuka Hivur Plain
kobart.on and
Idaho. SImulatud 20 Islutory
aarraolouqh (1973)
aquIfer.
of I o)IuLton.
Pollution of uhalluw aqulfur by saupayc from
1)8
Robaon (1914)
20, A
HOC
Do
..we ju troatauiit pandas 1 .rudI tu4 tutuia con—
contrutiona and Leatad alter .iatlvu water
ne.ia.jeimunt p1 ana.
TI. eo-se.j mcnl modal tar flaw, iiic1ulIn j abilIty
1)9
Robertson (1914)
2/31)
A/HOC
?
An,
8.1,
to sinulgia perched watu In thu unaturatud
80440 (4.4..ü also 1)?).
8i ulatu4 30
I .iMory of jrounJwatur
1110
fontIow (1977)
21), A
HOC
81
year
pulluI on by ct.loitdu from an unl lued dispu m1
pond *4410 thu un .Iur)yln ijj alluvial aquifor.
VOCuiWl at 116$ uuud as a coat-at lectivu
14*1
1 1a 1wu9 and Labadi.
20, A
HOC
Pr
An
-—
aulloiLy malla.)ulsonL technique for btruu*-
(1976)
a.juiter ayatuila.
-------
iAUL 5 0. (CaiitIsiuudI
Appi font lena/Co.vauisLa
Tiuns port of *uJuatrkal and )ow- evu1 rndto-
notivu uaut.se Into IntorbodAlod baualt flceua
lind un ’onuu1tdatu4 audlaunta.
?hruu-d Laiungluian& torau1it on. two-diannalonut
applicdtIon only.
Conaldurii aor Iton and iiachlingo of uuvurgl
*ucra- nJ icro-ioi...
DuucrlliuI aid pru4fetud futuru C cut rat lonu
of hoxavalusit cliroatsu. suuplnq troa a
diupotial pit into uodurlyinq glacial ouLwouh
a.)uj fur.
Duucrtho qround—watur pollution irca u.a)t
doia luacl igtu*.
Atual andul for uultilayucud aquaifuruyutuusj
ptudlctu .l cuncuatration chasnjua attat 4usa
Co. toot ton in th e alrouan Plain, ?unu ia.
V)t ,i pait ba,wJ on souuainuuq oguatlon,
duu ...rtbos puflut L ou by ft.Cl ho large unit
dwapu into alluvial oquitur Lu Uorthou .jtucn
V t .aiicu.
APPIIOd Lu .iulutu trassupurt i d Iht4tEWJUiiCOtHI
aqul for.
Vstr1 calIy LI .Le4JraLud .slLirp—iista*rtuce tialt
watur int uxtusi so..iul. Ito trunuport uquotton
Lu uolvutI.
Cuiculgttii .j thu pocitiun of Lisa suitwatur
front.
804WA 5Lü1 uii tuaCIiia øL Lu .stal aqutfuru.
Sturaqe of fiut.ii wEtur L underground
CetvOiI$ cuntatntn.j aa tnu Water.
CeIa.uLaLI .j thu tranatunt pouttlon of a
anitwatur front.
Nodal Nodo)
No. It’aturoncs
Gaoautry Nathod Typp Typo
at of of at
ifodul I £010Li0112 1 ( 5 )) aoil 4 )
Tyite of
CL a ical
£ntoract too )
Ad, I)u
£4, Cu
Ad, Cu
bU
C .rOVU l3lO)
20
Vu
U13
B14
015
iluddull and tlunela
0170)
Atslstroa and Lsca
41971)
Pinder 11373)
21), A,
20, A
20, £
c
sOC
IOC
HIS
?r
St/fr
St
Lu
Lu
An
tslL
017
Thoau at a l (l977
hurtiue* et . 11975)
IIu beu at at. (19761
20, A
2/10,
A
uta
E l)
St
Tr
-—
I.
Ad,
--
Da
1)11
Pried l375
Fried aôd UagomaoI.
‘I’ll)
20, A
ED
Ti -
——
--
019
I.uaat 11976)
20, C
UPS
Tr
I.
--
SALIWAflR INIRUSIOII
(4Oá LS
20, A
?VK
Tr
—
-
020
PinLar and Page
I 19TH
021
SugoL , Plndur and
Gray (1915)
20, C
SIPS
St
An
-—
022
023
Lou nt Chuag (1174)
Gruut and Con (1966)
25)
20
ISIC
bloC
Ti
Au
--
--
024
Ptnl.r and CoOpur
(1970)
20, C
bloC
St
“
(CuIILilnMk t*)
-------
TAtiI-K 5 0. (ContIa ud)
C UNSATURAtED-OlaLy
CL bra.Isr 419751
Ca Ilildubrand •nd
Ui,nwlbau (1971)
Illldubrand UllS )
C) Braslur 9731 3D
C I Wood sod Davidson ID
1. ..) (1915 ) 1 Davidson ft
0 ft• (1975) ,
CS Utys ft 1. (1971) 3D
CA uLia at ft. 41970) ID
C l Shah ft al. (1 5) 3D
C I Kirda CL .L (19111 I D
C I tanji ft !i. 4 19 )l 1
çanji ot a).
(1972)
do DuLL at aL. (19721 I D
CI I Robin anr aaua fl913 ) ID
CU van GUDUChLUR and ID
llndur 13977)
CL ) Guteyhian at ID
4 1917)
CII lUn.j ani flanks It)
fl973 , 1915)
ClS Gatitlut u a). (19 171 3D
rD Tr --
VD At I. Ai i,Ca
Dus tIbus two-dlaunslon s l transport of
solu tua LIOSOt U LC*Ck lU sourco.
tra. .sport of nIt ata in a •and coluan.
Cui ..p cu4 reeuLL with field data on
chIoridu tranw oct duriny inft ltratLoui.
kt ’ tieJ to peaticilu transport.
Coielmat J rialig. with observed fluid lata
on cIs)or du tr4aaport during Infiltration.
Applied to ranapoct of 2,1-B tib aoila.
1.ppilud to 1 bouphocus transport in aoiisg
aetuilus co,sstant diapurutun c uttIciunL -
ond kinetic sold for ihorploru ia4iorI LioJI.
A 1 ipliuJ to anion sovaisent i soil cokunna
Aiproxiisuta uolutiona fu r cation ctai jo
iii fluI d sollu.
$udui ltiJ of Luachate and soil InLuractions
In an aquifer.
SI isuIatioii of poLlutant tranulort in Long
I .i1a,id W. V.
Applied to it igatlon tuturn (Low quality
etujieg.. IiicIuJus punt root uptatu of
wetigi.
Apilluti to soiL wIth sobila and l.uuubllu
W LuC.
Model
lb.
•
$ 4.I
ku(urSncs
G.oaetry
of
Motel
Nutlod
21
Solution
Type
Of 3)
tiuw
Type
of I I
IlOIL
Type of
Chemical 5)
interactions A ppLIcUL IO i ) 5/C OSIIent iI
2D.C VU
ID VU
V U tr -- --
VU Tr —- Al
VU
V t )
3D SI
- - AJ .Du
Ad
a. Ad
b c )
1
0
_.A
0 0
V
I
--
Al,
Cu
L}
I.
Ad,
() ,
Cu
ll&N/VD
VU
Tr
I.
AJ ,I).i ,CO
VU
I C
--
Ad,DO ,Ce
(Coot. I jugad)
-------
TMILE 5O.(Cono )u*tod )
ci’
c l i
Selli. . .j.ll ii ) 10
Wggrtc i . a t. I D
0 1 7 1)
cie Pa jatral. 1 aL 3D
( 1 5 )
P ANAlYTICAl. T!MlaPOftT
l4OflEt .S
Dl kuo U1Th1. then fl 114 1 2 3D
Cleary, j. 0973)
Wang. e 1197))
Yeb an 3 Teal fll7 ),
Cleary (117 I
02 l.ap&4u. and Leaudeon I I)
(19 2) • Brunnuc 11912)
Lisniatro., at •
Lindetro. 1i bureau
(1971. 19 13), Lindetro.
and Stone 4)97O. hI.
Cloary and A4riaa (1973)
Marina fl174 ) Ogata
(l9b1) van Ganuckteu
and Wi.ran .ja (19)4).
Salt. .nd Man u1L
0974), Soyupnk , a
at. 0 ,19). rufler.
1t971 ). Preru, fl900)
Leonath and lM cbopu
(*990) . Davideon fl9$0)
3) 2)
ii) — on.-dlaenelona) A — Analytical
20 — two-dlcnilo gl Vt) — yinit. difference
3D - three dJen 4on 1 — Unaar unit. • le ,eent.
A — Area) 120 only) TV — Triangular t1ntt a eiamunts
C — Croseaucttona l (20 o Iy) • MIzud/highur ord.r finite
elnaunte
• $cithud of chaj-actvrhatlci
0 — Otbur
App lat i asa a/C c.a e nL.
Nudeliai .j of leachat. end act) inturactic.na
In an aquifer.
1ppro iIsaai. analytical aclution of traaia-
port uquetloni applied to field irrigaLtQo
atudy with chloride.
I4* cibla atseplaceisiant In w tl .
Variouia applto .tiona. Inoluatings
- zero end firet or.ier decay,
— linear uquilhlerIth. adeorptionj
— tireL order kinetic adsorpttonj
— ao lute trenetur between obtla cud
tanobilu watur
- ducey boundary condillona.
Model
Mo.
Modal
katferunc.
Gecantry Huthud
at of 2)
Nod 1 Solution
type
of 3)
flow
Type
4)
Soil
Type at
Cheaia.ial
Inter.ct lona
llY /VD
Tr
I .
Ad
ru/A
Tr
--
--
I’)
t.J
HOC
A
A
St —— (Ad. Pu) Variowa a .pllcetion. and aenueptiona
(14 , Pa)
mm
.‘
0
<0
0 0 .
a;’
p
3)
• Trene Least
St — Steady-atatu
4,
I . - Layered
An — AniaOLropIc
S)
Ad — Adaorpt Lou
Ce — Cation . changw
laulti—lon traneport)
0. — Decay
-------
3 dimensionS).
These additior.al classification schemes are described in
the following text.
Empirical or Conceptual Model .
The delineation between an empirical and conceptual model
is dependent upon whether or not the assumed physical processes
use input variables to produce output variables. Empirical
models are based completely upon observation and/or experimenta-
tion, while conceptual models use equations based upon the
conservation of mass, energy, and momentum. Examples of these
models are presented in Table 49.
Deterministic or 5t chastic Models
In a deterministic model, all input variables and system
parameters are assumed to have fixed mathematical or a logical
relationship with each other. These relationships completely
define the system (a unique solution is obtained). See Table 49
for examples of these models.
Stochastic (or probabilistic) models, on the other hand,
take into account the intrinsic randomness associated with any
of the system parameters, or the uncertainties associated with
the many mechanisms operating in the system, the system para-
meters themselves, or the input variables. Within the stochastic
mode]. category, two types exist. These are:
1. Stochastic models where the system parameters and.
jn ut variables are characterized by assumed probabji. .
itv istributioflS (normal, log-normal, etc.) which
are used to characterize the statistical behavior
of selected parameters (Freeze, 1975).
2. Stochastic models where the system parameters or
input variables are uncertain either because of
unreliable input data, or due to measurement errors.
uncertaintY may also result from the case of an
ever -simplified model where different mechanisms are
often bunched together, resulting in less well—
defined parameters. Tang and Pinder (1977) used a
mean and variance approach to describe the uncertain-.
ty of applicable parameters. Stochastic models
generally determine a confidence interval for
each of the ouput variables.
Static or Dynamic Model
Static models evaluate steady-state conditions, i.e.,
Where the input and ouput variables do not change with time.
222
-------
Conversely, a model whose input and output variables change with
time are termed dynamic or transient (see Table 49).
S atiaJ. Dirnensionality of the Model
The spatial dimensionality of the model refers to whether
it considers one, two, or three dimensions. A one—dimensional
model, for exam 1e, could be used to describe the vertical
transport of contaminants in the unsaturated zone between the
landfill and the ground-water table. Such a model, however,
will not provide accurate information regarding ground-water
pollution because the effects of ].eachate dilution by ground
water cannot be evaluated. In general, however, at least a.
two—dimensional cross—section model must be developed in order
to obtain the desired information. Two-dimensional models
may also be formulated on an areal basis whereby the system
parameters and input and output variables represent average
quantities along the vertical dimensions. Three-dimensional
models are utilized to obtain a more representative character i-
zation of an actual setting. Examples of one-, two—, and
three-dimensional models are given in Table 50.
MO DELS
Of the different types of models previously discussed,
conceptual-mathematical models appear to be the most promising,
but unfortunately, also the most complex ones for evaluating
ootential ground-”iater pollution from waste disposal sites
(van Genuchten, 1978). A compilation of the different types
of co ceptualmath atica.l models available for possible use in
ground—water quality evaluation studies is given in Table 50.
This compilation by Weston, Inc. (1978) is based largely on
contributory material prepared by van Genuchten (1978). The
list of models in Table 50 is not intended to be comp-lete; its
purpose is to demonstrate the existence of a wide variety of
models, to characterize their most important capabilities and
limitations, to identify their method of solution, and to show
some of their applications. The list of conceptual—mathematical
models in Table 50 is delineated according to the following
teg0rie
• Saturated and unsaturated transport models;
• unsaturated only transport models;
• Saturated only models; and
• Analytical transport models.
Each tYPe is described briefly in the following text.
223
-------
Saturated — Unsaturated Transport Models
Saturated - unsaturated transport models simulate either
a three—dimensional system (Model A2) or a two—dimensional
cross section. Except for Model A7, no cation exchange (multi—
ion transport) is considered in any of the models in this group,
though several include single—ion adsorption and decay (Mode ls
Al, A3, and AZ).
Saturated - unsaturated transport models are probably the
most appropriate ones for landfill simulation because they
consider the unsaturated conditions in and under the waste
disposal site. Another application of these models is the
determination of the dilution of lea hate (an important atten-
uation mechanism) by flowing ground water.
Saturated—Only Transport Models
These models deal exclusively with saturated flow and
ignore the dynamics of the unsaturated zone between the waSte
disposal site and the ground-water table. Thj approach j
most conducive to contaminant transport when the ground water
saturates the waste material. Many saturated-only models e
the method of characteristics (MOC) for solving the governing
transport eguatiori. Most of these models are adaptations of
mass flow approaches for fluid flow.
A special class of saturated—only transport models is pro-
vided by salt water intrusion studies (Models 320-924). These
models differ from the other (cross-sectional) models in this
group in that they consider density-dependent flow, and, as
such, are applicable to contaminant transport from water dis-
posal sites.
UnsaturatedOflly Transport Models
These one—dimensional models are useful for studying the
mechanisms of contaminant transport in the unsaturated zone.
Another important application of these models is obtained when
they are used simultaneously with saturated-only transport
models. The unsaturated models can then be used to redjct both
the amount and type of leachate reaching the ground-water table
information which can be used as input for the saturated-only
transport models. Models A6 and A7 represent examples of this
type of approach.
Analytical Models
Two— and three-dimensional analytical models appear to have
limited application to actual field ground-water contaminatjo
problems. Their application is generally restricted to one—
dimensional models based on constant porosity, dispersivity,
and conductivity. The different one-dimensional analytical
224
-------
models (Models D2 in Table Q) are potentially useful tools for
identification and quantification of waste—soil interactions
when used in conjunction with column leaching studies.
Conceptual-mathematical models are based upon a set of
equations which describe relationships between different input
and output variables as well as system variables. After suitable
5 j plificatiOflS, the equations generally reduce to a set of
coupled non—linear, second-order partial differential equations.
o these partial differential equations, one will describe fluid
flow while the others characterize the transport and behavior
of different chemical constituents associated with the waste
leachate.
uxnerous conceptual-mathematical models are resentlv
available with differences between the models arising from the
numbers of siinplifications made during the derivation of the
basic governing equations, the method of solving the equations,
and the stipulated boundary conditions.
The following partial differentail equations may be used to
simulate the (density independent) transport of water and dis-
solved constituents in a saturated-unsaturated three-dimensional
medium, such as occurs in a landfill and the underlying ground-
water system (van Cenuchten, 1978).
Mass transport ecuation :
3ec,
( D 3 )
(a) (b) (c)
- h — (q c ) ± m (am 3c )
(d) (e)
*
± ±Qck
m=].
f) (g) __ C 3 9)
fl.oW ecuatiOfl
( S + C) (K.. + K. ) + Q
n s t i.j 3x. L3
i, j 1,2,3) (40)
225
-------
Table 51 provides an explanation of the s nbo1s used in these
equations. Processes described in this constituent transport
equation are as follows:
• Changes in adsorbed constituent concentration;
• Changes in solution constituent concentration;
• Diffusion/dispersion effects;
• Convective transport of the constituent by the fluid;
• Production (#) or decay (—) reactions;
• Additional chemical/soil or chemical/chemical inter-
actions (precipitation, chemical transformations,
cation exchange reactions, volatilization, etc.);
and
• Constituent concentration changes resulting from
water sources (+) or sinks (-).
The volumetric fluid velocity, q., is required before the
transport equation can be solved. Fo this, it is necessary
to also solve the water flow equation. The water flow equation
may be solved to prowde either a steady state flow simu1at on
a Q), or repeatedly, in the case of a transient simulation.
Whatever approach is used, the vol mtetric fluid velocity, q.,
is obtained from Darcy’ s Law (Schwab, et al., 1966) in the
form of:
(41)
The transport equation is also coupled to the flow equatj
through the dispersion coefficient, The magnitude of D .
depends upon the fluid flow velocity, q , and the soil moistu. .e
content, G (which in turn is determined from the pressure head).
The constituent transport equation allows the incorporatjo
of an adsorption component describing the dependency of the
sorbed constituent concentration, 3, on the solution concentra
tjon, c, through the use of appropriate adsorption isotherms.
Models available for adsorption and/or ion exchange are class-
ified into either ecuilibrium models which assi. ne instantaneous
chemical adsorption, and kinetic models which consider the
approach towards equilibrium. Table 52 lists a few of the most
frequently used equilibrium and kinetic absorption models.
Most of the equilibrium models in Table 52 represent special
cases of kinetic models, in that, they are derived by setting
the time derivative equal to zero.
226
-------
51. EXPLANATION OF SY24BOLS USED IN T MASS TRANSPORT AND
FLOW EQUATIONS
Symbol Explanation
Solution concentration of chemical species k (NL 3 );
c Solution concentration of the fluid source or sink
k (ML 3 )
C Specific soil moisture capacity, C fl (L 1 );
Dispersion coefficient (L 2 T 1 );
Pressure head (L);
Soil hydraulic conductivity (LT 1 );
Porosity CL 0 );
Voli.m etric fluid velocity (LT );
Q Fluid source or sink term, Q Q (x — x ) (T );
Strength of fluid source or sink (L 3 T );
Rate term expressing soil-chemical and chemical-
chemical interactions (ML T );
5 1c Adsorbed concentration of chemical species I c
Specific storage coefficient (L );
Degree of fluid saturation (L 0 );
Time (T);
Distance in i-th coordinate direction CL);
i-th coordination of source or sink (I,);
m-th order rate constant for production or decay
(Ml mL 3 m 3 Tl);
Dirac delta function (L );
(Dry) soil bulk density (ML 3 );
Volumetric moisture content (L 0 ).
227
-------
TABLE 52. PARTIAL LIST OF EQUATIONS USED TO DESCRthE ADSORPTION REACTIONS*
MODEL
1.. qui1ibriurn
Li (linear)
1.2 (Langinuir)
1.3 (Freundlich)
1.4
1.5 (Modified
K )e l land)
‘- 2. Non-equilibrium
t . .)
2.1 (linear)
2.2 (Langmuir)
2.3 (Freundlich)
2.4
2.5
2.6
EQUATION
s = k 1 c +
k 1 C
= 1 + k 2 .c
k 2
s k c
1. .
k 2 8
a = k c a
— CS
— c + k 1 (1-c) exp
= k k 2 -
_kr(1+kc S
1k 2 (Cm 2 Cl
as
- =k (k 1 c -s)
k s —2k 2 s
as e 2
1€ = k
(k 1 c e
k(s -s) sinh
k k
as 1 ‘)
and are constants, kr represents a rate constant
initial and final (or maximum) adsorbed concentrations,
Genuchten and Cleary, 1978).
REFERENCE
Lapidus and Aiuundson (1952)’
Lindatrotn,,et al . (1967)
Tanji (1970)
BailauK and Peaslee (1975)
Lindstrorn and Boersma (1970)
Swanson and Dutt (1973)
Lindalrom ,et al. (1971)
van Genuchten,et al. (1974)
Lai and Jurinak (1971)
Lapidus and Amundson (1952)
Oddson,et al. (1970)
Hendricks, 1972
Ilornsby and Davidson (1973)
van Genuchten,et at. (1974)
Lindstron et al. (1971)
Fava and Eyring (1956)
Leenheer and Ahlrichs (1971)
Enfield,et al. (1976)
(T ), and Sj and 9 m represent
respectively (after van
—8)
-------
Other relations related to the geometry and boundary condi-
tions of the partial differential equations are also required.
These auxiliary conditions may include such information as:
• Thitial constituent concentrations distributions;
• Type and concentration of potential contaminants;
• Geometry of the waste disposal site;
• Aquifer configiiratiort (2 or 3 dimensional);
• PrecipitatiOn/evaPOratiOn data; and
• Location of surface water bodies.
Once these relationships have been determined, solutions
for constituent concentrations can be obtained by appropriate
mathematics. 1 manipulations.
MAT NATI SOLUTIONS FOR CONCEPTUAL MODELS
Mathematical solutions for conceptual models are generally
obtained by either analytical or ni.nnerical aoproaches, which are
briefly discussed in the following text.
.nalYtical Methods
Most analytical solutions require an assumed constant value
for the fluid velocity, dispersion coefficient, and the physical
parameters. The advantages of analytical solutions are the ease
f application and low-cost of operation, once derived. However,
the necessity of having to make simplifying assumptions severe’y
restricts the application of analytical solutions to waste dis-
posal ground-water contamination problems. Although analytical
methods have limitations, some of the available two— and three-
dimen3i0x analytical solutions (Kuo, 1976; Yeh and Tsai, 1976;
Larson and Reeves, 1976) probably could be applied to a few
weil_fi ed qeohydrological systems (van Genuchten, 1978).
erical Methods
A nwnerical approach is required in most situations because
most field problems have such complex physical and chemical
cact i5ti In most situations, numerical methods may be
applied advantageously. Numerical approaches exhibit the ad-
tageS of reducing the partial differential equations to a
5et of approximating algebraic equations. This approach pro—
a greater flexibility in portraying the complex physical
d chemical characteristics of the natural system. Of the
methods employed, finite differences, finite element,
t1 e method of characteristics, and miscellaneous models are the
229
-------
most common.
Finite DiffereflCe-
The finite difference numerical method approximates the
derivatives contained in the governing partial differential
ec’uatiOfls with appropriate difference equations. This approach
is normally, successfully employed in ground-water flow studies,
but has limited use in ground-water quality studies due to
several factors such as:
• possible introduction of numerical dispersion;
• occurrence of undesirable oscillations in calculated
concentration distributions; especially when t e dis-
persive transport is small when compared to convective
transport; and
• triability to accurately reproduce irregular boundary
conditions.
While these inherent limitations exist, finite difference tech-
niques represent the simplest numerical approach especially for
simulation of one-dimensional unsaturated transport problems.
Two_dimensional solutions are limited. Exanples of one- and
two_dimensional applications are shown in Table 50.
Finite Element
The finite element numerical method designates the Pressu 5
head and concentration as dependent variables. These dependent
variables are approximated and inserted into the governing
equation. Resulting errors are minimized through the use of
weighted_residual theorems (e.g., the Galerkin Model). As in
the case of the finite difference approach, the study area j
subdivided into elements which, unlike for finite differences,
may assume any configuration required. Finite element methods
have been used successfully in a number of field problems
involving mass transport, although, in some situations the
occurrence of numerical oscillations remain a problem. EX Xflples
of linear, triangular, and mixed/higher order finite element
applications are shown in Table 50.
Method of CharacteristiCs
This numerical approach uses a finite difference approach
for the flow equation while characteristic equations are used
for the constituent transport equation. Characteristics
equations are developed by eliminating corrective transport
terms from the governing equation and including them in separate
equations. The correcti re transport terms are reflected by the
insertion of markers 1 particles (or moving points) into each
finite cell. The marker particles are moved through the network
as prescribed by the local fluid velocities, thereby establishing
the effects of the convective transport terms. The method has
230
-------
been shown to produce acceptable results for a wide variety of
field problems for one-dimensional studies (Table 5Q). An
important disadvantage of this method is that it is generally
djffiCUlt to program in two or three dimensions.
Miscellaneous Numerical Models——
Numerous other numerical models exist which can be applied
to ground-water contamination problems. Many numerical
models exist which do not rely upon the direct solution of
governing partial differential equations. Many of these models
do not consider spatial variability in the dependent variables.
In this approach, a mass balance equation is generally applied
to the entire system, the different input and output variables
being only a function of time. (Hornsby, 1973; Gelhar and
Wilson, 1974; Donegan and Crawford, 1976; Mercado, 1976).
An approach incorporating more of the characteristics of
a distributive parameter scheme may be obtained by directly
applying the mass balance equations to a number of well defined
soatial cells. By this method, for an explicit time, a finite
difference approximation of the governing equations is obtained,
provided all significant transport mechanisms are included in
the mass balance equations. Tanji,et al. (1970) and Orlob and
oodS (1967) have presented examples of this method. Elzy
et al. (1974) applied a similar approach for the development
rvertical—horizontal routing model to the transport of
hazardous wastes from a landfill site. A two-dimensional
polygonal network has been described by Hassan (1974) in which
a u1ti-layered ground-water basin was modeled.
Each of the numerical techniques discussed above has
specific advantages and disadvantages associated with factors
affecting the accuracy, efficiency, and accessibility of the
particular method. It has been suggested by van Genuchten
(1978) that the accuracy of the numerical technique may become
secondarY when viewed against the uncertainty in the many
input parameters needed in most models. Additionally, the
prcgrafliflg efficiency, the general Setup of the mode]. and its
accessibilltY, are also important factors affecting the useful-
nesS of a particular solution scheme.
SZ 4MARY
The attenuation and migration of contaminants in the soul
water system are influenced by physical, chemical, and biologi-
cal. mechanisms. The most important physical properties of a
g jl system relative to leachate migration are diffusion and
dispersion, dilution, physical sorption, and straining. Impor-
tant chemical mechanisms include: precipitation/dissolution,
adsOrPti0fl/ 5orPtj0m , complexation, ion exchange, and oxidation!
reduction (redox). Adsorption, complexation, and ion exchange
processes are difficult to distinguish from one another;
231
-------
therefore, they are usually grouped together as if they were one
mechanism. Major effects attributed to microbial activities are:
oxidation/reduction, tninerali:ation, immobilization, and com lex—
ation.
The migration of trace metals in the soil/water system is
extremely complex. Thermodynamic and kinetic influences make
transport phenomena difficult to explain. Some of the major
mechanisms that influence mobility of trace metals are:
solubilization, sorption, coraplexation, and dilution.
The fate of chlorinated hydrocarbons in the environment is
very difficult to predict. In general, volatilization, microbial
degradation, chemical hydrolysis, oxidation, and sorption can
be involved. Various studies indicate that sorption may be the
dominant mechanism in soil/water sys ains.
Factors j fluenciflg the fate of pesticides in soil/water
systems include: adsorptiOfl/de50rPt 0fli microbial decomo s ition,
votalilization, soil moisture, and physical properties of the
soil. Adsorption/desorPti ofl is considered to be the prime factor
governing the interactioflS between pesticides and soil col1 jds.
Recent studies have pointed out that immobilization of
viruses by soil should not be equated with virus inactivation
because many adsorbed virus particles have been demonstrated to
be infectious for significant periods of time, virus survival
within a soil relate to soil moisture content, temperature, p ,
nutrient availability, and antagonisms. An evaluation of these
factors indicates that viruses which penetrate the soil surface
are expected to survive for prolong periods of time, as compared
to those retained near the soil surface.
Quantitative data relating to the attenuation of individual
constituents by specific mechanisms are not available, consequent-
ly, considerable effort has been placed on the development of
sophisticated leachate migration models. However, due to the
highly complex leachate/soil environments, no leachate migratjo
model exists that can simulate all of the physical, chemical, and
biological processes occurring in a typical landfill system.
conceptualntathematiCal models appear the most promising, but,
i. nfortunatel7, also the most complex ones for predicting the
performance of waste disposal sites.
232
-------
SECTION 6
AVAIL? 1 3LE CONTROL TECHNOLOGY
INTBODT CT ION
Environmental control technologies are presented in terms
f the following four topic areas:
• Leachate volume control;
• Leachate composition control;
• Leachate collection systems;
• Leachate treatment methods.
Leachate volume control, within the context of this report,
refers to the control of water entering the fill or Leachates
ating from the fill. Major water sources Lnclude contribu-
tions from groi.md water or surface water infiltration.
The largest contribution to landfill moisture, in most
5 jtuationsi is from precipitation. Precipitation can enter the
1 dfill: (1) during the actual filling process in the area of
exposed open working face, (2) as a result of infiltration
and percolation of surface runoff and overland flow, or (3) as
a result of infiltration from inadequately designed or maintain-
ed surface runoff channels either around or through the landfill.
GrOuX water can enter due to a fluctuating water table, or
due to a poorly selected and designed landfill which has
lateral ground-water migration through the fill
area. Capillary action through fine grain sands and silts for
1 4fill5 in close proximity to the ground-water table is
other pathway. The landfill operation may also contrjb te
relatid7 Y small amounts of water during periods of dust control
Available leachate volume control methods for each of
t1 eze sources are evaluated in terms of the three following
criteria:
• optimum site conditions most conducive for a
particular control approach;
• Limitations associated with the control method; and
233
-------
• System economiCs.
Detailed descr.Pt 0flS of control measures, in terms of these
criteria, are discussed in the following text as well as pre-
sented in tabular form.
LEACH.ATE VOLZrMZ CONTROL
Ground—water Control Measures
Ground-water control measures are presented in terms of:
(1) ground_water/leachate isolation, and (2) leachate plume
control.
Ground-water leachate/isolation methods can be defined as
those approaches by which contaminants are contained within the
fill and not permitted to migrate into the ground water. Leach—
ate plume control refers to those methods which attempt to
prevent the contact of the plume of a contaminated ground-water
body with additional ground water or surface water resources.
Available control techniques include the following:
• Groundwater/LeaChate Isolation
Artificial Liners
Bottom Sealers
Slurry Trench
Grout Curtain
Sheet Piling Cutoff Wall
• LeachatePluxne Control
Drains
Well Point Systems
Deep Well Systems.
Grouxld —waterlteaChate Isolation——
Artificial liners——Artificial or synthetic liners may be
utilized as both a collection device and as a means for isolating
leachate within the fill and keeping ground water out of the fill.
Numerous types of synthetic liners are available and a significa
volume of literature exists describing their characteristics
(Ewald, 1973; Lee, 1974; Geswein, 1975; Sanks, . , 1975;
Haxo, 1976a,b,c; Haxo, 1977a,b; Eaxo, 1978; Stewart, 1978;
234
-------
are and Jackson, 1978; Haxo, et al., 1979; Styron and Fry, 1979;
3rOWn and Anderson, 1980; Matrecon, Inc., 1980). An excellent
t ary of commonly used synthetic liners and their properties
compiled by uinar and Jedlicka (1973) is presented in Table 53.
structural and physical information for various synthetic liners
compiled by Ewald (1973) are provided in Table 54,,
To be effective in controlling leachate, liners must pos—
gess a number of physical properties. Kuxnar and Jed.licica (1973)
have listed 11 criteria for synthetic liner selection which are
Listed jfl Table 55. While abundant literature exists describing
these physical characteristics, many properties (e.g., liner
reaction to leachates, anticipated liner lifetime for a number
of environmental variables, etc.) have not been rigorously
quantified. Eaxo (1980) provides an excellent discussion of the
results of his studies regarding some of these properties.
The installation of a liner for containment purposes has
been thoroughly described by Schultz and Mikias (1980). These
researchers have categorized installation procedures according
to: (1) subgrade preparation, (2) liner placement, and (3)
special design considerations.
Subgrade preparation involves cortvoactjon and soil steriliza-
tion. Compaction of the subsoil may be accomplished with a
sheep’s foät roller attached to a bulldozer which is suitable
for steep slopes. A typical subgrade could consist of 24 in.
t 48 in. of compacted soil with a 2-in, layer of sand above the
soil. A vibratory roller or compactor may be used to grade
these layers. Other approaches have included the use of bull-
dozers or other earth moving devices for soil compaction, then
smoothing prior to liner placement. In those cases where rocks
may cause problems due to their size or jagged edges, they are
usually removed manually.
Soil sterilization is necessary for controlling the growth
oe plants which may puncture the liner. The literature indicates
that it is imperative that the proper effective regional herbi-
cide be selected for a particular geographic area. It is
recoxi nended to wait several days before placing the liner
directly over the treated area. Placement of the liner imxnedj—
ately after application of the herbicide is believed to
accelerate plant growth process due to increased moisture and
warmth (Schultz and Miklas, 1980).
Liner installation requires a number of considerations for
successful placement. Among these are site storage, securing
liners, bentonite application, seaming, sealing penetrations,
wind damage, soil cover on the liners, and quality assurance
programs.
235
-------
TAUL 5 PWL1tT1 CV ui*aax 00tD Ut& 6 ll ’s°’r Jixtlic$.g , 19231
lb1yut1iyloi h)yvthyl l1 )utJoato3 lt y— Outyl ILiturol
L . oity lIlqh Duootty C1*1orl o yloo IOOIIw ktI* r
l IcItIc9rav ity 0.92-0.94 0.94-0.96 1.20-1.S 1.35-1.39 0.9-0.9 L 1.00-1.4 0.92—*.2 5 0.91—1.25
? nst1e Mren th . psi 1.300-2,500 2.400-4900 3 ,500—i0 ,000 1900 mm. 4000-32000 9 .000-11.000 1.000-4006 1,000-3,500 1 .000-2,000
91u(V)otioO. i oo-ooo mo-iso 60-206 315-525 40-400 250-556
6toco A Iiigdwo. 15-Pc 20-100 55-95
c * istiiq tu .51w5t1*’.
ron’i , ‘V -10- to 190 -20 to 240 -60 to 200 -40 to 206 -60 to 220 -20 to 300 -50 to 325 -10 to 250 -45 to 250
to sold. P-C C G-E Gil o-e p
I glot . .i to Lgaous 0-5 G- 5 G G- i 6- 8
l.tsoou to oy ’)io .t P-G I ’-G 0 p C
I t .blbtO . 10 stCsk&(*Q -
. d holo . otød aulvuots P V U P 6 0 P P V
ILiolutofl(i to 51 I 115110
ut oIuu.) .o1 t. P —V V-C 0 6 G I i P P 6
W.*luc V .i )OC panuoabllity,
.A1i 1-14 1.9-2.3 3- I l 0.040-0.049 0.25-1 0.09-1.9 0.15 2.0
I4 Lh r hlIlLy P P 6 8 P p 6 V
1’.) ‘tLm to ccadi, hr. 900 300 lb iock 1.11 lb ttsx* 111 100 1,200 -
1.4 2.500 lit. 4,000 hr.
to di .1k , hr. I i) ottcc* til 600 300 DittO 600 lb of fuct 1.11
2,500 1w. 3.100 hr.
Thu to falu, hr. 300 300 100 DItto 900 200
Ibto* l)ots w L i r. ai .vaiLohl. to LtO auLl s. P - Poor, V - Volt, 6 - Qa I . £ - £ I loot.
Nrnl to*t 5.3110.1. t virious propurtlus; u1 .ucItta ovity, D151 Itoutlu Ct.rwiptli. D97-61T 1 oLui.j .tk.o , l14I2-bT S10ru ‘A’ lurdibS .,
1i6i6-SP w .tor vu r pet5. aI4iIty. 896-66.
-------
V4iclc aiwat
Warp Junker
VlUin .j Junker
riLur
Waive
WukjLt. oa.Jiq. yO
r 1 O.igtin thii on.s SUe A. all. -- -- --
C rnt*rnj tlddu .. ., 114* I . .11. — — —
1 *aI tg.ic$ iiii. aitI 17 20 20
0 lUtal wat.jlit oz./a j./p1. )4.S 17.0 111.0
lurni la eta-nn $h Iba.IIn.
• warn $ 5
f l l l*nij 30
LI 1 i 4 I ai..
Udi)) 30 35 20
1l l )Irq 30 35 20
Tuat etzenyth Urn. (lunç wa)
S S
I IIUI*J S S
Nil leas h drnut.t1o 100 125
20 2 * X I 30 I I — — 21
111 20 30 30 2) — —- 2)
TM11X 5¼ TIPICM. QNlk t 2tl t*4 (W 0YImI2 IC
U*301 tUad4. 1533*
Waee l diifon*jain
Ujiut. of lirpekin lIal caun
Un .** at *3914 IUI* win
L$i*e u of r Nrnlirnnu
VaI. IC
*al*ica**it
I$ S 11*0 22*fl 14*14 12.12
-—
22 11 311.21 21.21
Warp JunIor
100 100 210 410 $40
—- —— —
210 210 810
F* )iisJ denIer
210 2*0 420 440 $10
— —— --
420 2*0 840
rIlor
Wanve
*&ilght. os/v . 1.
, ‘*rn nylou nylon pIy .tor nylon
2 /I loon 2/ I 10 1*) iial plain tI u
0.4 0.4 1.4 1.11 2.1
nylon nylon nylon
—— — —
2 2 2
nylon nylon nylun
plain plain plein
1.4 2.0 4.9
•al aiin
Q Lth9 Udduawa-Sidu £ mi ii
CoaUn thtc*jw ..-8t4. I, all.
IS lS 14 14 I I
15 1S *4 14 14
—
— — ——
5 4 0-’
3 2
Tutal thickriuve. aft.
1 .itaI wat jir . oz./w . yJ.
Tundlu st ii h lb../ln.’
warp
fL l I in.j
30 45 20 28 28
33 4$ 28-32k 3O-32 2)1
4
15 120 120-115 1)0 2 1 10k
35 *20 120— 119 130 284)
30-34 44-41 11—15
— — -—
150 158 150
158 150 150
10 14 3 111
10 5 18
110 200 4 50
110 *90 400
Lkn at Ian, O . k ..
W1U3)
23 23 — —— —
30 20 2)
f ililncj
23 23 —— —- ——
30 30 30
— — ——
1 )iar .trwyt.b , )ba• 4ton uI
warp
I $ I
20 24 2S-50 10 20
25 29 29
40 11 96
tll)inj
*dlan I djU1ltat1c
20 24 25—50 10 70
175 175 22S-580 204) 279
25 29 29
208 200 200
40 11 00
500* 215 500
Uter, of Ikatyl NiUwr N .4 rrn.
—1
11*17 11*17 22.14 22*14 22*14 22*14 22*14 *2.12 12*12 21*21 22.21
210 210 210 210 2*0 2*0 210 010 $40 840 04 11
218 210 420 420 420 420 420 810 $10 010 810
nylon nylon nylon nylon nylon nylon pylon siyk.s nylon nylon nylon
plain plain plain plain plain plain plain plain *1cm plaIn plain
1.0 10 1.4 1.1 1.1 1.4 1.4 2.7 2.7 4.11 4.9
19.4
*49 100 1)2 *00
10 100 307 100
32 45 62
3* 39.0 41
-- -— S-la - -
- - — 5-la --
*2 1 2 16 45
— — I I —
100 100 225 ‘ —
100 100 200 -
150 229
375 2)0
S
*2
*2
12
*2
50
55
4 4 ) 50
S
I l
*2
10-I l
12
30
25
35 50
*25
l OS
1 44 )
1 85-I SO
200
350
—-
SOOt ——
-------
TUik Si thimsal*
b uu
1 4 g b
l.Lnuts (if
4U U I4..Mgaic
L3ii . tm
of Cit
wtait
Waip d aib
sa iv a
W.iijht, oz.Itq. y .1.
22 14
210
420
plain
1.4
22 z *4
2*0
420
oyk*i
plain
1.4
22 14
2111
120
uy*ou
plain
1.4
22 * P .4
2*0
420
nyhjii
p1 in
1.4
22 u I i
210
420
ivjli t
plahe
1.4
*2 12 12 * *2
4411 440
444 440
ny*oo fiyluo
pl*in plain
2.1 LI
II * 11
2*0
210
ny*(c
platii
1.0
15 * iS
440
140
i lyaiitai
iiIaLt
LI
12 * 12
444
410
uy1
plath
2 1
.
t*uq U k *.—SIdu A 11a
c o ating t1iidnoaw Si4u I, ails
i . . i .a
—
——
20
--
— —
32
--
—
2
—
35
—
—
45
—
——
32
--
——
45
4
4
I
14
14
20
14
11
24
ThLa$ ad$it. aj,yd
20
52
“
4.9
31 )32
32-34
T.ai iia **xwajth lbu.Ith.
wJtp
- tI l l i n.j
i 1C4V )4tiiA ), I a*
108
*06
100
100
104)
100
104 )
10 1 )
100
*00
225
20*)
225
204
254
202
130 *
130
IOS-424)
365—420
I,v
t illing
30
30
30
30
30
30
30
30
30
2)
3*)
1 )
30
34)
—
—
——
1 . .4r atranith ilba./thj
warp
10
14
10
10
*2
50
90
22
*0
54-80
ti tling
S
10
I
S
*0
20
20
22
30
59-80
Hullen $ykoatiitta
145
144)
1119
3119
*90
350
354)
IbIS
200
—
Ilibil is a 5-ply aaturta* fila/scti/ti Ia/aic,i*/f iIa, with 3 pllus cC 15-iU*-tldct ttIa aid 2 p I les of c ia
I Ib./ln. — 0.18 kgA a
is tic sac as aaddnu d*ructLcin ttlllnf ii tic a as cluas 4jiai bou.
In aUltirai to the 4 w iaity l nt&er linings, .iiaiIa linings of bill sill LyI-r* .r/b i*l bLu i a&u avullalilu.
I&iqits are witIuut dip.
thiusa ictasi , tout .cikds 0a A IL$ 1)-iSi ware .ini to& th*ckncas, total wu ght , taisibisi atiw .jth , 441C 5’jati(JB , toaL utrei&jLli, sad lI.illon I) )dauatai.ic tati.
lic list of P liacra La far C na 5114)1 .48 anouja to to i4saial gItetiv.i. IM ’ 1 1 1 52 ( 1 shA,lJ ss* L a ciad asilaas wvartd with lM ith wata to as . iat
rapid waatlcring aiat ion.
d . Iocinstud poi thy1anu, 11*14 - ethyl ykannir nu Ii aisj aid - piilyviaayl dalnaidu.
IDa
0 .0
0
a-
0
o
a
-------
r 3LE 53. SELZCTION CP ITERIA FOR SYNT TIC LINERS (Kurnar and
Jedlicka, 1973)
1. High tensile strength, flexibility, elongation
without failure.
2. Resists abrasion, puncture, chemical degradation
by leachate.
3. Good weatherability, manufacturer guarantees long
life.
4. Irune to bacterial and fungal attack.
5. Specific gravity (5) > 1.0.
6. Color: black (to resist tJV light).
7. Minimum thickness, 20 mils.
8. Membrane should have uniform compositions free of
physical defects.
9. Withstand temperatu.re variation and ambient cortditons.
io. Easily installed.
ii. Economical.
Proper storage of the liner is important to minimize deteri-
oratiOfl of the liner from the elements. Storage considerations
ghOUld be predicated upon the type of liner being stored, antici-
pated weather, length of time the material will be stored at the
5 jee, and ambient temperatures. Liner material may be covered
a light or reflective plastic to prevent possible degrada-
tjofl due to sunlight and massive heat buildup.
curing liners involves the anchoring of the liner to
either the subgrade or to penetrations or other concrete struc’-
tures found within the confines of a surface impoundment or
1 dfill. This may be accomplished by attaching liners in an
cbOr trench utilizing concrete. Another similar approach
ti1izes either a slit-trench or a blade—cut type which has a
5 tee edge adjacent to the liner slope and a gently sloping
from the liner slope. Fill material is then placed
0 ver the liner within the trench. Liner manufacturers usually
rovide useful approaches for securing liners (Asthalt Institute,
1966; Asphalt Institute, 1974; Portland Cement Assoc., 1974;
Industries, Inc., 1976; 3. F. Goodrich, Co., 1978, American
239
-------
Colicid Co., 1979; Burke Rubber Cc., 1980; Watersaver Cc., 1980)
Bentonite is often used for su.bgrade cover in areas where
compatibility or soil tests show that the proper application
cqi2.i. lower the permeability to the desired level. Common
application systems include discing, soil spreader, and a light
spring toothed rake.
Seaming of the liner materials is an extremely important
step. Problems associated with seaming arise when joining the
material on steep slopes; mnoveable support boards or ladders
may be used to aid field crew personnel. The use of improper
adhesive systems and adverse field conditions (e.g., low temper-
atures < 50°F) rain, high winds C> 20 mph) etc.) are other
common problems. Selection of the proper adhesive system for a
given climate and the identification of a possible back—up adhe-
sive can minimize seine of these difficulties. Seaming activities
should be halted when the weather conditions are deemed too
adverse.
Accumulation of excess liner materials in the landfill
corner is another common problem; excess material should be cut
and removed. Seaming in the corners and along the slope should
always be perpendicular to the toe of the fill (Shultz and
Mikias, 1980).
Special precautions must be observed when sealing an intan-
tional penetration (e.g., monitoring wells, grounding cables,
etc.) of a liner. A specially manufactured boot may be available
which can be sealed to the liner.
Wind damage usually occurs when winds exceed 20 miles/hour.
While preventative measures are largely ineffective for winds
in excess of 30 miles/hour, damage may be minimized through the
employmertt of various methods. These include anchoring any loose
edges with adequately weighted materials or the use of battens
(Burke Rubber Co., 1980).
Use of PVC liners require the placement of a soil cover OVeZ
the liner to minimize the deleterious effects of the sun’s
ultraviolet radiation. Other liners which are not as succeptible
to damage by ultraviolet radiation (Hypalon, EPE, and others).
utilize soil covers as a means of stabilizing the liner rather
than as a protective measure. Soil covering may also provide a
leachate drainage system. For example, the following added
sequence of subgrade, liner, and cover soils could serve as
drainage medit.mt: (1) smooth graded native material or 0.6 in
to 1.2 in (2 ft to 4 ft) of compacted soil; (2) 5 cm (2 in) sand;
(3) liner, (4) 5 cm (2 in.) sand; (5) 30 cm (12 in.) of 1—cm
(3/8—in.) diameter crushed rock; (6) 15 cm (6 in.) of 1.3—cm
to 2.2-cm (1/2-in, to 7/8—in.) diameter crushed rock or gravel.
and (7) 15 cm (6 in.) of rock dirt or fine sand (Schultz and
240
-------
MikiaS, 1980; MatreCon, Inc. , 1980).
The use of a low permeable barrier material generally
requires that a pipe collection network be installed over the
liner to collect leachate to prevent ponding and surface runoff.
Another function of overdrains is to relieve excessive head
buildup and pressure which could result in rupture of the liner.
An underdrain system may also be required to collect ground
water, preventing pressure buildup and possible rupture of the
liner in the event of a rising water table. Collection piping
placed beneath liners can also facilitate detection and collec-
eion of leachate leakage through a ruptured liner, thus
jnimizing downward percolation of leachate which could pollute
the ground water.
A quality assurance program is an important aspect of any
liner installation program. Such a program will include inspec-
tion of integrity of seams, proper liner storage methods, wind
damage, repair procedures and other activities. Considerations
of methods to protect liner edges are especially important if
the facility is to be expanded.
In those cases where materials such as asphalt, clay, or
bentonite are selected for liners, instatliation procedures will
deviate from the installation of a synthetic sheet membrane.
Three recommended construction layers for asphalt, clay, and
bentOflite are listed in Table 56. Specific installation pro-
cedures are available from the manufacturers. (Asphalt
Institute, 1966; Asphalt Institute, 1976; American Colloid Co.,
1979). The applicability and effectiveness of asphalt sealers
are discussed in detail in the following section. -
Tables 57 and 58 list the updated 1981 installation costs
of various liner materials (based on 1977 costs given by
frlatrecon, Inc. (1980). Of the polymeric membranes——polyethylene
(PE) and polyvinyl chloride (PVC) are the least extensive
liner materials. Of the soil, admix materials, anä asphalt
membrane liners-—clays, soil + bentonite, soil cements, soil
asphalts , and spray—on asphalts are the least expensive to
jnstall.
A cost range for ground-water control measures is presented
in Table S9which also su narizes optimum site conditions, and
limitations for the employment of artifical barriers.
Monitoring of liner performance is difficult although a
ni. ber of approaches have been commonly employed. These have
included piping laid beneath the liner and underdrains consisting
of clay, asphalt, and plastic. These collection systems facil-
itate detection and collection of leachate passing through a
ruptured liner. The collection system is generally drained to a
si P or a series of sumps from which the leachate is withdrawn.
241
-------
TABLE 56. SYNTHETIC LAYERS FOR ASPHALT, CLAY, AND BENTONITE (Asphalt Institute, 1966;
Middlebrooks, et al., 1978)
*
________ 24—48° of graded, compacted
soil.
Liner
2° of hot rolled asphalt.
24” of compacted,
impervious material
placed in one 12”
layer and two 6”
layers.
Mix bentonite with
clean soil in
desired ratio.
Disc to depth of
2—3” (no sheeps
foot allowed).
12° of 3/4° diameter crushed
rock.
6° of 1/4—3/4” diameter
crushed rock.
6° of rock dust or fine
sand.
12” of 3/4” diameter
crushed rock.
6” of 1/4” or 3/8”
diameter crushed rock.
6” of rockdust or tine
sand.
12” of compacted
imported earth.
Subgrade
Asphalt
c y Bentonite
24—48” compacted soil..12° of clean soil.
Cover
*
1 in. = 2.5 cm
-------
TABLE 57. COSTS OF FLEXIBLE ?CLY ERIC M BR NE, ?L.STC,
RUBBER LINERS, 1981 (after Matrecon, Inc. 1980)
Material
Nominal 1981 Installed Costs
Thickness $/yd 2 *
mu s
Dutyl rubber
30
4.90 — 5.20
Chlorinated polyethylene
(CPE)
ChlOrOSUlfOflated poly—
ethylene (CSPE)
30
30
4.50 - 5.00
4.50 — 5.00
Elasticized polyolefin
(ELPO)
20
3.00 — 3.25
.
Ethylene propylene rubber
(EPDM) covered
uncovered
30
30
4.90 - 5.00
4.15
Neoprene (CR)
30
8.40 — 9.70
polyethylene (PE)
10
1.60 — 2.00
polyvinyl chloride (PVC)
20
30
2.10 - 2.85
2.60 — 3.50
•
* $l.00/yd 2 = $1.20/n i 2
Because of the possibility of liner leakage, placement of a
small piece of the installed liner (coupon) in the fill may be
used to evaluate the effects of the waste on the liner material
properties over time intervals. Coupon placement should allow
for essentially the same exposure to the waste and exiviror menta1
conditionS as the installed liner.
Bottom sealers——A bottom seal is a method by which a sealing
material is applied on the bottom of a new landfill site or a
pressu.reinjected grout is pumped under an existing site. A
variety of materials, such as asphalt concrete, soil asphalt,
soil cement, sprayed asphalt membranes, and bituminous seals,
may be utilized for sealing purposes, depending upon soil prop-
erties (Haxo, 1977a; Tolman, et al., 1978; Middlebrooks, et al.,
1978; Matrecon, Inc., 1980). This approach requires a detailed
flowledge of both landfill boundaries and soil properties; ex-
tensive exploratory corings and soil testings are therefore
required with this method. Literature describing bottom sealing
243
-------
is extensive due to its use in construction (Xarcl, 1968;
Lenahazi, 1973; Halli urton Services, 1976; American Cyana’nid Co.,
1975).
TABLE 58. COST ESTINATES OF SOIL, ADMIX LATERIALS AND ASPHALT
MBPANE LINERS, 1981 (after Matracon, Inc. 1980) —
Liner Type 1981 Installed Costs $/yd 2 *
Compacted clay ,2-ft thick 2.30
Soil + bentonite 6—th. thick 1.43
9 lb/sq yd
Soil cements 6-in, thick + sealer 2.50
(2 coats — each 0.25 gal/sc yd)
Soil asphalt, 6—in, thick + 2.50
sealer (2 coats — each 0.25
gal/sq yd)
Asphalt cortcrete, dense - graded 4.65 - 6.45
paving with sealer coat (hot
mix, 4 - in. thick)
Asphalt concrete, hydraulic (hot 6.00 - 8.30
mix, 4-in, thick)
Bituminous seal (catalytically 3.00 - 4.00
blown asphalt) 1 gal/sq yd (with earth cover)
Asphalt emulsion on mat (poly- 2.50 - 3.50
propylene mat sprayed with
asphalt emulsion)
* $1.00/yd 2 — $1.20/rn 2
Installation costs for bottom sealers vary according to Soil
porosity. For pressure—injected sealing, a 1.2—itt (4-ft) thick
portland cement sealer for a 4-hectacre (10-acre) landfill with
a porosity of 20% may range in cost from $1,610,000 to
$4,010,000; at 30% porosity, from $3,470,000 to $6,020,000 in
1981 dollars. For the same area, a 1.8—rn (6—ft) thick liner
(20% porosity) may range in cost from $2,400,000 to $6,000,000;
at 30% porosity, $3,600,000 to $9,000,000 in 1981 dollars (after
Tolman et al., 1978).
Bottom sealing practices offer advantages over other alter...
natives (primarily excavation and re—burial) especially j coarse
grained soils and gravels. The seal can prevent ground water
244
-------
TAbLB I9.i OB?8 0 GIIOUHDW&TBII cQKTBOb $* *8URtS, 1981 4sttur ?o laae. g J,. 191 1i KgL(ncon• Ino. 19161
Opti.ua Bit. Conditiun .
Ground water near to the
retu.ii/ioll interface.
Underlying .otI poanu..
£ aoderata)y tepid to
very rapid perneablilty
1.1 iir 3 rjau/aeo to
1.4 a l0 ca/ n e c.
8. . . a. f ur art*tkciiil
liner and whoa a detailed
understanding of the aofl
ii know..
Near to surface ground
watCr end bedrock con-
dub.. Other conditions
aj.llar to tho.e tot an
artitiolal linac.
Gtout Curtain Baa. a. for arti(lol.l
liner as well a. a near
to aurfice bedrock
condition.
Bbu .t PiUng Bame a. tot grout curtain.
Cutoff U 1l
1981 lnntau latluu Co tij
41.80 to $9.10 put y4 2 ’
l.ia ltatiou n
Pequtras *,i .isilatiou prior
to lendtiliing. Lapacted
lila not sitobhiahad. Ii . —
109 ground w 5tet nay ruptura
th. liner, flay be danaqed
during )andtIlling
activitie.. Once installed.
eatreciuly difficult to ra—
pair.
Detailed undet .tandtwj of the $1.44 to $8.30 per y4 2
nail properties ar, required. (Bun ?able tar deaign
htuthods of deLeisI tng that thickneaa)
all voids have been effectively
grouted are hOt available.
Lcachate say have a deleterious
effect on grout integrity.
Coat. of shipping bunto.iit .
Lacavation difficult due to
soil condi Lion.. Deteriora-
tion of beotonita when
eaposed to high .ta ength
leachatü soy occur.
Corroaton potential .nd pro-
Lie., with driving the cheat
through rocky nulls, ability
to ..jlntain integrity of I ii.
pi ling.
Perceived i ftuct$vu uvs
Bttectiv. when used an a
leachatu ol lect ion
method or as a sean. (or
isolating loachiete with-
IA the fill and keeping
ground w*ter out of the
fill.
Beau a. for attiflulol
liner..
&tt.cttve uu4ur jtouiw1-
water barrier uaud to
divert ground-watec
flow around a landfill
site or ispedo the
Lorisomital aubaurtace
aoveaent of ieachatu
t aos the till.
I’urtorae aisilar func-
tion ci slurry trench.
Co i1d r 4 uftectivu In
soil, with pyrseabil-
itica of lo ca/ nec
Or greater.
BttuaLLvo ç ntrol ..e-
aura whaun a deep
vertical harrier La
required to retard time
flow of ground water
Qwier aol through a
landfill (no axcavelion
is required).
$l.00/yJ — $1.30/a 3
1 ft —0.31 a
Lineal ft coat. laaad on a I II a 41300 tt l lung ntru.ature. IS a (80 ft$ deep.
Conttol ?ectuilquu
Srtittctal Liner
Blurry Trench
I’ )
to 41*0 per lineal
ft. ’
Co,t. any he prohibitive
for large soda tale. 01 (1*-
cult La asses. the integrity
of the seal.
$880 to $1890 per lineal
ft.
$SS0 to 4620 per lineal
ft.
-------
from entering the landfill as well as impounding leachates formed
within the fill. Disadvantages cited for bottom sealing include
problems associated with exoloratory drilling, over-design to
compensate for the erratic nature of filling, and the uncertainty
of determining whether a complete seal has been formed (Tolman,
et al., 1978).
A wide variety of admix and asphalt sealing materials may
be used as bottom sealers. Primary advantages of asphalt
materials include universal availability, versatility in avail-
able forms, low cost, and the fact that asphalt materials are
some of the most logical engineering materials available for
large scale installations. Utilizing pure asphalt in membrane
form has proven effective in seepage control and cost effective-
ness. The following are examples of commonly used bottom
sealers.
Asphalt concrete is a mixture of asphalt cement and graded
aggregate that is placed and compacted at elevated temperatures.
This sealer is well developed for use in construction of lining
for all types of hydraulic structures. When properly mixed and
applied, asphalt concrete forms a durable, stable, and erosion—
resistant lining. Mix design of asphalt cement should f l1ow
guidelines set by the Asphalt tnstitute (1974). The expense for
this type of membrane is much higher than natural clay liners
(Table 58) ,but is comparable with synthetic types (Table 57).
Soil asphalt embrace a wide variety of soils, usually those
of low plasticity, mixed with a liquid asphalt. Soil as hajts
containing emulsified asphalts usually require a waterproofing
seal, membrane, or asphalt cement, placed on top of them. Other
types of asphalt have been used with soils to provide an effective
soil asphalt which is more flexible and crack resistant than
soil cement or asphalt cement, and more resistant. to aging, as
well (Matrecon, tnc., 1980). Soil asphalt is a relatively
inexpensive bottom sealer (Table 58).
Soil cement is prepared by mixing portland cement, water,
and a variety of soils. As the cement hydrates, the mixture
hardens resulting in an impermeable liner. The permeability of
the sealer is dependent upon the nature of soil used in the mix-
ture. The more granular the soil, the higher the permeability.
Usually, surface sealants are applied to soil cement to impart
waterproof capabilities. Aging and weathering characteristics
are usually good, especially those associated with the wet-dry
and freeze-thaw cycles. Some degradation of the cement is
expected in acidic environments (Stewart, 1978). Costs are
comparable to soil asphalt (see Table 58).
Sprayed asphalt membtane sealer (hot-sprayed type) usually
consists of a continuous layer of asphalt, often without filler
246
-------
or reinforcement, covered or buried to provide protection from
weathering or mechanical damage. Cover soils may include native
soil, gravel, or asphalt macadam. These membranes are then
dried into tough, pliable sheets that conform readily to soil
contour alterations. Useful lives of 15 years or greater have
been observed when these membranes are carefully applied and
covered with an adequate layer of fine grained soil (Middlebrooks,
et al., 1978). Sprayed asphalt membranes are one of the least
sive types of current bottom sealers (see Table 58).
Bituminous seals are most often used to sea], the pores of
an asphalt mixture serving as a liner in order to provide water-
proofing. There are basically two types of bituminous seals.
One is called catalytically-blown asphalt since it is produced
by means of air-blowing the asphalt in the presence of a catalyst
(usuallY ferric chloride) which produces a brownish tint. The
asphalt produced has a high softening point, yet remains flexible,
tough,and impervious to water. It is resistant to acids, but
not oils. Aging resistance is also good (Lutton, et al., 1979).
The second type of butiminous seal is the fabric plus asphalt
emulsion type. These form continuous filins of asphalt after
breaking of the emulsion and the evaporation of the water. These
films are less tough than filr .s of the aforementioned type;
therefore, toughness is achieved by spraying the asphalt on
supportive fabric such as glass fiber, Woven jute, etc. (Geswein,
et al., 1978). The combination has been used quite successfully
j ’lining for ponds. The cost is relatively low for this type
of lining (see Table 58).
Slurry—trench-—A slurry-trench is an underground water
barrier which may be used to prevent the horizontal subsurface
movement of leachate from a landfill or to impede the movement
of water into the fill (Tolman, et al., 1978). The slurry-
trench cutoff wall may isolate leachate from a landfill by
providing a complete seal extending to an impermeable liner or
by increasing the up-gradient length of the ground-water
flow path. The slurry-trench may also be utilized as a dawn-
gradient barrier from the landfill, in which case, leachate
impounded by the barrier may be collected for recycling and/or
treatment. Slurry-trench technology has been developed for use
in the construction industry and considerable literature exists
sCribing this application al with sanitary landfill appli-
cations (Seaman, 1966; Gerwick, 1967; Kapp, 1969).
Tolman., et al. (1978) developed estimated costs for slurry—
renching. A completed slurry—trench 18 in (60 ft) deep and 1 in
(3 ft) wide will range from $1,380 to $2,330/rn ($4.20 to $7.10!
lineal ft) based upon 1981 costs. To construct a semicircular
cutoff wall 518 in (1,700 ft) long around the up-gradient end of
a 4-heCtacre (10—acre) landfill would cost between $720,000 and
$1,200,000 (1981 dollars). Differences in price depend on wheth-
er the excavation is soil or rock and the type of bentonite
247
-------
slurry used. Emrich, et al. (1980) proposed a slurry-trench cut-
off wall 945 in (3100ft) long and. 44 in (143 ft) d p t help
protect ground waters from a sanitary landfill in Windhaxn,
Connecticut. A cost range of $3.7 to $5.6 million was proPosed
for the 10—ha (25—acre) site (1981 dollars).
Advantages of slurry-trenches include relatively simple
construction methods, availability of leachate-resistant
bentonites, low maintenance, and minimal liner deterioration.
Problems associated with slurry-trenches include the cost of
shipping bentonite, over-excavation in rocky ground due to boul-
ders, and the possible deterioration of certain bentonites when
exposed to high ionic strength leachates. Slurry-trenches are
most effective in those cases in which a shallow ground-water
condition exists and a impermeable soil or bedrock strata is
available for contact with the barrier (Tolman, et al., 1978).
Grout curtain——A grout curtain performs a similar function
as that of a slurry—trench in providing a vertical barrier to
either divert ground—Water flow around a landfill site or inter-
cept migrating leachate. A grout curtain is created by injecting
solutions under pressure into soils and underlying earth mate-
rials. The grout soi.ution fills soil voids preventing the
passage of leachates or groundwater flow (Tolman, et al., 1978).
Grout materials include cement, bentonite, epoxy resins,
silicone rubbers, lime, fly ash and bituminous compounds
(Raymond International, Inc. 1967; Tolman, et al., 1978). The
use of a particular grout material is a function of soil
characteristics encountered at the facility. American Cyariamjd
Co. has categorized the use of chemical grouts with various
soil properties (American Cyanaznid Co., 1975).
The variety of grout materials available in a particular
situation will result in a disparity of Costs. The relative
costs of grouts in comparison to portland cement are listed j
Table 60. Tolman, et al., 1978 developed estimated costs for
portland cement grout curtains. An installed grout curtain
18 m (60 ft) deep and 1.5 in (5 ft) wide would cost from $2,230
to $5,540/rn ($680 to $1,690/lineal ft) based on 1981 dollars.
Estimated costs for a 518-rn (1700—lineal ft) portland cement
grout curtain up-gradient of a 4—hectare (10-acre) landfill to
a depth of 18 in (60 ft) would cost between $1,150,000 and
$2,880,000 (1981 dollars).
Grouting technology is well developed and is applicable to
a wide range of soil properties. However, grouting is usually
considered effective only in soils with permneabilities of l0
cm/sec or greater (Tolman, et al., 1978). Grouting offers a use..
ful approach in consolidate or cohensionless soils although
costs may be prohibitive for large landfills. The use of grout
248
-------
curtains for diverting ground water around landfills or inter-
e ting leachates is undocintented; however, the principle
h uld be aoplicable for landfilling practices.
*
TABLE 60. RELATIVE COSTS OF GROUT (Xays, 1977)
y e of Grout
Portland Cement
Silicate Base—15%
Lignin Base
Silicate Base—30%
Silicate Base—40%
Urea Formaldehyde
Acrylamide CAin-a)
Resin
Basic Cost Figures
1.0
1.3
1.65
2.2
2.9
6.0
7.0
* Base unit = 1. t.Inder a given set of conditions, where
portland cement grouts costs 1.0 times S/unit, other types
of grout will cost the given figure times S/unit.
Sheet oiling cutoff wall——Sheet piling cutoff walls may
be used to retard the flow of water under and through a land-
fill. A sheet piling cutoff wall. consists of lengths of steel
sheet piling permanently driven into the ground. Sheet piling
is readily available in various shapes and weights, the
selection of which is based on costs and ease of installation.
The sheet piling is assembled before being driven into the
ground. The sheet piling wall is not watertight due to mill
tolerances in the interlocking edges. Cathodic protection
can prolong the useful life of the wall, which can range up to
40 years (Toitnan, et al., 1978).
Toi.man, et al. (1978) developed costs for a 518 ta (1700 ft)
long and 18 rn (60 ft) deep cutoff wall up—gradient of a 4-
hectacre (10-acre) landfill. Total costs to construct and
install a 20 ton sheet piling cutoff wall (including shipping
cOStS from Buffalo, New York to Philadelphia) are estimated
to be between $940,000 to $1,380,000 based on 1981 dollars.
installed costs range from $1,800 to $2,690/rn ($550 to $8201
lineal ft). The unit cost of the sheet metal plus shipping
ogts are major variables in any cost estimates for installed
sheet piling cutoff walls.
Sheet piling cutoff approaches offer several distinct
dvafltageS over slurry trenches and grout curtains. Among
these are ease of construction (no excavation is necessary),
availability of contractors to perform the work, maintenance
free nature of the wall, and the relative economical construc-
tion costs. Limitations of the sheet piling cutoff wall include
its corrosion potential, problems of driving the sheet through
249
-------
rocky ground, and that the sheet piling is not initially water-
tight (To] an, etal., 1978 ).
Leachate Pli ne Control-—
Leachate plume control refers to those efforts to prevent a
contaminated ground-water body from contacting additional ground-
water or surface water resources. Methods involve the removal
of water by pumping or drainage from around the landfill. These
approaches include drains, well point systems, and deep well
systems.
Plume control is similar in purpose to subsurface inf ii—
tration barriers by effecting changes in water table elevations,
and are often used in conjunction with subsurface barriers. The
control measures differ insofar as a continuing input of energy
is usually required for pumping water. Table 61 summarizes these
systems. These systems may be utilized either up—gradient or
down-gradient from a landfill.
Drains——The purpose of drains is to intercept up—gradient
ground water which can be redirected from the landfill to lower
the water table. They can also be used to collect ground water
and leachate down-gradient of the landfill. Drains are con-
structed by excavating a trench. to a desired depth, partially
backfilling the trench with highly permeable sand or gravel,
placing a plastic or ceramic drain tile in the sand and gravel
bed, and complete the backfilling. The drainage can be collect-
ed in an impermeable sump and. directed or pumped back to the
landfill.
Tolman, et al., 1978 developed estimated capital costs for
ground-water drainage by trenches or subgrade drains. 3ased on
1981 dollars, the costs would range from $1,780 to $2,490 oer
hectare ($720 to $1,000 per acre) or $11.30 to $15.90/rn ($3.40
to $4.85/lineal ft) , depending on the depth of placement and
materials used. Operating and maintenance costs are estimated
to be $2,230 per year or $3.48/rn ($1.06/lineal ft).
Drainage systems are especially useful in situations where
high ground-water conditions exist, providing a relatively
efficient control system in these cases. Disadvantages include
difficulties in trenching operations for coarse or rocky soils,
as well as problems associated with deep trenching.
Well point systems-—Well point systems can be placed up—
gradient from a landfill to prevent ground-water flow through
the fill. Well points may also be used for leachate collection
down-gradient of the fill, although care must be exercised to
prevent the system from accelerating the rate of leachate
migration from the fill.
250
-------
‘U
1 It — 0.31 a
Power coats * c1udud
?A L 6 . LEACI*ATt PUN4 c TWOL ToInanet ei•. 1$1 )
COHTUOL
?tCHN IQ1E
HUM
al$D)? OH2
L1MLTAT OH9
19 11 cowj!
PY$C IVk I)
£vI’!cfl_ ttwS8
Drain.
Kear—to .urface 9rOund
watar .oila which are
easily .sc avat.4 1 oonai .-
tent 9r0u04 water level
with little fluctuation.
Ground water aay be too
low for •tf ctive we. of
drain.. Ocila any not be
eseneble to .vcavation.
Perla.tor area required
for draInn9u nay be too
large an area tow
practical drainage pur—
poan ..
Cap$tal coaLl
$3.40 to $4.15 r i,.ual
5tI.ctlve when uuej
In conjunctioii with
ot*ier control ep—
proecie. audi a.
linern wwfl point..
etc.
$lO pur Ii per yr.
Well point
.y .t..
Iuep well
ayatea
&aas a. to w drum.
Water must, at greater
depth. tba a w l1 point
syatea can withdraw.
usually effective fur
hallow dewetering
purpoane ‘IO*I
Capgbl. of witlitrewing
approm .Iaat.Iy 40
vertical ft of water
in willows aanila.
CiftiLal coita
113i* 11 pr lineal ft.
I?iitat*n j a ll .aintenamce cosLa
$24 par iineal9ij ryr.
ft cIiv whun weul
to coiplinont another
cootrol eyeteuu e.g..
leechete treataunt).
Lower. the water
table In deeper
pplIcatione than a
well point syatow.
C j al cofle
22 pc 1Tna i ft.
eUftg Qima1ntanancd il!’
$22 per ineeil j
-------
; ell oints are short lengths of plastic (PVC) or telfon
well screen which can be placed in the saturated zone up— or
down-gradient frcm the fill. The well points are connected to
a pump which evacuates the air from well points creating a
vacuum which forces the ground water to flow through the well
points. Well point systems are usually effective for shallow
dewatering purposes up to 10 rn (32 ft), and may provide draw—
downs of up to 4.5 rn (15 ft) within the pumping zone of
influence (Nobel, 1963).
Based on costs reported by Tolinan, j. (1978), well point
systems are estimated to cost between $43 to $48/rn ($13 to $141
lineal drain—ft) for installation, plus $2,880 to $4,320 for a
19 to 24—liter/sec (300—400 gal/mm) pump, based on 1981 dollars.
To drain a 275—rn (900—liflealft) area up-gradient from a landfill
would cost approximately $17,300 (capital cost). This capital
cost, in addition to pumping tests, analyses and engineering
costs, could result in a $31,000 to $40,300 total capital ex en—
djture. Maintenance, power, and labor costs for this system are
estimated to be approximatelY $21,600 per year or $79/rn ($24/
lineal ft) cer year. Power costs represent the major expense
(i.e., $15,000/yr. based upon a 32 kw (24 ho) demand at $0.10,
kwh.
The advantages of a well point system are that installation
costs are moderate and the construction methods are relatively
simple. The disadvantage of this system is that the power costs
are high, continued maintenance is necessary, and a long term
financial commitment is required.
DeeO well systems DeeP well withdrawl refers to an instal—
latio in which a greater vertical depth of water can be removed
than can be handled by a vacuum-extracted well point system.
Deep wells are capable of lowering the water table as much as
12 m (40 ft) in uniform sand. Each well is usually equipped
with its own submersible pump and is, therefore, capable of
dewatering large areas.
Deep well dewatering systems generally require a higher
initial investment than well point systems, since they require
more materials, equipment, and creater installation costs. Based
on costs given by Tolman, et al. (1978), it is estimated that
deep well systems (1981 do l lars) would cost $72/rn ($22/lineal. ft)
for a l0— n (4-in) PVC well; a screen may cost $140 and a pump
$1,140 to $1440. Capital costs for a well system 8 m (60 ft)
deep to dewater a 275 m (900 ft) front is estimated to cost
approximately $25,900. Maintenance, power, and labor costs for
this system are estimated to be approximately $20,200 per year
or $73/rn ($22/lineal ft) per year. Power costs represent the
major expense (i.e., $15,000/yr based upon a 32 kw(24hp) demand
at $0.10/kwh).
252
-------
Emrich,et al. (1980) proposed a deep well dawataring system
to lower the water table surrounding a sanitary landfill. Esti-
mated capital costs for a ring of 80 wells, 19.5 in (65 ft) deep,
generating 25.1 1/sec (400 gprn) are $198,000 in 1981 dollars;
operation and maintenance costs for such a system are estimated
at $17,000 to $19,000 per year.
Surface Water Control
Surface water control is an approach by which infiltration
of water entering the landfill is reduced or prevented altogeth-
er. 3 utilizing control measures to Lfliniinize the quantity of
water entering the landfill via rainfall or runon, leachate
generation can be reduced. This section presents three methods
of reducing infiltration:
• Contour grading and surface water diversion;
• Surface sealing; and
• evegetatiOn.
Contour Grading and Surface Water Diversion--
Grading and compacting the landfill to a profile of a
of 12 percent and a minimum of 5 percent will allow
g .irface water to dram from the site and will minimize jrtflj.,-
tratiofl (Tolznan,et al., 1978). Most soils wil remain stable
at these grades; however, slopes greater than 12 percent may
accele te erosion of the landfill, depending on the soil type
.nd extent of vegetative cover (U.S. Dept. of Agriculture, 1975).
IdeallYs to minimize overland routing areas subject to infii.-
tration, the center of the landfill should be the highest
e1evati0 Exact slopes to be used should be designed on the
basis of soil type and slope stability at the site (Tolman,
al. , 1978)
Contour grading operations are simple and economical. Costs
fork contour grading are affected by the size of the landfill,
av’ailabilitY of cover material, and the percent of slope. Based
ofl costs developed by Tolman, et al. (1978), the total cost of
00 ntour grading a 4—hectacre (10—acre) landfill plus a 0.6 m
(2 ft) minixm.un thickness of soil cover would range from $181,000
$349,000 (1981 dollars).
Contour grading is a coimnon practice for sanitary landfills.
znrich, et al. (1980) studied seven possible remedial actions
o minimize or abate pollution from a closed landfill in
Connecticut. It was concluded that contour grading, in con-
junction with diversion ditches, a PVC membrane seal, and a
253
-------
vegetative cover, would be the most affective and economical
control measures. Monitoring of the landfill will continue
for two years following the installation of the remedial action
alternatives to deterriine their effectiveness.
Diversion ditches are usually constructed around the peri-
meter of a site for diversion of upland rainfall, as well as for
re—routing water from spring-heads. Costs for diversion ditches
vary according to ditch configuratiOn, placement of material
within the ditch, and the use of a liner material (e.g., plastic,
concrete culvert, etc.). Diversion ditches are generally
designed for a 10-year storm intensity (Tol.man, et at., 1978).
Figure 46 presents some design specifications for surface and
interceptor ditches.
tpdated 1981 cost estimates based on Tolman at al. (1978),
for a 0.6 m (2 ft) deep and a 1 ra (3 ft) wide trench range from
$54.00 to $86.40 per ra ($16.30 to $26.10 per lineal ft).
Included in this figure is an assumed cost of $11.40 to $16. 9O/r
($8.60 to $13.00/yd ) for trench excavation.
Diversion ditches and surface contouring represent two
mitigations which are extremely cost effective control approaches.
Table 62 summarizes these methods and other surface water
control measures for a 4-hectacre (10—acre) landfill.
Surface Sealing——
Surface sealing is a method of landfill closure y which
an impermea.ble material is used to cap or seal a landfill,
preventing water infiltration and minimizing leachate generation.
Caps and seals can be constructed of clays, fly ash, soil—cement,
lime-stabilized soil, membrane liners, bituminous concrete, and
asphalt-tar materials. Membrane and asphalt compounds may
require an earth cover to protect the cover seal and to provide
support (Toirnan, at , 1978). In addition to the above—
mentioned materials, 114 chemical additives which may be mixed
with a soil cover for various beneficial effects (e.g., stabili-
zation, repel water or resist water erosion, or resist dusting
or wind erosion) are presented (Lutton, at al., 1979).
Lutton, et al. (1979) provides a four step process for
selection an design of cover as well as the functions which
a cover provides. Specific functions of cover on waste disposal
sites include the following:
• Control insect emergence and entrance;
• Control rodent burrowing;
• Reduce bird and animal attraction;
254
-------
if Ciplol. DItch
Rock RI rap
Cloy Concrit.
Lumber Asphalt Ripropped Ditch
SURFACE DITCH LININGS
Figure 46.
Design specifications for surface and interceptor ditches (Lutton, et al. ,
1979).
Sod Lining
SURFACE a INTERCEPTOR DITCHES
Ui
Sodded DItch
2’ Dllch Where
Needed On Long
Bock Slopes.
Conslruci Check Darn When
Velocity Is Gr.ol Enough To
Couse Scouring
-------
t )
U’
0
S
Coats aru based on a 4-b.ctire lie-acre) situ
6 2 • 8URVAC IJATEI 1 NtkOL jyolaan , ti. • I97 )
Control ?echntquus
Opt& Site Conditione
Lisitattone
1981 CouLd
Percisived ttucLivaiiea*
Contour Grading
Surtace blatar
and
OLvacafoa
Landfill atabilizuds sufti-
dent borrowarul for soil
excavation available nearby.
Available coil covur
sh u14 be prucunt.
1 1tterenttal auttle-
sent eay be uxtancivu,
a-uquirin9 l ryu volu .e.
f coil.
$l8i.0u0
$349,000
to
An sffuotiv* control
budSuC@ used both durin9
lanJttliing and once
the fifl he. Lu n
stablilaud.
Surface Sealing
•
borrow aterLal ava*lubiu
)andt*U bias beau stgliiliaud
and/or graded.
duttabto cover auturiat
aay not be readily
available. Gee vuistinij
•ay alcu be required.
8 O2.0UO
$829,000
to
Eftuctivu if &UtLLUNMOL
lii ainlial end when uo4
in coijunction with
other control cuasurus.
Kuvequtation
Stabilized ff1 and available
cover coil 4’ 3 ttl. b4inl.uiui
of auttlea.ont. and gas oonLrb l
equt aent available at the
situ.
Decoapoettiosi aeua say
he toxic to vegetation
it not vunte) or con-
trollud.
$10,500
$17,300
to
Pruvidea a leut(ur to
aot 5tute untuciny the
fill, end tuducui crouton.
Is souL effective when
eaployud In conjunction
with other control
suaxures.
-------
• Minimize water erosion;
• Minimize water infiltration;
• Minimize or £na cimize gas contaimnent;
• Minimize fire hazard potential;
• Minimize blowing paper;
• Control noxious odors;
• Provide sightly appearance;
• Minimize settlement and ntaximize compaction;
• Provide for vegetative growth;
• Minimize wind erosion and dust generation;
• Resist cold climate deterioration and operational
difficulties;
• Preserve slope stability; and
• Resist cracking.
Emrich, et a].. (1980) reported the installation of a 20 mu
p c rnem.brane seal for a sanitary landfill which required 10 cm,
(4 in.) of fine grained sand as a subgrade. A cover for the mem-
brane consisted of a 46 cm—(6-in.) layer of sand and gravel
placed over the membrane. Lutton, et a].. (1979) describes the
dvafltageS of several cover materials T yered over a membrane
which is not obtained with a single material. A vertical
cross—section of such an approach is as follows: (1) refuse,
(2) sand (buffer), (3) silt (filter), (4) clay (barrier), and
(5) a loam. The cover profile for a particular site may vary
according to the climatic conditions, availability of materials,
esigfl considerations, etc.
Costs for sealing depends upon the amount of earth-moving
and grading as well as th selection of seal materials. To]man,
et al. (1979) developed costs for landfill covers for a variety
? al material for a 4-ha (10-acre) site. These costs (1981
dollars) are listed in Table 63.
Surface seals offer a number of advantages in surface
water control. Among these benefits are: (1) surface seal
placement is economical and is relatively simple to install,
(2) many of the cover materials can withstand minor settlement,
and (3) some covers such as soil—cement and bit inous avings
do not require soil covers and long service life is expected.
257
-------
TABLE 63. COSTS OF SUR3’ACE SEALS (Lutton, et al, 1979)
Material
Thickness
in.
1981 Cost
Range
Clay Cap
15
6
201,700
—
36a,900
Clay Cap
46
13
259,800
473,300
Bituminous Concrete
Cap
4
1.5
277,200
—
490,800
Bituminous Concrete
Cap
13
5
416,600
—
630,100
Fly—Ash Cap
30
12
196,000
356,100
Fly—Ash Cap
60
24
258,600
—
481,500
Soil—Cement Cap
13
5
301,600
462,800
Li 1e—StabiliZed Cap
30
5
301,600
—
462,800
PVC Membrane Cap
30
mU.
0.030
559,900
—
828,700
* Cost estimates reflect land. preparation costs for cover place—
ment, including:
• excavation of 26,700 tn 3 (34,950 yd 3 ) of common borrow
material;
• excavation of 7,700 (10,000 yd 3 ) of the waste during
grading operations; and
• eartbmoviflg operatiOflS grading and con oacting materials
equipment and procedures for constructing each seal
(estimated at $1440 per acre).
Costs are based on a 4—hectacre (10—acre) site.
258
-------
problems associated with cover seals include the unavailability
of certain cover material, such as natural clay deposits; large
quantities of borrow material may be recuired; and gas venting
is required with all surface seals.
Revege tat ion——
Revegetation provides an inexpensive and cost—effective
means of stabilizing the landfill cover, reducing infiltration,
and help minimize erosion by wind and water. Revegetation can
also reduce water runoff resulting in more water entering the
soil. This increased infiltration is offset at least partly by
transpiration from vegetation, through there is considerable
question with regard to the relative importance of these off-
setting processes (Rovers and Faquhar, 1973; Motz, 1974). A
subsidiary benefit is that revegetation enhances the appearance
of the landfill site. A cover layer of suitable soil at least
0.6 in (2 ft) thick is usually required for rooting purposes
(Tolznan, 1978).
The selection of the best revegetative cover for a parti-
ctj.lar site depends upon soil type, nutrient, pM levels climate,
and knowledge of plants that have the desired characteristics.
If soils do not possess sufficient nutrients to sup ort vigorous
plant growth, sludge might be used as a soil conditioner.
Bennett and Donahue (1975) and Lutton,.et al.(1979) provide
comprehensive descriptions of revegetation practices. Many
species of grasses and legumes are used for revegetation.
Commonly used species are described in Table 64.
One of the disadvantages of revegetation is that an imper-
vious clay or plastic barrier incorporated in the cover makes
the plant—root zone susceptible to Swamping after moderate
rains since verticle drainage is impeded (Lutton, et al, 1979).
Irrigation ‘nay be necessary during prolonged dry spells to
prevent loss of plant cover. Also, revegetation may be damaged
by 1 ndf ill gases which continue to vent for years after site
closure (Flower, et al, 1977). Tolinan, et al. (1979) prepared
cOStS for vegetating a completed fill. Based upon a slope of
less than 12% and a number of seeding and cover soil preparation
asgtitItPtiOflS, costs would range from $0.17 to $0.26 per in 2
($0.15 to $0.22 per yd2) in 1981 dollars. Nuiching with straw
or hay would 2 cost between $0.09 and $0.17 per in 2 ($0.07 and
$0.15 per yd ). From these cost estimates, a cost range of
i0,5OO to $17,300 for a 4—ha (10—acre) landfill was derived
(1981 dollars)
LEAC AD CO OSITION CONTROL
The major objective of leachate composition control is to
reduce the strength and contaminant flux of leachate. Leachate
comoosition is a function of many factors including those
inherent in the refuse mass and landfill location, and those
259
-------
6 s Gk&S S ANtI LIGt6*.S COIII4UIILY USED V0I4 kEvLc rA?lon Lutton , 19 O)
Annual ryu .jra .s
Timothy
Seed canarygrass
Alfalfa sally
vsri.tiosI
birdstoot tr.tolL
Sweet clover
Bed clover
Alaike clover
Boreen laspwla*ii
Se& ices lesiedeza
hairy votch
Whitu clover
Crarnv.tch
Strong. rhtso sgtous roots, puruiu ai
Long .iived UgbHL 5L
Annual, fibrone root., winter rapid growth
Alkaline soil., rapid grower, perennial
Slow to esteblish, l ng-llve4 perennial, good seeder
Sssfler tliaii tall, auscuptililu to leaf rust
More host tolerant, but less cold resistant than
uaooth brostzgrias or Kentucky bluegre*im
Not whltur hardy, poor dry land grass
Shallow rootS, bunch grass
Tell coarse, .o4 former, perunniol, resIsts
flooding and drouqht
Good on alkaline loam, requires good eanegesunt
Good on infertile soils, tolerant to acid soils
Good pioneer on non-acid soils
Plot droughst resistant, tolerant to sold soils
bistlar to red clover
Wet, acid soils, wars season
Daimp, cool sts.r., drought tusisIanL
Coi-nbelt eastward
North, busid, U.S. south to Tuflhius uu
Wi luly adapted, deap soil.
Cool to wers region., widuly adapted
Tosperatu U.S.
Moist southern U.S.
Not thorn U.S., cool, hussid aroma
Norhturn U.S., wet, cool arose
Widely adapted
Hoist, tusgiurat. U.S.
Widely a.t ptod
Cool, moist areas
Cool, aufut areas
Soulhurii U.S.
Southorn 1 1.9.
All of U.S.
All of U.S.
Northiera U.S.
.
Bust
variety
Seeding
Ti.. imiportant CharacteristIcs Area/Ceiidltionsot Adaptation
Grasses
ai p bentgras .
Sacotli brcs.yra$s
lield broae’jra .5
KentuCky blu .gra sa
Tall fescue
Meadow foscue
Orchard grass
0
rail
Spring
Spring
Vail
Vsl I
fall
Spring
Vail
Vail
Late
Su.aset
Late
Sprt q
Spring
Early
Spring
Sac ly
spring
far ly
sprIn j
Early
Spring
fall
Early
l ’s II
Early
raIl
Annual, widely adapted
Perennial, tell sect plsnt, widely adapted
Winter annual, survives below 0V, widely adapted
World-wide, many varietIes, doe, well on autst, acid
soils
Perenalal, creeping stalls and rhiiueue, acid tolerant
-------
created by landfill construction and operational features.
The chemical and biological properties of refuse are largely
uncontrollable; however, the codisposa2 . of sewage sludge or
industrial Wastes with municipal refuse may increase refuse
stabilization. Available leachate composition control tech-
niques include the following:
• Landfill construction and operational features;
• Leachate recirculation;
• Addition of municipal sludge or industrial wastes;
and
• Addition of selected sorbertts.
Landfill Construction and Operational Features
Landfill and operational features which are significant
in jnfluencing leachate composition are the physical character—
j 5 ticS of the refuse, including particle size (shredding) and
density (compaction and baling); rate of water application
controllable through the permeability of cover soils or
synthetic membranes; and landfill depth or lift height.
Simulated landfill studies by Hentrich,. at al. (1979) and
ernper and Smith (1981) indicate that shredding and baling of
refuse have significant effects on contaminant flux and
stabilization rates which are amenable to the following control
rneasures
• Baled refuse may be employed to enhance leachate
production both by decreasing the time period
before continuous production of leachate, and by
increasing the overall volume of leachate produced.
Baling results in large volumes of dilute leachate
with a long period of time required for stabili-
zation of the landfill;
• Shredding may be employed to promote rapid landfill
stabilization through the rapid generation of
concentrated leachate; and
• Shredding and baling both may be employed when the
slowest possible rate of leaching and longest
period of landfill stabilization are desired.
The amount of leachate generated in a landfill is directly
related to the amount of water that infiltrates the fill. In
generali the rates of biological and chemical stabilization in
a landfill increase with the amount of moisture present until
an upper limit is reached. Leckie and Pacey (1975; and Pohla.nd
261
-------
arid Gould (1980) have established that water application tec
niques are advantageous to accelerated landfill stabilization,
i.e., leachate production. Depending on whether rapid or
gradual landfill stabilization is desired, surface water
infiltration may be increased by the installation of a highly
permeable cover soil; or reduced by lartdf ill cOfltoU grading,
diversion ditches, and/or relatively iin errrteable surface seals.
Stabilization rates generally increase with tem erature
until an upper limit is reached. Temperature is usually not
a restricting factor in northern latitudes since the interior
of the refuse mass is insulated from the ambient temperature
and the degradation process is exothermic in character (Zononj
1972). tt is concluded that regulation of the refuse tempera—’
ture to decrease or increase stabilization rates would not be
an easily imolementad control technique. Available moisture
then beccmes the limiting factor.
The influence of refuse depth on leachate composition
is a factor of the effective contact time between percolate
and refuse. Co1’ mtn tests by Fuxigaroli, et al. (1979) showed
that upper levels of refuse stabilize earlier than deeper levels
This behavior indicates that shallow landfills will reach
stabilization much earlier than deeper ones; consequently, the
release of contaminants will be distributed over a longer period
for deeper fills (QasiuZl and Burchinal, 1970). These studies
indicate that the regulation of landfill depth may be a viable
method for controlling landfill stabilizati0n rates; however,
the economics of this approach has riot been established.
Two distinct refuse stabilization rates, resulting in two
distinct kinds of leachate and contaminant flux regimes are
recognized through manipulation of the above—mentioned construc_.
tion variables: (1) rapid refuse stabilization, conducive to
rapid contaminant extraction per unit time arid, ultimately,
a relatively dilute leachate if adequate soil attenuation
capacity is available, and (2) gradual refuse stabilization., co
ducive to suppressed contaminant release per unit time and
maintenance of concentrated leachate for extended periods. Each
of these two conditions and corresponding control measures
governing the appropriate leachate response are presented in
Figures 47 and 48. The leaching effects in each case are
realized only after a landfill has attained field capacity.
Leachates from landfills which have not reached field capacity
are largely controlled by conditions of permeability (including
channeling effects) which can greatly alter leachate trends.
Two points of view are expressed in the literature as to
whether a landfill site should be designed for gradual or rapid
stabilization. One viewpoint is that if the stabilization pro-
cess is retraded by restricting the infiltration of water, more
time will be allowed for the natural attenuation process to take
262
-------
Cover Soil :
- Highly permeable
Refuse:
Kept very moist,
with leachate recycle,
or water addition;
Small particle size
(shredded); loosely
packed and non—baled;
shallow depth or
short lifts
Underlying Condition:
Highly Impermeable
liner desired for
1 eachate coll ect ion
and recycle; other-
wise highly permeable
soil.
Figure 47. Landfill design and operational features promoting rapid landfill
stabilization.
263
-------
.,.:
-
— —
— —
.!
y..
Figure 48. Landfill design and operational features promoting gradual landfill
stabilization.
Cover’ Soil
Highly impermeabi e
liner or clay
Refuse:
Kept dry; unshredded;
highly compacted or
baled; deep or high
lifts.
underlying Condition:
Soil of low perme
.ability
264
-------
place. For this scenario, it is argued that a combination of
retarded degradation in a geologic formation of adacuate
attenuation provides the least likelihood of serious ground-
water pollution. It should be noted that the gradual stabili-
zation approach may require Long term post-closure landfill
monitoring and management. Advocates of this approach further
declare that increasing the stabilization rate by maximizing
the infiltration of surface water imposes a greater pollutional
load in a shorter time span on the attenuation capability of the
geologic formation which may result in significant degradation
of the ground-water quality. It shcu.ld be noted that the same
potential pollutiona]. load is involved in both cases.
The other viewpoint expressed by some is that a landfill
should be designed for optimum stabilization by permitting
surface water to percolate through the fill. For this approach,
it may be necessary to collect and treat the leachate prior to
jscharge to a surface water body, or possibly, recycle the
leachate through the fill. After a reasonable degree of
stabilization has occurred, the leachates will no longer be
collected but allowed to pass into the surrounding soil. In
this way, it is argued that the stabilization process can be
controlled as desired and the possibility of fut e pollution
problems is reduced considerably. It is further argued that
large quantities of landfill refuse are potential pollutional
time bombs” which could exist for many years after landfill
closure. There is no question. that landfill sites can be
esigfled and operated to speed up the degradation process and
àollect and treat the leachate; however, the economics of this
approach may favor other disposal methods.
Leachate Recirculation
The stabilization of landfill refuse and the quality of
j.eachate are largely the result of biological processes.
Leachate recirculation essentially uses the landfill as a
generallY uncontrolled anaerobic digestor for effective anaero-
bic treatment of its own leachate (Leckie, et a)., 1975).
ReCirCU!atiofl systems offer several advantages in terms of
leachate control, including acceleration of landfill stabili-
zation and quality control of the leachate. Also, treatment
facilities are not usually required for collected leachate.
Cited disadvantages are the high capital and maintenance
COStS, above—grade systems often result in odor problems,
and the possible increase in leachate volume production result-
ing in increased handling costs (Tolman, et a).., 1978).
Three types of leachate recircu.lation systems are recogniz-
ed: spray irrigation, overland flow, and subgrade irrigation.
A summary of these systems and leachate collection systems
is given in Table 65.
265
-------
.155 • tl ’° W J Oi .— L1CT O LCG2W eroLmaa. j . •
as M __________
iML .w O,als .
iti.U I j Pi a*
Svst
ltd wstSt4.4$ f t.a Liz t s
ws, U1 ,su.eattoe weti.
3&ta wsnwiz
:t . - sS v tth1 . 1W.
tsM iv t1aiU.L1V wtt
.scMW t .ICt.2.W.
iUr$ .qe1 t . - —
gal r.y11 .I •I swl
1,ot* nw.. 5.4 —
_______ a.tizaA qrzviti
Ui ’. t. , .M. 1 .L ri’.e?.Laa
aM ltd
trwsW*zu’J s r.s.* 4..a4S4.
C .1at Oa 4.4 L.acaSC*
4j t s .y.WIa
ata far tt. t3.LL.
iha.L i’ t . 4zws* to . 4ttiist
t. ___ ..• __ ws . .4
tLa.
Adadt.t* if
LaMLtLL aM — wsSaf.
pe.zeLo..L isJy i.a 4zy
.t at -“ w, .sMt
ewM .tiala — -‘— .m
•4fsct .vsas*u. P taf
•ds1 iI. toL .arta9n wu .La
a..i’ .ISM p& .oq .a
- . 4.
We. adu z atII
L.arhaS. cDL .Cr .iR £
4tatzs s1J.W lyIt5.
r . t . 4. Ca .4ua s
s .tftSr toMitoftu
f ez ipt U i ’.
ror i’s S .f!sitti’ WI.
toqE .x. L .s**Liatiiu
prior to
atffWtLt to cusir ears
• iSb Iit to
P*aq sta U. S
.S OI .al.s:.. $ V
.ar.Lauss C l i. ef
..a4sts ,iqr.tioU tr io
‘a. tti.i_..
; zytt.. ai*
S4 4tt CS
T flLa s.t
ft/yr.
Pew to .ti
3L4. . Ofl . iae4J .
ft/yr.
; ai 1l. Ceuti
ft.
ewritiig 4.4
l
S’s
m7 , U i.. . ).
ft/yr.
P . 5 W iItI
ft
Oa.r.ti tg SM
. w %. iaecs
pe5W tottl
e ut .i eni
$ .443 so
$4.iSl Uas.J.
ft.
C.sr .tt.q SM
ft/yr.
aot i seats
so
31.4. 40/U.toIJ.
ft.
a5WWi ir 1.4
s a int l e S i e w
ft/yr.
ft/yr.
t it •
g iosSa —t
I Reproduced from
best available copy
eaos:st ?ws qws . L .t l t .ttaa. Lilt Costa Perc.tv4 U i cttvio. . .
.S5. t. a,txwsL .t
lt.arsd. zrt ezo
ti.aoz .to pipied aM
4,Latri ws.ie. evIl’
—I d.
Ssf sfts ttw L i trig
s .qtaRS 01 t. 1.oq ‘ as
u r riota. sews
a. t if C
is s . tiqflss;.
en
1r,m nt acoaws vt.s
.a.4 La YSUsa).. 4.4
ai.i .atia agitiaa.
us avor5..Lu.
rend.. 40 sUi’tt, .
La.MftLl IWE.U.lss.io 5
— 4ata.. T .uqsty
ara4 .cr 14 by ttiosst
condj;toas. r .sew.or
.Uiei.i’y is 1.... t ,
far ‘prey JrLqat*,ot
or . 5szLs . 4 1. . 5.
4i •tfat). s
toU.clta. a.S . 44 s Iu.
saId ii
w%th .yili 1.J .a
ir bOIWI i’s)..r$.
Prentd.. en •ffsqt).v
wst/o4 for soUarr ) .. .
of L..cnat. diwn—
prediws s if tie LaM
fLU.
266
-------
Spray irrigation is accomplished by periodically u ping
collected effluent through spray nozzles placed at 15- to 30—in
(50 tO 100 ft) intervals on the landfill. Spray irrigation has
the advantage that some leachata treatnei t may be effected
during spraying as a result of both aeration and infiltratior.
rough the cover soil.
Overland flow irrigation is an inexpensive well-developed
tecbni Ue by which trenches, spreading basins, or grated pipes
are used to distribute the effluent. Leachate is periodically
pumped into the distribution system and allowed to infiltrate
jnto the ground. c ith both spray irrigation and overland
jrrigatiOfl, an impermeably lined pond is necessary to provide
gtorage of the leachate between irrigation periods. o leachate
overland flow programs are reported in the literature, although
use of the treatment processes inherent in overland flow
irrigation, represent a viable means of treating recycled
leachate.
Su.bgrade irrigation has been postulated as a potential
means to promote uniform stabilization of the fill. In these
aporOaChe5 leachate is distributed throughout the fill through
smai.ler subgrade tile fields or pipes. Very large tile fields
must be subdivided into smaller units and each unit fed leachate
separately. Su bgrade irrigation may be used continuously and
has the advantage of avoiding local odor problems.
Based on capital costs given by Tolman. et al. (1978) it is
estimated that spray irrigation, overland flow, nd subqrade
j rigatiOn systems would Cost, espectively, $70.00 per in
($21.30 per lineal ft); $10.40 per in ($3.20 per lineal ft); and
$24.50 per in ($7.46 per lineal ft), based cm 1981 dollars.
operati0fl maintenance, and power COStS for spray irrigation
are estimated to be $63.40 per in ($19.30 per lineal ft) per year.
power costs for this system represent the major expense, i.e.,
$55.00 per in ($16.90 per lineal ft) per year. Operation and
maintenance costs for overland flow and subgrade irrigation
gystems are estimated to be $5.67 per rn/$1.73 per lineal ft) and
$3.51 per in ($1.07 per lineal ft) per year, respectively.
p nnual power costs are not available for overland flow and sub—
grade irrigation systems.
Addition of Munici al and Industrial Sludges
The effects of adding municipal and industrial sludges to
municipal refuse landfills are detailed in Section 4, entitled
LeaChate Coinoosition. Evidence suggests that certain additions
can be beneficial in accelerating the stabilization of a landfill
(i.e., the concentration of organic pollutants and organometalljc
species appearing in leachate are reduced in several months rath-
er than years). Research has indicated that receipt of certain
alicalime or nutrient bearing wastes may promote a similar
267
-------
leachate response. An approach utilizing natural and synthetic
materials as a bottom layer of a landfill has been demonstrated
as a leachate ccmpositi0fl control measure. These techniques
are suarized as follows.
Municipal Wastewa.tar Treatment Plant Sludge—-
Sewage sludge may be added to municipal refuse for several
reasons. The most common is to provide disposal by incorporat-
ing it with municipal refuse in a landfill. An auxilliarv
purpose of adding sewage sludge to refuse, however, may be to
change the degradation processes within the landfill. it is
the latter application which is of interest from the standpoint
of leachate compositiOfl control.
co—disposal of dewatered sewage sludge with municipal ref-
use is a common practice at multipurpose landfills. Studies
by Stone (1973), Encon Associates (1974), and Pohiand and Gould
(1980) have indicated that sewage sludge can act as a buffering
agent, thereby, aintaifl g a landfill pH at levels (> 6.0)
which can increase stabilization. Additionally, the sludge can
enhance refuse decomposition by possessing a high microbial
population and high moisture content. There is concern, however,
that addition of sewage sludge may result in pathogens, and toxic
organics and trace metals in the leachates and will pose a
threat to ground-water supplies.
One method of co—disposal involves the placement of refuse
at the top of the working face along with the sludge. The
two wastes are mixed thoroughly and spread thinly and compacted
upon the working face. Techniques have been developed to
dewater and solidify the semi-solid sludge for conversion into
a granular solid suitable for landfilling. High costs will
likely prevent the use of this co—disposal method.
Industrial Wastes——
A distinct lack of data is evident for the technologies
associated with landfill disposal of industrial wastes. Little
is known on what effect adding industrial sludge to municipal
refuse has on the stabilization process and the quantity and
quality of leachate produced during decomposition. There is a
stronq concern that addition of industrial wastes, particularly
those high in heavy metals, will result in elevated metal
concentrations in the leachates arid potentially, in potable
ground-water supplies. Advocates of co—disposal of sludges and
municipal waste believe the presence of organics in the landfill
will irmnobilize heavy metals. They also believe the presence
of such sludges may accelerate the decomposition process and
shorten the time required for biological stabilization of the
refuse (Shoemaker and Ritterthouse, 1980).
Leachate stabilization is usually accelerated in a landfill
envirormient in which the pH is low, the chemical environment has
268
-------
become reduced and a primary determinent of the preser.ce or
absence of various constituents will be their relative solubili-
ties in the leachate.medium. The availabi ity of many heavy
metal species (e.g., Zn 2 , Pb 2 , Cd ’ 2 , CU ) is likely governed
by the presence of sulfide, which in most cases, exists in con-
centrations sufficient for precipitation. t3nder reducing
c ndjtiOfls and an increasing pH consequenced by the conversion
of volatile acids to methane in a biochemically active landfill,
metal sulfide solu.bilities become even lower. Thus, the co—
disposal of heavy metal—laden industrial wastes with municipal
refuse in a controlled manner may mitigate the environmental
impact of such disposal practices (Pohiand and Gould, 1980).
Pilot-scale investigations by Pchland and Gould (1980)
indicate that significant attenuation and reduction of contami-
nant concentrations (particularly of heavy metals) in leachate
from landfills receiving industrial waste is ossible provided
appropriate landfill conditions exist (i.e., reducing conditions,
‘high sulfide concentrations).
Among the environmental factors required for support of
active biological stabilization of leachate constituents are
the availability of substrate and nutrients. Leachate data
reported by Pohland and Gould (1981), indicate that phosphorus
may be limiting in a landfill environment. This premise tends
to be substantiated by the indicated 300..:N:P ratios of about
100:10:0.3 in leachate from a pilot-scala landfill cell. Thus,
additions of Wastes with phosphorous in the proper chemical
state may prove beneficial in restoring stabilization.
Because of the high moistu re content and the high alkalinity
of many industrial sludges, the inclusion of alkaline wastes
may prove beneficial in accelerating refuse stabilization. By
increasing leachate pH to levels which enhance methane fermenta-
tion, levels of leachate organics may be diminished. Additional
research is required to define the applicability and rovide
quantitative information to optimize this procedure.
Addition of Selected Sorbents
Laboratory studies by Liskowitz,et al. (1976) have demon-
strated the effectiveness of various natural and synthetic
materials in removing contaminants from leachates. Selected
materials included bottom ash, fly ash, vermiculite, illite,
ottowa sand, activated carbon, kaolinite, natural zeolities,
activated alumina, and cullite. This study showed that the
behavior of a sorbent as regards the removal of a particular
contaminant, varied with the type of leachate being treated.
This effect may result from the competitive ion exchange and
sorption processes which exist among the constituents of a
leachate and the sorbent involved.
269
-------
The results reported by Liskowit, . indjcat
that no single sorbent, of those examined, can significantly
reduce the concentration of all constituents in leachate to ac-
ceptable levels. The results, however, Suggest that combination 5
of different sorbents can be used to reduce most of the Contami-
nants in the leachates to acceptable levels. The most promisjng
use of sorbents are for leachate treatment following collection;
however, placement in landfills as a bottom layer to provide
substantial contaminant reduction in leachates prior to migratj
from the sites, may exhibit greater economical feasibility due to
the reduced contaminant load imposed Upon the leachate treatment
facilities.
St idies by Fuller and Artiola (1978) indicate that the use
of limestone as an additional attenuation layer on top of natural
or synthetic layers has the ability to alter the o of a leachate
solution, thereby, effecting pollutant removal. Crushed lime-
stone has proven effective as a low-cost aid in the migration
control of certain heavy metals. The limestone apparently ef-
fects the mobility of the hazardous Constituents of the leaciiate
by: (1) absorbing the metallic ion directly; (2) forming less
soluble calcium and carbonate compounds; and/or (3) raising the
pH of the particular leachate as it passes through the limestone.
Practical use of crushed limestone as liner material for land-
fills as an aid in reducing the migration rate of potentially
toxic metal pollutants from leachates is suggested (Chen and
Eichextherger, 1981).
t2ACHATE COLLECTION SYSTEMS
Shallow interceptor drains and well systems are most COmmon-
ly used for leachate collection. The equipment and technj ues
are identical to that described for leachate volume control,
though leachate rather than indigenous waters are handled. See
Table 55 for a s nmary of leachate collection technology and.
costs.
Shallow Drains
Drains may be employed to intercept leachate from a shallow
ground—water table such as in the case of trench and area fill
landfill operations. Drains may also be used to collect
leachate along a steep hill downgradient from a disposal area.
Lutton, et al. (1979) and Lutton (1980) cite the use of
an underground drainage system incorporated into a cover seal
for a sanitary landfill. Construction, methods, and basic Costs
of drains are identical to those used to intercept ground water
prior to entering a disposal site.
Steiner, et al. (1979) described a sanitary landfill col-
lection system in Pennsylvania which allowed the leachate to
collect by gravity at three locations equipped with manholes.
270
-------
LaaChate was pumped from these anhcles and to the treatment
facility via pressure lines.
Well Systems
Shallow well point systems located down-gradient from a
landfill, or a deeper well system utilizing an underdrain
collection system, have been implemented at sanitary landfills
oughout the U.S. The model Lycorning County landfill utilized
a performated pipe network installed Witflifl a 15.2 cm (6 in.)
sand filter for leachate collection above a membrane (Giddings,
1977). A second drain collection network was installed under
the membrane to protect ground waters from leachates leaving
the landfill via membrane leaks. Similar systems have been
described for which leachate was collected at contoured low
points of the landfill bottom through collection wells (Mang
arid Weaver, 1976).
LEAC ATE TREATMENT
Control systems which involve the interception and collec-
tion of municipal landfill leachates may also require treatment
of the collected leachate prior to ultimate disposal. Leachate
treatment has been the subject of considerable research over
the past decade, the results of which are discussed below.
Section 4, Leachate Composition, describes the chemical and
physical character of municipal landfill leachate. A cursory
review indicates that, in general, fresh leachates are character-
ized by very high SOD and COD Concentrations (i.e., in the range
10,000—50,000 mg/i or higher). These concentrations, however,
are likely to decrease with continued leaching and biostabiliza—
tiori. Righ concentrations of major cations and anions, arid
suspended solids are also present near the onset of leachating,
and similarly tend to decrease as percolation through the solid
waste continues. The presence of significant concentrations of
metals, including copper, cadmium, lead, mercury, and zinc, are
of particular environmental and public health concern. Unlike
the organic and major ionic concentrations of leachate, heavy
metal concentrations are generally unpredictable as leaching
contiriue5. flitrogen and phosphorus may be present in leachate
ough at levels below that required for uninhibited biological
utilization of the organic matter present. Although the acidity
or alkalinity of leachate varies, organic acids formed by acid-
forming bacteria during anaerobic decomposition of solid waste
may create an environment that is slightly acid.
In addition to the long—term changes in the composition of
1eaChat mentioned above, short—term variability of specific
components is observed (Steiner, et al., 1979). These short—term
anges are produced by seasonal effects such as increased infil-
tration or evapotranspiratiori, or through hydrologic events.
27].
-------
The combined effects of short- and long-term leachate variabil-
ity makes the design of a treatment system particularly
difficult.
The aforementioned information indicates that munjcj al
landfill leachate can be quite different from domestic waste—
waters, and thus may not be treatable through conventional
municipal wastewater treatment systems. Leachates may have a
much higher organic content than wastewater, as well as gener-
ally higher concentrations of toxic substances such as metals.
Also, whereas a particular domestic wastewater may exhibit
stable concentrations over time, leachates can be highly vari-
able. Further evaluation of the potential for leachate
treatment by conventional wastewater treatment plants is pro —
vjded later in the discussion.
Basic Leachate Treatment Methods
Leachate treatment techniques demonstrated in recent years
are comprised of the traditional treatment processes common to
wastewater systems. The suggested general approaches to treat-
ment are:
• Biological treatment—aerobic and anaerobic bio—
stabilization systems; and
• Physical and chemical treatment-precipitation,
adsorption, coagulation, chemical oxidation, and
reverse osmosis.
After an extensive literature survey of treatment methods
for landfill leachates, Chian and DeWalle (1976) concluded that
newly formed landfill leachate is best treated by biological
treatment processes. Leachate from stabilized landfills, i.e.
leachates of low organic content, are best treated with physjc
and chemical methods.
Biological Treatment of Leachate-—
Various researchers have investigated the potential of
aerobic and anaerobic processes for the stabilization of leach—
ate. The primary goal of these approaches is to reduce high
BOD and COD concentrations in leachate. A basic problem in
biological treatment is that the leachate metals and other
contaminants may exert toxic effects on the biological treatment
culture. Evaluation of the kinetic factors in leachate treat-
ment exhibits these inhibitory, effects.
Aerobic treatment LabOrat0rYSCale research has demon-
strated that activated sludge processes can effectively remove
organic matter and metals from leachate (Boyle and Ham, 1974;
Cook and Foree, 1974; Uloth and Mavinic, 1977; Palit and Qasim
1977) Removal of 90 to 99% of the leachate BOO and COD and.
272
-------
some metal removal, was ccrmnonly retorted. The MLVSS concen-
tration in these lab-scale reactors was higher than that
typicallY observed in wastawater treatment: (e.g., 5,000 to
10,000 mg/i for leachates compared to 1,000 to 2,000 mg/i for
typical municipal wastewater processes). Because of the high
suspended solids concentrations, the food to microorganism
ratio was low (0.02-0.06). Eydraulic detention times are
substantially higher for leachates then for municipal waste-
water (e.g., 1 to 10 days for leachate compared to several hours
for wastewater). Solids detention times were 30 to 60 days
for leachate compared to 5 to 15 days for municipal wastewater.
These operational characteristics indicate that the large amount
of organic matter in leachate is not readily oxidized, and
requires extensive biological activity for stabilization. The
long treatment times indicate that extensive aeration energy
requirements will be required for aerobic treatment of leachate.
Various other operational problems were observed during
aerobic treatment, including foaming, nutrient deficiencies,
and toxic it ibjti0fl. Uloth and Mavinic (1977) indicated that
exCeSSi aeration in conjunction with high concentrations of
metals contributed to foaming, and that mechanical mixing
independent of aeration and anti-foaming adxnixtures, could help
obviate this problem. Palit and Qasini (1977) indicated that
leaChate stabilization could be hampered by nutrient deficiences,
and that the addition of nutrients may be necessary in some
cases- The toxic effects of metals and other constituents appear
to jnhjbit biological removal of oxygen demanding material,
as indicated by the increased time required for bio stabilization
(Uloth and Mavinic, 1977).
With regard to stabilization kinetics, Uloth and Mavinjc
(1977) evaluated the kinetics of aerobIc treatment of high
5 trength leachate in comparison with municipal wastewaters.
‘rable 66 shows the kinetic values when evaluated according to
the Lawrence and McCarty (1969) Model. The values shown in Table
66 indicate a subdued level of microbial activity in the treat-
ment of leachates in comparison to the treatment of municipal
wastewaters.
Metals removed from leachate during aerobic treatment was
reported by 1oth and Mavinic (1977). Following a biological
detention time of 10 days, activated sludge digester effluents
showed less than 10 mg/i of iron, dropping from an original
concentration in raw leachate of 240 mg/i. More than 95% of
the mixed liquor aluminum, cadmium, calcium, chromium, manganese,
and ZinC were also removed by the settling biological floc.
A average of 85% of the lead and 76% of the nickel was associ-
ated with the sludge solids. Between 49% and 69% of the mixed
liquor magnesium was removed by settling. The authors noted
that the percentage removal for all metals is generally higher
273
-------
than that observed by other researchers for activated sludge
processes.
TABLE 66. KNETIC PARAMETERS FOR AEROBIC TREATMENT OF LEACHATE
(HIG STRENGTH) AND DOMESTIC WASTEWATER. (tfloth and
Mavinic, 1977)
k Day-i km mgt l kd Day 1
High Strength Leachate O.3 2 0.75 21375 Q. 0 025
Typical Municipal Wastewater 0.6 6.0 100 0.05
Y growth yield coefficient (biological solids produced!
COD destroyed) lb/lb
k — naxirnum rate of substrate utilization per unit weight
of biological solids
lcd = endogenous death rate of biological solids
k substrate concentration when specific substrate
utilization is one half the maximum
Anaerobic treatment——Anaerobic treatment of leachates has
been effective in reducing organic loads. Continuous culture
lab—scale reactors have demonstrated 90 to 99% organic removal
which is corn arable to that achieved by aerobic treatment (Boyle
and Ham, 1974; Pohland, 1975). Similar removal efficiency is
reported by Chian and DeWalle (1976) using a lab-scale anaerobic
filter.
Anaerobic treatment offers several advantages over aerobic
treatment. First, the build—up of biological solids in the
reactor is low, implying a reduction in sludge disposal requjre
ments. Second, treatment times demonstrated for anaerobic
treatment of leachate (about 10 to 15 days) are comparable t
the time required for anaerobic digestion of municipal waste
water. Finally, the absence of aeration requirements allows a
saving of power costs, though some energy may be required to
maintain reactor temperatures at the necessary 30—35 C. A
major product of anaerobic digestion is methane, which can be
used as an energy source for maintaining reactor temperatures,
or for other purposes.
Physical/Chemical Treatment of Leachata-—
Various physical/chemical treatment processes have been
applied to leachate treatment. Physical or chemical treatment
processes may be particularly useful in treating leachates from
older landfills whose organic content is negligible, or as a
274
-------
00 jshing step for leachatas previously treated by biological
ethOd3. Sonie of the major physical/chemical treatment processes
demonstrated are described below.
preci itatiOfl and coagulation——Chemical . precIpitation and
0 agulati0fl experiments have been fairly successful in removing
irofl, color, and suspended solids in leachates. Ho, et al. (1974)
d Thornton and Blanc (1973) have demonstrated that precipita—
tion with lime is effective in the removal of iron and other
multivalent ions, color, suspended solids, and COD. BOD con-
centrations, however, are ap arently unaffected. Iron precipita-
tion by lime is particularly pronounced, as nearly 100 percent
removal is consistently reported at lime concentrations over
300 mg/i. Precipitation by sodimn sulfide is reported by Ho,
et al. (1974). Iron removal occurred only at a very high chemi-
i ose (1000 ing/l), and. its effects on other constituents was
negligible. Sulfides are apparently not as promising for
leachate cation removal as lime.
Coagulation with alum or ferric chloride is reported by
each of the aforementioned researchers. Although some removal
0 f iron. and color was demonstrated, both coagulants were found
to be of limited value in removing COD, chloride, hardness and
total solids. Results were highly pH dependent, and chemical
sages were high, resulting in large emounts of solids produc-
tiOfl
Ion exchance——LeaChate treatment by ion exchange is reported
bY Po and (1973) for leachates wthich had been previously treated
jo1ogicallY. Pohiand reported good results using a coml ination
of cationic and anionic exchange resins, removing many ionic
5 pecie s as well as dissolved solids and nutrients. Very little
residu organic removal was reported; however, the use of mixed
reSi appears promising as a treatment approach for the non-
organic fraction of leachate.
Carbon adsorption-—The removal of leachate organics by
carb01 adsorption was studied by Pohiand. (1975), for leachates
,revioUSlY treated by ion exchange. Though moderately success-
ui., activated carbon was found to release solids which adversely
affected the total solids concentration of the leachate. Pohiand
5 gge 5ted that if leachate were to be treated with both carbon
a . mixed resins, the carbon treatment should precede the mixed
Activated carbon column tests performed by Ho, et al.
(1974) demonstrated complete color and odor removals at —
detention time of 20 minutes, along with 55 percent COD removal.
Reverse osrnosis-—Chian and DeWalle (1976) examined reverse. - -—
mosis treatment. of Ièachatés an round iffe5tive removal of
COD and dissolved solids. However, the potential for membrane
f 0 uling by suspended solids, colloidal material, and iron hydrox-
jdeS was noted. Thus, reverse osmosis is perhaps most effective
275
-------
as a cost—biological treatment step for removal of residual COD
and dissolved solids. Chian and Decqalle also note that membrane
efficiency is sensitive to pH.
Chemical oxidation-—Chemical oxidation was shown to be
reasonable effective in removal of COD, iron and color (Ho, at al.
1974), though only at high concentrations of the oxidizing ag t
Oxidizing agents included chlorine, calcium hypochlorite, potas-
sium permanganate 1 and. ozone.
As a means of si.munarizing the treatabilities of landfill
leachates, Figures 49 and 50..present the removal.efficiencies
of the various treatment processes reported in the literature.
Figure.49 portrays the COD removal efficiencies for raw leachatas
of a young landfill (or fresh refuse), and the COD removal
efficiencies following initial biological treatment. Each pro-
cess is subject to a number of design variables, which may in
part, account for the range of observed efficiencies. According
to Figure 49, the anaerobic filter, anaerobic digestion, and
activated sludge processes, show the greatest COD removal
efficiency for raw, non—biologically treated leachate. For
COD removal following initial biological treatment, activated
carbon and reverse osmosis treatment approaches provide the
highest efficiency. Coagulation and oxidation systems generally
remove less than 40 percent of the leachate org’anics. Figure so
presents the reported removal efficiencies for various inorganic
contaminants by aforementioned unit treatment processes. A
comprehensive summary of the effectiveness of chemical oxidation,
chemical precipitation, ion exchange, and reverse osmosis treat-
ments of sanitary landfill leachate is provided by Shuckrow,
Pajak, and Touhill (1981).
Leachate Treatment Systems
Because leachates can have high concentrations of both
organic and inorganicscontaminantSi treatment of leachates at
the field scale requires integration of the basic treatment
methods into a systematic approach. The design of such a
system should account not only for the volume and quality of
leachate to be treated, but also on the changes in leachate
quality over time. Several leachate treatment system options
are: complete treatment of leachate, pretreatment or combined
treatment with municipal wastewater, land application of leach—
ate, aerated lagoon, and leachate recirculation through the
landfill. The following text describes these basic treatment
systems.
Complete treatment of leachates refers to some sort of
combination of the basic treatment schemes discussed earlier.
The complete treatment approach is perhaps best suited for land-
fills remotely located from an existing sewage treatment Plant.
Steiner (1979) demonstrated a complete treatment system
276
-------
_____________________ — 4 I
ANAEROBIC FILTER I
AERATED LAGOON 1 1
ACTIVATED SLUDGE I I
ANAEROBIC DIGESTER I
ACTIVATED CARBON (COL.) I
ACT. C (BATCH) I
REVERSE OSMOSIS
LUM OR ILIfIE COAGULATION AND AERATION
LUM OR 1 LIME COA( ULATION
I LIME COAGULATION
OXIDATION OZONATION)
O IDATION (çHLORINATIgN)
0 20 140 60 80 OO
COD REMOVAL OF RAW LEACHATES
I ’
ACTIVATED CARBON I
REVERSE OSMOSIS
STRONG BASE ANION EXCHANGE RES IN j
WEAK BASE ANION EXHCANGE ______ RESIN
I I OZONATION
I LIME COAGULATION
I I AERATED LAGOON TREATMENT
I CALCIUM HYPOCHLORITE COAGULATION
0 20 40 60 80 100
Z COD REMOVAL FOLLOWING INITIAL 3IOLOGI AL TREATMENT
Fjqure 49. COD removal efficiencies for the treatment of leach—
ates (Boyle and aam, 1974: Cook and Foree, 1974;
Eo et al., 1974; Palit and Qasim, 1977; J1oth and
Mavinic, 1977; Steiner et al., 1979; Chian and
Dec4alle, 1977).
277
-------
SM
Q
‘5
I
A .
C-
I S
S S
20 40
% Removal
—
60 80
of
I
E+F
4 4
E+G
E+R
A
B_
—
—
S
I S
F—
E—
— —
—
E#G —
—
—D C
A
I —
I I I I
%
20 40 60 80
Removal of Fe
4 £ 4 4
—
— —
D
&F
E +C
B
I I
20 40 60 80
% Removal of i
20 40 60 80
% Removal of Zn
a a
B —
— B + F
— B 4 G
— B +
A
I 5 5 t
20 40 60 80
% Removal of Cd
Lagend:
A— Anaerobic Filter
B— Air Stripping
C- Activated Sludge
D— Activated Carbon
B— Lime Coagulation
a I I I
I
E + F
£ +G
- + if
A
I S I I
80
Cu
20 40 60
% Removal of
F- NaOC1 CoaguiLation
C— Alum Coagulation
if- NaOif Coagulation
I- Aerated Lagoon
Figure 50. % removal of NE , Fe, Zn, Ni, Cd, and Cu by v j j
trea ent alter atives (Thornton, and Blanc, 1973;
Boyle and ifam, 1974; Cock and Force, 1974; Ee,et al..,
1974; Pohinad, 1975; Chian and DeWalle, 1977;
Uloth and Navinic, 1977; and Ste ner,et al., 1979).
278
-------
utilizing physical, chemical, and biological unit processes.
Leachate produced by a landfill in Tullytown, Pennysivania is
collected by an asphalt liner. Several unit treatment process
sequences have been tried. The best treatment sequence, and
the only one that is capable of approaching effluent standards,
provides flow equalization, lime urecipitation and clarification,
onia stripping, activated sludge, and chlorination. The de-
sign is illustrated in Figure 51. A summary of the plant
operating data appears in Table 67.
TA3LZ 67. S NARY OF STST M 1 OPEBATION DATA (Steiner, 1979)
Raw Final Discha.rge
Leachats E f1uent Percent Standard
parameter ec/1iter g/1iter Removal, mg/liter
Aonia-N 758 75 90.1. 35
30D 5 11886 153 98.7 100
C a i 0.08 0.017 78.2 0.02
c rc i 0 • 26 0 • 07
73.1 0.1
COD 18490 945 94.9 *
CcppOZ ’ 0.40 0.11 72.5 0.2
Iron 333 2.7 99.2 7.0
Lead 0.74 0.12 83.8 0.1
Mercnr y 0.006 0.004 27.4 0.01
Nicice2. 1.76 0.75 57.4 *
zinc 19.5 0.53 97.3 0.6
*
No dizoharge standard for thi3 parameter
Because of low flow conditions of the receiving waters,
treated leachate ca ll be discharged only from December until
April. In other months, treated ].eachate is recycled to the
landfill. Leachate to be recycled is not chlorinated.
Major problems reported for this operation include leachate
flow and strength variability. Leachate COD has varied over a
wide range from day to day, though concentrations averaged
betWeen 10,000 and 15,000 mg/i during the first three years of
o eratiOfl. A lime precipitation stage was needed to reduce the
metalS concentrations to levels which would not inhibit the
279
-------
Figure 51.
Sc iematic of complete leachate treat nent plant
(Steiner, et al., 1979).
Sludge
Ho)d n - - WasteSlud__ —
21 cu
U’
35.8 Cu m
23.7 Cu tfl
Cu I
Landfl I I
280
-------
activated sludge process. Similarly, the toxic effect of
anonia was controlled by the .monia stripping process. After
lime treatment, sulfuric and phosphoric acids are added to lower
the pH and raise nutrient levels before the activated sludge
process. The activated sludge process reduced 30D and COD
concentrations by over 90 percent (Table 67), though a detention
time of more than one day is typical.
Steiner, et al., report that the system operation is sensi-
tive, and eff1U n standards are approached only during optimum
operation of the plant. Thus, although overall satisfactory
complete treatment of high strength leachate can be accomplished,
it is clear that a sophisticated treatment system is needed.
A comprehensive review of process train alternatives for hazard-
ous waste leachates is presented by Shuckrow, Pajak, and
ToUhill (1981).
For leachate collected from a Landfill located near a
wasteWater treatment system, a convenient method of treatment
would be to discharge the leachate to the sewer system. The
0 ssibility of mixing untreated leachate with domestic waste-
water for conventional biological treatment was studied by
Boyle and Ham (1974). Their findings indicated that high
strength sanitary landfill leachate (10,000 mg/i COD) can be
added to domestic wastewater in an extended aeration activated
sludge plant at a level of at least 5 percent by volume without
seriously impairing the treatment process or the effluent qual-
ity. Beyond 5 percent by volume, leachate additions resulted
in substantial solids production, increased oxygen uptake
rates, and poorer mixed liquor separation. Chian and DeWalle
(1977) also studied combined treatment, and found that leachate
j f1OWS greater than 4 percent reduced treatment efficiency.
Boyle and Ham (1974) suggested that the presence of metal and
aumtonia toxins and extremely high organic loads in leachate
0 ould cause severe upsets in biological reactors.
Where toxins are expected to be a problem, or for leachate
flows representing a large fraction of the domestic load, leach—
ates may be pretreated prior to mixing with domestic flows. A
9 jmple pretreatment system involving an aerated basin with a
detention time of 10 days, and a polishing basin with a 10—day
detention time, was reported by Vydra and Grj (1976). Actual
performance data was not available for the facility. In the
event that the pretreatment system can bring leachates to within
effluent guidelines, the leachates will be discharged to an
adjacent river rather than the treatment plant.
Land application of leachates through spray irrigation or
overland flow is another possible treatment alternative. Since
the soil is the media in which the treatment is to take place,
careful consideration must be given to selecting a site with
gøil properties which are suitable for the retention and
281
-------
degradation of the leachate to be treated. A detailed soil
survey coriducted according to the standard Soil Conservation
Service (SCS) tecbniq.ue is necessary for all land treatnene
facilities ( . W. Brown and AssociateS, 1980). A thorough know-
ledge of the hydrogeciogic setting of the site is also requir .
The possibility of land application of leachate from a
sanitary landfill on a sandy soil in a high water table area in
Manatee County, Florida was investigated by Nordstedt, et al.
(1975). Collected leachates were pumped to a detention o ,
where a sprinkler irrigation system discharged the leachate over
pasture grasses. Examination of soil solutions collected at 1—,
2—, and 3—ft (20—, 61—, and 91—ca) under leachate irrigated plots
after 6 months, revealed no unusual changes. The authors
suggest that land application of high strength leachates in
sandy, high water table areas is a promising method of treatment
and disposal. Among the advantages of land application are that
fluctuating flow rates and water quality may be easily accoz o—
dated, and that equipment used to spread the leachate may be
moved to another location when a site is completed.
Although riot considered a land application system, the
migration and attenuation of leachate contaminants by soils
underlying a landfill is similar to a land application system.
The land disposal gudelines promulgated under RCRA (EPA, 1979)
suggest full assessment and subsequent utilization of the hydro—
geologic capability of subsurface soils to attenuate leachate
pollutants as a major leachate control alternative. Fuller
(1978) performed a comprehensive investigation of Leachate
pollutant attenuation by soils, and suggests that soil attenu—
ative ability depends largely on its clay and silt content,
hydrous oxides of iron, and pH.
Aerated lagoon treatment of high strength leachate (COD
57,900 mg/i) was examined by Chian and De ia11e (1977) , who
utilized completely—mixed, laboratory—scale plastic tanks to
simulate a lagoon. The experiments showed that between 93% and
97% of the organic matter of the leachates could be removed at
detention times ranging from 86 days to as low as 7 days. Nutri-
ent addition was deemed essential to successfully maintain
effluent organic matter at low levels, and to prevent a decrease
in MLVSS and deterioration of sludge settling rates. !igh
removal of heavy metals was observed, especially for iron
(> 99.9%), zinc (99.9%), calcium (99.3%), and magnesium (75.9%).
Perhaps the most innovative and acclaimed method of Leachate
treatment is the recirculation of high strength leachate back to
the landfill, so that it may again percolate through the refuse.
By recirculating the leachate, the organic component of the
leachate can be reduced by the biological communities active
within the refuse mass. tlsing experimental landfill columns,
Pohland (1975) found that when leachate was continuously
282
-------
collected and reapplied to the landfill surface, the organic
load of the leachate decreased to a small fraction of its peak
value (i.e., from 20,000 ing/1 COD to less than 1,000 mg/i) in a
period of just over a year. A continued investigation of the
effects of leachate recycle using a pilot—scale fill Droduced
5 jmilar results (Pohiand, 1979). Figure 52 shows the results of
pohiand’s original study and results from one of the field scale
cells operated by Leckie, et al. (1979). The COD concentration
histories show somewhat si ilar trends, with the leachate recjr-
culation cells showing rapidly declining CO concentrations in
comparison to the non-recirculating control cells.
Recycling of leachate apparenti.y accelerates the stabiliza—
tior’. of the readily available Organic material in the landfill
.nd in leachate. Pohiand (1975) suggests the recirculation
provides for a more rapid development of an active anaerobic
bacterial population of methane forming bacteria. The rate of
removal of organic components was further enhanced by the initial
addition of sewage sludge and by pH control.
In addition to rapid decline in leachate COD concentrations
over time provided by recirculation, similar concentration his-
tories were observed for BOD, TOC, volatile acids, phosohate,
. 0 nja-nitrogen, and T!DS. Reduction in nitrogen and phosphorus
are not considered great enough to limit biological growth.
Recirculated leachate remains high in inorganic material.
Specific advantages of leachate recirculation include not
only a substantial reduction in leachate organic components, but
also a delay in the starting time for leachate treatment, and a
leachate volume reduction through evapotranspiration loss.
Leachate treatment costs may also be substantially reduced by
recirCUlati0 . as described below.
. rreatmetlt Costs
Costs associated with leachate treatment include leachate
0 ollecton costs and the costs of actual treatment. Leachate
collection costs include the design and construction of an
jmperme able barrier or liner beneath the landfill, shallow
draifl 5 ’ well points, and pumping equipment involved in leachate
collection A review of the costs associated with collection
5 ystems and pumping was provided in Table 65. It is generally
estimat that a liner and leachate collection system will
increase 30 to 40 percent the costs of an equivalent landfill
withOUt collection ( aldwin, 1979).
Costs associated with leachate treatment are largely un-
documented, primarily because few field—scale treatment Systems
have been constructed. Chian and DeWalle (1977) estimated the
total costS of leachate treatment (excluding collection costs)
for three promising biological processes, and for physical!
283
-------
22000
z
z
z
10000
C
C -,
C .)
70000
60000
—
50000
40000
30000
20000
Z 10000
C .)
0
Time since leachate Droduction began, months
Figure 52. Effects of added moisture on leachate COD.
18 000
14000
6000
2 10 18 26 34
Time since leachate r ducticn becan, months
Leckie, t al. (1979)
A Control
D Leachate Recycle
I?”
/
/
6
I
/
A
—
1
12
18
24
284
-------
chemical treatment DrocesseS used to treat the effluents from
the biological procesSes. The estimated costs are resented in
Table 68 and are based on 1981 dollars. Table 68 shows that the
treatment costs are a strong function of both leachate flow
rates and leachate strençth. In general, as the overall leachate
5 ength increases, the unit cost of treatment increases greatly.
as the leachate flow rate increases, the unit cost of
treatment decreases. The authors note that the circumstances
of a particular situation may drastically alter treatment costs
listed in Table 68. For example, for leachates discharged to a
jcipal wastewater treatment plant (activated sludge, combined
treatment)’ having excess capacity the cost (in terms of stir—
charges paid) could be sithstantially lower than. the est imates
presented.
Steiner, et al. (1979) estimated the operating costs associ-
ated with the complete treatment system illustrated in Figure 51
to be approximately $4.20/bOO gallons, based on 1981 dollars.
power costs connected with the aeration of activated sludge and
the monia stripping lagoon, amount to over 50 percent of this
ogt. Capital costs connected with the System were estimated
by Straub (1979) using Engineering ‘Tews Record and EPA cost
estiutatjon guidelines. tpdated 198]. costs are:
Equalization lagoon 202,000
Time precipitation tank 144,000
znmonia stripping lagoon 202,000
Activated sludge process 248,000
Disinfection tank 36,800
Total 832,800
Straub estimated total treatment costs of the system based
a 20 year design life. Making assumptions regarding the
total volume of leachate generated over the 20 year period,
energy costs, and an eight percent interest factor, Strai.th
estimated that landfilling costs (on a cost/ton of refuse basis)
are increased by 25—35% due to costs of collection treatment
and handling of leachate.
Vydra and Grimm (1976) estimated the capital costs of the
wastewater pretreatment system discussed earlier. The capital
and construction costs of this System (including collection) was
ported to be about 10 percent of the total construction cost
the landfill. Capital costs were kept low because of the use
o native clay soils for the liner. The cost of installing an
j ermeable liner could more than double the capital cost figure.
operational costs were not reported.
For landfills where leachate collection and treatment is
required, Straub (1980) indicated that leachate recirculation
an substantially reduce treatment costs, regardless of the
treatment system selected. The reduction stems from:
285
-------
TABLE 63. A St 1NARY OF COST ESTI 4-ATE5 FOR LEACHATE TREATMENT
(Chian and Dec 1al1e, 1977)
Leachate Typical Effli.zent 1981 Costs
(gal/rain) COD (mg/i) ($11000 gal 1eachate ’
Influent BOD,
rag / 1
25,000
5,000
25,000
5,000
Activated Sludge
(AS) (Combined
treatment)
20
2
30
30
30
30
34.0
59.6
8.6
17.1
Aerated Lagoon
20
2
500
500
100
100
•
25.8
45.5
5.9
14.4
Anaerobic Filter
(AF)
20
2
1500
1500
300
300
31.8
(25.8)
61.9
(55.9)
9.8
(8.5)
25.5
(24.2)
AL—Sand Filter
(SF) —Activated
Carbon (AC)
20
2
125
125
25
25
37.0
57 5
.
10.5
19 7
.
AL-SF—AC—Reverse
Osmosis (RO)
20
2
25
25
5
5
39.7
64.2
13.2
26.5
AF—S?—AC
.
AF—SF—AC—RO
20
2
20
2
375
375
75
75
75
75
15
13
47.2
(41.2)
78.0
(72.0)
50.0
(43.8)
84.8
(78.2)
‘
15.3
(14.0)
31.7
(30.4)
18.0
(16.6)
38.5
(36.6)
* 1 gal 3.79 litre
Ni ers shown in parenthesis indicate the costs of treatment
after deducting the credit for methane produced $1.50/boo
Cu ft.
After RO treatment the total dissolved solids (TDS) decreased
to 300 mg/i and 60 mg/i for influent leachate BOD concentra..
tions of 25,000 rag/i and 5,000 rag/i, respectively.
86
-------
1. Internal reduction of the initially high organic
load of the leachate; and
2. Reduction of the total volume of leachate due to
increased evapotranspiration.
For a landfill operating without leachate recirculation,
leachate COD can be extremely high at the onset of leaching
(e.g., 25,000 mg/i or greater) and would diminish with continued
eachiflg. The design of the leachate treatment system would need
to account for the initially high organic load. Conversely,
leachate drawn from a landfill with recirculation would have
3 ignificafltly Lower organic concentrations. In addition, inf ii—
trated precipitation from a landfill with recirculation would be
su JeCt to greater evaDotranspiration as opposed to Landfills
withoUt recirculation. The results of greater evapotranspira-
tion is to reduce the total volume of leachate to be treated.
Comparative annual treatment costs for leachate from recir-
cui.ated and non-recirculated landfills are presented in Table 9,
based on the treatment system suggested by Chian and Dewalle,
1377, w.hich. includes an ae±ated lagoon, slow sand filtration, and
activated carbon treatment. For each of the two hypothetical
landfills, estimated treatment costs for landfills with recir-
cui.ation is less than 25% of the estimated costs for the non—
recirculated case. The major factor for this reported cost
difference is the cost of treating the initially high COD present
in the non-recirculated leachate (Straub, 1980). Note that the
costs presented in Table 69 do not include the costs of a liner
and collection system.
TABLE 69 . ESTIMATED COMPARATIVE ANNUAL LEACHATE TREATMENT COSTS
FOR RECIRCULATED AND EQUIVALENT NON-RECIRCULATED
LANDFILLS (Straub, 1980)
2 Treatme 9
Landfill area = 218,00Cm Vo1t ne m /yr Cost/m’ Cost/yr
Recirculated Case 39,735 $2.28 $90,300
Non—recirculated Case ‘ 45,848 $8.83 $405,000
Landfill area 22,500 rn 2
Recirculated Case 4,099 $4.52 $18,600
Non—recirculated Case 4,730 $15.72 $74,300
2 7
-------
Landfills with leachate recirculation require a system for
the reapplication of leachate to the landfill surface, which
increase the costs of recirculation. By adding this cost to the
figures in Table 70, the comparative savings per year in treat-
ment of the recirculated leachate over the non-recirculated
leachate is given as follows.
TABLE 70. COMPARATIVE SAVINGS OF RECIRCIJLATED CASE OVER NON—
RECIRCULATED CASE ( Straub, 1980)
1981 Costs
Landfill rea Treatment ReappJ.i.cat .on Savings/yr for
218,000 rn Cost/yr Cost/yr Recirculated Case
Recirculated Case $90,300 $233,700 $80,500
Non-recirculated $405,000 0
Case
Landfill Area =
22,500 rn 2
Recirculated Case $18,600 $24,100 $31,700
Non—recirculated $74,200 0
Case
For the larger landfill, a savings of 20% is estimated for
leachate treatment and handling costs for the recirculated design
over the non—recirculated design. For the smaller landfill,
savings are estimated to be 43%. Thus, leachate recirculatjon
becomes more attractive as the size of the landfill decreases,
though savings are evident, regardless of size.
Leachate volume and quality estimates which serve as the
basis for these cost estimates were obtained from the unsat-
urated flow model (St.raub and Lynch, 1982b), described in the
Leachate Composition section. The ability to predict leachate
quality and quantity is invaluable in designing leachate treat—
inent systems and in estimating the costs of such systems.
SUMMARY
Leachate voli ne control invo’ves both ground-water and s —
face-water control measures. Subsurface infiltration barriers
(e.g., bottom sealers, slurry trenches, grout curtains, sheet
piling cutoff walls) are designed to either prevent ground water
from flowing through the landfill and generating leachate or
control the movement of leachate away from the fill. Drains,
288
-------
shallow well points, or deep wells are used to lower the water
table to either prevent the fornatior,. of leachate or contain its
soread. Surface water control measures suCh as contour grading
ad surface water diversion, surface sealing, and revegetation
are used to min nize the quantity of water entering the landfill
for purposes of reducing leachate generation.
The purpose of leachate Ccmpositiofl control is to reduce
the strength and contaminant flux of leachate. In general,
high water application rates, small refuse particle size, and
shallow landfill depths will increase refuse stabilization rates
and ui.tintately, the leachate strength. Leachate recirculation
offers several advantages including acceleration of refuse
stabilization, and treatment systems are not always recuired
for the collected leachate. Codisposal of sewage sludge or
alicaline wastes with municipal refuse can be beneficial in
accelerating the stabilization of a landfill. Laboratory
studies have demonstrated the effectiveness of various natural
and synthetic materials in removing contaminants from leachate.
The addition of crushed limestone as an additional attenuation
layer has also proven effective as a low-cost aid in the migra-
tion control of certain heavy metals.
The successful treatment of high strength leachates will
probably require different combinations of unit treatment
techniqUeS , the appropriate combination dictated by the character
of the leachate. For a high strength leachate containing high
concentrations of both organic and inorganic contaminants, a
combination of biàlogica1. and physical/chemical processes will
be needed.
The design of a leachate treatment system will depend not
only on the character of the leachate, but on landfill location.
For landfills located near a wastewater treatment facility,
leachates might be treated by the facility. For landfills dis-
tant t.o a wastewater treatment plant, an aerated lagoon, land
appliCati0n or complete treatment may be applicable. Each type
f system has been successfully demonstrated, either in pilot-
or field-scale testing. Regardless of the system selected,
leachate recirculatior ,. can be beneficial in reducing the organic
5 rength and voiwne of leachate, as well as reducing treatment
StS. Treatment related costs may represent a significant
portion of the total landfilling cost, perhaps 25 percent or
more.
289
-------
SECTION 7
ENVIRO 1ENTAL MONITORING
INTRODtCTION
The Envirox mental Protection Agency has published a manuaj.
titled “Procedures Manual for Ground-Water Monitoring at Solid
Waste Disposal Facilities” (EPA/530/SW—611) which provides a
guide for superwisory personnel of solid waste regulatory agen-
cies (Fenn, et al., 1977). The Manual establishes the need for
ground—water monitoring, assigns priorities for facilities to be
monitored and implements and. directs a cost-effective, on—going
monitoring program. The presented information is offered a
guidance and suggested methods only. This section cotnoi.emnents
the material presented in the manual with additional informatjo
which is useful for landfill designers and site operators.
Three topic. areas are presented in this section:
1. Monitoring in the vadose zone;
2. Monitoring in the zone of saturation; and
3. Approaches and considerations for monitoring.
The first two sections address available monitoring methods,
costs, and the advantages and limitations of the monitoring
equipment for these two hydrologic conditions. The final section
discusses selection of sampling areas, monitoring frequency,
monitoring parameters, and sample collection and preservation
technology.
VADOSE ZONE
The vadose zone is that area beneath the top soil and over-
lying the water table, in which water in pore spaces coexists
with air, or in which the geologic materials are unsaturated
(Figure 53). Perched. water tables may develop above interfaces
between layers having greatly different textures. Saturated
conditions may also develop beneath recharge sites as a result of
prolonged infiltration.
290
-------
C
‘a
0 0 •—
“.4 0
‘U ‘J3
.U
“ . 4
0
‘2)
‘4
Fiqure 53. Hydrogeologica.l cycle.
291
-------
r astes applied at the land surface may travel through sig-
nificant thicknesses of tc soil and vadose zone before reaching
the water table. Pollutants can be significantly retained or
attenuated in the top soil and vadose zone. The storage capac-
ity of the vadose zone for percolating waters may also be great.
The vadose zone is an integral component of any leachate moni-
toring effort; however, in contrast to th large number of
studies on water movement in the top soil, parallel studies in
the vadose zone have been few.
Monitoring Approaches
Monitoring within the vadose zone may be accomplished
through one of several methods: soil sampling, pressure/vacuum
lysiiueter sampling, and measurement of the volumetric water
content of soils in situ. The first two methods provide a means
by which water samples may be collected for chemical analyses.
Non—sampling methods (i.e., the third type of method) are
available for measuring the movement of water through the un-
saturated zone. These are especially useful in arid regions to
indicate the degree of water saturation and subsequent move tent
of water in the vadose zone. Non—sampling devices, in conjunc-
tion with samplers, can provide an optimum system for samoling
when water is flowing through the unsaturated zone. Vadose
samplers do not provide this ability to specifically sample
during wetting and drying cycles.
Available Monitoring Ecui rrtent and Costs
Monitoring equipment for the vadose zone may be classified
as sampling or non—sampling. Sampling approaches include pore—
water extraction from soil cores and deployment of pressure!
vacuum lysimeters to obtain samples of in situ soil moisture.
Mon-sampling equipment includes tensiometers, ;sychrcmeters,
moisture blocks and neutron moderation devices.
Water extracted from soil cores may be analyzed for such
parameters as major anions, trace metals, total organic carbon
(TOC), pM, specific conductance, and other specific constitu-
ents. Core samples may provide information on physical charac-
teristics such as soil texture and its change with depth, water
content, hydraulic conductivity, and bulk density or water-
release curves. These are important if non—sampling approaches
are to be used in conjunction with vadose water sampling
(Nielsen,et al., 1973).
Soil Sampling-—
Equipment and ao licability——For shallow sampling of soils
and vadose waters, traditional hand augers and bucket type
samplers may be used. Hand augers include the Eankenson,
292
-------
oaJ fie1d, and assorted hand piston samplers (Soiltest, Inc.,
1976). For deeper sampling, standard drilling equipment and
chniques are required (i.e., jetting, water and air rotary,
cable tool, augering, and air drilling). Continuous flight
augers and air drilling provide the most useable samples.
3lake (1969) and Barton (1974) furnish an excellent discussion
in their application and use. Methods which utilize drilling
fluids oreclude obtaining cohesive representative samples for
water extraction and complicate correlation of sample quality
with depth.
Advantages-—The advantages of soil sampling are suimnarized
(Everett, , 1976; Fenn, at aL, 1977).
• Ease of soil sample collection;
• Accurate vertical and area], sampling locations;
• Chemical species associated with soil solution can
be traced downward in a soil profile;
• 4ay be used to measure extent of leachate attenuation
at various soil depths;
• Intermittent leachate production permits long interval
between sampling periods; and
• Provides for laboratory measurement of soil texture,
water content, field capacity, and hydraulic conduc-
tivity.
Limitations——Limitations of soil sampling include the
following (Ev arett, at al., 1976; Fenn, at al., 1977):
• Extraction of geochemical fractions from soil samples
is a slow, tedious procedure more costly than soil
water analysis;
• Extraction procedures are not a proven standardized
sampling method for monitoring programs;
• wetting and drying cycles and changes in redox
potential can change chemical reactivity of some
soil constituents after collection;
• In situ conditions of soil samples are difficult to
maintain; and
• State—of—the—art not documented in leachate studies.
Costs of monitoring-—Soil sampling Costs vary primarily
a function of sampling death and sample number. For shallow
293
-------
soil sampling up to 4.5 rn (15 ft) in unconso1idat d sedi.-nents
various hand augers may be used. The cost of hand augers (1981
dollars) updated after Soilt st, nc. (1976) ranges from $70 to
$130. For deeper samples, a hand—driven soil sam 1er system
such as that designed by Veihmeyer (1919) can be used for uncon-
solidated and consolidated sediinentsT 3asic parts and costs f
a Veibmeyer soil sampler are presented in Table 71. A cost dia.—
gram for coring in consolidated and unconsolidated formations,
sampling every 3 in (10 ft) is given in Figure 54. The cost
figures are generalized to cover small-diameter coring of 5 to
7.5 (2 to 3 in.) for shallower depths and larger coring sizes
of 7.5 to 12.5 (3 to 5 in.) for deeper samples, since ].arg r
bore diameters are required when drilling to depths greater than
31 i (100 ft): For unconsolidated forrnat. .°ns, equipment set—up
constitutes a larger part of the sampling costs than for con-
solidated formation sampling because of more elaborate casing
requirements 1 slumping problems, and drilling fluid differences.
At greater depths, the time required for sample procurement
overrides the drilling time.
sampling equipment (e.g., hand augers, rotary rigs, bucket
type samplers, sampling tubes) should be thoroughly cleaned
prior to reuse to prevent cross—contamination of soil samples.
An effective procedure is to wash(scrub) with detergent followed
by a tap water rinse. Cost data for cleaning sampling equipment
is not available in the literature; however, an approximate
assessment can be made for sampling tubes. A conservative
estimate is that a technician at $15/hr (including overhead)
could clean and rinse ten 1.5—in (10—f t) or 3—rn (5—f t sampling
tubes in five hours at a total cost of $75. A comparison of thj 3
cost to the replacement cost of ten 1.3—in (10—f t) or 3—in (5—f t)
sampling tubes at $721. and $1212, respectively, (Table 71)
indicates that service cleaning could be an effective cost/bene...
fit procedure.
Pressure/Vacuum Lysilneters-—
Equipment and a vlicabilitY ?re55Ure/vaCUum lysimeter
provide a means for collection of in situ vadose waters. These
devices are an alternative to soil sampling for characterizati 0
of the water quality of unsaturated soils. These units were
developed by Parizek and Lane (1970) and modified by Wood (1973)
for sampling depths up to 36 in (118 ft). The design of the
Parizek and Lane sampler is shown in Figure 55.
The basic design of the sampling unit consists of a porous
ceramic cup capable of holding a vacuum, a rigid-plastic body
tube that is egual in length to the desired sampling depth, and
two sampling tubes leading to the surface. when placed in the
* unconsolidated sediments: sand and gravel
consolidated sediments: limestone, sandstone, and do1oinjt
294
-------
soil, the pores in the ceramic cup become an extension of the
core space of the soil so that the soil—water content in the
soil and the cup become equilibrated at the existing soil-water
ressure. After emplacement, a vacuum is applied to the cup.
Soil solution is drawn into the sampler under this gradient.
The vacui is then released and pressure is applied, forcing the
accumulated water sample to the surface through the outlet sam-
pling tube. Although pressure/vacuum samplers have limitations,
they are considered to be the best available device for sampling
within the vadose zone.
TP .BL 71. BASIC PARTS AND COSTS OF A VEI WIEY R SOIL SAMPLER
(Veibmever, 1919)
Part
—
Cost
Tube, 1.5 m (5 ft)
$72.10
Tube, 3 zn (10 ft)
121.20
Tip, Type A, general use
36.90
Drive head
41.50
Drop hammer, 6.8 kg (15
puller jack and grip
lb)
Total
102.75
231.50
$605.9
* One of each part is needed. Manufactured by Hansen Machine
Works, 334 M. 12th Street, Sacramento, CA 95815.
updated to December, 1981.
Recommended primarily for deep soil sampling.
Advantages——The advantages of pressure/vacui lysimeters
are as follows (Penn,et al, 1977; Everett, 1980; Wil o , 1980):
• Best available techniques for obtaining samples of the
soil solution in situ;
• Inexpensive sampling device of great reliability;
• Inexpensive insta 1ation; and
• Samples can be collected at a central point.
Limitations-—Limitations of pressure/vacuum lysizneters in-
clude the ol1owing (Hansen and Harris, 1975; Fenn,et , 1977;
295
-------
DEPTH (m)
Figure 54.
46
CORING DEPTH (ft)
61 76
Generaiized cost of coring in unconso1id ted
and consolidated formatior s, December, 198].
(after Everett, et al., 1976).
10
15 • 30
>4
z
0
z
a
C
C.)
z
C
C.)
0 50 100 150 200 250
296
-------
VACUUM PORT
AND GAUGE
3 ,16 -INcH
COPPER 11JBE
PLASTIC Pt
a4 INCHES LONG
6-INCH HOLE
WITH TAMPED
SiLICA SAND
SACKF ILL
POROUS CUP
PUMP
PLASTiC TUBE
AND CLAMP
8OTTLE
6ENTcNITE
TU8E
SENTONI TE
ig12.re 55. Pre5sure/vacu n lysiineter (after Parizek and Lar e,
1970).
297
-------
Wilson, 1980)
• Suction cups are capable of sampling only at pressure
greater than about —1.0 atmosphere; therefore, samples
cannot be obtained over the entire range of soil-water
pressures;
• Chemical transformations within the ceramic cup are
possible;
• Small quantities of samples are collected over long
periods of time preventing short-term analysis of
concentrations as a function of time;
• Macromolecules may be screened out altering the
sample and plugging of the ceramic cup;
• Repairs are difficult in situ, and
• Samples represent only a small area; a localized
region of pollutant leakage may occur undetected.
Costs of monitoring——Commercially available lysimeters
may have one of a variety of porous cups and plastic tube de-
signs. (Soiltest, Inc., 1976; Soil Moisture Equipment Corp.,
1978; Tiztco Manufacturing Company, Inc., 1980) Depending u on
the length of the ceramic tip and plastic tube, individual
units cost $50 to $60 (1981 dollars) based on 1978 costs (Soil
Moisture Equipment Company, 1978). Plastic withdrawal and air
injection tubing costs 15 to 20 per ft. Bentonite seals
and silica sand packing costs $2.00 to $3.00 per installed unit.
Figure 56 gives the capital costs of pressure/vacuum lysimete
for various depths and sample space densities. The costs pre—
sented in Figure 56 do not include installation.
Tensiometers-—
Ecui rnent and appljcability-—TensiotneterS are used to
measure soil-water pressures during unsaturated flow conditions;
soil—water pressures may be employed to estimate soil-water
content through the development of suitable soil-water char-
acteristic curves (Nielsen,et al., 1973). Tensiometers consist
of a porous ceramic cup connected to a rigid plastic tube which
in turn is connected to a manometer with small diameter tubing.
The internal volume of the system is filled with water except
for the mercury reservoir (Figure 57 ). When properly emplaced
in the soil, the pores in the ceramic cup form a continuum with
the pores in the soil. Water moves either in or out of the
tensiometer system until an equilibrii. n is attained across the
ceramic cup. Changes in pressure are measured by the mercury
manometer. Holines,et , (1967) described their use and oper-
ation. Watson (1967) presented a design for a tensiometer-
298
-------
4i
4 - 1
H
>
I
C)
It ’
100
0$
60
40
20
0
10
PRESSURE/VACUUM LYSIME’I’ER COSTS ($)
Figure 56. Cost of pressure/vacuum lysinietera for various depths and sampling
densities, December, 1981 (after Everett, el a l . , 1976 .
30
24
18
12
6
t%J
F ’
1 41
50 100 500 1000 5000
-------
MANOMETER
FILLING PLUG
TENS 1OMETER
acov (PVC)
SOIL
PAR TI CLE
CERAMIC -.- ’ WAT
C Jp
Figure 57. Schematic representation o f tensiometer and section
through the ceramic cup (after Nielson, et al., 1973).
pressure transducer system, in lieu of the mercury system, to
permit the conversion of soil water pressure into any equivalent
electrical resistance. Figure 58 illustrates this system.
Advantages—-The advantages of tensiorneters include the
following (Gairon and Hadas, 1973; Oaksford, 1978; Brakensiek,
et al., 1979; Wilson, 1980) :
• Provide continuous in place measurements of water
content;
• Successive measurements are obtained;
• Inexpensive and simple; and
• Transducer tinits respond fairly rapidly to water con-
tent changes.
Limitations--Limitations of tensiometers include the follow-
ing (Holmes, et al., 1967; Scbmugg’e, eta1. ,1980; Wilson, 1980;
Everett, 198117 —
• Units fail at the air entry value of the ceramic cup,
generally about 0.8 atmospheres;
ERCURY LEVEL IN MANOMETER
RESERVOIR
!RAMIC
z
300
-------
ELECTRI C
LOCK WASHER
ILR PRESSURE CONNECTION
t/8”STD. FLARE FITTING
STAINLZSS STEEL TUBING
7/8” OI& 3/16” WALL THICXNESS
GASKET
FERENCE PCRT
SENSING DIAPhRAGM
?igure 58. Cross section of tensiometer—pressure trarsducer
asseznbly (after Watson, 1967).
TRANSDUCER
RUBBER STOPPER
GASKET
PM 131 TC
CERAM IC
FIBER GASKET
301
-------
• Results are subject to hysterisis, i.e. , different
results are obtained for drying and wetting cycles;
• If proper contact is not made between cup and media,
units will not operate properly;
• Sensitive to temperature changes; and
• Difficult to install at great depth in vadose zone.
Costs of monitoring-—The cost of a tensiometer utilizing
a mercury manometer including a 1.8-rn (6-f t) rigid PVC tube is
approximately $54 (1981 dollars) based on 1974 costs (Everett,
et al., 1976). Figure 59 gives the capital costs of tensio-
meters for various depths and sample space densities. The Costs
presented do not include installation. -
P sychrometers--
Equipment and applicability-—Tensiometers cannot be used to
measure soil—water pressure below about —1.0 atmosphere because
of air-entry problems. In recent years, progress has been made
in developing the thermocouple psychrometer for this purpose.
According to Watson (1974), in situ pressure measurements down to
-30 atmospheres are possible with these units. The principle of
psychrometric measurement of soil-water potential as discussed
by Rawlens and Dalton (1967), is that a relationship exists
between soil-water potential and the relative humidity of soil
water.
Psychrometers employ a porous bulb with a chamber to measure
the relative humidity of the soil, a sensitive thermocouple, a
heat sink, a reference electrode, and associated electronic
circuitry. Such units use the principle of Peltier cooling to
lower the temperature of one junction of the thermocouple below
the dewpoint, thereby allowing an evaluation to be made of the
relative humidity (Wilson, 1980). Two types of thermocouple
psychrometers are available. One type, which is installed in
access tubes, is constructed by mounting psychrorneters in porous
cups at the base of tubing. This unit may be withdrawn for
recalibration. The second type, shown in Figure 60 , is called a
“sealed-cup” psychrometer by Merrill and Rawlins (1972).
Advantages-—The advantages of psychrorneters are as follows
(Rawlins and Dalton, 1967; Merrill and Rawlins, 1972; Enfield,
et al. , 1973; Schmugge,. et al. , 1980)
• In situ pressure measurements are possible down to
-30 atmospheres, permitting the determination of
water contents in the very dry range;
302
-------
50 100
30
24
18
12
6
10,000
COST OF TENSIOMETER ($)
59’
Cost of tensioxneter and electrical meter for various
depths and sampling densities, December, 1981 (after
Everett, et al., 1976).
100
80
60
40
20
I
a
C.’
500 1000
303
-------
ACRYL1C TUBING
CO PPER
i 0 COPPSR
TEFLON INSERT
THERMOCOUPLE WIRE
-CERAMIC BULB
• Permits continuous recording of pressures and water
contents;
• Can be interfaced with portable or remote data
collection systems; and
• Some units have been installed to great depth
(down to 91 in (300 ft).
ACRYLIC TUBING
EPOXY RESIN
COPPER LEAD WIRES’ LEAD WIRE
COPPER HEAT SINKS
HE. 1 T SINKS
TEFLON INSE]
THERMOCOUPLE WIRE
CERAMIC BULB
Figure 60.. Soil psychrometer (after Merrill and RawLins, 1972).
Limitations--Limitations of psychrometers include the fol-
lowing (Campbell, et àl., 1973; Hanks and Ashcroft, 1980;
Schinugge, èt al., 1980; Wilson, 1980):
• Results are subject to hysteresis;
• Good contact between bulk and surrounding media may
be difficult to obtain;
• Provide point measurements only;
• May be difficult to obtain accurate calibration
curves for deep regions of the vadose zone; and
• Fragile, requires great care in installation.
Costs of monitoring—-The cost of a psychrometer is approx-
imately $80 (1981 dollars) based on 1976 costs (Soiltest, Inc.,
1976). Figure 61 shows the capital costs of psychrometers
including access tubes for various depths and sample space
densities. The costs presented do not include installation.
, — 2cm— I
304
-------
100
4 . )
‘t 80
El
0
0
U)
( L I
0
‘4
Figure 61.
60
40
20
100
COST OF PSYCJIUOMETER 1$)
Cast of psychrometer and electrical meter for various depths and
sampling densities, Docember, 1981 (after Soiltest, Inc., 1976).
12
a
a
U’
30
24
18
6
500 1000 5000 10,000
-------
Electrical Resistance Blocks——
E uicment and a licabilitYE1eCtr1cai resistance blocks
(moisture blocks) consist of electrodes embedded in a suitable
porous material such as çypsuxn, fiberglass, or nylon cloth. Soil
moisture contacts are connected by rubber-covered leads to the
surface. A schematic representation of a gypsum moisture block
is shown in Figure 62. ResistanCe blocks are used to measure
either soil-water content or soil-water pressure. Water content
(or negative pressure) within the blocks responds to the water
content (or suction) of the soil with which the blocks are in
intimate content and the electrical resistance properties of
the blocks change correspondingly. Moisture blocks calibration
involves evaluating resistance readings against a range of soil—
water contents or negative pressures. Moisture blocks may be
used to measure soil-water pressure if suitable soil-water
characteristic curves are available (Everett, 1981).
Lead
Gypsum Block
Stainless
Steel Mesh
Figure 62
Gypsum electrical resistance block
Wigh (1971) studied the use of a MC-300A soil moisture meter
and block at a sanitary landfill and found that changes in the
Wire
306
-------
moisture content of the refuse could be dete riined. Use of
such data as part o a monitoring strategy has not been widely
utilized, although it appears promising. Scbmugge, et a l . ,
1980 suggested siinila.r networks as an early warning system
to trigger implementation of a synoptic vadose sampling
progr 5 .
Advantages--The advantages of electrical resistance blocks
are ai ollcws (Holmes, 1967; Brakensiek, al., 1979;
5c1unugge , etal.,1980; ierett, 1981):
• Inexpensive and simple;
• Can be calibrated for either suction or water content;
• Can be used at soil water pressures less than -0.3
atmospheres;
• Can be interfaced with portable data collection systems;
and
• Precision is good.
Limitations-—Limitations of electrical resistance blocks
include the following (Phene, et al., 1971; Gairon and Eadas,
1973; Wilscn, 1980; Everett, l TTT
• May be difficult to install at great depth in vadose
zone and maintain good contact;
• Requires calibration for each textural type in pro-
file;
• Lack of sensitivity in wet range;
• Sensitivity to soil salinity;
• Gypsum blocks deteriorate badly in certain media;
• Calibration curves of some units shift with time; and
• Time lag in response.
Costs of monitoring-—Moisture blocks provide an inexpensive
means of developing a grid to determine the moisture content of
a ojl. Because of their low cost, a dense array of moisture
blocks may be placed under a landfill to detect leakage. The
St of an electrical resistance block is approximately 8 dollars
(1981 dollars). The cost of a soil moisture meter is approxi-
mately $250 based on 1981 costs (Everett, et al., 1976). The
capital costs, excluding emplacement, of eI c ica1 resistance
blocks, leads, and a soil moisture meter for various depths and
307
-------
sample space densities are given in Figure 63.
Neutron Moisture Logging--
Equipment and ap ijcability NeUtrOn moisture logging is
employed to measure changes in the volumetric water content
within a soil horizon. It is also used to delineate perched
water zones and to estimate flow rates.
The neutron thermalization method is based on the principle
that high energy neutrons are slowed down (therinalized) in
their trajectories primarily by collisions with light atomic
nuceli, especially those in hydrogen atoms. Instrumentation
used to measure water content by neutron thermalization requires
three Dr±ncipal components:
(1) A source of fast neutrons;
(2) A detector of slow neutrons; and
(3) An instrument to determine the count rate from
the detection equipment (Wilson, 1981).
The basic components of a neutron moisture logger are shown in
Figure 64.
In operation, source of high energy neutrons (e.g.,
amerecium-beryllium) in a down—hole tool is lowered into an
access well. Water in the vadose zone slows down the fast
neutrons which are captured by a detector in the tool. Counts
measured by a ratemeter are converted into volumetric water
content by an appropriate calibration relationship. Successive
readings show temporal changes in water storage at different
depths (Everett, 1981; Wilson, 1981).
The use of the neutron moisture logger normally requires a
cased well. The inside diameter of wells should be as close
as possible to the outside diameter of the probe. During
installation, it is essential to establish a tight fit between
the borehole wall and the casing, to minimize water migration
via the annulus. Drilling mud should be used since it interfers
with water content observations. In situations requiring a
tight fit, drilling techniques which do not require a drilling
mud (jetting or augering) are used (Everett, 1981).
Advantages——The advantages of neutron moisture logging
are as follows (Holmes, etal.,l967; Brakensi k,et al., 1979
Mct owan and Williams, 1980; Wilson, 1980; Everett, et a].., 1981):
• Rapid;
• Water content profiles can be obtained in situ;
308
-------
100
80
C -,
7 ’
60
C’,
-4
C ’ ,
40
20
0
0
COST OF SOIL MOISTURE TESTING EQUIPMENT ($)
jgii.re 63. Cost of multiple electrical resistance blocks and soil
moisture meter for various depths and. sampling densi-
ties, December 1981 (after Everett, et al., 1976).
a
250 500 750 1000 1250
309
-------
AND
POWER
SCU RCE
WHEN BERYLLIUM IS
BOMBARDED WITH ALPHA
PARTICLES FROM
AMERICIUM NEUTRONS
ARE EMIT1’ED
AVERAGE ENERGY LOSS
PER COLLISION:
HYDROGEN 63 e/.
OXYGEN IZ /.
and principles of neutron moisture logging
Figure 64. Equipment
(after Keys and MacCary, 1971).
NEUTRON
100
NEUTRON MODERATION
NEUTRONS
ENERGY
EMENT
NEUTRON CAPTURE
310
-------
• Can be used to locate perched ground-water zones
together with their growth and djssj atjon;
• Water content changes at a given depth in a succes-
sion of wells may provide data on lateral flow
velocities; and
• Can be interfaced with portable data collection
systems.
Limitations-—Limitation of neutron moisture logging include
the f fl.owing (Brakensick ‘_ _. 1 . . 979; Schmugge, et al., 1980;
Wilson, 1980; Everett, 1981): —
• Expensive, requiring the purchase or leasing of
equipment;
• Cannot relate results exactly to a specific depth;
• Fast neutrons are moderated by other constituents
besides hydrogen in water, e.g., chloride and boron.
Accuracy may be affected;
• An indirect method requiring a difficult calibration
procedure; and
• Cannot be used to infer water movement in regions
where storage changes do not occur.
Costs of monitoring-—The cost of a neutron moisture logger
c1uding the neutron source, slow neutron detector, and surface
0 ounter is about $5! 90 (1981 dollars), updated after Everett, -
1976. Figure 65 qives the cost of neutron moisture loqqinq
jth various ntuv.bers of well loggings performed, at depth inter-
vals of 2 feet. The Costs in Figure 65 include Ofl ,ly the time
required to conduct the log; they do not include ca ita1 costs
for the equipment and installation.
zONE OF SATTJRATION
The zone of saturation is that portion of the ground-water
5 ystem in which available pore space is occupied by water
(ricure 53). The zone of saturation may include both permeable
and jmpermeable earth layers.
Monitorinc Approaches
Monitoring within the zone of saturation has traditionally
been accomplished with screened wells. A serious 1 njtatjon of
thi9 method is its inability to provide a representative vertical
cr° 5 section of contamination within the aquifer due to vertical
jxing within the well casing. More recently developed equipment
311
-------
30
100
23
“-4
F-’
15
F’ -“ 14
I—’ 14
t’ ) 0
0
a
0
25 8
0 0
10 50 100 500 1000 5000
LOGGING COSTS AT DEPTH INTERVALS OF 2 FEET ($)
Figure 65. Cost of neutron logging for various depths and sample space densities,
December, 1981 (after Everett, et al., 1976).
-------
allows for multiple sa ipling points throughout the aquifer.
Irrespective of the selection of equipment and the location
of samplers, accurate hydrogeologic information is required.
a minimum 4 , the hydraulic gradient must be known. The hydraulic
gradient is obtained by measurement of not less than 3 wells
whiCh form a triangle on the perimeter of the site. Water level
measurements can then be used to determine approximate flow
patterns. A more complete hydrologic understanding of the site
will usually be obtained with additional wells and water level
measurements. Fignre 66 was developed to illustrate the use of
this method.
Determination of ground—water flow patterns represents the
minimum data required for establishing a monitoring program.
Other critical data are the characteristics of the geological
formation, water level fluctuations, and soil types. various
researchers have provided detailed procedures describing the
j formation required and the approaches by which the necessary
physiograPhic data. for a disposal site may be obtained (Todd,
et 1976; Sendleiin and Yazicigil, 1981). The success of
e monitoring program is usually closely related to the results
of these pre-monitoring site characterizations.
Available Monitoring Eq 4pment and Costs
For purposes of discussion, monitoring equipment within the
zone of saturation is described in three categories:
.. Wells with the capacity to sample at a single depth;
2. Multi-sampling wells for sampling’ at different depths;
and
3. Piezometers which are designed to obtain samples
utilizing air lift methods.
Well Construction——
Wells are constructed by the following principal methods
(Everett, etal., 1976):
• Driven;
• Augered;
• Jetted;
• Cable tool (percussion); and
• Rotary drilled (i.e., direct-circulation, reverse—
circulation, air-circulation).
313
-------
LEGEND
WATER LEVEL
IN WELLS
DISPOSAL
AREA
-
WATER LEVEL
D IRECT ION
OF
—
CONTOURS
GROUNDWATER FLOW
Figure 66. DeterminatiOn of approximate ground-water flow
direction (Morrison and Brewer, 1981).
314
90:7 ’
99.5,
5’
-------
Driven-—Driven wells (i.e., sa.ndpoint wells) consist of a
casing 3 to 5 cn (1¼ to 2 in.) in diameter attached to a steel
drive Point which is slotted. These wells are driven to depths
generally up to 9 in (30 ft). Wells points are most useful in
unconsolidated soils devoid of rocks or pebbles or deposits of
hard materials. The chief disadvantage of driven wells are
that coflst.rUCtiOfl is slow and difficult when tightly compacted
materials are encountered; however, they are useful for monitor-
ing shallow aquifers in unconsolidated rock.
Augered--ShallOW wells may be augered by hand or power
equipment. Hand augers are useful in penetrating unconsolidated
deposits less than 4.5 in (15 ft). Moxu.tor .ng wells to depths
f 46 to 61 rn (150 to 200 ft) with holes diameters ranging from
to 81 . (2 to 32 in.) can be constructed in unconsolidated
formations by use of a drilling rig equipped with a power auger.
casing is usually required for deeper holes and during drilling
into the saturated zone. Where hollow—stem augers are used, a
casing and well point can be set inside the hollow stem to the
required depth.
Jetted--Water jetting and associated hydraulic approaches
use a high-velocity stream of water or jetting fluid for drill-
ing. The water stream Loosens the material and transports it
to the surface. The jetting method is most successful in sandy
soils with shallow water tables. The technique Ls ineffective
against hard rock and boulders; compacted clay and hardpan also
presents problems. The casing is usually sunk as the jetting
proceeds and if too much resistance is encountered, the casing
may be driven. Water jetted wells are usually of small diameter
and less than 46 in (150 ft) deep.
Cable tool (percussion)——In cable tool (percussion) drill-
ing, he hole is formed by the percussion and cutting action of
a drilling kit that is alternatively raised and dropped. The
drill cuttings are removed at intervals by bailer or sand pump.
Casing is not required in consolidated rock, but in unconsoli-
dated formations, casing is driven down the hole during
drilling. The bottom of the casing is fitted with a hardened
steel drive shoe. As the casing is driven in unconsolidated
formations, frictional forces increase until the casing cannot
be driven further without the use of special tools. Small
cable tool rigs can usually drill a 13—cu (5in.) diameter hole
to a depth of about 91 in (300 ft).
Rotary drilling- — Rotary drilling has virtually no depth
limit for most ground-water systems. In direct-circulation
(conventional) rotary drilling, the drill stern rotates downward
as drilling mud is p nped through the inside of the drill oipe
and out through the openings in the kit. Biodegradable rnu s
are presently in use which do not contaminate the waterbearing
formation. The principal advantages of direct-rotary drilling
315
-------
are speed, great depth capability, and the ability to ri
electric logs in an open hole prior to casing installation.
In the reverse—circulation rotary method, the drilling
fluid is introduced down the hole outside of the drill pipe.
The drill cuttings are pumped up through the drill pipe as
drilling proceeds and are separated from the drilling fluid at
the surface.
The air-circulation rotary method is similar to the other
rotary methods, with the exception that compressed air is used
instead of drilling mud to bring the cuttings to the surface.
En consolidated rock, the conventional drill kit can be replaced
with a pneumatic drill kit to speed up the drilling.
Costs of Drilling——
In a well-constructed program, the cost of drilling can be
separated from well development and can be approximated. In
general, drilling costs range from $75 to $200/hr or $16 to
$66/rn ($5 to $20/ft) The cost of drilling will vary primarily
with hole diameter, type of geologic formation and depth of the
well. Drilling costs for 12- to 61-rn (40- to 200- ft) depths
in unconsolidated formations for various well diameters are
given in Figure 67. Similarly, drilling costs for consolidated I
formations are given in Figure 63.
Well Development-—
Well casing——Campbell and Lehr (1973) summarized five types
of casing material used for monitoring wells:
1. Standard pipe;
2. Line pipe;
3. Reamed and drifted pipe;
4. Drive pipe; and
5. water-well casing.
Monitoring well economics and possible interference with
identified pollutants will usually determine material selection.
For purposes of water sampling, plastic (e.g., PVC, fiberglass—
reinforced plastic pipe) well screen and casings are preferred
because of their chemical inertness which minimizes contamination
of collected water samples.
* Recent survey (1981) by Calscierice Research, Inc., in Southern
California area.
316
-------
DEPTH ( in)
cn
C13
U
z
6000
5000
4000
3000
2000
1000
00 200
DEPTH OF WELL (ft)
‘1gure Costs of well drilling in inconsolidated formations,
December, 1981 (after Everett, et al., 1976).
60
50 100 150
317
-------
DEPTH (m)
45 60
0 50 100 150
DEPTH OF WELL Cf t)
TIgtire 68.
Costs of well drilling in consolidated formations,
December, 1981 (after Everett, et al., 1976).
15
30
a
C
z
-4
-4
3
8000
7000
6000
5000
4000
3000
2000
1000
0
200
318
-------
Other well casing materials are ceramic tile, concreta, and
asbestos cement. Less coirmton materials such as stainless steel,
cupro—nickel alloys, silicon bronze, aluminum, and other non-
ferrous metals, can be used for casing in special situations
when the natural soil and water quality conditions dictate their
emp1Oyr eflt (EPA, 1975).
Polyvinyl chi.oride (PVC) casing has increased in use in
recent years. Maximum installation depths for PVC casing are
normally less than 61 m (200 ft). PVC casing costs for diameters
of 5 to 25 cm (2 to 10 in.) and lengths of 6 to 61 m (20 to 200
ft) are given in Fig e 69.
Well screens-—In unconsolidated materials, openings in the
casing must be provided to permit entrance of water. The size
of perforations, slots, or screen openings are chosen with re-
spect to particle size distribution of the water bearing-zones.
Entrance velocity into the screen is usually recommended to be
less than O..l to 0.2 feet per second. Well screens are available
jn iron, stainless steel, brass, fiberglass, and PVC (Everett,
1976; EPA, 1975).
Gravel packing— — After casing and screen have been in-
stalled, a gravel pack is placed around the screen. The object
of gravel packing is to provide a zone of material coarser than
the natural material. The gravel pack in conjunction with
ap ropriateSiZed screen openings, retain fine material while
pe mitting water to enter without excessive head loss (EPA,
1975)
Gravel packing is also used for formation stabilization.
5jnce drilling by the rotary method through unconsolidated
materials results in a hole somewhat larger than the outside
diameter of the casing, the annular space around the well
screen is gravel packed to prevent silt and clay above the water
table from caving or slumping into the well producing zone
(Everett, et al., 1976).
Well sealing (grouting ) -—Well sealing or grouting isolates
the screened zone. Grouting materials include portland cement,
bentOflite, pazzolana, perlite, diatomacecus earth, Gilsonjte,
and mixtures of these. rcm a sampling standpoint, well sealing
jg vital to protect the producing zone from contamination, with-
out the seal, rainwater would infiltrate the backfill which could
result in dilution of the collected samples, or fluids could move
downward, from other levels causing the samples to be unrepre-
sentative (Everett, 1976 Fe , 1977).
Total Well Construction Costs——
While general data have been presented for drilling and
casing costs, additional costs such as for screens, gravel
319
-------
DEPTH (m)
45 60
3500
3000
2500
q .
— 2000
cJ
rj
C
L 1500
z
cI
C-,
1000
500
0
0 50 100 150 200
DEPTH OF CASING (ft)
Figure 69. Costs of PVC casing, December, 1981, (after
Everett, et al., 1976).
15 30
320
-------
oacking, grouting, well clean—up, etc. are highly dependent on
local conditions. It is therefore, appropriate to prov.de sep-
arate cost data for completed wells in consolidated and
unconsolidated formations. Total well construction costs
(1981 dollars) based on 1974 costs (Everett, et al., 1976) for
consolidated and unconsolidated formations are presented in
Figures 70 and 71, respect .vely..
Monitoring Well tnstallations
sampling equipment is designed to meet specific monitoring
requirements. The applicability and costs of single screened
wells,, well clusters, and air-lift samDler wells are discussed
in the following sub—sections. The advantages and disadvantages
of each well system are s mnarized in Table 72.
Single screen wells——A PVC well slotted at a designated
level within an aquifer constitutes the most cotmnon ground—water
sam ling approach (Figure 72). A serious.limitation of this
method; however, is its inability to provide a. representative
vertical distribution within the aquifer due to vertical mixing
within the well casing. Point source sampling at discrete depths
is a desirable characteristic which presently is not obtainable
with a single screen well (Morrison and Brewer, 1981).
Costs of single screen wells for diameters ranging from 5 to
15 cm (2 to 6 in.) and installed depth of 3 to 61 in (10 to 200
£t), are presented in Figure 73. Costs include PVC casing,
PVC slotted well monitoring screen, and bentonite seals. See
FigureS 67 and .69 for d .rilling.costs.-
Well clusters-—A variety of devices have been proposed which
are capable of samp] .i.ng at predesignated levels within an aquifer.
These approaches are inherently more useful than a single screen-
ed well, especially if only a single boring is required. One
approach to the problem is the placement of several email-
diameter wells within a single boring (Figure 74). Each cfl ster
00 nsist of a group of spaced small-diameter wells installed at
&jfferent depths. These wells provide for water samples that
are representative of the different levels within the aquifer.
Judicious placement of well clusters at a landfill site and the
5 rrounding area will allow delineation of both vertical and
areal distribution of contaminants (Fenn, et A2• 1977; Morrison
and Brewer, 1981). Costs of cluster wells depend On bore—hole
jiling costs, the number and diameter of the wells in each
cluster, and the depth of each monitoring well. Drilling costs
(Figures 67 and 68) and thonitoring well costs (FLgure 73) c3n-
be u3ed to develop cluster well costs for site—specific situa-
tions.
Air—lift samplers-—Air—lift samplers utilized for monitoring
in the zone of saturation generally refer to any method in which
321
-------
DEPTH r )
5000
a
cl
3000
8
2000
1000
Note:
15
30
45
Total costs include the fo11 w—
ing:
0 50 100
150
DEPTH 0 WELL (ft)
Figure 70.
Total costs of 4-, 5—, and 6-inch diameter
wells in consolidated formations,• December
1981 (after Everett, et al., 1976).
Setting up and removing
drilling equipment.
• Drilling.
• Installing casing only in
the unconsolidated section
of the bore hole.
• Grouting and sealing.
• Installing well
screens and
fittings.
• Developing
the iel1.
4000 —
*
1 in. = 2.5 n
—
—
200
322
-------
DEPTH (In)
15 30 45 60
5000
4000
3000 —
2000
1000
I I
I I
0
0 50 100 150 200
DEPTH OF WELL (ft)
Figure 71. Total costs of 4-, 5—, and 6—inch diameter wells
in unconsolidated formation, December, 1981 (after
Gibb, 1971)
Note: Total costs include those items
listed for cor .solidated forma’
tions (Figure 18) with the
exception that casing is
installed to the
bottom of the bore
hole
*
1 in 2.5 cm
a
cI
C
C.)
p
323
-------
TABLE 72. ADVANTAGES AND DISADVANTAGES OF ZONE OF SATURATION MONITORING DEVICES
( Everett,et al., 1976; Fenn,et al., 1977; Wilson. 1980: Morrison and Brewer, 1981 )
‘U
Monitoring
Advantages
DiBadvantage s
Device
Single
Shallow wells are cost effective
Provides samples from only one
Screened
and provide easy installation,
depth.
Wells
Variety of drilling methods are
available for installation,
Improper screening may provide
erroneous results.
Easy to rep iir if damaged.
Excessive screening may contrib-
ute to vertical leachate move-
ment.
Multi-Probe
Wells (Well
Detailed vertical distribution
of a contaminant may be obtained,
Difficult to repair (e.g.,
single well, multi-part config-
Clusters)
.
Established methodology for verti—
cal and areal distribution of
contaminants,
uration),
Sealing between screens may be
difficult for multi-cluster
wells.
Easily installed.
Proper well construction and
sampling procedures are recjuired
for successful application.
Air—Lift
Sample collected from a specific
Difficult to repair if damaged.
Samplers
section of the aquifer.
Relatively inexpensive and easily
installed if the casing diameter
is small.
Improper placement of the device
may cause problems in contamin-
ant determination.
Provides vertical sampling of the
aquifer If properly engineered.
Tubing susceptible to clogging
or breakage.
-------
Flush Joint Riser
azuped Bentonite Seal
Flush Joint
Slotted Monitor
Screen With
Plugged Bottom
______ Protective Cover
• Optional
Screw -on Black Iron
Vented Protective Cap
Concrete Surface
Seal
Tamped Virgin 9ack i1l
Clean Washed Sand
Or Pea Gravel
Figure 72. Single screened groundwater monitoring well (Ti co,
1980).
Typical Bore Role
325
-------
DEPTH (in)
15
45
DEPTH OF MONITORING WELLS (ft)
60
Figure 73. Costs of 2—, 3-, 4-, 5-, and 6—inch diameter monitor...
ing wells, December, 1981 (after Timco Mfg. Co., Inc.,
1980).
30
25
200
1500
cI
rj
C
C.,
ca
z
0
E.
0
z
1000
0 50 100 150
200
326
-------
Locking Cover .- -
I n F ILartd
:t. • r 1 J/ :1 ;
:-••. .. ...-•... b •• • r Mcrii.tcr rt tiells
- - -. .. I- — t- . .
. ..•.;.. . : • :::..::. .•.:-.• ••. . :-
- around Wat r Leve’,. - ..
.: • / -
—. - •—I. “ . — — U 1 lIPU I ___
—
S S — t S. - S
I’ & r — -
—
• - -- : ‘ ‘ . — 3ac i -
t __ : : .
•: •. ..
a i’: : . a
j 1
— .. — S — — t .. S
a • _ . ‘—
• ‘ . .—
‘ • •
• - s — . .. .. ?• i• : : ?_ -. • _ • t - •:- •. . •
. V + i-: • .. — —
E •: •. . . •.‘ • .• . •• L
• • -ç — — S.
Figure 74. Well c lt sters (Morrison and Brewer, 1981).
[ Reproduced from f7’
Lbest available
327
-------
a cressurized carrier gas is used to t ansport a sample to the
surface. while a variety of designs exist, a check-valve
arrangement, which allows water entry and closes when a carrier
gas is injected into the sampler, is a ccz imon feature of these
devices. The ability to collect samples at various de ths
within a single boring is a valuable characteristic of an air-
lift sampler; many investigators have employed air—lift samolers
exclusively for this purpose (Morrison and Brewer, 1981).
The installation of air-lift samplers consists of two
approaches. The selection depends upon whether one sampler
or a series of samplers is placed within the boring. In the
case where one sampler is used (Figure 75), the sampler can...
be threaded onto a well casing and placed at the desired depth.
The input/output tubes are situated within the well casing and
lead to the surface. Costa of a single air-lift sampler for
3.2 cm (1¼ in.), 3.8 t (1.5 in.) and 5.1 cm (1½ in.) diameters
and installed depths of 3 to l m (10 to 200 ft), are presented
in Figure 76. Costs include slotted PVC air—lift sampler,
PVC casing and collection tubing, and bentonite seals. See
Figures 67 and 68 for drilling costs.
Where several air-lift samplers are required in a vertical
section, the casing is omitted and the tubing collected at the
surface as shown in Figure 75. This array provides greater
flexibility, especially in three dimensional, volume monitoring.
Costs of multiple a±r—lift samplers depend on the diameter and
n mther of installed units, and the depth of the sinqle boring.
Drilling costs (Figures 67 and 68)and air—lift sampler costs
(Figure 76) cam be used to develôpmultiple air-lift sampler..
Costs for site—specific situations.
MONITORING APPROAC S AND CONSIDERATIONS
Selection of Sam 1ing Locations
Vadose Zone——
Monitoring within the vadose zone underlying a. landfill
site affords an early warning of potential ground-water po].-
lution. If remedial measures are implemented prior to the
onset of pollut .on, the associated renovation costs will be
eliminated or reduced. Concomitantly, vadose zone monitoring
may reduce the need for extensive ground—water monitoring, i.e.,
if a vadose zone monitoring program fails to detect the movement
of contaminants, the requirements for ground—water monitoring
may be reduced or largely precluded (Wilson, 1981). The savings
in costs for constructing ground-water wells could be signjf j-
cant, particularly in western regions where water taMes are
often hundreds of feet deep. Beca ise of the limited use of
vadose samplers; however, little literature concerning their
placement is available. For this reason, general guidelines
are presented for the vadose zone.
328
-------
Gas Entry/Collection
Tuties
Bentonite Seal
..—rr .
9
j er
I,
Figure 75. Single (A) and multiple (B) installation configurations for air-lift
samplers. (Morrison and Brewer, 1981).
Locking Cover
P
-------
Depth ( in)
30
45 60
DEPTH OF AIR LIFT SAMPLERS (ft)
200
Figure 76.
Costs of 1¼-, 1½—, and 2—inch diameter air lift
samplers, December, 1981 (after Timco Mfg. Co., Inc.
1980)
20
15
Mote:
Costs include PVC slotted
air lift sampler, PVC
collection tubing, and
bentonite seals
* 1 in.
10
a
rfj
0
C.)
c;4
0 50 100 150
330
-------
Selection of an optimum location for sampling and non—
sampling devices within the zone of aeration is largely site—
soecific. Figure 77 was d velo ed. for several h qthetica1
gi .oundwater situations which may be encountered at disposal
facilities and the corresponding placement of vadose monitoring
devices within each system. In all cases, the water collection
devices are situated below the landfill rather than within the
refuse because of the need for developing and maintaining contact
between the interstitial waters and the porous ceramic cup. In
models 0 and E, non—sampling devices are also included in the
array. In model 0, these devices may be used to identify
optimum periods for water collection. In model E, the non—
sampling devices and any ground—water wells determine the
exact depth to the zone of saturation. The vadose zone sampling
devices may actually be employed to collect samples which are
within the zone of saturation. Figure 78 illustrates an actual
monitoring program as described by Johnson and Cartwright (1978)
at the Genesco landfill in Illinois, in which vadose samplers
and ground—water wells were successfully used in an integrated
system.
zone of Saturation——
A monitoring program is established to give a prompt indic-
ation of ground-water contamination. The size of the landfill,
ydrogeolOgic environment, and budgetary constraints are factors
which will dictate the actual number of installed wells. The
spacing and depths of monitoring wells depends on the particular
pattern of groundwater flow, making it extremely difficult to
specify national minimums or maximums in this area. It is
recotnmended that the following minimal monitoring well network
be implemented in order to detect and evaluate potential
grouzidwater contamination at a sanitary landfill (Fenn, et al.,
1977; EPA, 1980). — —
• One well upgradient from the landfill to character-
ize ground-water ality in the Uppermost aquifer.
The owner or Operator must assure that the upgradien’t
samples represent true background conditions and
are not contaminated by the facility;
• One well placed adjacent to the downgradient waste
boundary to give a prompt indication of ground-
water contamination; and
• A minimum of three wells downgradient of the landfill
to provide representative samples capable of detecting
migration of contaminants from the facility. Once
contamination is detected, additional lines of wells
can be constructed downgradient to gauge dispersion
and attenuation of the leachate, thereby-providing
the necessary information for predicting the ultimate
331
-------
I1J f 7Y3 ffAI
;. F! IF!’! 3 SJW IIcN
Q ‘ ATI 4
-------
M B_ D
SIANITMY tJNDFILL
GR iCWATER L V
SN UNG ceitcz S
ING DEVtC 0
BLCCX, ETC.)
Figure :• (Continued).
I Reproduced from
best availabfe copy.
D.
WEWIICAL TO MGDE .. C JiJ’
E, MELD S !PTTCN
o e m ro ZONE CF SATURATtCN
FUJCTUATES SIGNtFtCM4TLY
333
-------
—4 .1.)
(a
0
—4
I_I h
4C
C,
— a
C)
C,
C,:
.p4
w
0
0
— C,
C,
“4
0
00
- p4
C)
‘-4
N
334
-------
fate of the plume. This approach will be an expen-
sive and tine—consuming process, the necessity for
which will depend on regulatory requirements.
Monitoring Frecuencv
Monitoring frequency is greatly influenced by many factors,
jncluding (Fenn, at al., 1977):
• Purpose for monitoring (e.g., compliance versus
investigatory; routine versus synoptic);
• Background quality of ground water at the site;
• Characteristics of ground—water flow;
• Monitoring network utilized;
• Location and purpose of the particular monitoring
well;
• Other sources of ground-water contamination;
• Trends in the monitoring data;
• Legal and institutional needs;
• Climatologica3. characteristics; and
• Site history.
The monitoring frequency should be flexible, allowing for
modification to meet the individual needs of each site. One
tiseful monitoring approach combines synoptic and routine sam-
1ing. The synoptic sampling provides data during events
representing the greatest pollution potential (e.g., heavy pre-
cipitation, snow melt, flooding). The scheduled sampling
measures seasonal and other long term phenomena. Because of
site—specific requirements, it is not possible to delineate
national maximums or minimums in this area.
nitor±flg Parameters
The parameters selected for study will be determined large—
y by the purpose of the monitoring program. Many researchers
have developed lists of key indicator parameters (Goerj.itz and
Brown, 1972; EPA, 1973; EPA l975 Fenn, etal., 1977; EPA, 1979,
EPA, 1980). Such lists can be referred to for the purpose of
3 jloring site—specific monitoring programs. EPA (1980)
estabhi5 guidelines for ground-water monitoring of three
gets Of parameters, each of which serve a different ouroose.
These key parameters may be modified, as required, t m et
335
-------
site—specific situations. The three sets of rameters are
summarized as follows:
1. The first set reflects the aquifer’s suitability
as a drinking water supply. These parameters,
are those specified in the Interim Primary Drink-
ing Water Regulations established under the Safe
Drinking Water Act. The purpose of the initial
sampling for drinking water parameters is to
identify facilities that may be severely degrading
present and future drinking water supplies. It
should be noted that the specified parameters
do not comprise a complete list of contaminants
that define an aquifer’s potential as an accept-
able drinking water supply; however, these
constituents will be useful in establishing
monitoring priorities;
2. The second set of parameters includes chloride,
iron, manganese, phenols, sodium, and sulfate.
These parameters are generally recognized as
useful for cha acteriZiflg ground—water quality.
These contaminants are ubiquitous in the environ-
ment and are often used to characterize a
ground-water supply’s suitability for a variety
of uses. Information Ofl these parameters will
be useful in any assessment of ground—water
contamination that follows the determination
that a facility is leaking. Such information
will, for example, assist in determining the
extent to which contamination of the aquifer
may be coming from sources other than the dis-
posal facility; and
3. The third set of parameters consists of four
indicators that will be used to determine whether
a facility is leaking. It is important for a
facility to answer the threshold question of
whether hazardous waste constituents are entering
the aquifer underlying the facility. The four
indicators-—specific conductance, pH, total
organic carbon, and total organic halogen——
reflect changes in the organic and inorganic
makeup of the ground water. A statistically
significant change (increase or decrease for
pH, increase only for the others) in these
indicators between the initial background
concentration or value and those from down—
gradient wells suggests that organic or inorganic
substances are being introduced into the aquifer
by the facility. Increases in specific conductance
indicate the presence of inorganic substances in
336
-------
the ground water. Likewise, increases or decreases
in pH suggest the presence of inorganic contamination.
Total organic carbon (TOC) and . total organic halogen
(TOX) concentrations in ground water tend to increase
as a result of organic contributions from a solid
waste facility. The ntethodologv to sample and
analyze for these indicators is presently available.
Monitoring these indicators will be sufficient
to make the threshold assessment of whether a
facility is leaking.
Sam 1e Collection Methods Containers and Sample Preservation,
Analytical Procedures .
sample Collection——
Either a vacuum or pressure device may be used to withdraw
samples from lysimeters (Wood, 1973). For Lysimeters placed
deeper than 20 ft, pressure devices utilizing nitrogen as a
carrier gas have been found effective (Morrison and Brewer,
1981). For non—sampling devices, differences in the moisture
content may be recorded continously or at selected times.
The method chosen for sampling within the zone of satura-
tion depends on the depth of the water table. A Suction—lift
is installed above a well. The maximum suction lift is
iimlted by atmospheric pressure, head losses due to friction,
and the required inlet head of the pump. At sea level, the best
signed pumps usually achieve a suction lift of 7.6 m (25 ft),
while the suction lift of an average pump varies from 4.6 to
5.5 ttt (15 to 18 ft) (Everett,- .etal., 1976). Where the depth
o the water table is greater than the practical suction ijinit,
peristaliCi centrifugal,or submersible pumps may be
p1cyed.
Submersible pumps require a minimum casing diameter of 10
(4 in.). A submersible pump is feasible where a number of
0 nitoring wells are located in a small area and only periodic
0 ping is required (Everett, et al., 1976). Submersible pumps
are considered the best, but most exoensive, method of collect-
ing water samples (Penn, . ,1977). M Mil1ian and Zeeley
(1968) reported the use of a truck portable submersible puma
capable of withdrawing water samples at depths up to 91
(300 ft) inside a casing as small as 11.4 cm (4½ in.). With
jg equipment, cross-contamination between wells from the
5 ampling equipment will be virtually eliminated by the flushing
3 cti°fl during pre-pumping.
Water samples may also be taken by bailing. Where pumping
6 qUipme t is not available, or where a pump cannot be installed
because the diameter of the well too small, a container such
aS a weighted bottle or a short section of pipe capped at the
bottom can be lowered into the well to collect a sample. This
337
-------
sample will give an indication of the chemical quality of the
water in the aquifer, but should not be used for bacteriological
or detailed chemical analysis because of the likelihood of
contamination. This method is not recommended except where other
methods of sampling are unavailable (Everett, et al., 1976).
Air lift sampling can be useful in monitoring wells that
need to be pumped only at periodic intervals. The advantage
of this sampler is that the apparatus can be designed for perma-
nent installation in the monitoring well. The air-lift sampler
eliminates the possibility of cross-contamination between wells
as can occur with portable pumping and sampling devices. This
method is also of value in obtaining bacteriological samples
where external sources of contamination must be avoided (Fenn,
et a].., 1977). Johnson Division UOP (1971) presents data on
the relation between total lift, submergence of air lines, and
water discharged per cubic foot of air delivered by an air
compressor.
Where it is desired to collect water from a specific depth
within a well, a special cylinder known as a “thief sampler,”
can be lowered by cable into the well and closed at a predeter-
mined depth. Most types consist of a cylinder that can be less
than 5 cm (2 in.) in diameter. A bar with a plunger or cork at
each end rums through the cylinder. When the sampler is lowered
below the water table to the desired depth, the two ends are
closed. The most cor non type of “thief sampler” is spring
loaded. With this type, a messenger is lowered along the cable
to release the two end cups, which in turn trap the water sam-
ple. Many “thief samplers” are custom made for a specific use
(Everett, et a].., 1976). Additional detailed descriptions of the
operation and application of different types of ground-water
samplers are available in the literature (Cherry, 1965; Trescott
and Pinder, 1970; Son,merfeldt and Campbell, 1975).
Containers and Sample Preservation——
EPA (1979) and APHA (1981) provide a detailed presentation
of appropriate containers for various chemical species and sam-
ple preservation techniques. Methods of preservation are
relatively limited and are intended to:
(1) Retard biological action;
(2) Retard hydrolysis of chemical compounds and corn—
plexes; and
(3) Reduce volatility of constituents.
Preservation methods are generally limited to pH control,
chemical addition, refrigeration, and freezing. Recoimnended
containers, preservatiVeS and holding times for various samples
are given in Table 73 (EPA, 1979).
338
-------
TABLE 73. RECOMNENDATIONS FOR SAMPLING AND PRESERVATION
OF SAMPLES (EPA, 1979).
Sample V
olume
Holding
Measurement
Reciijred (ml) Container
Preservative Time
* .
Acidity 100 P, C Cool, 4°C 24 Mrs.
Alkalinity 100 P,. C Cool, 4°C 24 Mrs.
A rsenic 100 P, G ENO 3 to pM<2 6 Mos.
1000 P, G Cool, 4°C 6 Mrs.
rornide 100 P, C Cool, 4°C 24 Mrs.
COD SO ?, G H 2 S0 4 to pH<2 7 Days
chloride 50 P, C None Req. 7 Days
Color 50 P, C Cool, 4°C 24 Mrs.
cyanides 500 P, G Cool, 4°C 24 Mrs.
NaOM to pH 12
Dissolved Oxygen
probe 300 C only Det. on site Mo
Holding
winicler 300 C only Fix on site 4—8 Mrs.
‘j .uoride 300 P, C Cool, 4°C 7 Days
ardneSS 100 P, C Cool, 4°C 7 Days
O3 to pE<2
1 odide 100 P, C Cool, 4°C 24 Mrs.
250 P, C Cool, 4°C 24 Mrs.
Metals
Dissolved 200 Filter on site 6 Mos.
ENO 3 to pH<2
Suspended Filter on site 6 Mos.
Total 100 MNO 3 tO pH 2 6 Mos.
MercurY
Dissolved 100 P, C Filter 38 Days
HN0 3 to pE<2 (Class)
13 Days
(Hard
Plastic)
(continued)
339
-------
TABLE 73 (continued)
Sample Volume Holding
Measurement Recuired (ml) Container Preservative Time
Total 100 P 1 C O3 to pH<2 38 Days
(Glass)
13 Days
(Hard
Plastic)
Nitrogen
knxnonia 400 P, C Cool, 4°C 24 Hrs
E2504 to pH<2
Ejeldahi, 500 P, C Cool, 4°C 7 Days
total H 2 50 4 to pH<2
Nitrate 100 P 1 C Cool, 4°C 24 Ers.
H 2 S0 4 to pH<2
Nitrite P, C Cool, 4°C 24 Ers.
NTA 50 P, C Cool, 4°C 24 Hrs.
Oil & Grease 1000 G only Cool, 4°C 24 Ers.
H 2 S0 4 or
Ed to pH<2
Organic Carbon 25 P, C Cool, 4°C 24 Ers.
H 2 50 4 to pH<2
25 P, G Cool, 4°C 6 His.
Det. on site
phenolics 500 C only Cool, 4°C 24 His.
H 3 P0 4 to pH<4
1.0 g Cu50 4 /1
Phosphorus
Ortho-
phosphate, 50 P, C Filter on site 24 His.
Dissolved Cool, 4°C
Hydrolyzable 50 P, C Cool, 4°C 24 His.
H 2 S0 4 to pE<2
Total 50 P, C Cool, 4°C 7 Days
Total,
Dissolved 50 P, C Filter on site 24 His.
Cool, 4°C
Residue
Filterable 100 P, C Cool, 4°C 7 Days
(continued)
340
-------
TABLE 73. (concluded)
Measurement
Sample Volume
Recuired (ml)
Container
PreservatIve
Holding
Time
Non-
Filterable
100
P,
G
Cool, 4°C
7 Days
Total
100
,
Cool, 4°C
7 Days
Volatile
100
P
C
Cool, 4°C
7 Days
Settleable
Matter
1000
P,
C
None Req.
24 Mrs.
Selenium
50
P..
C
HNO 3 to pM<2
6 Mos.
i1ica
50
P
Only
Cool, 4°C
7 Days
Specific
Conductance
100
P,
C
Cool, 4°c
24 Mrs.
Sulfate
50
P,
C
Cool, 4°C
7 Days
Sulfide
500
P,
C
2 ml Zinc
24 Mrs.
gulfite
so
acetate
Det. on site
No
Molding
Temperature
1000
P,
C
Det. °fl Site
No
.
Holding
Threshold
Odor
200
C
only
Cool, 4°C
24 Mrs.
TurbiditY
100
P,
C
Cool,4°C
7 Days
* p plastic (polyethylene or
t glass
equivalent)
341
-------
Analytical ?rocedures and Costs- —
EPA (1979) and APHA (1981) describe detailed analytical
rocedures which, in general, are quite comparable. The methods
are applicable to both water and wastewaters, and both fresh
and saline water samples. Tests procedures are given for the
measurement of physical, inorganic, and selected organic con-
stituents and parameters. Instrurtental methods have been
selected in preference to manual procedures because of the
improved speed, accuracy, and precisicn. Precision and accuracy
statements are provided where such data are available.
Analytical costs vary considerably in the U.S. The required
sensitivity, degreee of accuracy, n .m ber of samples, time
constraints, and laboratory location are major considerations.
Water quality analyses costs (1981 dollars) for a coxmr ercial
laboratory are given in Table 74 (West Coast Technical Services,
Inc., 1981).
SUMMARY
Monitoring equipment for the vadose zone may be classified
as sampling or non—sampling. Sampling approaches include pore—
water extraction from soil cores and deployment of pressure!
vacuun lysimeters to obtain in situ water samples for chemical
analyses. Non—sampling equipment includes tensiometers,
psychrozneters, electrical resistance blocks, and neutron
moisture logging. Non-sampling methods provide for the deter-
mination of water content and water movement in the vadose zone.
A zone of saturation monitoring program is established to
give a prompt indication of grouna-water contamination. The
size of the landfill, hydrogeologic environment, and budgetary
constraints are factors which will dictate the actual number of
wells. The spacing and depths of monitoring wells depends on
the particular pattern of grounduwater flow. Monitoring within
the zone of aeration has traditionally been accomplished with
single screened wells. More recently developed installations
such as the cluster well and the air-lift sampler well allows
for multiple sampling points throughout the aquifer. Multi—
sampling wells provide for the vertical distribution of contami-
nants in the aquifer. This information is of great value for
the determination of leachate migration and changes in ground-
water quality.
Monitoring frequency should be flexible, allowing for
modification to meet the individual needs of each site. One
useful monitoring approach combines synoptic and routine sam-
pling. Monitoring parameters selected for study will be deter-
mined largely by the purpose of the monitoring program. Many
researchers have developed lists of key indicator parameters.
The selection of key monitoring parameters should be modified,
as required, to meet site—specific situations.
.342
-------
TABLE 74. COSTS OF WATER QUALITY ANALYSES, DECEMBER, 1981.
(West Coast Technical Service, Inc., 1981)
Price per sample
priority Pollutants Wastewater Sludge
Organic $825 $1080
Metals (Be, Cd, Cr., Cu, Pb,
Ni., Ag, Th, Zn, Sb, As,
Se, Hg) 430 500
Samples submitted at one tine
(price per sample)
1—3 4—7 8+
pesticides and Herbicides by —
GC & LC
Organochiorirte $120 $100 $ 75
Triazthe 120 100 75
Organophosphorous 130 110 85
Fluralins 120 100 75
Carbamate and Urea 180 145 130
polychlorinated Biphenyls
(PC3’ s)
In waste water $150 $120 $100
In transformer oil 120 100 85
galcethers $120 $100 $ 75
Chlorinated Hydrocarbons 170 145 120
p htha lates 145 120 95
polynuclear Arontatics (PNA’s)
b GCMS 220 195 185
phenols by GcNS 200 180 170
3 enzidifle s 240 210 180
itrOarOmatiCS and
Isophorone 170 145 120
jtrosamifles 170 145 120
GC.’ . S Confirnation when
possible 180 180 180
343
-------
REFERENCES
Ahlstroxn, S. W., and R. C. Baca. 1974. Transport Model User’s
Manual. Batelle Pacific Northwest Laboratories, Report
BNWL—1716, UC—70. 25 pp.
Alexander, M. 1971. Microbial Ecology John Wiley & Sons, Inc.,
New York, N.Y. 421 pp.
American Colloid Co. 1979. Use of Bentonite as a Soil Sealant
for Leachate Control in Sanitary Landfills, Voiclay Soil
Laboratory, Eng. Report 280—E, Skokie, Ill. 36 pp.
American Cyanainid Co. 1975. All About Cyanaznid A5-9, Chemical
Grout. Wayne, New Jersey. 95 pp.
American Public Health Association. 1980. Standard Methods
for the Examination of Water and Wastewater, 15th Edition.
American Public Health Association, New York, N.Y. 1193 pp.
American Society of Civil Engineers. 1960. Design and Construct-
ion of Sanitary and Storm Sewers. Manuals of Engineering
Practice, No. 37, or Water Pollution Control Federation
Manual of Practice, Nc. 9.
Anderson, J. R. and J. N. Dornbush. 1967. Influence of Sanitary
Landfill on Ground Water Quality. American Water Works
Association J. 59 (4) : 457—470.
Apgar, M. A. and I D. Langutuir. 1971. Ground—Water Pollution
Potential of a Landfill Above the Water Table. In: Pro-
ceedings; National Ground Water Quality Symposium, LI. S..
Environmental Protection Agency, and the National Water
Well Association, Denver, Colorado, 1971. U. S. Government
Printing Office, Washington, D. C. pp. 76-96.
Asphalt Institute. 1966. Asphalt Lining for Waste Ponds. The
Asphalt Institute Information Series No. 136 C15—136.
College Park, Maryland.
Asphalt Institute. 1974. Specifications for Paving and Indus-
trial Asphalts. Specification Series No. 2 (55—22)
Asphalt Institute. 1976. Asphalt in Hydraulics (M542) College
Park, Maryland. 65 pp.
-------
Baily, G. W. , and 3. L. White. 1970. Factor Affecting the
Adsorption, Desorption, and Movement of Pesticides in
Soils. Residue Reviews Vol. 32, Springer—Verlag, N.Y.
Baldwin, V. A. 1979. Enviror.mental and Resource Conservation
Considerations of Steel Industry Solid Waste. EPA—600/2-
78—074, u.S. Environmental Protection Agency, Cincinnati,
Ohio. 78 pp.
BallauX, 3. C., and D. E. Peaslee. 1975. Relationship Between
Sorption and Desorption of Phosphorous by Soils. Soil
Sci. Soc. Amer. Proc. 39:275—278.
Barton, C. M. 1974. Bore Hole Sampling of Saturated tSrtcemented
Sands and Gravels. Groundwater 12 (3) :170-181.
Bayer, L. D. 1956. Soil Physics (4th ed.). John Wiley and
Sons, Inc., ew York, N. Y. 498 pp.
Bear, F. E. 1955. Chemistry of the Soil. Reinhold Pub. Co.,
New York, N. Y. 373 pp.
eluche, R. 1968. Degradation of Solid Substrate in a Sanitary
Landfill. Ph.D. Thesis, t3niversity of Southern California,
Los Angeles, California.
Bennet, F. W., and R. L. Donahue. 1975. Methods of Quickly
Vegetating Soils of Løs Productivety. EPA—440/9—75—0026.
u.S. Environmental Protection Agency, Washington D.C.
35 pp.
Bernard, M. 1932. Discussion of Runoff Rational Runoff Formu-
las. Trans. Am. Soc. Civil Engineers, 96:1161.
Besbes, N. E.., E. Ledoux, and G. de Marsily. 1975. Modeling
of the Salt Transport in Multi-layered Aquifers. In:
System Stimulation in Water Resources. van Steenkiste,
G. C (ed.). North-Holland Pub. Co., Amsterdam. pp. 229—
245.
Blake, G. R. 1965. Bulk Density. In: Methods of Soil Ana].y—
see, C. A. Black (ed.) Agronomy Mo. 9, American Society of
Agronomy, Madison, Wisconsin. pp. 375-390.
BlaflflOfl, J. z. and fli. L. Peterson. 1974. Survival of Feca].
Coliforins and Fecal Streptococci in a Sanitary Landfill.
News of Environmental Research in Cincinnati. U.S.
Environmental Protection Agency, Cincinnati, Ohio. 25 pp.
345
-------
Boucher, F. R., and G. F. Lee. 1972. Adsorption of Lindane
and Dieldrin Pesticides on Unconsolidated Aquifer Sands.
Environ. Sci. & Tech. 6:538—543.
Boyle, W. C. and R. K. Earn, 1974. Biological Treatability of
Landfill Leachate. Journal of the Water Pollution
Control Federation, 46(5):860—872.
Brakensiek, ID. L., H. 3. Osborn, and W. J. Rawis. 1979. Field
Manual for Research in Agricultural Hydrology Agricultural
Handbook No. 224. U.S. Dept. of Agriculture, Science and
Education Administration.
Bredehoeft, J. ID., and G. F. Pinder. 1973. Mass Transport
in Flowing Groundwater, Water Resources 9(1):194—210.
Brenner, H. 1962. The Diffusion Model of Longitudinal Mixing
in Beds of Finite Length: Numerical Values. Chern. Eng. Sci.
17: 229—243.
Bresler, E. 1973. Simultaneous Transport of Solutes and Water
Under Transient Unsaturated Flow Conditions. Water Re-
sources Research, 9(4):975—986.
Bresler, E. 1975. Two—DimensiOnal Transport of Solutes During
Nonsteady Infiltration from a Trickle Source. Soil Sci.
Soc. amer. Proc. 39(4):604—613.
Broad.bent, F. E. 1973. Organics. Proceedings of the Joint
Conference on Recycling Municipal Sludges and Effluents
on Land. Sponsored by the Environmental Protection Agency,
The U.S. Dept. of Agriculture, and the National Assoc. f
State Univ. & Land-Grant Colleges. Champaign, Ill., July,
9—13.
Brown, K. W. and Associates, Inc. Hazardous Waste Land Treat-
ment. EPA—SW-874, U.S. Environmental Protection Agei cy,
Cincinnati, Ohio. 891 pp.
Brown, K. W., and ID. Anderson. 1980. Effect of Organic Cheinj—
cals on Clay Liner Permeability. In: Disposal of Hazardous
Wastes. Proceedings of the Sixth Annual Research Symposj
EPA—600/9—80—OlO. U.S. Environmental Protection Agency,
Cincinnati, Ohio. pp. 123-124.
Brunner, ID. R. 1979. Forecasting Production of Landfill
Leachate. In: Municipal Solid Waste: Land Disposal.
EpA—600/9—79—023a, U.S. Environmental Protection Agency,
Cincinnati, Ohio. pp. 268—282.
346
-------
3urke Rubber Co. 1980. Installation Procedure for Burke
H alon 45 Pond/Pit Liners. San Jose, California. 11 pp.
Burrows, c . D. and R. S. Rowe. 1975. Ether Soluble Constituents
of Landfill Leachate. J. Water Pollution Control Federation,
47(5) :921.
Caffrey, R. P., and R. K. Ram. 1974. The Role of Evaporation
in Determining Leachate Production from Milled Refuse
Landfills. Compost Science, March/April: 11—15.
California State Water Pollution Control Board. 1961. Effects
of Refuse Dumps on Ground-Water Quality. Publication No.
24. Sacramento, California. 107 pp.
Campbell, S. C., C. S. Campbell, and W. R. Barlow. 1973. A
Dewpoint Iygrometer for Water Potential Measurement.
Agricultural Meteorology. 12: 113—121.
Campbell, M. D., and J. if. Lehi. 1973. Water Well Technology.
McGraw—Rill, Tew York, N.Y. 681 p.
Carison, J. F. Jr., F. S. Woodward, D • F. Wentworth and 0. J.
Sptoul. 1968. Virus Inactivation on Clay Particles in
Natural Waters. J. Water Pollution Control Federation.
40:R89—R106.
carroll, 0. 1959. ton Exchange in Clays and Other Minerals.
Geological Society of ? znerica Bulletin. 70:749—780.
chang, A. C., and A. L. Page. 1979. Rydraulic and Water Quality
Considerations of Artificial Ground Water Recharge Proceed-
ings of the Tweith Biennial Conference on Ground Water,
Sacramento, California, Sept. 20—21. California Water
Resources Center, University of California, Davis Report
No. 45. pp. 88—99.
chan, ic Y., B. C. Davey, and if. R. Geering. 1979. Adsorption
of Magnesium and Calcium by a Soil with a Variable Charge.
Soil Sci. SOC. Pm. J. 43:301—304.
Charlie, W. A., R. E. Wardwell, and 0. 3. Andersiand. 1979.
Leachate Generation from Sludge Disposal Area. J. Environ-
mental Engineering, Div. Proceedings of ASCE, 105(EE 5):
947—960.
Chen, K. Y., i. C. Lu, and 3. A. Eichenberger. 1980. Evaluation
of Potential Water Quality Impacts from Coal Utilization
Waste Disposal under the National Energy Plan. Report Sub-
mitted to Energy and Environmental Systems Division,
Argonne National Laboratory, Argonne, Illinois. 120 pp.
347
-------
Chen, K. Y. and B. A. Eichenberger. 1981. Evaluation of Costs
for Disposal of Solid Wastes from High—3TU Gasification.
Energy and Environmental Systems Division. Argonne National
Laboratory, Argonne, Illinois. 127 pp.
Chen, Y. S. R, J. N. Butler, and W. Stumin. 1973. Kinetic Study
of Phosphate, Reaction with Aluminum Oxide and Kaolinite.
Environ. Sci. & TeCh. 7:327—332.
Cherry, R. N. 1965. A Portable Sampler for Collecting Water
Samples from Specific Zones in Uncased or Screened Wells.
Professional Pater 525—C. U.S. Geological Survey. p. 214—
216.
Chian, E. S. K. and F. B. DeWalle. 1975. Compilation of Method-
ology Used for Measuring Pollution Parameters of Sanitary
Landfill Leachate. EPA-600/3-75—011. U.S. Environmental
Protection Agency, Cincinnati, Ohio. 125 pp.
Chian, E. S. and F. B. DeWalle. 1976. Sanitary Landfill
Leachates and Their Treatments. J. of the Environmental
Engineering Division, ASCE, 102 (EE2) :215-239.
Chain, E. S. K. and F. B. DeWalle. 1977. Evaluation of Leachate
Treatment, Volume 1, Characterization of Leachate. EPA—
600/2—77—186a. U.S. Environmental Protection Agency,
Cincinnati, Ohio.
Chian, E. S. K. and F. B. DeWalle. 1979. Effects of Moisture
Regimes and Temperature on MSW Stabilization. EPA-600/9—7g —
023a, U.S. Eiwironmental Protection Agency, Cincinnati, Ohio.
pp. 32—40.
Chow, V. T. 1964. (Editor-in-Chief). Handbook of Applied
Hydrology. McGraw-Hill Book Co., New York.
Chun, N. J., R. H. F. Young, A. S. Kawatachi, and P. R. Buldoc.
1975. Ground-Water Pollution From Sanitary Landfill Leach-
ate, Oahu, Hawaii. Technical Report No. 87, U.S. Dept. of
the Interior, Washington, D.C. 80 pp.
Cleary, R. W. 1976. Unsteady State Multi-Dimensional. Analytical
Modeling of Water Quality. Proceedings of the Conference
on Environmental Modeling and Simulation. Wayne Ott (ed.).
EPA 600/9—76—016. pp. 434—438.
Cleary, R. W., and D. D. Adrian. 1973. Analytical Solution of
the Convective-Dispersive Equation for Cation Adsorption
in Soils. Soil Sci. Soc. Amer. Proc. 37(2):197—l99.
348
-------
cleary, R. • • c- O 1 and • Short. 1973. Unsteady
State Three-Dimensional Model of Thermal Diffusion in
Rivers. In: Water, inerican Institute of Chemical Engineers
S posium Series, 129(69):422—431.
Colman, E. A., and C. B. Bodman. 1944. Moisture and Energy
Conditions During Downward Entry of Water into Moist and
Layered Soils. Soil Sci. Soc. Am. Proc. 9(3):
cook, E. N. and E. C. Foree. 1974. Aerobic Biostabjljzation of
Leachate. Journal of the Water Pollution Control Federation,
Vol. 46:380—392.
Cook, H. H., 0. C. Cromwell, and H. A. Wilson. 1967. Micro-
organisms in Household Refuse and Seepage Water from Sanitary
Landfill. Proc. W. Va. Acad. Sci. 39:137-114,
Coocer, R. C., S. A. Klein, C. J. Leong, J. L. Potter, and C. C.
Golueke. 1974a. Effect of Disposabi Diapers on the Cota—
position of Leachate from a Landfill. SER.L Report No. 74—3.
Sanitary Engineering Research Laboratory, College of Engineer-
ing and School of Public Health, University of California,
Berkeley, California. pp. 93.
cooper, R. C., J. L. Potter, and C. Leong. 1974b. Virus Survival
in Solid Waste Treatment Systems. In: Virus Survival in
Water and Wastewater Systems, J. F. Maline and B. P. Sagik,
eds. Water Resources Symposium No. 7, Center for Research
in Water Resources, University of Texas, Austin, Texas.
pp. 218—232.
County of Los Angeles, Department of County Engineers, and
Engineering-Science Inc. 1969. Development of Construction
arid Use Criteria for Sanitary Landfills; an Interim Report,
U.S. Department of Health Education and Welfare, Cincinnati,
Ohio. 267 pp.
Dass, P., C. R. Tamke, and C. M. Stoffel. 1977. Leachate
Production at Sanitary Landfill Sites. J. Environmental
Engineering Division, Proceedings of AScE, 103(EEG):
981—988.
vaison, J. M. 1978. Adsorption, Movement, and Biological
Degradation of High Concentrations of Selected Pesticides
in Soils. Proceedings of the Fourth Annual Research
Symposium. EPA600/9780l6e US. Environmental Protection
Agency, Cincinnati, Ohio. p. 233-244.
349
-------
Davidson, J. M. 1980. Movement and Biological Degradation of
Large Concentrations of Selected Pesticides in Soils. In:
Disposal of Hazardous Waste. Proceedings of the Sixth
Annual Research S posium. EPA-600/9-80—010. U.S. Environ—
mental Protection Agency, Cincinnati, Ohio. pp. 93-107.
Davidson, J. M., C. H. Brusewitz, D. R. Baker, and A. L. Wood.
1975. Use of Soil Parameters for Describing Pesticide
Movement Through Soils. EPA—600/275-009. U.S. Environ-
mental Protection Agency, Corvallis, Oregon. 187 pp.
Degens, E. T. 1964. Biochemical Compounds in Offshore California
Sediments and Seawater. Geochim. Cosmochim. Acta, 28 (45):
220—225.
De Wiest, R. j. M. 1967. Geohydrology. John Wiley and Sons,
Inc., New York, N.Y. 366 pp.
Dilaj, M., and J. F. Lenard. 1975. Leachate Control at Landfills
Based on Hydrogeological Studies. Public Works, April, 1975:
91—93, and 122.
Doqett, R. M., M. K. O’Farrell, and A. L. Watson. 1980. Fore-
casts of the Quantity and Composition of Solid Wastes.
EPA—600/5—80—001. U.S. Environmental Protection Agency,
Cincinnati, Ohio. 159 pp.
Doner, H. E. 1978. Chlorine as a Factor in Mobilities of Ni,
Cu, and Cd in Soil. Soil Sci. Soc. Am. J. 42:882—885.
Donigian, A. D., and N. H. Crawford. 1976. Modeling Pesticides
and Nutrients on Agricultural Lands. EPA-600/2-76-043.
Office of Research and Development, U.S. Environmental
Protection Agency, Athens, Georgia. 312 pp.
4
Donnelly, J. A. and P. V. Scarpino. 1981. Recovery of Fecal-
Indicator and Pathogenic Microbes from Landfill Leachate.
In: Land Disposal: Municipal Solid Waste, EPA600/981002a.
US. Environmental Protection Agency, Cincinnati, Ohio,
pp. 37—54.
Duguid, J. 0., and M. Reeves. 1976. Material Transport Through
Porous Media: A Finite—Element Galerkin Model. Oak Ridge
National Laboratory. ORL4L-4928. 198 pp.
Duguid, 3. 0., and M. Reeves. 1977. A Comparison of Mass Trans-
port Using Average and Transiet Rainfall Boundary Conditions.
In: Finite Elements in Water Resources. W. C. Gray, G. F.
Pinder, and C. A. Brebbia (eds.) Pentech Press, London.
pp. 225—235
350
-------
tt, G. R., M. J. Shaffer, and W. J. Moore. 1972. Computer
Model of Dynamic Biophysico-Chemical Processes in Soil.
Univ. of Arizona. Technical Bulletin 196. 101 pp.
Duvel, W. A., Jr., R. A. Atwood, W. R. Gallagher, R. G. Knight,
and R. J. McLaren. 1979. FGD Sludge Dis csal Manual.
Research Project 786—1, Electric Power Research Institute,
Palo Alto, California.
dens, M. R. 1978. Hydrogeologic Study of a Landfill Site Con-
sisting of Eighly Compressed Solid Waste. Master Thesis,
University of Minnesota.
zlzy, E., T. Lindstrom, and L. Eoersma. 1974. Analysis of the
Movement of Eaza.rdous Waste Chemicals In and From a Landfill
Site via a Simple Vertica1- orizontal Routing Model. Agr.
Exp. Sta. Special Report 414. Oregon State University,
Corvallis, Oregon.
ncon Associates. 1974. Sonoma Count Solid Waste Stabilization
Study. EPA-SW 530 65D. U.S. Environmental Protection Agency,
Washington, D.C. 228 p .
rich, C. H. and R. A. Landon. 1969. Generation of Leachate
from Landfills and Its Subsurface Movement. In: Proc. North-
eastern Regional Anti-Pollution Conference, University of
Rhode Island. pp. 57—63.
mrjch, C. H., W. W. Beck, Jr., and A. L. Tolinan. 1980. Top—
Sealing to minimize Leachate Generation. In: Proceedings
of the Sixth Annual Research Symposium on Disposal of
Hazardous Wastes. EPA—600/9—80—0l0. U.S. Environmental
Protection Agency, Cincinnati, Ohio. pp. 274—283.
4
nfjeld, C. C., C. C. Harlin, and B. E. Bledsoe, 1976. Com-
parison of Five Kinetic Models for Orthophosphate Reactions
in Mineral Soils. Soil Sci. Soc. Amer. J. 40:243—249.
Enfield, C. C., J. C. Hsieh, and A. W. Warrich. 1973. Evalua-
tion of Water Flux Above a Deep Water T&ble Using Thermo-
couple Psychrorneters. Soil Sci. Soc. of America Proceed-
ings. 37(6):968—970.
ngelbrecht , R. S. 1973. Survival of Viruses and Bacteria in a
Simulated Sanitary Landfill. NTIS/P3-234 589, National
Technical Information Service, Springfield, Virginia.
nge1brech, R. S. and P. Amirhor. 1975. Inactivation of Enteric
Bacteria and Viruses in Sanitary Landfill Leachate. NTIS!
PB—252 973/AS, National Technical Information Service,
Springfield, Virginia.
351
-------
Encelbrech, R. S., M. J. Weber, P. Anirhor, D. H. Foster and
D. LaRossa. 1974. Biological ?roperties of Sanitary
Landfill Leachates. In: Virus Survival in Water and
Wastewater Systems. J. F. Malina and B. P. Sagik, (eds.).
Water Res u.rces Symposium No. 7, Center for Research in
Water Resources, University of Texas, Austin, Texas. pp.
210—217.
Engers, L. V. 1978. Mineralization of Organic Matter in the
Subsoil of a Waste Disposal Site: A Laboratory Experiment.
Soil Science 126:22—28.
Environmental Protection Agency. Criteria for Classification
of Solid Waste Disposal Facilities, Federal Register,
Thursday, September 13, 1979.
Everett, L. C. 1981. Monitoring in the Vadose Zone. Ground-
Water Monitoring Review. Summer 1981. pp. 44-51.
Everett, L. C., K. D. Schmidt, R. M. Tinlin, and D. K. Todd.
1976. Monitoring Ground-Water Quality: Methods and Costs.
General Electric Co.-TEMPO Center for Advanced Studies,
Santa Barbara, California. 140 p.
Ewald, C. W. 1973. Stretching the Life of Synthetic Pond Liners.
Chemical Engineering. 80(22):67—69.
Farmer, W. J. 1978. Land Disposal of Hexachlorobenzene Wastes:
Controlling Vapor Movement in Soil. Proceedings of the
Fourth Annual Research Symposium. EPA-600/9—78—0l6. U.S.
Environmental Protection Agency, Cincinnati, Ohio. 424 p .
Farquhar, C. J. and F. A. Rovers. 1973. Gas Production During
Rufuse Decomposition. Water, Air, and Soil Pollution.
2(10):483—489.
Farquhar, C. J., J. F. Sykes, and S. Soyupak. 1978. Modeling
Contaminant Attenuation in Soil. Microbial Decomposition
of Organic Matter. In: Land Disposal of Solid Waste,
Proceedings of the Fourth Annual Research Symposiu n EPA .
600/9—78—016. U.S. Environmental Protection Agency,
Cincinnati, Ohio. pp. 243—254.
Fava, A., and H. Eyring. 1956. Equilibrium and Kinetics of
Detergent Adsorption: A Generalized Equilibration Theory.
J. Phys. Chem. 65: 890—898.
Fenn, D., E. Cocozza, J. Isbister, 0. Braids, B. Yare, and p.
Roux. 1977. Procedures Manual for Ground Water Monitoring
at Solid Waste Disposal Facilites. EPA/530/SW —611, u.s.
Environmental Protection Agency, Cincinnati, Ohio. 269 pp.
352
-------
0. G., K. J. anley, and T. V. Degeare. 1975. Use of
the Water Balance Method for Predicting Leachate Generation
from Solid Waste Disposal Sites. EPA—530/SW—168, U.S.
Environmental Protection Agency, Cincinnati, Ohio. 40 p .
Fisher, E. A. 1927. Some Factors Affecting the Evaporation of
Water from Soil, II. 17:407—419.
Flower, F. B., I. A. Leone, E. F. Gilman, and J. J. Arthur.
1977. Vegetation Kills in Landfill Environs. In: Manage-
ment of Gas and Leachate in Landfills. EPA-600/9—77-026,
U.S. Environmental Protection Agency, Cincinnati, Ohio.
pp. 218—236.
Fox, R. L., and E. J. Kamprath. 1970. Phosphate Sorption
Isotherms for Evaluating The Phosphate Requirements of
Soils. Soil Sci. Soc. of America. Proc. 34: 902—907.
Freeze, R. A., A. Stochastic. 1975. Conceptual Analysis of
One-Dimensional Ground-Water Flow in Nonuniform omoqeneous
Media. Water Resources Research 11(5):725-741.
Frere, M. H. 1980. The Nturient Submodel. In: CREAMS. A
Field Scale Model for Chemicals, Runoff, and Erosion from
Agricutural Management Systems. W. C. Knisel, (ed.). U.S.
Department of Agriculture, Conservation Research Report
No. 26. pp. 65—87.
Friest, J. .3., and P. 0. Ungemach. 1971. A Dispersion Model
for A Quantitative Study of A Ground-Water Pollution by
Salt Water Research. 5:491-495.
Fried, .3. .3. 1975. Ground-Water Pollution: Theory, Methodology,
Modeling, and Practical Rules. Developments in Water
Science. Elsevier, 330 pp.
Frost, R. R., and R. A. Griffin. 1977. Effect of pH on Ad-
sorption of Arsenic and Selenium from Landfill Leachate
by Clay Mineral. Soil Sci. Soc. Pm. .3. 41:53—57.
Fuller, W. . 1976. Attenuation Mechanisms of Pollutants
Through Soils. In: Gas and Leachate from Landfills:
Formation, Collection, and Treatment. EPA-600/9—76—004.
U.S. Envirnomental Protection Agency, Cincinnati, Ohio.
pp. 111—122.
‘ul1er, W. if. 1977. Movement of Selected Metals, Asbestos,
and Cyanide in Soil. Applications to Waste Disposal
Problems. EPA—600/277-020, U.S. Environmental Protection
Agency, Cincinnati, Ohio. 257 pp.
353
-------
Fuller, W. H. 1978. Investigation of Landfill Leachate Pollut-
ant Attenuation By Soils. EPA-600/2-78-158 , U.S. Environ—
n’.ental Protection Agency, Cincinnati, Ohio. 212 pp.
Fuller, W. H., A. A. Fa.rd, E. E. Niebis, and M. Boyle. 1980.
Influence of Leachata Quality on Soil Attenuation of Metals.
In: Disposal of Hazardous Wastes. Proceedings of the Sixth
Annual Research SympositmL. EPA—600/9—80—010, U.S. Environ—
mental Protection Agency, Cincinnati, Ohio. pp. 108-117.
Fuller, W. H., and J. A.rtiola. 1978. Use of Limestone to Limit
Contaminant Movement from Landfills. In: Land Disposal of
Hazardous Wastes. Proceedings of the Fourth Annual Research
Sym osiuin. EPA—600/19-78 0 l 6 , U.S. Environmental Protect-
ion Agency, Cincinnati, Ohio. pp. 282-298.
Fungaroli, A. A. 1971. Pollution of Subsurface Water by
Sanitary Landfills. SW-l2rg, U.S. Environmental Protection
Agency, Washington, D. C.
Fungaroli, A. A., and R. L. Stiner. 1973. Investigation of
Sanitary Landfill Behavior. EPA—R2/2—7, Office of Research
and Development, U.S. Environmental Protection Agency,
Cincinnati, Ohio. 313 pp.
Fungaroli, A. A., and R. L. Steiner. 1979. Investigation of
Sanitary Landfill Behavior. Vol. I. Final Report. EPA—
600/2—79—053a, U.S. Environmental Protection Agency,
Cincinnati, Ohio. 331 pp.
Gaby, W. L. 1975. Evaluation of the Health Hazards Associated
with Solid Waste/Sewage Sludge Mixtures. EPA-670/2—75—023.
U.S. Environmental Protection Agency, Cincinnati, Ohio.
44 pp.
Gadde, R. R., and H. A. Laitinen. 1974. Studies of Heavy Metal
Adsorption by Hydrous Iron and Manganese Oxides Anal. Chexu,
46: 2022—2026.
Gairon, S., and 3. Hadas. 1973. Measurement of the Water Status
in Soils. In: Arid Zone Irrigation, Y. B. Dantois and y.
Vaadin (eds.). springer-Verlag, New York, N.Y. pp. 215—266.
Garcia-Miragaya, J. and A. L. Page. 1977. lnfluence of Exchange...
able Cation on the Sorption of Trace Amount of Cadxnii. n by
MontmorillOflite. Soil Sci. Soc. Am. 3. 41:718—721.
Garrels, R. M., and C. L. Christ. 1965. Equilibria. Harper an
Row, New York, N.Y. 415 pp.
354
-------
Gaudet, J. P., t . Jegat, G. Vachaud, and P. r, erenga. 1977.
Solute Transfer with Exchange Between Mobile and Stagnant
Water Through Unsaturated Sand. Soil. Sci. Soc. ? mer. J.
41: 665—671.
Geiger, R. 1963. The Climate ‘Iear the Ground, revised ed.,
Harvard University Press, Cambridge, Mass.
Geihar, L. W. and J. L. Wilson. 1974. Ground Water Quality
Modeling. Ground water. 12:399:408.
Gerwick, B. C. 1967. Slurry—Trench Techniques for Diaphragm
Walls in Deep Foundation Construction. J. Civil Eng. Div.,
ASCE. 37(1):70—72.
Geswein, A. L. 1975. Liners for Land Disposal Sites: An
Assessment. EPA/5301SW137, U. S. Environmental Protection
Agency, Cincinnati, Ohio. 66 pp.
GeSweifl, A. J., R. E. Landreth, and H. Haxo. 1978. Use of
Liner Materials for Land Dispoal Facilities. U.S. Environ-
mental Protection Agency, Office of Solid Waste (WW—562).
18 pp.
jddings, T. 1977. The Lycoming County, Pennsylvania Sanitary
Landfill: State-of-the—a.rt in Ground-Water Protection.
Groundwater. 15(13) :5—13.
Gilber, R. C. 1976. Virus and Bacteria Removal from Wastewater
by Land Treatment. Applied and Environmental Microbiol.
32: 333—338.
GlotzbeCker, R. A. and A. L. Novello. 1975. Poliovirus and
Bacterial Indicators of Fecal Pollution in Landfill Leach—
ates. News of Environmental Research in Cincinnati, U.S.
Environmental Protection Agency, Cincinnati, Ohio.
Goerlitz, D. F., and E. Brown. 1972. Methods of Analyses of
Organic Substances in Water. Chapter A-3, Techniques of
Water-Resources Investigations, Book .5, Laboratory Analyses,
U.S. Geological Survey. 40 pp.
Goodrich, B. F. Company. 1978. Typical Installation Details
for B. F. Goodrich Flexseal -Pond and Pit Liners. Bulletin
EL—l.5—775, Marietta, Ohio. 5 pp.
Gray, D. M., and D. I. Norum. 1968. The Effect of Soil Moisture
on Infiltration as Related to Runoff and Recharge. n:
Soil Moisture, Proceedings of Hydrology Syrnoosium, No. 6,
National Research Council of Canada. pp. 133—150.
355 -
-------
Green, D. W., and R. L. Ccx. 1966. Storage of Fresh Water in
underground Reservoirs Containing Saline Water, Phase I,
Completion Report 3. Kansas Water Resources Research
Institute, Manhattan, Kansas. 24 pp.
Green, K. M., and D. 0. Oliver. 1975. Removal of Virus from
Septic Tank Effluent by Sand Columns. Proceedings of a
National Home Sewage Disposal Symposium. American Soc.
of Agricultural Engineering, St. Joseph, Michigan.
Greenland, D. J. 1971. InteractiOnS Between Humic and Fulvic
Acids and Clays. Soil Science. 111:34—41.
Griffin, R. A., K. Cartwright, and N. L. Shimp. 1976. Attenua——
tion of Pollutants in Municipal Landfill Leachate by Clay
Minerals. Environmental Geology Notes, No. 78. Illinois
State Geolgoical Survey, Urbana, Illinois. 34 pp.
Griffin, R. A., and N. F. Shiinp. 1978. Attenuation of Po11u .
tants iii Municipal Landfill Leachate by Clay Minerals.
EPA—600/2-78-15 7 , U.S. Environmental Protection Agency,
Cix cinnati, Ohio. 157 pp.
Griffin, R. A. 1978. Disposal and Removal of Polychiorinated
Biphenyls in Soils. Proceedings of the Fourth Annual
Research Symposium. EP? -600/9-78-O16, U.S. Environmental
Protection Agency, Cincinnati, Ohio. 424 p .
Griffin, R. A., and S. F. J. Chou. 1980. Disposal and Removal
of Halogenated Hydrocarbons in Soils. Proceedings of
the Sixth Annual Research Symposium. EPA—600/9—80-010,
U.S. Environmental Protection Agency, Cincinnati, Ohio.
290 pp.
Grove, D. B., 1970. A Method to Describe the Flow of Radioactive
Ions in Ground Water-Final Report, Dec. 1, 1966 through
June 30, 1968. Springfield, Virginia, Nat’l. Tech. Inf.
Serv. 41 pp.
Gupta, S. K., K. K. Tanji, and L. Luthin. 1975. A Three—
Dimensional Finite Element Ground-Water Model. University
of California, Water Resources Center, Contribution Series
No. 152. ii pp.
Gureghian, A. B., and R. W. Cleary. 1977. Three—Dimensional
Modeling of Pollutant Transport. 208 Project, Final Report
Nassau Suffolk-County Regional Planning Board, Hauppauge, ‘
New York.
356
-------
Gureghian, A. B., R. W. Cleary, and S. Ward. 1977. One—
Dimensional Modeling of unsaturated Pollutant Transport-
208 Project, Final Retort, Nassau-Suffolk County Regional
Planning Baord, Eauppauge, Mew York.
Halliburton Services. 1976. Grouting in Soils. Vol. 2, Design
and Operations Manual. NTIS/PB—259044, National Technical
Information Service, Springfield, Virginia. 29 pp.
Ham, R. K. 1975. The Generation, Movement, and Attenuation of
Leachates from Solid Waste Land Disposal Sites. Waste Age.
6: 50—112.
Ham, R. IC, and C. R. Anderson. 1975. Pollutant Production y
Refuse Degradation in Test Lysimeters. Waste Age. pp. 31-
36.
Ham, a. K., M. A. Anderson, R. Stanforth, and R. Steçmann. 1978.
The Development of a Leaching Test for Industrial Wastes.
In: Land Disposal of Hazardous Wastes. EPA —600/9—78-01G,
U.S. Environmental Protection Agency, Cincinnati, Ohio.
pp. 33—46.
Ham, R. K., M. A. Anderson, R. Stegmann, and R. Stanforth. 1979.
Background Study on the Development of a Standard Leaching
Test. EPA-600/2—79-l09, C..S. Environmental Protection
Agency, Cincinnati, Ohio.
HaztkS, R. J., and C. L. Ashford. 1980. Applied Soil Physics.
Springer-Verlag, New York, N.Y. 293 pp.
Hassan, A. A. 1974. Mathematical Modeling of Water Quality for
Water Resources Management, Vol. 1, Development of the
Ground-Water Quality Model. District Report. Dept. of Water
Resources, Southern District, Sate of California.
Hansen, E. A., arid A. R. Harris. 1975. Validity of Soil-Water
Samples Collected with Porous Ceramic Cups. Soil Sci.
Soc. Amer. Proc. 39:528—536.
Haxo, H. E., Jr. 1976a. Assessing Synthetic and Admixed Materi-
als for Lining Landfills; Formation, Collection, and Treat-
ment. In: Gas and Leachate from Landfills. EPA—600/9-76—
004, U.S. Environmental Protection Agency, Cincinnati, Ohio.
pp. 130—158.
Haxo, H. E., Jr. 197Gb. Evaluation of Selected Liners When
Exposed to Hazardous Wastes. Proceedings of Hazardous Waste
Research Symposium, Tucson, Arizona. EPA-600/9—76-015, U.S.
Environmental Protection Agency, Cincinnati, Ohio. 269 pp.
357 -
-------
Haxo, H. E., Jr. 1976c. Evaluation of Liner Materials Ex osed
to Leachate: Second Interi Report. EPA-600/2-76—255,
U.S. Environmental Protection Agency, Cincinnati, Ohio.
54 PP.
Haxo, H. E., Jr. 1977a. Compatability of Liners with Leachate.
n: Management of Gas and Leachate from Landfills. Pro—
ceed.ings of the Third Annual Municipal Solis Waste Research
Symposium. EPA—600/277081, U.S. Environmental Protection
Agency, Cincinnati, Ohio. pp. 149-159.
Haxo, H. E., Jr. 1977b. What’s New in Landfill Liners. The
American City and County Magazine. pp. 54-56.
Haxo, H. E., Jr. 1978. InteractiOn of Selected Lining Materials
with Various Hazardous Wastes. In: Land Disposal of
Hazardous Wastes. Proceedings of 4th Annual Research
Symposium. EPA—600/9-78—016, U.S. Environmental Protection
Agency, Cincinnati, Ohio. pp. 256—272.
Ha.xo, H. Z., Jr. 1980. Interaction of Selected Lining Materials
with Various Hazardous Wastes. In: Disposal of Hazardous
Waste, Proceedings of the Sixth Annual Research Syrnposju .
ERA—600/9—80—0l0, U.S. Environmental Protection Agency,
Cincinnati, Ohio. pp. 160-180.
Eaxo, H. E., Jr., R. S. Haxo, and I. F. I(ellog. 1979. Liner
Materials Exposed to Municipal Solid Waste Leachate. EPA—
600/2—79-038. U. S. Environmental Protection Agency,
Cin.c .nnati, Ohio. 57 po.
Hays, B. D. 1977. Potential for Parasitic Disease Transmission
with Land Application of Sewage Plant Effluents and Sludges.
Water Research. 11(7):583595.
Helweg, 0. J., and J. W. Labodie. 1976. A Salinity Management
Strategy for Stream—Aquifer Systems. Hydrology Paper No.
84, Colorado State University.
Hentrich, R. L. , J. F. Swartzbaugh, and J. A. Thomas. 1979.
Influence of MSW Processing on Gas and Leachate Production.
In: Municipal Solid Waste; Land Disposal. Proceedings of
the Fifth Annual Research Symposium, Cincinnati, Ohio.
pp 98—137.
Hendricks, D. W. 1972. Sorption in Flow Through Porous Media.
In: Fundamentals of Transport Phenomena in Porous Media.
Developments in Soil Science, Elsevier Pub. Co., Amsterday.
pp. 384-392.
35
-------
Hjldebrand, M. A. 1975. The Transport of Nitrate During Un-
steady Water Flow Through Unsaturated Sand Media. Ph.D.
Dissertaton. University of Te cas, Austin, Texas.
Hjld.ebrand, M. A., and D. M. Ei ne1bau. 1977. Transport of
Nitrate Ion in Unsteady Unsaturated Flow in Porous Media.
AICLE Jounral. 23(3):326—335.
, s., w. C. Boyle, and R. K. Ham. 1974. Chemical Treatment
of Leachates from Sanitary Landfills. Journal of the
Water Pilution Control Federation, 46(7):1776—1791.
Holmes, 3. N., S. A. Taylor, and S. 3. Richards. 1967. Measure-
ment of Soil Water. In: Irrigation of Agricultural Lands,
Agronomy No. 9. Americal Society of Agronomy, Madison
Wisconsin. pp 275—298.
Eornsby, A. G., and J. M. Davidson. 1973. Solution and Adsorbed
Fluometuron Concentration Distribution in a Water-Saturated
Soil. Soil Sci. Soc. Amer. Proc. 37:823—828.
Hornsby , A. C. 1973. Prediction Modeling for Salinity Control
in Irrigation Return Flows. EAP—R2-73—168. U.S. Environ-
mental Protection Agency, Cincinnati, Ohio. 153 pp.
Hughes, C. M., R. A. Landon, and R. N. Farvolden. 1971. Hydro-
geology of Solid Waste Disposal Sites in Northeastern
Illinois. SW—12d, U. S. Environmental Protection Agency,
Washington, D.C. 115 pp.
Ishiwateri, a. 1971. Molecular Weight Distribution of Humic
Acids from Lake and Marine Sediments. Geochemical Journal.
5:121.
Jen, S. W. 1949. Infiltration. In: Hydrology Handbook, Manual
of Engineering Practice No. 28, American Society of Civil
Engineers, N.Y. pp. 33—63.
Jenne, E. A. 1968. Controls on Mn, Fe, Co, Ni, Cu, and Zn.
Concentrations in Soils and Water. The Significant Role of
Hydrous Mn and Fe Oxides. Trace Inorganics in Water, Ad-
vances in Chemistry Series NO. 73. American Chemical
Society, Washington D. C. pp. 337—388.
Johansen, 0. J., and D. A. Carlson. 1976. Characterization of
Sanitary Landfill Leachates. Water Research. 10:1129-1134.
Johnson Division, UOP. 1971. Ground Water and Wells: A Refer-
ence Book for the ater—tie11 Industry. Edward E. Johnson,
Inc., St. Paul, Minnesota. 440 pp.
359
-------
Johnson, T. M., and K, Cartwright. 1978. Implications of
Solid-Waste Disposal in the Unsaturated Zcne. In: Pro-
ceedings of the First Annual Conference of AO ljed Research
and Practice on Municipal and Industrial Waste, Madison,
WisconSin. 185 pp.
Jorgenson, D. C. 1975. Analog-Model Studies of Ground-Water
Hydrology in the Houston District. Texas Water Develop—
inent Board Report 190. 84 pp.
Kapp, M. S. 1969. Slurry-Trench Construction for Basement
Wall of World Trade Center. J. Civil Eng. Div, ASCE.
39(1) :36—40.
Karol, R. H. 1968. Chemical Grouting Technology. 3. Soil
Mechanics Foundation Div. ASCE, Vol. 94, No. SMI. Proc.
Paper 5748. pp. 175—204.
Kays, W. B. 1977. ConstructiOn of Linings for Reservoirs,
Tanks, and Pollution Control Facilities. John Wiley and
Sons, Inc., New york, N.Y. 310 pp.
Keen, B. H., E. M. Crouther, and J. R. H. Coutts. 1926. The
Evaporation of Water from Soil, III. A Critical Study of
the Techniques. 3. Agric. Sci. 16:105—122.
Kemper, 3. M. and R. B. Smith. 1981. Leachate Production by
Landfilled processed Municipal Wastes. In: Land Disposal
Municipal Solid Waste. Proceedings of the Seventh Annual
Research Sy nposium. EPA-600/981—002a, U.S. Environmental
Protection Agency, Cincinnati, Ohio. pp. 18-36.
Keswich, B. H., and C. P. Gerba. 1980. Viruses in Ground-Water
Environ. Sci. & Tech. 14:1290—1297.
Keys, W. S., and L. M. MacCary. 1971. Application of Borehole
Geophysics to Water Resources Investigations. In: Tech-
nical Techniques of Water—Resources. Investigations of
the u.s. Geologic Survey, Book 2. 126 pp.
Khare, M. , and N. C. Dondero. 1977. Fractionation and Con-
centration of Volatiles and Organics on High Vacuum Syst .
Examination of Sanitary Landfill Leachate. Evironmental
Science and Technology. ll(8):814—819.
King, L. C., and R. 3. Eanks. 1973. Irrigation Management
for the Control of Quality of Irrigation Return Flow.
EPA—R2—73-265, U.S. Environmental Protection Agency,
Washington, DC. 307 pp.
360
-------
King, L. G., and R. J. Hanks. 1975. Management Practices faot-
ing Quality and Quantity of Irrigation Return FLOW. EPA—
660/2—75—005, t3.S. Environmental Protection Agency, Corvallis,
Oregon. 267 pp.
jnni urgh, D. G., M. L. Jackson, and J. K. Syers. 1976. Ad-
sorption of Alkaline Earth, Transition, and Heavy Metal
Cations by Hydrous Oxide Gels of Iron and Alumin n.
Soil Sci. Soc. zn. J. 40:796—799.
Kirda, C., D. R. Nielsen, and J. W. 3iggar. 1973. Simultaneous
Transport of Chloride and Water During Infiltration.
Soil Sci. Soc. ? ter. Proc. 37(3):339—345.
njsel, 4. C. Editor. 1980. CREAMS: A Field—Scale Model for
Chemicals, Runoff, and Erosion from Agricultural Management
Systems. U.S. Department of Agriculture, Conservation
Research Report No. 26. 640 pp.
Kor.ikow, L. F., and J. D. Bredehoeft. 1974. Modeling Flow and
Chemical Quality Changes in an Irrigated Stre t-Aquifer
System. Water Resrouces Rsch. 10:546-562.
KOfljkCW, J. F. 1977. Modeling Chloride Movement in the Alluvial
Aquifer at Rocky Mountain Arsenal, Colorado. Geological
Survey Water Supply Paper 2044. U.S. Government Printing
Cf fice, Washington D.C.
ortonGva, M. N. 1961. Soil Organic Matter. Pergamon Press,
New York. pp.
rte, N. E., J. Skopp, W. H. Fuller, E. E. Niebla, and B. A.
Alesh. 1975. Trace Element Movement in soils: Influence
of Soil Physical and Chemical Properties. Soil Science.
122: 350—359.
N. E. 1976. A Baseline Study on Trace Metal Elution from
Divers Soil Types. Water, Air, Soil Pollution. 5:149—156.
K., and P. H. Jones. 1979. Complexaton Characteristics of
Sanitary Landfill Leachates. Water Research. l3(ll):839-
846.
nan, 3., and 3. A. Jedlicka. 1973. Selecting and Installing
Synthetic Pond Linings. Chemical Eng. 80(5):67—70.
E. Y. F. 1976. Analytical Solutions for 3-D Diffusion
Model. J. Environ. Eng. Div. ASCE. 102:805—820.
çq Brown and Associates, Inc. 1980. Hazardous Waste Land Treat-
ment. EPA Sw—874, U.S. Environmental Protection Agency,
Cincinnati, Ohio. 913 op.
361
-------
Lai, Sung-Ho, and J. J. Jurinsk. 1971. Numerical Approximation
of Cation Exchange in Miscible Displacement Throught Soil
Columns. Soil Sci. Scc. Amer. Proc. 35:894—899.
Lapidus, L., and N. R. xnundson. 1952. Mathematics of Adsorption
in Beds. J. Phys. Chem. 56:984—988.
Larson, N. M., and M. Reeves. 1976. Analytical Analysis of Soil-.
Moisture and Trace-Contaminant Transport. Oak Ridge National
Laboratory. ORNL/NSF/EATC-12. 180 pp.
Latimer, W. M. 1952. Oxidation Potential. Prentice—Hall, Inc.
N.J. 392 pp.
Lawrence, A. W. and P. L. McCarty. 1969. inetics of Methane
Fe entatiOfl in Anaerobic Treatment. Journal of the Water
Pollution Control Federation, Vol. 41, No. 2, Part 2.
Leckie, J. A., and R. 0. James. 1976. Control Mechanizns for
Trace Metals in Nautral Waters. Aqueous—Environmental
Chemistry for Metals. A. J. Rubin (ed.) Ann Arbor Science
Pubi. Inc., Ann Arbor, Michigan. pp. 1-76.
Leckie. J. 0., J. P. Pacey, and C. Halvadakis. 1975. Accelerated
Refuse Stabilization Through Controlled Moisutre A p1ictjon.
In: Second Annula Natonal Conference on Environmental
Engineering Research, Development, and Design, Gajnsvjlie,
Florida.
Leckie, J. 0., J. G. Pacay, and C. Halvadakis. 1979. Landfill
Management with Moisture Control. J. Environ. Eng. Djvj n,
ASCE, EE2. pp. 337-355.
Lee, C. H., and R. T. Cheng. 1974. On Seawater Encroachment
in Coastal Aquifers. Water Resources Research. 10(5):
1039—1043.
Lee, J. 1974. Selecting Membrane Pond Liners. Pollution Eng.
6(1) :33—40.
Leenheer, J. A., and J. L. Ahlrichs. 1971. A inetic and
Eqi ilibrium Study of the Adsorption of Carbaryl and
Parathion Upon Soil Organic Matter Surfaces. Soil Sci.
Soc. amer. Proc. 35:700—704.
Lenahan, T. 1973. There is a Place for Grouting in Underground
Storage Coverns. Bulletin Assoc. Engineering Geologist.
10(2) :137—144.
362
-------
Leonard, R. A., G. W. Baily, and R. R. Swank. 1976. T:ansocrt,
Deto: ificatiOn, Fate, and Effects of Pesticides in Soil
and Water. Land Applicaton of Soils Material. Soil Con-
servation Society of America. 253 .
Leor.ard, R. A. and a. D. Wauchopi. 1980. The Pesticide Model.
In: CREAMS A Field Scale Model for Chemistry, Runoff, and
Erosion from Agricultural Management Systems. W. G. Knisel,
(ed.). U.S. Dept. of Agricutlure, Conservation Research
Report No. 26. pp. 88-112.
LeSSj, L. 1976. Simulation Nurnerique de la ?rc agation d’ ur.e
Pollutant Dams Un Milliue Poreux Sature. These de Docteur-
Ingenisur, Universite de Mecanique des Fluides de Strasbourg,
Universite Louis Pasteur, Strasbcurg, France. 157 op.
Lin, Y. H. 1966. Acid and Gas Production from Sanitary Land-
fills. Ph.D. Thesis, West Virginia University, Morgantown,
West Virginia. 95 pp.
Lindetrom, F. T., R. Hague, V. H. Freed, and L. Boersina. 1967.
Theory on the Movement of Scme Herbicides in S iis. Env.
Sci. & Tech. 1:561—565.
tindstrom, F. T., and L. Boersma. 1970. Theory of Chemical
Transport with S u1taneous Sorption in a Water-Saturated
Porous Medium. Soil Science. 110:1—9.
jndstrom, F. T., L. Boersma, and 0. Stockard. 1971. A Theory
on the Mass Transport of Previously Distributed Chemicals in
a Water-Saturated Sorbing Porous Medium: Isothermal Cases.
Soil Science. 112:291—300.
Lindstrorn, F. T., and L. Boersma. 1971. A Theory on the Mass
Transport of Previously Distributed Chemicals in a Water
Saturated Sorbing Porous Medium. Soil Science. 111(3):
192—199.
Lindstrom, F. T., and L. Boersina. 1973. A Theory on the Mass
Transport of Previously Distributed Chemicals in a Water
Saturated Sorbing Porous Medium. Exact Solution for First-
Order Kinetic Sorption. Soil Science. 1l5(1):5—10.
Limdstrcm, F. T., and W. M. Stone. 1974a. On the Start Up or
Initial Phase of Linear Mass Transport of Chemicals in
a Water Saturated Sorbing Porous Medium. SLAM Applied Math.
26(3) :578—591
3.3
-------
Lindstrcm, F. T., and W. M. Stone. 1974b. On the Start Up or
Initial Phase of Linear Mass Transport of Che .icals in a
r ,.later Saturated Sorbing Porous Mediun. Integral Eouation
Approach. SIAM J. Appi. Math. 26(3):592-606.
Linsley, R. D., and J. B. Franzini. 1972. Water Resources
Engineering. McGraw—Hill Book Co., Inc. New York, N.Y.
682 Pp.
Liskowitz, J. w., P. C. Chan, and R. B. Trattner. 1976. Eval-
uation of Selected Sorbents for the Removal of Contaminants
in Leachate From Industrial Sludges. In: Residual Manage-
ment by Land Disposal. Proceedings of the Hazardous Waste
Research Symposium. EPA 600/9—76-015. U.S. Environmental
Protection Agency, Cincinnati, Ohio. p . 162-176.
Lu, J. C. S. 1976. Studies Ofl the Long-Tertt Migration and
Transformation of Trace Metals in the Po1lut d Manned
Sediment-Seawater System. Ph.D. Thesis, University of
Southern California.
Lu, J. S. C. 1981. Chemical Speciation of Flue Gas Desulfuri—
zation (FGD) Sludge Constituents. EPA—600/S2-81—062
U.S. Environmental Protection Agency, Municipal Environ-
mental Research Laboratory. 315 pp.
Lu, J. C. S., R. J. Stearns, 3. Eichenberger, and. R. D. Morrison.
1980. A Critical Review of Wastewater Treatment Plant S1udg
Disposal by Landfilliflg. Contract No. 68-03—2886,
U.S. Environmental ProteCt .Ofl Agency, Cincinnati, Ohio.
273 pp.
Lutton, R. J. 1980. Evaluation Cover Systems for Solid and
Hazardous Waste. EPA SW867. U.S. Environmental Protection
Agency, Cincinnati, Ohio. 57 pp.
Lutton, R. J., C. L. Regan, and L. W. Jones. 1977. Design and.
Construction of Covers for Solid Waste Landfills. EPA—
600/2—79—165. U.S. Environmental Protection Agency,
Cincinnati, Ohio. 249 pp.
Malcolm. a. L. 1970. OrganiC Matter in Natural Waters. 0. W.
Hood. (ed.). University of Alaska, Institute of Marine
Science, Pub. No. 1. p. 479.
Mang, J. L., J. C. S. Lu, R. L. Lofy, and R. P. Stearns. 1979.
A Study of Leachate from Dredged Material in Upland Areas
and/or in Productive Uses. Waterways Experiment Station,
u.s. Corps of Engineers, Vicksburg’, Mississippi. 428 pp.
364
-------
Mang, J., and D. t’leaver. 1976. Methods for Control of Ground-
water Contamination from Sanitary Landfills. In: Third
National Ground Water Quality Symposium, Las Vegas, Nevada.
224 pp.
MariflO, M. A., 1974. Distrthution of Contaminants in Porous
Media Flow. Water Resources Research. 10(5):1013—1018.
Martinez, 7. D., D. H. Kupper, and R. L. Thomas. 1975. A.n
Investigation of the Utility of Gulf Coast Salt Domes for
the Storage or Disposal of Radioactive Wastes. Institute
for Environmental Studies, LSTJ, Baon Rouge, Louisiana.
Matrecon, Thc. 1980. Lining of Waste Irn cundment and Dis osal
Facilities. EPA SW—870, U.S. Environmental Protection
Agency, Cincinnati, Ohio.. 385 p .
ateigod, S. V., and C. Sposito. 1977. Estimated Association
Constants for Some Complexes of Trace Metals with Inorganic
Ligands. Soil Sci. Soc. Amer. J. 41:1092—1097.
Mc3ride, M. 3. 1978. Retention of Cu, Ca, Mg, and Mn by
Amorphous Alumina. Soil Sci. Soc. Amer. J. 42:27—31.
McCarty, P. L., and 3. E. Rittznan. 1981. Trace Organics in
Ground Water. Environ. Sci. & Tech. 15:40—49.
McGowan, M., and J. B. Williams. 1980. The Water Balance of
an Agricultural Catchment: Estimation of Evaporation from
Soil Water Records. J. Soil Science. 31:217—230.
McMillon, L. C., and 3. W. Keeley. 1968. Smapling Equipment for
Ground-Water Investigations. Ground Water. 6(2):9—11.
MeichtrY, T. M. 1971. Leachate Control System. Presented at
Los Angeles Regional Forum on Solid r aste Management. 20
pp.
Mercado, A. 1976. Nitrate and Chloride Pollution of Aquifers:
A Regional Study with the Aid of a Single-Cell Model.
Water Resources Research. 12:731-747.
Merrill, S. D., and S. L. Rawlens. 1972. Field Measurement of
Soil Water Potential with Thermocouple Psycbrometers.
Soil Science. 1l3(2):102—109.
Merz, R. C. 1954. Final Report on the Investigation of Leaching
of a Sanitary Landfill, Publication No. 10. California
State Water Pollution Control Baord, Sacramento, California.
90 pp.
365
-------
Merz, R. C., and R. Stone. 1962. Landfill Settlement Rates.
Public Works. 93(a):l03—210.
Merz, R. C. and R. Stone. 1968. Special Studies of a Sanitary
Landfill. EPASWSR6—70, U.S. Public Health Service, Bur-
eau of Solid Waste Management.
Middlebrooks, E. J., C. D. Perinan, and I. S. Dunn. 1978. Waste—
water Stabilization Pond Linings. EPA-MCD- 54. cr.s. Environ-
mental Protection Agency, Washington, D.C. 115 pp.
Ministry of Housing and Local Government. 1961. Pollution of
Water by Tipped Refuse. Report Tech. Committee on Experi-
mental Disposal of House Refuse in Wet and Dry Pits. He:
Majesty’s Stationary Office, London, England. 141 p.
Mitchell, R. A. 1974. Introduction to Environmental Micro-
biology. PrenticeHall, Inc., Englewood Cliffs, New
Jersey. 324 pp.
Mo1 , F. L. 1974. TranspiratiOn Drying of Sanitary Landfills.
Ground Water Vol. 12. pp. 394—398.
Molz, F. L., S. R. Van Fleet, and V. D. Browning. 1974. Trans-
piration Drying of Sanitary Landfills. Ground Water.
12 C 6) : 394—397.
Moore, 3. E., B. P. Sagik, and J. F. Maljna, Jr. 1975. Viral
Association with Suspended Solids. Water Research.
9(2):197-203.
Morrison, R. D., and P. E. Brewer. 1981. Air—Lift Samplers
for Zone—of-Saturation Monitoring. Ground Water Monitoring
Review, Spring 1981. pp. 52—55.
Murrmann, R. P., and F. R. Koutz. 1972. Role of Soil Chemical
Processes in Reclamation of Wastewater Applied to the
Land. Wastewater Management by Disposal on the Land.
U.S. Army Corps of Engineers, Cold Regions Research and
Eng. Lab., Speciality Report 171.
Myers, T. E., J. C. Duke, Jr., P. G. Malone, and D. W. Thompson.
1979. Gas Production in Sanitary Landfill Simulators.
In: Municipal Solid Waste: Land Disposal. EPA—600—9—79—
023a, U.S. Environmental Protection Agency, Cincinnati, Ohio.
pp. 58—73.
Nielsen, D. R. , 3. W. Bigger, and K. T. Erb. 1973. Spatial
variability of Field-Measured So il-Wate Properties.
Hilgaria. 42(7) :205—269.
366
-------
Nisbet, I.C. T., and A. F. Sa.rofin. 1972. Rates and Routes
of Transport of ?C3s in the inviror .enz Environ. ealth
Perspectives. 1:21—38.
Nobel, D. G. 1963. Well Points for Dewatering Ground Water.
Ground Water. l(3):21-36.
Norstedt, R. A., L. B. Baldwin, and L. M. Rhodes. 1974. Land
Disposal of Effluent from a Sanitary Landfill. Journal of
the Water Pollution Control Federation, 47(7):1961-1970.
Novello, A. L. 1974. Poliovirus Survival in Landfill Leachate
and Migration through Soil Solumns. M. S. Thesis, Univer-
sity of Cincinnati, Ohio.
aksford, E. T. 1978. Water- omcrae r Ter.siometers Thstalled
and Read from the Land Surface. Geotechnical Testing
Journal. l(4):199—202.
Qddson, J. K., J. Letey, and L. V. Weeks. 1970. Predicted
Distribution of Organic Chemicals in Solution and Adsorbed
as a Function of Position and Time for Various Chemical
and Soil Properties. Soil Sci. Soc. Amer. Proc. 34:412—417.
ogata, A. 1961. Trnasverse Diffusion in Saturated Isotropic
Granular Media. Geological Survey Professional Pacer 411—B,
U.S. Government Printing Office, Washington, D. C. 8 pp.
ogata , A. 1970. Theory of Dispersion in a Granular Medi im.
U.S. Geological Survey Professional Paper 411—t. 42 pp.
Orlob, G. T., and P. C. Woods. 1967. Water-Quality Management
in Irrigation Systems. 1. Irr. Drain Div. ASCE 93:49—66.
palit, T., and S. R. Qasim. 1977. Biological Treatment Kinetics
of Landfill Leachate. Journal of the Env. Eng., Div. ASCE,
103 (EE2) :353—366.
parizek, R. R., and B. E. Lane. 1970. Soil Water Sampling Using
Pan and Deep Pressure Vacuum Lysiin+Ters. Journal of
Hydrology, Vol. II. pp. 1—21.
parkhurst, J. D. 1978. Landfill Disposal Demonstration Project,
Draft Report. Los Angeles County Sanitation Districts,
Los Angeles, California. 102 pp.
patrick, W. H., and R. A. Khalid. 1974. Phosphate Release and
Sorption by Soils and Sediments: Effect of Aerobic and
Anaerobic Conditions. Science. 186:53-53.
367
-------
Penman, i. L. 1948. Natural Evaporation From Open Water, 3are
Soil and Grass. Proceedings of the Royal Society, Series
A. 193:120—145.
Perez, A. 1., W. C. Hu.ber, J. P. Heaney, and E. E. Pyath. 1974.
A Water Quality Model for a Conjuctive Surface-Ground Water
System. EPA—600/5—74—0l 3 , U.S. Environmental Protection
Agency, Washington, D.C. 323 pp.
Perrier, E. H., and A. C. Gibson. 1980. Hydrologic Simulation
on Solid Waste Disposal Sites. EPA/SW—863, u.s. Enviro men —
tal Protection Agency, Cincinnati, Ohio. 11]. pp.
Perry, H. H. 1976. Engineering Manual, 3rd Ed. McGraw—Hill Book
Co., New York, N.Y. 946 pp.
Peters, D. G., J. N. Hayes, and G. M. Hieftje. 1974. Chemical
Separations and Measurements. W. B. Saunders Cc., Phil.
Pennsylvania. 749 pp.
Peterson, N. L. 1971. The Occurrence and Survival of Viruses
in Municipal Solid Waste. Ph.D. Thesis, UnIversity of
Michigan, Ann Arbor, Michigan.
Phelps, D. H. n.d. Solid Waste Leaching Model. Department of
Civil Engineering, University of British Columbia. 48 pp.
Phene, C. J., .G. J. Hoffman, and S. J. Rawlins. 1971. Measuring
Soil Matric Potential in Situ by Sensible Heat Dissipation
Within a Porous Body. Soil Sci. Soc. Amer. Proc. 35:27—33.
Philip, J. H. 1969. Theory of Infiltration. In: Advances in
HydroscieflCe Vol 1 5, V. T. Chow, (ed.). Academic Press,
New York. pp. 215—296.
Pickens, J. F., and W. C. Lennox. 1976. Numerical Simulation
of Waste Movement in Steady Ground Water Flow Systems.
Water Resources Research. 12(2):171-’180.
Pjnder, C. F. 1973. A GalerkinFinite Element Simulation of:
Ground Water Contamination of Long Island, New York.
Water Resources Research. 9(6):1657—1669.
Pjnder, G. F., and I. H. Cooper. 1970. A Numerical Technique
for Calculating the Transient Position of the Saltwater
Front. Water Resources Research. 6(3):875-882.
Pirider, C. F., and R. H. Page. 1977. Finite Element Simulation
of Salt Water Intrusion on the South Fork of Long Island.
In: Finite Elements in Water Resources, W. C. Gray, C. F.
Pinder, and C. A. Brebbia (eds.), Pentech Press, London.
368
-------
pohiand, F. G. 1975. Sanitary Landfill Stabilization with
Leachate Recyc’e and Residual Treatment. EPA-600/2-75-043.
tJ.S. Environmental Protection Agency, Cincinnati, Ohio.
105 pp.
pohiand, F. G. and J. P. Gould. 1980. Stabilization at Munici-
pal Landfills Containing Industrial Wastes. In: Dis osa3.
of Hazardous Waste. Proceedings of the Sixth Annual Research
Syi osiuZn, EPA—600—9—80—O10, U.S. Environmental Protection
Agency, Cincinnati, Ohio. pp. 242-253.
pohland, F. G., J. P. Gould, R. Ramsey, 3. J. Spiller, and W. R.
Esteves. 1981. Containment of Heavy Metals in Landfills
with Leachate Recycle. In: Land Disposal Municipal Solid
Waste. Proceedings of the Seventh Annual Research S posium
EPA-600/9—81-002a, U.S. Environmental Protection Agency,
Cincinnati, Ohio. pp. 179—194.
Portland Cement Assoc. 1974. Soil-Cement for Water Control
Skokie, Illinois. 102 pp.
preul, H. C., and C. J. Schroefper. 1968. Travel of Nitrogen
in Soils. J.W.P.C.F. 40: 30—48.
Qasim, S. R. 1965. Chemical Characteristics of Seepage Water
from Stimulated Landfills. Ph.D. Thesis, West Virginia
University, Morgantown, West Virginia.
Qasirn, S. R., and J. C. Burchinal. 1970. Leaching of Pollutants
from Refuse Beds. J. Sanitary Eng. Div. ASCE. 96(SA.1)
49—58.
Qasima, S. R., and J. C. burchinal. 1970. Leaching from Siinulat-
ed Landfills. J.W.P.C.F. 43(3):371—379.
Rawlens, S. L., and F. N. Dalton. 1967. Psychromatric Measure-
ment of Soil Water Potential without Precise Temperature
Control. Soil Sci. Soc. Amer. Proceedings. 31:201—297.
Raymond International, Inc. 1967. Sjroc Grout Technical
Manual, New York. 25 pp.
Reddell, D. L., and D. K. Sunada. 1970. Numerical Simulation
of Dispersion in Ground Water Aquifers. Hydrology Paper
41, Colorado State University, Fort Collins, Colorado.
79 pp.
369
-------
Reir.hardt, 3’. 3., and Ham, R. K. 1973. Final Report on a
Demonstrated Project at Madison, consii, to Investigate
Milling of Solid Wastes Between 196 and 1972, Vol. 1.
U.S. Environmental Protection Agency, Office of Solid
Waste Management Program, Washington, D.C.
Remson, I., A. A. Fungaroli, and A. W. Lawrence. 1968. Water
Movement in an Unsaturated Sanitary Landfill. 3. Sanitary
Engineering Division, Proceedings of ASCE. 94(SA2):307—
317.
Reuter, 3. H., arid E. M Perdue. 1977. Importance of Heavy
Metal-Organic Matter Interactions in Natural Waters.
Geochim. et. Cosmochilfl. Acta. 41:325.
Revah, A., and Y. Avnimeleih. 1979. Leaching of Pollutants
from Sanitary Landfill Models. 3’. Water Pollution
Control Federation. 51(11)2705-2716.
RicCiO, 3. F., and L. W. Hyde. 1971. Hydrogeology of Sanitary
Landfill Sites in Alabama; Preliminary Appraisal. Geologj
cal Survey of Alabama, Circular No. 71. 23 pp.
Riehi, H. 1972. Introduction to the Atmosphere, 2nd Ed.,
McGraw—Hill Book Co., Inc., Net ” York, N.Y. 89 pp.
Riley, R. D., R. C. Benedict, R. W. Seabloorn, and D. .;. Carison.
1977. Chemical and Microbiological Studies on Leachates
from a Waste Tip. 3. Applied Microbiology. 42:285—283.
Ringborri, A. 1973. Coinpiexation in Analytical Chemistry
IntersCieflCe Pubi., New York, N.Y. 493 pp.
Robertson, . 3. 1974. Digital Modeling of Radioactive and
Chemical Waste Transport in the Snake River Plain Acuifer
at the National Reactor Testing Station, Idaho, U.S.
Geological Survey Open File Report IDO-22054.
Robertson, 3. M., C. R. Toussaint, and M. A. Jorque. 1974.
Organic Compounds Entering Ground Water from a Landfill.
EPA 660/2—74—077, U.S. Environmental Protection Agency,
Washington, D.C.
Robertson, 3. 3., and 3. T. Barraclough. 1973. Radioactive and
Chemical Waste Transport in Ground Water at National
Reactor Station, Idaho: 20 Year Case History and Digital
Model. In: Underground Waste Mamagexnent and artificial
Recharge, Amer. Assoc. of Petroleum Geologists. 1:291—322.
370
-------
Robson, S. G. 1974. Feasibility of Water—Quality Modeling
Illustrated by Application at Barstow, California, Water
Resource Investigation. Report 46—73, U.S. Geol. Survey,
Menlo Park, California.
Rose, C. W. 1966. Agricultural Physics. Pergarnon Press, New
York, N.Y. 393 pp.
Rovers, F. A., and G. J. Farquhar. 1973. Infiltration and
Landfill Behavior. J. Environmental Engineering Div,
proceedings of ASCE. 99(EER):671-690.
Roulier, M. . 1977. AttenuatiOn of Leachata Pollutants by
Soils. EPA—60019—77026, U.S. Environmental Protection
Agency. pp. 127-138.
Rubin, A. J. 1974. Aquecus—EflViornmefltal Chemistry of Metals.
Ann Arbor Sicence Pub., Inc., Ann Arbor, Michigan. 390 pp.
Rabin, J., and R. V. James. 1973. Dispersion—Affected Trans-
port of Reacting Solutes in Saturated Porous Media:
Galerkin Method Applied to Equilibrium-Controlled Exchange
in Unidirectional Steady Water Flow. Water Resources
Research. 9(5) :1332—1361.
Salvato, J. A., W. G. Witkie, and 3. E. Mead.. 1971. Sanitary
Landfill Leaching Prevention and Control. J. W.P.C.E’.
43: 2084—210.
Sartks, R. L., J. H. LaPlante, and B. F. Gloyna. 1975. Survey
Suitability of Clay Beds for Storage of Industrial Solid
Wastes. Report ot Texas Water Quality Board. Technical
Report E -76—04, CRWR—L28. 95 pp.
sanitillafl-Medraflo, J., and J. J. Jurinak. 1975. The C iemistry
of Lead and Cadmium in Soil: Solid Phase Formation.
Soil Sci. Soc. Amer. Proc. 39:851—856.
savage, G. M. and G. J. Trezek. 1980. Significance of Size
Reduction in Solid Waste Management. EPA—600/2—80-1l5.
u.s. Environmental Protection Agency, Cincinnati, Ohio.
138 pp.
scarpino, P. V., and J. A. Donnelly. 1979. Pathogen Content of
of Landfill Leachate. In: Municipal Solid Waste: Land
Disposal. EPA600/979023a, U.S. Environmental Protection
Agency, Cincinnati, Ohio. pp. 138—167.
SchaUb, S. A., and B. P. Sagik. 1975. Association of Entero—
viruses with Natural and Artifically Introduced Colloidal
Solids in Water and Infectivity of Solids-Associated
Virions. Applied Microbiology. 30(3) :212-222.
371
-------
Schmugge, T. J., T. J. Jackson, and H. L. Mc in. 1980. Survey
of Methods for Soil Moisture Determination. Water Resources
Research. l6(6):961—979.
Schnitzer, M. 1971. Metal-Organic Matter Interactions in Soil
and Water. In: Organic Compounds in Aquatic Environments.
S. J. Faust and J. V. Hunter (eds.) , Marcel Dekker, Inc.
New York, N.Y. pp. 297—315.
Schultz, D. W., and M. P. Mikias. 1980. Assessment of Liner
Installation Procedure. In: Disposal of Hazardous Waste,
Proceedings of the Sixth Annual Research Symposium. EPA—
600/9—80—010, U.S. Environmental Protection Agency,
Cincinnati, Ohio. PP. 135—159.
Schwab, G. 0., et al. 1966. Soil and Water Conservation Eng-
ineering. John Wiley & Sons, New York, N.Y. 683 pp.
Schwartz, F. W. 1975. On Radioactive Waste Management: An
Analysis of the Parameters Controlling Subsurface Contamin-
ant Transfer. Journal of Hydrology. 27:51—71.
Schwartz, F. W. 1977. On Radioactive Waste Management: Model
Analysis of a Proposed Site. Journal of Hydrology. 32:
257—277.
SCS Engineers. 1976. The Selection and Monitoring of Land
Disposal Case Study Sites. Volume I. Project Description
and Findings. Contract No. 68-01—2973, U.S. Environmental
Protection Agency, Washington, D.C. 360 pp.
SCS Engineers. 1978. Investigation of Ground-Water Contamination
from Subsurface Sewage Sludge Disposal. Volume 2, Case Study
Reports. EPA 530/SW—167C, U.S. Environmental Protection
Agency, Washington, D.C. 533 pp.
Seaman, W. K. 1966. Revised Bibliography on Chemical Grouting.
Third Progress Report of the Conunittee on Grouting. J.
Soil Mechanics Foundation Div. ASCE, Vol. 91, No. SM6,
Proc. Paper 4469. pp. 39—66.
Segol, G. 19 76. A Three—Dimensional Galerkin-Finite Element
Model for the Analysis of Contaminant Transport in Variably
Saturated Porous Media. Dep. of Earth Sciences, Univ.
Waterloo. 171 pp.
Segol, C. 1977. A Three—Dimensional Galerkin-Finite Element
Model for the Analysis of Contaminant Transport in Saturated
Unsaturated Porous Media. In: Finite Elements in Water
Resources, W.G. Gray, et al. (eds.), Pentech Press, London.
pp. 2:123—2:144.
372
-------
segol, G., G. F. Pinder, and W. G. Gray. 1975. A Galerkin-
Finite E1er ent Technique for Calculating the Transient
position a: the Salt-Water Front. Water Resources Research.
11(2) :343347.
seliIn, H. M.,, J. N. Davidson, and P. 5. Rao. 1977. Transport
of Reactive Salutes through Multilayered Soils.
Soil Sci. Soc. Amer. J. 41(1):3—l0.
Seliia, H. M., R. S. Mansell, and A. Zizeftawy. 1976. Distribut-
ion of 2, 4-D and Water in Soil during Infiltration and
Redistribution. Soil Sci. 121:176—183.
Selim, H. M. and H. S. Mansell. 1976. Analytical Solution of
the Equation for Transport of Reactive Salutes through
Soils. Water Resrouces Research. 12:328532.
Sendlein, L. V., and H. Yazicigil. 1981. Surface Geophysical
Methods for Ground Water Monitoring. Ground Water Monitoring
Review, Fall 1981. pp. 42—46.
Shaf far, M. J, H. W. Ribbens, and C. W. Huntley. 1977. Pre-
diction of Mineral Quality of Irrigation Return Flow.
EPA600/277179e. Office of Research and Development,
U.S. Environmental Protection Agency, Ada, Oklahoma.
345 pp.
shah, D.. 3., G. A. Coulman, L. T. Novak, and 3. G. Ellis. 1975.
A Mathematical Model for Phosphorous Movement in Soils. J.
Environmental Quality. 4(l):87-92.
Shapiro, J. 1964. Effect of Yellow Organic Acids on Iron and
Other Metals in Water. J. American Water ! ork3 Assoc.
56(1062)
Shen, H. T. 1976. Transient Dispersion in Uniform Porous
Media Flow. J. Hydraulics Division, ASCE. 102(HY6): 707-
716.
Shoemaker, N. 3., and 3. L. Rittenhouse. 1980. Current Research
on Land Disposal of Hazardous Wastes. EPA-600/9—80010l.
U.S. Environmental Protection Agency, Cincinnati, Ohio.
pp. 1—14.
ShuckTOW, A. J., A. D. Pajak, and C. J. Touhill. 1981. Manage-
ment of Hazardous Waste Leachate. SW-87l, U. S. Environ-
mental Protection Agency, Cincinnati, Ohio. 147 pp.
373
-------
Shi. man, L. M. 1976. Adsorption Isotherms for Soil Clays with
and without Iron Oxides Removed. Soil Sci. Soc. ? mer. J.
40: 349—352.
Shuman, L. M. 1977. Adsorption of Zn by Fe and Al Hydrous Oxides
as Influenced by Aging and pH. Soil Sci. Soc. Amer. J.
41: 703—706.
Siegel, A. 1971. organic Compounds in Aquatic Enviror .rnents.
S. D. Faust and J. V. Hunter (eds.), Marcel Dekker,
New York. p. 265.
S1J.len, L. G. and A. E. Martell. 1964. Stability Constants of
the Metal-Ion Complexes, Spec. Pubi. No. 17, The Chemical
Society, London.
Sillen, L. G., and A. E. Martell. 1971. Stability Constants of
the Metal-Ion Complexes, Spec. Pu.bl. Nc. 25. The Chemical
Society, London.
Singer, I. A., and R. M. Brown. 1956. The Annual Variations of
Sub—Soil Temperatures about a 600—foot Circle: Transactions
of the American Geophysical Union, Vol. 37. p. 743.
Singer, P. C. 1972. Anaerobic Control of Phosphate by Ferrous
Ion. J. W.P.C.F. 44:663—669.
Skinner, F. A. and C. 3. Squesnel. 1978. Streptococci. Academic
Press, London, New YOrk. 415 pp.
Smajstrala, A. G., D. L. Redell, and A. E. Hiler. 1975. Simula-
tion of Miscible Displacement in Soils Using the Method of
Charzcteristics. ASAE Trans. 18(2):281.
Snoeyink, V. 0. and D. Jenkins. 1980. Water Chemistry, jo
Wiley and Sons, New York. 463 pp.
Sobsey, M. 0. 1975. Studies on the Survival and Fate of
Enteroviruses in and Experimental Model of a Municipal Solid
Waste Landfill and Leachate. J. Applied Microbiol.
30 (12) :565—574.
Sobsey, M. 0. 1978. Field Survey of Enteric Viruses in Solid
Waste Landfill Leachtes. Am J. Public Health. 68(9):85g—
863.
Soil Conservation Service. 1972. National Engineering Handbook
Sec. 4, Supplement A. Hydrology. U.S. Dept. of Agricu1tu ,
Washington, D. C. 667 pp.
374
-------
soil Moisture Equi.prrtent Corp. 1978. Operating Instructions for
the Model 1900 Soil Water Sampler, Santa arbara, California.
goiltest, Inc. 1976. Soil Testing Equipment, Evanston, Illinois.
157 pp.
staff Industries, Inc. 1976. General Instruction for Installa—
tjon of Staff Liners for Preventing Seepage from Ponds,
Resevoirs, Cnala, Lagoons, etc., Detroit, Michigan. 2 pp.
Steiner, R. L., J. D. Keenan, and A. A. Fungaroli. 1979. De-
monstrating Leachate Treatment Report on a Full-Scale
Operating Plant. EPA-SW-758, U.S. Environmental Protection
Agency, Washington, D. C. 143 pp.
Stevenson, F. J. 1976. Sta i1ity Constants fo Cu, Pb, and Cd
Complexes with Huinic Acids. Soil Scj. Soc. Amer. J.
40: 665—672.
steward, W. 5. 1978. State-of-the—art Study of Land Impoundment
Techniques. EPA 600/2—78-196, U.S. Erwironmnental Protection
Agency, Cincinnati, Ohio. 76 p .
stone, a. 1974. Disposal of Sewage Sludge into a Sanitary
Landfill. EPA—SW—71d, U.S. Environmental Protection Agency,
Cincinnati, Ohio. 418 pp..
strahier, A. M. 1971. The Earth Sciences, 2nd Ed., Harper and
Row Publishers, Inc., Mew York, M.Y. 120 pp.
straub, W. 1979. Survey of the Approaches and Alternatives
for Treatment of Leachates from Sanitary Landfills.
DSD #199. Resource Policy Center, Thayer School of
Engineering, Dartmouth College, Hanover, New Hampshire.
21. pp.
trath W. A. 1980. Development and Application of Models of
Sanitary Landfill Leaching and Landfill Stabilization. ap.
#259. Resource Policy Center, Thayer School of Engineering,
Dartmouth College, Hanover, New Hampshire. 180 pp.
traub , W. A. and D R. Lynch. 1979. Preliminary Modeling of
Moisture Movement and Contaminant Generation in Sanitary
Landfills. DSD # 158, Resource Policy Center, Thayer
School of Engineering, Dartmouth College, Hanover, Mew
Hampshire. 26 pp.
5 traub, W. A. and D. R. Lynch. 1982a. Models of Landfill and
Leaching: Moisture Flow and Inorganic Strength. J.
Environ. Eng. Division. ASCE, April pp. 95—119.
375
-------
Straub, W. A., and D. R. Lynch. 1982b. Models of Landfill
Leaching: Organic Strength. J. Environ. Eng. Division.
CE, April. pp• 95—119.
Streng, 0. R. 1976. The Effects of the-Disposal of Industrial
Wste Within a Sanitary Landfill Environment. In: Residual
Management by Land Disposal. EPA-60019-76—015.
u.s. Environmental Protection Agency, Cincinnati, Ohio.
pp. 51—70.
St nm, W., and J. J. Morgan. 1981. Aquatic Chemistry. John
Wiley & Sons, New York, N.Y. 780 p .
Sykes, J. F. 1975. Flue Gas Cleaning Sludge Leachate/Liner
Compatability Investigation: Interim Report. EPA-600/2-79—
136, u.s. Environmental Protection Agency. 61 pp.
Sykey, J. F. 1975. Transport ?henomean in Variably Saturated
Porous Media. Ph.D. Thesis. Dept. of Civil Engineering,
Uiversity of Waterloo.
Swanson, R. A. , and G. R. Dutt. 1973. Chemical and Physical
Processess that Affect Atrazine Movement in Dsitribution
in Soil Systems. Soil Sci. Soc. Amer. Proc. 37:872—876.
Tang, D. H., and G. F. Pinder. 1977. Simulation of Ground Water
Flow and Mass Transport under Uncertainty. Advances in
Water Resources. 1:25—30.
Tanji, K. K., C. R. Dutt, J. L. Paul, and L. 0. Doneen. 1967.
Quality of Percolating Waters. A Computer Method for
Predicting Salt Concentration in Soils at Variable
Moisture Contents. Hilgardia. 38:307—318.
Tanji, K. K. 1970. A Computer Analysis on the Leaching of Boron
from Stratified Soil Co1t xnns. Soil Scj. 110:44-51.
Tanji, K. K., L. 0. Doneen, C. V. Ferry, and R. S. Ayers. 1972.
Computer Simulation Analysis on Reclamation of Salt-Affected
Soils in San Joaquin Valley, California. Soil Sci. Soc.
Pmer. Proc. 36(l):127—133.
Taylor, R. M., and R. M. McKinzie. 1966. The Association of
Trace Elements with Manganese Minerals in Australian Soils.
Aust. J. Soil Res. 4:29—39.
Thomas, R. L. C. G. Smith, and J. D. Martinez. 1977. Domal
Salt Pitunes Ground Water. In: Finite Elements in Water
Resources, w. C. Gray, G. F. Pinder, and C. A. Brebbia
(eds.). Pentech Press, London. pp. 237—249.
376
-------
Thornton, R. J., and F. C. Blanc. 1973. Leachate Treatment by
Coagulation and ?:acipitation. Journal of the Environmental
Engineering Division, ASCE, 99(EE4) :535—544.
Thornwaite, C. W., and J. R. Mather. 1955. The Water Balance
Pubicatioris in Climatology, Lab. of Climatology, Drexel
Inst. of Tech. Centerton, New Jersey. 3(l):1-104.
Thornthwaite, C. W., and J. R. Mather. 1957. Instructions and
Tables for Computing potential Evapotranspira.tion and the
Water Balance. Publication in Climatology, Lab. of China—
tology, Drexel Inst. of Tech., Centerton, New Jersey.
10(3) :185—311.
Timco Manufacturing Co., Inc. 1980. Timco Geo-Technical
Products, Prairie du Sac, Wisconsin.
Todd, S. K., R. M. Tinhin, K. D. Schmidt, and L. G. EVerett.
1976. Monitoring Ground-Water Quality: Monitoring
Methodology. Gnerai. Electric Co. TEMPO Center for Ad-
vanced Studies, Santa Barbara, California. 154 pp.
Toi.man, A. L., A. P. Bahlestero, Jr., W. W. Beck, Jr., and
G. H. nrich. 1978. Guidance Manual for Minimizing
Pollution from Waste Disposal Sites. EPA—600/2—78—].42,
U.S. Environmental Protection Agency, Cincinnati, Ohio.
82 pp.
5on merfeldt, T. G., and D. B. Campbell. 1975. A Pneumatic
System to Pump Water from Piezotneters. Ground Water.
13(3) :293
TresCOtt, P. C., and C. F. Pinder. 1970. Air Pump for Small
Diameter Piezometers. Ground Water. 8(3):10—15.
Uloth, V. C. and D. S. Mavinec. 1977. Aerobic Treatxnent of a
High Strength Leachate. Journal of the Env. Bag. Div.
ASCE, 103(EE4) :647—745.
ungs, M., R. W. Cleary, L. Boersina, and S. Ynzjajaval. 1976.
The Quantitative Description of Transfer of Water and
Chemicals through Soils. In: Land as a Waste Management
Alternative. R. C. Loeki (ed.). 298 pp.
urquhart, L. C. (ed.). 1959. Civil Engineering Handbook, 4th
Ed. McGraw—Hill Book Co., Inc., New York, N.y. 1146 pp.
377
-------
U.S. A y Engineer Waterways Experiment Station. 1978. Chemi-
cal and Physical Effects of M rnicipal Landfills on
Underlying Soils and Gro’ md Waters. EPA—600/2-78-096.
U.S. Environmental Protection Agency, Cincinnati, Ohio.
103 pp.
U.S. Department of Agriculture. 1975. Soil Conservation
Service. Standards and Specifications for Soil Erosion
and Sediment Control in Developing Areas. College Park,
Maryland. 72 pp.
U.S. Public Health Service. 1976. ClassificatiOn of Etiologic
Agents on the Basis of Hazard. Dept. of Health, Education,
and Welfare. Center for Disease Control, Atlanta, Georgia.
U.S. Soil Conservation Service. 1971. A Guide for nterpreting
Engineering Uses of Soil. U.S. Conservation Service. U.S.
Printing Office Stock No. 0107-0332.
U.S. Environmental Protection Agency. 1973. Handbook for
Monitoring Industrial Wastewater. TechlloloTJ Transfer
Series. u.s. Government Printing Office, 732—349/414.
175 pp.
u.s. Environmental Protection Agency. 1974. Methods for
Chemical Analysis of Water and Wastes. EPA—625/6—74—003,
Methods Development and Quality Assurance taboratory,
Cincinnati, Ohio. 298 pp.
U.S. Environmental Protection Agency. 1975. Manual of Water
Well Construction Practices. EPA—570/9—75-00l, Office
of Water Supply. 136 pp.
U.S. Environmental Protection Agency. 1979. Handbook for
Analytical Quality Control in Water and Wastewater Labora-
tories. EPA—600/4-79 0 l 9 . Environmental Monitoring and
Support Laboratory, Cincinnati, Ohio. 115 pp.
U.S. Environmental Protection Agency. 1979. Criteria for
Classification of Solid Waste Disposal Facilities. Federal
Register, Thursday, Sept. 13.
U.S. Environmental Protection Agency, 1980. Hazardous Waste
arid Consolidated Permit Regulations. Federal Register
Vol. 45, No. 98, nook 2. pp. 33063—33285.
van Gerxuchten, M. Th., J. M. Davidson, and P. 3. Wierenga. 1974
An Evaluation of Kinetic and EquilibriUlLl Equations for the
Prediction of pesticides Movement Through Porous Media.
Soil Sci. Soc. Artier. Proc. 38:29—35.
378
-------
van Genuchten, M. Th., and P. J. Wierenga. 1976. Mass-Transfer
Studies in Sorbing Porous Media. Analytical Solutions.
Soil Sci. Soc. Amer. J. 40:473—480.
Van Genuchten, M. Th., G. F. Pinder, and W. F. Saukin. 1979.
Modeling of Leachate and Soil Interactions ilran Aquifer.
In: Proceedings of the Third Annual Syrnposiim on the
Management of Gas and Leachate from Landfills. EPA—600/9-7-
025, J.S. Environmental Protection Agency, Cincinnati, Ohio.
pp. 95—103.
van Genuchten, M. T.,and G. F. Pinder. 1977. Mass Transport
in Saturated-Unsaturated Media. Research Retort 77-WR-)
Water Resources Program, Dept. of Civil Engineering.
Princeton, New Jersey.
van Genuchten, M. Th. 1978. •Simulaticn Models and their
Application to Landfill Disposal Siting; A Review of
Current Technology. Land Disposal of Hazardous Wastes
Proceedings of the Fourth Annual Research Symposium.
EPA-600/9-78-Q16, U.S. Environmental Protection Agency,
Cincinnati, Ohio. pp. 191—214.
Genuchten, M. Th., and R. W. Cleary. 1978. Movement of
Solutes in Soil: Contouter-Sjznuiated and Laboratory Results
Soil Chemistry. C. H. Bolt and M.G. M. Bruggewert (eds.),
Chap. 10. Developments in Soil Science, Elsevier Pu.bl.
Co., Amsterdam.
eihmeyer, F. J. 1919. An Improved Soil-Sampling Tube. Soil
Science 27(2) 147—152.
Vejhmeyer, F. J. 1964. Evapotranspiration. In: Handbook of
Applied Hydrology, Section 11, V. T. Chow, (ed.), McGraw-
Hill Book Co., New York, N.Y. pp. 11—1 to 11—38.
ejhmeyer, F. J. and A. H. Henderickson. 1955. Rates of Evapora-
tion from Wet and Dry Soils and Their Significance. Soil
Sci. 80:61—67.
Viessman, çq•, Jr ., J. W. Knapp, C. L. Lewis, arid T. E. Harbaugh.
1972. Introduction to Hydrology (2nd Ed.) Harper and Row,
Pubi.’,: Mew York, N.Y. 704 pp.
vyda, 0. N. and A. Grixnra. 1976. County Treats a Shredfill
Leachate. Civil Engineering, ASCE, 102 (EES) :44-49.
qaijcnafl, S. A. 1952. Soil Microbiology. John Wiley & Sons,
New York, N.Y. 362 pp.
379
-------
Walsh, 3. 3. and R. N. Kinmari. 1979. Leachate and Gas Production
Under Controlled Moisture Conditions. In: Municipal Solid
Waste: Land Disposal. EPA—600/9—79—023a, U.S. nviror.znental
Protection Agency, Cincinnati, Ohio. pp. 41-57.
Walsh, 3. J., and R. N. Kinxaan. 1981. Leachate and Gas From
Municipal Solid Waste Landfill Simulators. In: Land
Disposal: Municipal Solid Waste, D. W. Shultz, (ed.). EPA—
600/9—81-002a, U.S. Environmental Protection Agency,
Cincinnati, Ohio. pp. 67—93.
Wang, S. T., A. F. McMullan, arid B. H. Chen. 1977. Analytical
Model of Dispersion in Tidal Fjords. J. Hydraulics
Division, ASCE 103(HY7):737—751.
Ware, S. A., and C. S. Jackson. 1978. Liners for Sanitary
Landfills and Chemical and Hazardous Waste Disposal
Sites. EPA—600/9—78—00S, U.S. Environmental Protection
Agency, Cincinnati, Ohio. 77 pp.
Ware, S. A. 1980. A Survey of Pathogen Survival During
Municipal Solid Waste and Manual Treatment Processes. EPA—
600/8-80-034, U.S. Environmental Protection Agency,
Cincinnati, Ohio. 100 pp.
Warrick, A. W., 3. W. Biggar, arid D. R. Nielsen. 1971. Simulta-
neous Solute and Water Transfer for an Unsaturated Soil.
Water Resrouces Research. 7(5) :1216-1225.
Watersaver Co., Inc. 1980. Standard Specification: Polyvinly
Chloride Plastic Linings: Data Sheet SPVC-74, Denver,
Colorado. 4 pp.
Watson, K. K. 1967. A Recording Field Tensionxneter with Rapid
Response Characteristics. Journal of Hydrology. 5:33-39.
Watson, K. K. 1974. Some Applications of Unsaturated Flow—
Theory. In: Drainage for Agriculture. 3. van Schifgaarde
(e d .) , Agronomy No. 17, P xaerican Society of Agronomy,
Madison, Wisconsin.
Weber, W. 3., and H. S. Posselt. 1976. Equilibrium Models and
Precipitation Reactions for Cadniiwn (II). Aqueous Environ-
mental Chemistry of Meals, A. 3. Rubin (ed.) Ann Arbor
Science Pub., Inc., Ann Arbor,Nichigan. pp. 255—289.
Wellings, F. M. 1974. Virus Survival Following Wastewater
Spray Irrigation of Sandy Soils. Virus Survival in Water
arid Wastewater Systems. 3. F. Malina and B. P. Sagik, (eds.)
University of Texas, Austin.
380
-------
e1lings, F. M. 1975. Virus is Gro d Water, Proceedings of
the 2nd National Conference on Individual Or.site Wastewater
Systems. National Sanitation Foundation, Ann Arbor,
Michigan.
West Coast Technical Service, 1981. Fee Schedule, Cerritos,
California. 6 pp.
Weston, Inc. 1978. Pollution Prediction Techniques for Waste
Disposal Siting. EPA—SW—162C, U.S. Environmental Protection
Agency, Cincinnati, Ohio. 440 pp.
wjgh, R. 1971. Evaluation of the MC-300A Soil Moist e Meter
to Determine In-Place Moisture Ccntent of Refuse at Land
Disposal Sites. P3—256359. 85 pp.
R. J., 1979. Boone County Field Site Interiin.Report.
EPA—600/2-79—058, U.S. Environmental Protection Agency,
Cincinnati, Ohio. 202 pp.
wigh, R. J. 1979. Boone County Field Site Interim Report, Test
Cells 2A, 23, 2C, and 2D. EPA—600/2—79—058, U.S. Environ—
mental Protection Agency, Cincinnati, Ohio. 97 op.
igh, R. J. 1981. Stmunary of Landfill Research Boone County
Field Site. In: Land Disposal: Municipal Solid Waste,
0. J. Shultz, (ed.) EPA-600/9/81—002a, U.S. Environmental
Protection Agency, Cincinnati, Ohio. pp. 209—242.
jgh , R. J., and D. R. Brunner. 1981. S mimary of Landfill
Research, Boone County Field Site, In: Land Disposal:
Municipal Solid Waste EPA600/9—81 002a. U.S. Environmental
Protection Agency, Cincinnati, Ohio. pp. 209—242.
Wi)claflder, L. 1974. Cation and Anion Exchange Phenomena.
Chemistry of the Soil. F. E. Bear (ed.) Reinhold PUb1.
Corp, New York. pp. 163—205.
çjilson, L. C. 1980. Monitoring in the Vadose Zone. General
Electric Co.-TEMPO Center for Advanced Studies, Santa
Barbara, California. EPA-600/7—80—134. U.S. Environmental
Protection Agency, Las Vegas, Nevada. 168 pp.
jj1son, L. C. 1981. Monitoring in the Vadose Zone. Part I:
Storage Changes. Ground Water Monitoring Review, Fall, 1981.
pp. 32—41.
WOOd, 1. W. 1973. A Technique Using Porous Cups for Water
sampling at any Depth in the Unsaturated Zone. Water
Resources Research ((2):486—488.
3 1
-------
Wood, A. L., and J. M. Davidson. 1975. Fluometuron and Water
Content DistributiCfl: Measured and Calculated. Soil
Sci. soc; ? er. Proc. 39:820—325.
Ysh, G. and Y. Tsai. 1976. Ana1ytic 1 Three-Dimensional
Transient Modeling of Effluent Discharges. Water Resources
Research. 12:533—540.
Zononi, A. E., 1972. Ground-water Pollution and Sanitary Land-
fills-A Critical Review. Ground Water. 10(1):16 pp.
Zenone, C., D. E. Donaldson, and J. J. Grunwaldt. 1975. Ground-
Water Quality Beneath Solid Waste Disposal Sites at Anchorage
Alaska. Ground Water (13(2):182—190.
382
-------
A??ENDIX A. T ORNTEWAITE TA3L S
TABLE A1. r r /ALUSS OF j C R ESP JN0 I G TO NT LY ZA TE ?E ATu. 53 (‘C)
*
T’C .0 .1 .2 .3 .4 .5 .5 .7 .8 .9
fi
1 . 51 4
roducedf
3
1
2
3
4
.09
.23
.46
.11
.10
. 37
.48
.74
.01
.12
.29
.51
.77
.01
.13
.31
.53
.30
.02
.13
.33
.36
.32
.33
.13
.35
.38
.35
.04
.15
.37
.51
.58
.35
.20
.39
.63
.31
.05
.21
.42
.65
.94
.07
.23
.44
.69
.97
5
5
7
3
9
1.30
1.32
1.65
2.04
2.44
1.33
1.35
1.73
2.03
2.48
1.06
1.39
1,74
2.12
2.32
1.09
1.42
1.77
2.15
2.55
1.12
1.45
1.61
2.19
2.50
1.15
1.49
1.35
2.23
2.64
1.19
1.52
1.39
2.27
2.99
1.22
1.56
1.32
2.31
2.73
1.25
1.59
1.36
2.35
2.i7
1.29
1.33
2.00
2.39
2.31
10
11
12
13
14
2.85
3.30
3.76
4.25
4.73
2.30
2.24
2.31
4.30
4.31
2.34
3.39
3.35
4.33
4.36
2.39
3,44
3.91
4.’0
4.91
3.33
3. 5
3.96
4.4S
4.96
3.33
3.53
4.00
4.30
5.01
3.12
3.53
4.05
4.55
5.37
3.16
3.32
4.10
4.50
5.12
3.21
3.67
4.15
4.55
5.17
3.25
3.72
4.20
4.70
5.22
15
15
17
18
12
5.23
5.82
8.38
8.95
7,55
5.33
5.37
5.44
7.01
7.61
5.38
5.93
8.49
7.07
7.67
5.44
5.98
6.35
7.13
. 73
5.49
5.34
3.51
7.19
7.79
5.35
S. 0
8.55
7.25
7.55
3. 0
S.iS
6.72
1.31
7. 51
5. 5
5.21
5.73
7.37
7. 97
Se 71
5.25
6.34
7.43
9.03
5.75
5.32
5.30
1.49
3.10
20
21
22
23
24
3.16
8.73
9.42
10.03
10.75
8.22
3.35
9 49
10.15
10.52
8.23
8.91
9.35
10.21
10.89
3.34
8.91
9.62
10.23
10.95
3.41
9.04
9.53
10.35
11.02
3.47
9.10
9.75
10.41
11.09
8.33
9.17
9. 2
10.43
11,15
3.59
9.23
9.38
10.55
11.23
8.56
9.29
9.95
10.92
11.30
8.72
9.38
10.01
10.58
11.37
25
26
27
28
29
11.44
12.13
12.85
13.58
14.32
11,30
12.21
12.52
13• 5
14.39
11.57
12.23
12.39
13.72
14.47
11.54
12.35
13.37
13.80
14.54
11.71
12.42
13.14
13.37
14.62
11.78
12.49
13.21
13.94
14.69
11.35
12.56
13.29
14.02
14.77
11.92
12.53
13.38
14.09
14.34
11.99
12.70
13,43
14.17
14.92
12.06
12.78
13.50
14.24
14.99
30
31
32
33
34
15.07
13.34
18.62
17.41
13.22
13.13
13.52
16.70
17.49
18.30
13,22
15 ,99
15.78
17.57
18.38
15.30
16.07
16.85
11.85
18.46
13,33
18.15
13,93
17.73
18.54
15.45
16.23
17.01
17.31
13.52
15.33
16.30
17.39
17.89
18.70
15.61
16.38
17.17
17.37
18.79
15.63
16.46
17.25
18.05
18.57
15.75
16.34
17.33
18.13
18.95
•
35
36
37
38
39
40
19.03
19.36
20.70
21.56
22.42
23.30
19.11
19.35
20.79
21.64
22.31
19.20
20.03
20.87
21.73
22.59
19.29
20.11
20.96
21.81
22.50
.
19.35
20.20
21.04
21.50
22.77
19.45
20.23
21.13
21.39
22.86
19.53
20.36
21.21
22.37
22.95
19.51
20.43
21.30
22.16
23.03
19.69
20.53
21.38
22.25
23.12
19.73
20.52
21.47
22.33
23.21
3 3
-------
T 3L A-2.
3 3.0 0.0
.25 0.1 0.3
.50 3.1 3.1
.73 0.2 0.2
130 0.2 3.2
1.23 3.3 0.3
1.30 0.3 0.3
1.75 0.4 0.4
2.33 0.4 0.4
2.25 3.5 0.5
2.50 3.5 0.3
2.73 3.5 2.6
3.30 0.6 0.3
3.25 0.7 0.7
3.50 3.7 2.7
3.75 0.3 0.7
4.30 0.3 0.3
4.25 0.3 0.3
4.50 0.9 0.3
4.75 1.3 0.3
5.30 1.0 1.0
5.25 1.1 1.3
5.50 LI 1.1
3.73 1.2 1.1
1.2 9.2
1.2 9.2
1.3 9.2
1.3 1.3
1.4 1.3
1 4 9.4
1.5 1.4
1.3 1.4
1.6 9.5
1.5 1.5
1.5 1.5
1.7 1.1
1.7 1.5
Li 9.7
1 . 5 1.7
LI 1.5
1.3 1.3
1.3 1.3
2.0 1.3
2.3 2.0
2.1 2.0
2.1 2.0
2.1 2.1
2.2 2.3
2.2 2.2
2.3 2.2
2.3 2.3
2.3 2.3
2.4 2.3
2.3 0.0 0.3 0.0 0.0 0.0
0.0 0.0 0.0 0.3 3.5 0.3
0.1 0.1 0.9 3.1 0.1 3.3
0.9 0.1 0.1 0.1 0.4 0.1
0.2 0.2 0.2 3.1 0.1 0.1
0.2 0.2 0.2 3.1 0.1 0.1
0.2 0.2 0.2 0.2 0.2 0.2
0.3 0.3 0.3 0.2 3.2 0.2
0.3 2.2 0.3 0.2 3.2 0.2
0.4 0.3 2.3 0.3 3.3 0.3
0.4 0.4 0.4 0.3 0.3 0.3
0.5 0.4 0.4 0.4 0.4 0.4
3.5 3.5 3.3 2.4 0.4 3.4
3.5 0.5 2.5 0.5 0.4 0.4
0.5 0.5 0.5 3.3 0.5 3.5
0.5 0.3 3.5 3.3 2.5 0.2
0.7 0.5 0.5 0.6 0.5 0.5
0.7 0.7 0.5 0.5 0.5 2.3
0.3 0.7 0.7 0.5 2.5 0.5
0.3 0.3 3.7 0.7 0.5 0.5
0.3 0.3 0.3 0.7 0.7 0.7
0.3 3.3 0.3 0.3 3.7 0.7
1.0 0.3 0.3 0.3 0.3 0.7
1.3 1.3 0.3 0.3 0.3 0.3
1.0 0.3 0.3 0.3
1.3 1.0 3.9 3.3
1.1 1 ,3 0.3 0.3
9.1 1.0 1.3 0.3
1.! 9.1 1.0 1.3
1.2 1.1 1.0 1.3
1.2 1.2
1.3 1.2
1.3 1.3
1.4 9.3
9.4 1.4
1.5 1.4
1.1 1.3
1.1 hI
1.2 hI
1.2 1.2
1.3 9.2
1•3 9.5
9.3 9.3
1 4 1.3
1.4 9.4
9.5 1.4
1.3 9.4
9.3 1.5
1.1 1.3
1.1 1.5
1.7 1.6
hi 1.5
1.7 1.7
1.3 1.7
9.3 1.1 1.3 1.3
2.0 9.9 1.5 1.3
2.0 2.0 i.3 1.5
2.1 2.0 2.0 9.3
2.1 2.1 2.0 2.0
0.3 0.0 3.0 0.0
0.3 0.0 0.0 2.0
0.0 0.0 0.3 0.0
0.1 0.1 0.0 0.0
0.1 0.1 0.1 0.1
0.1 0.3 0.1 0.1
0.2 0.1 0.9 5.1
0.2 0.2 0.2 3.1
3.2 0.2 0.2 0.2
0.3 2.2 0.2 0.2
0.3 0.3 3.2 3.2
0.3 0.3 0.3 3.3
2.3 0.3 0.3
2.4 0.4 0.5
2.4 0.4 3.4
2.4 0.4 3.4
3.5 0.4 3.4
3.5 0.5 3.4
0.5 0.5 0.3
2.5 0.5 0.5
0.1 2.5 0.5
0.3 0.! 3.3
0.5 0.6 3.5
0.1 0.1 0.6
3.4
0. 4
3. 4
0.2
0.5
3.3
0.5
0.5
0.5
3.1
0.7
0. 7
0.3
0.3
0. 0
0.3
0.9
I.0
0.7 3.7 0.7
0.7 0.7 0.7
0.3 0.3 0.7
0.3 0.3 0.5
0.3 0.3 3.3
0.3 0.3 0.3
1.3 0.3 0.9 0.3
1.3 1.0 1.0 0.3
1.1 1.0 hO 1.0
hI 1.1 1.1 1.3
1.2 1.1 1.1 1.0
1.2 1.1 1.1 1.1
1.2 1.2 1.2 1.1
1,3 1.2 1.2 1.1
9.3 t.3 1.2 1.5
1.4 9.3 9.3 1,2
9.4 1.4 1.3 1.3
1.4 9.4 1.4 1.3
1.4 1.4 1.4 1.4
1.3 9.5 1.5 1.4
1.6 1.3 1.5 1.4
1.5 1.5 1.3 1.5
1.3 1.5 1.5 1.5
1.7 1.1 9.1 1 3
1.7 1.7 1.7 1.3
1.3 1.7 1.7 1.1
1.3 9.3 1.3 1.7
1.9 1.3 1.3 1.7
1.3 1.3 1.3 1.3
* Yeirly total of Tnonth y heat 1r ex
t ilable PY.
VALUL Of UNA3 . UZTZ0 3A L ?OT7. flAL VAPOTRA SPIRATICN (r i)
FOR OIFF(RLNT i A TZrPE ATU0C5 ( ) 0 YALIJ(3
23.0 27.5 20.0 32.5 25.0 37.5 40.0 42.5 43.3 47.5 50.0 52.5
9.1
1.9
1.1
9.2
1.2
‘.3
1.3
‘.4
1.4
1.4
9.3
1.3
1.5
1.5
1.1
1.7
1.7
9.3
1.3
1.3
1.3
1.3
2.0
2.0
L00
6.25
5.30
‘75
7.30
7.20
7.30
7.75
5.00
6.25
3.30
‘73
3.00
3.23
3.33
5.73
10.00
10.25
10.50
10.75
91.00
11.23
01.50
11.75
92.00
12.23
12.50
12.75
93.00
0.1
0.9
1.0
1 . 5
1.3
1.1
1.1
1.2
1.2
1.3
1.3
1.3
1.4 1.4
1.5 1.4
9.5 1.5
LI 1.4
1.5 1.5
1.7 1.5
9.7 1.1
1.5 1.7
1.5 1.7
9.3 1.3
1.3 1.3
1.9 1.3
1.5
1.5
1.5
1 . 5
1.7
1 ,7
‘.3
1.3
‘.3
9.9
1.9
1.3
2.1 2.0
2.1 2.0
2.2 2.2
2.2 2..l
2.3 2.2
384
-------
Reproduced
TA3LZ
A-2
(C3r t ued)
,•.
,
13.23
13.50
13.73
14.00
14.23
14.30
23.3 21.5
2.4 2.4
2.3 3.4
2.3 2.4
2.3 2.3
2.1 2.3
2.3 2.3
30.3
2.3
2.3
2.4
2.4
2.4
2.3
32.5
3.2
2.3
2.3
2.3
2.4
2.4
33.0
2.2
2.2
2.3
2.3
2.3
2.4
31.5
2.1
2.2
2.2
2.2
2.3
2.3
40.3
2.1
2.1
3.1
2.2
2.2
2.3
42.3
3.0
3.1
2.1
2.2
2.
2.3
45.3
2.3
3.3
3.3
2.1
2.1
2.2
•i.
1.9
2.3
3.3
2.3
2.1
2.1
so.o
1.3
3.0
3.3
2.3
2.1
2.1
1.3
1.9
1.3
1.3
2.3
3.3
14.75
15.00
15.25
13.30
15.73
13.30
2.7 3.1
2.7 2.5
2.3 3.7
2.3 2.7
2.3 2.3
2.3 2.3
2.3
2.5
2.3
2.7
2.7
2.3
2.3
2.5
2.5
2.3
2.7
2.1
2.4
2.5
2.3
3.5
3.3
3.7
7.4
2.4
2 .3
2.3
3.3
2.3
2.3
3.4
2.4
2.3
2.3
2.5
2.3
2.3
2.4
3.4
2.3
7.3
2.2
2.3
2.3
3 4
2.4
2.3
2.2
2.2
2.3
2.3
3.4
7 4
2.2
2.2
2.3
2.3
2.4
2.4
2.1
2.2
2.2
2.2
3.3
2.3
23.25
13.50
13.75
17.30
17.25
17.50
3.3 2.3
3.3 3.9
3.0 2.9
3.0 3.3
3.1 3.3
3.1 3.3
2.3
2.8
2.9
2.3
2.3
3.0
2..?
2.3
2.3
2.3
2.3
3.3
2.7
2.7
2.3
2.3
2.9
3.3
2.5
3.7
2.7
2.3
2.3
2.9
2.5
2.5
2.7
2.?
2.3
2.3
2.3
2.5
2.5
2.7
2.7
2.3
3.5
2.5
2.5
2.5
3.7
2.?
2.4
2.3
3.5
3.3
2.3
2.7
2.4
3.3
2.5
2.5
2.3
2.7
2.4
3.4
2.3
2.5
2.3
2.1
17.75
13.00
11.29
13.30
18.75
13.30
3.2 3.1
3.2 3.1
3.2 3.2
3.3 3.2
3.3 3.2
3.3 3.3
3.3
3.1
3.1
3.2
3.2
3.2
3.0
3.1
3.1
3.1
3.2
3.2
3.3
3.3
3.2
3.1
3.1
3.2
2.9
3.0
3.3
3.1
3.1
3.1
2.3
2.9
3.0
3.3
2.2
3.1
2.3
2.3
3.9
3.3
3.0
3.3
2.3
2.9
2.3
3.0
3.3
3.3
2.8
2.3
2.3
2.3
2.9
3.0
2..?
2.3
2.3
2.9
2.9
3.3
2.1
2.7
2.8
2.3
2.9
2.9
13.23
13.33
13.15
20.30
20.25
20.50
3.4 3.3
3.4 3.4
3.3 3.4
3.5 3.5
3.5 3.3
3.5 3.3
3.3
3.3
3.4
3.4
3.5
3.3
3.2
3.3
3.3
3.4
3.4
3.3
3.2
3.3
3.3
3.4
3.4
3.3
3.2
3.2
3.3
3.3
3.4
3.4
3.1
3.2
3.2
3.3
3.3
3.4
3.1
3.t
3.2
3.2
3.3
3.3
3.1
3.1
3.2
3.2
3.3
3.3
3.3
3.1
3.1
3.2
3.2
3.3
3.3
3.1
3.1
3.2
3.2
3.3
3.3
3.0
3.1
3.1
3.2
3.,2
20.75
21.30
21.25
21.50
21.75
22.30
3.? 3.1
3.7 3.3
3.7 3.5
3.3 3.7
3.3 3.7
3.2 3.3
3.3
3.5
3.5
3.5
3.7
3.7
3.3
3.3
3.5
3.3
3.1
3.7
3.5
3.5
3.3
3.5
2.5
3.?
3.5
3.3
3.3
3.5
3. .l
3.7
3.4
3.5
3.3
3.5
3.5
3.3
3.4
3.4
3.4
3.3
3.5
3.5
3.4
3.4
3.4
3.3
3.3
3.5
3.3
3.4
3.4
3.5
3.3
3.5
3.3
3.4
3.4
3.3
3.3
3.5
3.3
3.3
3.4
3.4
3.5
3.5
22.50
22.75
23.20
23.25
23.30
3.3 3.3
4.0 3.3
4.0 3.9
4.1 4.3
4.1 4.0
2.3
3.3
3.3
3.9
4.3
4.0
3..3
3.3
3.3
3.9
3.3
4.0
3.3
3.4
3.3
3.3
4.3
3.7
3.3
3.3
3.3
3.3
4.0
3..?
3..?
3.3
3.5
3.3
3.3
3.7
3.7
3.8
3.3
3.3
3.3
3.5
3.7
2.?
3.3
3.3
3.3
3.5
3.7
2.1
3.3
3.8
3.3
3.3
3.7
2.7
3.3
3.3
3.9
3.1
3.1
2.7
3.4
3.3
3.3
23.75
24.00
24.23
24.30
24.75
23.00
4.5 4.0
4.2 4.1
4.2 4.1
4.2 4.1
4.3 4.2
4.3 4.2
4.3
4.1
4.1
4.1
4.2
4.2
4.0
4.3
4.1
4.1
4.2
4.2
4.0
4.3
4.1
4.1
4•3
4.2
4.0
4.0
4.1
4.1
4.2
4.2
4.3
4.3
4.1
4.1
4.2
4.2
3.9
4.0
4.3
4.5
4.1
4.2
3.3
4.0
4.0
4.1
4•3
4.2
3.3
4.3
‘.3
4.1
4.1
4.2
3.3
4.3
4.0
4.1
4.1
4.2
3.3
3.1
4.0
4.1
4.1
4.2
25.25
25.50
23.73
23.00
25.25
25.30
4.3 4.3
4.4 4.3
4.4 4.4
4.5 4.4
4.5 4.5
4.3 4.5
4.3
4.3
4.4
4.4
4.3
4.3
4.3
4.3
4.4
4.4
4.5
4.5
4.3
4.3
4.4
4.4
4.5
4.5
4.3
.3
4.4
4.4
4.5
4.3
4.3
4.3
4.4
4.4
4.5
4.5
4.2
.3
4.4
4.4
4.3
4.5
4.2
.3
4.4
4.4
4.3
4.3
4.2
4.3
4.4
4,4
4.5
4.3
4.2
4.3
4.4
4.4
4.3
4.3
4.2
4.3
4.4
4 ,4
4.3
4.3
385
-------
-3I2 A—2 ((Contiflued)
r d rom
5.3 51.5 60.0 62.5 62.3 51.5 70.0 72.5 15.3 71.5 00.0 02.5
•
0
.25
.50
.75
1.20
1.25
3.0
0.0
0.0
0.0
0.1
3.0
0.0
0.0
3.0
0.0
0.1
0.3
0.3
3.0
0.0
0.3
0.0
0.0
0.0
0.3
0.0
0.0
0.0
0.3
0,0
0.0
0.0
0.0
0.3
0.0
0.0
0.0
0.0
0.0
0.3
0.0
3.0
0.0
0.2
0.2
0.0
0.5
0.0
0.0
0.0
0.0
0.3
0.0
0.3
0.0
0.0
0.0
0.0
0.3
0.0
0.0
0.0
3.0
0.0
0.0
0.0
0.0
0.3
0.3
0.0
0.3
0.3
0.0
3.0
0.0
0.0
0.0
1.50
1.75
2.00
2.25
2.50
2.75
0.1
3.1
3.1
0.2
3.2
0.2
0.1
3.1
3.1
3.1
0.2
0.2
0.1
0.0
3.1
0.1
0.2
0.2
0.0
3.0
0.1
0.1
0.1
0.1
0.0
0.0
0.1
0.2
0.1
0.1
0.0
0.0
0.1
0.1
0.0
0.0
0.0
0.0
0.0
3.1
0.1
0.1
0.3
0.0
0.0
0.1
0.1
0.1
0.0
0.0
0.0
0.1
0.1
0.1
0.0
0.0
0.0
0.1
0.0
0.1
0.0
3.0
0.3
0.1
0.1
3.1
0.0
0.2
0.0
0.0
‘3.1
0.0
33
3.25
3.30
3.75
4.20
4.25
0.3
0.3
0.3
0.3
3.4
0.4
0.2
0.1
0.3
0.3
0.3
0.4
0.2
0.2
0.3
0.3
0.3
0.3
0.2
0.2
3.2
0.2
0.3
0.3
0.1
0.2
3.2
0.2
0.2
0.3
0.1
0.1
0.2
0.2
0.2
0.2
3.1
0.1
0.1
0.2
0.2
3.2
0.2
0.1
0.1
0.1
3.2
3.2
0.2
3.1
0.1
0.1
0.2
3.2
3.1
0.1
3.1
0.1
0.2
0.2
0.1
0.1
0.1
0.1
0.1
0.2
0.0
0.
3.1
0.1
0.1
0.0
4.50
4.75
3.20
5.25
5.30
2.75
0.4
0.4
0.3
0.5
3.5
3.5
0.4
0.4
3.4
0.5
0.3
0.5
0.4
3.4
0.4
0.4
0.3
0.3
0.3
0.3
0.4
0.4
0.4
0.5
0.3
0.3
0.3
3•4
3.4
0.4
0.3
0.3
3.3
3.3
0.4
0.4
0.2
0.3
0.0
0.3
0.3
0.4
0.2
0.2
0.3
0.3
0.3
0.3
0.2
0.2
0.3
3.3
0.3
0.3
0.2
0.2
0.3
0.3
0.3
3.3
0.2
0.2
0.2
2.2
0.3
0.3
0.2
0.2
0.2
0.2
3.3
0.3
1.30
5.25
5.50
5.75
1.00
1.25
0.5
0.5
0.7
0.7
5.7
0.5
0.5
3.5
0.3
0.7
0.1
0.7
3.3
0.5
0.1
5.6
0.7
0.7
0.3
0.5
0.5
0.1
0.6
0.7
0.3
0.5
0.5
0.5
0.5
0.6
0.4
0.4
0.5
0.3
0.3
0.5
3.4
0.4
3.4
0.3
3.5
0.3
0.4
3.4
0.4
.4
0.3
0.5
0.4
0.4
0.’
O•4
‘3.3
0.5
0.3
0.4
0.4
0.4
0.4
0.5
0.3
3.3
3.4
0.4
0.4
1.5
0.3
0.3
0.4
3.4
0.4
0.4
1.50
7.75
3.30
3.25
5.50
3.75
3.6 0.3
0.1 0.6
0.3 0.9
0.6 0.9
0.0 0.3
1.3 1.0
0.7
0.3
0.3
0.3
0.6
0.3
0.7
0.7
5.3
0.3
0.6
0.3
0.7
0.7
0.7
0.3
0.3
0.0
0.6
0.6
0.7
0.7
0.5
0.1
0.5
0.5
0.5
0.7
0.7
0.7
0.5
0.5
0.5
0.5
0.7
0.?
0.3
0.5
3.1
0.5
0.7
0.7
0.5
0.5
0.5
3.5
0.5
0.7
3.5
0.5
0.3
0.5
0.6
0.5
0.4
0.3
0.3
0.5
0.5
0.5
9.00
9.25
S0
75
10.30
10.25
1.0
1.1
.1
.2
1.2
3.3
1.0
1.0
1.1
1.1
1.2
1.2
0.1
1.0
1.0
1.1
1.1
1.2
0.9
0.9
0.0
0.3
1.1
1.1
0.9
0.9
0.3
1.0
1.0
1.1
0.1
0.3
0.3
0.9
1.0
1.3
0.3
0.3
0.9
0.3
0.3
.0
0.7
0.0
0.3
0.3
0.9
0.9
0.7
0.!
0.3
0.0
0.3
0.9
0.7
0.7
0.3
0.5
‘3.5
0.3
0.7
0.7
0.7
0.3
0.3
0.3
0.5
0.7
0.7
0.7
3.3
0.1
10.50
10.75
11.30
01.25
lt.30
11.75
0.3
1.3
1.4
1.4
1.4
1.5
1.2
1.3
1.3
1.3
1.4
1.4
1.2
0.2
0.3
1.3
1.3
1.4
1.2
1.2
1.3
1.3
1.3
1.4
1.3
1.2
1.3
1.2
1.3
1.3
1.0
1.1
1.1
1.2
1.2
0.3
1.0
2.1
1.1
1.1
1.2
1.2
1.3
1.3
1.0
2.1
1.1
1.1
1.0
1.0
1.3
1.1
1.1
1.1
0.3
0.3
1.0
1.0
1.0
1.1
0.5
0.3
1.0
1.0
0.0
0.1
0.5
0.3
0.9
0.3
1.o
0.0
12.00
12.25
12.50
12.75
*3.30
1.3
1.1
1.5
1.7
1.7
1.3
1.3
1.1
1.5
1.7
1.4
1.5
1.5
1.1
1.1
1.4
1.3
1.3
1.3
1.1
2.4
2.4
1.5
1.5
1.5
1.3
0.4
0.4
1.3
1.5
1.3
1.3
1.4
1.4
1.5
2.2
1.2
1.3
1.4
1.5
1.2
1.2
1.3
1.3
1.3
1.1
1.2
1.2
1.3
0.3
1.1
1.2
1.2
1.2
1.3
1.3
1.1
0.2
1.2
1.2
386
-------
‘ LBLZ A—2 (C ntixiuad)
53.0 57.5 50.0 52.5 53.3 57.5 10.3 73.5 75.0 77.3 50.3 52.5
13.25 1.3 1.7 1.7 1.3 1.5 1.3 1.5 1.5 1.4 1.4 1.3 1.3
13.30 1.3 1.3 1.7 5.7 1.5 1.3 1.5 1.3 1.4 1.4 1.4 1.3
13.75 1.3 1.3 1.3 1.7 1.7 1.1 1.3 1.3 1.3 1.3 1.4 1.4
14.30 1.3 1.3 1.3 1.7 1.7 1.7 1.5 1.3 1.5 1.3 1.3 1.4
14.25 2.0 1.3 1,3 1.3 1.7 1.1 1.1 1.7 1.5 1.3 1.3 1.3
14.34 2.3 1.3 1.3 1.3 1.3 1.3 1.1 1.7 1.3 1.3 1.5 1.3
14.75 2.1 2.3 2.0 1.1 1.5 1.3 5.3 I.? 1.7 1.5 1.3 5.5
15.33 3.1 2.3 3.3 1.9 1,3 1.3 1.3 1.3 1.7 1 .7 1.7 1.5
15.25 2.2 2.1 2.1 2.0 1.3 1.9 1.3 1.5 1.3 1.7 1.7 1.7
15.33 2.2 2.1 2.1 2.0 3.3 1.3 1.3 1.3 1.3 1.3 1.5 1.7
15.75 2.3 2.2 2.2 2..! 2.2 2.3 3.0 1.3 1.9 1.3 1.3 1.3
15.00 3.3 2.2 2.2 2.1 2.! 2.0 3.0 2.0 2.3 1.3 1.3 1.3
11.25 2.3 2.2 2.2 2.1 2.1 2.0 3.0 2.3 2.3 1.3 1.3 1.3
11.30 3.4 2,3 2.3 2.2 2.1 2.1 2.1 2.3 2.3 1.3 1.3 1.3
15.75 2.4 2.3 2.3 2.2 2.2 3.! 2.1 2.1 2.1 2.0 2.3 2.3
11.00 2.3 2.4 2.4 2.3 3.3 2.2 2.2 3.1 2.1 2.0 2.0 2.3
11.25 2.3 2.5 2.4 2.4 2.3 2.3 2.2 2.2 2.2 2.1 2.1 3.1
17.50 2.5 2.5 2.3 2.4 2.4 2.3 2.3 2.3 3.2 2.2 2.2 2.1
17.75 2.3 2.3 3.3 2.5 2.4 2.4 2.4 2.3 2.3 2.3 2.2 2.2
13.00 3.7 3.5 2.3 2.5 2.3 2.4 3.4 2.4 3.3 2.3 3.3 3.3
15.23 2.7 2.1 2.3 2.5 2.3 2.3 2.5 2.4 2.4 2.4 2.4 2.4
1.30 2.3 2..? 2.7 2.5 2.5 2.5 2.3 2.3 2.4 2.4 2.4 3.4
15.75 2.3 2.3 3.7 2.7 2.3 2. 2.3 2.5 2.3 2.5 2.3 2.3
13.00 2.3 2.3 2.3 2.7 3.7 2.5 3.5 2.3 2.3 2.5 2.3 2.3
13.25 2.3 2.3 2.3 2.3 2.? 2.7 2..? 2.3 2.3 2.3 2.5 2.3
13.30 3.0 2.9 2.3 2.3 2.3 3.7 2.7 2.7 2.7 2.1 2.5 2.5
13.73 3.0 3.3 2.3 2.9 2.3 2.3 3.3 3.7 3.7 3.7 2.7 2.7
20.00 3.1 2.0 3.0 3.0 3.3 3.3 2.3 3.3 2.3 2.3 2.3 2.3
20.23 3.2 3.3 3.1 3.0 3.0 3.3 2.3 2.3 2.3 2.3 2.3 2.3
24.30 3.2 3,2 3.2 3..! 3.1 3.0 3.3 3.0 2.3 2.3 2.3 2.3
20.75 3.3 3.2 3.2 3.1 3.1 3.1 3.3 3.0 2.3 3.3 3.3 3.3
21.00 3,3 3.3 3.3 3.2 3.2 3.1 3.1 3 ..1 .3 3.0 3.3 3.3
21.23 3.4 3.3 3.3 3.2 3.2 3.2 3.1 3.1 3.1 3.1 3,1 3.1
21.30 3.4 3.4 3.4 3.3 3.3 3.2 3.2 3.2 3.2 3.1 3.1 3.1
21.75 3.3 3.4 3.4 3.3 3.3 3.3 3.2 3.2 3.2 3.2 3.2 3.2
22.00 3.3 3.5 3,4 2.4 3.4 3.3 3.3 3.3 3.3 3.3 3.3 3.3
22.25 3.5 3.5 3.3 3.4 3.4 3.4 3.3 3.3 3.3 3.3 3.3 2.
22.10 3.3 3.5 3.3 3.5 3.5 3.5 3.4 3.4 3.4 3.4 3.4 3.4
22.75 3.7 3.7 3.3 3.5 3.3 3.3 3.5 3.3 3.5 3.5 3.5 3.5
23.00 3.? 3.7 3.7 3.5 3.1 3.5 3.5 3.5 3.5 3.3 3.5 3.3
23.25 3.3 3,3 3.3 3.7 3.7 3.7 3.1 3.7 3.7 3.7 3.1 3..?
23.30 3.5 3.3 3.3 3.1 3.3 3.3 3.3 3.7 3.7 3.7 3.7 3.7
23.15 3.1 3.3 3.3 3.3 3.3 3.3 3.5 3.3 3.3 1.3 3.3 3.3
24.04 3.9 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3
24.25 4.0 4.0 4.0 3.9 3.3 3.3 3.3 3.3 3.3 3.3 3.9 3.3
24.30 4.0 4.0 4.3 4.0 4.3 4.3 4.0 3.9 3.3 3.3 3.3 3.9
24.75 4.1 4.1 4.0 4.0 4.3 4 3 4.0 4.0 4.3 4.0 4.0 4.0
25.00 4.2 4.1 4.1 4.1 4.1 4.1 4.1 4.1 4.1 4.3 4.0 4.3
23.25 4.2 4.2 4.2 4.2 4.2 4.2 4.1 4.1 4.1 4.1 4.1 4.1
25.30 4.3 4.3 4.3 4.3 4.3 4.2 4.2 4.2 4.2 4.2 4.2 4.2
23.75 4.3 4.3 4.3 4.3 4.3 4.3 4.3 4.3 4.3 4.3 4.2 4.2
25.00 4.4 4.4 4•4 4.4 4.4 4.4 4 4 4 4 4•4 4,4 4.3 4.3
21.25 4.5 4.5 4.3 4.5 4.5 4.3 4.3 4.5 4.5 4.5 4.4 4.4
25.30 4.3 4.3 4.5 4.5 4.5 4.3 4.5 4.5 4.3 4.3 4.3 4.5
r !rticrom
387
-------
TABLE A-2 (C 1tino2ed)
7,:
3 0.0 0.0 0.0 0.0
.25 0.0 3.0 0.3 0.0
.30 0.0 0.0 0.0 0.3
.75 0.0 0.0 0.0 0.3
1.00 0.0 0.0 0.0 0.3
1.23 0.0 0.3 0.0 0.3
1.50 0.3 0.3 0.0 0.3
5.75 0.0 0.0 0.0 0.3
2.00 0.0 0.3 0.0 0.0
2.2.3 3.3 0.0 0.0 0.0
2.50 0.1 0.1 0.0 0.0
2.75 0.1 0.1 0.1 3.0
3.00 0.1 0.1 0.1 0.1
3.23 3.1 3.1 0.1 0.1
3.50 0.1 0.1 0.1 3.1
2.71 3.1 0.1 0.1 0.1
4.00 0.1 0.1 0.1 0.1
4.2.5 0.1 0.1 0.1 0.1
4.30 0.2 0.1 0.1 0.1
4.75 0.2 0.2 0.1 0.1
0.00 0.2 0.2 0.2 0.1
3.25 0.2 0.2 0.2 3.2
1.50 0.2 0.2. 0.2 0.2
5.75 0.3 0.2 0.2 0.2
3.00 0.3 0.3 0.2 0.2
3.25 0.3 0.3 0.2 0.2
3.50 0.3 0.3 0.3 0..1
3.75 0.3 0.2 0.3 0.3
1.00 0.4 0.3 0.3 0.3
7.25 0.4 0.4 0.3 0.3
7.50 0.4 0.4 0.4 0.4
7.75 0.5 0.4 0.4 0.4
5.00 .0.5 0.3 0.4 0.4
3.25 0.3 0.3 0.5 0.4
3.00 0.6 0.5 0.5 3.3
3.75 0.5 0.5 0.5 0.5
5.00 0.1 0.5 0.5 0.5
1.25 0.5 0.5 0.1 0.3
3.50 0.7 0.6 0.5 0.5
3.75 0.7 0.7 3.6
10.00 0.7 0.7 0.7 0.1
10.25 0.3 0.7 0.7 0.7
10.30 0.8 0.3 0.1 0.7
10.75 0.3 0.3 0.3 0.3
11.00 0.9 0.5 0.6 0.8
11.25 0.9 0.1 0.6 0.8
11 ,50 0.3 0.3 0.5 0.1
11.15 1.0 0.1 0.1 0.5
12.30 1.0 1.0 0.1 0.1
12.25 1. 1.3 1.0 1.0
12.30 1.1 LI 1.0 2.0
12.75 1.2 1.1 1.2 1.1
13.00 2.2 1.2 1.1 1.2
0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0
0.0 0.0 0.3 0.0
0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0
0.0 0.0 0.3 0.0
0.3 0.0 0.0 0.0
0.0 0.3 0.0 0.3
0.3 0.0 0.0 0.0
0.3 0.3 0.0 0.3
0.0 0.0 0.0 0.3
0.3 0.0 0.0 0.0
0.3 0.0 0.0 0.0
5.3 0.0 3.0 2.0
3.3 0.3 5.0 3.3
0.0 0.0 0.3 0.0
0.3 0.0 0.3 0.0
0.0 0.0 0.3 0.0
0.1 0.1 3.1 0.1
0.1 0.1 0.2 0.2
0.1 3.1 0.1 0.1
2.1 0.1 0.1 0.1
0.1 3.2 0.1 0.1
0.1 3.1 0.1 0.1
0.1 3.1 3.1 3.1 0.1
3.1 3.1 0.1 0.1 0.1
0.1 0.1 0.1 0.1 0.7
0.2 0.2 0.2 .0.2 0.1
0.2 0.2 2.2 3.2 0.2
0.2 0.2 3.2 0.2 0.2
0.3 0.2 0.2 0.2 0.2
0.3 0.3 0.2 0.2 0.2
0.3 0.3 3.3 0.3 0.2
0.3 0.3 0.3 0.3 0.3
0.0 0.3 0.3 0.3 0.3
0.4 0.4 0.3 0.3 0.3
0.4 0.4 0.4 0.3 . 5.3
0.4 0.4 0.4 0.4 0.3
0.4 3.4 0.4 0.4 5.4
3.5 0.5 0.4 0.4 0.4
3 0.3 0.4 0.4 3.4
0.3 0.5 0.5 0.5 0.4
0.5 3.5 0.3 0.3 0.5
0.8 0.1 0.3 0.5 0.5
0.1 0.5 0.5 0.5 0.5
0,1 0.6 0.1 0.5 0.3
Q.7 3.7 0.1 0.5 0.1
0.7 0.7 0.? 0.1 3.1
3.7 0.7 , 5.1 0.7 0.5
3.1 0.3 0.8 0.7 0.1
0.6 0.3 0.3 0.7 0.1
0.9 0.3 0.9 0.8 0.7
0.1 0.1 0.3 0.! 0.5
roduced rorn
I
85.3 87.3 00.0 22.5 95.3 97.5 100.3 102.5 105.0 07.5 110.0 112.5
0.3 0.0 0.0
0.0 0.3 0.3
C.0 0.0 0.3
0.3 0.3 0.0
0.0 0.0 0.0
0.3 3.0 0.0
0.0 0.0 0.0
0.5 3•3 0.0
0.0 3.0 0.3
0.3 0.3 0.0
0.0 3.0 0.0
0.3 0.0 0.0
0.0
5.0
0.0
3.0
0.0
0.0
0.1
3.1
0.2
0.1
0.1
0.1
0.3
0.3
0.3
0.3
0.0
0.0
0.3
0.0
0.0
0.0
0.3
0.0
3.0 0.3 0.0
0.1 0.2 0.0
0.1 0.2 0.1
0.1 0.1 0.1
3.1 0.1 0.1
0.1 0.1 0,1
2.1 3.1 0.1
0.1 3.1 3.1
0.1 0.1 0.2
0.2 0.1 3.1
0.2 0.2 0.1
3.2 0.2 2.2
3.2 0.2 0.2
3.2 0.2 0.2
3.2 0.2 0.2
0.3 0.3 0.2
3.3 0.0 0.2
0.3 3.3 0.3
0.3 0.3 0.3
0.0 0.3 0.3
0.4 0.3 0.3
0.4 3.4 0.4
3.4 0.4 0.4
0.4 0.4 0.4
0.5 0.4 0.4
0.5 2.5 0.5
0.5 3.5 0.1
:.s 0.5 0.5
0.6 0.1 0.5
0.1 0.5 0..1
0. l
0.5
0.7
0.7
0.?
0.1
0.1
0.3
0.3
0.3
1.3
0.1 0.7
0.7 0.7
0.? 0.7
3.5 0.1
0.3 0.3
0.5 0.3
0.9 0.8
0.1 0.3
0.3 0.9
1.3 1.3
1.0 1.0
388
-------
TA3LZ .‘.-2 (Cti u d)
I
53.0 T.5 50.0 92.3 35.3 37.5 100.3 502.5 105,3 107.5 110.0 112.5
33.25 1.2 1.2 1.2 1.1 1.1 1.0 5.0 1.0 0.3 5.9 0.3 3.3
13.50 1.3 1.2 1 1.2 1.1 1.1 1.3 1.3 1.0 0.3 0.3 0.3
13.73 1.3 1.3 1.2 1.2 1.3 1.1 1.1 1.1 1.0 1.3 0.3 3.9
14.00 1.4 1.1 1.3 1.2 3.2 1.1 1.1 1.1 1.0 1.3 1.3 ‘3.3
14.25 1.4 1.3 1.3 1.3 1.2 1.2 1.2 1.1 1.1 1.3 1.0 1.3
14.30 1.3 1.4 3.4 3.3 1.3 1.2 1.2 1.2 1.1 1.1 1.3 1.3
14.75 1.5 1.4 1.4 1.3 1.3 3.3 1.3 1.2 5.1 5.1 1.1 1.1
15.00 1.5 1,3 1.3 1,4 1,4 1,3 1.3 1.3 3.3 1,3 3,1 3,1
15.25 1.3 1.5 3.5 1.5 1.3 1.4 2.4 1.3 1.2 1.2 3.2 1.2
15.53 1.7 1.3 1.5 1.1 1.3 1.3 3.4 1.4 1.3 1.3 1.3 1.3
35.75 1.7 2.7 1.7 1.1 1.3 1.5 1.5 1,4 1.3 1.3 3.3 1.3
15.00 1.3 1.7 3 ,1 1.7 1.1 1.5 1.5 1.3 1.4 1.4 1.4 1.3
15.25 1.3 1.3 1.3 1.7 1.7 3,3 1.5 1.5 l 4 1,4 3,4 1.4
15.30 1.3 3.3 3.4 1.7 1.7 1.7 1.5 1.9 1.5 1,3 1.3 1.4
55.75 1,3 1.3 1.3 2.4 1.5 1.7 1.7 1.5 1.5 3.3 1.3 1.5
37.30 2.0 2.3 1.3 1.3 1.3 1.3 1.7 1.7 I .? 1,5 1.3 1.5
11.25 2.1 2.3 2.0 2.9 1.3 3.3 1.3 5.7 5.7 1.5 1.5 1.5
11.50 2.1 2.3 2.0 2.0 1.3 3.3 3.3 1.3 1.7 1.7 1,7 1,3
11.75 2.3 2.1 2..1 2.3 2.0 1,3 2.9 2.3 1.3 3,3 1.3 1,1
13.03 2.3 2.2 2.1 2.1 2.3 2.3 2.0 1.9 1.3 3.3 1.3 1.3
15.25 2.3 2.2 2.2 2.2 2.2 5.3 2.1 2.3 3.3 1,3 1.9 1.3
13.30 2.4 2.3 2.2 2.2 2.2 2.2 2.2 2.3 2.0 2.3 1.3 1.9
15.75 2,4 2.3 2.3 2.3 3.3 3.2 2.2 2.1 2.3 2.3 2.3 2.0
13.30 2.3 2,4 2.4 2.3 2.3 2.3 5.3 2.1 2.1 2.3 2.3 3.3
11.25 2.3 2.4 2.4 2.4 2.4 2.3 2.3 2.2 2.1 2.1 2.3 2.1
19.30 2.5 2.5 2.5 2.4 2.4 2.4 5.4 2.2 2.2 2.2 2.2 2.1
11.75 2.5 2.3 2.5 2.5 2.5 2.4 2.4 2.3 2.2 2.2 2.2 2.2
20.00 2.7 2.3 2.5 2.4 2.3 2.5 2.3 2.4 5.3 2.3 2.3 2.3
20.25 2.3 2.7 2.7 2.5 2.5 2.3 5.3 2.5 2.4 2.4 2.4 2.4
20.30 2.3 2.3 2.3 2.7 2.7 2.7 2.5 2.3 2.5 2.3 2.5 2.3
20.75 2.1 2.8 2.3 2.5 2.3 5.7 2.7 2.1 2.5 2.5 5.1 2.1
21.03 3.3 2.3 2.3 2.3 2.3 2.3 2.4 2.7 2.1 2.1 2.7 2.5
21.25 3.3 2.9 2.9 2.3 2.9 2.3 2.5 2.3 2..? 2.7 2.7 2..?
21.30 3.1 3.3 3.0 2.3 2.3 2.9 2.3 2.3 2.3 2.4 2.3 2.7
21.75 3.2 3.3 3.1 3.0 3,0 3.3 3.3 2.3 2.9 2.3 2.3 5.3
22.00 3.2 3.1 3.1 3.3 3.5 3.1 3.1 3.3 3.0 3.3 3.3 2.3
22.23 3.3 3.2 3.2 3.2 3.2 3.2 3.2 5.1 3.3 3.0 3.0 3.0
22.30 3.4 3.3 3.3 3.2 3.2 3.2 3.2 3.1 3.1 5.1 3.1 3,1
22.75 3.5 3.4 3.4 3.3 3.3 3.3 3.3 3.2 3.2 3.2 3.2 3.2
23.03 3.5 3.5 3.5 2.4 3.4 3.4 3.4 3.3 3.3 3.3 3.3 3.3
23.25 3.1 3.3 3.5 2.3 3.5 3.3 3.3 3.4 5.3 3.3 3.3 3.3
23.30 3.? 3.4 3.5 3.5 3.5 3.5 3.5 3.3 3.4 3.4 3.4 3.4
23.75 3.7 3.7 3.1 3.5 3.5 3.3 3.5 3.3 3.3 3.3 3.3 3.4
24.00 3..1 3..? 3.1 3.7 3.7 3..? 3.1 3.1 3.3 3.3 3.5 3.3
24.25 3.1 3.3 3.4 3,1 3.3 3..! 3.7 3.7 3.4 3.1 3.1 3.5
24.50 3.3 3.4 3.3 3.3 3.3 3.5 3.3 3.0 3 ..? 3.7 3.? 3.7
24.75 4.0 3.3 3.3 3.1 3.3 3.3 3.9 3.3 3.3 3.8 3.3 3.1
25.00 4.0 4.0 4.3 4.3 4.3 4.0 4.3 4.3 3.3 3.3 3.3 3.3
25.25 4.1 4.1 4.1 4.1 4.1 4. 4.1 4.3 4.0 4.3 4.0 4.3
23.30 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.2 4.1 4.1 4.1 4.1
25.15 4.3 4.3 4.3 4.3 4.3 4.3 4.3 4.3 4.2 4.2 4.2 4.2
25.03 4.4 4.4 4.4 4.4 4.4 4.4 4.4 4.4 4.3 4,3 4.3 4.3
25.25 4.5 4.3 4.3 4.5 4.5 4.5 4.3 4.3 4.4 4.4 4.4 4.4
25.5.0 4.5 4.3 4.3 4.3 4.3 4.3 4.3 4.5 4.5 4.5 4.5 4.3
L roducrd from
389
-------
TA3LZ. A- 2 (C ntirn2ea)
I
115.0 117.3 330.0 122.5 125.3 127.5 330.3 122.5 125.3 137.5 140.3
——
0 0.3 5.3 0.0 0.3 3.3 0.0 3.0 0.0 0.0 0.0 0.0
.:s 0.3 0.3 0.0 0.4 0.3 0.0 0.0 3.3 0.0 3.3 5.0
.30 0.5 5.0 0.3 5.0 0.3 0.0 0.3 0.0 5.0 3.0 0.0
.:s 0.0 0.3 o.o 0.0 0.0 0.0 5.0 0.3 3.0 0.0 0.3
1.00 0.0 0.0 5.0 0.0 0.3 0.0 0.0 3.0 5.0 0.3 0.3
1.25 3.0 0.0 0.3 3.0 0.0 0.0 5.5 3.4 0.0 0.0 0.0
3.50 0.3 0.3 0.3 0.3 0.3 0.0 5.4 0.3 0.0 0.0 0.4
1.75 0.3 0.0 0.0 0.0 0.0 0.0 0.3 0.0 5.0 0.0 0.3
5.00 0.3 0.0 3.0 5.0 0.0 0.0 0.0 0.0 0.0 5.3 0.0
2.25 0.3 3.0 0.3 0.0 3.0 0.3 5.3 0.0 0.0 0.2 0.0
2.50 3.0 5.0 5.0 0.3 0.3 3.3 5.0 0.0 0.3 0.3 0.0
3.75 0.0 0.0 0.3 5.3 0.0 0.0 0.0 0.3 5.0 5.0 0.0
3.20 3.3 0.0 5.0 0.0 0.0 0.3 0.0 0.0 0.0 3.3 5.0
2.25 0.3 0.0 0.3 5.3 0.3 0.0 0.3 5.0 5.0 3.0 0.9
3.30 0.0 0.0 0.3 5.3 0.0 0.0 0.0 0.3 0.3 3.3 5.4
3.75 0.0 0.0 3.0 0.2 0.0 0.0 0.0 0.0 0.0 0.3 .3
4.00 0.0 0.0 0.0 0.0 0.0 0.0 0.3 0.0 0.0 0.0 5.0
4.25 0.3 0.0 5.2 0.3 0.3 0.0 3.0 0.0 3.0 5.0 0.0
4.30 0.4 0.0 5.0 0.3 0.0 0.3 0.0 0.0 0.3 3.3 0.3
4.75 0.1 0.0 5.3 0.0 0.0 0.0 5.3 5.0 3.9 3.0 0.3
5.00 0.1 0.1 0.0 3.3 0.0 0.3 3.0 0.0 0.3 0.0 0.3
5.2.5 0.1 0.1 5.0 0.0 0.0 0.0 3.3 0.0 0.0 3.0 0.3
5.50 0.1 0.1 0.1 0.1 0.1 0.3 0.0 5.0 0.0 0.3 0.0
5.75 0.1 0.1 3.1 0.1 0.1 2.1 0.3 0.3 0.0 0.0 3.0
5.00 0.2 3.1 0.1 0.1 0.1 0.3 0.1 0.0 3.0 0.3 0.0
5.2.5 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.0 0.3 0.0
5.50 0.3 0.1 0.1 3.1 0.1 0.2 3.1 0.1 0.0 0.0 0.0
5.75 0.1 0.1 0.1 0.1 0.0 3.1 3 ,1 0.1 0.1 0.1 0.0
7.00 0.1 0.1 3.1 0.1 3.1 0.1 0.1 0.1 0.1 0.1 5.0
7.25 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
1.30 0.2 0.1 0.1 0.1 0.0 0.1 0.1 0.1 0.1 0.1 0.1
7.75 0.2 0.2 0.2 0.1 0.1 0.1 0.1 5.1 0.1 3.1 0.1
5.00 0.2 0.2 0.2 0.2 0.1 0.3 0.1 0.1 0.1 0.1 0.1
8.25 0.2 0.2 0.2 0.2 0.2 0.2 0.1 0.1 0,1 3.1 0.1
5.50 0.3 3.2 0.2 0.2 0.3 0.2 0.2 3.1 0.1 0.1 0.1
5.75 0.3 3.2 0.2 0.2 0.2 0.2 0.2 0.2 0.1 0.1 0.1
3.30 0.3 0.3 0.2 3.2 0.2 0.2 0.2 0.2 0.1 0.1 0.1
3.25 0.3 0.3 3.3 5.2 0.2 0.2 0.2 0.2 0.1 0.1 0.1
3.50 0.3 0.3 0.3 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.1
3.75 0.4 0.3 0.3 0.3 0.3 0.2 0.2 3.2 0.2 0.2 0.2
10.30 0.4 0.3 0.3 0.3 0.3 0.3 0.2 0.2 0.2 0.2 0.2
10.25 0.4 0.4 0.3 0.3 0.3 0.3 0.3 0.2 0.2 0.2 0.2
10.50 0.4 0.4 0.4 0.3 0.3 0.3 0.3 0.3 0.2 3.2 O.
10,75 0.5 3.4 0.4 0.4 0.4 3.3 0.3 0.3 0,2 0.2 3.2
11.00 0.5 0.5 3.4 0.4 0.4 0.4 0.3 0.3 0.3 0.2 0.2
11.23 0.3 0.5 0.4 0.4 0.4 3,4 0.4 3.3 0.3 0.3 0.3
11.50 0.5 0.3 0.5 0.4 3 ,4 0.4 0.4 5.3 0.3 3.3 0.3
11.75 0.5 0.5 0.5 0.4 0.4 0.4 0.4 0.4 0.3 0.3 0.3
12.30 0.5 0.5 0.3 0.5 0.5 0.4 0.4 0.4 0.3 0.3 0.3
12.25 0.5 3.6 3.6 0.5 0.5 0.5 0.3 0.4 0.4 0.3 0.3
12.00 0.7 0.6 0.6 3.5 0.5 0.3 0.3 0.5 0.4 0.4 0.4
12.75 0.1 0.6 0.6 0.5 0.1 0.6 0.3 0.3 0.4 0.4 0.4
13.00 0.7 0.7 0.7 0.6 0.5 0.0 0.1 0.5 0.5 6.4 0.4
TReproduced from
390 I _ best available copy.
-------
TA3LZ A-2 (C r .ciuded)
I ‘0
13.25
13.30
13.75
14.00
14.23
14.30
14.15
13.00
13.25
15.50
15.75
16.23
16.50
16.73
17.00
17.23
11.50
17.75
11.30
15.25
15.30
16.75
16.00
19.25
16.3*
15.15
20.30
20.23
20.50
20.73
21.00
21.25
21.50
21.73
22.00
22.23
22.30
22.75
23.00
23.25
23.50
23.73
24.03
24.23
24.50
24.73
23.03
0.7 3.1 0.7 0.3
0.3 0.3 0.7 0.1
0.3 0.3 0.3 0.1
0.3 0.3 0.1 0.7
6.3 0.3 0.6 3 ,3
0.3 0.3 0.3 0.3
1.3 1.3 0.1 0.3
1.3 1.0 1.0 0.3
5.7 1.1 1.3 1.3
1.2 1.1 1.1 1.3
1.2 1.5 1.1 1.1
1.3 1.2 1.2 5.1
1.3 1.2 1.2 1.2
1.4 1.3 1.2 1.2
1 ,4 I•3 1.3 1.2
1.3 1.4 5.3 1.3
1.4 1.5 1.4 1.3
1.6 1.3 5.3 1.4
1.1 1.6 1.5 1.3
1.3 1.7 1.6 1.3
1.3 1.7 1.3 1.3
1.9 1.3 1.7 1.5
1.3 1.6 1.3 1.7
1.9 5.5 1.3 1.7
2.3 LI 1.3 1.5
2.1 2.3 1.3 1.3
2.2 2.0 2.0 1.3
2.2 2.1 2.1 2.3
2.3 2.2 2.2 2.1
2.4 2.3 2.3 2.2
3.3 2.4 2.3 2.3
2.5 3.3 2.4 2.4
2.1 2.5 2.3 2.4
2.7 2.6 2.3 2.3
2.8 2.7 2.5 2.1
2.3 2.3 2.7 2.1
3.0 2.3 2.3 2..?
3.1 3.0 3.8 2.3
3.2 3.1 3.3 2.3
3.3 3.2 3.1 3.0
3.3 3.3 3.2 3.1
3. 4 3.4 3.3 3.2
3.3 3.3 3.4 3.3
3.3 3.5 3.4 3.3
3.3 3.5 3.3 3.4
3.? 3.7 3.3 3.3
2.3 3.. ! 3.? 3.9
3.3 3.1 3.1 3.3
0.6 0.5
0.5 3.5
3.7 3.3
0.7 0.7
0.7 0.1
3.3 0.7
0 . 5 0.1
3.3 0.3
0.3 0.3
5.3 3.3
5.0 1.3
1.1 1.3
1.1 1.1 1.1
1.2 1.1 1.1
1.2 1.2 1.2
1.3 1.3 1.2
1.3 1.3 1.3
1.4 1.4 1.3
5.5 1.4 5.4
LI 1.5 1.3
1.3 1.5 1.5
1.5 1.1 1.5
1.7 1.? 1.5
1.7 1.7 1.7
1.3 1.3 1.7
1.3 1.3 L3
1.3 5.3 1.3
2.0 2.3 2.0
2.1 2.1 3.1
2.2 2.2 2.2
2.3 2.3 3.3
2.3 2.3 2.3
2.4 2.4 2.4
2.4 2.4 2.4
2.5 2.3 2.3
2.5 2.3 2.6
2.7 2.7 2.7
2.! 2.3 2.3
2.8 2.3 2.9
3.0 3.0 3.0
3.1 3.1 3.1
3.2 5.2 3.2
3.2 3.2 3.2
3.3 3.3 1.3
3.4 3.4 3.4
3.5 3.5 3.5
3.3 3.3 3.5
3.3 3 ,.! 3.3
0.5 5.3 0.3
5.5 0.3 3.5
0.5 0.5 0.3
3.5 0..1 0.5
0.1 0.3 0.3
5.7 0.1 0.5
0.7 0.7 0.5
0.3 0.? 0.7
0.3 3.5 0.7
0.3 3.3 0.3
5.3 5.3 0.3
.3 0.3 0.3
5.0 3.3
1.3 0.3
5.3 .0
1.1 1.5
1.3 1.1
5.2 1.2
1.3 1.2
1.4 1.3
1.4 1.4
7.3 5.4
1.3 1.3
1.5 1.5
5.7 1.5
1.7 1.7
1.3 L3
1.9 1.3
3.0 2.3
2.1 2.1
2.2 2.1
2.2 2.2
2.3 2.3
2.3 2.3
3.4 2.4
2.5 2.5
0.5
3.5
0.5
0,5
0.5
0.$
0.5
0.7
0.1
0.3
3.3
5.3
3.3
0.3
1.3
1.0
t . 1
1.2
1.2
1.3
1.4
1.4
1.5
1.3
3.5
3.5
3 .7
1.5
1.3
2.3
2.1
2.2
3.3
2.3
2.4
2.5
2.5
2.7
2.3
2.3
3.3
3.1
3.1
3.2
3.5
3.4
3.3
3.7
I Reproduced from
[ test available copy .
A
113.3 117.5 120.3 122.3 125.2 527.5 120.3 132.5 135.3 137.5 140.2
0.5
3.,
0.1
0.7
0.3
0.3
0.3
0.3
0.9
1.3
1.1
1.1
‘.3
7.’
I.’
5.2
1.2
5.3
1.4
‘.3
1.3
3 .3
‘.5
‘.7
3.1
3.3
1.3
3.3
2.1
2.2
2.2
2.3
2.3
2.4
2.3
2.3
2.?
2.!
2.3
3.0
3.5
3.2
2.5
2.7
2.3
2.9
3.0
3.1
3.2
3.3
3.4
2.5
3.6
2.5
23.25
25.30
22.15
26.00
26.25
29.30
2.7
2.3
2.3
1.0
2.1
3.2
3.2 3.2
3.3 3.3
3.4 3.4
3.5 3.5
3.3 3.5
3.3 3.3
3.5 3.9
4.3 4.0
4.2 4.2
4.3 4.3
4.4 4.4
4.3 4.3
4.0 4.0 3..1 3.3
4.1 4.1 4.0 4.0
4.2 4.2 4.2 4.2
4.3 4.3 4.3 4.3
4.4 4.4 4.4 4.4
4.5 ,.5 4.3 4.5
3.5 3.3 3.9
4.0 4.0 4.0
4.2 4.2 4.2
4.3 4.3 4.3
4.4 4.4 4.4
4.5 4 5 4.3
3.9 3 ,.!
4•3 3.
4.2 4.1
4.3 4.2
4.4 4.3
4.3 4.4
391
-------
TABLE A-3
VAUJES OF IJNAOJUSTEO OML’f P TENTIA(. 5VAPO1 M P R .TI0N
FOR A1 T £ ATUR53 A9OV 26.5°C
U AaJu3TLD POTCITIAL ZYA?OTRAN3PI*ATsON IN
T0 0.0 0.1 0.2
0.3 0.4 0e5
0.5 0.7 0.8
0.9
26
27’ 4.6 4.7 4.7
28 4.9 5.0 5.0
29 5.2 5.2 5.2
30 5.4 5,4 5.4
31 5.6 5.6 5.6
32 5.3 5.8 5.8
33 5.9 5.9 5.9
34 5.0 5.0 5.0
35 6.1 6.1 6.1
4.5
4.7 4.8 4.8
5.0 5.0 5.1
5.2 5.3 5.3
5.5 5.5 5.5
5.5 5.7 5.7
•4 •W
5.9 5.0 6.0
6.0 6.1 6.1
5.1 6.1 6.1
5.7 5.7 5.7
5.9 5.9 5.9
.0 6.0 6.0
5.1 8.1 5.1
6.1 6.1 6.1
5.8
5.9
6.0
6.1
6.1
36 5 .1 5.1 6.2
37 6.2 6.2 6.2
38 6.2
6.2 6.2 6.2
6.2 5.2 6.2
6.2 6.2 6.2
6.2 5.2 8.2
6.2
6.2
4.5
4.5
4.6
4.5
4.8
4.3
4.3
4.3
5.1
5.1
5.1
5.2
5.3
5.3
5.4
5.4
5.5
5.5
5.5
5.6
392
-------
TA3LZ A-4
1p.rt,crom.
rtA$ .2313%.L t3MTI4L.Y ULLnaN CF S MLiC IT r) T4CFN ts .at
C RPR S3 3 IM WIlT! C l 12 43WI3
j
4417M11 U n
3 31.2
1 31.2
2 31.2
3 30.1
4 30.3
riftS
29.2
29.2
25.2
25.2
21.3
31.2
31.2
31.2
30.1
30.3
30.3
30.3
20.3
30.3
30.5
31.2
31.2
31.3
31.5
31.3
20.3
30.3
30.5
30.5
30.3
31.2
31.2
31.2
31.5
31.3
31.2
31.2
31.2
31.2
31.3
20.3
30.3
33.3
30.3
33.3
31.2
31.2
31.2
31.2
33.3
30.3
30.3
33.0
30.0
30.3
31.2
21.2
10.9
30.9
30.5
9 30.3
I 23.5
1 30.3
5 30.3
9 20.0
27.3
27.3
21.5
27.5
27.3
30.3
30.3
30.3
30.3
30.3
20.5
23.5
30.3
33.9
33.3
ii • a
31.3
32.3
32.1
32.4
30.3
31.2
31.2
31.3
31.5
31 • 3
31.3
32.1
32.1
32.4
31.3
31.5
31.3
11.3
31.3
30.3
30.3
30.3
30.5
30.3
:3.3
30.3
30.3
33.9
30.5
23.
29.7
29.7
29.4
29.4
30.3
33.3
:0.3
30.3
10 23.3
I I 25.7
12 29.7
13 29.4
14 23.4
27.3
22.3
V.3
21.3
27.3
30.3
30.3
30.3
20.1
30.3
33.3
23.3
31.2
21 • 2
31.2
32.4
32.7
32..?
33.0
33.3
31.3
31.3
32.1
33.1
32.4
32.4
32..?
33.3
3.3
33.3
12.3
32.1
32.1
32.4
32.4
33.3
30.5
20.5
30.9
30.3
30.5
30.3
33.3
30.3
13.3
29.4
29.1
39.1
29.3
23.3
25.5
29.7
2?.’
29.4
29.1
29.1
13 29.1
13 23.1
11 25.3
II 25.3
1, 23.3
27.3
21.3
27.3
27.0
21.3
30.3
33.1
30.3
30.3
30.9
31.2
31.2
31.5
31.5
31.5
33.3
33.3
33.3
33.3
33.3
32.’
12.7
32.7
33.3
23.3
33.5
33.3
33.3
3 .3
34.2
32.4
32.7
22.7
33.0
33.0
10.3
30.5
30.3
30.3
30.1
30.3
10.3
30.3
30.0
30.0
23.3
19.2
25.2
27.3
27.1
33.3
29.3
19.5
23.3
29.2
20 23.3
21 29.2
22 23.2
22 21.3
24 27.3
27.3
27.3
29.?
23.?
25.7
10.3
30.3
30.3
20.3
20.3
31.3
31.5
31.3
31.1
31.5
33.3
33.3
34.2
34.2
34.3
33.3
33.3
33.3
33.9
34.2
34.2
34.3
34.3
34.3
34.5
33.3
33.3
33.3
33.3
33.3
33.3
30.5
30.3
10.5
30.3
30.0
30.0
23.7
23.7
25..?
23.3
27.3
2’.3
27..)
27.3
23.2
27.3
21.3
27.!
21.3
zS 27.9
25 21.3
2? 27.5
25 27.3
23 21.3
23.7
23.4
23.4
23.4
25.1
30.3
30.3
10.3
30.3
30.3
31.3
22.1
32.1
32.1
32.2
34.3
34.3
34.3
33.3
35.1
34.2
34.5
24.3
34•3
34.3
21.1
19.1
35.4
35.4
35.7
33.3
23.5
33.3
33.3
33.9
33.3
23.3
20.5
30.3
30.9
23.7
29.?
29.4
29.4
27.3
27.3
27.3
23.7
2!.?
27.3
27.3
27.3
23.7
25.4
3*3 27.3
31 27.3
32 23.7
33 39.4
34 23.4
23.1
29.1
23.3
25.1
23.3
30.3
30.3
20.3
30.9
30.3
32.4
32.4
32.4
32.7
32.?
35.4
25.4
35.7
33.?
33.3
35.1
35.1
33.4
35.7
33.0
35.3
39.0
3 3
33..)
33.3
34.2
34.2
34.5
34.5
34.5
33.3
30.3
30.3
33.3
30.3
29.4
29.4
23.1
29.1
33.4
29.4
25.1
35.1
21.3
23.4
23.1
23.3
25.3
23.3
._33...25.1
35 23.1
31 29.3
• 25.5
33 25.3
21.5
25.5
29.3
23.2
23.2
30.9
33.3
33.9
30.3
30.3
32.7
33.0
23.
33.0
33.3
33.3
36.3
9
. 3
35.3
39.3
36.6
3 3
37.2
31.2
39.9
37.2
37.3
32.5
31.3
34.3
34.3
33.1
25.1
35.4
30.3
30.3
30.3
31.2
31.2
23.1
29.1
29.5
23.3
23.3
23.3
25.2
23.2
24.9
29.2
4.9
24.3
24.5
24.3
40 29.2
41 24.3
42 24.3
43 24.3
44 24.3
24.9
34.3
24.5
24.5
24.3
30.3
33.3
30.3
33.5
30.5
33.3
33.3
33.5
33.3
33.3
37.2
37.3
37.5
31.3
39.1
37.3
37.3
35.1
33.4
35.?
39.1
33.1
39.4
35.1
39.3
33.4
35.7
33.1
35.0
39.0
31.2
31.2
31.2
31.2
31.2
23.3
29.3
23.3
29.5
29.3
24.3
24.5
24.3
24.0
23.7
24.0
23.7
23.1
22.3
22.5
45 24.0
43 23.7
41 23.1
45 22.5
49 22.5
24.3
24.0
24.0
23.7
23.7
30.5
30.3
30.3
30.3
30.5
33.3
33.9
34.2
34.2
34.5
24.3
33.4
39.7
33.0
39.3
33.5
39.3
35.7
39.0
39.5
39.3
40.2
40.F
33.3
33.6
39.3
40.2
40.5
41.1
33.3
33.3
33.3
35.9
37.2
21.3
31.2
31.2
1.9
31.5
31.3
31.3
23.2
23.2
27.9
27.9
27.6
21.5
23.1
23.4
23.1
22.3
22.3
22.2
21.9
21.3
21.3
21.3
393
-------
TABLE A—S
PRCVIS CflAL wATE ‘ CL3INt1 CAPACIT1 S w 1’H 3iFFZRE IT C0 1? ATIC S
CF SOIL ANO VECZTATICN
SLLOW OOrL0 C 0P$ (s s*c .
Fin 3 .IRG 100
FluE SANOY LOAM 1 50
Sat .? LOAM 200
CLAY I.OAM 250
CLAY 300
ROOF Zout
N 11
PEAS, StAllS, BEETS. A1ROT3 ,
1.2 .50 1.57
1.3 .50 1.67
2.4 .52 2.06
3.0 .40 1.33
.25 .33
(coui , corron, roe*cco, cs c t . a s
1.2 .75 2.50
l.a 1.00 3.33
2.4 1.00 3.33
3.3 .30 2.57
3.6 .50 1.87
50 2.0
75 3.3
125 5.0
100 4,0
75 3.0
Soil. TYPE AVAILAII.i ‘,IATER
P IN/N iu/rT
APPLICABLE SOIL
RLTEITI ON
i STUU
TABU
I I
1O BUA TILT
FINE SAND
FINE BANDY
Su.T I.OAIV
CLAY LOAM
CL.i V
I ?-1O0TtD *OPS
100
LOAN 150
200
250
300
CEcI’—400TED CROPS (ALFALfA. PASTURES, SHRUBS)
FINE 3AN3 100 1.2
FIRE SANDY t..oAri 150 1.3
SILr LOAN ZOO 2.4
CLAY LOAN 250 3.3
C2.AY 300 3.6
100 1.2
1 50 .1.3
200 2.4
250 3.0
300 3.6
C 0 )1* A OS
Fm 54)13
Fin SANDY LOAM
SILT LOAM
CLAY LOAM
CLAY
cLOSED MATURE FOREST
Fsit SAND
Fmu SANDY LOAN
SILT LOAM
CLaY LOAN
CLAY
75
I C
I—
200
200
50
100
150
250
250
200
150
250
300
250
200
250
300
400
400
350
1 • 00
1 00
1.25
I • 00
.6 ?
I • 50
1.67
1.50
1 • 00
.57
2.50
2.00
2.00
1 • $0
1.1?
3.3
6.0
5.0
6.0
6.0
4.0
5.0
10.0
10.3
3.0
6.0
10.0
12.0
10.0
8.0
10.0
12 .0
16.0
1 5.3
14.3
3.33
3.33
4.17
3.33
2.22
5.00
5.55
5.00
3.33
2.22
8.33
5.56
6.66
5.33
3.30
100
150
200
250
300
1.2
1.3
2.4
3.0
3. 6
TNUL FI URLS ARE FOR IATONC VE ETATIOR. Y0UN C IL?IY&?I3 CROPS 3LtOLIR B,
OTHER IMMATURE YEDETATION WILt. NAVE SHALLOWER ROOT 7 •• , $LRCL, HAVE LESS WATER
AVAILABLE FOR THE JSL OF THE YLZTAT IO*. AS THE PLANT SEVILOPS FROM A SEES OR A YOUUI
SPROUT TO THE lIATUAC PORN, THE ROOT ZONE WILL INCREASE PROGRESSIVELY FROM ONLY A PEW
INCHES TO THE VALUES t.ISTED ABOVE. USE OF A SERIES OF SOIl. NOISTURE RETENTION TABLES
WITH SUCCISSI VILY INCREASI ND VALUES OF 4YAILABU MOISTURE PERMITS THE SOIL MOISTURE TO
SE OETE I NED YHROUDNOUT THE RO’4IM SEASON.
Reproduced from
best available copy.
394
-------
TABLZ A-6
SOIl. P ISTURE R T TIOM TABLE — 25
SOIL $013TU1( RtTAIMC A1T R jI CRENT A eOUIT3 OF IOTt ?pA . cvapoyF ,spIRArIoa
NAYC OCCU* (a 1ArU NOLOINO CAPACITY 01 S0s is 25
0 1 2 3 4 5 6 7 3 3
WaTt* RC?AIJ,CO ii Soti.
0 25 24 23 22 21 20 19 18 17 16
10 16 15 15 14 13 13 12 12 11 ii
20 10 10 9 9 3 8 8 3 7
30 7 6 8 6 5 5 5 5 5 4
40 4 4 4 4 3 3 3 3 3 3
50 3 3 3 2 2 2 2 2 2 2
60 2 2 2 2 1 1 1 1 1 1
70 1 1 .1 1 1 1 1 1 1 1
80 1 1 1 1 1 1 1 1 1 1
90 1 1 1 1
-------
TABLE A-7
SOIL. t ISTUR R T NTI N TA8L. — 50 —
5 ii. o, yuAC *cria c *,rcl ifF I( T £,1OU r3 Q PQTCITIAL £Y*JOTIAM3PIRATIOI
IAf( OCCUI Q.
3 1 2 3
WATU IOL2tl C P*CITY ai sait. iS 50 MM .
4 5 $ 7 5 9
WAT!* Rt1AIUO i SOIL
0 50 49 48 41 46
10 41 40 39 38 37
20 33 32 32 31 30
30 2T 25 25 25 24
40 21 21 20 20 19
17 IT 17 16
14 14 13 13
11 11 11 10
9 9 9 5
7 7 7
6 6 6
5 5 5
4 4 4
3 3 3
3 3 3
2 2 2 2
2 2 2 2
I •1 1 1
1 1 1 1
1 1 1 1
45 44 43 42 41
36 36 35 34 33
30 29 28 28 27
24 23 23 22 2.2
19 19 18 18 18
16 16 15 15 15 14
13 13 12 12 11 11
tO 10 10 9 9 9
8 8 8 8 8 8
7 7 1 6 6 6
2 2 2 2 2 2
2 2 2 2 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1
1 1 1 1 1
50
60
70
80
90
100
110
120
130
140
150
160
170
180
190
200
210
220
6 6 6 5 S 5
5 4 4 4. 4 4
4 4 4 4 4 3 3
3 3 3 3 3 3 3
3 2 2 2 2 2 2
4
1 1 1 1 1
1 1 1 1 1
1 1 1 1 1
/ f frorn
396
-------
TABLE A-S
SOIl. t ISTUR TZNTlQH TAGL — 75
Sou’. O $TUt !CT&ILO FTU O rrUtsT AlIOuMr! OF OTt TIA £,APOT13FI A1I0I
HAYC ocCuaRC . WATCR NOLOII; cir c r or :os 75
0 1 2 3 4 5 6 1 3
‘ ATZ* R*T&,NCO * Sotz.
0 75 74 73 72 11 10 69 68 67 56
10 65 64 63 92 61 60 59 58 56 57
20 57 56 55 54 53 53 82 51 51 50
30 50 49 48 41 46 46 45 45 44 44
40 43 43 42 41 40 40 39 39 36 33
50 38 37 35 35 35 35 35 34 34 33
60 33 23 32 31 31 31 30 30 29 29
70 29 28 28 27 27 27 25 25 25 25
80 25 24 24 23 23 23 23 22 22 22
90 22 21 21 21 20 20 20 20 19 19
100 19 19 18 18 18 18 17 17 17 17
110 16 16 16 16 15 15 15 15 15 14
120 14 14 14 14 13 13 13 13 13 13
130 12 12 12 12 12 12 11 11 11 11
140 11 11 11 10 10 10 10 10 10 10
1 50 10 10 9 9 9 9 9 9 9 8
160 8 8 8 8 8 8 8 8 7 7
170 1 7 7 7 7 7 7 1 6 6
180 6 5 5 6 6 6 6 6 6 6
ISO 6 6 5 5 5 5 5 5 5 5
200 5 5 5 4 4 4 4 4 4 4
210 4 4 4 4 4 4 4 4 4 4
220 4 4 3 3 3 3 3 3 3 3
230 3 3 3 ‘3 3 3 3 3 3 3
240 3 3 3 3 3 3 3 3 3 3
250 2 2 2 2 2 2 2 2 2 2
250 2 2 2 2 2 2 2 2 2 2
270 2 2 2 2 2 2 2 2 2 2
280 2 2 2 2 2 2 2 2 2 2
290 1 1 1 1 1 1 1 1 1 1
300 1 1 1 1 1 1 1 1 1 1
310 1 1 1 1 1 1 1 1 1
320 1 1 1 1 1 1 1 1 1 1
330 1 1 1 1 1 1 1 1 1 1
340 1 1 1 1 1 1 1 1 1
350 1 1 1 1 1 1 1 1 1 1
360 1 1 1 1 1 1 1 1 1 1
rodUced
397
-------
TABr2 A-9
I 2
aTII 31T& UI ii 301$.
si Ia 37 34 30
II 36 33 31 65
31 30 71 79 TI
73 72 71 70 70
35 63 64 34 53
33 se 37
3.3 53 22 52 53
46 43 47 47 46
44 43 43 42 42
39 33 35 33 33
35 35 23 34 34
32 32 31 31 31
29 29 2! 2! 23
26 26 2! 23 23
24 23 23 23 23
21 21 21 21 20
19 19 19 19 33
17 17 57 17 1?
11 15 15 I ! 15
14 14 14 4 14
13 12 12 $2 12
31 11 11 II 11
30 15 30 50 I C
9 3 9 6 9
I 3 3 3 3
3 3 7 7 1
7 7 1 7 7
S 6 6 5 I
6 6 6 6 3
3 5 5 5 5
5 4 4 4 4
4 4 4 4 4
4 4 4 4 4
3 3 3 3 3
3 3 3 3 3
1 3 3 3 3
2 2 2 2 2
2 2 2 2 2
2 2 2 2 2
2 2 2 2 2
3 2 2 2 2
2 2 2 2 1
1 1 1 1 1
I I I I I
I 1 1 1 1
I 1 1 3 1
I I I I 1
1 1 1 1 I
1 1 1 I 1
1 1 1 1
1 1 3 1 1
94 93 32 31
35 44 53 32
71 75 75 74
SI 66 55 57
62 32 51 60
35 56 55 34
31 50 50 49
46 4! 45 44
41 41 40 40
31 37 25 35
34 33 33 33
30 33 30 33
21 27 21 27
25 24 24 24
22 22 22 22
23 20 20 20
13 18 II 13
I I 11 II IS
15 15 14 14
13 13 13 $3
12 12 12 12
11 11 11 11
to to to 10
3 9 9 1
I 5 3 I
7 7 7 7
S S S S
1 5 3 6
5 5 S 5
5 3 3 5
4 4 4 4
4 4 4 4
4 4 4 4
3 3 3 3
3 3 3 3
3 2 3 2
2 2 2 2
2 2 2 2
2 2 2 2
2 2 2 2
2 2 2 2
1 1 1 1
1 I 1 1
I I 1 1
1 1 1 1
1 1 1 1
1 1 1 1
I I I I
1 1 1 1
I I I I
1 1 1 1
I Reproduced from
best available copy.
331%. VISI1IRE tUE TION TA6LZ — ICO
SOIL IISTVI( 4 (741 1 ( 1 iSYII 4i FU(IY 8.OOIV1 f POT(1T 14L (vh 4TJ4lS IM&? 4I
a&vC CC1II (I. W 1t( lCt( 1 14 CAP8C&1’ 4S SOIl. II 335
3 4 5 I 7 9 1
Pt
0
10
20
30
40
SQ
50
70
50
30
too
110
120
130
140
1 30
160
570
130
too
200
210
223
230
240
250
253
270
230
zoo
300
313
320
330
340
330
350
310
310
330
400
410
420
430
440
430
450
473
430
430
500
9
300
30
91
74
U
53
34
49
44
40
35
32
29
26
24
22
‘9
16
16
14
‘3
12
33
9
S
3
7
S
S
S
S
4
4
3
3
3
2
2
2
2
2
2
398
-------
ABL 1-10
3011 . 7CI3TUR( i TZ 4TiCfl 7A4LE — 123 -.
Seii. MISTUl LLNUIS ?t1 11 111 141? £i UlTt U PO?EITU1. 1?CTAI$1iIi?ilU
s s xiaist,. iics 4ot ,3 laS : psc r, i : i is 125
P5 0 1 2 3 4 5 $ 7 8 3
wau mlIcs 1 3ous.
0 125 324 123 122 121 120 319 113 11? 116
10 113 114 133 112 111 210 139 106 101 106
20 101 205 104 103 102 102 101 300 39 39
30 39 1? 99 93 94 34 33 91 90
4 3 90 U U 37 31 38 35 34 34 33
30 63 52 32 31 00 50 73 79 73 Ti
30 75 15 73 14 74 73 73 72 12 11
70 70 70 49 59 53 18 37 67 66 53
40 93 34 54 43 53 52 52 61 61 60
30 60 30 39 54 23 37 51 56 56 35
100 95 55 34 54 53 33 33 52 32 51
110 91 91 50 40 43 49 43 43 43 47
120 47 47 46 46 45 45 45 44 44 43
130 43 43 ‘2 42 41 41 41 41 40 40
240 40 40 39 33 39 33 35 33 33 37
250 3? 31 33 33 35 33 33 33 45 34
150 34 34 33 .13 33 32 32 32 32 31
170 31 31 31 30 30 30 30 30 30 23
150 29 23 29 25 23 23 21 21 27 21
130 25 25 25 23 25 25 25 25 25 25
ZOO 24 24 24 24 24 23 23 23 23 23
210 22 22 22 22 22 22 22 21 21 21
220 21 23 21 21 20 20 20 23 23 20
230 13 19 19 19 13 15 13 13 18 16
240 18 13 17 17 17 27 17 17 17 17
250 15 15 11 15 15 15 15 16 15 IS
260 13 13 13 15 13 14 14 14 Ii 14
270 14 14 14 14 44 13 13 13 13 13
280 13 13 13 13 13 12 12 12 12 12
290 12 12 12 12 12 11 11 11 II 11
300 11 11 11 31 11 10 10 13 13 10
310 10 10 10 10 10 10 10 10 9 9
320 9 9 3 3 3 3 9 9 9 9
330 3 S 3 8 9 S 3 3 3 3
340 3 $ 3 3 3 7 7 7 7 7
S • •• 55•SS•S •I S S • S
0 5 0 3 0
350 7 7 450 3 3 530 1
360 7 5 460 3 3 560 1
370 5 $ 410 3 3 310 1
380 S 3 460 2 2 3 50 1
330 5 S 450 2 2 330 1
400 5 3 900 2 2 300 1
410 4 4 510 2 2 510 1
420 4 4 520 2 2 620 1
430 4 4 530 2 2 630 1
440 3 3 540 2 1 540 1
l r0d ed om
-------
A LZ A-LI
143 142 *41
133 122 131
125 124 123
113 115 114
109 103 107
112 101 100
95 94 33
39 53 37
3 33 52
77 77 19
12 12 71
53 51 57
S3 53 32
9 sa
55 54 54
92 51 51
45 45 47
45 45 44
‘2 42 41
39 39 39
37 :s 35
34 34 34
32 32 32
30 30 30
22 2! 23
25 26 25
24 24 24
22 23 22
22 21 21
20 23 23
13 18 15
11 17 17
*5 15 16
13 15 15
14 14 14
13 13 13
12 12 12
12 11 11
11 11 11
10 10 10
10 3 9
3 9 3
S S S
6 3 3
7 7 1
Reproduced from
best aveiIab e copy.
SOI l. ICISTURL 3275i 1T 1C11 ?ASU — 1 50
Si ll. $$ I9T IC iCTllU SflI • 1V9 11 11? £flØVIVS l POttITl l. (,IPOTi I3PI4ITIOI
. v5 co aiss. i& i isaa c uclTT 9ill. 3 130
Pt 0 1 2 3 4 5
41?II !tna II I a Soil.
6 7 5 3
3 150 149 148
1* 140 *39 133
:a 131 130 *29
30 122 122 121
44 114 *13 l13
30 101 106 106
60 100 39 08
73 53 32 92
30 97 35 56
90 32 31 31
100 76 75 75
113 11 71 70
120 58 56
130 62 52 51
140 33 55 37
150 54 53 93
180 51 51 50
113 47 41 41
160 44 44 44
190 41 4* 41
200 25 35 35
210 38 36 39
220 34 24 33
220 32 31 31
240 34 29 21
220 29 21 27
290 26 25 25
279 24 24 24
250 22 22 22
290 21 21 21
300 20 19 13
310 13 13 16
320 11 Il 11
330 11 15 16
340 13 15 13
330 14 14 14
360 13 13 13
370 1? 12 12
350 11 11 11
350 11 11 11
400 13 10 10
410 1 3 9
420 9 1 9
430 5 5 3
440 5 I S
141 141 143 144
*37 *36 135 *34
*25 127 121 126
120 *13 *15 111
112 *11 111 110
10! *04 103 133
97 31 31 56
31 00 90 39
33 34 34 34
80 79 79 75
73 74 74 73
70 63 59 55
65 64 54
31 60 60 50
51 56 96 93
13 32 12 52
20 10 49 49
46 45 46 43
43 43 ‘3 ‘2
40 40 43 40
31 37 37 37
35 a :5 33
33 33 33 33
31 31 31 34
29 29 23 25
27 27 21 25
23 25 23 25
23 23 23 23
22 22 22 22
20 20 23 23
19 19 19 19
13 11 18 11
11 *1 17 17
16 15 15 18
13 13 13 14
14 *4 14 14
13 13 *3 13
12 *2 12 12
11 II 11 11
10 10 10 10
13 10 10 10
9 3 9 9
S I S S
9 5 5 3
7 7 7 1
400
-------
TABLE A- il (Continued)
PE 0 1 2 3 4 5 6 1 5 9
ATC* R 1Ai CD tw SOIL
450 1 7 7 7 7 7 7 7 7
460 7 7 1 7 6 6 6 $ 6 6
470 5 6 6 6 5 6 6 6 6 6
460 6 6 6 6 6 6 6 6 5 5
490 5 5 5 5 5 5 5 5 5 5
500 5 5 5 5 5 5 5 5 5 5
510 5 5 5 5 5 5 5 5 4 4
520 4 4 4 4 4 4 4 4 4 4
530 4 4 4 4 4 4 4 4 4 4
540 4 4 4 4 4 4 4 4 4 4
550 4 4 4 4 4 4 4 3 3 3
560 3 3 3 3 3 3 3 3 3 3
570 3 3 3 3 3 3 3 3 3 3
560 3 3 3 3 3 3 3 3 3 3
550 3 3 3 3 3 3 .3 3 3 3
500 3 3 3 3 3 2 2 2 2 2
610 2 2 2 2 2 2 2 2 2 2
620 2 2 2 2 2 2 2 2 2 2
630 2 2 2 2 2 2 2 2 2 2
640 2 2 2 2 2 2 2 2 2 2
550 2 2 2 2 2 2 2 2 2 2
660 2 2 2 2 2 2 2 2 2 2
570 2 2 2 2 2 2 2 2 2 2
680 2 2 1 1 1 1 1 1 1 1
690 1 1 1 1 1 1 1 1 1 1
700 1 1 1 1 1 1 1 1 1 1
110 1 1 1 1 1 1 1 1 1 1
120 .1 1 1 1 1 1 1 1 1 1
730 1 1 1 1 1 1 I 1 1 1
740 1 1 1 1 1 1 1 1 1 1
a.... •....... S •• •• S I
0 5 0 5 0 5
750 1 1 190 1 1 630 1 1
750 1 1 800 1 1 840 1 1
170 1 1 810 1 1
780 1 1 820 1 1
Reproduced from
esI ava Iable copy .
401
-------
TABLI A-12
ii. TJø5 55 75 1 17 1 0 * TASU — ZOO r
3G 15. nOISTUU( ,iThiuL iFYSI £r4 IIVS OYIlTI I. (v&POTl a PlIaVIQI
vI SC *I(I, 4a1U . l.IIl4 1 ?&CITY S 2 t iS 200 i
p 5 3 1 2 3 4 5 5 7 5 3
&vct .qg?s,acS ii SoiL
0 200 199 133 197 196 195 134 193 132 591
10 190 133 133 161 153 ¶53 534 153 132 152
23 161 150 173 173 VT? 175 175 174 173 ¶73
30 112 171 170 153 163 155 157 153 163 ¶54
40 113 162 192 151 560 159 133 155 iS? ¶ 55
53 155 ¶54 133 5J 152 ¶51 ¶51 ¶30 143 546
50 146 147 ¶43 145 145 144 ¶43 142 142 541
70 140 40 139 134 ¶35 13? 136 135 535 134
90 133 133 132 131 133 130 129 125 ¶25 ¶27
90 127 126 129 122 124 124 ¶23 122 122 121
100 120 120 139 119 118 11! 111 115 II I ¶15
110 115 114 113 ¶13 112 112 111 110 110 109
120 109 106 ¶09 ¶01 10? 136 105 105 104 104
130 104 103 102 102 02 101 100 100 39 93
140 3! 13, 3? 37 35 96 33 94 34 44
150 34 3 33 92 92 91 31 30 50 59
163 33 35 5! 3! 31 31 16 56 55 55
170 35 34 4 33 33 32 52 32 31 61
180 50 50 50 79 73 75 75 75 7? Ti
190 75 TI 75 75 75 14 14 74 73 73
2 0 73 72 72 71 71 71 TO T O TO 95
210 69 43 53 5! 34 17 11 56 35 55
220 55 55 35 55 54 54 54 53 53 63
233 62 62 52 61 61 51 50 50 63 50
240 53 53 59 54 53 93 5! 57 57 37
230 55 35 56 54 55 55 55 54 54 34
250 54 53 53 52 52 52 52 51 91
210 51 91 50 50 50 50 49 49 49 43
250 43 43 45 44 41 47 41 41 44 46
290 46 46 45 45 45 45 45 44 44 44
:_ 44 44 43 43 43 43 42 42 42 42
42 41 45 41 41 41 40 40 40 43
320 40 39 39 39 33 33 36 33 38 33
330 35 37 37 37 31 37 31 38 35 35
340 36 .35 35 35 33 35 35 34 34 34
350 34 34 34 33 33 33 23 33 22 32
350 32 32 32 32 32 22 33 31 31 31
310 31 30 20 30 30 30 30 29 29 23
320 29 21 29 29 29 23 23 2! 25 23
350 2! 26 27 21 27 27 27 2? 27 21
400 21 25 26 25 23 25 24 25 25 25
410 25 25 23 29 25 24 24 24 24 24
420 24 24 24 23 23 25 23 23 23 23
430 23 22 22 22 22 22 22 22 22 22
440 22 21 21 21 21 21 21 21 21 21
u1 rom
402
-------
TABLE A-12 (Continted)
PC 0 1 2 3 4 5 6 7 9 9
.ATLI UTAINC ii
450 20 20 20 20 20 20 20 20 20 20
450 20 19 19 19 19 19 19 19 19 19
470 18 19 19 18 18 18 18 18 18
480 18 18 17 17 17 11 1 7 17 17
490 17 17 16 16 16 16 15 16 16 16
500 16 16 16 16 15 15 15 15 15 15
510 15 15 15 15 15 15 15 15 14 14
520 14 14 14 14 14 14 14 14 14 14
530 14 14 14 13 13 13 13 13 13 13
540 13 13 13 13 13 13 13 12 12 12
550 12 12 12 12 12 12 12 12 12 12
560 12 12 12 12 12 11 11 11 11 11
570 11 11 11 11 11 11 11 ii ii 11
580 11 10 10 10 10 10 10 10 10 10
550 10 10 tO 10 10 10 10 10 10 10
600 10 9 9 9 9 3 9 3 9 9
610 3 9 9 9 9 9 9 9 9 9
620 9 9 9 3 8 8 8 8 3 8
630 3 8 8 8 3 9 8 8 8 3
640 8 8 6 8 3 8 8 9 8 8
650 1 7 7 7 7 7 7 7 1 7
660 1 7 7 7 7 7 7 7 7 7
570 7 7 7 7 7 7 6 6 8 6
680 $ 5 6 8 6 6 8 6 6 3
660 6 6 6 6 6 5 6 6 6 8
700 5 6 6 6 6 6 6 6 6 6
710 8 6 5 5 5 5 5 5 5 5
720 5 5 5 5 5 5 5 5 5 5
730 5 5 5 5 5 5 5 5 5 5
740 5 5 5 5 5 5 5 5 5 5
750 5 5 5 5 5 5 5 5 5 5
7 60 5 5 5 5 5 5 5 5 5 5
170 4 4 4 4 4 4 4 4 4 4
7 30 4 4 4 4 4 4 4 4 4 4
790 4 4 4 4 4 4 4 4 4 4
900 4 4 4 4 4 4 4 4 4 4
810 4 4 4 4 4 4 4 4 4 4
820 4 4 3. 3 3 3 3 3 3 3
330 3 3 3 3 3 3 3 3 3 3
840 3 3 3 3 3 3 3 3 3 3
•850 3 3 3 3 3 3 3 3 3 3
360 3 3 3 3 3 3 3 3 3 3
870 3 3 3 3 3 3 3 3 3 3
380 3 3 3 3 2 2 2 2 2 2
890 2 2 2 2 2 2 2 2 2 2
403 çrom
-------
TABLE A-12 (Concluded)
0 1 2 3 4 5 6 7 8 9
‘. rc* Rgr&i’.C i Zoii
900 2 2 2
910 2 2 2
923 2 2 2
930 2 2 2
940 2 2 2
1 1 1 1
1 1 1 1
1 1 1 1
1 1 1 1
1 1 1 1
1 1 1 1
2 2 2 2 2 2 2
2 2 2 2 2 2 2
2 2 2 2 2 2 2
2 2 2 2 2 2 2
2 2 2 2 2 2 2
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
9 0
2
2
2
2
2
2
2
2
960
2
2
2
2
2
2
2
2
2
2
970
2
2
2
2
2
2
2
2
2
2
980
990
2
2
2
2
2
2
2
2
2
2
2
1
2
1
2
1
2
1
2
1
1000
1010
1020
1030
1040
‘ SW
1060
1070
1080
1090
1100
0 5
I I
I I
1 1
I I
1 1
• S S • • • •*•••.•• S S S • • S
0 5
1110 1 1
1120 1 1
1130 1 1
1140 1 1
1150 1 1
0 5
1170
1
1
1180
1
1
1190
1
1
I Reproduced Irom
best available copy.
404
-------
A3LZ A-U
SOft 1 * 5TURI 4tT8NTI0 7Aa • 250 ‘i”
Sin. ,zTumI Itl$i.CI &UtR S1SPUUV ti OUl?S O , ?tuT! (, POTR SSP,aatI 1
uaU 4C V tO. ‘ &1E t3iI$ C4P $i! .ç SQn. i 250 .
o 1 2 3 4 5 6 7 I I
VAfl fliU(S ii 4 it
0 250 243 24 241 241 245 a 243 242 241
¶0 240 239 233 231 275 233 234 233 232 231
20 231 230 229 22! 227 223 225 224 223 222
30 222 231 223 219 21! 21? :16 715 214 213
40 213 212 211 210 203 20! 73! 207 205 205
50 204 204 203 732 201 200 ¶95 195 197 194
50 191 195 154 193 552 191 ¶31 ¶50 139 15!
10 153 133 137 131 155 165 ¶34 133 132 531
30 131 130 171 17! *7? 171 113 175 175 174
50 174 173 172 571 171 *70 170 169 169 163
100 117 167 II I 165 159 154 164 153 562 ¶51
110 160 160 149 159 156 557 157 166 155 155
120 154 ¶54 153 132 152 121 151 ¶20 149 145
130 145 147 141 145 145 145 144 144 143 143
140 142 142 141 140 140 136 139 133 137 131
150 135 135 135 135 134 134 ¶33 ¶22 132 131
150 131 ¶30 130 129 129 12! 125 121 12? 12S
¶70 ¶25 125 125 124 124 123 ‘23 122 122 121
150 121 129 120 119 119 115 118 11? 11? ¶16
190 115 115 Ii! ¶14 114 *14 113 113 112 112
200 1*1 111 110 110 109 109 105 10! IC! 101
210 107 ¶31 106 105 105 105 105 104 104 ¶03
220 103 103 102 102 101 101 101 100 100 39
230 99 65 95 9! 31 ST 37 35 96 35
240 93 35 54 34 33 33 33 32 52 91
250 91 91 50 35 39 59 69 68 55 57
250 37 81 57 31 86 36 3 ! 55 85 34
230 34 54 53 63 52 82 52 32 51 31
230 61 51 SO SO 79 79 13 79 15 74
290 13 73 77 77 73 75 75 15 75 75
300 74 74 14 73 7 1 73 13 72 72 71
310 71 TI TI 70 70 70 7* 99 99 53
323 59 58 68 58 57 57 3? 65 65 66
330 96 56 65 65 54 55 94 64 54 53
340 53 53 53 62 62 62 61 31 51
34* 51 51 60 30 so so 59 si 93 5 .
360 53 58 58 5? 57 57 5? 51 51 31
370 51 55 55 2.5 55 53 54 34 54 54
350 54 54 33 53 53 53 53 52 52 52
330 52 52 51 51 51 51 51 50 50 30
400 50 50 49 49 49 49 49 45 46 45
410 46 45 47 47 47 47 47 46 46 41
420 46 46 45 45 45 45 45 44 44 44
430 44 44 43 43 43 43 43 42 42 42
440 42 42 42 42 41 41 41 41 41 41
405
-------
450 41 41 40
460 39 39 39
470 37 37 37
420 36 36 36
490 34 34 34
600
510
520
630
640
550 18
560 17
670 17
680 15
620 15
700 15
710 14
720 14
730 13
740 13
33 33 33
32 32 32
31 31 30
29 29 29
23 23 25
TABLE A—13 (C r.tiU e )
27 27 27 27
28 25 26 25
25 25 25 25
24 24 24 24
23 23 23 23
3 4 5 6 7 a a
40 40 40 40 40 40 40
39 39 38 38 38 38 38
37 37 37 37 36 35 36
35 35 35 35 35 35 25
34 34 34 34 34 33 33
33 33 32 32 32 32 32
32 31 31 31 31 31 .31
30 30 30 30 30 30 30
29 29 29 29 29 29 28
25 29 28 28 23 27 27
27 25 26 26 25 28
25 26 25 25 25 25
25 24 24 24 24 24
24 24 24 23 23 23
23 23 23 23 22 2.2
12 12 12 12
12 12 12 12
11 11 11 11
11 11 11 11
10 tO 10 10
8 8 8
8 3 8
7 7 7
7 7 7
7 7 7
12 12 12 12 12 12
11 11 11 11 11 11
11 11 11 11 11 11
10 10 10 10 10 10
10 10 10 10 10 10
3 8 8 3 8 8 8
8 8 8 8 5 7 7
7 7 1 7 1 7 7
1 7 1 7 7 7 7
7 7 r 7 7 7 7
406
I Reproduced Irorn
best ava abIe copy.
0 1 2
W &? * ETuI(1 ii 3oi .
500
510
520
530
540
550
560
570
580
520
22 22 22 22 22 22 22 22 21 21
21 21 21 21 21 21 21 21 20 20
20 20 20 20 20 20 20 20 20 20
20 20 20 19 19 19 19 19 19 19
19 16 19 19 19 18 18 18 18 18
18 18 18 18 18 18 18 17 17
17 17 IT 17 17 17 17 17 17
17 17 17 16 16 16 15 16 16
15 18 15 16 16 16 16 15 15
15 15 15 15 15 15 15 15 15
15 15 15 15 14 14 14 14 14
14 14 14 14 14 14 14 14 14
14 14 14 13 13 13 13 13 13
13 13 13 13 13 13 13 13 13
13 13 12 12 12 12 12 12 12
750
760
770
780
7 60
300
810
820
830
840
850
360
870
880
390
10
3
3
9
3
10 10 10 10 10 10 10 9 9
9 9 9 9 9 9 9 9
9 9 9 9 9 9 9 9 9
9 9 9 3 8 8 8 3 8
3 8 6 8 8 3 8 8 8
-------
TABLE A-13 Conc1uded)
0 1 2 3 4 5 5 a S
WATER CTAl (5 ii $csz.
900 $ 6 6 6 5 $ 6 5 5 6
910 6 6 6 6 6 6 6 6 6 6
920 6 6 6 6 6 6. 6 5 5
a 8 6 6 6 5 5 9 5 6
940 6 $ 6 6 6 6 6 5 5 5
950 5 5 5 5 5 5 5 5 5 5
950 5 5 5 5 5 5 5 5 5 5
970 5 5 5 5 5 5 5 5 5 5
980 5 5 5 5 5 5 5 5 5 5
990 5 5 5 5 5 5 5 5 5 5
1000 4 4 4 4 4 4 4 4 4 4
1010 4 4 4 4 4 4 4 4 4 4
1020 4 4 4 4 4 4 4 4 4 4
1030 4 4 4 4 4 4 4 4 4 4
1040 4 4 4 4 4 4 4 4 4 4
1050 4 4 4 4 4 4 4 4 4 4
1060 3 3 3 3 3 3 3 3 3 3
1070 3 3 3 3 3 3 3 3 3 3
1080 3 3 3 3 3 3 3 3 3 3
1090 3 3 3 3 3 3 3 3 3 3
1100 3 3 3 3 3 3 3 3 3 3
1110 3 3 3 3 3 3 3 3 3 3
1120 3 3 3 3 3 3 3 3 3 3
1130 3 3 3 3 3 2 2 2 2 2
1140 2 2 2 2 2 2 2 2 2 2
1150 2 2 2 2 2 2 2 2 2 2
1160 2 2 2 2 2 2 2 2 2 2
1170 2 2 2 2 2 2 2 2 2 2
1180 2 2 2 2 2 2 2 2 2 2
1190 2 2 2 2 2 2 2 2 2 2
1200 2 2 2 2 2 2 2 2 2 2
1210 2 2 2 2 2 2 2 2 2 2
1220 2 2 2 2 2 2 2 2 2 2
1230 2 2 2 2 2 2 2 2 2 2
1240 2 2 2 2 2 2 2 2 2
1250 2 2 2 2 2 2 2 2 2 2
S SI•S••S• S S U • 5S••S•S
0 5 0 5 0 5
1250 2 2 1310 1 1 1400 1 1
1270 1 1 1320 1 1 1450 1 1
1280 1 1 1330 1 1 1500 1 1
1290 1 1 1340 1 1 1550 1 1
1300 1 1 1350 1 1
[ geproduced from
407 I best available. copy.
-------
A3LZ A-14
301*. rCi5*U t .ICTENTION TA3U — 300 ,v
.5OS NUSYVII UUIIIS FYtJ $I Ft*(IT UVCIIItZ V 1UTI1I. j,sF1 a3PiIa? e
I&V( ocgUIU . W*T1 l. IIS * C*T7 l Z IL ts 300 .
P3 3 1 2 3 4 5 5 1 5 3
siii ii Son.
3 300 213 213 29? 294 215 224 293 292 291
10 290 295 283 237 286 235 284 233 232 251
20 280 219 211 27! 217 275 273 274 273 272
30 271 273 253 255 253 257 286 255 254 253
40 262 251 250 250 259 255 237 216 235 284
10 234 213 252 211 210 243 243 245 24? 246
50 245 244 244 243 242 241 240 240 233 223
70 237 23! 235 233 234 233 232 232 231 220
50 229 223 228 227 ! 225 225 224 223 222
90 231 220 219 219 21! 217 216 213 213
100 214 214 213 212 212 211 210 239 209 203
*10 20? 207 235 205 234 204 203 202 202 201
120 230 200 199 19! 198 197 *96 19! *95 194
133 194 193 *22 *92 191 191 192 139 159 153
140 15? 157 195 195 135 154 154 153 152 152
130 181 151 150 173 111 115 17 ! 177 175 175
160 171 175 174 173 173 172 112 171 171 173
120 173 159 135 155 167 157 165 166 153 164
130 164 163 163 182 162 161 150 160 139 139
110 156 155 151 t57 155 136 133 135 134 154
230 153 233 152 132 151 15* 150 130 149 149
210 148 *43 14? *47 146 145 145 146 144 144
220 143 143 142 *42 141 141 140 140 133 133
220 134 13! 135 13? 137 135 135 133 133 134
240 134 133 733 132 132 132 131 131 130 130
250 133 129 *28 12! 123 12? 121 *23 125 125
210 125 125 124 124 124 *23 123 122 132 121
270 $21 121 120 120 113 119 113 113 118 117
280 117 11? 115 115 115 113 115 114 114 114
290 113 *13 112 112 112 111 111 110 110 110
300 109 *09 101 10 ! 108 106 10? 101 106 106
310 135 103 l OS 105 *04 104 104 103 *03 103
320 102 102 102 101 101 *01 100 100 100 33
330 29 38 93 98 91 31 9? 97 96 95
340 36 95 96 3! 34 94 94 51 33 23
350 32 33 92 91 31 31 90 98 90
350 59 53 53 53 35 i 85 37 87 3?
313 95 IS 95 56 53 55 45 34 54 54
330 64 53 e3 53 32 62 32 52 51 31
390 I I 50 50 60 50 30 79 73 79 73
400 7 ! 73 13 71 7? 7? 17 75 76 7 5
410 15 75 7 ! 15 14 74 74 74 74 73
420 73 73 12 72 72 12 72 11 71 71
430 11 10 70 73 10 70 53 19 59 65
440 11 54 63 58 57 5? 5? 57 56 61
e r0m.
va ab 0PY.
408
-------
TA3LE A-13 (Ccntinued)
3 1 2 3 4 5 6 8 9
WATtR RLyA,uL II 301L
450 55 65 55 63 65 63 65 64 64 54
463 64 54 63 63 63 63 63 62 52 62
470 62 62 61 61 61 81 51 50 60 60
480 60 60 59 59 59 59 59 58 58 58
460 58 58 51 57 57 57 57 55 55 56
500 56 56 55 55 35 55 55 54 54
510 54 54 54 53 53 53 53 53 52 52
520 52 52 52 52 51 51 51 51 51 50
530 50 50 50 50 50 50 49 49 49 49
540 49 49 48 48 45 48 46 48 47 47
550 47 41 47 47 46 46 46 46 46 46
550 46 45 45 45 45 45 45 44 44 44
570 44 44 44 44 44 43 43 43 43 43
580 43 42 42 42 42 42 42 42 42 41
590 41 41 41 41 41 41 40 40 40 40
530 40 40 40 39 39 39 39 39 39 39
910 35 38 38 38 38 38 38 35 35 37
520 37 37 37 37 37 37 36 36 35 36
630 36 36 39 36 35 36 35 35 35 35
640 35 35 35 34 34 34 34 34 34 34
650 34 34 33 33 33 33 33 33 33 33
660 32 32 32 32 32 32 32 32 32 32
670 32 31 31 31 31 31 31 31 31 31
580 30 30 30 30 30 30 30 30 30 30
660 30 29 29 29 29 29 29 29 29 29
700 23 23 29 28 28 28 23 28 28 28
710 28 27 27 27 21 27 27 27 27 27
120 27 25 26 25 25 26 26 as 25 26
730 25 26 26 26 25 25 25 25 25 25
740 25 25 25 25 25 24 24 24 24 24
150 24 24 24 24 24 24 24 24 23 23
760 23 23 23 23 23 23 23 23 23 23
770 22 22 22 22 22 22 22 22 22 22
7 50 22 22 22 22 22 21 21 21 21 21
790 21 21 21 21 21 21 21 21 20 20
800 20 20 20 20 20 20 20 20 20 20
810 20 20 20 20 20 19 19 19 19 19
820 19 19 19 19 19 19 19 19 19 18
330 18 18 18 18 18 18 18 18 18 18
840 18 18 18 18 18 18 15 17 17 17
950 17 17 17 17 17 17 17 17 17 17
960 17 17 17 1? 16 16 16 16 16 16
310 16 16 16 16 16 16 15 15 16 16
890 16 15 16 16 15 15 15 15 15 15
890 15 15 15 15 15 15 15 15 15 15
[ Reproduced from
409 best available copy .
-------
TABLE A-14 (Conc1uded)
a 1 2 3
4 5 9 7 3 9
ArE* (ZTAI 3 ii
1000
1010
1020
1030
1040
1050
1060
1070
1060
1090
1100
1110
1120
1130
1140
1150
1160
1170
1180
1190
1200
1210
1220
1230
1240
1250
1260
1270
1280
1290
9 9 9
a 8 8
3 8 6
3 8 8
S S 8
S S
S
a S
6 S
6 6
0 5
S $
5 5
5 5
5 3
5 5
4 4
4 4
4 4
4 4
4 4
12 12 12 12
12 12 12 12
11 11 11 11
11 11 11 11
11 11 11 11
1300 4
1310 4
1320 4
1330 3
1340 3
1350 3
1350 3
1310 3
1380 3
1390 3
900
910
920
930
940
15
14
14
13
13
15
14
14
13
13
14
14
14
13
13
14
14
14
13
13
14
14
13
13
12
14
14
13
13
12
14
14
13
13
12
14
14
13
13
12
14
14
13
13
12
14
14
13
13
12
950
960
970
12
12
12
12
12
12
.
12
12
11
12
12
11
12
12
11
12
12
11
980
990
11
11
11
11
11
11
11
10
11
10
11
10
10
10
10
9
9
10
13
10
9
9
10
13
10
9
9
10
10
10
9
9
10
10
10
9
9
10
10
10
9
9
10
10
10
9
9
10
10
10
9
9
10
10
10
9
9
10
13
10
9
9
9
8
8
8
3
9
8
3
8
6
9
3
8
8.
3
9
8
8
8
8
9
8
8
8
8
3
8
8
a
8
9
6
8
a
a
7
7
7
I
6
7
7
1
T
5
1
7
7
1
6
7
1
7
7
6
7
7
7
7
3
7
7
7
7
6
7
7
7
1
5
7
7
7
1
6
1
7
7
7
5
1
1
1
7
S
6
S
5
6
6
6
6
6
6
6
6
6
5
6
6
6
5
6
s
5
5
6
6
5
6
S
S
5
5
6
8
5
6
8
6
5
3
0
5
4
4
4
3
3
1400
1410
1420
1430
1440
3
3
2
2
2
3
3
2
2
2
3
3
3
3
3
besi
1450 2
1460 2
1470 2
1480 2
1490 2
Reproduced from
2
2
2
2
2
• S. •• S • S •a u • e S. S S S • S
0 5
410
-------
T LZ ? .-L5
sou. n3ssr t lcw rios r .ie — 21o
Søn. i,aisTu t c1AIIg VY*I 3IVV( IIt AM I1I a ‘OttNVIAI. (y POT .$PIaIfIOI
liii )CCl lt$. .4?ta Q iS &P Cl?Y s zot is 3.50 P
P8 0 1 2 3 4 5
Wi,ta fliuLI ii Soit.
ro b py.
4 13 .
8
1
3
3
0
3.50
343
341
341
3.46
343
344
343
342
341
10
340
333
333
337
338
334
334
333
333
331
20
330
323
323
327
325
326
324
323
332
321
30
3.21
323
319
318
311
318
318
315
314
3.13
40
312
311
310
3.09
3.08
301
307
306
3.05
204
33
3.03
302
3.01
300
299
231
293
237
216
230
30
204
294
233
202
201
200
239
aS
ai
281
70
258
254
254
293
282
232
231
2.53
219
2.?!
30
273
277
2.75
273
274
2.74
273
272
271
2.70
90
3.70
251
258
237
263
256
26.5
254
253
253
100
282
252
261
280
259
3.59
253
257
233
25.5
110
253
254
253
2.52
251
2.51
2.50
230
249
243
120
241
247
246
245
244
244
243
243
242
341
130
241
2.40
239
235
237
231
238
236
225
234
140
224
233
23]
232
231
231
230
223
223
221
190
32?
225
225
225
224
224
223
223
222
221
150
221
220
220
219
213
213
2.17
217
215
215
170
215
214
214
2.13
212
211
211
210
203
209
130
:08
208
201
207
333
205
20.5
205
204
203
190
203
202
202
201
200
200
139
199
196
197
200
107
.594
198
133
194
194
103
193
152
111
210
131
830
130
169
139
138
185
III
13?
135
220
185
155
184
134
183
153
132
132
181
131
230
161
130
179
119
178
178
178
177
173
116
240
175
175
174
174
173
173
173
133
111
173
220
170
170
169
163
163
153
157
151
156
161
260
166
135
165
164
184
15 ]
183
162
162
151
270
161
851
160
160
153
139
153
153
137
517
250
156
156
133
155
154
154
154
153
113
152
230
152
152
151
151
150
150
150
149
149
143
300
148
146
141
841
148
145
845
144
144
.543
310
143
143
143
112
141
141
141
140
540
13.9
320
330
133
135
131
133
138
135
13!
134
13?
134
131
134
137
133
136
133
!31
13]
135
132
340
132
132
131
131
130
130
130
129
123
12!
350
123
121
127
827
128
126
126
125
125
124
350
124
124
123
123
122
122
122
122
121
121
310
121
121
120
120
120
119
113
819
Ii!
Ill
380
117
111
111
115
118
816
816
115
113
115
310
114
114
814
113
113
112
112
112
111
111
400
111
Ill
110
110
110
101
109
109
105
10$
410
108
103
107
10?
107
106
106
108
105
105
420
104
104
104
103
103
t03
103
102
.502
102
430
102
102
101
101
101
100
100
100
39
99
440
39
31
33
38
33
31
97
9?
98
I I
-------
TABLE A-iS (Conti1U d)
a
1 2 3
5 3 7 5 3
WATtS RzTaanE 1 3O .
450
460
410
480
490
96 95 95
93 93 93
90 93 93
88 88 88
85 35 as
300 35
810 34
820 33
830 32
840 31
54 54 83 53
52 52 52 52
51 51 51 50
50 50 49 49
48 48 48 47
40 40 40
39 39 39
38 38 38
37 37 37
35 35 36
35 35 35
34 34 34
33 33 33
32 32 32
31 31 31
30 30 30
30 30 29
29 29 2!
23 28 23
21 27 27
35 94 94 94 94 93
92 92 92 91 91 91
89 89 89 39 58 83
87 8? 87 88 83 56
84 84 84 84 54 33
82 92 31 81
80 19 79 79
77 17 76 76
15 75 74 74
73 73 72 72
53 53 83 52 52
52 52 51 31 51
30 50 50 50 50
49 49 49 48 45
41 47 47 47 41
40 40 40 40 40 40
39 39 39 39 39 38
38 38 38 38 38 37
37 37 31 37 37 36
36 36 35 36 35 35
35 34 34 34 34 34
34 34 34 33 33 33
33 33 33 33 32 32
32 32 32 32 32 31
31 31 31 31 31 30
30 30 30 30 30 30
29 29 29 29 29 29
28 28 28 28 28 23
27 21 27 27 27 21
27 27 27 26 23 26
412
lieproduced from
L st avaiIab e copy .
35
92
9,
87
55
83
81
15
76
14
83 83 83 32 82
81 81 0 80 80
78 13 15 17 77
76 76 75 75 78
74 74 73 73 73
72 71 71 71
TO 59 89 69
63 67 67 67
66 65 65 65
64 54 63 63
62 32 62 51
60 60 SO
59 58 58 58
37 57 56 56
55 55 55 55
12 72
70 70
63 68
66 56
64 64
62 52
60 60
59 59
57 57
55 56
500
510
520
530
540
550
860
570
580
550
600
610
620
630
640
6.50
680
670
360
590
700
110
720
130
740
150
760
770
150
190
71 71 10 70
59 69 68 88
57 67 66 53
65 55 65 54
53 53 63 53
61 61 61 61
60 59 59 59
33 58 56 57
56 56 56 56
55 55 54 54
54
32
51
50
48
47 47
45 45
44 44
43 43
42 42
47 46 46
45 45 45
44 43 43
43 42 42
41 41 41
46 48 46 46 48
45 45 45 44 44
43 43 43 43 43
42 42 42 42 42
41 41 41 41 41
40
39
33
3?
36
850
560
870
680
590
30
30
29
28
27
-------
TABLE A13 (Cor c1ided)
0 1 2 3 4 5 S 7 3 9
WAlE! R.cTLIIu ii Soit,
900 26 26 26 25 26 25 26 25 25 25
910 26 26 26 25 25 25 25 25 25 25
920 25 25 25 25 24 24 24 24 24 24
930 24 24 24 24 24 24 24 24 24 23
940 23 23 23 23 23 23 23 23 23 23
950 23 23 23 23 22 22 22 22 22 22
960 22 22 22 22 22 22 22 22 22 22
970 22 22 22 21 21 21 21 21 21 21
980 21 21 21 21 21 21 21 21 20 20
990 20 20 20 20 20 20 20 20 20 20
1000 20 20 20 20 .19 19 19 19 19 19
1010 19 19 19 19 19 19 13 19 13 19
1020 19 19 19 18 18 16 18 18 18 18
1030 18 18 18 iS 18 18 1 16 18 18
1040 18 16 18 18 17 17 iT 17 17 17
1050 17 17 17 17 17 17 17 17 17 11
1060 17 17 17 16 16 16 16 16 18 16
1070 16 16 16 16 15 15 16 16 16 16
1080 16 16 18 15 15 15 15 15 15 15
1090 15 15 15 15 15 15 15 15 15 15
1100 15 15 15 15 15 15 15 15 14 14
1110 14 14 14 14 14 14 14 14 14 14
1120 14 14 14 14 14 14 14 14 14 14
1130 14 14 14 13 13 13 13 13 13 13
1140 13 13 13 13 13 13 13 13 13 13
. .. S • • S S •S S S S • S S S S S •
0 5 0 5 0 5
1150 13 13 1300 8 8 1450 5 5
1160 12 12 1310 8 8 1500 5 5
1170 12 12 1320 5 8 1550 4 4
1180 12 12 1330 8 5 1600 4 4
1190 11 11 1340 8 8 1650 3 3
1Z I I 11 11 1350 7 7 1700 3 3
1210 ti 11 1360 7 1 1750 2 2
1220 10 10 1370 1 7 1500 2 2
1230 10 10 1380 7 1 1850 2 2
1240 10 10 1390 6 6 1800 2 2
1250 10 10 1400 5 6 1950 1 1
1260 9 9 1410 6 6 2000 1 1
1270 9 9 1420 S $ 2100 1 1
1280 9 9 1430 6 6 2200 1 1
1290 5 8 1440 6 6
413
I Reproduced from
besi available copy.
-------
3I2 -i5
SOIL ITU1E RET T 3N T. 3U — 400
3ei ae$Tu urislCS &17Z1 u 7ULAT £? U I5 • Tt Y At. tTh OV &l$ tL&TIø
I VL 0C5411(3. .ATU AQ lt3 C*P’.C T’Y O 3 It. 3 400
3 400 393 331
10 330 331 318
20 360 379 373
30 371 370 369
40 352 331 363
:52 333
30 :44 344 343
73 335 335 334
80 327 325 225
30 119 313 317
100 311 310 239
113 333 302 302
120 298 235 294
130 288 263 237
140 281 231 260
150 274 274 273
¶60 267 257 233
110 261 230 250
180 254 214 213
190 248 246 247
233 242 242 241
213 235 235 233
223 230 233 229
230 224 224 223
240 213 211 218
253 213 213 213
260 208 208 231
273 233 202 212
260 138 197 197
250 ¶33 112 ¶02
300 158 188 187
313 163 133 132
320 179 179 7!
330 174 174 174
340 170 170 151
330 163 151 153
360 162 162 111
313 153 153 13?
380 134 154 153
390 150 150 149
400 143 141 145
413 *43 142 42
420 126 139 139
433 136 135 13.5
440 132 132 132
337 355
337 338
371 375
353 357
353 :33
330 349
342 341
333 332
324 323
119 115
308 307
301 230
233 252
265 235
271 273
212 271
233 294
253 235
252 262
246 243
240 233
234 223
233 22!
223 222
211 218
212 211
201 236
201 230
136 135
131 191
167 136
¶62 ¶31
:;a I ??
113 173
113 161
153 *64
*61 ¶80
137 *56
153 153
149 148
145 *45
142 141
933 13!
¶33 134
131 131
414
331
3 61
312
263
354
:45
336
327
319
311
303
29 i
268
252
275
268
251
2
243
242
3.36
231
225
213
214
233
203
193
133
¶88
164
‘73
175
171
‘U
152
133
154
150
‘4’
143
140
136
133
129
Reproduced
S
FE 9 1 2 3 4 3
4 T*a 8t &i.( ‘a
5 7 a 3
153 394 393 332
385 384 363 382
315 373 374 373
356 366 355 264
337 351 358 353
346 345 347 346
340 239 328 337
331 330 229 325
323 322 321 320
315 314 313 312
307 306 335 204
300 299 296 297
2 291 290 239
3.65 234 233 262
275 277 275 275
271 273 299 265
234 254 253 262
251 237 255 3.55
261 251 230 249
245 244 244 243 -
239 233 226 231
233 233 232 231
237 227 226 223
222 221 221 220
211 218 215 215
211 210 210 209
235 203 204 204
ZOO 200 ¶99 139
135 195 114 114
190 130 139 ¶69
165 165 135 184
151 131 150 *30
117 171 115 *75
172 172 172 171
153 168 167 167
154 154 153 ¶63
160 150 139 *59
135 155 153 153
132 132 151 131
143 146 147 14?
144 144 144 143
141 141 140 *40
137 137 137 138
134 ¶34 133 133
131 133 ¶30 ¶30
-------
TABLE A-16 (Continued)
o 1 2 3 4 6 3 9
WaTLR RcraI*L II $OIL
450
460
470
480
450
129
126
123
120
117
129
125
122
120
117
123
125
122
119
116
128
125
122
113
115
123
124
121
119
116
127
124
121
118
115
127 127 126 126
124 124 123 123
121 121 120 120
118 113 117 11?
115 115 114 114
500
510
520
530
540
114
111
108
105
103
114
111
108
105
103
113
110
108
105
102
113
110
107
105
102
113
110
107
104
102
112
110
107
104
102
112 112 111 111
109 109 109 108
10? 106 106 106
104 104 103 103
101 101 101 101
550
560
570
580
590
100
98
95
93
31
100
98
85
93
91
100
98
95
93
90
100
91
95
92
90
99
91
94
92
90
99
97
94
92
90
99 99 99 93
95 96 96 96
94 9 93
92 91 91 91
89 59 89 89
600
610
820
630
640
88
85
84
82
80
88
86
84
52
80
88
86
54
82
80
88
86
54
82
80
53
86
84
82
80
ST
35
83
81
79
87 87 87 87
55 35 85 85
33 83 83 63
81 31 61 81
19 73 19 79
550
660
670
680
690
78
76
74
72
70
78
76
74
72
70
18
76
74
72
TO
78
75
74
72
70
18
78
74
72
70
77
75
13
71
70
77 77 17 77
75 75 15 75
73 13 13 73
11 71 71 71
69 59 69 69
700
110
720
730
740
53
67
65
64
62
63
67
65
54
52
58
67
55
64
62
68
67
65
63
62
68
66
65
63
61
53
66
65
33
61
58 63 61 57
86 65 66 56
34 64 64 64
63 63 62 62
61 91 61 61
750
760
110
180
790
61
59
58
56
55
50
59
58
56
55
60
59
58
56
55
60
59
57
56
5.4
50
59
57
56
54
60
58
57
56
54
50 60 50 59
58 56 58 58
57 57 57 56
55 55 55
54 54 54 5+
52
51
80
48
47
46
45
44
43
42
r fa rom
53
51
50
49
46
800
810
620
830
840
54
52
51
50
46
54
52
51
50
48
53
52
51
49
48
53
52
50
49
48
53
52
50
49
48
53
52
50
49
46
53
52
50
49
48
•
350
360
370
980
880
47
46
45
44
43
47
46
45
44
43
41
46
45
44
43
47
45
44
43
42
47
45
44
43
42
46
45
44
0 43
42
46
45
44
43
42
46
45
44
43
42
46
45
44
43
42
52
51
50
49
47
415
-------
TABLE A—16 (Conti. ued)
0 1 2 3 4 5 6 1 8 9
AE RLTAIIt LOU.
900 42 42 42 41 41 41 41 41 41 41
910 41 40 40 40 40 40 40 40 40 40
920 39 39 39 39 29 39 39 39 39 39
930 38 38 38 38 38 38 23 38 38
940 38 38 37 37 37 37 3 1 31 37 37
950 31 37 31 36 36 36 36 36 36
960 36 36 36 35 35 35 35 35 3 ,5 35
970 35 35 25 34 34 34 34 34 34 34
980 34 34 34 34 34 34 34 33 33 33
990 33 33 33 33 33 33 33 33 22 32
1000 32 32 32 32 32 32 32 32 32 32
1010 32 31 31 31 31 31 31 31 31 31
1020 31 31 31 30 30 3’) 30 30 30 30
1030 30 30 30 30 30 30 30 30 29 29
1040 29 29 29 29 29 29 29 29 29 29
1050 28 26 23 28 28 2! 28 25 23 28
1060 28 23 23 27 27 27 27 27 27 27
1070 27 27 27 27 27 27 27 27 25 28
1080 26 26 28 26 26 25 26 26 25 26
1090 25 26 25 25 25 ‘ 25 25 25 25 25
1100 25 25 2! 25 25 25 25 25 25 25
1110 24 24 24 24 24 24 24 24 24 24
1120 24 24 24 24 24 24 24 24 24 23
1130 23 23 23 23 23 23 23 23 23 23
1140 23 23 23 22 22 22 22 22 22 22
1150 22 22 22 22 22 22 22 22 22 22
1160 . 22 22 22 21 21 21 21 21 21 21
1173 21 21 21 21 21 21 21 21 21 21
1180 21 21 20 20 20 20 20 20 20 20
1190 20 20 20 20 20 20 20 20 20
1200 20 20 19 19 19 19 19 19 19 19
1210 19 19 19 19 19 19 19 19 19 19
1220 19 19 19 18 18 18 18 18 18 18
1230 18 18 18 18 18 18 16 18
1240 18 Ia 18 11 17 17 11 17 17 17
1230 17 17 17 17 17 11 17 17 17 11
1260 17 IT 11 17 11 17 17 15 16 16
1270 18 16 16 16 16 16 16 16 16 16
1280 16 16 16 16 16 16 16 16 16 16
1290 16 16 1 5 15 15 \15 is is is 15
1300 15 15 15 15 15 15 15 15 is is
1310 15 15 15 15 15 15 15 15 15 15
1320 14 14 14 14 14 14 14 14 14 14
1330 14 14 14 14 14 14 14 14 14 14
1340 14 14 14 14 14 14 14 14 14 14
A 1 Reproduced From
best available copy .
-------
TABLE A-lG (Concluded)
0 1 2 3 4 5 6 8 9
WATER MC?AI*Ea Soit.
1350 13 13 13 13 13 13 13 13 13 13
1360 13 13 13 13 13 13 13 13 13 13
1370 13 13 13 13 13 13 13 13 13 13
1380 12 12 12 12 12 12 12 12 12 12
1390 12 12 12 12 12 12 12 12 12 12
1400 12 12 12 12 12 12 12 12 12 12
1410 12 12 12 12 12 11 11 11 11 11
1420 11 11 11 11 11 11 11 11 11 11
1430 11 11 11 11 11 11 11 11 11 11
1440 11 11 11 11 11 11 11 11 11 11
1450 10 10 10 10 10 10 10 10 10 10
1460 10 10 10 10 10 10 10 10 10 10
1410 10 10 10 10 10 10 10 10 10 10
1480 10 10 10 10 10 10 10 10 10 10
1490 . 9 9 9 9 9 9 9 9 9 9
1500 9 9 9 9 9 9 9 9 .9 9
1510 9 9 9 9 9 9 9 9 9 9
1520 9 9 9 9 9 9 9 9 9 9
1530 8 8 8 8 9 8 8 8 8 8
1540 8 8 8 3 8 8 8 8 8 8
1550 8 3 8 8 8 8 8 8 8 5
....••• ...,u S.. 55S•S
0 5 0 5 0 5
1560 8 8 1750 5 5 2250 1 1
1570 3 3 1800 4 4 2300 1 1
1580 8 3 1850 4 4 2350 1 1
1590 7 7 1900 3 3 2400 1 1
1600 7 7 1950 3 3 2450 1 1
1620 7 7 2000 3 3 2500 1 1
1640 8 6 2050 2 2 2550 1 1
1660 6 8 2100 2 2 2600 1 1
1830 8 6 2150 2 2 2650 1 1
1700 8 6 fl O O 2 2
Reproduced from
417 best available copy.
-------
APPENDIX B. CASE STUDY SITES USED FOR EVALUATION OP LEi ClIAPE GENERATION MODELS.
JAthi 6-I. SII4AMI 01 SILLCILO COISUIIIONS Al CASt 51W! 511(5 (SLS INGINCII IS. )
• Operst.d as S landfill Iros Its Iuceptiw,.
P OrigInally operated as a burning diamp now opsr sled as a lanJ(Ill.
• filling activities co*plete; SIte 00W tlo cJ.
f.r.eability vaI e st.o as are fur soil used as cover . aerial. ca/Sec.
I — e 44cnlij l, C — Co.mercial. I • inJust.i sl.
si
Site
A
a)_tocatioa
Northwe l
(Stat, of Washington)
p raljh k.l_1eaturs ’
$urI ce Area
21 58
Average Pcpth
of 1111
N it
10.0 33
Osily Vests Ad.
Tnne
901 1000
Ags of
Stt.
IO
Wisi. ly e
recdtl
* dill (est)
C 30 (esi)
I IU (. 5 1)
j trin
area
Sate part of glacial outwesh
vallcy streligraphy cosposed
of glacial tills. outwa is
revels. en t silty cl y .
pets eabIlily • 10
Average surlac. slop. • 2.51
West
(Stats of C.li( rsis)
Area of .*tavu lcanlc ürigla
with well de(i ed Ledroch.
Soils consist of cl.yey
gravels md outws b send/
gravsl s
Soil pcr*cabIllty — t0 ;
Averags surfacs slope — 21
tO 26
6.) 20
91
tOO
.
2O
6 66 (e t)
C 26 last)
1 10 (est)
ca iyun
c
Midwest
Area cooposed of bedrock of
blue shale /I iae tone 1 over-
lai, by silty clsysi
Soil pesseability • 10
Average urtacs slope — 2.6%
0.4 I
6.1 20
IS
I?
3
I SO
C SO
I 0
trench
area
I
fast
(Slate of Penosylvasis)
Ares of igneous intrusion
with dI aLa i and lisestono
bedrock’ silty loss soi)
Soil per. ablIity — to’
Average surface s4upe — 101
4 II
2.6 6
81
96
2’
1 48
C 21
I 31
area
Soutis
(Stat. of Ieen . s.e)
Azia of fractured Cavabrian
lial. bedrocks overla1 by
wettkered sissIes and cltrs;
Soil per.vabIlity — 1w
Aver.ye surface Iope - 101
3 1
6.1 20
116
95
2
6 88
c g
I 3
caiiyosa
a o l
tresich
= — —-.—.-.--
a 0
<
0 ’
no
o
-------
3ARfl 0-2. HFASL*(0 0*1* OF 1LA IIAI1 GEUIRAIIWI. PKLCIP11AII0 . *tm (VAZ UltA1I0N. (SCS Iugloeers, 1976)
A B C I I I
S It. I I ,, .
I4istk L P 1 i. P 1 L P 1 i. P 1 I. P
January . -. --- --- 0.13 3.42 1.05 0.81 1.19 0 --- --- ---
February 0.06 1.34 --- 0.16’ 12.22 0.10 0.10 4.04 0 0.01 0.78 0 0.08 4.35 1.17
March 0.10 6.19 2.02 O.20 14.20 1.99 0.10 5.44 0 0.02 2.68 0 0.19 9.12 3.07
A wI1 0.18 4.32 2.46 0.23 2.43 3.10 0.10 3.22 2.93 0.01 2.52 1.90 0.08 2.311 4.29
Hay 0.13 2.45 4.34 0.02 0.28 8.36 2.33 2.01 6.61 0.04 2.48 4.82 0.01 4.63 4.41
June 0.03 2.41 4.44 0.14 0.22 9.20 I . 17 6.25 6.13 0.04 5.55 5.60 0.13 3.53 6.00
July 0.03 0.25 8.20 0.08 0.61 10.18 0.68 3.51 2.43 0.03 6.41 6.48 0.04 2.90 6. i6
Au iJust 0.01 5.43 1.10 0.01 0.45 9.16 0.39 4.02 5.62 0.06 3.11 6.37 0.01 2.35 5.111
Septeuber 0.03 0 3.02 0.02 0 1.61 0.49 3.15 2.92 0.01 3.01 3.46 0.13 3.33 3.00
October 0.21 --- --- 0.22 5.13 2.80 0.58 6.24 3.51 0.03 3.05 2.30 0.01 3.29 2.38
Total (lackas)0.95 28.99 25.24 1.26 39.02 58.01 6.81 40.4? 35.1? 0.2? 35.22 29.91 0.14 35.118 3?.
on
L. vs. per .0.0.11 -.- •-- 0.13 --- --- .68 --- --- 0.03 --- 0.082
All units In keasure4 teackate Fl 1 1 es/u.o.
t Me data avalIa • I Measured precipatatlen. Iisclies/.o.
I I t$.ated II Meuured preclpatattuii. iecIse /.o.
p.
-------
APPENDIX C. MUNICIPAL LANDFILL LEACHATE
CO POSITICN
1’ 3L C.1. MUN1C1P I. LAZICFILL L! AC.4ATZ c3 1PO IT1C1
(Values in nq/1 exCIOt as
ueir t .r
HUqhCS at 41.
( 1571r —
81acX ..11
iiag es etjL .
(1971)
L 53 Cupage
gnes nl.
(197TT —
LA 63 Cu qe
P fl1and
c t l
(1975)
.ii
Age f efusa
initial
6 years
17 years
i year
Iy . of Study
landfill
1 ndfi1l
1 ndf i11
FTC
4n ti t *fltS
9.300
003
19.680
3, G0 (sic)
.0 (sic)
5 .S10 (sici
14.080 (sIc)
225 SIC)
rac
(p 11 unit)
‘3
i.502
(i i11ivo1ts)
IS
‘ ‘
5,794
1.198
1
155
,2 3
802
1S5
ris
F5
I
a. Cun4. ( ias1c . )
1.500
ATh. as CacO 3
I,Z5 5
. 1 5 9
1.3)1
1,575
as
7,330
2.230
Tat 1 P
0—?
3.90
Q —inor.
l$4
6.4
1.70 nitrate
Org...’ l
0.70 nitrate
1.50
T1 (
F
C .
:
Mg
41
900
130
.
:
o dU
2
2
C l
1,697
1,330
135
143
F (totel)
•
6.500
. Ô
5.3
U. s6
.5
0.06
292
10
M n
‘
in
Cu -
u.
‘u. o
0.13
<0.05
<0.05
<0.1
-------
8LE C-i. UNtCP. t. LMI0 tLL LE. ATE cP:r 0 i
(V ii uss fr lTq/1 •xca s
Resear er
?onhind (1975)
contr 1 csIl
fund (1975)
c ntroi ci ii
Ministry
(1961)
4erz
(1954)
cu 0f R.f *.
2 years
3 years
ri.1.
1.5
T yp, of Study
rrc
F TC
T indfl l l
FTC
C in s :ltu.ntz
I_______________
COO
3,490
. ,l75
aop
3. . c0
1. 337
5.49 1
31—33,100
TOG
I . 98
1 .320
L
(p unic)
5.3
s.3
7.4
5.6—7. 9
iii111vøit )
f
;
rz
r o
7 55
340
36
YSS
200
1 5
rvs
FS
So. CGnd. nnos/ ,)
1—
Uk. u Cai .0 3
7th)
730—9.900 I
s cica.
;,; z
•
950—3,120
?
0.2-29
a—?
• nor.
4$. 4
Lj
od
0.2-690
$0 3 sMO 2 - I
Orq.. 1
34
U. .b
25
T4 .Q
101
T N
C i
150
L12
248
l1 5—3 . 70
ul I
22
20
Mi
so
15
35—1. 505
K
sO
I
Ci
200
30
1845
‘ ° ‘
Fe
co
6.5—3 6
Mn - I
4.5
5.40
Zn
13
18.7
Cij•
Cd
-Th
1
-
5. .
.
Cr
I
•
7ur . ( . acxsan un ts)
421 F produced from
est available copy .
-------
r. aL! C-I. IU. ’ 1CP .L L I0FILL L C14 TE CC.MPC ITICl
(Va’ues in . gj1 axceot as oted)
es.irc .r
F .iic n (157 )
c:ntr 1
Con (197 ) LC n i1
( 7)
I1 * 1niver lty of
f o 1
e nnar a ’i
(1971)
MUNC
Ag. of efusa
2 y. ri
2 yuri
0.25 years
0.33 yurs
Ty . of Study
FiC
F7C
I L ndf111
t. Oor ry
1
Constftuent
0
3,250—20.300
22.700—89.520
49.300
71,580
3CQ
3.250—19.200
15.500—33.300
24.700
67.00
TOC
17,360
27 ,700
fl fp s sni
3.5—4.5
5.53
5.57
(U11lvoIt )
40
32
23359
53.348
5
7. ’14,35
I4. 0C —21.3l0
T S
s—zoo
22-400
129
202
“ss
rvs
J ,
9.3a6
22.395
So. Conc ( nnos/
i
13.700 16,300
41k. Ci 3
240-3320
I
J
—,
4ir . u CiCO,
I
3—2.8
2 .3
98
—?
15.3
29
1’Q •frtor.
1 —i
0—31
154— 4 80
392.5
028
3 ’ 2•
arg. -
0—0.50
3—d.34
‘3.5
10.29
345
545
i IQl
C i
20—1,332
551.1,300
3.1 90
3300
Mg
2.5 — 608
316—472
350
1,140
M a
0—33 •1
408—1,010
1,360
T o
Z 3
48.143
zso—;io
1.14.0
2 , 3 00
20-250
267.1,040
1,110
l ,; 3
C l
S—4S0
520—1 .5155
.480
Z, 4o1
e [ total)
2Z- ,Qa*7
ó —3GQ
Z 2CQ
I, 4p
Mn
-—-
Zn
a.5 —;5
U 4
370
Cu
<0.2-0.4
0.33
,75
C4 I
1 —<• 3 S
<0.03—0.15
0.1
0.375
Pb •
<0.5—1.81
‘o.o—zo
.25
19
0— ).013
‘
S ..
Cr •
8
0.5 —
N i”
Tur3 , ( 1ac son uts
13 j
rs
30
422 Er0 b py
-------
T 8LZ C.i. MWflCIP t. L.V 0FIU. LE . CiATE C .MP0SIT 0N
(Va u*s In ig/1 e . caot as iota)
2 asaart1 er
Funqar
(1971)
Ii
Qi Iin and
8ur ina1
Zanon, a1.
(1974)
‘ rouo n
C3wn
(1976)
Aqt of Rifus.
2 y*irs
0.33 ‘ean
1$ years
tiltial
Ty i of Study
L4borat ry
PTC
t ndff1 1
Laboratory
Const1t sanc s
j
1
coo
4 1 ,000 —5
1 .000
2250
3005
14,760-33.350
1
1 .QC
o .’ ao
( i rnit)
.7,.3.3
9.38..6.4 .8
9.0—6.9
( ni111vo1t )
T5
0— 2.300
21.140—59.200
2750
r o
. 1
2530
fS . 5
10 .26,500
127
yss
95
T’I S
11,383—32.250
1270
So. C nd . UnO 5/caJ
24.60—5350
950
:
A1 . as CaC0 .
3—9 , 700
10,530—2 0.3So
izso
1ar . as C.aC0,
7 5 0 . I3 100
1410
Tot ?
0.04..).13
•3
o—?
—,nor.
L27—154
0— 4 3 2
473—flc .a
2.1—5.9
0rg.. l
3.2-0.3
2 8 — . 4i5
0 —1.4
TXN
J —li7
3.4—9.2
16.5
C4
2790—4080
150.440
1q
1 75—420
54 .-ICO
91-495
li i
0—7.7 0 .0
584—1439
75.140
0 .37
1050.3770 1
33
Q4
2 6—450
615—1002
C l
4.7—2.340
951—2310 j
110-t7 —
Fe ( taI )
0.1,716
175—350
34—75
32
Mn
Zn
0—167
4.d— 3.3
0.38-3.260
31.39
3. -3.4
Cu T
3.010
3.7-3.1
C4
I_____
3
p .
‘g _
.
,.
.
1
0—0.316
Cr
I
0.01-3.02
N1
I______
Turn. ( . ac&son units)
Fiproduced from
423 [ t avatlable COpy.
-------
7 8LE C-i. U $tCIPAi. L 10FIU. LL C:- ATZ C0?IPCSITI0 1
(laluu In ag/i ixceOt is noted)
a a sasrtJ er
•1nhar an4
4a (1973)
oflanssn S
Carlsøn (1976)
Yggei.th
Johaniere &
C4rison (1576)
Iaranrod
4IqPt (1979)
C iii zo
Age of
0.33 years
3.5 year,
2.5 yeari
tiltIal
lype of S dy
Liborawry
Lindfilt
Undfl i l
iC
C nst1tuants
—— — —
o. . oU
3,425
3,L55
4,394
( tai(
2.C3
0C
1,:ao
300
; (pr uni 5 39
3.5
3.. :
5.4
( 111 v&t )
—____________
I
75 7 ,330
,160
3.160
r;
755
192
T,a
‘1 5.5
302
ris
,670
F5
3•375
s . (Ja as/cs)
I
s.. zo
2. 70
Aik. as CaC0.
-234
i4ard. as CiCO,
oui P
7.7
0 —?
35
I
r0 4 -tncr.
227
O i_,1 357.4
N00
I is.s
n
31
T iQ4
c
3iz
250
400
156 I
2L 223 1
14g
.zp j 54
40
42
Na
j 206
297
sas
K
9
137
214
504
77
100
100
30
I
C l
Fe (total)
474
91
370
234
340
50
M,e
•
15
Zn
I
C a
0.022
0.021
c
[
0.0009
0.0008
Pb
0.01
0.015
—
,g
li
Cr
0.00
0.17
1
i i ’
3. 2
lurO. ( ..icx ofl unIte)
,. ,. ,
424
I Reproduced from
best available copy.
-------
T a E C -i. MUNICIPAL LMOFILL UAC IATZ C POSITON
(Values In ugh exci t as o ad)
esurChqr
gII (1979)
Call 20
Ign (1979)
Ciii 20
Fungar iI and
Steiner (1979)
Fung r ii and
Stetner (1979)
i
I y.ar*
S y.sr
2 years
4.3 years
Type of Study
j FTC
FTC
Laboretory
L abcrItory
Cinsti tuints
coo
25.064
15,484
24,520
1
3005
15540
roc
.
H (p. uniC)
5,4
5.3
3.33 I
°
( vi1111volt )
.
:
r
I
T 3
“
VS
I
rvs
1
Sa. C nd. ( sunos/ ?
13.300
7.16 3
Alk. as CaC0 3
4J90
3. 130
I
i4ard. as CaCa
3.480
2.484
rocal ?
o—
14.3
o 4 .inor.
3.3
0
. * j_M
s a
212
185
1.3
$0 3 $M0 2 .
13
158 j 3.5
TXN
57
244 1
C a ) 1, i4O
453
Mg
I OU
Ha
600
704
71
5O
558
96
299
0
C l
777
366
35
Fe (toul)
775
1 , 3Q
306
100
Mn •
50
1
Zn
26
2.1
11.1
0.93
Cu 1
h O
C4 • M.Q.
Pb M D
I
Hg 1
5 ’ ,
Cr ‘ 1
M i ’
c l.ø
Turn. (JacI son unity
425
-------
T. 8L c-i. u itct t. LA WFttL LZAc: Arz c: :rrc i
(Values ?t igjl xc! t d Ottd)
.sear ar
Fuller (1973)
4.!.$. (1973)
Landfill A
4. .5. (1973)
Undlil1 3
.Z.Z. C1978
Landfill C
Age of Refuse
6 ‘onths
15 years
4 years
3 years
7y9i 1 Study
Uboruory
Landfill
Landfill
Landfill
C n ti tuancs
coo
QQ 6
T OC
21. 5 0
p2-148
unit)
.11
•5
05
T 3
VZ
115
F5
C . na. ( ffernsJ)
z, o— .soo
AT I . as
ard. as CICO
rotal ?
•37•9
0—?
PQ •tnor.
tH . M
10 .dSO
0 3 W4Q 2 44
Org. - it
.05—47
033-0 3
033 . 305
rx
Ca 90—27S
—
34.427
170—213
10.37
—
Mg
17.48.5
6.3-130
.4-56
a
20—72
32—75
6.7-220
‘
1 50—950
23—47
44-240
8-24
Cl
93-3,900
5
<542
5.415
• çtoul)
•
48—LZ0
3.5—18.5
0.2-20
0.38—18.20
Mn
o. —La
3—5
1.11.3
3.1942.30
Zn
0.1—3.4
0.2 4—L36
.Q3—0.7
Cu
i .3 ,
‘0.02
‘0.02
‘0.32
C4
•
14.0.
<0.03
<0.03
Pb
14 • .
<0.1
0.14—0.37
H g
•
14.0.
<0.0302
‘0.CC O3
0.304—0.007
3.3012-0.3318
S e
Cr
-
14.3.
<0.03
0.5 1—0.63
<0.33
i
. cx on ufl1t$
1 4. .
0. 4— .3. i
‘
3.37.1.51
426
I Reproduced from
best available copy.
-------
T. .BLZ C-I. IUNZCr? L L. 0FILL LL C:iArZ
(1è1 In g/1 s noted)
ov r 3nQ
Farqufler (1973)
C.3 £nqlneers SCI Zngine.rs
SCS nqineers
efu ye ar
Active
Active
Active
5 y FTC
Landfill —
L ndf1l1
Landfill
0—44,33Q
2,251-30 . 333
7g .423
1,4 10. 17 , 4 03
s—4a
9.390-25,5 34
fl 3-U,3QQ
5.3—11.5
3.37—5.11
3.34— 7.10
5,tQ0—7,3
5.54— 5.7
— 0— 17 , 3ca
1,;z3 .34,;z 4 —
,352— i , 33 1
7. 4O.11,595
ii o — ip
19.55
—
—
•1
I
—I
I
C. ca 3
tov—s,. j
s a — •h Q
— j• . —E —
.j —j
3 . 4 1 — i d
o—sso
1.7.5;3
0.33—12
0.37.11
1.2.36
o—ssa
0.031—11
61—752
5—31
1 0—2,3 30
—
162-1,705
105—251
34—57
0—530
37—324
T,5O5_2.IY.3
0—900
5 0 - 52 I
232-372
—
340-700
90-620
0-i ,330
60.390
1-200
1—140
j 0—300
.
74-1,126
0.8—L20
3 .647
111—1,100
I , .33
0.7—4.3 1
0.38-47
f
0.01—1.13
0.31—0.38
1
Q.001—0.Qi
0.001—0.31
0.001-0.35
0.36—0.32
0T—0.Q08
0.001—0.32
0.19-0.50
0.C05 . . 3. 1p
0.001—0.11
I
T5 ozs .2.o4
0.001-0.19
0.01—0.46
I_
3.2-0.7
0.06—0.27
0.22 .09
unitu t
427
-------
TA8LE C-h U CtPAL L lCFtU. LZJ,Q1ATE c os:iic
(VaIu.s in nq/1 exceot as noted)
as.ercher
5C nqin.ers
(1976)
S Ite 0
50 Engineers
(1976)
SIte 8
.i r1cn and E rlch and
Landon (1969) Landon (1969)
InItial 3 y*erl
Landfill
J
7S0.000
72 .X0 -
I
I
— 11,253
—
I
Age of Refuse
2 YCIII
ctIve
1yoe øf Study j
Canstituents I
7 j5Q ..l7,SS4j
I
24S—8.669
G0
TOC
wnt)
,C40—5. .00
5.3— .S3
5.7S . .42
fl i
j
E, (re111 volt )
r
n5
r Ss
4420—3,307
60- 4Z
43_7.503
44.479
1 53
TvS
I
, 5
3.
20- .3A0O —
.aso-;o.c o
£ . Cartd. ( u as/c t
Alii. as CaCO 1 I
larl. as C.aC0
Total?
— 0—3,323
0
T_________
J .3Z5
0.4 1 11
0 irior.
0. . a O— 1 4 8
38 0 . .7 19
1H 4 -l
Q. 4 7—LZ
o.z4-al.z3
Qrg.- 4
r s
94 — . 0
16—936
;a— zo
I
Ca
Mg
M i
171—224
96—106
6
162—462
750—1,130
I
a 6
C 1
1
10—35
4—814
2.000
1 .000
Fe (tot i)
66-475
6.4—35
1,000
M i ,
Zn •
13—1,550
0.32—27
“
0.01—0.65
0.016—0.4.8
Ca -
0.001-0.07
0.004—0.01
?b
0.17—0.31 0.02-0.33
o. 0l—0. 6 0.0003—0.16
I
51
0.01—0.33 0.002—0.21
Cr 3.09—0.29 0.06—0.13
0.07—0.56 0.08—0.21
Tur . (ac sofl un1 S>
t
[ eproduced from
428 [ st availab)e copy
-------
TA8L C.i. MUt1IC?A(. L NQF LL U.AC) ATE CCMP0SIT1O I
(Valuu tn itg/1 axceot s
.sear er
E nr1e1 nd
I
(1969)
pgar anU
L ngnuir
1962 a11
(1971)
•
-
A gar no
Linqmu r (1971)
1967 c&1 j
ncerion
Oorit usn
(1967)
Age of *1us.
15 yeari
<1 year
5.3 jean j
n.t.
Iy . of Study
Undfil l J
TC
Landfill
Conu ltu. nta
C OO
Z.C62
sao_
,
1- Oc
940
piI wilt)
7.1
i (m1l11vo1t )
•T3
r5
Z. 91
i
2.375
163
T S
vS
rvs
1,294
.
S . Cond. h oflO$/ )
. o
Alk. ac C aC0 .
Q2
14.3 71
9ard. 5
aU
73
Tot al ?
0.7 3
0—?
I
0.5
thor.
•*4 . I
18
5*3
Qrg.-. 4
0 O
4.4
rx
C
1
1 . 40
293
Mg
37
75
Ma
I
I
51
16*3
2.3
4
24
CT
2 O
225
557
320
Fe (tout)
1a
215
4.3
Mn
Zn
f
C .
C4
r
Hg
5 *
Cr
I
Mi
7uro. (.jIcxs
on unitu
429 best ava Iable copy.
Reproduced from
-------
A2LE C - ). MU.IXCI?AL. LANCFIU. LEAC lA7E CP0 IT1C I
(Va1ue in ritg/1 exceQt as ote )
ese r er
4yde
(1971)
Me icltry I
(1971)
j
Meic try
(1971)
—Z4—71
C. un y of
1.35 Angeles
(1559)
Age of ReI ase
n.f.
3 years
> years
n. i.
r 79 , of Study
C rtfl1tuent. s
Landfill
L ndfil1
Landfill
,
CCO
—s,aoo
.a
3005
io. oc
103
oc
P :2 Ufl1C
5•75
i
‘ -5
J4.900
13. C9 9 .190
T SS
172
220
,iss
I
rvs
I
r5
$ . nd. ( a no5 c )
-
1k. as CaC3,
9.3 60
3.330
3.2 6 3
4Ir . as
320
22.300
3.24
0.53
Total
3— ?
04• inor
Z 0
3
S
0 3 ’ 4) 2 .M
0rg.- 1
TX.N
,,zc
zis
;05
c. j
I 3Q
5, 1 14
Mg
130
1
60
353
550
,
j
520
9
53
s .s
o oQ
C I
710
4. 13
re (t t.si)
3
Mn
C .
‘
C4
?b
ng
—
.5 .
Cr
I
I
Jr . .. CXSOfl jfl1 )
430
I torn
-------
Cu
‘Pb
‘-F
3IS C.T. JN1CI7AL L. DF1LL L C AT! CCNPOSITION
(Values tn ‘ 19/1 sxce t s ot.d)
CsUrd er
.iginser ngtne,r cx qineer
(1978) (1978) (1973)
SIte 1 SIte 2 SIte S
— ,gThiiFr
(1973)
SIts 5
Aqs f
7 w, 20 so ‘ears
20 years
Iy e of S ,ijdy
Landfill Landfill
Landfill
Landfill
C 0 7789 723 1193 .t79 2
5005
0C 3440 315 Z0 3000
;$ (ps unit) 5.3 3.2 —
-—
(wfl1fy l
1 5
.
r 5
155
,
155
rdS
S . Cong. a nos/c ;
.lk. as CacO 3
I
4ird. ss C4C0 3
o.?
lump.
111
11.1
3
0.1
U..&
Qrg..N
c 0.4
.
nci
132
124
12 73
Ca
1171
6
36 103
Mg
I
Ma
I
:
504
220
170
IZQ
Cl
270
IZ
4T
e (v . talJ
li. e
a. u
u
.
U. Q
—
— ..
,,
cO. 03
0.0 c 0.01
< 0.01
-
‘ 0.20
0.4 0.2
c 0.2
‘ 0.30 < 0.30 1.30 ‘ 0.20
ui,,
.ac son snit
431
I Reproduced from
[ es available copy .
-------
1 8LE C-i. MU UCI?AL LANDFiLL LE C 1ATZ c P051TCN
(Values in ugh excs c s nCt )
C. unty Of I Zenone et l.
Los Angeles (i97iT
mendorf
5C igineer St. siary ange
(197 5 ) of I iuei
Sfte 7
Age 1 Refu*.
- • 13 yeers
25-30 jean
Typ, of Study
L andf il l
Landfill
C ns 1 tuenu
I
0
36.31
3005
—37.0C0
rac
I
1 C0
5.3—27.700
li (pM un1 )
3.3
7.1 —7.6
3.7.11.3
Th nitiiivo1ts)
3Z3.602
7 5
T 5
I
0—59,200
2.000
0 -44. , 00
ia—i .243
VSs
25—602
ris
1270—32.250
P S
5 . Cond. ( os/ )
I
360—16,300
AlIt, CEC3
.‘ Q
4ar4. as C4C
3 ,11
3—22,300
To 3i ?
3.0—.3.1
?
0_.35
0 4 —jncr.
_ 4
0
160
0—I .106
30 3 ’M0 -.3
0.29—0.32
‘ I
0—27.2
0.08—0.09
0 1,4 16
fl
-------
APPENDIX D
ADDITIONAL RESEARCH ‘TEEDS
An examination of existing literature addressing production
and management of leachate from municipal landfills has revealed
a number of important deficiencies. In order to identify these
areas, this section will (1) list those areas requiring addition-
al research, (2) suggest specific objectives for research in
these areas, and (3) reconimend generalized approaches to meet
these research needs.
Suggested research approaches will be designated by one or
more of the following categories:
• Basic research — identification of the processes and
mechanisms occurring in a sanitary landfill and their
role in leachate generation and transport.
• Bench scale - experiments performed within the labora-
tory, utilizing apparatus and materials which simulate
landfill conditions.
• Field studies — studies at actual landfill sites
which can be used in conjunction with bench scale
experiments to verify results or improve their
viability.
In addition to the suggested topics and methods of approach,
approximate costs and time appropriations considered necessary
for completion of potential projects are estimated.
A comprehensive evaluation of the literature concerning
leachate generation and assessment revealed a number of areas
where additional research is warranted. For purposes of clarity,
these research needs are presented in terms of the following
five subject areas:
• Leachate generation
• Leachate composition
• Leachate migration
433
-------
• Available control technology, and
• Environmental nit rir g.
LEACEATE GENEP.AT ION
An evaluation of the literature reveals that the water bal-
ance model was used as the major tool for leachate generation
estimation. Over a hundred techniques are available for leach-
ate generation estimation. by using the water balance principle.
Thus far, few techniques have been verified under field condi-
tions. The applicability, accuracy, and sensitivity of leachate
generation techniques are, therefore, greatly lacking. The
following research needs are suggested for the techniques deemed
promising. No priority is established for such research needs.
In our opinion, all such techniques should be compared by field
experiments. The hydrologic simulation model HSSWDS suggested
by EPA, which selects SCS curve number for surface runoff, and
the Penman equation for evapotranspiratiofl estimation, should
also be verified under the landfill conditions, since the data
used were mainly from agricultural lands, not landfills.
1. Determine the amount of water contributed by refuse decrada—
tion
Obj ectiVe-—
The water contributed by refuse degradation, although not
significant, can affect the refuse water absorbing capability and
hasten the generation of leachate. A thorough literature search
only revealed one set of data related to refuse biological water.
Additional research s suggested for gathering information for
d fferent refuse and environmental conditions.
Recommended Approach——
Bench scale--Construct laboratory lysirneters packed with
municipal refuse. Variables to be considered include refuse
composition, refuse age, initial refuse moisture content, degra-
dation temperature, and. other environmental and chemical factors
such as pR, nutrients, toxins, etc. Experiments should be con-
ducted at selected time intervals over a period of one to two
years. Data gathered can be used to estimate as discussed
in the generalized water balance model (Table 2, Section 3).
2. Determine the ermeability of refuse
Obj ective— -
A review of the literature found that most of the refuse
permeability experiments were performed under bench testing at
refuse densities of less than 700 pounds per cubic yard.
field conditions, the refuse density is usually in the range of
800 tO 1200 pounds per cubic yard. Since the refuse permeability
434
-------
affects the rate of water percolation in refuse, such icrnatjon
is essential and should be obtained at conditions si nilar to
field situations. Due to the difficulty in conducting field
testing for refuse pe neability, only bench scale testing is
suggested.
Recommended A roach--
ench scale—-Construct laboratory lysimeters which are
fjlle jth refuse having different particle sizes (shredding),
and density (compaction and balIng) characteristics. Measure
permeability values and develop r mographs for specific refuse
physical parameters.
3. Deternjne the surface runoff characteristics under different
landfill surface conditions .
Objective--
A review of the literature indicates that the Rational Meth—
0 d and, Curve Number Method approaches may be the best techniques
for calculating surface runoff. However, the existing informa-
tion for these two methods (e.g., runoff coefficients, antecedent
moisture contents, hydrologic soil groups, runoff curve numbers)
are mainly derived from areas other than landfills. In order
to make accurate estimations, information from actual landfills
should be obtained. Due to the wide varieties of landfill sur-
face conditIons (e.g., precipitation, soil type, surface slope,
vegetation type), both pilot and field testing are suggested.
Pilot testing can cover all the variables which are not easily
obtained in the field testing (e.g., a wide range of selected
controlled precipitation). Field testing can be used to verify
the pilot data for larger landfill areas.
Recommended Approach--
Pilot scale——Test cells can be constructed in the laboratory
or in the landfill. Variables to be considered should cover all
the factors discussed in the main text for each specific tech-
nique (e.g., Rational or Curve Number Methods). Tables or
i omographs should be prepared to cover all the possible landfill
conditions.
Field scale——Construct test plots in selected landfills to
study surface runoff. Methods, as described in most hydrology
literature, can also be used for setting up the field test site.
The testing method as discussed under “Surface Measurement”,
(Section 3 of this report) can be followed. Artificial rainfall
frequency, duration, and quantity can be used to facilitate the
tests. It is suggested that experimental factors be presented
jn a matrix or nomograph for interpretatjo of test results.
Experimental variables should covar most of the possible
435
-------
landfill conditions. These data should be compared to the pilot
testing results for verification or refinement of the experi-
mental approach. Data deemed reliable can be incorporated
into existing models (e.g., HSSWDS).
4. Determine eva otranspiratiOn rates under different landfill
conditions .
Obj ective——
A review of the literature indicates that both the
ThornthWaite and Penman approaches may be used satisfactorily
for landfill conditions. However, data obtained by both
approaches were not generated from landfills. Some important
variables (refer to Section 3) may also be omitted in both
amproaches. Since the evapotranspiration estimation is one
of the most important steps for the water balance calculation,
data generated, from landfills for evapotranspiration are
considered most necessary for accurate leachate generation
estimations.
Recommended Approach--
Pilot scale-—Pilot scale testing can be conducted by follow-
ing the set—ups used in hydrological or agricultural fields.
For ease in cor.troliing the experimental variables (e.g., soil
type and moisture, vegetation type, temperature, humidity,
radiation, etc. as discussed in Section 3 of this report),
lysimeter testing may be desired. Statistical analysis should
be used for the data evaluation.
Field scale——Field scale testing for evapotranspiration is
very difficult. However, in connection with other testing (e.g.,
surface runoff, infiltration, and soil moisture changes versus
time), the evapotranspiration cart be estimated. Such data can
then be used for correlation with other evapotranspiration
factors (e.g., temperature, soil type, radiation, etc.) by
statistical means. The data should also be compared t the pilot
testing results for verification or modification of the experi-
mental approach. Tables and ncniographs should be prepared to
represent most landfill and environmental conditions. The
effects of additional variables on evapotranspiratiorl obtained
in this research may be used to derive a modified technique
(from either Thornthwaite, Penman, or other methods).
LEACHATE COMPOSITION
While considerable information has been gathered regarding
landfill behavior and leaching patterns, efforts to model the
quality of leachate have revealed some information deficiencies.
Though the models reviewed in Section 4 demonstrate promising
436
-------
predictive capabilities, their application to field-scale
situations requires a broadened empirical base. Aside from the
empirical support needed for modeling efforts, some basic
information deficiencies are evident regrading the viability
of microorganisms in landfills and leachate, as well as factors
governing the leaching of heavy metals from solid waste. These
research needs are swnmarized below.
1. Determine the erisistence and viability of enteric bacteria
viruses, and parasites jn municipal landfills and municioal
landfill leachates .
Objective-—
Recent evidence has indicated the presence of bacterial
species in 9 year old landfill leachates, suggesting that even
older landfills pose a potential public health threat. Though
several comprehensive investigations on the presence of viruses
in leachate has shown viruses are rarely present, the data base
should be expanded. The presence of parasites have yet to be
studied. The objective of this research is to identify the
public health threat from enteric bacteria, viruses, and para-
sites in solid waste and sewage sludge/solid waste codisposal
leachatas from older landfills.
Recommended Approach- -
Field scale——Perform a field survey of 20-30 municipal solid
waste landfills throughout the U.S., collecting refuse core
samples and leachates for microbiological examination. The
landfills should range from 10 to 30 years in age. Some of the
landfills should represent sewage sludge/refuse codisposal oper-
ations. The microbiological examination should include major
indicator groups, such as total and fecal coliform and fecal
streptococci, as well as select viruses (e.g., poliovirus) and
parasites (e.g., Ascaris) .
2. Determine the importance of ionic strength, chemical activity,
com lexatjon, adsor tjon, and precipitation on the removal
of heavy metals from landfills by leaching .
Obj active—-
The concentration histories of heavy metals in leachate from
several studies indicate that the leachi.ng behavior of heavy
metals does not follow the pattern established by other refuse
constituents. The mobility of heavy metals is known to be
influenced by various physica1/chemi a]. phenomenon which are
mentioned above. The objective of this research is to identify,
and if possible, quantify the effects of each of these factors
on the mobility of heavy metals in landfills.
Recommended Approach- -
437
-------
3ench sca1 — —ConStrUCt a series (4 or 3) of lahoratory
lvsimetar columns. One column should contain only municjoal
refuse, while the others contain refuse plus varying nounts of
sludge containing iron, zinc, chromium, copper, ca&aium, nickel,
and lead from electroplating industries. Subject each column
to equal moisture regimes. Collect and analyze leachates as a
function of cumulative leachate volume. Identify metal species
and organic and inorganic ligands. From the data, explain the
presence or absence of metals in leachate in terms of equilibrium
concepts modified by consideration of ionic strength and con—
plexation.
3. Determine the potential for leachate organic compounds to
cornolex with heavy metals and the extent to which metal
concentrations can be increased above normal solubilities
by cornplexation .
Obj ective—-
Determine the stability formation constants of organo-
metallic complexes based on the organic compounds likely to be
present in leachate.
Recoxtznended Approach-—
Bench scale——Collect leachatas from several active municipal
landfills of various ages, and identify types and levels of var-
ious organic species in the leachates. Fractionate the organic
phase into their respective molecular weight fractions using
ultrafiltration techniques and gel permeation chromatography,
as described by Chiam, et al. (1976). Determine the heavy metal
distribution in each molecular weight fraction. Develop sta-
bility formation constants of the organometallic complexes based
on thermodynamic equilibria and the empirical data gathered in
this and previous studies.
4. Determine the limits of industrial waste heavy metal loadings
that can be sustained during codisposal without deleterious
effects on leachates on the biochemical processes of stabili-
zation. Also, identify those landfill conditions or
externally added mediators which are capable of minimizing
the deleterios effects on microbial populations from added
metals .
Obj ective-—
Identify the industrial sludge loadings at which the quality
of municipal landfill leachates is impaired, and whether combina-
tions of sludges and refuse can alleviate these effects.
Reco nended Approach- —
Bench scale-—Construct a series of laboratory lysimeters
packed with shredded and unshredded refuse and seed with heavy
438
-------
metal—laden indi trial sludge in various increments. Dete ifle
the sludge/refuse ratios resulting in increased rates of metal
appearance in leachate. sludges should include common industrial
metals, such as cadmium, copper, nickel, lead, chomium, and zinc.
4onit rir1g of 1ysin o qas, and 1e chate Ia ssen—
tial to evaluate the effects of added sludge on stabilization
processes. Lysimeter testing should also include combinations of
sludges for possible synergistic effects.
5. Dotnrmine the hydraulic pro erties of municical solid waste
to assistmodel develootnent efforts .
Obj ective--
Investigate the hydraulic behavior of specific components
of solid waste to identify the influence of solid waste composi-
tion on moisture movement. In this way, the hydraulic properties
of specific landfills can be more accura ely characterized.
Reconmiended Approach-—
Bench scale—-Construct a series of laboratory lysimeters and
fill with varying proportions of moisture holding materials
(e.g., paper products, rags) and impermeable material (e.g.,
glass, metal products). Identify the hydraulic relationships
among these materials and the macro—void space associated with
them. Measurements of moisture content and moisture flow within
the refuse would supply information that may directly support
formulations of refuse hydraulic properties. Also observe th
determinants of field capacity in the lysixrteter.
6. Investigate the recise nature of the solubilization of
contaminants in solid waste for model develooment .
Obj ective—-
Develop a more broad empirical base for the generation of
specific contaminants, (e.g., major organic indicators such as
BOD, COD, and TOC, and inorganic containiants such as nutrients
and heavy metals) from solid waste. By garnering more empirical
data, description of specific contaminant generation through
models is better supported.
Recommended Approach——
Bench scale—-Fill a series of tall (20 ft) laboratory columns
with municipal refuse of varying density and composition. Place
intermediate cover soils between successive lifts. Apply differ-
ent moisture regimes to replicate columns. Collect and analyze
leachates for major organic indicators, nutrients and heavy metals
over a period of 4 to 5 years. A series of curves plotting
concentration and mass removal as a function of cumulative leach—
ate volumes and time should be developed.
439
-------
7. Investiçate the occurrence and behavior of ref e and leach-
ate microbial ooulaticns to better ur.derztand their im act
on leachates and landfill stabilization .
Obj ectiv --
A major proc zuodel described in Secti.on 4 assuxne mi-
crobes in solution, and described their action using the Monod
Model. Experimental work is needed to verify or justify these
assumptions.
Recommended Approach-—
Bench scale-—Fill a series of laboratory flasks with land-
fill leachates (or synthetic leachate) of various ages. Spike
each with volatile fatty acids (e.g., acetic acid, propionic
acid) and a bacterial seed from a landfill slime layer (at
bottom of landfill). Determine degradation rates of the acids
and cell concentrations at various intervals. Determine whether
Monod Xinetic Model adequately describes substrate utilization
in the landfill environment. If not, suggest other models of
microbial activity for use in process modeling efforts.
LEAC ATE MIGRATION
A review of the existing literature indicates that addition-
al research is required for the development and evaluation of
leachate migration models which can be used to predict the extent
and rate of biodegradation. Additional studies are also required
to determine the effects of flow rate (or solution flux) and
soil chemical and physical properties on the predictive capabil-
ities of leachate migration models. Such efforts may lead to the
development of a simplified model which can be utilized by land-
fill designers and site operators in estimating the attenuation
of leachate constituents. These research needs are si m arized
below.
1. Develop a leachate migration model which incorporates a
biodegradation rate constant which can be used to predict
the extent and rate of bioloqical decomposition .
Objective——
Numerous models are available for predicting chemical and
physical migration; however, the integration of biodegradation
models with mass transport approaches are greatly lacking.
Additional research is required for the development of bio-
degradation models which would be of significant value in predict-
ing the migration of biological contaminants..
Recommended Approach——
Basic research——Select several potential biodegradation
models (e.g., Monod, Teissier, Contois, Moser, and various
440
-------
ccmpetitive/nor.—cQrnPetitive inhibition forms) and evaluate them
in terms of various fluid velocity and mass transport approaches.
Combir.e the most Dromising combinations of mass transport and
biodegradation models which would rovide the most definitive
predictive Capability.
Bench scale—-Construct a laboratory test column containing
a soil having a low orcanic content. Operate the column anaerob-
ically under continuous saturated flow conditions. Aerobiosis,
may be achieved by sealing the system and, flushing with nitrogen.
after each feed addition. Apply a Continuous flow of leachate of
known composition to the column and measure selected biological
parameters at various time intervals. Determine by statistical
methods the predictive capability of different models conthina—
tions for various leachate strengths.
2. Determine the predictive capability of single—ion and multi-
ion simulation models for different flow rates and constizu-
exit concentrations .
Objective--
Leachate migration models, in general, are developed by
selecting the flow rate (or solution flux) as an independent
variable because available data indicate that attenuation is
sensitive to solution flux. The application of such models for
different flow conditions is largely unknown. Additional re-
search is required to determine if parameters estimated at one
flux are comparible to those at another flux. Model validation
over a range of flow rates is considered essential for practical
applicability.
Recommended Approach—-
Basic research--Review the literature and identify those
single-ion models which may be applicable for integration with
a mass transport equation. Models which may be used for a
group of single or multi-ions should also be identified. and
evaluated for compatibility with mass transport equations.
Bench test-—Construct laboratcry lysimeters containing soils
of varying permeabilities (e.g., clay, silty loam, sandy loam,
sand). Operate the columns under continuous saturated flow
conditions. Apply leachates of known composition and measure
the effluent concentrations of selected. water quality parameters
(based on ion specific model) at various time intervals. Deter-
mine the predictive capability (at selected confidence levels)
for different flow rates and solution concentrations. Based on
the results of this study, modify (refine) the model(s) for
ixaporved predictive capability.
3. Develop a simplified model based on soil key Parametric
I dicators. Assess the effects of such simplificatj n upon
He predictive capability of the develo ed model .
441
-------
Ob j ective——
Most leachate migration models require complex a proaches
which are both unwieldly and impractical. Therefore, addition-
al research is recci tended for the development of simplified
models based upon soil parametric indicators. Such development
is essential for increased utilization of predictive migration
models.
Recommended Approach-—
Basic research——Review the literature for data correlations
between soil chemical and physical properties, and model develop-
ment. Construct a matrix or nomographs for data interpretation
and selection of soil properties to be evaluated in bench scale
testing.
ench scale——Construct laboratory lysimeters containing
soils having a range of chemical and physical properties which
are known to affect the accuracy of the model. Key parameters
may include texture (clay content), content of hydrous oxides,
type and content of organic matter, cation exchange c apacity,
and permeability. Operate the columns under continuous satu-
rated flow conditions. Apply solutions of known composition
and determine the extent and rate of attenuation for selected
constituents. Quantify adsorption and decay constants, dis-
persion coefficients, etc. Correlate these results with
applicable soil characteristics. Develop a simplified model
based on key controlling parameters. Verify the model over
a range of flow rates and solution concentrations for different
soils. Evaluate, stastically, the effects of model simplifica-
tion on the accuracy of the obtained results.
LEACHATE CONTROL
A review of the literature indicates that additional
studies are required to develop the technology and determine the
effectiveness of various leachate control measures, particularly
in the areas of overland flow leachate recycling, leachate
extraction and well injection, nutrient and alkaline industrial
waste addition, natural and synthetic sorbent materials, and
leachate recirculation techniques. Additional research in
these areas will provide for alternative leachate control
approaches which can be utilized by landfill designers and
operators. Research needs are summarized below.
1. Investigate and evaluate the efficiency of overland flow
leachate recirculating systems in reducing the organic
loading of landfill leachate .
Objective—-
Operational overland flow leachate recirculating systems are
not reported in the literature; however, overland flow
442
-------
irrigation is an inexpensive well-developed technique. Research
is suggested for the development of such a system which may
represent a viable means of treating recycled leachate.
Recommended Approach-—
Field study-—Construct an overland flow spreading basin at
a sanitary landfill facility which is eguipped with recycling
capabilities. valuate the time—dependent removal efficiency
of biological indicators (e.g., COD, TOC). Relate the results
to the rate and frequency of recyle, and soil cover material.
Estimate the rate of landfill stabilization. Assess feasible
design configuration and operating procedures together with
associated costs for implementation.
2. Investigate the removal efficiency and effectiveness of
a leaèhate extraction and deeD well injection system .
Objective——
Very little information is available in the literature
concerning leachate extraction and deep well, injection. Addition-
al research is required to perform detailed hydrogeologic
investigations to ensure that leachate is contained and does not
migrate to cause contamination problems.
Recommended Approach--
Field study-—Install a recycling and injection well system
at an existing landfill which is producing large quantities of
concentrated leachate. Determine the zone of influence for
various pumping rates. Install a cluster of monitoring wells
(the number dependent on the pumping sphere of influence) on
the parameter of the injection well in a number of concentric
rings. Monitor the migration and removal efficiency of trace
metals and other applicable water quality criteria.
3. Determine the effect of injection of major nutrients (i.e.,
nitrocen, phos horous) on the biological stabilization of
leachate constituents .
Objective-—
Available information in the literature indicates that
phosphorus may be limiting in the stabilization of landfill
refuse. Additional research is required to determine what
nutrient ratios are required to restore or increase biological
stabilization of leachate constituents.
Recommended Approach-—
Bench scale-—Construct a lysimeter test cell containing
refuse front a landfill exhibiting a slow rate of stabilization.
Develop and maintain anaerobic conditions within the cell.
443
-------
Inject varying amounts of nutrients (e.g., nitrogen, hosphcrous)
and measure the BOD_: :P ratios at selected time intervals
to indicate a possi le limiting nutrient concentration. Deter—
mine what nutrient additions are required to restore or increase
refuse stabilization as measured by the rate of reduction of
COD or TOC levels.
4. Determine the effects of codisposal of municipal refuse and
alkaline industrial wastes on the biodegradation of creanics
in landfill leachate .
Obj ective—-
Because of the high moisture content and alkalinity of many
industrial sludges, the inclusion of alkialine wastes may prove
beneficial in accelerating refuse stabilization. Additional
research is required to define the applicability and provide
quantitative inforamtion to optimize this process.
Recommuended Approach--
Bench scale-—Construct lysimeter test cells and fill with
municipal refuse. Add varying amounts of alkaline industrial
wastes for p I adjustment. Dete iine the alkaline waste/refuse
ratios resulting in increased refuse stabilization rates as
indicated by reduction of effluent COD and TOC concentration
levels. Estimate the effectiveness, application range, costs,
and practicability of this method.
5. Determine the effectiveness of various natural and s thetjc
materials in removing conta.rninants from Leachates .
Objective-—
Laboratory studies have demonstrated the effectiveness of
various sorbent materials in removing contaminants from leach—
ates. The results indicate that combinations of different
sorbents can be used to increase the attenuation of leachate
contaminants. Placement of sorbents in landfills as a bottom
layer to provide substantial contaminant reduction prior to
migration from the site may exhibit economical feasibility due
to the reduced contaminant load imposed upon the Leachate
treatment facilities. Additional research is required to define
the applicability and optimize this approach.
Recommended Approach--
Beach scale——Construct laboratory lysimeters which are
filled with a layer of various natural and synthetic materials.
Selected materials should include bottom ash, fly ash, vermi-
culate, illite, ottowa sand, activated carbon, kaolinite,
natural zeolities, activated alumina, and cullite. Apply
leachates of known composition and operate the column under
continuous saturated flow conditions. Measure the effluent
444
-------
concentration levels of trace metals and organic indicators at
various t ne intervals and determine the removal efficiency
of s ecific sorbent materials. Based on the results of this
preliminary study, select sorbent combinations which may be
more effective than individual sorbents in reducing the concen-
tration levels of contaminants. Verify the removal efficiency
of selected combinations of sorbent materials by additional
bench-scale testing.
6. Investigate the effects of municical and industrial sludge
additions to rnunicioal refuse on leachate concentration
histories durina leachate recircuJ.ation .
Obj ective——
Laboratory and field test cell investigations have demon-
stratec. the benerits of leachate collection and recirculatjon
back to the landfill: the rapid decline in leachate organic
concentrations with time of leaching, a delay in the starting
time for leachate treatment, and a leachate volume reduction
through evapotranspiration. The objective of this research
is to determine the adverse or beneficial effects of numicipa
and industrial sludge addition to refuse landfills subject - -
to leachate recirculation.
Reconm ended Approach--
Bench scale——Construct a series of 5-6 laboratory coli mns,
and provide a leachate recirculation system for each. One
column would be filled with municipal refuse only, and the
others refuse plus varying amounts of municipal and metal-laden
industrial sludges. Subject each to identical moisture regimens,
and collect periodic samples of recycled leachate for analysis.
Analyze for general organic indicators (e.g., COD, BOD, TOC
volatile acids) as well s heavy metals (e.g., Cu, Cd, ‘e, Cr,
Pb, Ni, Hg, Zn). Evaluate the effects of sludge addition on
leachate recirculation and refuse stabilization.
ENVIRONMENTAL MONITORING
An evaluation of the existing literature has revealed some
important deficiencies with regard to the performance of passive
vadose zone monitoring devices, obtaining representative samples
for analysis by use of pressure/vacuum lysimeters, and the
potential environmental impacts of surface run—off from sanitary
landfills. Additional studies are required in these areas to
provide for the implementation of a more effective monitoring
network. Research needs are suxmnarized below.
1. Investigate and evaluate passive, non—sampling vadose zone
monitoring devices underlying a sanitary landfill .
445
-------
Obj ective—
Passive, ncn—s pl g vadose zone m itorin; devices have
received limited attention which reflects, to a large exter.t,
the complexity of flow in the vadose zone compared to saturated
flow. Also, these devices experience hysteresis effects in
repor.se to wet-dry cyclas and require frequent recalibaration
depending on exposure conditions. Additional research is
required for the development of improved technology for monitor-
ing in the vadose zone.
Reco ended Approach- -
3ench sca --ConstruCt laboratory lysimeters which are
filled with a layer of various soils (e.g., clay, silty loam,
sanding loam, fine sand) , overlaid by municipal refuse, to
simulate the refuse/soil interface at the bottom of a landfill.
Va the time sequence of water addition to simulate various
wet/dry cycles. Calibrate the vadose zone monitoring devices
against a range of soil-water contents (i.e., soil-moisture
curves). Assess the sensitivity and accuracy of available pas-
sive, non—sampling devices (e.g., tensiorneters, psychroizeters,
moisture blocks), with regard to their response to varying
volumetric flow conditions.
9e ch scale——Construct a laboratory lysiineter and fill with
a soil having a moderate permeability (i.e., silty loam, sandy
loam). Emplace moisture blocks having different encapsulation
media (e.g., gypsum, fiberglass, polymeric resins). Flush the
lysimeter with high ionic strength leachates to simulate the mass
rate of application encountered in a 3 to 10 year period.
Evaluate (by visual observation and weight-loss measurements)
the effect of the applied leachate upon the lead wires and
encapsulation materials. Assess the simulated time effect of
the applied leachate on the accuracy and sensitivity of the
respective devices by comparison of calibration curves obtained
at selected time intervals.
2. Determine the validity of oressure/VaCUUIfl lvsimeters for
obtaininc reoresentative in situ vadose zone leachate
samDles. Identify water quality parameters which maybe
attenuated during oassage through the Dorous cup .
Objective-—
A significant problem in vadose zone monitoring is obtaining
representative leachate samples. Limited laboratory studies
indicate that passage of leachate through the porous cup of a
pressure/Vacuum lysimeter will result in attenuation of certain
water quality constituents. Additional research is required to
determine the extent of such attenuation for various porous
cup materials.
4”
-------
Reccz ..Iflended A orcach--
Bench scale-—Construct an experimental test cell d em lace
press e/vacu lysimeters constructed of different porous cup
materials (e.g., cqrarnjc, PVC, teflon). Pass a raoresentativ
ni .ber of simulated or collect d 1 ach t s. mpl through the
selected porous cups. Identify, (by statistical methods),
applicable water quality parezneters attenuated by the cup mater-
ial. Relate the results to the feasibility of obtaining
representative in situ leachate samples.
3. Investigate the enviroiunental imPact of surface run—off
from various municip l landfill oPerations .
Objective——
Surface runoff from landfill sites may result in contaznina-
tion of surface water supplies. Quantitative information is
generally lacking in the literature. Additional research is
required to determine the effects of ty e of cover soil, landfill
surface conditions, vegetation, and precipitation on the genera-
tion and composition of surface runoff.
Recommended A proach-—
Field studv—-lnstall surface runoff collection devices at
selected landfills. Select several key monitoring parameters,
e.g., organic indicators, heavy metals, pathogens, and nutrients,
and study their concentration and flow rate of surface runoff
waters from area fill and trench disposal operations. Perform
routine and synoptic storm related monitoring. Relate the
results to the tyoe of refuse, landfill surface conditions
(e.g., slope, type of cover soil, vegetation), and intensity
and duration of input water.
SZ MNARY
The additional research needs for leachate production
and associated areas are summarized in Table D-l. The specific
research objectives as listed include estimates of the period
of performance necessary for project completion as well as
corresponding costs. The subjective cost and performance time
estimates are based upon anticipated man—hour requirements and
capital investment for the recommended approaches to the
specific objectives as presented in the previous text.
447
-------
T*UI 0-1 TINAtUI TINl MW C0 f t 1UIIItU$tt11 ION Ctl*4P1. 1T1011 4)1 Itrry.wr,Aa. petu ci
jr4LM) of E et4a.4Lcd Cost
Ka&iuarch Araa AI.Ib u4ch I urtor ai . i i •Ooo
I.oa hate Gencrat Ion
Duturmino tIW OUlIt Ut WAt . )(
conttibuted by 111(138 . ) du taI.L*o,i. Conalluct laboratory Iyiibautui cusitoiiiin.i I - 2 yuar. i0U- ISu
i fu e tri. ru rusun atLvu lanJttlIa. Quantify
thu 1384013 1,1 of 1381111 9 .3 134 .181101 1 wIth retu u
COI8(40* Itti0Ib. 113(41811 49U 1*1111131 U13 1 181 11 1U C013
to n I 4 11384) . )ratur . ). 1.11. nulrlonta. to*iua . 41t0.
tlnturtno pur. . abtItty of
municipal cotu.).). Co uL1uct laboratory tyotmutura which ur tili. .d I yoar
with [ 13(13.113 Iiaviinj dltfuiu, ,t f .articl . ) ulzo .i
• and .l .iu .i&ty (cuaq .utt 1 . 3 (4 404 hal iI4 j
clI4rdcLorIlltIc.1. Hcauuru P°” Y value .. 40(1
dovolop jraplia f..r .)r.oci tic ratu .o i.bytilcal
cx
bot . )rahIw . oufaca unotf
characterl.otlca (or ditfurent Cusu .truct l.oratucy or landfill *0.41 Ili. I year 141 4)
landilil surtacu conditiona. Nea uiu ah 1i4 .8ble igEAaIuLuru tot Icula-
I i i , . .t ur tacu runoff I.y u..u ul H*t
or CUIVU 4141841 )4 *1 N th J .
Co.iatiuct test pluto at uIe .ctu.1 luiidtiil . .. I to 2 yaar.. IU O-1SO
lolluw teutl, .’J actitodsi di ci ,t uud w .des urtacu
Huaaurcmug4t l ) 4oct Ion i of 11.1.1 rep.3rt . Itpply
Ii*nilatu ..I t Intu3I of varyluq tsc. uoncy. dU14tiU0
and intunaity. Cun tiuct a ..atrix or I . )o .Jrq9 .a
for lnteijrutatlo,i (it $u t 14311.1110,
c Heturminu .V84*atraa*lf)kr.t bit I - 2 ydora IOU- I SO
1814.0 undar ditfuront landfill Co,i..truct lab oratory Iyala*utera liii . )vaIodt lois of
condition.. ux aurionta) vat IabIu . . 113.9. • .n .il ty ,u And mola-
tUtu cOaltunt . Vtt’Jotat ion. tUlq1314 1U1 1 3 . hisaldity.
raddalloit). Utill o tatiotica1 a . .stbo(1a t.*&
8 44(4 11V4IUAL IL3I I.
tCOIsI 111134.41
a
-------
po*j j
I)oteraine the peraleteucu Led
viability at ontario bacteria.
vjrueeii, end poriiiiteg in
lendtIIt and landfill loechetea.
fleturine the laportance of
iouit rnuçjth, cc*.p1e mtlo .
adaiorption and prac*pltetion
oh the remuvel of heavy aotol
from lainIftilo by leochipiq.
Dutoratiw the abiIity
tor.at ion coiiota t at
organouluta ) lie coapluno beoud
on thu ocyania compounds fowul
in iuacbjtu .
l)uturainu thu induaitrtal slud’ju
loadings at which the quality
of landfill leaclintei Is i.peireJ
and whutliur cumbinat Ions at
uludyos can aflevietu ihusu
eftocts.
Fluid
Conigtruct lust pI at solucztud landfills.
Eatimete evapoirenspiratlon rOtue by use of
water balancu equntlans. Corrulatu dote with
oIlier ova Lranai iIra t io n taclorit (u.’j..
t..mperaturo, soil typ o 0 vugutelinu, radiation)
by statistical muttiods. Coeperu d Ca with
Lunch scale atudius end conatiuct a mains or
somograptie for in rutution ond application
at test results
Fluid caiu
Survey 70-30 municipaL aoLtd wastu lassitill .
tlizouqhout thu U.S., culluctintj rufusu cure
samples end luecIi tu ton miobtalagicel
examinot ton. £*amliiu omplus (or buciurlal
indicator giuu(ie , virusu . end pe .oiltus.
bunch. Bcol
Construct a eur1 of I bo, atu& y columns
cu .itaintsiq varying aeou,its of aui .icliwl
tufu u cii i uluctraplotlug iludgo. E pIain the
concaninetionu of heavy suloli in ( 1 w Iaa. .h .ateu
In terms at aquilibrina col icupto nodiflud by
cunuldurattuis of ionic triiii jth and cuasiplosaiion.
bun.. ) bcu lu
Culluct lueeh.st s fro.. suvural ectivu asnssilcipal
landfills of varlona ego end tracliunetu thu
urqaiiic phiasu into ruspoctivu iecui .sr weight
Crac llonii. Duvulop stability furmuilois
CoisutanLi of thu coepLusu basio.I Os th...rmo-
dynamic uquiiibria aid umpinical data.
butch tlu
Cosuirsici a siorios of tahursitury lysissoturs
nd till wiih lutussu and huavy—isolel Isidun
sludges in variou L,scrus.uu la. Uu Ius.si,su
silud’je/rufuou ratios rovulting Its iiicioaauJ
luvuisi of usulalsi 40 Iue hatus , .nd doiiga o..l
ratios of staIsi1$ atioii.
Nosaerch Ar.. SP ioroec I s re . Lad of
Eat i.ited Cost
0
em
V
2 yuero I S O
1 yuar 125
2 yoars 70 -
a - mouths 50 — 00
2 yours 70 — 00
(COil I ituod)
-------
?AIII.E f l — i Itosit iitucd
imrit)4 of
I’er 1ormanc
Li ttu aI ud t oiL
mm
0
0
Dotmrim. thu Iiydiaulio buh.vlor
of .pecific solid ws.te coupononts
to idantify thu Influencu of
solid wastu CO )O 5ItLOII On
.iotsLure .ovu.iuot for odal
davaiopu mnt.
Invautigat. thu .olublllzation
of conta. 1nant. i n solid wustu
to brosdun the eapirical baua
for aodul duvulopaunt.
invastiqats thu behavior of
rutuua and loachate aicrobial
pol)uldLtonI to da rains their
luspacta on iaschatus and landfill
utobii lzutlon for .iodei duvUlOp
ao OL.
Duiich St ulu
construct a sertus of laboratory culuans awl till
with varying prol Lions of isupoisuablu nd ..aisturo
holding rufusu serial. N. asuru autsisiro t.oIstun(
asid solature flow. and esitasatu fluid capacities.
I iiscIs Scala
(‘111 a unties of tall laboratory columnss wiLls
•unicillaL rufu o of varying duassity asid coisipoaltion.
A pIy different .aotu tu&e rssqiaus. Col*us L od
analyze iua icliatus 5 s,J sluvsulop cuncentrat ion
hiutory curvos.
B..,ncl s Scab
Using laboratory fl ka tinaa with machates and
spiked with a bacterial suoti . doturalnu (lie
ducjratlatloss rates of volatile acids sii dotuisine
cull concuntrstioiis at virlouv listuivile.
floteralasu whether the nonod Elnotic bio.iul
duacritsus ub tritu utilizatioss iii iii.. lassJtifl
usivirussanot
Ru uarch Arus
Ap i) iolch
U’
C)
luact iAte Hi ro t ion
Develop a luacliatu aigrat ton
odeI which incorporatus a blo-
do.jra .lmt ion factor.
6 auntIss 40 -
4 — S years Ion
6 suonthia 0 - J O
I year I SO
I - 2 yuars 100 - l O
Uauj. I4osiuarchi
Ausussi varluus lsiodeqralatlon lMwidIs toid evaloetu
lit torasu of applIcabIlity whoss i,s(urqrated wills
sass transport suotiuls.
Uosscts Scale
Construct test COIUS OS contali1Iu j low sjiadu uojl
and dutusalne iiredicitlvu capability to terse of
(Isa blodoyratlet Lou 1usd I catsir fin dl (let-usa (low
rates aisti luacluatu cmiconl rat bass.
ICulit I lows 1 I
-------
fl LS Il-i ConhlR*e4i
R...stch kt.a Approach Period of
i.tiasted Coat
Develop Lou specific module
which may be uae4 in leachato
migration prediction studies.
Develop simplified models
based on soil parametric
indicator..
leactiste Control
Litaminu removal efficiency of
icachate contamiflanta by overland
flow.
invest i9ace the removal efficiency
and effectiveness of a leachate
withdrawal aiid infection well
system.
lIjslc U. arclI
Review literature lidsotilylun specific
a.oduis which are applicable iii a mess tra,iepert
equation.
bench Scale
Construct lysimeters cuIIts iiIisj soils of various
permeahilitlea. Apply luachates of known
composition at saturated flow cunditluije.
£valuat the accuracy at the model tot different
flow rates ai4 solution conceut.mtiuus.
Basic Ilesuercl.
Review the literature jar correlations bCtweun
soil parameters indicators dial a o.lol tlevulopmui.t.
Deiscl. Scale
Construct laboratory lysimotors coiitainhiaj coils
of different chemical and physical piupcit lee.
Apply leacliates of knuwi, composition. Refine
the model bided on the tOut, results und
determine the if ct of slmp)itlcat lon on
predictive capability.
Construct overland flow plot at a sanitary landfill
equipped with recycling capabilities. lvalujte
removal efficiency of bioLO jical indicatois. Kel te
the reault to the rate and frequency of recycle.
auiil sail cover for treata&uiit.
install a recycti .j system at an uxfstiii .j laiidtill.
Monitor the isiajratiun and removal efficiency
of key tnd*c tor parameters tlsrowjhout tIns entire
system. install a moiiitorin.j well network on thu
perimeter of the ililuction woll. Determine ties
onu of influence fur various piisiping rates.
i -n
I-a
6 months
1 — 2 years lee - i s o
1 year RU
- 2 years 100 - I S O
I - 2 years 100 - 1 50
- 2 years 1 50 - 250
ico lit iiWe.I)
-------
TAUI.E I>— t (Cuist inauJi
tur Lad of
Par fots ancu
uaatuJ Cut.
i$t.oou,
flutaratilu LIi affect. at 1,.gul*oo
of aa Ot nutrientl an btaLoqtc I
atabtltzation In4
,,utrtant oon uiitrgttoit ..
flutu liltiW the ettuctit ot COitC& 0i3 .L
at CUIILCLII.t tufulu IRJ ikaltn
tadutitria.) wa teb on bto)a jtca)
btdbt I Izatloil.
U’
th t ai• Iilu LIsO uftuct.tvuiWuli at
uuiect.J earbent etar1uIe to
atluouat Iriq Luachtata co,itaustnant .
I}at. retilw the IdVercu Or
uttuct of .uaiclpel and 1, duetrLal
lIud’ju. A4JttIUnil La rutuae 1 n4-
tUtu aubjoct to )uachat.n ruetrcuia
t I .ii.
Iioil.h catu
Cun.trucL Lallolatory Iy.taeturi caniatntiwj retuau.
Naintatit gnaurabit cundittone wtthitit the te L
. uII. flutur*Inu the uttuct*voiluaa at nuttleilt
eddittuna on rufuuu utaht)Izattun aS U41UIUJ hy
the reduatton Lii eftluuiit UU or ‘hOC lovute.
hsuuichi g.eIc
CtiiiuttuCt teh.iratOly tyui*uturu cuotutnin j
rutuue o d va•iOuii aeot lIlLu at gikehtne
inJu&triat waCtac tar IU adiuuteuut.
Duturision mhIjItnu wa tu/rufu&u r4t10u
rneutttnq to ii iu u L (nfllui3 bilizatiun.
Uuiit h t cehu
Conutiuct lahoraLory %yut*utecu iuiitatiitii’j
dtttutei.t cortuuit eatertulit. Ap1 1 y
hecb. tuu Ut ki Uwit 1 .uu*t Iwi. ..rLnu
the rueuvet ufttctuiicy at trucU NiaLelu all 4
ur.jaiiIc tntttcatarai (or hiu4iv idu.t eurhalital.
Sclu t uu iheHt co h4ilJt tuna tot hiuiprüvod
attessuaL lu of coot ausi lillit a.
I U,l( hl I u
Ci.iautruct a aortae of IatQ(àtury culisaile
uqut .uJ wilts a tuechustu rucLrCt4tgLtuU ayetu .
I1I cutucue wills vaiytit ’J a .ouilLa of
sausitutisal a s s d tisduatriul alo 4 ju ussil,.J Oslo
coltisali au C control. Coltoet rucyctud
I achsgtu j uIk..ItcChIy (ui .siialyu lb. tvaIu. Lu
utO at eIud’ju 5 t.itt Irsis oh luachsutu
rucircul ut I oil.
* - 2 yuara 1 10 — 1511
I — 2 years tOO - 1 5 1 )
2 yours 100
a-,o
su e
u s°
‘C
&i)
cr 0 -
icula I isueuit
-------
?LMI-* 0-1 IContlausdi
s..arch Lisa Approaah Period at &stimaIsJ COal
P.rto r.anc (U_. 1 000I
I r umotd.a! N tOf !“J
i 5t
<0
1
a
liwosliqate ,id evaluate poasivo.
non- uaauipt taiij vadoso zone • monitor -
ing devices.
Ilututalila the validity of pressure!
vacuum ly.i.etors for obtaining
reprusuhtat*ve vadosa zone
leachats samples.
Investigate the impact of surface
runoff from active landfill situ,.
UusIcl• tics to
Cunstfuct ly moteta aol place vadose monituiAu.j
devices below simulated retueelsoll inter taco.
Assess sensitivity and accuracy lot various wet!
dry cycles.
ftant b 8c lu
Emplace moisture blocks of .Iitterent eawapsulallon
malarial, witlilsi lysisicturs tilled with neutral
pH soil. Flush with high tuuic strengths leachalo
and assess lisa sisul ted lonj term u< 1 u 1( issint
put formance.
Uuiichi ti.ele
Select commercially available Lysiumuturu construct—
cd cit various porous cup mjturlala. Pa u retire-
sesitatlue luachalu otut ion throusub Lisa porouS
cups to dutermine the attenuation cit waler quality
I sra1s et ura.
Fluid Study
Install suttacu runot7 coT luctioji devices a t.
selected landfills. Perform r utinu ai&d uyisoplic
stora related mossiteuincj. Relate the ru u ts to
the type cit COVO soil. landfill surface cous.litionsi .
vequtatlosi. isitcinalty. and duration of in t u it water.
La
U t
Li i
I year Ho
munt hs
1 year 50
I —2yuacs 80 120
------- |