PB85-181071
STATL-OF-THE-ART AQUIFER RESTORATION: VOLUME I. SECTIONS I
THRU VIII
National Center for Ground Water Research
Norman, OK
Nov 84
U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
NI1S
-------
EPA/600/2-84/182a
November 1984
PB85-161071
STATE-OF-THE-ART
OF
AQUIFER RESTORATION
Volume I. Sections I thru VIII
by
R.C. Knox
L.W. Canter
D.F. Kincannon
E.L. Stover
C.H. Ward
National Center for Ground Water Research
University of Oklahoma, Oklahoma State
University, and Rice University
Norman, Oklahoma 73019
CR-806931
Project Officer
James McNabb
R.S. Kerr Environmental Research Laboratory
Ada, Oklahoma 74820
ROBERT S. KERR ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
ADA, OK 74820
-------
TECHNICAL REPORT OATA
1Please read Instructions on the reverse before completing)
1. REPORT NO. 2.
EPA/600/2-84/182a
3. RECIPIENT'S ACCESSIOI^NO.
PB8 5 1 81 071 /AS
4. title and subtitle
State-of-the-Art Aquifer Restoration:
Volume I. Sections I thru VIII
S. REPORT DATE
November 1984
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
R.C. Knox; L.W. Canter; D.J. Kincannon; E.L. Stover;
C.H. Ward
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
National Center for Ground Water Research
University of Oklahoma
Norman, Oklahoma 73019
10. PROGRAM ELEMENT NO.
CBPC1A/CCUL1A
1^. Aact/grant no.
CR-806931
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Robert S. Kerr Environmental Research Laboratory
P. 0. Box 1198
13. TYPE of report and period covered
Final Report - 1979-1984
14. SPONSORING AGENCY CODE
EPA/600/15
15. SUPPLEMENTARY NOTES
volume report presents a summary of the state-of-the-art of aquifer restor-
ation. Included are eight sections and seven appendices. The text includes sections
on: (1) ground water pollution control through institutional measures, source control
stabi1ization/solidifioation methods, well systems, interceptor systems, capping and
liners, sheet piling, grouting and slurry walls; (2) treatment of ground water via
air stripping, carbon adsorption, biological treatment, chemical precipitation, and
other treatment techniques; (3) in-situ chemical treatment and biological stabili-
zation; (4) a protocol for aquifer restoration decision-making; and (5) techniques
for aiding the decision-making process. The appendices (volume II) include: (1)
case studies of aquifer restoration; (2) considerations regarding an aquifer restor-
ation information center; (3) information for public participation in aquifer restor-
ation decision-making; and (4) an annotated bibliography of 225 selected references.
The stat-of-the-art of aquifer restoration is a rapidly changing technology, with
many uses of single or combined techniques in planning or recently implemented.
Unfortunately, few if any systems have yet reached shut-down levels for contaminants.
Thus, effectiveness, duration and cost data are as yet incomplete but highly encour-
aging. A major need is for a systematic and comprehensive study of the cost-effect-
iveness of aquifer restoration technologies.
17. KEY WORDS AND DOCUMENT ANALYSIS
1. DESCRIPTORS
b.lOENTlFIERS/OPEN ENDED TERMS
c. COSATI l ield/Gxoup
Ground Water Water Treatment
Groutld Water Recharge Water Reclam-
Aquifers ation
Geology Abatement
Contaminants Waste Disposal
Water Pollution
Decontamination
Clean-up
Remedial Actions
Restoration
In-Situ Treatment
Air-stripping
Biological Clean-up
Slurry Walls
68D
91A
50B
68C
13. DISTRIBUTION STATEMENT
Release to Public
Iff? iliJmiYVt ASS* (This fit port)
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
«PA Form 2220-1 (••71)
1
-------
DISCLAIMER
The information in this document has been funded wholly or in part
by the United States Environmental Protection Agency. It has been
subject to the Agency's peer and administrative review, and it has been
approved for publication as an EPA document. Approval does not signify
that the contents necessarily reflect the views and policies of EPA, nor
does mention of trade names or commercial products constitute
endorsement or recommendation for use.
-------
FOREWORD
EPA is charged by Congress to protect the Nation's land, air and water
systems. Under a mandate of national environmental laws focused on air and
water quality, solid waste management and the control of toxic substances,
pesticides, noise, and radiation, the Agency strives to formulate and imple-
ment actions which lead to a compatible balance between human activities and
the ability of natural systems to support and nurture life.
The Robert S. Kerr Environmental Research Laboratory is the Agency's
center of expertise for investigation of the soil and subsurface environment.
Personnel at the laboratory are responsible for management of research pro-
grams to: (a) determine the fate, transport and transformation rates of
pollutants in the soil, the unsaturated zone and the saturated zones of the
subsurface environment; (b) define the processes to be used in characterizing
the soil and subsurface environment as a receptor of pollutants; (c) develop
techniques for predicting the effect of pollutants on ground water, soil and
indigenous organisms; and (d) define and demonstrate the applicability and
limitations of using natural processes, indigenous to the soil and subsurface
environment, for the protection of this resource.
This project was Initiated to provide a state-of-knowledge document
of ground-water restoration methods describing available technologies and
their applicability, and comparing methodologies and relative costs vs
alternatives to restoration. The information found has been grouped into
two volumes and includes sections on: (1) ground-water pollution control,
(2) treatment of ground water, (3) in situ treatment, (4) a protocol for
aquifer restoration decision-making, and (5) techniques for decision-making
processes. The appendices include case studies of aquifer restoration and
an annotated bibliography of 225 selected references. This compilation
should be useful to those having a need for Information about the cost-
effectiveness of aquifer restoration technologies.
Clinton W. Hall
Director
Robert S. Kerr Environmental
Research Laboratory
ill
-------
ABSTRACT
This two-volume report presents a summary of the state-of-the-art
of aquifer restoration. Included are eight sections and seven
appendices. The text includes sections on: (1) ground water pollution
control through institutional measures, source control,
stabilization/solidification methods, well systems, interceptor systems,
capping and liners, sheet piling, grouting and slurry walls; (2)
treatment of ground water via air stripping, carbon adsorption,
biological treatment, chemical precipitation, and other treatment
techniques; (3) in-situ chemical treatment and biological stabilization;
(4) a protocol for aquifer restoration decision-making; and (5)
techniques for aiding the decision-making process. The appendices
(volume II) include: (1) case studies of aquifer restoration; (2)
considerations regarding an aquifer restoration information center; (3)
information for public participation in aquifer restoration decision-
making; and (4) an annotated bibliography of 225 selected references.
The state-of-the-art of aquifer restoration is a rapidly changing
technology, with many uses of single or combined techniques in planning
or recently implemented. Unfortunately, few if any systems have yet
reached shut-down levels for contaminants. Thus, effectiveness,
duration and cost data are as yet incomplete but highly encouraging. A
major need is for a systematic and comprehensive study of the cost-
effectiveness of aquifer restoration technologies.
iv
-------
CONTENTS
ABSTRACT
LIST OF TABLES . . .
LIST OF FIGURES. . .
ACKNOWLEDGEMENTS . .
Section
I INTRODUCTION
Classification of Methodologies
Overview Studies ^
Objective of this Study 11
Scope of Study H
Organization of Report 11
Selected References. ................. 12
II CONCLUSIONS 13
III RECOMMENDATIONS 17
IV GROUND WATER POLLUTION CONTROL 22
Institutional Measures 22
Regional Ground Water Management Districts. ... 22
Source Control Strategies 23
Land Zoning 23
Effluent Charges/Credits 23
Aquifer Standards 24
Guidelines 24
Source Control Strategies 24
Stabilization/Solidification Strategies 25
Well Systems 31
Construction 31
Design 34
xiii
xx
xxv i
l
V
-------
Section Page
Application in Hydrocarbon Recovery 46
Costs 48
Advantages and Disadvantages 51
Interceptor Systems 51
Construction 61
a. Collector Drains. . 61
b. Interceptor Trenches 65
Design . 67
a. Collector Drains. . 67
b. Interceptor Trenches 69
Costs 70
Advantages and Disadvantages 70
Surface Water Control, Capping and Liners 72
Construction — Surface Capping 77
Surface Capping — Selection and Design
Considerations 78
Costs 91
Sheet Piling 91
Construction 91
Cos ts 93
Advantages and Disadvantages 93
Grouting 94
Construction 94
Design 99
Costs 101
Advantages and Disadvantages 106
Slurry Walls 106
Construction 106
Design. ........ 112
Costs . 118
Advantages and Disadvantages 118
Selected References 122
V TREATMENT OF GROUND WATER 127
vi
-------
Section Pa8e
Air Stripping 127
Description of Process. ............. 129
Design Parameters and Procedures 131
Applications to Ground Water 133
Costs 136
Carbon Adsorption 136
Description of Process 141
Design Parameters and Procedures 144
Applications to Ground Water 147
Cos 147
Biological Treatment 149
Description of Process 150
Design Parameters and Procedures 152
Applications to Ground Water 153
Costs 156
Chemical Precipitation 156
Description of Process 163
Design Parameters and Procedures 164
Applications to Ground Water. 164
Costs 166
Other Treatment Techniques for Inorganics 166
Treatment Trains 176
Selected References 176
VI IN-SITU TECHNOLOGIES 182
In-situ Chemical Treatment ... 182
Description of Process 182
Advantages and Disadvantages 186
In-situ Biological Stabilization 186
Enhancement of Indigenous Population 191
Addition of Acclimated Microorganisms 206
Conclusions 218
Selected References 219
vii
-------
Section Page
VII PROTOCOL FOR AQUIFER RESTORATION DECISION-MAKING ... 227
Background Information 227
Conceptual Framework for Aquifer Restoration
Decision-Making 233
Preliminary Activities ........... 233
Multi-disciplinary Team 233
Problem Definition and Characterization 236
a. Anticipated Problems 237
b. Existing Problems 238
Preliminary Study 238
a. Plume Delineation 239
b. Hydrogeologic Characteristics 243
c. Site Characterization 246
d. Water Use and Requirements 247
e. Human Health Costs and Risk
Assessment 248
f. Land Use Patterns and Growth
Projections 249
g. Regulations and Institutional
Constraints 250
h. Funding 252
Data Evaluation 253
Data Needs 253
Development of Alternatives 254
Definition of Goals 255
Goals Identification Matrix 256
Technology-Decision Factor Matrix 256
Preliminary Screening 259
Scope Design 262
Iteration 262
Evaluation of Alternatives and Selection of
Aquifer Restoration Strategy .... 263
Economic Considerations 263
Economic Evaluation 270
Environmental Evaluation 274
Risk Assessment 275
vi i i
-------
Section Page
Selected References 278
VIII TECHNIQUES FOR DECISION-MAKING 281
Conceptual Framework for Trade-off Analysis 283
Importance Weighting for Decision Factors 285
Systematic Comparison of Structured Importance
Weighting Techniques 286
Ranking Techniques for Importance Weighting 292
Public Preference Approach 293
Nominal Group Process Technique • • 293
Rating Techniques for Importance Weighting 294
Pre-defined Importance Scale 295
Group Approach Using Scales of Importance .... 295
Multi-attribute Utility Measurement
Technique 299
Computerized Version of MAUM Technique 302
Paired Comparison Techniques for Importance
Weighting 312
Unranked Paired Comparison Techniques 313
Ranked Paired Comparison Techniques 324
Delphi Technique for Importance Weighting 337
Scaling/Rating/Ranking of Alternatives ........ 344
Paired Comparison Technique for
Scaling/Rating/Ranking 346
Functional Curves for Scaling/Rating/Ranking. . . 353
Development of Decision Matrix 354
Selected References 363
App. A CASE STUDIES OF AQUIFER RESTORATION MEASURES A-l
Introduction A-2
Case A: Tar Creek in Oklahoma A-3
Sources of Contaminants A-6
ix
-------
Section Page
Studies to Determine Remedial Actions A-9
Proposed Remedial Action. ... A-17
Case B: Gilson Road in New Hampshire. A-17
Sources of Contaminants A-21
Studies to Determine Remedial Actions A-23
Remedial Actions Implemented A-34
Case C: Rocky Mountain Arsenal A-37
Sources of Contamination A-41
Studies to Determine Remedial Actions A-43
Remedial Actions Implemented A-48
New Jersey Department of Environmental Protection
Aquifer Restoration Program A-54
Case Studies of Aquifer Restoration via
Biological Processes A-62
Selected References A-73
App. B CONSIDERATIONS RELATED TO AQUIFER RESTORATION
INFORMATION CENTER B-l
Introduction B-2
Conceptual Framework for Aquifer Restoration
Information Center B-3
Summary B-7
App. C AIR STRIPPING OF VOLATILE ORGANICS FROM GROUND
WATER C-l
Introduction C-2
Mass Transfer Process: Theory C-3
Choice of Air Stripping Method C-7
Diffused Aerator C-8
Packed Tower C-8
Coke Tray Aerator C-8
Cross-flow Tower C-ll
Summary of Methods C-ll
Design of a Packed Tower C-ll
X
-------
Section Page
Case Study Related to Ground Water C-15
Costs of Aeration C-20
Desorption Rate of Air Strippable Organic8
in Ground Water C-20
Air Pollution Considerations C-23
Selected References C-25
App. D ACTIVATED CARBON ADSORPTION OF ORGANICS FROM
GROUND WATER D-l
Introduction • D-2
Adsorption Isotherms D~2
Single and Multistage Adsorption D-5
Fixed Bed Adsorption D-ll
Predictive Model for Design D-16
Economic Considerations D-19
Selected References D-19
App. E PUBLIC PARTICIPATION IN AQUIFER RESTORATION
DECISION-MAKING E-l
Introduction E-2
Basic Definitions ................ E-2
Advantages and Disadvantages of Public
Participation E-3
Public Participation in the Decision Making
Process. E-6
Objectives of Public Participation E-9
Identification of Publics E-18
Recognition of Types of Publics E-18
Pragmatic Approaches for Identifying Publics. . . E-24
Selected Techniques for Communicating
with Identified Publics E-31
xi
-------
Section Page
Selection of Public Participation Techniques E-31
Techniques Classification According to
Function. E-33
Techniques Classification According to
Communication Characteristics and Potential
for Meeting Stated Objectives E-37
Catalog of Additional Techniques E-48
Practical Plan for Implementing Public
Participation Program E-67
Practical Suggestions for Implementing
Program E-68
Incorporation of Results in Decision-Making . . . E-71
Selected References E-71
App. F RISK ASSESSMENT AS RELATED TO AQUIFER RESTORATION
PLANNING F-l
Introduction F-2
Background Information F-2
Elements of Risk Assessment F-9
Environmental Risk Analysis F-ll
Risk Assessment for Hazardous Waste Disposal
Sites F-22
Hazard Ranking System F-27
Site Rating Methodology F-30
Air Quality Risk Assessment System F-36
Summary F-39
Selected References F-40
App. G ANNOTATED BIBLIOGRAPHY FOR AQUIFER RESTORATION .... G-l
xii
-------
LIST OF TABLES
Table Page
1.1 Methodologies for Aquifer Clean-up 2
1.2 Aquifer Clean-up Survey (Lindorff and Cartwright,
1977)^
1.3 Examples of Aquifer Clean-up Projects
(Organization for Economic Cooperation and
Development, 1983) 9
III.l Examples of Aquifer Restoration Research Needs ... 18
IV. 1 Potential Source Control Strategies 26
IV.2 Water Well Drilling Methods. ....... 33
IV.3 Analytical Solutions of Steady Flow Applicable
to Corrective Actions (Cohen and Miller, 1983) ... 35
IV.4 Analytical Solutions of Transient Flow Applicable
to Corrective Actions (Cohen and Miller, 1983) ... 38
IV.5 Methods for Calculating Radius of Influence
(Kufs, et al., 1983) 41
IV.6 Advantages/Disadvantages of Well Systems 60
IV.7 Advantages/Disadvantages of Subsurface Drains
(U.S. Environmental Protection Agency, 1982) .... 71
IV.8 Advantages/Disadvantages of Capping and Liners
(Smith, 1983) 76
IV.9 Ranking of USCS Soil Types According to
Performance of Cover Functions (Fung, 1980) 79
IV.10 Costs for Various Sanitary Landfill Liner
Materials (Haxo, 1973) 92
IV.11 Advantages/Disadvantages of Steel Sheetpiles
(Tolman, et al.t 1978) ............... 95
IV. 12 Materials Commonly Used for Grout 96
IV.13 Properties of Admixtures Used with Cement Grouts . . 97
xiii
-------
Table Page
IV.14 Interactions Between Grouts and Generic Chemical
Classes (Hunt, et al., no date) 102
IV.15 Interactions Between Grouts and Specific Chemical
Groups (Hunt, et al., no date) 103
IV.16 Predicted Grout Compatibilities (Hunt, et al.,
no date) 104
IV. 17 Relative Costs of Grout 105
IV. 18 Advantages/Disadvantages of Grout Systems 107
IV.19 Comparison of C-B and S-B Slurry Trenches
(Ryan, 1980) Ill
IV.20 Funnel Viscosity for Common Types of Soil
(Xanthakos, 1979) 119
IV.21 Control Limits for the Properties of Slurries
(Xanthakos, 1979) 120
IV.22 Approximate Slurry Wall Costs as a Function of
Medium and Depth (Spooner, Wetzel and Grube,
1982) 121
IV.23 Advantages/Disadvantages of Slurry Trenches 123
V.l Packed Column Air Stripping of Volatile Organics . . 135
V.2 Activated Carbon Adsorption of Organics 148
V.3 Removal Mechanisms of Toxic Organics 154
V.4 High Temperature Air Stripper Extract Water
Qualities for Biological Studies 157
V.5 Design Summary for Activated Sludge System 158
V.6 Design Summary for RBC System 159
V.7 Project Effluent Qualities from Activated Sludge
Design 160
V.8 Project Effluent Quality from RBC Design ...... 161
V.9 Results of Specific Compound Studies . 162
V.10 Removal Chemical Precipitation Data for Metals . . . 165
xiv
-------
Table Page
V.ll Comparison of Precipitation Treatment Processes. . . 167
V.12 Ion-Exchange System - Cost Estimate Summary 168
V.13 Lime-Softening System - Cost Estimate Summary. . . . 169
V.14 Chemical Precipitation-Direct Filtration System
Cost Estimate Summary 170
V.15 Monetary Comparison of Alternative Treatment
Systems 171
V.16 Treatment Alternatives for Inorganics 172
V. 17 Summary of Suitability of Treatment Processes. . . . 177
V.18 Ground Water Treatment Results from a Selected
Treatment Train 178
VII.1 Site Contamination and Liability Audit Phased
Structure (Housman, Brandwein and Unites, 1981). . . 232
VII.2 Potential Sources of Information 240
VII.3 Aquifer Characteristics 245
VII.4 Examples of Ground Water Threatening Activities. . . 251
VII.5 Potential Aquifer Restoration Strategies for
Chronic Pollution Problems 257
VII.6 Goals Identification Matrix 258
VII.7 Technology-Decision Factor Matrix 260
VII.8 Average U.S. Low and High Costs of Unit Operations
for Medium-Sized Landfill Sites (Rishel, et al.,
1983) 267
VII.9 Average U.S. Low and High Costs of Unit Operations
for Medium-Sized Surface Impoundment Sites
(Rishel, et al., 1983) 268
VII.10 Summary of Estimated Costs and Characteristics
of Remedial Measures (Tolman, et al., 1978) 269
VII.11 Cost Components of Well Systems (Lundy and
Mahan, 1982) 272
XV
-------
Table Page
VII.12 Example Trade-off Matrix for Cost-Effectiveness
Assessment of Remedial Actions at Uncontrolled
Hazardous Waste Sites (St. Clair, McCloskey and
Sherman, 1982) 273
VIII.1 Trade-off Matrix 284
VIII. 2 Importance Weighting Techniques 287
VIII.3 Partial Paired Comparisons I (Eckenrode, 1965) . . • 290
VIII.4 Importance Scale (Linstone and Turoff, 1975) .... 296
VIII.5 Example of the Development of Impact Area
Importance Weighting (Rau, 1980) 298
VIII.6 Illustration of the Use of DECIDE (Rugg and
Feldman, 1982) 303
VIII.7 Program Listing for DECIDE (Rugg and Feldman,
1982) 307
VIII.8 Use of Paired-Comparison Technique for Importance
Weight Assignments 314
VIII.9 Example Comparison Matrix (Ross, 1976) 317
V11I.10 Permuted Example Comparispn Matrix (Ross, 1976). . . 318
VIII.11 Permuted Example Comparison Matrix (Inconsistent)
(Ross, 1976) 320
VIII.12 Environmental Categories and Parameters for
Waterway Navigation Project (School of Civil
Engineering and Environmental Science and Oklahoma
Biological Survey, 1974) 333
VIII.13 Group Communication Techniques (Linstone and
Turoff, 1975) 338
VIII.14 Information for Trade-off Analysis 347
VIII.15 Scaling/Rating/Ranking of Alternatives Relative
to F1 348
VIII.16 Scaling/Rating/Ranking of Alternatives Relative
to F2 349
xvi
-------
Table Page
VIII.17 Scaling/Rating/Ranking of Alternatives Relative
to F3 350
VIII.18 Scaling/Rating/Ranking of Alternatives Relative
to F4 351
VIII.19 FIC and ACC Values for Example Decision Problems . . 358
VIII.20 Product Matrix for Trade-off Analysis for Example
Decision Problem 359
VIII.21 Basis for Impact Rating (Rau, 1980) 360
VIII.22 Summary of Some Weighting-Scaling/Rating/Ranking
Methods for Trade-off Analyses 364
VIII.23 Delineation of a Methodology for Trade-off
Analysis Involving Environmental Impacts 366
A. 1 Summary of Capital Costs for Tar Creek
Alternatives I, II, and III (Hittman Associates,
Inc., 1981) A-18
A.2 Summary of Annualized Costs for Tar Creek
Alternatives I, II, and III (Hittman Associates,
Inc., 1981) A-19
A.3 Additional Studies Needed for Tar Creek (Knox,
Stover and Kincannon, 1982) A-20
A.4 Primary Contaminants of Concern in Ground Water
at the Gilson Road Site (Stover and Kincannon,
1982) A-22
A.5 Test Results from Metals Removal Studies of Ground
Water from the Gilson Road Site (Stover and
Kincannon, 1982) A-27
A.6 Volatile Organics Removal During Stripping
Studies of Ground Water from the Gilson Road Site
(Stover and Kincannon, 1982) . A-29
A. 7 Removal of Extractable Organics by Steam
Stripping of Ground Water from the Gilson Road
Site (Stover and Kincannon, 1982) A-31
A.8 Test Results from Various Activated Carbon Studies
of Ground Water from the Gilson Road Site
(Stover and Kincannon, 1982) A-32
xvi i
-------
Table Page
A.9 Test Results from Batch Activated Sludge Studies
of Ground Water from the Gilson Road Site
(Stover and Kincannon, 1982) ..... A-33
A. 10 Summary of Treatability Studies of Ground Water
from the Gilson Road Site (Stover and Kincannon,
1982) A-36
A. 11 Ground Water Contaminants at North Boundary of
Rocky Mountain Arsenal (Hager and Loven, 1982) . . . A-42
A. 12 Information on Interim Granular Activated Carbon
Treatment System (1/7/78 through 6/30/81) Used
at North Boundary of Rocky Mountain Arsenal
(Hager and Loven, 1982) A-47
A. 13 Simulation Results of a Digital-Transport Modelling
Study at Rocky Mountain Arsenal (Warner, 1979) . . . A-49
A.14 Information on Permanent Granular Activated Carbon
Treatment System Used at North Boundary of Rocky
Mountain Arsenal (Hager and Loven, 1982) A-53
A. 15 Ground Water Contamination Investigations in
New Jersey — By Source (Althoff, 1983) A-57
A. 16 Ground Water Contamination Investigations in
New Jersey — By Contaminant Type (Althoff, 1983). . A-58
A. 17 Aquifer Restoration Programs in New Jersey —
January, 1983 (Althoff, 1983). . A-59
C.l Henry's Constants for Pollutants Which Might
be Found in Ground Water C-17
C.2 Pollutant Levels in Ground Water C-19
D.l Organic Compounds and Their Removal D-3
D.2 Freundlish Isotherm Absorbtion Characteristics
for Organics in Ground Water D-6
D.3 Operating Results — Influent Contaminants at
mg/1 Levels D-17
D.4 Operating Results — Influent Contaminants at
ug/1 Levels D-18
xviii
-------
Table Page
E.l Public Involvement Objectives at Various Decision
Making Process Stages (Bishop, 1975) E-13
E.2 Citizen Involvement Checklist (Schwertz, 1979) . . . E-27
E.3 Scheme for Identifying Publics (Bishop, 1981). . . . E-29
E.4 Effectiveness of Different Media on Various
"Publics" (Bishop, 1975) E-32
E.5 Participation Techniques Classified by Function
(Schwertz, 1979) E-34
E.6 Capabilities of Public Participation Techniques
Bishop, 1975) E-38
E.7 Rough Cost Guide to Most Frequently Used Public
Involvement Techniques (Delli Priscoli, 1981). . . . E-69
F.l Actions Increasing Risk of Death by One in a
Million (Conway, 1982) F-14
F.2 Mechanisms Affecting Environmental Fate of
Chemicals (Conway, 1982) F-20
F.3 Candidate Tests for Screening Ecological Impact
of New Products (Conway, 1982) F-21
F.4 Summary of Guidelines Used in Rating Severity of
Damage at Evaluated Sites (Tusa and Gillen, 1982). . F-23
F.5 Comprehensive List of Rating Factors in Hazard
Ranking System (Caldwell, Barrett and Chang,
1981) F-29
F.6 Rating Factors and Scales for the Site Rating
Methodology (Kufs, et al., 1980) F-32
F.7 Guidance for Additional Points System in the
Site Rating Methodology (Kufs, et al., 1980) .... F-35
xix
-------
LIST OF FIGURES
Figure Page
IV.1 Pumping by Suction Lift (Johnson Division,
UOP Inc., 1975) 43
IV.2 Principle of Withdrawal Wells 44
IV.3 Principles of Pressure Ridge System 45
IV.4 Schematic of One Pump System Utilizing a
Submersible Pump and Float Controls (Blake and
Lewis, 1982) 47
IV.5 Schematic of Two Pump System Utilizing Two Small
Diameter Wells (Blake and Lewis, 1982) 49
IV.6 Schematic of Two Pump System Utilizing One
Recovery Well (Blake and Lewis, 1982) 50
IV.7 Well Completed in Unconsolidated Sediments
(Sand and Gravel) (Campbell and Lehr, 1977) 52
IV.8 Cost of Cased Wells Finished in Sand and Gravel
(Campbell and Lehr, 1977) 53
IV.9 Gravel-Packed Well Completed in Sand and Gravel —
Unconsolidated to Partly Consolidated Sediments
(Campbell and Lehr, 1977) 54
IV.10 Cost of Gravel-Packed Wells Finished in Sand and
Gravel (Campbell and Lehr, 1977) 55
IV.11 Shallow Well Completed in Consolidated Sediments
of Sandstone, Limestone or Dolomite (Bedrock Well)
(Campbell and Lehr, 1977) 56
IV.12 Cost of Shallow Sandstone, Limestone or Dolomite
Bedrock Wells (Campbell and Lehr, 1977) 57
IV.13 Deep Well Completed in Consolidated Sediments
(Campbell and Lehr, 1977) 58
IV.14 Cost of Deep Sandstone Wells (Campbell and Lehr,
1977) 59
IV.15 Hydraulic Gradient Toward Interceptor System .... 62
IV.16 Collector Drain System 63
XX
-------
Figure Page
IV. 17 Collector System Layouts 64
IV.18 Shape of Water Table Adjacent to Tile Drains
(a) Relatively Permeable Soil; (b) Less Permeable
Soil (Spangler and Hardy, 1973) 68
IV.19 Moisture Components of Landfills (Smith, 1983) ... 74
IV.20 Four Combinations of Capping and Liners (Smith,
1983) 75
IV.21 Typical Layered Cover Systems (Fung, 1980) 87
IV.22 Two Types of Gas Vents for Layered Cover Systems
(Fung, 1980) 89
IV.23 Soil Gradation Limits for Grout Injection (from
American Cyanamid Company) 100
IV.24 Cross Section of Landfill Before (a) and After (b)
Slurry-Trench Cutoff Wall Installation (Tolman,
et al., 1978) 108
IV.25 Isolation of Existing Buried Waste (Ryan, 1980). . . 109
IV.26 Relationship Between Permeability and Quantity of
Bentonite Added to SB Backfill (Ryan, 1980) 114
IV.27 Permeability of Soil-Bentonite Backfill Related to
Fines Content (Ryan, 1980) 115
IV.28 Stability of a Trench for Arbitrary Slurry and
Natural Water Level (Xanthakos, 1979). ....... 117
V.l Air Stripping Equipment Configurations 130
V.2 Mass Transfer Coefficients as a Function of
Water Velocity 134
V.3 Comparison of Costs for Packed Column Aeration
(Dyksen, et al., 1982) 137
V.4 Adsorption Isotherm 140
V.5 Fixed-Bed Absorption System 142
V.6 Moving-Bed Absorption System ............ 143
V.7 Typical Breakthrough Curve 146
•xxi
-------
Figure Page
V.8 Effluent Phenol versus Phenol Loading Rate ..... 155
VI.l Vyredox Concept 184
VI.2 Vyredox Plant with Two Supply Welle Complete with
Aeration Wells and Oxygenator Building ....... 185
VII.1 Remedial Action Process (Bixler, Hanson and
Langner, 1982) 229
VII.2 Master Site Schedule (Kaschuk and ttadeau, 1982). . . 231
VII.3 Integrated Site Restoration Process (Dawson
and Brown, 1981) 234
VII.4 Flowchart for Aquifer Restoration Decision-
Making 235
VIII.1 Rating Scale (Eckenrode, 1965) 289
VIII.2 Design and Analytical Phases of the SAGE Method
(Hyman, Moreau and Stiffel, 1982) 321
VIII.3 Environmental Evaluation System (Dee, et al.,
1972) 327
VIII.4 Flow Diagram of Environmental Analysis Procedure
(School of Civil Engineering and Environmental
Science and Oklahoma Biological Survey, 1974). . . . 331
VIII.5 Functional Curve for Species Diversity
(Dee, et al., 1972) 355
VIII.6 Functional Curve for Dissolved Oxygen in Water
(Dee, et al., 1972). « . . • . . * . . . * . • . • « 356
A.1 Hydrogeology of the Tar Creek Area (Hittman
Associates, Inc., 1981). . . . A-4
A. 2 Conceptual Design of Alternative I (Hittman
Associates, Inc., 1981) A-ll
A.3 Acid Mine Drainage Treatment Plant — Unit
Processes (Hittman Associates, Inc., 1981) A-12
A.4 Conceptual Design of Alternative II (Hittman
Associates, Inc., 1981) A-14
*xii
-------
Figure Page
A.5 Conceptual Design for Collector and Monitoring
(Observation) Wells (Hittman Associates, Inc.,
1981 ) A-15
A.6 Conceptual Design o£ Alternative III (Hittman
Associates, Inc., 1981) A-16
A.7 pH Titration Characteristics of Raw and Treated
Ground Water from the Gilson Road Site (Stover
and Kincannon, 1982) A-26
A.8 TOC Removal by Air and Steam Stripping of Ground
Water from the Gilson Road Site (Stover and
Kincannon, 1982) A-28
A.9 Location of Rocky Mountain Arsenal (Warner, 1979). . A-38
A.10 Vicinity Map of Rocky Mountain Arsenal (Shukle,
1982 ) A-39
A.11 Ground Water Contours and Bedrock Highs at the
Rocky Mountain Arsenal (Warner, 1979) A-40
A. 12 Extent of Ground Water Contamination at Rocky
Mountain Arsenal (Shukle, 1982) A-44
A.13 North Boundary Area of the Rocky Mountain
Arsenal (Hager and Loven, 1982) A-45
A.14 Permanent Granular Activated Carbon Treatment
Facility at North Boundary of Rocky Mountain
Arsenal (Hager and Loven, 1982) A-52
A. 15 Containment System Locations at Rocky Mountain
Arsenal (Shukle, 1982) ..... A-55
B.l Conceptual Framework for Aquifer Restoration
Information Center (ARIC) B-4
B.2 Conceptual Framework for Information/Technology
Transfer Program B-5
B.3 Data/Information Flow Involving ARIC ........ B-6
C.l Schematic Representation of Interfacial Mass-
transfer Process with Controlling Resistance
Residing in Liquid Phase C-5
C.2 Typical Turbine Aerator Installation . . C-9
xxiii
-------
Figure Page
C.3 Packed Tower C-10
C.4 Process Schematic for Concurrent Packed Tower. . . . C-13
C.5 Schematic of Volative Desorption Apparatus C-22
C.6 Intalox Saddle Desorption Apparatus C-24
D.l Single-Stage Adsorption D-8
D.2 Two-Stage Crosscurrent Adsorption D-10
D.3 Two-Stage Countercurrent Adsorption D-10
D.4 The Adsorption Wave D-12
D.5 Idealized Breakthrough Curve D-12
D.6 The Adsorption Zone D-14
D.7 Arrangements of Fixed Bed Adsorption Systems .... D-15
E.l Objectives of Public Participation (Hanchey, 1981) . E-10
£.2 Example of a Diffusion Process (Bhsop, 1985) .... E-16
E.3 An Interaction Process (Bishop, 1981) E-17
E.4 Correspondence of Communications Objectives and
Models (Bishop, 1981) E-19
E.5 Example of Multiple Public Associations (Bishop,
1981) E-21
E.6 A Temporal Perspective of Identification of
Publics (Bishop, 1981) E-30
F.l Extrapolations Required in Determining Risk
(Schweitzer, 1981) F-6
F.2 Pathway Concept (Ess and Shih, 1982) F-8
F.3 Risk Quantification Using an Event Tree
(Lee and Nair, 1979) F-10
F.4 An Event Tree Analysis of LNG Spills (Lee and
Nair, 1979) F-12
x.xiv
-------
Figure Page
F.5 Generalized Decision Tree Structure (Shih and
Ess, 1982) F-13
F.6 Deterministic Approach to Environmental Risk
Analysis (Conway, 1982) F-16
F.7 Overview of the Three Basic Elements of
Environmental Risk Assessment (Cornaby, et al.,
1982) F-17
F.8 Environmental Pathways from a Waste Site
(Schweitzer, 1981) F-18
F.9 Overall Environmental Risk Assessment Scheme
(Cornaby, et al., 1982) F-19
F.10 Preliminary Site Evaluation Assessment Schematic
(Unites, Possidento and Housman, 1980) F-24
F.ll Sample Diagram from Migration of Hazardous Wastes
in the Environment (Murphy, 1982) F-26
F.12 Guidelines for Interpreting the Level of Potential
Hazard Based on the Site Rating Methodology
(Kufs, et al., 1980) F-37
.XXV
-------
ACKNOWLEDGEMENTS
The authors of this report wish to acknowledge the assistance of
James NcNabb and Marion R. Scalf of the R.S. Kerr Environmental Research
Laboratory of EPA located at Ada, Oklahoma. In addition, a panel of
experts provided periodic advice, consultation, and review comments
during the conduction of this study. The panel included:
William Althoff, New Jersey Department of Environmental Protection
Rich Barteldt, Region V Superfund Coordinator, EPA
Richard Conway, Union Carbide Corporation
George Dixon, Office of Solid Waste, EPA
Lawrence Greenfield, Office of Solid Waste, EPA
Jerry Kotas, Ground Water Protection Branch, EPA
Roy Murphy, Office of Waste Programs Enforcement, EPA
Dr. Rick Standford, Office of Emergency Response, EPA
The authors also wish to acknowledge the typing accomplishments of
Ms. Leslie Rard in the assemblage of this two-volume report. The able
assistance of Mrs. Wilma Clark and Mrs. Mittie Durham is also
acknowledged.
XXVi
-------
SECTION I
INTRODUCTION
Pollution of ground water can result from many activities,
including leaching from municipal and chemical landfills and abandoned
dump sites, accidental spills of chemicals or waste materials, improper
underground injection of liquid wastes, and placement of septic tank
systems in hydrologically and geologically unsuitable locations. In
recent years incidents of aquifer pollution from man's waste disposal
activities have been discovered with increasing regularity.
Concurrently, demands for ground water usage have been increasing due to
population growth and diminishing opportunities to economically develop
surface water supplies. Until recently the general viewpoint held by
many ground water professionals and policy-makers was that once an
aquifer had become polluted its water usage must be curtailed or
possibly eliminated. However, this viewpoint is changing as a result
of increasing needs for ground water utilization and the development of
appropriate methodologies for aquifer clean-up. The focus on
methodologies has been heightened by current hazardous waste site clean-
up efforts funded by "Superfund".
CLASSIFICATION OF METHODOLOGIES
Table 1.1 lists methodologies for aquifer clean-up organized
according to whether the pollution problem is acute or chronic. Acute
pollution may occur from inadvertent spills of chemicals or releases of
undesirable materials and chemicals during a transportation accident.
Acute pollution events are unplanned and are characterized by their
-1-
-------
Table 1.1: Methodologies for Aquifer Clean-up
Pollution
Problem
Goal
Methodologies
Acute
Abatement
1. In-situ chemical fixation.
2. Excavation of contaminated
soil
with subsequent backfilling
with
"clean" soil.
Restoration
1. Removal wells, treatment
of
contaminated ground water,
and
recharge.
2. Removal wells, treatment
of
contaminated ground water,
and
discharge to surface drainage.
3. Removal wells and discharge
to
surface drainage.
Chronic
Abatement
1. In-situ chemical fixation.
2. Excavation of contaminated
coil
with subsequent backfilling
with
"clean" soil.
3. Interceptor trenches to collect
polluted water as it moves
laterally away from site.
4. Surface capping with impermeable
material to inhibit infiltration
of lisachate-produclng precipita-
tion.
5. Subsurface barriers of impermeable
materials to restrict hydraulic
flow from sources.
6. Modify pumping patterns at
existing wells.
7. Inject fresh water in a series of
wells placed around source or
contaminant plume to develop
pressure ridge to restrict
movement of pollutants.
-2-
-------
Table 1.1 (continued)
Pollution
Problem
Goal
Methodologies
Chronic
Restoration
1. Removal wells, treatment of
contaminated ground water, and
recharge.
2. Removal veils, treatment of
contaminated ground water, and
discharge to surface drainage.
3. Removal wells and discharge to
surface drainage.
4. In-situ chemical treatment.
5. In-situ biological treatment.
-3-
-------
emergency nature. Chronic aquifer pollution may occur £rom numerous
point and area sources and may involve traditional pollutants such as
nitrates and bacteria, or unique pollutants such as gasoline, metals,
and synthetic organic chemicals.
Methodologies for aquifer clean-up can also be considered in terms
of the goals of abatement and restoration. Abatement means "to put an
end to"; therefore, abatement as a goal refers to the application of
methodologies which will aid in preventing pollutant movement into
ground water, or preventing contaminated plume movement into usable
aquifer zones. The latter example of abatement is also called plume
management. Aquifer restoration as a goal refers to the restoration of
water quality to its normal quality, usually by removing both the source
of pollution and renovating the polluted ground water. If the pollution
source has already been dissipated by time, restoration may only involve
renovation of the polluted ground water.
It should be noted that a given aquifer clean-up project may
involve usage of several methodologies in combination. For example, for
an acute problem, excavation and backfilling may be used in conjunction
with removal wells, treatment of contaminated ground water, and
discharge to surface drainage. A chronic pollution clean-up project may
include surface capping, subsurface barriers, and in-situ chemical
treatment.
OVERVIEW STUDIES
In 1977, Lindorff and Cartwright surveyed the nation for case
histories of aquifer clean-up. Information on 116 cases of aquifer
-4-
-------
pollution was summarized, with most of the pollution either caused by
industrial wastes or by leaching from municipal landfills. Table 1.2
summarizes the clean-up methodologies used in 32 of the cases (Lindorff
and Cartwright, 1977). Removal wells are best applied when the
pollution has not traveled far from the source, and they have most often
been used to remove oil and oil products. This technique has also been
used to control water soluble chemicals. For oil pollution two or more
pumping locations in each well may be used, a deep pump to maintain a
drawdown cone and a shallow skimming pump to remove the oil (Todd,
1973). For pollutants that have traveled very far from the source or
have dispersed, the use of interceptor trenches is often more effective
than the drawdown cone in a removal well. The interceptor trench (or
pit or ditch) is placed across the pollution plume to capture the ground
water flow. The water is then pumped out and treated. This method is
obviously most useful for shallow aquifers (Committee on Environmental
Affairs, 1972).
From May to October, 1980, a survey of on-going and completed
remedial action projects was conducted by Neely, et al. A total of 169
hazardous waste sites were identified; however, because controls for
ground water clean-up are more expensive and time consuming than those
for surface water, complete remedial measures were implemented at only a
few sites. Details of remedial measures were reported for nine selected
sites, with the measures appearing to be effective at only two sites
(Neely, et al., 1981).
An expert seminar was held in Paris, France in November, 1980,
under sponsorship of the Organization for Economic Co-operation and
-5-
-------
Table 1.2: Aquifer Clean-up Survey (Lindorff and Cartwright, 1977)
Remedial Action State
Removal wells California
Delaware
Georgia
Illinois
Indiana
Kentucky
Michigan
Minnesota
New Jersey
New Mexico
New York
Ohio
Oregon
Pennsylvania
Wisconsin
Removal wells California
and treatment
Connecticut
New Jersey
Ground Water Pollutant or Source
gasoline
landfilling
gasoline
acrylonitrile
chemical wastes
cyanide
acetone cyanohydrin
oil
hydrochloric acid
industrial waste pond
solvents and acids
chromium
heavy metals
chlorides
gasoline
brines
chlorides
aluminum and mine tailings
fuel oil
gasoline
insecticides
terpenes and acids
organic wastes
sulfite liquor
phenols (a)
styrene (b)
petrochemicals (b)
-6-
-------
Table 1.2 (continued)
Remedial Action
State
Ground Water Pollutant or Source
Interception
trench
Georgia
Idaho
Pennsylvania
South Dakota
gasoline
cadmium, lead, and zinc
fuel oil
landfill
landfill
(a) treatment with chlorine dioxide
(b) treatment with activated carbon filters
-7-
-------
Development (OECD). Table 1.3 summarizes remedial actions used at 13
sites, including 5 from the USA, 4 from Germany, and one each from
France, The Netherlands, Sweden, and the United Kingdom (Organization
for Economic Co-operation and Development, 1983). One conclusion from
the seminar was that a careful evaluation of the cost-effectiveness of
possible remedial measures is essential before decisions are made with
regard to implementation of a particular methodology. All components of
the cost of a remedial program must be taken into account. These
include, in addition to the actual engineering cost, the physicochemical
site evaluation required prior to implementation of remedial measures
and the subsequent monitoring. It is possible that the cost of prior
evaluation and subsequent monitoring may equal that of the remedial
methodology used.
As a result of these overview studies, three generalizations about
aquifer restoration strategies can be made. First, they are costly.
Restoration often costs tens of millions of dollars. Secondly, they are
time-consuming. Because most problems have been recognized only after
the pollutant has been moving and spreading for a number of years, the
areal extent of pollution is often quite large. Consequently, it takes
a long time to clean up these large problems. Third, aquifer
restoration strategies are not always effective. Although the success
ratio of recent years has increased, there are a number of instances
where the clean-up strategy has been ineffective and second and third
attempts to restore the aquifer have been made.
-8-
-------
Table 1.3: Examples of Aquifer Clean-up Projects (Organization for
Economic Cooperation and Development, (1983).
Brief Site Description
Nature of "Problem"
Remedial Action
Comments
1. West Germany. Hamburg Soil and groundwater con-
Contamination, mainly heavy metals tamination shown by drill-
detected in 1976 at a central de- ing and sampling,
contamination plant in an ^
Industrial area. The 100 m area
contained 100 ra^ of contaminated
material.
Removal of contaminated soil
to hazardous waste disposal
site. Pumping of ground-
water.
I
SO
I
2. La Bounty Dump. Charles City,
Iowa. USA
Chemical waste dump operated in
river flood plain over major
aquifer, 1952-1977. Contained
over 3 million tons of feed
additives and veterinary
pharmaceuticals including 28
priority pollutants.
3. Windham. Connecticut, USA
Landfill to tfillimatic Township
Pollution of river.
Groundwater pollution
and threatened public
water supply reservoir.
Monitoring and closure of
site with a "capping".
Some in situ stabilization
may follow.
PVC Cap Installed to prevent
rain water inflow to refuse.
Site closed.
If remedial action is not
effective enough, an up gradient
cut-off wall may be Installed
to reduce water flow.
90% of water to be intercepted
and routed away.
4. Clin Corporation Site.
Saltville. PA.. USA
Mercury wastes from nearby
alkali Industry.
Mercury globules visible
at surface. River
polluted. Vapour hazard.
Temporary cosmetic treatment Only satisfactory solution is a
to reduce immediate hazard by $23 million removal operation,
covering exposed waste near Recovery of mercury currently
river with plastic and sand not economic because of price
to reduce pollution to river, slump.
5. Firestone Tire and Rubber Site,
Pottstown, PA.. PSA
Tires, scrubber waste, organic
waste, pigments, and PVC sludge.
Unknown.
Recovery wells to intercept 100Z efficient solution is the
polluted water and recycle it problem anticipated. Length
through plant. of pumping and costs not quoted.
6. Various locations in USA
Sites containing solvents, oils,
paint wastes, heavy metals, acid
sludges and PCBs.
Toxic liquid lagoons.
Leachate surface migration.
Leachate spread.
Techniques include: Many cases had received
removal of waste at the site; temporary remedial treatment
containment of waste at the and were awaiting funds before
site; emptying of pits/ complete remedial work could be
lagoons; oil absorption in carried out.
fuller's earth; mobile active
carbon unit.
-------
Table 1.3: (continued)
7. City of Hamburg, Germany
Toluene contamination from «
700 of waste reached 100 m
of subway in the aarketlng area
of the city.
Damage to concrete of
subway. Soil and ground-
water contamination.
Toluene vapour.
Reconstruction of part of
tunnel shown by drilling and
sampling to be in contaminat-
ed area. Removal of soil to
proper landfill site. Three
years ground water pumping.
flrenzach Uyllea, Germany
Disposal of chlorinated
hydrocarbons , chemical war-
fare agents in gravel pit.
Chlorinated hydrocarbons
drinking water. Highly
toxic material.
in
Development of special
investigation and analyses
methods.
9. Marals de Ponteau, France
30,000 m^ of oil waste on salt
Surface water contamina-
tion aesthetic insult.
Removal, incineration and
solidification.
Cost - 6 million francs.
Successful solidification.
0
1
10. Lekkerkerk, Netherlands
Houses built on chemical waste
with high solvent content.
11. R,T. Keml, Sweden
Large-scale disposal of
pesticides in agricultural
region near surface water.
Contamination of drinking
water and air.
Surface water contamina-
tion, odors, and possible
public health impacts.
Total removal, incineration
and disposal.
High cost, evacuation of
residents necessary.
Water management, excavation Experience in groundwater
and incineration. purging. Heavy participation
by public and media.
12. Malklns Bank, United Kingdom
General industrial waste tip in
rural area.
Noxious odors, surface
water contamination.
Encapsulation and restoration Private land acquired by
to public golf course. government.
13. Cendorf, Germany
Organics and other pesticides
on production site.
Soil, groundwater and sur-
face water contamination.
Encapsulation, surface seal-
ing and solidification.
Excavation too dangerous.
Permanent restrictions on land
-------
OBJECTIVE OF THIS STUDY
The major objective of this study was to conduct a state-of-the-art
survey of aquifer restoration in the United States. Emphasis was given
to the assemblage of information on technologies available for the
prevention, abatement, clean-up and restoration of polluted ground water
formations* In addition, a structured approach or protocol for the
selection of aquifer restoration strategies was developed.
SCOPE OF STUDY
The scope of the study reported herein primarily involved the
conduction of a review and analysis of published literature on aquifer
restoration. In addition, visits were made to several ground water-
oriented organizations to gather information. A nation-wide survey of
consulting firms was also conducted. Probably the single most
constructive activity, in terms of information accumulation, was the
attendance at, participation in, or conduction of, several conferences
dealing in whole or in part with aquifer restoration.
ORGANIZATION OF REPORT
This report consists of eight sections and seven appendices
organized into two volumes. The first volume contains the eight
sections, while the second volume contains the appendices. In addition
to this introductory section, the other section titles are as follows:
Sec. II — Conclusions
Sec. Ill — Recommendations
Sec. IV — Ground Water Pollution Control
Sec. V — Treatment of Ground Water
-11-
-------
Sec. VI — In-situ Technologies
Sec. VII — Protocol for Aquifer Restoration Decision-Making
Sec. VIII — Techniques for Decision-Making
The topical coverages of the appendices are as follows:
App. A
App. B
App. C
App. D
App. G
App. F
App. 6
SELECTED REFERENCES
Committee on Environmental Affairs, "The Migration of Petroleum Products
in Soil and Ground Water - Principles and Countermeasures", Publ No.
4149, 1972, American Petroleum Institute, Washington, D.C.
Lindorff, D.E. and Cartwright, K., "Ground Water Contamination:
Problems and Remedial Actions", Report No. 81, May 1977, Illinois State
Geological Survey, Urbana, Illinois.
Neely, D., et al., "Remedial Actions at Hazardous Waste Sites, Survey
and Case Studies", EPA 430/9-81-05, Jan. 1981, U.S. Environmental
Protection Agency, Cincinnati, Ohio.
Organization for Economic Co-operation and Development, "Hazardous Waste
Problem Sites", 1983, Paris, France.
Todd, D.K., "Ground Water Pollution in Europe - A Conference Summary",
Contract No. EPA-68-01-0759, 1973, Center for Advanced Studies, General
Electric Co., Santa Barbara, California.
-12-
Case Studies of Aquifer Restoration Measures
Considerations Related to Aquifer Restoration
Information Center
Air Stripping of Volatile Organics from Ground Water
— Activated Carbon Adsorption of Organics front Ground
Water
Public Participation in Aquifer Restoration
Decision-Making
Risk Assessment as Related to Aquifer Restoration
Planning
Annotated Bibliography for Aquifer Restoration (a
total of 225 references are included).
-------
SECTION II
CONCLUSIONS
As recently as the late 1970's, the four most common perceptions
concerning aquifer restoration measures were: (1) they were costly; (2)
they were time-consuming; (3) they were not always effective; and (4)
pertinent information was unavailable. Although these perceptions have
not been totally reversed, the state-of-the-art is significantly
different. There has been an ever-increasing amount of information
becoming available concerning aquifer restoration and ground water
cleanup. The majority of this recent information has been presented in
a wide variety of conferences and symposia. However, the amount of
information published in the refereed literature remains sparce, and a
significant amount of information also remains unavailable because it is
associated with litigation. This section summarizes the findings and
conclusions of this state-of-the art study of aquifer restoration.
Many different measures, ranging from institutional mandates to
physical technologies, have been proposed for the protection and/or
cleanup of ground water resources. Institutional measures actually
implemented consist mainly of legislated authority to enforce cleanup
mandates. In addition, a number of states have developed preventive
policies such as requiring liners and/or surface water control at waste
disposal facilities. Several Federal institutional measures have
provisions for addressing the protection or cleanup of ground water,
including the Comprehensive Environmental Response, Compensation and
-13-
-------
Liability Act (Superfund), the Safe Drinking Water Act, the Underground
Injection Control Program, and others.
The large number of physical technologies proposed for the cleanup
of polluted ground water have come from several disciplines. The
traditional hydrogeological technology of pumping wells has been
supplemented by technologies from civil engineering and construction
industries such as grouting and slurry walls; and from the agricultural
industry which has provided subsurface drainage technologies. At any
one site, the remedial program employed will usually consist of a
combination of technologies. Not all of the technologies have been
tried or proven for ground water pollution applications, and performance
data for these new applications is just now becoming available.
In spite of the innovative technologies now being promoted, the
most popular ground water cleanup measure consists of removal and
treatment. The mechanics of ground water flow to removal wells have
been studied for a number of years. This knowledge is most often
combined with traditional wastewater treatment technologies to clean up
a polluted aquifer. In fact, this mode of cleanup has become so common
that there is increased interest and study toward developing compact,
mobile wastewater treatment units; especially units for organics removal
by adsorption or air stripping.
Treatment of polluted ground water by in-situ techniques is still
in its infancy; however, it is receiving increased attention.
Successful in-situ treatment is highly dependent on both the
characteristics of the pollution(s) and the subsurface hydrogeology.
Examples of studies with adequate controls to determine the
-14-
-------
effectiveness of in-situ techniques are limited, but it is recognized
that in-situ techniques will be applicable only to sites meeting highly
specific requirements.
The costs of aquifer restoration measures are dependent on many
factors. Published information concerning cost of remedial techniques
has most often been reported as unit cost data or national average
costs. However, from the few specific reports on economics and
consideration of several case studies, one significant conclusion can be
drawn. The conclusion is that the cost of cleaning up an aquifer will
not always be in the tens of millions of dollars. Not all aquifer
restoration projects will fall under the Superfund category which has
received so much publicity for its large expenditures. In fact, some
examples of successful and economic aquifer cleanup projects using
private funds have been identified.
Development of feasible strategies (alternatives) from potential
aquifer restoration measures also depends on many factors. In
developing a set of alternatives, it is necessary to consider the total
system and to include future preventive measures in addition to current
cleanup activities. Because ground water pollution is not strictly a
hydrogeologic problem, proposed solutions require the consideration of a
multi-disciplinary team. A comprehensive preliminary study is necessary
to catalog existing information, thus eliminating duplication of effort;
and to prevent the immediate implementation of poorly planned
alternatives which can be ineffective. The development of a list of
alternative remedial measures, following the preliminary study, should
be based on an iterative process. By iterating through the selection
-------
process, measures can be refined in order to minimize design flaws which
may go undetected in the subsurface environment.
Selection of a single aquifer restoration strategy from a list of
alternative measures requires balanced consideration of the economic,
environmental and risk implications of each alternative. In evaluating
the economics of a remedial measure it is important to include all costs
and to convert all costs to a common time period. Environmental
evaluations should recognize any irretrievable commitments of the
subsurface environment, and the finite life expectancy of some measures
such as liners. Risk assessment is a new field of study and as such,
there are few available methodologies and a dearth of risk data. A
proper risk assessment methodology should utilize minimal amounts of
estimated input data and produce replicable results.
A myriad of techniques are available for aiding decision-making
based on a balanced and systematic consideration of the economic,
environmental and risk features of feasible alternatives. Any technique
used should include public participation as an integral component.
Because of adverse publicity given such sites as Love Canal, Valley of
the Drums, and Times Beach, the public has become extremely concerned
over sites with the potential for contaminating their drinking (ground)
water supplies. A number of procedures for incorporating public input in
decision-making are available.
-16-
-------
SECTION III
RECOMMENDATIONS
Based upon this study of the state-of-the-art of aquifer
restoration, seven recommendations have been formulated as follows:
(1) Systematic Research Program — There is a need for development
of a systematic research program aimed at increased
understanding of the behavior and effectiveness of various
remedial measures. Outlined in Table III.l is an example list
of the most immediate research needs concerning aquifer
restoration measures.
(2) Aquifer Restoration Information Center — The need for a
centralized aquifer restoration information dissemination
center cannot be overemphasized. Currently, information
directly related to ground water quality control and cleanup
is being generated from a large number of widely dispersed
activities. Additionally, extant and informative material can
be found in sources somewhat related but not directly applied
to ground water. An aquifer restoration information center
could centralize available and pertinent information. The
centralized information could be categorized according to
sources, pollutants, remedial measures employed, costs, and
effectiveness. The centralized and categorized information
could be disseminated much more efficiently.
In addition to information on cases of ground water pollution
cleanup there is a need for cataloging the growing number of
professionals and/or companies providing services related to
ground water quality management. This catalog should also be
categorized according to area of experience and expertise.
In addition to national conferences and symposia, there should
be developed intensive short courses or workshops devoted
solely to the technical aspects of aquifer restoration.
Although geographically dispersed, these courses could be
conducted through the aquifer restoration information center.
(3) Monitoring — One aspect of remedial measure design that is
also receiving increased interest is ground water quality
monitoring. Most remedial measures are designed, at least in
part, on the basis of pre-existing monitoring data. All
remedial measures should include, as an integral component of
their design, provisions for monitoring the long-term
effectiveness of the measures employed.
-17-
-------
Table III.l: Examples of Aquifer Restoration Research Needs
Topical Area
Specific Needs
Surface Capping and
Liners
Slurry Walls
1. Development of standardized tests for
assessment of in-place integrity.
2. Development of accelerted testing
procedures for assessing the long-term
performance of liners and seals.
3. Effects of root-penetration on surface
caps.
1. Effects of organics on bentonite —
change in bentonite properties and migra-
tion of organics through bentonite over
time.
Air Stripping of
Volatile Organics
from Ground Water
3.
2. Effects of organics on various bentonite-
cement mixtures — change in properties
of mixtures (e.g., shrink-swell poten-
tial) and migration of organics through
the mixtures over time.
Development of techniques for assessing
the in-place integrity of slurry walls,
(permeability, bottom connections, seep-
age, etc.).
1. Stripping of single compounds vs. mix-
tures of compounds.
2. Stripping characteristics of different
classes of volatile organics.
3. Direct or indirect (fouling of packing
media) interferences of ground water
constituents on stripping. Examples in-
clude metals (Fe, Mn, Cr), inorganic
salts (TDS), X saturation of oxygen, non-
volatile organics, and microbial activity.
-18-
-------
Table III.l: (continued)
Topical Area Specific Needs
4. Effectiveness of batch vs. continuous vs.
combination (batch and continuous)
stripping.
5. Effectiveness of natural desorption vs.
mechanical stirring.
6. Use of Total Organic Carbon (TOC) and
COD as a surrogate indicator of volatile
organics.
7. Effects of concentrations of volatile
organics on stripping.
8. Use of simulated vs. actual ground water
for treatability studies.
9. Optimization of packed tower design in
terms of packing type and size, detention
time, air/water ratio, height of packing,
temperature and variations, moving gas
(air, oxygen, ozone), stripping enhance-
ment via chemical additions, counter-
current vs. cross-current vs. co-current
flow, no packing, single tower vs. towers
in series, scale-up from laboratory
studies to pilot plants to full design,
and development of computer-based design
program.
10. Atmospheric dispersion of released
organics.
11. Cost-effectiveness studies of existing
facilties.
Removal of Organics
via Activated
Carbon Treatment 1. Adsorption of single compounds vs. mix-
tures of compounds.
2. Adsorption characteristics of different
classes of compounds.
-19-
-------
Table III.l: (continued)
Topical Area Specific Needs
3. Direct or indirect (fouling of activated
carbon) interferences of ground water
constituents on adsorption. Examples
include: metals (Fe, M.n, Cr), inorganic
salts (TDS), % saturation of oxygen, non-
volatile organics, and microbial activity.
4. Use of TOC and COD as surrogate indicator
of adsorbable organics.
5. Effects of concentrations of adsorbable
organics on adsorption.
6. Use of simulated vs. actual ground water
for treatability studies.
7. Optimization, of activated carbon, column,
design in terms of: size of carbon,
detention time, column diameter, depth of
carbon, temperature and variations, ad-
sorption enhancement via chemical addi-
tions, single column vs. columns in
series, scale up from laboratory studies
to pilot plants to full design, and de-
velopment of computer-based design pro-
gram.
8. Cost-effectiveness studies of existing
facilities.
-20-
-------
(4) Costs — The costs of remedial measures is an area that is
severely lacking in comprehensive and transferrable
information. Information on the over-all cost-effectiveness
of remedial measures, including the long-term operation,
maintenance and monitoring costs, is needed. A research
effort based on a case study approach should be conducted to
develop cost-effective information.
(5) Remedial Measure Selection — All future remedial measures
should be designed and selected based on the application of a
structured decision-making methodology or protocol. The
methodology should include provisions for public
participation. The approach utilized should be documented and
technically defensible. Progress in the field of remedial
measure design and selection cannot proceed based only on an
analysis of previous cases developed by ad hoc procedures.
(6) Risk Assessment Methodologies — There is a need for
development of risk assessment methodologies that can aid in
evaluating the need for and effectiveness of cleanup measures.
Additionally, methodologies need to be developed that are
simple in theory, easy to apply, and utilize available data.
Methodologies involving complex stochoastic analysis usually
require data that is not available. The utility of the
results from these methodologies is minimal.
(7) Product Recovery — Future remedial action design should be
required to explore the possibility of recovering and
utilizing the ground water pollutants. This has been widely
practiced in cases involving hydrocarbon leakage from storage
tanks. There may exist other instances in which the
pollutants may be economically recovered, especially those
involving spills. Recovery and utilization of the pollutants
could help to defray costs of remedial measures.
-21-
-------
SECTION IV
GROUND WATER POLLUTION CONTROL
The most important information required for any aquifer restoration
study is related to the technical options available for ground water
pollution control. In other words, what, if anything, can be done to
prevent, abate or clean-up polluted ground water. Presented in this
section is information on institutional measures and source control
strategies for preventing/minimizing the occurrence of ground water
pollution. The major section of this chapter addresses physical control
measures such as well and interceptor systems, surface capping and
liners, and physical containment barriers such as sheet piling,
grouting, and slurry walls.
INSTITUTIONAL MEASURES
Institutional measures are primarily useful as preventive measures.
The measures enumerated herein will be most useful to those groups who
are interested in protecting ground water resources. Institutional
measures include regional ground water management districts, source
control strategies, land zoning, effluent charges/credits, aquifer
standards, and guidelines.
Regional Ground Water Management Districts
The establishment of a regional ground water management district
could be useful in prevention of long term quality and quantity-
threatening activities. Possible functions of such a group could
include information dissemination and some regulatory enforcement.
-22-
-------
Information dissemination could include such things as proper use of
agricultural chemicals, waste disposal guidelines, and lists of
reputable ground water professionals such as well drillers. Minimal
enforcement capabilities could help prevent such threatening activities
as overuse or abuse of chemicals, indiscriminant dumping, or overpumping
of certain formations.
Source Control Strategies
Implementation of certain source control strategies can reduce the
threat to ground water resources. Source control measures could include
source reduction steps such as recycling or waste pretreatment prior to
disposal. Another source control strategy would be optimum site
selection, i.e. locating potential threatening activities in the least
vulnerable areas. The next section of this chapter will provide more
information on this subject.
Land Zoning
Related to optimum site selection would be the use of prudent land
zoning. One such measure would be to disallow industrial or waste
disposal activities in ground water recharge areas.
Effluent Charges/Credits
One means of protecting ground water resources is by using
economics as an incentive. By applying charges to effluents applied to
or disposed on the soil, one can indirectly improve the quality of
effluent from certain facilities. Credits can be applied to those
-23-
-------
facilities that move to improve the ground water resources such as those
employing artificial recharge measures.
Aquifer Standards
A more authoritative institutional measure would be to establish
ground water quality standards for aquifers. Beneficial use standards
that are currently used for surface waters could be modified and applied
to ground water resources.
Guidelines
Probably the most immediately effective measure would be to
establish guidelines or standards for construction and operation of
ground water-threatening activities. Many states are now adopting
standards for construction of waste disposal facilities, and this could
be expanded to include oil well drilling, mining and minerals
extraction, on-site sewage disposal, and other activities.
SOURCE CONTROL STRATEGIES
Source control strategies represent attempts to minimize or prevent
ground water pollution before a potential polluting activity is
initiated. The objectives of source control strategies are to reduce
the volume of waste to be handled, or reduce the threat a certain waste
poses by altering its physical or chemical makeup.
Source control strategies are most applicable in the design of new
facilities as a component of one or more ground water pollution
prevention steps. However, some of the strategies can be used as
abatement measures for existing facilities.
-24-
-------
The development and design of each of Che strategies listed below
are site and situation specific; depending very heavily on the type of
waste to be treated. Probably the only criteria universally applicable
to these strategies is that they will require extensive monitoring and
recordation. Most certainly, indiscriminate dumping must become a thing
of the past. Some general advantages of source control strategies
include:
(1) reduces the threat to the ground water environment;
(2) accelerates the time for "stabilization" of waste disposal
facilities; and
(3) offers opportunities for economic recovery.
Two general disadvantages of source control strategies are:
(1) increased capital and maintenance costs; and
(2) monitoring and skilled operator requirements.
Table IV. 1 lists some of the source control strategies that can be
applied for ground water pollution prevention; usually at waste disposal
facilities. The list is not totally comprehensive. It should be
emphasized, however, that the possibility for implementation of source
control strategies should be examined for all future waste disposal
facilities. The following section contains additional information on
stabilization and solidification strategies.
STABILIZATION/SOLIDIFICATION STRATEGIES
One means of reducing the threat to ground water from land disposal
of waste materials is to structurally isolate the waste material in a
solid matrix prior to landfilling. This process, known as
stabilization/solidification, is becoming increasingly popular for
-25-
-------
Table IV.1: Potential Source Control Strategies
I. Volume Reduction Strategies
A. Recycling
B. Resource Recovery
1. materials recovery
2. waste-to-energy conversion
C. Centrifugation
D. Filtration
G. Sand Drying Beds
II. Physical/Chemical Alteration Strategies
A. Chemical Fixation
1. neutralization
2. precipitation
3. chelation
4. cementation
5. oxidation-reduction
6. biodegradation
B. Detoxification
1. thermal
2. chemical — ion-exchange, pyrolysis, etc.
3. biological — activated sludge, aerated lagoons, etc.
C. Degradation
1. hydrolysis
2. dechlorination
-26-
-------
Table IV.1: (continued)
3. photolysis
4. oxidation
D. Encapsulation
E. Waste Segregation
F. Co-Disposal
G. Leachate Recirculation
-27-
-------
hazardous and radioactive waste disposal (U.S. Environmental Protection
Agency, 1979). The objective of solidification/stabilization processes
is to chemically fix the waste in a solid matrix. This reduces the
exposed surface area and minimizes leaching of toxic constituents.
Effective immobilization includes reacting toxic components chemically
to form compounds immobile in the environment and/or entrapping the
toxic material in an inert stable solid. Thus stabilization and
solidification have different meanings, although the terms are often
used interchangeably. From a definitional perspective, stabilization
refers to immobilization by chemical reaction or entrapping (watertight
inert polymer or crystal lattice); while solidification means the
production of a solid, monolithic mass with sufficient integrity to be
easily transported. It will become apparent that these processes may
overlap or take place within one operation. An example is cementation
where the process both stabilizes by producing insoluble heavy metal
compounds, and solidifies into a formed mass while entrapping the
pollutants.
Chemical stabilization is designed to provide a substance which is
more resistant to leaching, and also more amenable to the solidification
process. By chemically fixing the hazardous waste constituents, their
release will be minimized in the event of a breakdown of the solid
matrix. Probably the simplest stabilization process is pH adjustment.
In most industrial sludges, toxic metals are precipitated as amorphous
hydroxides that are insoluble at an elevated pH. By carefully selecting
a stabilization system of suitable pH, the solubility of any metal
hydroxide can be minimized. Certain metals can also be stabilized by
-28-
-------
forming insoluble carbonates or sulfides. Care should be taken to
ensure that these metals are not remobilized because of changes in pH or
redox conditions after they have been introduced into the environment.
Where possible, it is desirable to co-dispose of wastes which stabilize
without the addition of extraneous chemicals.
Stabilized wastes can be solidified into a solid mass by
microencapsulation or macroencapsulation. Microencapsulation refers to
the dispersion and chemical reaction of the toxic materials within a
solid matrix. Therefore, any breakdown of the solid material only
exposes the material located at the surface to potential release to the
environment. Macroencapsulation is the sealing of the waste in a thick,
relatively impermeable coating layer. Plastic and asphalt coatings, or
secured landfilling, are considered to be macroencapsulation methods.
Breakdown of the protective layer with macroencapsulation could result
in a significant release of toxic material to the environment.
Stabilization techniques have concentrated on the containment of
toxic inorganic compounds. This is because many of the techniques
originated as methods for treating radioactive wastes which consist
primarily of inorganic isotopes. Also, organic compounds may interfere
with the stabilization/solidification process, although small amounts
may be incorporated under test conditions. Chemical oxidation or
incineration have been found to be the most successful treatment methods
for the majority of dangerous organic chemicals.
Stabilization/solidification processes should produce a material
whose physical placement will not render the land on which it is
disposed unusable for other purposes. The resultant stabilized/sol-
-29-
-------
idified material should be impervious, with good dimensional stability
and load-bearing characteristics. It should also have satisfactory wet-
dry and freeze-thaw weathering resistance. These properties plus
optimum size and shape will make the material easily transportable under
Department of Transportation (DOT) regulations, when compared with the
precautions necessary when shipping wet wastes or sludges.
The nature of most stabilized industrial wastes precludes their use
on agricultural land. However, they could be suitable for building
subfoundations or as subgrade materials under roads or runways. For
waste products expected to be scarce in the future (chromium, nickel,
and manganese), it may be desirable to make them reclaimable. There is
no optimum stabilization/solidification process applicable to every type
of hazardous waste. Each individual waste must be characterized and
bench tests and pilot studies performed to determine the suitability of
a given process.
Present solidification/stabilization systems can be grouped into
seven classes or processes:
(1) solidification through cement addition;
(2) solidification through the addition of lime and other
pozzolanic materials;
(3) techniques involving embedding wastes in thermoplastic
materials such as bitumen, parafin, or polyethylene;
(4) solidification by addition of an organic polymer;
(5) encapsulation of wastes in an inert coating;
(6) treatment of the wastes to produce a ceraentitious product with
major additions of other constituents; and
(7) formation of a glass by fusion of wastes with silica.
-30-
-------
WELL SYSTEMS
Well systems for ground water pollution control are based on
manipulating the subsurface hydraulic gradient through injection or
withdrawal of water. Well systems are designed to control the movement
of the water phase directly, and subsurface pollutants indirectly. This
approach is often referred to as plume management. There are three
general classes of well systems. Two of these systems, which involve
withdrawal of water, are well point systems and deep well systems. The
other technology involves injection of water and can be referred to as a
pressure ridge system. All three of the systems may require the
installation of several wells at selected sites. These wells are then
pumped (or injected) at specified rates so as to dictate the movement of
the water phase and associated pollutants.
Construction
The single most important construction aspect of well systems is
the installation of the wells. The sequence for installation of wells
involves the following steps:
(1) setting up the drilling equipment;
(2) drilling the well hole;
(3) installing casings and liners;
(4) grouting and sealing the annular spaces between casings and
boreholes;
(5) installing well screens and fittings;
(6) gravel-packing and placing materials; and
(7) developing the well.
-31-
-------
Not all of the above steps are required for each well. Conversely,
in addition to the above steps, the process could also require
construction of above-ground facilities (pumphouses, etc.) and testing
(pump tests).
Several methods exist for the initial drilling of the borehole, and
these methods can usually be listed under one of the headings shown in
Table IV.2. After the borehole is drilled, the next series of steps
involves setting of casings, grouting and sealing of casings, and
installing well screens. Excellent discussions are available on these
procedures (Campbell and Lehr, 1977; and Johnson Division, UOP Inc.,
1975).
The next step in well installation is well development or
completion. Development is the process by which the finer particles of
the water-bearing formation around the well are removed, thus resulting
in enlarged passages for moisture travel. Three beneficial results of
well development are:
(1) correction of damage to, or clogging of, water-bearing
formation which occurs as a side effect from drilling;
(2) increases the porosity and permeability of the formation in
vicinity of the well; and
(3) stabilizes the sand formation around a screened well so that
the well will yield water free of sand (Johnson Division, UOP
Inc., 1975).
The final activity would be the construction of the above-ground
facilities associated with the well. This could include pumphouses,
pumps and piping networks.
-32-
-------
Table IV.2: Water Well Drilling Methods
Equipment Type
Procedure Description
1. Cable Tool
The ctble tool drills by lifting and dropping
a ttring of tools suspended on a cable. A
bit at Che bottom of Che tool string strikes
the bottom of the hole, crushing and breaking
the fornation material. Cuttings are removed
by bailing or recirculation of a slurry.
Casing is usually driven concurrent with
drilling operations (Campbell and Lehr,
1977).
2. Hydraulic Rotary
Drilling
3. Reverse Circulation
Drilling
4. Air Rotary Drilling
5. Air-Percussion
Rotary Drilling
Hydraulic rotary drilling consists of cutting
a borehole by means of a rotating bit and
removing the cuttings by continuous
circulation of a drilling fluid as the bit
penetretes the formation materials (Johnson
Division) UOP Inc., 1973). Drilling fluid is
usually a water-based slurry containing some
fine material (clay) in suspension.
Ssme as (2) except the fluid is forced
downhole outside of the casing and returns up
the inside of the casing. For (2) the route
is down inside the casing and returning up
outside the casing.
Same as (2) only forced air becomes the
drilling fluid.
Rotery technique in which the main source of
energy for fracturing rock is obtained from a
percussion machine connected directly to the
bit.
6. Hollow-Rod Drilling
7. Jet Drilling
8. Driven Wells
9. Hollow Stem Auger
Drilling
Similar to (1) except with more rapid and
shorter strokes. Ball and check valve on bit
allows water and cuttings to enter and travel
up the casing (hollow-rod).
Similar to (6) except water is pumped down
inside casing (hollow-rod) and returns up
outside the casing.
Pipe with a driving plug on the end is driven
into the ground by repeated blows from a
driving weight.
Hells can be drilled in unconsolidated
material and sampled by coring, with no
drilling fluid or air required. Limited to
shallow (less than 100 ft) unconsolidated
material.
-33-
-------
Design
The design of any of the three well systems should be preceded by a
preliminary investigation. This preliminary investigation should
involve a hydrogeologic study of the area. The hydrogeologic study
should generate two categories of information. First, the contaminant
plume dimensions (width, length, depth and general shape) and the
hydraulic gradient across the plume should be determined (Lundy and
Mahan, 1982). Second, the hydrogeologic characteristics of the aquifer
should be identified. This information can be used along with the
concepts of well hydraulics to establish the number and spacing of
needed wells.
Some equations available for analyzing the flow to wells under
different aquifer conditions are given in Tables IV.3 and IV.4 (Cohen
and Miller, 1983). Table IV. 5 summarizes the methods that can be used
to calculate the radius of influence under various aquifer conditions
(Kufs, et al., 1983). Simplistically, the design of a well system
involves an iterative procedure of utilizing a set pumping rate,
calculating the radius of influence associated with that rate and
checking the radius of influence against the plume dimensions.
Lundy and Mahan (1982) identify three constraints which well
recovery systems must satisfy:
(1) the recovery system must be located near enough to the
downgradient plume boundary to reverse the hydraulic gradient
at that boundary;
(2) total discharge is large enough to create limiting flowlines
that bound the plume up to its widest part upgradient of the
well system; and
-34-
-------
Table IV.3: Analytical Solutions of Steady Flow Applicable to
Corrective Actions (Cohen and Miller, 1983)
Equation
Assumptions
Steady flow from a line
source to a line sink
in a confined aquifer
with no recharge
Steady flow in an
unconfined aquifer
bounded by fully
penetrating drains
(h0"hL) with recharge
Steady flow in an
unconfined aquifer
bounded by drains
with recharge
Potential theory model
of steady flow to drains
in an unconfined aquifer
with recharge
USEPA steady state
landfill drainage
model for horizontal
liners or covers
USEPA steady state
landfill drainage
model for sloping
liners or covers
Steady state model for
leachate collection
with a sloping liner
Kirkham's 1967 D-F
model of steady flow
to drains in an
unconfined aquifer
with recharge
1.
fully
line source and sink are
penetrating
2. no recharge occurs to confined aquifer
3. isotropic and homogeneous
4. horizontal flow
5. aquifer thickness varies linearly with x
1. D-F assumptions
2. isotropic and homogeneous
3. constant and uniform infiltration rate
4. horizontal impermeable base
5. fully penetrating drains with equal head
1. D-F assumptions
2. isotropic and homogeneous
3. constant and uniform infiltration rate
4. horizontal impermeable base
1. isotropic and homogeneous
2. constant and uniform infiltration rate
3. horizontal impermeable base
1. D-F assumptions
2. isotropic and homogeneous
3. constant and uniform infiltration rate
4. horizontal impermeable base
5. drain at the impermeable base
1. D-F assumptions
2. isotropic and homogeneous
3. constant and uniform infiltration rate
4. sloping impermeable base
1. isotropic and homogeneous
2. constant and uniform infiltration rate
3. impermeable base
1. D-F assumptions
2. isotropic and homogeneous
3. constant and uniform infiltration rate
4. horizontal impermeable base
5. fully penetrating ditches
-35-
-------
Table IV.3: (continued)
Equation
Assumptions
Hooghoudt's 1940 steady
state drain spacing
formula for an
unconfined aquifer
with recharge
Hooghoudt's steady
state drain spacing
equation for a two
layer unconfined
aquifer
Youngs' steady state
inequality drain
spacing model
Bouwer's model of
steady flow to a
drain in an
unconfined aquifer
with zoned recharge
Steady flow to a
fully penetrating
drain in a leaky
and infinite
confined aquifer
Steady flow to a
fully penetrating
drain in a leaky
and infinite two
layered confined
system
1. D-F assumptions
2. isotropic and homogeneous
3. constant and uniform infiltration rate
4. horizontal impermeable base
5. drains at equal depth
1. D-F assumptions
2. constant and uniform infiltration rate
3. horizontal impermeable base
4. drains fully penetrate upper layer
1. isotropic and homogeneous
2. constant and uniform infiltration rate
3. horizontal impermeable base
1.
2.
3.
1.
2.
D-F assumption
isotropic and homogeneous
constant and uniform infiltration rate
in the recharge zone
horizontal impermeable base
isotropic and homogeneous in each layer
leaky confined aquifer with horizontal
flow
before pumping, the initial artesian
and phreatic water levels are horizontal
and coincident at h^
during pumping, the phreatic water level
is assumed to remain constant at h^
isotropic and homogeneous in each layer
leaky confined aquifer with horizontal
flow
before pumping, the initial artesian and
phreatic water levels are horizontal and
coincident at h^
during pumping, only the phreatic water
level is assumed to remain horizontal
at h^
drain fully penetrates upper confined
aquifer and is of great length
-36-
-------
Table IV.3: (continued)
Equation
Assumptions
Steady flow to a
fully penetrating
drain in an
infinite leaky
unconfined aquifer
1. D-F assumptions
2. isotropic and homogeneous in each layer
3. constant and uniform infiltration rate
4. drain fully penetrates unconfined
aquifer
5. artesian water level is horizontal or
varies linearly with distance; it
remains constant despite changes in the
phreatic water level
Steady flow to
partially penetrating
drains in a leaky
unconfined aquifer
Steady flow to a
fully penetrating
drain in a leaky
confined-unconfined
aquifer system
1. D-F assumptions
2. isotropic and homogeneous in each layer
3. constant and uniform infiltration rate
4. artesian water level remains constant
5. horizontal impermeable base
1. horizontal flow in both aquifers
2. isotropic and homogeneous in each layer
3. constant and uniform infiltration rate
4. horizontal impermeable base to the
confined aquifer
-37-
-------
Table IV.4s Analytical Solutions of Transient Flow Applicable
to Corrective Actions (Cohen and Miller, 1983)
Equation
Assumptions
Transient flow to a
fully penetrating
drain in a semi-
infinite aquifer
caused by a unit
step change in head
Transient flow to a
fully penetrating
drain with constant
discharge in a semi-
infinite aquifer
Transient flow to a
fully penetrating
drain in a finite
aquifer caused by a
unit step change in
head
Transient flow to a
fully penetrating drain
in a semi-infinite
confined aquifer caused
by a unit step change
in head
Transient flow to a
fully penetrating
drain with constant
discharge in a semi-
infinite confined
aquifer
1. D-F assumptions if unconfined
2. isotropic and homogeneous
3. no infiltration
4. horizontal impermeable base
5. instantaneous head change at xs0
6. storage and transmissive properties
constant with time
7. fully penetrating drain
1. D-F assumptions if unconfined
2. isotropic and homogeneous
3. no infiltration
4. horizontal impermeable base
5. constant discharge from drain
6. fully penetrating drain
7. storage and transmissive properties
constant with time
1. D-F assumptions if unconfined
3. isotropic and homogeneous
3. no infiltration
4. horizontal impermeable base
5. storage and transmissive properties
constant with time
6. instantaneous head change at fully
penetrating drain
7. finite aquifer or parallel drain at 2L
1. isotropic and homogeneous confined aquifer
2. no infiltration or leakage
3. instantaneous head change at fully
penetrating drain
4. horizontal impermeable base
1. isotropic and homogeneous confined aquifer
2. no infiltration or leakage
3. constant discharge from fully
penetrating drain
-38-
-------
Table XV.4: (continued)
Equation
Assumptions
Transient flow to a
fully penetrating
drain in an unconfined
aquifer above a leaky
confined aquifer
Glover'8 model of
transient flow to
partially penetrating
parallel drains in an
unconfined aquifer
Brook's modification
of Glover's transient
flow model
Dumm, Tapp and Moody's
model of transient
flow to partially
penetrating drains in
an unconfined aquifer
Maasland's model of
transient flow to
partially penetrating
drains in an
unconfined aquifer
with recharge
Kirkham's potential
theory model of
transient flow to
partially penetrating
drains in an unconfined
aquifer with recharge
1. D-F assumptions for the unconfined aquifer
2. isotropic and homogeneous in each layer
3. no infiltration
4. water level of leaky confined aquifer is
horizontal and constant with time and is
coincident with the initial phreatic water
level
5. storage and transmissive properties are
constant with time
1. D-F assumptions
2. isotropic and homogeneous
3. no infiltration
4. horizontal impermeable base
5. drains at equal elevation
6. storage and transmissive properties are
constant with time
1. D-F assumptions
2. isotropic and homogeneous
3. no infiltration
4. horizontal impermeable base
5. drains at equal elevation
1. D-F assumptions
2. isotropic and homogeneous
3. no infiltration
4. horizontal impermeable base
5. drains at the same elevation
6. initial water level configuration is a
parabola
1. D-F assumptions
2. isotropic and uniform infiltration rate
3. horizontal impermeable base
4. drains at the same elevation
5. storage and transmissive properties are
constant with time
1. isotropic and homogeneous
2. constant and uniform infiltration rate
3. horizontal impermeable base
4. storage and transmissive properties are
constant with time
5. initial phreatic water level taken as the
equilibrium condition during steady state
-39-
-------
Table IV.4s (continued)
Equation
Assumptions
Terzidis' drain spacing
1. D-F assumptions
model for transient
2. isotropic and homogeneous
flow to partially
3. no infiltration
penetrating drains in
4. horizontal impermeable base
an unconfined aquifer
5. storage and transmissive properties are
constant with time
6. initial water table is horizontal
Van Schilfgaarde's D-F
1. D-F assumptions
drain spacing equation
2. isotropic and homogeneous
for transient flow to
3. no infiltration
partially penetrating
4. horizontal impermeable base
drains in an
5. storage and transmissive properties are
unconfined aquifer
constant with time
6. initial water table is an ellipse
Van Schilfgaarde1s D-F
1. D-F assumptions
drain spacing equation
2. isotropic and homogeneous
for transient flow to
3. no infiltration
fully penetrating
4. horizontal impermeable base
drains in an
5. fully penetrating drains
unconfined aquifer
6. initial water table is an ellipse
7. storage and transmissive properties
constant with time
Transient flow to
1. D-F assumption
fully penetrating
2. isotropic and homogeneous
drains in a sloping
3. no infiltration
unconfined aquifer
4. sloping impermeable base
with no recharge
5. initial water table parallel to the
sloping base
6. storage and transmissive properties
are constant with time
7. vertical seepage faces are located
at each drain
Hantush's 1967 D-F
1. D-F assumptions
model of the growth
2. isotropic and homogeneous
and decay of water
3. infinite in areal extent
table mounds in
4. horizontal impermeable base
response to uniform
5. rectangular recharge area receiving
percolation
constant and uniform infiltration
6. storage and transmissive properties
remain constant with time
-40-
-------
Table IV.5: Methods for Calculating Radius of Influence (Kufs, et al.,
1983)
Pumping Condition Method
Equilibrium
„ _ K(H2-hw2) . , _ T(H-hw) . ,
Exact In R0 =¦ —q— + In Rq ¦ + In rw
- Approximate Rq » 3(H-hw) (0.47K)^
Non-Equilibrium
- Exact Drawdown vs. log distance plots
' Tt ,
'4790 S;
- Approximate Rq - rw + (,-%!: J**
Ro " radius of influence (ft)
K ¦ permeability (gpd/ft2)
H ¦ total head (ft)
hw ¦ head in well (ft)
Q * pumping rate (gpm)
rw ¦ well radius (ft)
T * transmissivity (gpd/ft)
t ¦ time (min)
S * storage coefficient (dimensionless)
-41-
-------
(3) drawdowns of water levels resulting from withdrawals from the
system do not exceed limits that are a function of aquifer
saturated thickness.
Well point systems involve a number of closely-spaced, shallow
wells each connected to a main pipe (header) which is connected to a
centrally-located suction lift pump. Well point systems are used only
for shallow, water table aquifers because the maximum drawdown
obtainable by suction lift is the difference between the suction limit
of the pump and the distance from the static water level to the center
of the pump. The upper limit for lifting water by suction lift is
around 25 feet. Pumping by suction lift is depicted in Figure IV. 1
(Johnson Division, UOP Inc., 1975). Well point systems should be
designed so that the radius of influence (drawdown) of the system
completely intercepts the plume of contaminants as shown in Figure IV.2.
Deep well systems are similar to well point systems except they are
used for greater depths and are usually pumped individually. Once
again, the deep well system should be designed so that its radius of
influence completely intercepts the contaminant plume as shown in Figure
IV. 2.
Pressure ridge systems can be conceptualized as the inverse of the
previous two systems. Pressure ridges use the injection of water to
form an upconing of the water table which acts as a barrier to ground
water flow. Figure IV. 3 depicts the phenomena of a pressure ridge
system. These systems have found their widest application in coastal
areas to prevent saltwater intrusion.
The basis of design for all three of the well systems is that
movement of the ground water flow and pollutant be completely
-42-
-------
T
Depth to
static water
level
Static level
Suction
head
Drawdown
Pumping level
t
_ well
Screen
Note: When pumping wells by suction lift,
the maximum drawdown obtainable is
the difference between the suction
limit of the pump and the distance
from the static water level to the
center of the pump.
Figure IV.1: Pumping by Suction Lift (Johnson Division, UOP Inc., 1975).
-43-
-------
To Surface
Treatment
and/or
Discharge
Source of Pollutant
7777777//
Original Water Table
Drawdown que to pumping
Note: System
also draw# an
(excess) of non-
polluted water»
Figure IV.2: Principle of Withdrawal Wells
-44-
-------
Injection
of f
fresh watery.
Source of Pollutant
Upconlng of water table
Original water table
Ground water flow
Figure IV.3: Principles of Pressure Ridge System
-45-
-------
controlled. Through the use of data obtained from the initial
hydrogeologic study, and the principles of well hydraulics, the number,
depth, spacing, and pumping or injection rate of the wells can be
specified. In addition, estimates of the time required for the remedial
actions can be made.
Application in Hydrocarbon Recovery
The discussion of the three types of well systems above has been
general so as to be applicable to a wide variety of situations.
However, there is one situation in which the purpose of the well system
is very specific. That situation is when the contaminant is not water
soluble and tends to float on top of the water table; this most often
involves hydrocarbons. A unique aspect of this type of problem is that
the contaminant provides a strong economic incentive for ground water
cleanup since the product recovered from the water table can be
utilized. Because of this economic incentive and the large number of
instances of hydrocarbon leakage from storage tanks, considerable work
has been directed toward developing hydrocarbon recovery systems.
Blake and Lewis (1982) have outlined four design options available
for hydrocarbon recovery systems: (1) single pump systems utilizing one
recovery well; (2) single-pump systems utilizing multiple wells; (3)
two-pump systems utilizing two recovery wells; and (4) two-pump systems
utilizing one recovery well. The single-pump, single well system is
depicted in Figure IV.4. The advantages of this system include low
equipment and drilling costs. However, this system produces an oil-
water mix requiring separation of the surface. One-pump systems
-46-
-------
.FLOAT
SWITCH
OIL " WATER
SEPARATOR
METAL ROD
NATIVE
BACKFILL
CASING
LO<
PROOUCT
SUBMERSIBLE
PUMP
CONTINUOUS SLOT
SCREEN
Figure IV.4: Schematic of One Pump System Utilizing a Submersible
Pump and Float Controls (Blake and Lewis, 1982).
-»47-
-------
utilizing multiple well points are plagued by the same problems as the
single~well arrangements, but they may be the most feasible option in
low permeability formations (Blake and Lewis, 1982).
The two-pump, two-well system is depicted in Figure IV.5. The
deeper well serves the purpose of producing a gradient toward the
shallower (product recovery) well. The advantage of this system is the
ability to keep the product and the produced water separate; one
disadvantage is the increased cost of multiple wells. Additionally,
this type of system can require fairly sophisticated surface monitoring
equipment to insure proper operation.
The most desirable product recovery system is that involving two
pumps in one well as depicted in Figure IV.6. The basis of this system
is similar to the previous system, i.e., the lower pump produces a
gradient toward the well and the upper pump removes the accumulated
product. This system separates the product from the produced water, but
only involves a single well. This system can also be fully automated.
The disadvantages of this system include: the increased well diameter
to house two pumps; expensive operational monitoring equipment; the
system requires careful start up operations and calibration; and the
requirements for frequent inspections.
C08tS
The total cost of a well system is dominated by two individual cost
items: 1) the cost of the installation, and 2) operation and
maintenance costs. In this time of rising energy costs, the power
required to pump the well system will probably be the single most
¦48-
-------
H / ' H Mtmmoi tpocinQ
Product
Control
ttorooo
ditenoroo
V/
WATER TABLE
DEPRESSION WELL
PRODUCT RECOVERY
WELL
/n— Sool (o« rtquiftd)
Notivt Bochfill
.1 -t^GfOvti Pock
PRODUCT
Product Pump
ftoduc* D«ttc'iOft
Pro6f
Product Ottoction
^OM
Continuou* Slot Sertoli
Wotor Pump
Figure IV.5:
Schematic of Two Pump System Utilizing Two Small
Diameter Wells (Blake and Lewis, 1982).
-49-
-------
Wo»«r ^
Oitchorgc ——'¦
Wottr Pump JCj—Jm
CgHtroli — |
Stol (ot rtqwiftri)
Noiivt Backfill
BO
Product D«ttct>on
Prob«
Product Detection
Probe
Corrti«go«» Slot
Screw
Figure IV.6: Schematic of Two Pump System Utilizing One
Recovery Well (Blake and Lewis, 1982).
50-
-------
dominating cost factor. A cost analysis study for a variety of well
systems was conducted by Gibb (1971) and included in Campbell and Lehr
(1977). The actual cost figures are somewhat dated, but the principles
employed would be applicable today. A schematic of a well completed in
unconsolidated sediments (sand and gravel) is shown in Figure IV.7, with
corresponding cost relationships shown in Figure IV.8. Similar
information for a gravel-packed well in consolidated to partly
consolidated sediments (sand and gravel) is shown in Figures IV.9 and
IV.10; for a shallow well in consolidated sediments (sandstone,
limestone, or dolomite) in Figures IV.11 and IV.12; and for a deep well
in consolidated sediments in Figures IV.13 and IV.14 (Campbell and Lehr,
1977).
Advantages and Disadvantages
The use of well systems is presently, and probably will continue to
be, the most utilized method of ground water pollution control. This is
not without good reason. Well systems represent the most assured means
of controlling subsurface flows contaminants. Engineers and
hydrogeologists have a somewhat firmer understanding of the mechanics of
well hydraulics than they do of some potential aquifer restoration
strategies. However, well systems are not without their limitations. A
comparison of some of the advantages and disadvantages of well systems
is in Table IV.6.
INTERCEPTOR SYSTEMS
Interceptor systems involve excavation of a trench below the water
table, and possibly the placement of a pipe in the trench. Subsurface
-------
CLAY
o
CN
Q.
0)
"O
E
3
E
SAND
LENSES
c
t>
"O
k.
3
-O
k.
0)
>
o
>>
*
TS
*»
• FINE SAND .* •
12"
CEMENT
OR
CONCRETE
GROUT
STEEL
CASING
LEAD • *
PACKER* *
SCREEN
SAND £ GRAVEL
o • • •••*/• • * ' X /
© . . ® . • . •• ^ ( \° • •
• / _ . . A % ^ ? O • •
' o / V ' «WE3V . • . .
Figure IV.7: Well Completed in Unconsolidated Sediments
(Sand and Gravel) (Campbell and Lehr, 1977).
-52-
-------
100
80
"I 1 1 !—I I I I
SAMPLE cost estIHATE
Assumptions:
We I I depth * 90 feet
Desired yield ¦ 50 9pm
WeM diameter • 6 inches
iSatMif 411
Est imated we 11 Cos t:
Expected range $2950 - $3200
"Best Pit" $3080
BOTTOM BORE HOLE
DIAMETERS IN INCHES
12-15
W.C.* 850d
BEST PIT LINE
80% CONFIDENCE LIMITS
W.C.« 800d° "
I I
60 80 100 200
DEPTH OP WELL (d) IN FEET
Figure IV.8: Cost of Cased Wells Finished in Sand and Gravel
(Campbell and Lehr, 1977) .
-53-
-------
SAND £
LENSES- o-
. .-o
• * •§
C
"i
e
"j
*o
u
3
JO
u
O
>
o
' /
I
\
/
\
/
CLAY
• - •
• • •
o 0
o
/ •
* . °0
• 1
• . ,0-'
FINE SAND
• / • •
• ©
/ . • • ®
• • • o
• .* \ * -
• • • °o
/. ' •»
-1-
* t • • .
•
2V*i
20'
4
V '
drill
CUTTINGS
OR
CLAY
STEEL
CASING
(20* min.)
CEMENT
OR
CONCRETE
GROUT
(20' min.)
STEEL
CAS ING
COUPLING
GRAVEL
PACK
SCREEN
Figure IV.9: Gravel-Packed Well Completed In Sand and Gravel - Uncon-
solidated to Partly Consolidated Sediments (Campbell and
Lehr, 1977).
-54-
-------
100
80
60
1 1—i—i—i—m
SAMPLE COST ESTIMATE
Assumptions:
We 11 depth * 110 fee t
Desired yield • 200 gpm
Well diameter - 8 inches
Gravel pack annulus * 6 inches thick
Bore hole diameter » 20 inches
Est i mated We 11 Cos t:
Expected range S'<250 - 55000
"Best fit" S"«600
BOTTOM BORE HOLE
DIAMETERS IN INCHES
36 -« 2
W.C.- 890d
2 - 3 ••
16-20
0.<>82
W.C. » 680d
O.M)8
W.C." 680d
BEST FIT LINE
80t CONFIDENCE LIMITS
60 80 100 200
PERTH OF WELL (d) IN FEET
Figure IV.10: Cost of firavel-Packed Wells Finished in Sand
and Gravel (Campbell and Lehr, 1977).
-55-
-------
— in
0
/
o
•
ID *A
~ \ o «**
' • \ m V
V V
M C
.GLACIAL * w
-o u *
DRIFT £
c —
IQ X
* * \ c
O - '-X
- • u
¦
«J <-»
\ ./ , / _
o *-
O £
u —
~ • y • N "
¦o u
O -O
• o • o•
jO w •
O >*
^ 0 • "O
c c
\ 'W
•— 3
— r — E
Z_ ' /¦ I
D> M
/ / ¦-
(A ff W ,
- 7 . =
10 — t>
U U t>
-j SHALE • e
.
4-» .c
« W o
/, / i
t> — f*\
, * 5 A
T
/
• SANDSTONE,
LIMESTONE J"
«-lO'
DOLOMITE
D
/ . / . / ; /
/ ~ / , /
/ * / /
-------
i—i—i—r~n 1
sample cost ESTIMATE
11
BOTTOM BORE HOLE
diameters IN INCHES
Assumptions:
Oolomi te Wei I
Well depth * 290 feet
Desired yield ¦ 500 9pm
Well diameter ¦ 10 inches
Es t i ma ted We I I Cost
Expected range SJ3QQ - S'OQO
Best fit" 53700
I.Ml
W.t.« 0.578d
BEST FIT 11NC
80* CONFIDENCE LIMITS
60 BO 100
200 WO 600 800 1000 2000
DEPTH OF WELL (d> IN FEET
Fieure IV.12: Cost of Shallow Sandstone, Limestone, or Dolomite
Bedrock Wells (Campbell and Lehr, 1977).
-57-
-------
O X
O / j
' "DRIFT oN i
\ ° V ' ' S ^
— DOLOMITE r
/ / /
/ / / r
/ / / /\<
SHALE
• I nrw rsu i re I
DOLOMITE If
AND
sT^LIMESTONE • ~
'.'jlUsTLlZ
* • "*..** •
*. SANDSTONE * *
M'-j! STEEL
Mr*" CASING
[J DRILL
CUTTINGS
L OR CLAY
^-STEEL
CASING
CEMENT
OR
-CONCRETE
GROUT
STEEL
CASING
OPEN
HOLE
«• 19"-
/ » / * /
DOLOMITE
/ / AND
• . SANDSTONE .
/ t / ,
/ • • * 7—-*¦ .
» * y • • • •
SANDSTONE • —*
• * . • / • *
• ' • >
J 6'
STEEL
CASING
b/
OPEN
^~HOLE
>5"
Figure IV.13: Deep Well Completed in Consolidated Sediments
(Campbell and Lehr, 1977).
-58-
-------
I I
SAMPLE COST ESTIMATE
Assumpt tons:
Well depth ¦ 1380 feet
Desired yield » 2000 gpm
Well diameter • IS inches
Estimated Wei 1 Cost
Expected range SJ^.BOO - S')7,000 - . .
'Best fit" S""0,000
BOTTOM BORE HOLE
DIAMETERS IN INCHES
15-19
W.C.- 0.0z9d
BEST FIT LINE
— SO* CONFIDENCE LIMITS
100
200
-------
Table IV.6: Advantages/Disadvantages of Well Systems
Advantages
1. Efficient and effective means
of assuring ground water
pollution control.
2. Can be installed readily.
3. Previously installed monitoring
wells can sometimes be employed
as part of well system.
4. Can sometimes include recharge
of aquifer as part of the
strategy.
5. High design flexibility.
6. Construction costs can be lower
than artificial barriers.
Disadvantages
1. Operation and maintenance costs
are high.
2. Require continued monitoring
after installation.
3. Withdrawal systems necessarily
remove clean (excess) water
along with polluted water.
4. Some systems may require the
use of sophisticated mathematical
models to evaluate their
effectiveness.
5. Withdrawal systems will usually
require surface treatment prior
to discharge.
6. Application to fine soils is
limited.
-60-
-------
drains or trenches function similarly to an infinite line of extraction
wells, that is, they effect a continuous zone of depression which runs
the length of the drainage trench (Kufs, et al., 1983). Figure IV. 15
depicts an interceptor trench. Usually the trench receives perforated
pipe and coarse backfill material for efficiency of collection and
transport of incoming water. There are several names applied to
interceptor systems; examples include their classification as preventive
measures (leachate collection systems) or abatement measures
(interceptor drains).
Construction
a. Collector Drains
Construction of interceptor systems is relatively simple and
involves digging a system of trenches, placing perforated pipes in the
trenches, and backfilling with a coarse material such as gravel.
Usually collector systems will consist of a series of feeder pipes
(laterals) flowing into a main collecting pipe (header) as shown in
Figure IV. 16. From the header, the collected water will flow to a sump
where it is then pumped to the surface for treatment and/or discharge.
Collector systems used in a preventive mode, such as a leachate
collection system, usually include a series of 4 to 6 inch laterals
draining into 8 to 12 inch headers, all arranged in either a herringbone
or gridiron design as shown in Figure IV.17.
Interceptor systems used in an abatement mode can be classified as
relief drains or interceptor drains. Relief drains are installed in
areas where the hydraulic gradient is relatively flat. They are
-61-
-------
. original. water
hydraulic gradxen
I '/ I 4
Interceptor system
lrT~ ^ .. 1 V-- -1 ^ VV1^
Figure IV.15: Hydraulic Gradient Toward Interceptor System
-62-
-------
r
lateral
f
to »ump
Plan
Figure IV.16: Collector Drain System
-63
-------
I
CJN
¦>
I
(a) Herringbone
(b) Gridiron
Figure IV.17: Collector System Layouts
-------
generally used to lower the water table beneath a site, or to prevent
contamination from reaching a deeper, underlying aquifer. Relief drains
are installed in parallel on either side of the site such that their
areas of influence overlap and contaminated ground water does not flow
between the drain lines. They can also be installed completely around
the perimeter of the site (Kufs, et al., 1983).
Interceptor drains are used to collect ground water from an
upgradient source in order to prevent leachate from reaching wells
and/or surface water located hydraulically downgradient from the site.
They are installed perpendicular to ground water flow. A single
interceptor drain located at the toe of the landfill, or two or more
parallel interceptors may be needed, depending upon the circumstances
(Kufs, et al., 1983).
b. Interceptor Trenches
Interceptor trenches can be either active (pumped) or passive
(gravity flow). Active systems will have intermediately-spaced vertical
removal wells or a perforated, horizontal removal pipe (collector drain)
in the bottom of the trench. Active systems are usually backfilled with
a coarse sand or gravel to promote wall stability. Passive systems are
used almost exclusively for collection of pollutants that flow on the
surface of water (e.g., oil and hydrocarbons). Passive systems are
usually left open with the installation of a skimming pump for removal
of the pollutant only.
All interceptor trenches require excavation to at least 3 or 4 feet
below the water table to prevent the escape of inflowing pollutant and
-65-
-------
to speed up the inflow of further free pollutant (American Petroleum
Institute, 1980). In active systems, the pumping capacity should be
great enough to keep water drawn down to the bottom of the ditch.
Additionally, pumping (active system) or skimming (passive system) must
be continuous, or the collected pollutant will tend to seep into the
trench walls and continue downgradient (American Petroleum Institute,
1980).
The main construction activity associated with interceptor trenches
is excavation of the trench. This can be accomplished with conventional
equipment. The trench width should be wide enough to accomodate pipes
or pumps if needed, yet it should be kept narrow to minimize the amounts
of soil removed. Active (pumped) systems will require greater depths,
once again to accomodate pipes or pumps.
It has been suggested (American Petroleum Institute, 1972, 1980)
that the downgradient wall of the trench be lined with an impermeable
material such as polyethylene film to prevent floating pollutants from
passing through the trench and continuing downgradient. However,
Pastrovich, et al. (1979) indicates that the pollutant will tend to find
its way around the ends of the barrier and penetrate the ground beyond
while, at the same time, equalization of the hydrostatic pressure on
both sides of the film tend to cause it to float away from the wall. As
noted above, continuous removal of the pollutant and water from the
trench will cause drawdown on both sides of the trench, thus preventing
further migration of the pollutant.
After excavation of the trench, the passive system will usually
require installation of a skimming pump. Active systems will require
-66-
-------
the installation of removal wells or collector drains, with subsequent
backfilling of the trench with coarse sand or gravel.
Design
a. Collector Drains
The primary design parameters for a collector drain systems are (1)
the size of the pipe; and (2) the spacing of the pipe. To determine the
size of pipe required, the carrying capacity of the pipe is assumed to
be equal to the design seepage, and the pipe velocity is described by
Mannings equation. The resulting equation in a rearranged form is
(Luthin, 1957):
d - 0.892 (SC)°«375 (A)0»375 a - 0.1875
where
d ¦ inside diameter of pipe (inches)
A m drainage area (acres)
SC - seepage coefficient (inches/day)
s ¦ hydraulic gradient
For purposes of design of drain spacing, from the Dupuit-Fochheimer
theory the ground water surface between two tile lines may be taken as a
portion of an ellipse as shown in Figure IV.18 and described as follows
(Spangler and Handy, 1973):
S2 _ 4k
(b2 - a2) R
where
S "tile line spacing (meters)
a " depth to impermeable barrier under tile (meters)
b ¦ depth to impermeable barrier halfway between tiles (meters)
-67-
-------
Ground surface
Water table
/
¦o-" «
Ground surface
Water table
t m t m r; t i n 111 it [ n i m i n m h m 1111 f^i 11
(b) Impermeable
layer
Figure IV.18: Shape of Water Table Adjacent to Tile Drains.
(a) Relatively Permeable Soil; (b) Less Perm-
eable Soil (Spangler and Handy, 1973).
-68-
-------
k = hydraulic conductivity of soil (meters/day)
R * infiltrating rainfall (meters/day)
A wide variety of equations for parallel subsurface drains have
been developed for varying aquifer conditions. These equations are
outlined in Tables IV.3 and IV.4.
b. Interceptor Trenches
Design of interceptor trenches is different from that of relief
drains. In order to decide where to position the trench to intercept
the desired flow, the relationship between depth and flow, and the
upgradient and downgradient influence of the trench must be known. The
upgradient influence can be determined from the following (Kufs, et al.,
1983):
Du a 4/3 H tan
where
Du ¦ effective distance of drawdown upgradient (meters)
H * saturated thickness of the water-bearing strata not affected
by drainage (meters)
cf) » angle between the initial water table or ground surface and
the horizontal plane.
The theoretical expression for the downgradient influence is as
follows:
Dd ¦ ^ ^an 4* (hj - + D2)
where
Dd ¦ distance downgradient from the drain where the water table is
lowered to the desired depth (meters)
K - hydraulic conductivity (meters/day)
q ¦ drainage coefficient (meters/day)
-69-
-------
» effective depth of the drain (meters)
h£ ¦ desired depth to the water table after drainage (meters)
J>2 m distance from the ground surface to the water table before
drainage at the distance downgradient from the drain
(meters)
COS t3
Shallow drainage by trenches or collector drains has been estimated
to cost on the order of $1235 to $1730 capital cost per hectare drained
($500 to $700 per acre), depending on the depth of placement and
materials used. Labor and maintenance costs are about $1500 per year
(Tolman, et al., 1978).
Advantages and Disadvantages
Listed in Table IV. 7 are some of the advantages and disadvantages
particular to subsurface drains (interceptor systems). The most obvious
advantage of interceptor systems is their relatively simple construction
methods. Just as obvious is the disadvantage of requiring continuous
monitoring and maintenance. Other advantages of interceptor systems
are:
(1) not only easy but also inexpensive to install;
(2) useful for intercepting landfill side seepage and runoff (U.S.
Environmental Protection Agency, 1982);
(3) useful for collecting leachate in poorly permeable soils (U.S.
Environmental Protection Agency, 1982);
(4) large wetted perimeter allows for high rates of flow (U.S.
Environmental Protection Agency, 1982);
(5) possible to monitor and recover pollutants; and
(6) produces much less fluid to be handled than well-point
systems.
-70-
-------
Table IV.7: Advantages/Disadvantages of Subsurface Drains (U.S. Environ-
mental Protection Agency, 1982)
Advantages
1. Operation costs are relatively
cheap since flow to underdrains
is by gravity.
2. Provides a means of collecting
leachate without the use of
impervious liners.
3.
4. Systems fairly reliable,
providing there is continuous
monitoring.
5. Construction methods are simple.
Disadvantages
1. Not well suited to poorly
permeable soils.
2. In most instances it will not
be feasible to situate under-
drains beneath an existing
site.
3. System requires continuous and
careful monitoring to assure
adequate leachate collection.
Considerable flexibility is avail-
able for design of underdrains;
spacing can be altered to some
extent by adjusting depth or
modifying envelope material.
-71-
-------
Disadvantages of interceptor systems include:
(1) when dissolved constituents are involved, it may be necessary
to monitor ground water downgradient of the recovery line
(Quince and Gardner, 1982);
(2) open systems require safety precautions to prevent fires or
explosions;
(3) interceptor trenches are less efficient than well-point
systems;
(4) operation and maintenance costs are high;
(5) not useful for deep disposal sites (U.S. Environmental
Protection Agency, 1982); and
(6) may interfere with use of land.
SURFACE WATER CONTROL, CAPPING AND LINERS
Surface water control, capping and liners are three different
technologies that are usually used in conjunction with each other. The
three technologies serve different purposes in preventing ground water
pollution. Surface water control measures are designed to minimize the
amount of surface water flowing onto a site, thus reducing the amount of
potential infiltration. Capping of a site is designed to minimize the
infiltration of any surface water or direct precipitation that does come
onto a site. Impermeable liners provide ground water protection by
inhibiting downward flow of low quality leachate and/or attenuating
pollutants by adsorption processes. These technologies are most often
used as preventive measures, although surface water control and capping
have also been used successfully to abate ground water pollution
problems (Emrich, Beck and Tolman, no date). Technologies also exist
for placement of liners after a polluting activity has been initiated;
-72-
-------
these usually involve the use of grouts which are discussed in a
subsequent section.
Surface water control measures represent a relatively inexpensive
means of minimizing future ground water pollution problems, and as such,
should be part of any current site design. Surface water run-on is
easily minimized by using standard civil engineering techniques such as
diversion berms and drainage ditches (Smith, 1983). The design of a
site utilizing surface caps or impermeable liners, or both, is based on
a water balance for the site. Depicted in Figure IV.19 are the
individual components of moisture at a site.
There are four combinations of capping and liners that could be
used (in conjunction with surface water control measures) at a waste
site, with these combinations depicted in Figure IV.20. The first type
is called a natural attenuation site and utilizes neither capping or
lining; it relies entirely on the pollutant attenuating capacity of the
native soil. The second type of system utilizes an engineered liner
(impermeable liner) to maximize the amount of leachate collected and
minimize the amount escaping. The engineered cover (impermeable cap)
sites are designed to minimize infiltration into the waste source by
maximizing surface runoff and evapotranspiration. The combination
engineered cover and liner (total containment) sites are designed to
minimize infiltration and maximize the leachate collected, thus all but
eliminating the escape of any leachate. The advantages and
disadvantages of each type of system are outlined in Table IV.8.
Emrich, Beck and Tolman (no date) summarize a study of top-sealing
a landfill to minimize leachate generation. Applications of liner
-73-
-------
WATER BALANCE FOR ENGINEERING DESIGN
I P i PRECIPITATION
ET « EVAPO - |
TRANSPIRATION f R/O « RUNOFF
GWj =
GROUND WATER
INFLOW
t
PERC « PERCOLATION
WT=WASTE MOISTURE STORED
QUANTITY
COLLECTED
QUANTITY
RELEASED
MASS BALANCE
PERC + GWr -WT =Qc + Qr
^EXISTING
GROUND
Figure IV.19: Moisture Components of Landfills (Smith, 1983).
-74-
-------
PERMEABLE
COVER
Qe ¦ 0
PERMEABLE
COVER
IMPERMEABLE
LINER
Natural attenuation
Leachate control minimal
Address impacted environment -
hydrogeological environment must
be able to assimilate leachate
without adverse impact
Percolation very high, very
high, Qc very small
QearlOO%
QrsO
Impede flow - collect and treat
leachate
Address impacted environment -
impact of hydrogeological environ-
ment small. Place large emphasis
on leachate treatment
Percolation very high, very
small, very large
Permeable
cap
Oe • 0
Natural attenuation
Address impacted environment -
lessens impact of natural atten-
uation design
Percolation very low, Qr very low,
Q reduced
K
DUCEO
Permeable
cap
impermeable
liner
Qe 100%
Qf,«GREATUr REDUCED^0
Impede flow - collect and treat
leachate
Address impacted environment -
impact on hydrogeologic environ-
ment low and also lessens leachate
treatment impacts
Percolation very low, Q low,
greatly reduced
F*gure IV.20: Four Combinations of Capping and Liners (Smith, 1983).
-75-
-------
Table IV.8: Advantages/Disadvantages of Capping and Liners (Smith, 1983)
System (a)
Advantages
Disadvantages
Natural Attenuation
"No leachate collection, transport
and treatment costs.
"Reduced construction costs.
"Requires unique hydrogeologic setting.
"Regulatory acceptance difficult to obtain.
"Long-tennliabilities.
Engineered Liner
"Lessens hydrogeologic impact.
""Clay-bowl" effect.
"Allows waste to stabilize quickly.
"Increased construction costs.
"Chance for surface discharge.
Engineered Cover
"Lessens hydrogeologic impact after
closure.
"Reduces construction costs relative
to liners.
"Increases closure costs.
"No leachate control during site operations.
"Long-term monitoring and land surface
care.
Engineered Cover
and Liner
"Lessens environmental impacts.
"Minimizes post closure leachate
collection, transport and treatment
costs.
"Politically/socially acceptable.
"High cost for engineering and construction.
"Need high quality clay or synthetic material.
"Lengthens time for waste stabilization.
(a) See Figure IV.21
-------
technology are numerous and grow every year. Shultz and Miklas (1982)
give an excellent discussion of the activities at fourteen lined
facilities. Due to their potential for use in aquifer restoration
projects, the emphasis herein will be on surface caps.
Construction — Surface Capping
In general, the steps involved in the construction of an engineered
cover site include:
(1) closure of facility for future receipt of waste materials;
(2) compaction of waste material and laying and working of
suitable subbase;
(3) placement of impermeable membrane (surface cap);
(4) placement and working of suitable cover material over cap;
(5) revegetation; and
(6) regrading.
The key steps involved with surface capping are the placement of
the impermeable membrane over the source of contamination, and the
revegetation and regrading of the site. The regrading of the site
should be done so as to maximize surface runoff and channel it away from
the site. Revegetation allows for evapotranspiration of water that
soaks into the cover soil, with surface capping being designed to
prevent any remaining water from infiltrating into the waste and moving
down to the ground water. Additional measures that should be included
at most sites include installation of gas vents, and monitoring
facilities for both surface water runoff and ground water.
Fung (1980) provides some general recommendations for layering the
surface cap, assuming it is some type of soil or workable material, from
-77-
-------
the bottom up. These recommendations are:
(1) make buffer layer below barrier (cap) thick and dense enough
to provide a smooth, stable base for compacting;
(2) compact all layers except topsoil and top lift of upper
buffer;
(3) in barrier layer, strive for 90% of maximum density according
to 5- or 15-blow compaction test;
(4) cover barrier layer soon enough to prevent extensive drying;
(5) provide sufficient design thickness to assure performance of
layer function; specifying a 6 in. to 12 in. minimum should
prevent excessively thin spots resulting from poor spreading
techniques;
(6) construct in units small enough to allow rapid completion; and
(7) consider seeding topsoil at time of spreading.
Surface Capping — Selection and Design Considerations
The materials used for covering or capping are many and varied.
Fung (1980) groups these materials into natural soils, commercial cover
materials, and waste materials. Soil has traditionally been used as a
cover material at waste disposal sites. The cover has to fulfill
several functions, and Table IV.9 summarizes the types of soil
classified according to the Unified Soil Classification System (USCS) in
terms of several functions (Fung, 1980). The rating of soils
accomplished in Table IV.9 is based on the following points, with the
functional ratings from I (best) to as many as XIII (poorest).
Quantitative parameters are also provided wherever available, so that
the absolute position of a particular soil rating can be seen (Fung,
1980).
(1) The first column in Table IV.9 concerns the Go-No Go aspect of
trafficability. The ratings of I to XII from coarse granular
-78-
-------
Table IV.9: Ranking of USCS Soil Types According to Performance of Cover Functions (Fung, 1980).
(1)
USCS Go-No Go
Symbol Typical Soils (RCI Value)*
GW Well-graded gravels, gravel-sand I
mixtures, little or no fines (>200)
GP Poorly graded gravels, gravel- I
sand mixtures, little or no (>200)
fines
GM Silty gravels, gravel-sand-silt III
mixtures (177)
GC Clayey gravels, gravel-sand-clay V
mixtures (ISO)
SW Well-graded sands, gravelly I
I sands, little or no fines (>200)
--J SP Poorly graded sands, gravelly I
I sands, little or no fines (>200)
SM Silty sands, sand-silt mixtures II
_ _ (179)
SC Qayey sands, sand-clay mixtures IV
(157)
ML Inorganic silts and very fine IX
sands, rock flour, silty or (104)
clayey fine sands, or clayey
silts with slight plasticity
CL Inorganic clays of low to medium VII
plasticity, gravelly clays. (Ill)
sandy clays, silty days, lean clays
OL Organic silts and organic silty X
clays of low plasticity (64)
MH Inorganic silts, micaceous or VIII
diatomaceous fine sandy or silty (107)
soils, elastic silts
CH Inorganic days of high plasticity, VI
fat days (145)
OH Organic days of medium to high XI
plasticity, organic silts (62)
PT Peat and other highly organic soils XII
(46)
(2)
(3)
(4)
(S)
(6)
(7)
TrafficaMity
. . Water Percolation.. .
. . . Gas Mwation . . .
Stick HIM
Slipparinaw
Impede
Assist
Impede
Assist
(Ctay.%)
(Sand-Gravel. %)
(k, cm/s)*
Ik, cm/s)*
(Hc, cm)*
(He, cm)*
1
1
X
III
X
1
(0-5)
(95-100)
(Iff1)
(Iff3)
(6)
(6)
1
1
XII
.1
IX
II
(0-5)
(95-100)
(Iff1)
(Iff1)
-
-
III
III
VII
VII
IV
(0-20)
(60-95)
(5 x Iff* >
(5 x iff4)
(68)
(66)
VI
V
V
VIII
IV
VII
(10-50)
(50-90)
(Iff*)
(10^)
II
II
IX
IV
VIII
III
(0-10)
(95-100)
(Iff3)
(iff1)
(60)
(60)
II
II
II
VII
IV
(0-10)
(95-100)
(5 x Iff1)
(5x iff5)
_
IV
IV
VIII
V
VI
V
(0-20)
(60-95)
(Iff1)
(Iff3)
(112)
(112)
VII
VI
Y1
VII
V
VI
(10-50)
(50-90)
(2 x Iff4)
(2x Iff*)
_
V
VII
IV
III
VIII
(0-20)
(0-60)
(Iff5)
(Iff )
(180)
(180)
VIII
VIII
II
XI
II
IX
(10-50)
(0-55)
(3 x Iff")
(3x Iff*)
(180)
(180)
V
VII
_
_
_
_
(0-20)
(0-60)
—
—
_
_
IX
IX
U1
X
_
—
(50-100)
(0-50)
(icri
(Iff7)
-
-
X
X
i
XII
1
X
(50-100)
(0-50)
(iff*)
(Iff*)
(200-400*)
(200-400
z
I
—
—
-------
Table IV.9: Continued
18)
(91
(10)
(11)
(12)
(13)
(14)
Erosion Control
Crack
Imped*
uses
Water
Wind
Fast FreeM
Saturation
Resistance
Vector
Support
Symbol
(K-Factor)*
(Sand-Gravel, %)
IHe, cm)*
(Heave, mm/day)
(Expansion, %)
Emergence
Vegetation
GW
1
1
X
1
1
X
X
K0.05)
(95-100)
-
(0.1-3)
(01
—
-
GP
1
1
IX
I
(
X
X
_
(95-100)
—
(0.1-3)
(0)
—
-
GM
IV
III
VII
IV
III
VIII
VI
-
(60-95)
-
(0.4-41
-
—
-
GC
Ill
V
IV
VII
V
V
V
-
(50-90)
-
(1-8)
-
—
-
SW
II
II
VIII
II
1
IX
(X
(0.051
<95-1001
(0.2-2)
(0)
-
-
SP
II
It
VII
II
1
IX
tx
—
(95-1001
-
(0.2-21
(01
-
-
SM
VI
IV
VI
V
II
VII
II
10.12-0.27)
(60-95)
-
(0.2-7)
-
-
-
SC
Vlt
VI
V
VI
IV
IV
1
(0.14-0.27)
(50-90)
-
(1-7)
—
—
-
ML
XIII
Vlt
Ill
X
VI
VI
Ill
(0.60)
(0-60)
-
(2-271
-
—
—
CL
XII
VIII
u
Vltl
VIII
Ill
VII
(0.28-0.481
(0-55)
-
(1-6)
(1-10)
-
-
OL
XI
VII
_
VIII
VII
VI
IV
(0.21-0.29)
(0-60)
—
-
—
—
—
MH
X
IX
-
IX
IX
II
IV
(0.25)
(0.50)
-
-
-
-
—
CH
IX
X
I
Ill
X
1
VIII
(0.13-0.29)
(0-50)
-
10.8)
Of0)
—
—
OH
VIII
-
-
-
IX
-
VIII
PT
V
_
>11
(0.13)
-
-
-
—
—
—
"RCI it filing cone index, k is coefficient of permeability, Hc it capillary head, and K-Faetor is the soil erodibility factor.
The ratings I to XIII are for best tfuou^i poorest in performing the specified cower function.
-------
Table IV.9: Continued
Not*: Additional parameter! are as follows:
I
00
Fire resistance
Oust control
Side slope stability
Side slope seepage
Side slope drainage
Discourage burrowing
Discourage birds
Future use as natural settings
Future use as foundation
Same as Impede Gas Migration
Same ranking and values as for Wind
Erosion Control
Determine on basis of laboratory testing
Same ranking and values as for Impede
Water Percolation
Same ranking and values as for Assist
Water Percolation
Same ranking and values as for Slip-
periness Trafficability
All soils are suitable
Same ranking as for Support Vegetation
Same ranking and values as for Go-No Go
Trafficability
-------
to fine peaty soils are based on the rating cone index values
(RCI). The RCI expresses the probable strength of the
specific site under repeated traffic. In column 2, stickiness
trafficability is based on the weight percent of clay in the
soil. Ranking for slipperiness trafficability in column 3 is
similar to ranking in the previous column though based on a
different parameter, sand and gravel percentage. Sand and
gravel components decrease the slipperiness, i.e., improve the
quality of soil from that functional point of view, by
increasing frictional resistance. Attention is called to the
fact that all three trafficability functions provide
approximately the same soil ranking.
(2) Water percolation is addressed in columns 4 and 5. The best
parameter for judging relative effectiveness for water
percolation is the coefficient of permeability k, although it
specifically reflects water movement in a saturated condition.
The rankings in columns 4 and 5 are diametrically opposed,
since one reflects impeded percolation and the other assisted
percolation.
(3) Gas migration is addressed in columns 6 and 7 from contrasting
viewpoints of impeded and assisted flow. The parameter on
which the ranking is partly based is the capillary head (Hc)
of the soil. This parameter reflects the tenacity with which
the soil holds water.
(4) Two types of erosion control are considered in columns 8 and
9. Soils are rated for resistance to water erosion on the
basis of values of the USDA soil erodibility K- factor. (K is
the average soil loss in tons per acre per unit of rainfall
and runoff erosivity factor (R), where R usually equals the
pertinent rainfall index which is predictable from
meteorological data.) For wind erosion resistance, soils are
rated according to the sand and gravel content.
(5) Columns 10 and 11 concern cover functions of reducing freeze
action. The fast freezing/fast thawing aspects, which also
involve the tendency to freeze to greater depths, give soil
ratings of I to X based on the capillary head. Ranking for
crack resistance involves considerable judgment. It is,
however, well-known that the extreme cracking is almost
restricted to soils with high clay content, and only those
soils rated VIII or worse should cause much concern.
(6) In order to impede vector emergence, a soil with a high clay
content or well-graded combination of clay and silt, with or
without sand, is preferred. Loamy soils are somewhat more
suitable for supporting vegetation than soils at the sand,
silt, and clay extremes of grain-size distribution. Thus, a
ranking of cover soil for supporting vegetation in column 14
-82-
-------
gives the high positions, somewhat subjectively, to loam and
other well-graded mixtures. The fire resistance ranking is
the same as that for impede gas migration in column 6, since
fire resistance is increased if gas movement is impeded.
(7) With dust control, fine-grained particles are essential for
dust, so that the appropriate quantitative rating parameter is
considered to be the sand and gravel content. The ranking,
therefore, is the same as for wind erosion control.
(8) Important aspects of the side slope cover are stability,
seepage, and drainage. The seepage and drainage aspects have
rankings and rating values identical to those for impede water
percolation and assist water percolation, respectively. The
stability of side slopes may in some cases be a very important
function of the cover. In such cases, the choice of soil
should be carefully considered and based on laboratory
strength tests.
(9) The effectiveness of soils for discouraging burrowing animals
is believed to increase with the percentage of sand content.
Accordingly, the same ranking is used as was used for
slipperiness trafficability, since in both cases the important
parameter is sand and gravel content. Soil type is apparently
not a pertinent consideration in discouraging birds, so no
preference is given in the table.
(10) For the long-range future use of a landfill site for natural
or parklike settings, the ranking should be the same as for
supporting vegetation. Where the future use of the landfill
will be to support pavement or light structures, a major
concern is for high modulus. This aspect of cover soil is
adequately reflected in RCI values as for go-no go
trafficability. It should be remembered, however, that the
supportive capability of the underlying solid waste is
commonly the critical factor.
After selection of the soil for covering the waste, efforts should
be directed to the most effective placement and treatment. Soil cover
can be improved in several ways as it is constructed. Materials may be
Added for better gradation, hauling and spreading equipment can be
°perated beneficially, and the base or internal structure of the cover
system may be improved. Soil blending increases the expense of surface
capping operations and, therefore, has usually not been duly considered.
-83-
-------
The benefits derived from blending, however, are sometimes dramatic
through the alteration of grain-size distribution or average grain size.
Blending may be done in place using a blade or harrow. Examples of
materials used for blending include gravel, sand, silt, clay (Fung,
1980).
Soils to be used for surface capping may also be improved through
blending with additives and/or cements. Additives and cements are
defined as synthetic materials added in relatively small amounts to soil
to achieve beneficial effects. Classifications include strengthening
cements or stabilizers, dispersants, freeze-point suppressants, water
repellants, and dust palliatives. Factors that enter into determining
the cost effectiveness of additives and cements are their relatively
high unit cost, manner of addition or incorporation, and duration of
effect. A major concern in using any cover system with increased
strength is the remaining susceptibility to cracking as a result of
differential settlement or environmental deterioration, such as freeze-
thaw; plans for patching repairs are essential. Examples of additives
and cements which could be used are as follows (Fung, 1980):
(1) Soil-Cement — Soil cement is composed of sandy soil, portland
cement, and water that have reacted and hardened. About 8% of
dry cement is incorporated in place and compacted after water
is added to about optimum content. Pre-mixing and placement
wet is another method that seems less amenable to covering
waste because of extra costs, except possibly on steep slopes.
Somewhat higher cement content is needed in poorly graded
soils or where fines are lacking, but correspondingly less for
well-graded mixtures.
(2) Soil-Bitumen — Bitumen stabilizes sandy soils by cementing
and water-proofing. An optimum content increases the cohesion
of the soil, yet does not separate grains enough to reduce
frictional strength. From 4 to 8Z bitumen (soil dry weight
basis) is recommended as a first cut for sandy loam to be
-84-
-------
refined subsequently by testing mixtures o£ various contents.
Sand should require somewhat less bitumen.
(3) Cement-Treated Soil — Incorporation of as little aB 1% of
Portland cement has been shown to have stabilizing effects on
granular soil.
(4) Lime-Treated Soil — Lime (calcium oxide or hydroxide) is
principally mixed into clayey soils for its effects as a
flocculant or base exchanger. However, lime also promotes
some beneficial pozzolanic reaction in fine, cohesive soils,
and strengthening may continue for many weeks of curing.
(5) Fly Ash-Lime (or Cement-)-Treated Soil — Fly ash is
attractive as an additive to soil because of its pozzolanic
properties (i.e., although not a cement itself, fly ash
develops cementitious properties with lime or cement and
water). Accordingly, a small amount of lime or Portland
cement should be included unless self-hardening has been
confirmed by test or previous experience with the particular
fly ash.
(6) Fly Ash-Treated Soil — Some fly ashes possess moderate self-
hardening characteristics when moistened and compacted and,
therefore, offer promise alone as major additions to cover
waste. This phenomenon is partly due to the free lime content
of the ash itself. Laboratory tests or field trials will be
necessary to establish such useful peculiarities of specific
fly ash sources where previous experience is lacking.
(7) Fly Ash-Lime (or Cement)-Sulfate-Treated Soil — This mixture
is proposed as a locally viable option for covering waste
following research on the uses of fly ash-lime-sulfate (or
sulfite).
(8) Dispersants — Several soluble salts act as dispersing agents
and are recommended for improving soil compaction and reducing
k. The treatment is primarily for fine-grained soils
containing clay minerals. Better compaction can be achieved
on coarse-grained soils by other means, such as blending in
fine material. Dispersion (deflocculation) involves the
breakdown of clayey aggregates into the individual mineral
particles. The effect is to increase dry unit weight, to
lower permeability, and incidently, to facilitate compaction.
Three dispersing agents have been found notably effective.
Common sodium chloride produces satisfactory results in many
cases and is inexpensive. Tetra-sodium pyrophosphate and
sodium polyphosphate appear to have slight advantages in many
cases over sodium chloride.
-85-
-------
(9) Swell Reducers — Lime is the principal additive for reducing
shrink/swell behavior in clay-rich soils. About 5% is optimum
for effecting ion exchange (calcium for sodium), flocculation,
and cementation. Quicklime contains more calcium and,
therefore, is preferred to hydrated lime by some users
provided a mellowing period is allowed for homogenization.
(10) Freeze-Point Suppressant — Poor compaction can result when
soil pore water is frozen during cold weather operations.
Ordinarily soils construction for roads, etc., is curtailed in
the winter to avoid the poor results. Some improvements can
be achieved by adding a freeze-point suppressant and may be
advisable where covering operations are expected to continue
during cold spells. Calcium chloride can be effective, either
in solution or in dry, flaked form.
Layering is perhaps the most promising, yet presently
underutilized, technique for designing final solid waste cover. By
combining two or three distinct materials in layers, the designer may
mobilize favorable characteristics of each together at little extra
expense. The systematics are briefly set forth there. Figure IV. 21
schematically illustrates a layered system (Fung, 1980). A layered
system may include topsoil, barrier layer, buffer layer, water drainage
layer, filter and gas drainage layer. Comments on each of these layers
are as follows (Fung, 1980).
(1) Topsoil — A topsoil or a subsoil made amenable to supporting
vegetation frequently forms the top of a layered cover system.
Untreated subsoils are seldom suitable directly, so it has
been often necessary to supplement subsoil with fertilizers,
conditioners, etc., to obtain the desired result.
(2) Barrier Layer or Membrane — The primary feature in many
layered systems is the barrier. This layer functions to
restrict passage of water or gas. Barrier layers are almost
always composed of clayey soil that has inherently low
permeability; USCS types CH, CL and SC are examples of soil
types. Soil barriers are susceptible to deterioration by
cracking when exposed at the surface, so that a buffer layer
is recommended to protect the clay from excessive drying.
Synthetic membranes may be used in place of soil barriers.
Costs are generally high in comparison to available soils, and
-86-
-------
LO AM (fOm VEGETATION)
issiiiiisisssisiis o«*vel (gas channel l!iiiiiil||!iill
llllg?|go|cggggSgS;;8gjSEi8iSiig»gggsry»«rfi8lg»j»llcll
waste ®Wffl»SW|
iiiaiwiii
7»')'» i i
^ Clay (Barrieri
^ li'il i'iTi'i'i hfi
iILT iriLTE*!
I I I ' I I ' ' ' '
(AND (BUFFER)
"**T t Mv2;ife^iSS^
Figure IV.21: Typical Layered Cover Systems (Fung,
1980).
-87-
-------
problems in placement or with deterioration may arise.
Nevertheless, membranes are a viable option.
(3) Buffer Layer — A buffer layer may be described as a random
layer having a subordinate covering function and
characteristics in comparison with the adjacent layer. The
principal service of a buffer is to protect a barrier layer or
membrane located above or below. The buffer shields a
vulnerable, thinner barrier or sheet from tears, cracks,
offsets, and punctures. Below a barrier, the buffer layer
also provides a smooth, regular base. Any soil type will
serve as a buffer ordinarily, but it should be free of clods.
(4) Water Drainage Layer or Channel — A water drainage layer,
blanket, or channel may be designed into the cover in numerous
ways to provide a path for water to exit rapidly. Poorly
graded sand and gravel are possibilities as effective drainage
materials, i.e., soils classified GP and SP.
(5) Filter — Where layers with grossly discordant grain sizes are
joined, there may be a tendency for fine particles to
penetrate the coarser layer. As a result, the effectiveness
of the coarse layer for water drainage may be reduced by
clogging of pores. Removals from the fine layer may promote
additional bad effects, such as internal erosion and
settlement. A filter layer with a material such as silt may
be desirable.
(6) Gas Drainage Layer and Vents — A gas drainage layer has
consistency and configuration similar to those of the water
drainage layer or channel. Both layer types function to
transmit preferentially. The position in the cover system is
the main distinction. The gas drainage layer is placed on the
lower side to intercept gases rising from waste cells, whereas
the drain for water is positioned on the upper side to
intercept water percolating from the surface. Gas vents are
frequently used in the present state-of-the-art in municipal
waste landfilling. It has been found that dangerous
concentrations of flammable gases can accumulate if not vented
properly to the atmosphere. Figure IV.22 illustrates two
recommended gas vents.
Commercially-available capping materials or membranes may be
preferable to soil-based systems over wastes that pose a high risk of
serious pollution (where not properly shielded). Examples of commercial
materials include bituminous and Portland cement concrete barriers, and
membranes of various types functioning alone or in conjunction with soil
-88-
-------
••••••«••»••••¦«••••••»•••*••«••••
• ••••*••••«•( )••«««• ••%••>« •••••••«*•••
• ••••••••«•• »vVV •••••••••••«<•••••• ••••••«
••••••••••••>••
*****
P»PE VENT
vm
iff
z
*
paOQ0»K»
GRAVEL TRENCH
Figure IV.22: Two Types of Gas Vents for Layered Cover Systems
(Fung, 1980).
-89-
-------
or granular waste layers to exclude water. A summary of commercial
materials and their characteristics is presented elsewhere (Fung, 1980).
A number of factors should be considered in the selection of a
commercial cover material. Important characteristics of flexible
membranes and rigid barriers include vulnerability to tearing or
cracking, respectively. A smooth buffer of sand should reduce membrane
puncture during construction, but extra long-term planning will be
necessary against stressing effects of differential settlement. The
rigid barriers sometimes have an advantage because cracks can be
exposed, cleaned, and repaired (sealed with tar) with relative ease.
Deterioration is another major problem for serious consideration.
Concrete barriers are susceptible to chemical deterioration in harsh
environments such as those rich in sulfates. Sunlight, burrowing
animals, and plant roots increase the deterioration of membranes.
Twenty years is about the extent of experience with membranes. Longer
design service life may be somewhat risky in adverse environments and
conditions. The cost of commercial coverings will often be the
overriding consideration (Fung, 1980).
The other category of nonsoil materials that is available for
covering waste ranges through a long and diverse list of granular wastes
(some are in fact soils). An economic advantage may accrue from using
these inexpensive, locally generated wastes in place of scarce,
indigeneous soils. Examples of waste materials used for covers include
fly ash, fly ash modified with lime to enhance cementation, bottom ash
and slag, furnace slag (from iron steel), incinerator residue, foundry
sand, mine and pit wastes, mine mill tailings, plant sludges, reservoir
-90-
-------
and channel silt, dredged material, and composted sewage sludge (Fung,
1980).
C08ts
The general range for costs of surface caps is $20,000 to $30,000
per acre covered, while the cost for installing a liner is more
expensive, being in the range of $40,000 to $50,000 per acre. Cost
estimates for various liners are outlined in Table IV.10.
SHEET PILING
Sheet piling involves driving lengths of steel that connect
together into the ground to form a thin impermeable permanent barrier to
flow. Sheet piling materials also include timber and concrete, however,
their application to polluted ground water situations would be doubtful
due to corrosive actions and costs, respectively.
Construction
Sheet piling requires that the steel sections be assembled prior to
being driven into the ground. The lengths of steel have connections
along both edges. The connections may be either slotted or ball- and
socket-types. The sections are then driven individually into the ground
by use of a pile hammer. The types of pile hammers include: drop;
single-acting steam; double-acting steam; diesel; vibratory; and
hydraulic. For each type of hammer listed the driving energy is
supplied by a falling mass, which strikes the top of the pile (Bowles,
1977). After the piles have been driven to their desired depth, the
remaining above-ground portions are cut off. Initially, sheet piles are
-91-
-------
Table IV.10: Costs for Various Sanitary Landfill Liner Materials
(Haxo, 1973)
Installed Cost®
Material ($/sq yd)
Polyethylene (10-20b milsc)
0.90
- 1.44
Polyvinyl chloride (10-30^ mils)
1.17
- 2.16
Butyl rubber (31.3-62.5^ mils)
3.25
- 4.00
Hypalon (20-45^ mils)
2.88
- 3.06
Ethylene propylene diene monomer
(31.3-62.5® mils)
2.43
- 3.42
Chlorinated polyethylene (20-30^ mils)
2.43
- 3.24
Paving asphalt with sealer coat (2 inches)
1.20
- 1.70
Paving asphalt with sealer coat (4 inches)
2.35
- 3.25
Hot sprayed asphalt (1 gallon/yd^)
1.50
- 2.00 (includes
earth cover)
Asphalt Sprayed on Polypropylene fabric
(100 mils)
*0
CM
•
H
- 1.87
Soil-bentonite (9.1 lbs/yd^)
0.
,72
Soil-bentonite (18.1 lbs/yd^)
1.
,17
Soil-cement with sealer coat (6 inches)
1.
.25
®Co8t does not include construction of subgrade nor the cost of earth
cover. These can range from $0.10 to $0.50/yd2/ft of depth.
^Material costs are the same for this range of thickness.
c0ne mil ¦ 0.001 inch.
-92-
-------
not totally impermeable because of small gaps in the connections. As
time passes, these gaps are closed as ground water flow carries fine
particles into the gaps and closes them by clogging.
Steel sheet piling can be considered permanent because experience
has shown that corrosion is not a factor in causing failures (Bowles,
1977). One strong advantage of sheet piling is that the sections are
reusable and do not have to be left in place permanently. If conditions
permit, steel piles may be removed and reused.
C0818
In situations where the function of the sheet pile is just to
restrict ground water flow, i.e., no significant load resistance is
required, a light weight steel will be adequate. Construction costs of
a 1700 foot long and 60 foot deep light-weight steel cut-off wall was
reported to range from $650,500 to $956,000 (Tolman, et al., 1978). To
date, sheet piling has been proposed as a means of ground water
pollution control, but specific applications have been minimal, if any.
Advantages and Disadvantages
Construction of steel sheet piles as a means of ground water
control can potentially be effective and economical in specific cases.
In general, however, this is probably an over-elaborate technique to
achieve a relatively simple result. As the size of a project increases,
sheet piling will become uneconomical because of high material and
shipping costs. In addition, pile driving requires a relatively
uniform, loose, boulder-free soil for ease of construction. Other
-93-
-------
advantages/disadvantages are listed in Table IV.11 (Tolman, et al.,
1978).
GROUTING
Grouting is the process of injecting a liquid, slurry, or emulsion
under pressure into the soil. The fluid injected will move away from
the point of injection to occupy the available pore spaces. As time
passes, the injected fluid will solidify, thus resulting in a decrease
in the original soil permeability and an increase in the soil-bearing
capacity. Grouts are usually classified as particulate or chemical.
Particulate grouts consist of water plus particulate material which will
solidify within the soil matrix. Chemical grouts usually consist of two
or more liquids which will gel when they come in contact with each
other. Listed in Table IV.12 are materials commonly used for grouts.
Table IV.13 identifies the properties of cement grout additives.
Construction
Two of the more popular methods of grout installation are the stage
and packer methods. In stage grouting, holes are drilled to the
geologic seam closest to the land surface and the grouting fluid is
injected. The holes are then cleaned, drilling continues down to the
next seam and grouting continues. The process is repeated until a
sufficient depth has been obtained. In general, the stage method
proceeds downward, utilizing increasing injection pressures. In the
packer method, holes are drilled to the maximum planned depth. A zone
of specified thickness is then partitioned off by placing packers at the
top and bottom of the zone. Grout is then injected into the zone
-94-
-------
Table IV.11: Advantages/Disadvantages of Steel Sheetplles (Tolman, et al., 1978)
Advantages
Disadvantages
1. Construction is not difficult; no excavation is
necessary.
1.
The steel sheet piling initially is not
watertight.
2. Contractors, equipment, and materials are
available throughout the United States.
2.
Driving piles through ground containing
boulders is difficult.
3. Construction can be economical.
3.
Certain chemicals may attack the steel.
4. No maintenance required after construction.
5. Steel can be coated for protection from
corrosion to extend its service life.
-------
Table IV.12: Materials Commonly Used for Grout
cement, water
cement, rock flour, water
cement, clay, water
cement, clay, sand, water
asphalt
clay, water
chemical
-96-
-------
Table IV.13: Properties of Admixtures Used with Cement Grouts
Admixture
Property
Calcium chloride
Sodium hydroxide
Sodium silicate
Accelerates setting time
Gypsum
Lime sugar
Sodium tannate
Retards setting time
finely ground bentonite
Increases plasticity
Reduces grout shrinkage
Clay
Ground shale
Rock flour
Reduces cost of grout
Reduces strength of grout
—
-97-
-------
between the two packers. The mechanism is then moved up to the next
zone to be grouted and the process is repeated. The packer method moves
upward from the bottom utilizing decreasing injection pressures.
Advantages of the packer method include: grouting pressures can be
adjusted specifically to a particular foundation depth; walls of the
borehole remain smooth and an excellent seal with packers can be
achieved; and high pressures used in the stage method which may cause
fracturation are avoided. However, these advantages can be offset by
increased equipment needs and time for installation (Bowen, 1981).
Another method of grout injection is the driven-rod method. In
this method, a perforated rod is driven to a desired depth and grout is
injected as the rod is slowly withdrawn. This method is limited to
shallow depths and relatively boulder-free soils.
Problems that might arise during construction of grout systems
include: leakage of grout around the injection pipe of the hole being
grouted; loss of presssure below the water table resulting in sand and
water being forced into the pipe; and grout surfacing in outlying areas
due to lateral migration (Bowen, 1981). It is also desirable to deposit
cement grouts in clean seams from which any clay or unconsolidated
materials have been removed, thus adding the burden of washing boreholes
(Peurifoy, 1979).
Construction of grout cutoff walls requires certain equipment in
addition to normal borehole or drilling equipment. The list includes
one or more air compressors; one or two grout mixers; one agitator-type
reservoir tank; one or more grout pumps; and grout discharge pipes or
hoses, valves, and pressure gauges (Peurifoy, 1979).
-98-
-------
Design
The first consideration in the design of a grout cutoff system is
the actual composition of the grout. The composition will be a function
of several variables including: the soil type to be injected into; the
pollutant to be inhibited; the time since pollution started; and the
time for installation. In general, chemical grouts must be used in
fine-grained soils. However, chemical grouts (usually silicate) are not
suitable for highly acidic or alkaline environments because their gel
formation is an acid-base reaction. For coarse or gravel soils,
particulate grouts are suitable. The amount of cement or bentonite in a
particulate grout varies widely. The quantity of bentonite which can be
incorporated in a grout is dependent upon the following considerations:
the workability of the mixture, increased bentonite concentrations
increase the stiffness of the slurry until it may become unpumpable;
adding bentonite to cement slurries decreases their compressive
strength; increased bentonite concentrations yield lower specific
gravity slurries showing a reduced tendency to migrate through the soil
after placement; and increased bentonite concentrations reduce
settlement or sedimentation of the slurries before injection (Jones,
1963).
A general guide for the selection of grouts is shown in Figure
IV.23. Knowing the soil type in the area to be protected, Figure IV.23
can be entered from the top to determine both the applicable types of
grouts and the available grouting procedures. The range of soil types
to which a particular type of grout is applicable is covered by the
¦olid white horizontal bars. The range of soil types to which a
-99-
-------
mmii i i r
AM 9 CHEMICAL GROUT
COARSE
SILT
SILT (NON-PLASTIC)
GRAVEL
COARSE
MED. F NE
CLAY - SOIL
Mill I I I
_JCHROME - LIGNIN
I I I I I I •
)RESINS
It
IS ILICATES
I—I
BENTON ITE
"TEMEHT
COMPRESSED AIR
22222ZasuB-AQUEOUS Excy
zz
zzzz
I nh i f r f I ln i FREEZING
EXCAVATION- HEAVY PUMPING'YIELD
I I i ii i t i t i i ii « $ • i
IF QUICK CONDITIONS EXISTS
I I IIII II I I I I
WELLPOI NTS-THEORETICAL LIMITS FOR GRAVITY DRAINAGE
i
WELLPO I NTS-VACUUM SYSTEM NEEDED
)j j 11 j '
1111 >111 I 1111 I I I I I
VELLPOINTS-GRAVITY DRAINAGE VERY SLOW
' ¦ ' j' 'I" Mill l
ZZ7ZZZ21 wellpoint vacuum system
[lectro-osmosis
PpSSIgLE
10.0
1.0 0.1 0.01
GRAIN SIZE IU MILLIMETERS
0.001
Figure IV.23: Soil Gradation Limits for Grout Injection
(from American Cyanamid Company).
-100-
-------
particular grouting procedure is applicable is indicated by the cross-
hatched bars. Tables IV.14 to IV.16 summarize the compatibility o£
grouts with several contaminant types.
The second design consideration is the pressure at which the grout
is to be injected. The use of excess pressures may weaken the strata by
fissuring the rock or by opening fissures in otherwise closely-jointed
rock. Pressure-induced fissures will result in the wastage of grout.
In contrast, the pressure should be kept sufficiently high to ensure
penetration of the grout and decrease the time required for grouting.
The allowable grouting pressure is best determined by conducting
hydraulic fracture tests in the strata to be grouted (Morgenstern and
Vaughan, 1963).
A number of chemical grouting agents have been banned or their use
discontinued because of their toxicity. Huibregtse and Kastman (1981)
note that one of the most successful grouts, AM-9, is highly toxic and
has been removed from most markets because of its potential undesirable
effects on ground water. Other grouts considered to be toxic and
presenting a potential for ground water pollution are the lignin- and
formaldehyde-based grouts.
Costs
The costs for grout cutoff systems are high, hence they will be
applicable only to small localized cases of ground water pollution.
Costs have been reported to range from $142 to $357 per installed cubic
foot (Lu, Morrison and Stearns, 1981). Table IV.17 lists the relative
costs of different types of grout.
-101-
-------
Table IV.14: Interactions Between Grouts and Generic Chemical Classes (Hunt, et al., no date)
Portland
Polymers
Grout Type
Chemical Group
Cement
Clay (Bentonite)
-o
>->
Bitumen
Type I
Tyne II and V
Clay-Cement
i
i
Silicate
1 Acrylamide
!
Phenolic
1
Urethane
X,
a>
rH
e
*-<
0
Mh
1
03
-------
Table IV.15: Interactions Between Grouts and Specific Chem-
ical Groups (Hunt, et al., no date)
X.
Portland Cam ant
Polymer*
Ny a rout Typa
CKamical Qroup
8
i
««
3
I
>
>
1
I
l-
a
1
i
•
*
j
o
i
i
i
•
•
1
(A
1
?
u
<
3
i
a.
2
|
•
5
>
1
: 1
5 2
>
¦
i
•
8
>
&
Organic Compounds
Alcohol* and Glycol*
'a
?d
>d
'd
>d
?
>d
7
3a
?
Ja
>a
Aldahydat and Katonaa
>d«
7
7
>d
?
?
'a
3a
>d
>
>
AMtatie and Aromatic
Hvdrocarbona
'd
2*
2?
'd
?
?
'a
'd*
2j
>d
>d
Amidaa and Aminaa
7
?
7
7
?
>
7
7
3'
7
?
Chtonnatad Hydrocarbona
>d
2d
2d
7
>
?
>a
»d
'a
a
>d
>d
Ettiar* and Epoaidaa
7
?
7
7
>
?
'a
7
'a
>d
>
Hatarocyclic*
¦>
7
7
>d
>
7
'a
>
>
>
»
NitrWaa
¦>
7
7
7
>
7
7
7
7
Organic Acida and A ad
Chlondaa
>•
id
id
'd
>d
>•
2a
'a
2a
7d
>d
Organomatallica
7
7
7
7
>
7
7
7
?
Phanots
?d
1d
7
'd
>d
7
2a
>C
>
>
Organic Eatare
j
>
7
7
la
7
7
7
?
>
Inorganic Compounda
Haavy Maul Sana and
Comp»a*aa
M
2c
2a
>d
2c
3?
2?
7
7
'a
?
inorganic Acida
7,4 a
id
la
>e»
>c
3*
2c
>*»
2c
id
>a*
>a*
Inorganic Saaaa
'a
la
la'
7C»
>d
2c
3d
'd
'd
2c
7a
>d
Inorganic Salts
>d
2c
2a
2d
>d*
3?
3d
V
>d
'a
'a
'a
KEY: Compatibility Indaa
Erf act on Sat Tima
1 No wgmficant affact
2 Incraaaa m wt tima llangthan or pravant from tamngl
3 Oacraaaa in wt tima
Effact on Durability
• No significant aHact
b Incraaaa durability
c Oacraaaa durability (daatructtwa action Oagina within
a than tima oanod)
d Oacru«a durability Idaatructtva action occur* ovar a
long tima oanod)
* Eicapt urtataa. which ara 'c
t Eacaot KoH and NaOH. which ara id
t Low motaeolar waight ootymar* only
I Non-oaidwng
* Non-onidiling, ocaot HF
0 Eacaot concantratad acid*
* Eacaot tWahydat which ara 'a
1 Eacaot Waachaa which ara 3d
* For modrfiad Bamonitaa. 'd
' Data unavailitila
-103-
-------
Table IV.16: Predicted Grout Compatibilities (Hunt, et al., no date)
Qrout Typo
Chemical Group \
•
•
£
Polymer*
!
J
>
to
u
<
1
e
•
£
!
|
3
a
*
>
1
. 1
II
»¦
m
i
m
i
i
>
*
a.
Organic Compound*
Alcohol* and Glycol!
1a
1 -
3b
-
1
1 -
t -
Aldehyde* and Ketone*
la
-
-
ta
ta
ta
Aliphatic and Aromatic
Hydrocarbon*
1d
1 -
-
t -
-
1-
t -
Amtdee and Amine*
3a
3d
3b
-a
1a
1a
3a
Chionnated Hydrocarbone
id
1 -
1 -
-
1
1 -
Ether* and Epoiidea
la
I -
ta
i-
la
la
Hatarocyclica
Id
l -
la
1a
1a
la
1*
Nitniaa
la
3-
ta
ta
ta
1a
la
Organic Acida and Acid
Chlonae*
1-
-
3-
2-
-
-
1 -
Organomeuilici
la
3a
-
-
ta
la
3*'
Phenol*
ta
ta
-
2-
1a
la
1>
Organic Estar*
-
>
>
>
>
td
Inorganic Compound*
Heavy Metal Salt* and
Comple«ea
-a
-
-
-
-
3-
37
Inorganic AckM
-
-
2-
-
-
1 -
t -
inorganic Baaoa
-
-
3-
-
-
-
1 -
Inorganic Salt*
-d
-
-
-
-
-
3* -
KEY: Compatibility tndai
Effact on Sat Tim#
1 No ngrvficam affect
2 Incraaw in mi time (lengthen or prevent from letting)
3 Oecreeee in let time
Effect on OuraMity
a No wgnificant effect
b Incraaaa duraMity
c Oacraaaa duraMity Ideatructive action begin* within
a ihort time oenodl
d Oacraaaa duraMity Ideatructive action occuri over a
long time period)
* If metal lata thai are accelerator*
* If matai « capable of acting aa an accelerator
? Data unavailable
-104-
-------
Table IV. 17:
•k
Relative Costs of Grout
Type of Grout Basic Cost Figure
Portland cement
1.0
Silicate base - 15 percent
1.3
Lignin base
1.65
Silicate base - 30 percent
2.2
Silicate base - AO percent
2.9
Urea formaldehyde resin
6.0
Acrylamide (AM-9)
7.0
* Base unit = 1.0. Under a given set of conditions, where portland
cement grout costs 1.0 times $/unit, other types of grout will cost
the given figure times $/unit. (Tolman ,etal. , 1978).
-105-
-------
Advantages and Disadvantages
The technology of grouting has been used in the construction
industry for years. To date, most applications of grouting technology
have been for increasing a soils' bearing capacity (to aid in tunnel
construction for example) or to decrease the permeability of a soil to
inhibit water movement (such as a cutoff wall for a dam). The
applications of this technology to ground water pollution control are
very recent. For example, Huibregtse and Kastman (1981) have analyzed
the feasibility of mobile grouting units for protecting ground water
threatened by hazardous spills on land. A number of physical/chemical
advantages and disadvantages for grouting are listed in Table IV.18.
SLURRY WALLS
Slurry walls represent a technology for encapsulating an area to
either prevent ground water pollution or restrict the movement of
previously-contaminated ground water. The technology involves digging a
trench around an area and backfilling with an impermeable material.
Slurry walls can either be placed upgradient from a waste site (Figure
IV.24) to prevent flow of ground water into the site, or placed around a
site (Figure IV.25) to prevent movement of polluted ground water away
from a site. Usually, slurry walls will require a complementary
technology such as surface capping or purge wells.
Construction
The most common type of slurry wall construction is the trench
method. In this method a trench is excavated, in the presence of a
bentonite-water slurry, to a desired depth. After excavation, the
-106-
-------
Table IV.18: Advantages/Disadvantages of Grout Systems
Advantages
(a)
Disadvantages
1. When designed on basis of thorough preliminary
investigations, grouts can be very successful.
2. Grouts have been used for over 100 years in
construction and soil stabilization projects.
3. Many kinds of grout to suit a wide range of
soil types are available.
1. Grouting limited to granular types of soils
that have a pore size large enough to accept
grout fluids under pressure yet small enough
to prevent significant pollutant migration
before implementation of grout program.
2. Grouting in a highly layered soil profile
may result in incomplete formation of a
grout envelope.
3. Presence of high water table and rapidly
flowing ground water limits groutability
through;
a.
b.
extensive transport of contaminants,
rapid dilution of grouts^
Some grouting techniques are propietary
(a)
Procedure requires careful planning and
pretesting. Methods of ensuring that all
voids in the wall have been effectively
grouted are not readily available.
(a) Tolman,et al.,1978.
(b) Huibregtse and Kastman, 1981.
-------
4 ftacrsrti MOacrtil
•CDAOCK
/
i2-
-------
COMPACTED CLAY COVER.
' / ' ' /
SLURRY TRENCH
CUT-OFT
*>
&
V
A 4
A.
K
^yrjj_\vAST
¦Xx
&
*7^7 V?r7777
77/vy
AQUACLUDE
__l
:A ^///
Figure IV.25: Isolation of Existing Buried Waste (Ryan, 1980).
-109-
-------
trench can be solidified by backfilling with a mixture of bentonite and
the excavated material, or it can be allowed to solidify itself by
incorporating cement in the original slurry. The backfilled trench is
usually called a soil-bentonite (S-B) trench, and the other method is
called a cement-bentonite (C-B) trench. Comparative advantages of each
method are outlined in Table IV.19 (Ryan, 1980).
Another method of slurry wall construction is the vibrating beam
method. In this approach, a beam with a pressure hose attached is
driven into the ground, then slurry is injected under pressure as the
beam is gradually withdrawn. This is similar to grouting and possesses
similar advantages/disadvantages. This approach can be more economical
and can proceed faster than the trench method. However, the continuity
of the wall cannot be guaranteed, the in-place thickness of the wall is
considerably smaller, and the presence of boulders in the soil hinders
this method more than the trench method.
Construction of slurry trenches is generally simple and consists
only of excavating, recirculating the slurry, and backfilling.
Excavation can be accomplished by any one or combinations of the
following: backhoe; draglines; clamshells; bucket scrapers; or rotary
drilling equipment. The choice of the specific type of excavation
equipment is generally governed by the depth and width of excavation
(Xanthakos, 1979). The backhoe is usually desirable when depths
required are shallow. For depths of 30 meters or more, draglines are
required. Because dragline bucket widths usually exceed 2 meters, they
are not economical for C-B trenches due to high material costs.
-110-
-------
Table IV.19: Comparison of C-B and S-B Slurry Trenches (Ryan, 1980).
Cement-Bentonite
( C-B )
Soil-Bentonite
( S-B )
1. Independent of availability or
quality of soil for backfill.
2. More suitable for limited access
areas.
3. Cement sets quickly. Can cut
trenches or allow traffic over
wall in just a few days.
4. Can be constructed in sections.
S-B requires continuous trenching
in one direction.
1. Lower material costs
2. Can achieve higher permea-
bility than C-B.
-Ill-
-------
Clamshells can be cable-mounted or, like the bucket scraper, mounted to
a rigid sliding bar and used for deep trench excavation (Ryan, 1980).
Recirculation of the slurry is important for maintaining the
integrity of the slurry. During excavation the slurry will be subjected
to losses through seepage and changes in density through addition of
excavated material. Control of these changes is achieved by
continuously recirculating the slurry through a central mixing unit
which may have provisions for separating excavated materials from the
original slurry. Backfilling operations may require mixing of different
soil types prior to placement in the trench. Mixing can be accomplished
by discing and blading. The mixed soil is then bulldozed into the
trench, partially mixing with and displacing the slurry. Backfilling
follows trenching after an interval of time sufficient to prevent
interference between the two activities.
An important final aspect of slurry trench construction is keying
into an underlying impervious zone. Trenches will require 2 or 3 feet
for keying into clay materials, and will require a grout connection when
keying into impervious bedrock (D'Appolonia, no date).
Pes ign
The important design considerations of any slurry-trench are the
composition of the slurry and the ensuing impermeable wall. Design
procedures for slurry walls range from general rules-of-thumb to
detailed mathematical analyses of all aspects of the system. A general
guideline approach is probably the most helpful.
-112-
-------
D'Appolonia (no date) suggests that viscosities of both C-B and S-B
slurries should be such that drain times from a Marsh Funnel range from
40 to 50 seconds. It is also recommended that the specific gravity of
the slurry be at least 15 pounds per cubic foot less than the unit
weight of the backfill. The permeability of the S-B backfill will
depend on the soil and the amount of bentonite blended in. These
characteristics are depicted in Figures IV.26 and IV.27. The ideal
consistency for backfill placement is a paste having a water content
slightly above the liquid limit of the sand-clay-bentonite backfill mix.
This usually corresponds to a slump cone reading of 2 to 6 inches.
The durability of the slurry-trench refers to its resitance to
attack from contaminants. In the presence of clean ground water, both
C-B and S-B trenches show little deterioration and can be considered
permanent. However, C-B trenches show poor performance records where
acids or sulphates are present. Similarly, exposure of an S-B trench to
certain contaminants can lead to increased permeability through pore
fluid substitution or the increased solubility of barrier minerals in
the contaminant fluid. Pore fluid substitution can be the result of
high concentrations of salts, which attract the waters of hydration, or
it can result from ion substitution within the clay matrix.
Permeabilities of S-B trenches have been shown to increase in the
presence of certain organics, calcium, magnesium, heavy metals and
solutions of high ionic strength (D'Appolonia, no date).
Xanthakos (1979) suggests that a slurry system should be designed
based on the functions of the slurry. These functions are:
-113-
-------
.-4
WELL GRADED
COARSE GRADATIONS
(30-70% +20 SIEVE )
W/ 10 TO 25 % NP FINES
CO
POORLY GRADED
SILTY SAND V.'/
30 TO 50% N? FINES
_l
o
-o
CLAYEY SILTY SAND
W/ 30 T 0 £ 0 % FINES
0
3
2
5
4
%BENTONITE EY DRY WEIGHT OF SB EACXFILL
Figure IV.26: Relationship Between Permeability and Quantity of
Bentonite Added to SB Backfill (Ryan, 1980).
-114-
-------
80i
70
LU 60
>
u
w 50
o
S 40
CO 20
Z>
1 20
10 -
, ¦PLASTIC
• FINES
\*
\
*\9
•V
o o
N0N - PLASTIC OR LOW
PLASTICITY FiNES
j CL
I0"9 IC~8 I0"7 I0"6 I0"5
SB BACKFILL PERMEABILITY, cm/sec.
IC'
Figure IV.27: Permeability of Soil - Bentonite Backfill Rela-
ted to Fines Content (Ryan, 1980).
-115-
-------
(1) support the face of the excavation, and also prevent the soil
from sloughing and peeling off;
(2) seal the formation and form the filter cake, preventing slurry
loss to the ground;
(3) suspend detritus, thereby preventing sludgy unconsolidated
layers from accumulating on the bottom of the trench; and
(4) carry the cuttings in the slurry volume, thereby preventing
sedimentation in the mud circuit.
All of the above functions can be controlled by manipulating the
physical properties of the slurry, i.e., density, viscosity, etc.
Xanthakos (1979) has outlined a series of simple steps and procedures
for proportioning the materials that comprise a slurry. These are as
follows:
(1) Determine the density required for trench stability. The
density can be controlled by the presence of colloid and non-
colloid solid materials. If the depth to an impermeable
formation (H) is known, the required slurry density can be
calculated by the following (see Figure IV.28).
y(l-m2)K <* + Y^ni^R** +
y f s 2
nz
where
Yf » required slurry density (gm/m^)
y = unit weight of soil (gm/m^)
K
-------
Slurry
level i
Ground
level
— (l-m)H
Natural
water level
nH
mH
Figure IV.28: Stability of a Trench for Arbitrary Slurry and
Natural Water Level (Xanthakos, 1979).
-117-
-------
(2) Select the funnel viscosity by reference to Table IV.20.
(3) Establish any applicable control limits from Table IV.21.
(4) Determine whether control agents (peptizers, polyelectrolytes,
fluid-loss-control materials, etc.) are necessary and
economically justified.
(5) Proportion the constituent materials (water, bentonite,
control agents, and non-colloid solids). This phase merely
consists of a quantitative estimation. The proportioning may
be empirical and depend on experience if the properties of the
materials selected are known, or it may have a technical basis
of tests and estimations.
In general, slurry-trenches are attractive alternatives when an
impervious natural barrier exists at a reasonable depth and the waste
area is relatively large. A clayey sand or sandy clay containing 30 to
60% fines blended with the bentonite slurry is usually satisfactory for
most waste isolation applications (D'AppoIonia, no date).
C08 tS
Table IV.22 shows the comparative costs of S-B and C-B slurry
walls.
Advantages and Disadvantages
There are many instances where slurry wall technologies have been
employed. Slurry systems have been employed for their impermeability as
cutoff and diaphragm walls, and slurries have been used to aid in the
construction of load bearing elements and foundations (Xanthakos, 1979).
To date, there has been very little information provided on the
performance of slurry walls applied to ground water pollution control
problems. This does not mean the technology has not been applied.
Slurry walls have been constructed at the Rocky Mountain Arsenal in
-118-
-------
Table iv.20: Funnel Viscosity for Common Types of Soil (Xanthakos,
1979)
Funnel Viscosity
, s/946 cm^
Type of Soil
Excavation in
Dry Soil
Excavation with
Groundwater
Clay
27-32
Silty sand, sandy clay
29-35
Sand, with silt
32-37
38-43
Fine to coarse
38-43
41-47
And gravel
42-47
55-65
Gravel
46-52
60-70
-119-
-------
Table IV.21: Control Limits for the Properties of Slurries
(Xarithakos, 1979).
Property
Function
Average
bentonite
concentration,
X
Density,
lb/ft 3
sp gr
Plastic
viscosity,
cP
Marsh
cone
viscosity
10-rain gel
strength (Fann),
lb/100 ft2
pH
Sand
content,
X
Face support
>3-4
>64.3
>1.03
....
Limits
%
>15
Sealing process
>3-4
established
1
Suspension of
> 3-4
by soil type
>12-15
detritus
Displacement
<15
<78
<1.25
<20
<12
<25
by concrete
Separation of
<30
noncolloids
Physical
<15
<78
<1.25
<25
cleaning
Pumping of
Variable
slurry
Limits
>3-4
>64.3
>1.03
<20
>12-15
<12
>1
<15
<78
<1.25
<25
*Controls are not considered necessary for apparent viscosity and yield stress. Whereas fluid loss commonly is judged by
standard filtration test and a maximum film thickness of 2 mm, better control limits are established by stagnation-
gradient tests.
Should be expected to vary widely because of different bentonite brands.
TThe shear strength of filter cake is more applicable to peel-off control (also the time required for its formation).
§Optional.
-------
Table IV.22: Approximate Slurry Wall Costs as a Function of Medium
and Depth (Spooner, Wetzel and Grube, 1982)
SOIL TYPE
SOIL-BENTONITE
WALL*
CEMENT-BENTONITE
WALL*
DEPTH
<" 30
FEET
DEPTH
30-75
FEET
DEPTH
75-120
FEET
DEPTH
<60
FEET
DEPTH
60-150
FEET
DEPTH
>150
FEET
SOFT TO MEDIUM SOIL
N i40
2-4
4-8
8-10
15-20
20-30
30-75
HARD SOIL
N 40-200
4-7
5-10
10-20
25-30
30-40
40-95
OCCASIONAL BOULDERS
4-8
5-8
8-25
20-30
30-40
40-85
SOFT TO MEDIUM ROCK
N -200 SANDSTONE, SHALE
6-12
10-20
20-50
50-60
60-85
85-175
BOULDER STRATA
15-25
15-25
50-80
30-40
40-95
95-210
HARD ROCK
GRANITE, GNEISS, SCHIST*
95-140
140-175
1 75-235
~NOMINAL PENETRATION ONLY
FOR STANDARD REINFORCEMENT IN SLURRY WALLS AND $8.99 PER SQ.FT.
FOR CONSTRUCTION IN URBAN ENVIRONMENT ADD 252 TO 50% OF PRICE
* In 1979 dollars per square foot.
-121-
-------
Colorado and the Gilson Road Hazardous Waste Dump in New Hampshire. A
list of the advantages/disadvantages of slurry-trenches compared to
grouting, sheet piling, pumping or other techniques is shown in Table
IV.23.
SELECTED REFERENCES
American Petroleum Institute, "The Migration of Petroleum Products in
Soil and Ground Water - Principles and Counter Measures", Publication
No. 4149, Dec. 1972, Washington, D.C.
American Petroleum Institute, "Underground Spill Cleanup Manual", API
Publications 1628, June 1980, Washington, D.C.
Blake, S.B. and Lewis, R.W., "Underground Oil Recovery", Proceedings of
the Second National Symposium on Aquifer Restoration and Ground Water
Monitoring. 1982, National Water Well Association, Worthington, Ohio,
pp. 69-76.
Bowen, R., "Specifications for Grouting", Grouting in Engineering
Practice, 2nd ed., Applied Science Publishers LTD, Great Britain, 1981,
pp. 197-239.
Bowles, J.E., "Sheet-Pile Walls - Cantilevered and Anchored", Foundation
Analysis and Design, 2nd ed., McGraw-Hill, 1977, pp. 412-444.
Campbell, M.D. and Lehr, J.H., "Well Cost Analysis", Water Well
Technology, 4th ed., McGraw Hill, 1977.
Chapman, T.G., "Modeling Ground Water Over Sloping Beds", Water
Resources Research, Vol. 16, No. 6, 1980, pp. 1114-1118.
Chauhan, H.S., Schwab, G.O. and Harady, M.Y., "Analytical and Computer
Solutions of Transient Water Tables for Drainage of Sloping Land", Water
Resources Research, Vol. 4, No. 3, 1968, pp. 573-579.
Childs, E.C., "Drainage of Ground Water Resting on a Sloping Bed", Water
Resources Research, Vol. 7, No. 5, 1971, pp. 1256-1263.
Cohen, R.M. and Miller, W.J., "Use of Analytical Models for Evaluating
Corrective Actions at Hazardous Waste Disposal Facilities", Proceedings
of the Third National Symposium on Aquifer Restoration and Ground Water
Monitoring, 1983, National Water Well Association, Worthington, Ohio,
pp. 85-97.
-122-
-------
Table IV.23: Advantages/Disadvantages of Slurry Trenches
Advantages
1. Construction methods are
simple(a).
2. Adjacent areas not affected
by ground water drawdown^3).
3. Bentonite (mineral) will not
deteriorate with age^a^.
4. Leachate-resistant bentonites
are available^*).
5. Low maintenance require-
ments .
6. Eliminate risks due to
pump breakdowns, or power
failures ^).
7. Eliminate headers and other
above ground obstructions^).
Disadvantages(a)
1. Cost of shipping bentonite
from west.
2. Some construction procedures
are patented and will require
a license.
3. In rocky ground, over-
excavation is necessary
because of boulders.
4. Bentonite deteriorates when
exposed to high ionic
strength leachates.
(a) - Tolman, et al., 1978
(b) - Ryan, 1980
123-
-------
D'Appolonia, D.J., "Slurry Trench Cutoff Walls for Hazardous Waste
Isolation", Engineered Construction Internation, Inc., Pittsburgh,
Pennsylvania.
Emrich, G.H., Beck, W.W. and Tolman, A.L., "Top-Sealing to Minimize
Leachate Generation: Case Study of the Windham, Connecticut Landfill",
SMC-Martin, King of Prussia, Pennsylvania.
Ferris, J.G., et al., "Theory of Aquifer Tests", USGS Water Supply Paper
1536E, 1962.
Fung, R. , editor, Protective Barriers for Containment of Toxic
Materials, 1980, Noyes Data Corporation, Park Ridge, New Jersey.
Gibb, J.P., "Cost of Domestic Wells and Water Treatment in Illinois",
Ground Water, Vol. 9, No. 5, Sept.-Oct. 1971, pp. 40-49.
Hantush, M.S., "Growth and Decay of Ground-Water Mounds in Response to
Uniform Percolation", Water Resources Research, Vol* 3, No. 1, 1967, pp.
227-234.
Haxo, H.E., Jr., "Evaluation of Liner Materials", Oct. 1973, U.S.
Environmental Protection Agency, Cincinnati, Ohio.
Huibregtse, K.R. and Kastman, K.H. , "Development of a System to Protect
Groundwater Threatened by Hazardous Spills on Land", EPA-600/2-81-085,
May 1981, National Technical Information Service, Springfield, Virginia.
Huisman, L., "Ground-Water Recovery", Winchester Press, New York, 1972,
336 pp.
Hunt, G.E., et al., "Collection of Information on the Compatibility of
Grouts with Hazardous Wastes", no date, U.S. Environmental Protection
Agency, Cincinnati, Ohio.
Johnson Division, UOP Inc., Ground Water and Wells, 4th ed., St. Paul,
Minnesota, 1975.
Jones, G.K., "Chemistry and Flow Properties of Bentonite Grouts",
Symposium on Grouts and Drilling Muds in Engineering Practice.
International Society of Soil Mechanics and Foundation Engineering,
1963, pp. 22-28.
Kirkham, D., "Seepage of Steady Rainfall through Soil into Drains", AGU
Trans., Vol. 39, 1958, pp. 892-908.
Kirkham, D., "Explanation of Paradoxes in the Dupuit-Forchheimer Seepage
Theory", Water Resources Research, Vol. 3, 1967, pp. 609-622.
-124-
-------
Kufs, C., et al., "Procedures and Techniques for Controlling the
Migration of Leachate Plumes", Ninth Annual Research Symposium - Land
Disposal, Incineration and Treatment of Hazardous Waste, 1983.
Lu, J.C.S., Morrison, R.D. and Stearns, R.J., "Leachate Production and
Management from Municipal Landfills: Summary and Assessment", Land
Disposal: Municipal Solid Waste - Proceedings of the Seventh Annual
Research Symposium, U.S. Environmental Protection Agency, 1981, pp. 1-
17.
Lundy, D.A. and Mahan, J.S., "Conceptual Designs and Cost Sensitivities
of Fluid Recovery Systems for Containment of Plumes of Contaminated
Groundwater", Proceedings of the National Conference on Management of
Uncontrolled Hazardous Waste Sites, 1982, Hazardous Materials Control
Research Institute, Silver Spring, Maryland, pp. 136-140.
Luthin, J.N., Drainage of Agricultural Lands, 1st ed., American Society
of Agronomy, Madison, Wisconsin, 1957.
Marei, S.M. and Towner, G.D., "A Hele-Shaw Analog Study of the Seepage
of Ground Water Resting on a Sloping Bed", Water Resources Research,
Vol. 11, No. 4, 1975, pp. 589-594.
McBean, E.A., et al., "Leachate Collection Design for Containment
Landfills", American Society of Civil Engineers Journal of Environmental
Engineering Division, Vol. 108, No. 1, 1982, pp. 204-209.
Morgenstern, N.R. and Vaughan, P.R., "Some Observations on Allowable
Grouting Pressures", Symposium on Grouts and Drilling Muds in
Engineering Practice, International Society of Soil Mechanics and
Foundation Engineering, 1963, pp. 36-46.
Mualem, Y. and Bear, J., "Steady Phreatic Flow over a Sloping
Semipervious Layer", Water Resources Research, Vol. 14, No. 3, 1978, pp.
403-408.
Pastrovich, T.L., et al., "Protection of Groundwater From Oil
Pollution", Report No. 3/79, CONCAWE, The Hague 1979.
Peurifoy, R.L., "Foundation Grouting", Construction Planning, Equipment,
and Methods, 3rd Ed., McGraw-Hill, New York, 1979, pp. 491-505.
Pye, V.I., Patrick, R. and Quarles, J., Groundwater Contamination in the
United States, University of Pennsylvania Press, Philadelphia, 1983, p.
57.
Quince, J.R. and Gardner, G.L., "Recovery and Treatment of Contaminated
Ground Water", The Second National Symposium on Aquifer Restoration and
Ground Water Monitoring", May 1982.
125-
-------
Ryan, C.R. , "Slurry Cut-Off Walls Methods and Applications", Mar. 1980,
Geo-Con, Inc., Pittsburgh, Pennsylvania.
Shultz, D.W. and Miklas, M.P. , "Procedures for Installing Liner
Systems", Land Disposal of Hazardous Waste: Proceedings of the Eighth
Annual Research Symposium, U.S. Environmental Protection Agency, 1982,
pp. 224-238.
Singh, S.R. and Jacob, C.M., "Transient Analysis of Phreatic Aquifers
Lying Between Two Open Channels", Water Resources Research, Vol. 13, No.
2, 1977, pp. 411-419.
Smith, M.E., "Sanitary Landfills", Conference on Reclamation of
Difficult Sites, 1983, University of Wisconsin Extension, Madison,
Wisconsin.
Spangler, M.G. and Handy, R.L., "Soil Water", Soil Engineering, 3rd ed.,
Intext Educational Publishers, New York, 1973, pp. 191-209.
Spooner, P.A., Wetzel, R.S. and Grube, W.E., "Pollution Migration Cut-
off Using Slurry Trench Construction", Management of Uncontrolled
Hazardous Waste Sites, Hazardous Materials Control Research Institute,
1982, pp. 191-197.
Tolman, A.L., et al., "Guidance Manual for Minimizing Pollution from
Waste Disposal Sites", EPA-600/2-78-142, Aug. 1978, U.S. Environmental
Protection Agency, Cincinnati, Ohio.
Towner, G.D., "Drainage of Ground Water Resting on a Sloping Bed with
Uniform Rainfall", Water Resources Research, Vol. 11, No. 1, 1975, pp.
144-147.
U.S. Environmental Protection Agency, Survey of Solidification/Stabili-
zation Technology for Hazardous Industrial Wastes, EPA-600/2-79-056,
July 1979.
U.S. Environmental Protection Agency, "Handbook for Remedial Action at
Waste Disposal Sites", EPA-625/6-82-006, June 1982, Cincinnati, Ohio.
Xanthakos, P., Slurry Walls, 1st edition, McGraw-Hill, 1979.
-126-
-------
SECTION V
TREATMENT OF GROUND WATER
Concepts for treatment of contaminated ground water following its
pumpage from the subsurface are well established. These concepts have
been developed from years of experience treating industrial wastewaters.
However, confusion may occur when these concepts are applied to site
specific contaminated ground waters. The type of treatment required
depends primarily on the contaminants being removed. Treatment systems
may be relatively simple when a single chemical is the only contaminant,
or extremely complex for cases involving numerous contaminants. In most
cases, treatability studies should be conducted with representative
samples to determine appropriate treatment components. This section
addresses air stripping, activated carbon, and biological treatment for
organics in ground water, and chemical precipitation for inorganics in
ground water. Due to the growing importance of the first two processes,
Appendix C contains more detailed information on air stripping, and
Appendix D does similarly for activated carbon adsorption.
AIR STRIPPING
Air stripping is a mass transfer process in which a substance in
solution in water is transferred to solution in a gas. The rate of mass
transfer depends upon several factors according to the following
equation.
M - Kl a(CL - Cg)
where:
-127-
-------
M = mass of substance transferred per unit time and volume
(gm/hr/m3)
K-l = coefficient of mass transfer (m/hr)
a = effective area (m^/m^)
(CL~Cg) = driving force (concentration difference between liquid phase
and gas phase, gm/m^)
The driving force is an important aspect in the success or failure of an
air stripping process. The driving force is the difference between
actual conditions in the air stripping unit and conditions associated
with equilibrium between the gas and liquid phases. If equilibrium
exists at the air-liquid interface, the liquid phase concentration is
related to the gas phase concentration by Henry's law which states:
Cig ¦ HCi£
where
Cig = equilibrium concentration in gas phase (gm/m^)
^i£ a equilibrium concentration in liquid phase (gm/m^)
H = Henry's law constant
another form of Henry's law is
PA a H'C
where
Pa = partial pressure of substance in the air mixture in contact
with the water at equilibrium (atm.)
H' = Henry's law constant (atm./gm/m^)
C ¦ concentration of substance in the water at equilibrium (gm/m^)
The Henry's law constant can be used to predict the strippability of a
chemical. A compound with a high Henry's law constant generally is more
easily stripped from water than one with a lower Henry's law constant.
-128-
-------
The mass of a contaminant that can be transferred not only is a
function of the driving force, but is also a function of the mass
transfer. Due to the difficulty in separating these two parameters,
they are usually evaluated as one coefficient, KLa. The coefficient
is a function of the geometry and physical characteristics of the air
stripping equipment, the loading rate (air-to-water ratio), and the
temperature. These are factors that must be considered in the design of
an air stripping process.
Description of Process
As shown in Figure V.l, there are four basic equipment
configurations used for air stripping, including diffused aeration,
counter-current packed columns, cross-flow towers, and coke tray
aerators. Diffused aeration stripping uses aeration basins similar to
standard wastewater treatment aeration basins. Water flows through the
basin from top to bottom with the air dispersed through diffusers at the
bottom of the basin. The air-to-water ratio is significantly lower than
in either the packed column or the cross-flow tower.
In the counter-current packed column, water containing one or more
impurities is allowed to flow down through a column containing packing
material with air flow counter-currently up through the column. In this
way the contaminated water comes into intimate contact with clean air.
Packing materials are used which provide high-void volumes and high-
surface area. In the cross-flow tower, water flows down through the
packing as in the counter-current packed column, however, the air is
pulled across the water flow path by a fan. The coke tray aerator is a
-129-
-------
EXITjAIR
INFLUENT-
nnnmifiJ
-DISTRIBUTOR
PACKING
MATERIAL
SUPPORT
PLATE
t=Z— INCOMING
AIR
EFFLUENT)
PACKED COLUMN
RAW WATER
INLET-<=D
~
~ ~~~
DISTRIBUTING
SPLASH
APRONS
OUTLET
INFLUENT
AIR SUPPLY
DIFFUSER-
GRID
ar
-EFFLUENT
DIFFUSED AIR BASIN
AIR .OUTLET
WATER
INLET
AIR
INLE
WATER
OUTLET
WATER
INLET
INLET
COLLECTION
BASIN
COKE TRAY AERATOR
CROSS-FLOW TOWER
Figure V.l: Air Stripping Equipment Configurations
-130-
-------
simple, low-maintenance process. The water being treated is allowed to
trickle through several layers of trays. This produces a large surface
area for gas transfer.
The counter-current packed tower appears to be the most appropriate
equipment configuration for treating contaminated ground waters for the
following reasons:
(1) It provides the most liquid interfacial area.
(2) High air-to-water volume ratios are possible due to the low
air pressure drop through the tower.
(3) Emission of stripped organics to the atmosphere may be
environmentally unacceptable; however, a counter-current tower
is relatively small and can be readily connected to vapor
recovery equipment.
Design Parameters and Procedures
The design of an air stripping process for stripping volatile
organics from contaminated ground water is accomplished in two steps.
The cross-sectional area of the column is determined and then the height
of the packing is determined. The cross-sectional area of the column is
determined by using the physical properties of the air flowing through
the column, the characteristics of the packing, and the air-to-water
flow ratio. A key factor is the establishment of an acceptable air
velocity. A general rule of thumb used for establishing the air
velocity is that an acceptable air velocity is 60% of the air velocity
at flooding. Flooding is the condition in which the air velocity is so
high that it holds up the water in the column to the point where the
water becomes the continuous phase rather than the air. If the air-to-
water ratio is held constant, the air velocity determines the flooding
-131-
-------
condition. For a selected air-to-water ratio, the cross-sectional area
is determined by dividing the air flow rate by the air velocity. The
selection of the design air-to-water ratio must be based upon experience
or pilot-scale treatability studies. Treatability studies are
particularly important for developing design information for
contaminated ground water.
The height of column packing may be determined by the following
equation:
Ix2 - Yi/H)
£n
Z =
(Xx - Yi/H)
(1-A) + A
KLa C (1-A) (1-X) M
where
Z = height of packing, feet
L = water velocity, lb moles/hr/ft^
X2 = influent concentration of pollutant in ground water, mole
fraction
Xj = effluent concentration of pollutant in ground water, mole
fraction
K^a = mass transfer coefficient, 1/hr
C = molar density of water = 3.47 lb mole/ft^
H = Henry's law constant, mole fraction in air per mole fraction
in water
G = air velocity, lb mole/hr/ft^
A -
A HG
(l-X)M = the average of one minus the equilibrium water concentration
through the column
Yi = influent concentration of pollutant in air, mole fraction
In most cases, the following assumptions can be made:
(1) Yj * 0, there should be no pollutants in the influent air.
-132-
-------
(2) (l-X)M = 1, the influent concentrations should be too small
when converted to mole fraction to shift this term
significantly from 1.0.
The equation for determining the column packing height is then
reduced to:
X? .
L
Z =
Xo
jq- (1-A) + A
KLa C (1-A)
All variables except the mass transfer coefficient are set by the design
conditions. The selection of a suitable mass transfer coefficient must
be based upon pilot scale treatability studies and the judgment of the
design engineer. The mass transfer coefficient is a function of:
(1) Type of volatile organic to be removed
(2) Air-to-water ratio
(3) Temperature of the ground water
(4) Type of packing
(5) Geometry of the tower
A plot of experimental data as shown in Figure V.2 can be helpful in
selecting the proper mass transfer coefficient.
Applications to Ground Water
Air stripping has been successfully used for removing volatile
organics from contaminated ground waters. These successful operations
include an industrial park investigated by Stover (1982), a case study
reported by Pekin and Moore (1982), and Rockaway Township, New Jersey,
reported by McKinnon and Dyksen (1982). A summary of the removals
achieved in these operations for various organic contaminants at various
air-to-water ratios is given in Table V.l.
-133-
-------
1 I ! I I 111] —T-—r~l—I l u II | 1 1—rr
200 -
100
•0
•0
70
»0
*0
40
so
* 20-
PACKING B
1'RINQ
9/0*RING
PACKING A
' ' I I I i i i I I L
» I ¦ "I
J I I 1
1,000
10.000 so >0 40 S0.000
WATER VELOCITY.
-------
Table V.l: Packed Column Air Stripping of Volatile Organics
Organic Contaminant
Ai r:Water
Influent
Effluent
Ratio
ug/£
y9 /i
1,1,2-Tnchloroethylene
9.3
80
16
96.3
80
3
27.0
75
16
156.0
813
52
44.0
218
40
75.0
204
36
125.0
204
27
1 ,1 ,1-Trichloroethane
9.3
1200
460
96.3
1200
49
27.0
90
31
156.0
1332
143
1,1-Dichlorethane
9.3
35
9
96.3
35
1
1,2-Dichloropropane
27.0
50
<5
146.0
70
<5
156.0
377
52
Chloroform
27.0
50
<2
146.0
57
<2
Diisopropyl ether
44.0
15
7
75.0
14
6
125.0
4
4
-135-
-------
It has been pointed out that temperature has an effect on the mass
transfer coefficient. This becomes important when the contaminated
ground water contains compounds that are very soluble. This makes their
removal by ambient temperature air stripping almost impossible.
Lamarre, McGarry, and Stover (1983) have reported on such a case. It
was found in methyl ethyl ketone removal, that at 54°F and an air-to-
water ratio of 490, only 43% removal occurred; at 90°F and an air-to-
water ratio of 513, removal was 922; and at 136°F and an air-to-water
ratio of 469, the removal was 99Z. Removal efficiency increased
dramatically with temperature and less sharply with the air-to-water
ratio. Therefore, high temperature air-stripping, or steam stripping,
should be investigated for each case encountered.
Cos ts
Although the variation in the design of packed column air stripping
systems results in various costs, the major components of an air
stripping system for removing organic contaminants from a ground water
would include the packed column, the air supply equipment, and re-
pumping. Annual costs per 1,000 gallons of water treated has been
presented by Dyksen, et al. (1982), and Figure V.3 provides a summary.
The costs are 1982 costs and are based upon preliminary designs for
achieving 90 percent removal of trichloroethylene (TCE).
CARBON ADSORPTION
Adsorption occurs when an organic molecule is brought to the
activated carbon surface and held there by physical and/or chemical
forces. The quantity of a compound or group of compounds that can be
-136-
-------
s
<
111
K
flC
U
t-
<
u.
O
<
o
o
o
o
ae
lu
a.
(0
i-
z
IU
o
t-
(0
o
o
1000
100
10
J J— 1
»
-
RANGE OF
—
COSTS FOR
PACKED COLUMNS/ ^
-
*
1
0.01
0.1
SYSTEM SIZE - MOD
1.0
6.0
NOTES:
t. ANNUAL COSTS MCLUOES AMORTIZED CAPITAL
COSTS AND ANNUAL OPERATING COST8
2. SYSTEM SIZE REPRESENTS AVERAGE PLANT CAPACITY.
9. RANGE OP COSTS FOR PACKED COLUMN ACCOUNTS FOR
DEFERENCES M MATERIALS OF CONSTRUCTION AS
DISCftteCD IN THE TEXT
Figure V.3:
Comparison of Costs for Packed Column Aeration (Dyksen
et al., 1982)
"137-
-------
adsorbed by activated carbon is determined by a balance between the
forces that keep the compound in solution and the forces that attract
the compound to the carbon surface. Factors that affect this balance
include:
(1) Adsorptivity increases with decreasing solubility.
(2) The pH of the water can affect the adsorptive capacity.
Organic acids adsorb better under acidic conditions, whereas
amino compounds favor alkaline conditions.
(3) Aromatic and halogenated compounds adsorb better than
aliphatic compounds.
(4) Adsorption capacity decreases with increasing temperature
although the rate of adsorption may increase.
(5) The character of the adsorbent surface has a major affect on
adsorption capacity and rate. The raw materials and the
process used to activate the carbon determine its capacity.
When activated carbon particles are placed in water containing
organic chemicals and mixed to give adequate contact, the adsorption of
the organic chemicals occurs. The organic chemical concentration will
decrease from an initial concentration, CQ, to an equilibrium
concentration, Ce. By conducting a series of adsorption tests, it is
usually possible to obtain a relationship between the equilibrium
concentration and the amount of organics adsorbed per unit mass of
activated carbon.
The Freundlich isotherm and the Langmuir isotherm are most often
used to represent the adsorption equilibrium. One form of the
Freundlich isotherm is:
- = KCe1fn
m
where
X = mass of organic adsorbed (gm)
-138-
-------
ni = mass of activated carbon (gm)
Ce = equilibrium concentration of organics (gm/m^)
K, n = experimental constants
The Langmuir isotherm has the following form:
X = a K Ce
m 1 + K Ce
where
a = mass of adsorbed organic required to completely saturate a
unit mass of carbon (gm)
K = experimental constant
From an isotherm test it can be determined whether or not a
particular organic material can be removed effectively. It will also
show the approximate capacity of the carbon for the application, and
provide a rough estimate of the carbon dosage required. Isotherm tests
also afford a convenient means of studying the effects of pH and
temperature on adsorption. Isotherms put a large amount of data into
concise form for ready evaluation and interpretation. Isotherms
obtained under identical conditions using the same contaminated ground
water for two or more carbons can be quickly and conveniently compared
to determine the relative merits of the carbons.
Figure V.4 represents a typical Freundlich isotherm for the
comparison of two carbons removing the organics from a contaminated
ground water. The isotherm for carbon A is at a high level and has only
a slight slope. This means that adsorption is large over the entire
range studied. The isotherm for carbon B is at a lower range and with a
steeper slope. This means lower adsorption than carbon A. However,
-139-
-------
CARBON A
CARBON B
In Ce
Figure V.4: Adsorption Isotherm
-140-
-------
adsorption improves at higher concentrations. In general, the steeper
the slope of its isotherm, the greater the efficiency of a carbon in
column operation.
Description of Process
Activated carbon adsorption may be accomplished by batch, column,
or fluidized-bed operations. The usual contacting systems are fixed bed
or countercurrent moving beds. The fixed beds may employ downflow or
upflow of water. The countercurrent moving beds employ upflow of the
water and downflow of the carbon since the carbon can be moved by the
force of gravity. Both fixed beds and moving beds may use gravity or
pressure flow.
Figure V.5 shows a typical fixed-bed carbon column employing a
single column with downflow of the water. The column is similar to a
pressure filter and has an inlet distributor, an underdrain system, and
a surface wash. During the adsorption cycle the influent flow enters
through the inlet distributor at the top of the column, and the ground
water flows downward through the bed and leaves through the underdrain
system. The unit hydraulic flow rate is usually from 2 to 5 gpm/ft^.
When the head loss becomes excessive due to the accumulated suspended
solids, the column is taken off-line and backwashed.
Figure V.6 shows a typical countercurrent moving-bed carbon
column employing upflow of the water. Two or more columns are usually
provided and are operated in series. The influent contaminated ground
water enters the bottom of the first column by means of a manifold
system which uniformly distributes the flow across the bottom. The
-141-
-------
INFLUENT
INFLUENT
DISTRIBUTOR
IX1
« s
c z
< <
O
kS
z <
u ce
ft o
kCB /
TRANSPORT
WATER
777777/
'#///£/,
TRANSPORT
WATER
-01 * EFFLUENT
WA8H WATER
Figure V.5: Fixed-Bed Adsorption System
-142-
-------
CARBON FILLING
CHAMBER
FRESH CARBON
WLET
CARBON
VALVE
I
WA8TE
-FLUENT
OUTLET
.CARBON
VALVE
¦N >
PORT
4>4-
jKf8i
F
CARBON
^EFFLUENT
-Hi
SXk
23T
Figure V.6: Moving-Bed Adsorption System
-143-
-------
ground water flows upward through the column. The unit hydraulic flow
rate is usually 2 to 10 gpra/ft^. The effluent is collected by a screen
and manifold system at the top of the column and flows to the bottom
manifold of the second column. The carbon flow is not continuous but
instead is pulse-wise.
The fluidized bed consists of a bed of activated carbon. The water
flows upward through the bed in the vertical direction. The upward
liquid velocity is sufficient to suspend the activated carbon so that
the carbon does not have constant interparticle contact. At the top of
the carbon there is a distinct interface between the carbon and the
effluent water. The principal advantage of the fluidized bed is that
waters with appreciable suspended solids content may be given adsorption
treatment without clogging the bed, since the suspended solids pass
through the bed and leave with the effluent. This should not be a
concern with ground waters.
Design Parameters and Procedures
Although the treatability of a particular wastewater by carbon and
the relative capacity of different types of carbon for treatment may be
estimated from adsorption isotherms, carbon performance and design
criteria are best determined by pilot column tests. Design-related
information which can be obtained from pilot tests includes:
(1) Contact time
(2) Bed depth
(3) Pretreatment requirements
(4) Breakthrough characteristics
-144-
-------
(5) Headloss characteristics
(6) Carbon dosage in pounds of pollutants removed per pound of
carbon.
The design of an activated carbon adsorption column can be
accomplished by using a kinetic equation which is based on a derivation
by Thomas (1948). The method requires data obtained from a breakthrough
curve. A typical breakthrough curve obtained from a pilot test is shown
in Figure V.7. The expression by Thomas for an adsorption column is as
follows:
co . (AqM - CQV)
— = 1 + exp Q ^ °
where
C = effluent pollutant concentration (gra/m^)
CQ = influent pollutant concentration (gm/m^)
K-i « rate constant m-* per day per gm
Q = flow rate, m^ per day
Aq = adsorption capacity, gm per gm
M =¦ mass of carbon, gm
V = throughput volume, m-*
Rearranging and taking the logarithms of both sides yields
, C0 Ki AqM KiCoV
In (- - 1) - —
K1 AqM
This is the equation of a straight line in which — is the y-
*1 CD
intercept and —— is the slope of the line. The pilot column tests
C0
provides all parameters except and Aq. If In (— - 1), is plotted
versus V, the values of K} and Aq can be determined by a graphical
solution. Then the mass of carbon needed for a selected breakthrough
volume may be determined.
-145-
-------
BREAKTHROUGH
THROUGHPUT VOLUME
Figure V.7: Typical Breakthrough Curve
-146-
-------
Applications to Ground Water
Activated carbon adsorption has been successfully used for removing
organics from contaminated ground waters. Kaufman (1982) reported on 17
separate cases. The aquifers had been contaminated by leachate from
lagoons or dumpsites, industrial accidents, and chemical spills due to
railroad or truck accidents. Activated carbon successfully treated
these contaminated aquifers. McDougall, Fusco, and O'Brien (1980)
reported on the successful treatment of the Love Canal landfill leachate
by activated carbon adsorption. O'Brien and Fisher (1983) have reported
on 31 separate cases where activated carbon adsorption has been
successful in treating contaminated ground waters. The reasons for
treatment were to prevent the spread of contamination through an
aquifer, for process water reuse, and for decontamination of potable
water. Table V.2 provides some results obtained in these operations.
In all cases, the activated carbon was very successful in removing
priority pollutants.
Costs
The cost of treating ground water by activated carbon adsorption is
dependent on a number of factors such as: flow rates, concentrations,
type of contaminant; type of application; and site requirements. These
factors determine equipment and carbon usage; thus it is difficult to be
specific on costs. However, Kaufmann (1982) reported that for municipal
potable treatment the costs have been in the $0.40 to $1.00 per person
per month range. O'Brien and Fisher (1983) have reported that if the
influent concentration is in the mg/£ range the costs have ranged from
-147-
-------
Table V.2: Activated Carbon Adsorption of Organics
Contaminants
Influent
concentration
P9/£
Effluent
Concentration
pg/*
Phenol
63,000
<100
2,400
< 10
40,000
< 10
Carbon tetrachloride
61,000
< 10
130,000
< 1
73,000
< 1
1,1,2- Tetrachloroethane
80,000
< 10
Tetrachloroethylene
44,000
12
70,000
< 1
1,1,1-Trichloroethane
23,000
N.D.
1,000
< 1
3,300
< 1
12,000
< 5
143,000
< 1
115
< 1
Benzene
2,800
< 10
400
< 1
11,000
<100
2,4- Dichlorophenol
5,100
N.D.
N.D. - Nondetectable
-148-
-------
$0.48 to $2.52 per 1000 gallons treated. However, if the influent
concentrations are in the Ug/£ range, the costs have varied from $0.22
to $0.55 per 1000 gallons treated.
BIOLOGICAL TREATMENT
In biological treatment of contaminated ground water, the objective
is to remove or reduce the concentration of organic and inorganic
compounds. Because many of the compounds that may be present in
contaminated ground water are toxic to microorganisms, pretreatment of
the ground water may be required. When a ground water containing
organic compounds is contacted with microorganisms, the organic material
is removed by the microorganisms through metabolic processes. The
organic compounds may be used by the microorganisms to form new cellular
material or to produce energy that is required by the microorganisms for
their life systems. It has been found by Tabak, et al. (1981) and
Kincannon and Stover (1981a, b, c, no date); Kincannon, Stover and Chung
(1981); and Kincannon, et al. (1981; 1982a,b) that many organic
compounds that are considered to be toxic are biodegraded by
microorganisms when the proper environment is provided.
Heterotrophic microorganisms are the most common group of
microorganisms providing the metabolic process for removing organic
compounds from contaminated ground water. Heterotrophs use the same
substances as sources of carbon and energy. A portion of the organic
material is oxidized to provide energy while the remaining portion is
used as building blocks for cellular synthesis. Three general methods
-149-
-------
exist by which heterotrophic microorganisms can obtain energy. These
are fermentation, aerobic respiration, and anaerobic respiration.
In the case of fermentation, the carbon and energy source is broken
down by a series of enzyme-mediated reactions which do not involve an
electron transport chain. In aerobic respiration, the carbon and energy
source is broken down by a series of enzyme-mediated reactions in which
oxygen serves as an external electron acceptor. In anaerobic
respiration, the carbon and energy source is broken down by a series of
enzyrae-mediated reactions in which sulfates, nitrates, and carbon
dioxide serves as the external electron acceptors. The three processes
of obtaining energy forms the basis for the various biological waste
water treatment processes.
Description of Process
Biological treatment processes are typically divided into two
categories; suspended growth systems and fixed-film systems. Suspended
growth systems are more commonly referred to as activated sludge
processes, of which several variations and modifications exist. The
basic system consists of a large basin into which the contaminated water
is introduced, and air or oxygen is introduced by either diffused
aeration or mechanical aeration devices. The microorganisms are present
in the aeration basin as suspended material. After the microorganisms
remove the organic material from the contaminated water they must be
separated from the liquid stream. This is accomplished by gravity
settling. After separating the biomass from the liquid, the biomasa
increase resulting from synthesis is wasted and the remainder returned
-150-
-------
to the aeration tank. Thus, a relatively constant mass of
microorganisms is maintained in the system. The performance of the
process depends on the recycle of sufficient biomass. If biomass
separation and concentration fails the entire process fails. The
process requires the skills of well-trained operators.
Fixed-film biological processes differ from suspended growth
systems in that microorganisms attach themselves to a medium which
provides an inert support. Biological towers (trickling filters) and
rotating biological contactors are the most common forms of fixed-film
processes. The original trickling filter consisted of a bed of rocks
over which the contaminated water was sprayed. The microbes, forming a
slime layer on the rocks, would metabolize the organics, with oxygen
being provided as air moved counter current from the water flow. A
major drawback to the rock-bed trickling filter was the inefficient use
of space and poor oxygen transfer.
Biological towers are a modification of the trickling filter. The
media, which is comprised of polyvinyl chloride (PVC), polyethylene,
polystyrene, or redwood is stacked into towers which typically reach 16
to 20 feet high. The contaminated water is sprayed across the top and,
as it moves downward, air is pulled upward through the tower. A slime
layer of microorganisms forms on the media and removes the organic
contaminants as the water flows over the slime layer.
A Rotating Biological Contactor (RBC) consists of a series of
rotating discs, connected by a shaft, set in a basin or trough. The
contaminated water passes through the basin where the microorganisms,
attached to the discs, metabolize the organics present in the water.
-151-
-------
Approximately 40 percent of the disc's surface area is submerged. This
allows the slime layer to alternately come in contact with the
contaminated water and the air where oxygen is provided to the
microorganisms.
Removal efficiencies are generally the same for fixed-film and
suspended growth processes. However, fixed-film processes have the
potential to be lower in cost, due to the absence of aeration equipment,
and are easier to operate. Both systems may be operated under anaerobic
conditions, which may offer advantages for certain contaminated waters.
Design Parameters and Procedures
The design of biological treatment processes is usually
accomplished by using some type of kinetic model. The most widely used
models have been developed by Eckenfelder, McKinney, Lawrence and
McCarty, and Gaudy. However, Kincannon and Stover (no date) have found
that a great amount of variability exists in the biokinetic constants
for these models, and concluded that these models were not ideal for
waters containing priority pollutants; therefore, they have developed
the following models that are reliable for these types of waters:
Activated Sludge
FSi/X
V =
umax si
KB
Si - se
Biological Tower and Rotating Biological Contactor
FSi
A »
umax si
-------
where
V = volume of aeration tank (m^)
F = flow rate (m^/day)
X = mixed liquor volatile solids (mg/1)
Si = influent BOD, COD, TOC, or specific organics (mg/1)
Se = effluent BOD, COD, TOC, or specific organics (mg/1)
Umax an<* Kb = biokinetic constants (day~l)
A " surface area of biological tower or rotating biological
contactor (m^)
The biokinetic constants are determined by conducting laboratory or
pilot plant studies. After the biokinetic constants are determined, the
required volume of aeration tank or the required surface area for a
biological tower or rotating biological contactor can be determined for
any flow rate; influent concentration of BOD, COD, TOC, or specific
organic, and a required effluent concentration of BOD, COD, TOC, or
specific organic.
Applications to Ground Water
Biological treatment of contaminated ground water has not been
actively employed in the past. However, pilot plant studies by
Kincannon and Stover (1983) showed that biological treatment of a
contaminated ground water was feasible. In addition, Kincannon and
Stover (no date) have conducted extensive research on the fate of
priority pollutants in activated sludge treatment systems. The removal
mechanisms of 24 priority pollutants are shown in Table V.3. It is seen
that a majority of the 24 organic priority pollutants are removable by
biological treatment. Figure V.8 shows the effluent phenol
-153-
-------
Table V.3: Removal Mechanisms of Toxic Organics
Compound
Percent Treatment Achieved
Stripping Sorption Biologica
Nitrogen Compounds
Acrylonitrile -- -- 99.9
Phenols
Phenol -- -- 99.9
2,4-DNP — — 99.3
2,4-DCP — — 95.2
PCP — 0.58 97.3
Aromatics
1.2-DCB 21.7 — 78.2
1.3-DCB
Nitrobenzene — — 97 „ 8
Benzene 2,0 — 97.9
Toluene 5.1 0.02 94.9
Ethyl benzene 5.2 0.19 94.6
Halogenated Hydrocarbons
Methylene Chloride 8.0 -- 91.7
1,2-DCE 99.5 0.50
1,1,1-TCE 100.0
1,1,2,2-TCE 93.5
1,2-DCP 99.9
TCE 65.1 0.83 33.8
Chloroform 19.0 1.19 78.7
Carbon Tetrachloride 33.0 1.38 64.9
Oxygenated Compounds
Acrolein -- " 99.9
Polynuclear Aromatics
Phenanthrene — -- 98.2
Naphthalene — -- 98.6
Phthalates
Bis(2-Ethylhexyl) — -- 76.9
Other
Ethyl Acetate 1.0 — 98.8
-154-
-------
500
400 -
Ln
Ln
I
a
%
O 300 -
iu
X
a.
IU
3 200H
b.
u.
ui
100-
PHENOL LOADING, lbs ph«nol/day/1000 ft'
Figure V.8: Effluent Phenol versus Phenol Loading Rate
-------
concentration achieved by an RBC as a function of the phenol loading.
It is seen that very low levels of effluent phenol can be achieved by
the RBC process.
Kincannon and Stover (1983) have suggested alternative designs for
biological treatment of contaminated ground water in which treatability
studies were conducted. In this case, the contaminated ground water
undergoes metals removal and high temperature air stripping before
biological treatment. The quality of the water after metals removal and
stripping is shown in Table V.4. Activated sludge and RBC pilot plant
studies were conducted and designs suggested for both systems. These
designs are shown in Tables V.5 and V.6. The projected effluent
qualities from the two systems are shown in Tables V.7 and V.8. It is
seen that an excellent effluent is expected. Table V.9 gives the
expected priority pollutant levels for various biological treatment
processes.
Costs
There are no good data available at this time in regards to the
cost of biological treatment of contaminated ground waters. However,
the costs should be comparable to those for treating industrial
wastewaters.
CHEMICAL PRECIPITATION
Chemical addition for the removal of inorganic compounds is a well-
established technology. There are three common basic types of chemical
addition systems which depend upon the low solubility of inorganics at a
specific pH; (1) the carbonate system, (2) the hydroxide system, and (3)
-156-
-------
Table V.4: High Temperature Air Stripper Extract Water Qualities
for Biological Studies
Parameter Well #1 Well #2
PH
6.7
7.2
Specific Conductance, U MHOg
450
1300
TOC, mg/1
250
590
COD, mg/1
700
1960
BOD5. mg/1
450
1025
Total Phenols, mg/1
to
•
4.0
SS, mg/1
25
30
VSS, mg/1
10
16
TDS, mg/1
380
1860
TVDS, mg/1
120
1220
Fe, mg/1
1.7
0.4
Mn, mg/1
0
2.3
NH3-N, mg/1
5.5
9.0
Ortho-P, mq/1
0.2
0.2
-157-
-------
Table V„5r Design Summary for Activated Sludge System
Reactor Volume
MLVSS
Maximum BOD Removed
Oxygen Requirements
Energy Oxygen
Endogenous Oxygen
Total Oxygen
Nutrient Requirements
Nitrogen
Phosphorus
Alpha (a)
Beta (3)
Water Temperature
Elevation
Oxygen Transfer Characteristics
No
N
Diffused Air
Net Sludge Production
Secondary Clarifier
Overflow Rate
Underflow Solids Concentration
Recommended Recycle Pump Capacity
Sludge Dewatering Characteristics
Vacuum Filter Yield
Cake Dry Solids Content
High Molecular Weight
Cationic Polymer
Ferric Chloride
60,000 gallons, or
two-30,000 gallons each
1500 mg/1
600 lbs B0D5/day
0.4 lbs 02/lb BOD5 removed
6.3 lbs 02/hr/l,000 lbs MLVSS UA
710 lbs 02/day
30 lbs/day, as N
6 lbs/day, as P
0.61
0.99
90°F (32°C)
300 ft
3.0 lbs 02/Hp-hr
1.9 lbs 02/Hp-hr
500 Scfm
0.42 lbs/lb BOD5 removed
250 lbs/day
500 gpd/ft^
10,000 mg/1
Variable up to 50 gpra
1.0 to 1.5 lb/hr/ft2
12 to 18%
20-40 lb/ton
70-100 lb/ton
-158-
-------
Table V.6: Design Summary for RBC System
Surface Area
Maximum BOD Removed
Nutrient Requirements
Nitrogen
Phosphorus
Net Sludge Production
Secondary Clarifier
Overflow Rate
Underflow Solids Concentration
Sludge Dewatering Characteristics
300,000 ft2
Three Standard Density RBC Shafts
Two Stages
Two Shafts/First Stage
600 lbs B0D5/day
30 lbs/day, as N
6 lbs/day, as P
0.25 lbs/lb BOD5 removed
150 lbs/day
500 gpd/ft2
10,000 mg/1
Similar to Activated Sludge
(Based on CST Comparisons)
-159-
-------
Table V.7:
Projected Effluent Qualities from Activated Sludge Design
Soluble Effluent
Qualities, mg/1
Flow = 25 gpm
x = 1500a x = 2000
Flow = 50 gpm
x = 1500 x = 2000
Well #1 Quality
V = 30,000 gallons
BOD5
COD
TOC
10
53
28
9
52
26
13
57
33
11
55
30
Well #1 Quality
V = 60,000 gallons
BOD5
COD
TOC
8
51
25
7
50
24
10
53
28
9
52
26
Well #2 Quality
V = 30,000 gallons
BOD5
COD
TOC
12
148
88
5
125
74
28
231
137
19
191
114
Well #2 Quality
V = 60,000 gallons
BOD5
COD
TOC
1
104
60
1
95
55
10
145
85
5
126
75
ax denotes mixed liquor volatile suspended solids (mg/1)
-160-
-------
Table V.8:
Projected Effluent Quality from RBC Design
Soluble Effluent
Quality Flow = 25 gpra Flow = 50 gpm
Well #2 Quality
3 RBC Shafts
BOD 5
1
2
COD
94
96
TOC
60
65
-161-
-------
Table V.9: Results of Specific Compound Studies
Specific
Oorfxxird
frms
Fxtract
Ttrtal Fhenols
Arid Extractables
2-Ch lorcphenol
Etienol
2, 4,-Dimethylphenol
Pentacholrcphenol
Base Neutral Rxtractables
1, 4-Dichlorcbereene
1, 2-Dichlorohenzene
Naphthalene
2500(4000)
1000(200)
675(990)
<10(5300)
t®(510
105(85)
1920(430)
715(55)
A.S.
#1
Well #1
80(130)
00(<10)
£10(710)
00(00)
ND(95)
<10(00)
335(425)
70(50)
A.S.
#2
Well #2
60(115)
<10(00)
00(00)
00(00)
M)(80)
00(00)
335(155)
80(25)
A.S.
#3
(Well #2
A.L.
Water)
RFC
50(100) 50(70)
00(00)
00(00)
<10(00)
fJD(70)
00(00)
110( Z70)
25(<10)
00(00)
00(00)
00(00)
ND( <10)
00(00)
100(<10)
25 (<10)
-(100)
"(<10)
-(<10)
-(<10)
M5T90)
-(<10)
-(<10)
-(<10)
Malatile Organics
N.D. (N.D.)
All analysis in ntLiTujrais/li ter
N.D. - Ncne Detected
A.S. #1 - High Ffo Acti\«ted Sludae
A.S. #2 - Medium F/W Activated Sludge
A.S. #3 - low F/M ?Vsti\ated Sludge
A.L. - Aerated Lagoon
RBC. - Rotating Biological Contactor
-------
the sulfide system. In reviewing the basic solubility products for
these systems, the sulfide system removes the most inorganics, with the
exception of arsenic, because of the low solubility of sulfide
compounds. This increased removal capability is offset by the
difficulty in handling the chemicals and the fact that sulfide sludges
are susceptible to oxidation to sulfate when exposed to air, resulting
in resolubilization of the metals. The carbonate system is a method
which relies on the use of soda ash and pH adjustment between 8.2 and
8.5. The carbonate system, although workable in theory, is difficult to
control. The hydroxide system is the most widely used inorganics/metals
removal system. The system responds directly to pH adjustment, and
usually uses either lime (CaOH) or sodium hydroxide (NaOH) as the
chemical to adjust the pH upwards. Sodium hydroxide has the advantage
of ease in chemical handling and low volume of sludge. However, the
hydroxide sludge is often gelatinous and difficult to dewater.
Description of Process
Chemical precipitation can be accomplished by either batch or
continuous flow operations. If the flow is less than 30,000 gpd, a
batch treatment system would be the most economical. In the batch
system, two tanks are provided, each with a capacity of one day's flow.
One tank undergoes treatment while the other tank is being filled.
When the daily flow exceeds 30,000 gpd, batch treatment is usually
not feasible because of the large tankage required. Continuous
treatment may require a tank for acidification and reduction, then a
mixing tank for chemical addition, and a settling tank.
-163-
-------
Design Parameters and Procedures
The important design factors that must be determined for a
particular water during treatability studies include;
(1) Best chemical addition system
(2) Optimum chemical dose
(3) Optimum pH conditions
(4) Rapid mix requirements
(5) Flocculation requirements
(6) Sludge production
(7) Sludge flocculation, settling, and dewatering characteristics
Laboratory-scale test procedures consisting of jar test studies have
been used for years, and the test methodology developed is such that
full-scale designs can be developed from these studies with a high
degree of confidence. Some general design considerations include:
(1) The retention time in the reduction tank should be at least
four times the theoretical time for complete reduction.
(2) Twenty minutes will usually be adequate for flocculation.
(3) Final settling should not be designed for an overflow rate in
excess of 500 gal/day/ft^.
Applications to Ground Water
Chemical precipitation has been successfully used for removing
heavy metals from various waters. Kincannon and Stover (1983) reported
the results of treating a contaminated ground water by various
processes. The results of chemical precipitation to remove metals is
given in Table V.10. They found that all the metals found in that
particular ground water were removed to acceptable levels.
-164-
-------
Table V. 10: Removal Chemical Precipitation Data for Metals
Concentrations in Groundwater, mg/Jl
Compound
Raw Water
Lime-
-Treated Water
pH 9.1
pH 9.9
pH 11.3
Arsenic
0.12
0.03
0.03
0.03
Barium
0.24
0.17
0.15
0.19
Cadmium
0.003
<0.001
<0.001
<0.001
Chromium (total)
0.09
0.006
0.006
0.006
Lead
0.03
0.006
0.006
0.006
Mercury
<0.001
<0.001
<0.001
<0.001
Selenium
<0.001
<0.001
<0.001
<0.001
Si 1ver
<0.001
<0.001
<0.001
<0.001
Copper
0.10
<0.001
<0.001
<0.001
Iron
352
0.07
0.07
1.05
Manganese
90
0
0
0
Nickel
1.95
0.05
0.30
0.45
Zinc
0.69
0.36
0.09
0.61
-165-
-------
Brantner and Cichon (1981) compared the three precipitation
processes. The results of this work is shown in Table V.ll. Their work
also shows that chemical precipitation is effective in removing metals.
However, all methods are not equally as effective. In addition, they
found that the carbonate system was the simplest process to operate, the
hydroxide process the most reliable, and the sulfide precipitation
process was the most complex to operate.
Costs
There is very little information in the literature regarding costs
of chemical precipitation. However, Krause and Stover (1982) have
presented an excellent cost comparison for removing barium from water
supplies. Their cost evaluations for ion exchange, lime softening, and
chemical precipitation - direct filtration are shown in Tables V. 12
through V.14. All costs are April 1980 dollars. Table V.15 shows a
cost comparison of the three processes.
OTHER TREATMENT TECHNIQUES FOR INORGANICS
Chemical precipitation has traditionally been a popular technique
for the removal of heavy metals and other inorganics from wastewater
streams. However, a wide variety of other techniques also exist.
Listed in Table V.16 are some available technologies for the removal of
specific inorganic contaminants. The technologies in Table V.16, which
also include chemical precipitation, were compiled from Patterson (1978)
and represent viable techniques. Patterson (1978) treats each inorganic
in more detail by addressing such topics as: sources of the given
inorganic; its major industrial uses; proven treatment technologies;
-166-
-------
Table V.ll: Comparison of Precipitation Treatment Processes
Influent Clarifier Effluent Filtered effluent
(PP"0 (pp™) (ppm)
Parameter MeanRange MeanRange Mean Range
HYDROXIDE PRECIPITATION DATA SUMMARY
Suspended Solids
42
20-63
22
9-33
9
4-14
pH
7.5
7.0-8.1
9.9
9.7-10.2
9.8
9.5-10.4
Total Cadmium
1.66
0.13-4.30
0.05
0.03-0.10
0.04
0.02-0.06
Total Chromium
1.11
0.07-2.90
1.04
0.07-2.80
0.97
0.06-2.90
Total Copper
0.29
0.12-1.50
0.03
0.02-0.03
0.03
0.02-03
Total Lead
1.7
0.8-2.6
0.2
0.1-0.3
0.2
0.1-0.3
Total Zinc
31
6-91
0.40
0.23-0.75
0.28
0.10-0.66
CARBONATE PRECIPITATION DATA SUMWRY
Suspended Solids
43
16-75
27
14-52
6
2-10
pH
7.1
6.7-7.7
8.3
8.1-8.4
8.1
7.8-8.5
Total Cadmium
1.37
0.26-2.90
0.14
0.02-0.27
0.04
0.02-0.06
Total Chromium
0.67
0.23-1.80
0.62
0.17-1.8
0.60
0.14-2. 00
Total Copper
0.18
0.06-0.27
0.04
0.03-0.06
<0.03
<0.02-0.04
Total Lead
1.4
0.7-2.1
0.2
0.2-0.4
-------
Table V.12: Ion-Exchange System - Cost Estimate Summary
Item Cost
Capital cost (dollars)
Aerator $ 25,000
Wet well 9,500
Transfer pumps 12,700
Ion exchangers 137,500
Caustic soda system 5,600
Brine pit 16,300
Building addition 52,300
Outside piping and sitework 25,900
Electrical and instrumentation 34,200
General and special conditions 22,300
Construction contingencies 51,200
Total construction cost $392,500
Engineering and estimating
contingencies 58,900
Financial cost 13,500
Bond interest 404,600
Total capital cost $869,500
0 & M costs (dollars/year)
Labor 14,000
Chemicals 23,200
Power 2,400
Waste disposal 1,200
Total annual 0 & M cost $40,300
-168-
-------
Table V.13: Lime-Softening System-Cost Estimate Summary
Item Cost
Capital cost (dollars)
Aerator $ 25,000
Solids contact reactor 113,600
Gravity filter 222,900
Recarbonation system 51,900
Transfer pumps 12,700
Alum system 10,100
Lime system 40,000
Lime sludge dewatering 93,800
Building addition 209,800
Outside piping and sitework 78,000
Electrical and instrumentation 102,900
General and special conditions 67,200
Construction contingencies 154,200
Total construction cost $1,182,100
Engineering and estimating
contingencies 177,300
Financial cost 40,800
Bond interest 1,218,400
Total capital cost $2,618,600
0 & M cost (dollar/year)
Labor 28,000
Chemicals 27,500
Power 4,600
Waste disposal 19,100
Total annual 0 & M cost $ 79,200
-169-
-------
Table V.1.4: Chemical Precipitation-Direct Filtration System-Cost
Estimate Summary
Item Cost
Capital cost (dollars)
Aerator $ 25,000
Rapid mix tank 7,900
Flocculation basin 36,100
Gravity filter 286,600
Recarbonation system 62,200
Transfer pumps 12,700
Potassium hydroxide system 25,600
Gypsum system 22,400
Polymer system 5,000
Building addition 239,000
Outside piping and sitework 70,500
Electrical and instrumentation 93,000
General and special conditions 60,800
Construction contingencies 139,300
Total construction cost $1,068,100
Engineering and estimating
contingencies 160,200
Financial cost 36,800
Bond interest 1,100,900
Total capital cost $2,366,000
0 & M cost (dollars/year)
Labor 28,000
Chemicals 116,600
Power 4,900
Waste disposal 6,400
Total annual 0 & M cost $ 155,900
-170-
-------
Table V.15: Monetary Comparison of Alternative Treatment Systems
Item
Alternative Treatment Systems
Ion
Exchange
System
Lime
Softening
System
Chemi cal
Precipitation-
Direct
Filtration
System
Capital cost
(dollars)
Capital annual
equivalent
cost
(dollars/year)
Annual 0 & M
cost
(dollars/year)
Total annual
equivalent
cost
(dollars/year)
$869,500
116,400
40,800
$157,200
$2,618,600
350,000
79,200
$ 429,800
$2,366,000
316,800
155,900
$ 472,700
-171-
-------
Table V.16: Treatment Alternatives for Inorganics
Inorganic Treatment Method
Arsenic Charcoal Filtration
Lime Softening
Precipitation with lime + iron
Precipitation with alum
Precipitation with ferric sulfate
Precipitation with ferric chloride
Precipitation with ferric hydroxide
Precipitation with sulfide
Ferric Sulfide Filter Bed
Iron or Lime Coagulation + settling + dual media
filtration + carbon adsorption
Barium Iron or Lime Coagulation + settling + dual media
filtration + carbon
adsorption
Precipitation as sulfate
Precipitation as carbonate
Precipitation as hydroxide
Ion Exchange
Boron Evaporation
Reverse Osmosis
Ion Exchange
Cadmium Precipitation as hydroxide
Precipitation as hydroxide + filtration
Precipitation as sulfide
Coprecipitation with ferrous hydroxide
Reverse Osmosis
Freeze Concentration
Chloride Ion Exchange
Electrodialysis
Reverse Osmosis
Other (holding basins, evaporative ponds, deep well
injection)
-172-
-------
Table V.16: (continued)
Inorganic
Treatment Method
Chromium
(hexavalent)
Ion Exchange
Freeze Concentration
Activated Carbon
Cementation
Reduction (Reduce Cr+6 to Cr^ and
precipitation of hydroxide)
Reduction with sulfur dioxide
Reduction with bisulfite
Reduction with bisulfite + hydrazine
Reduction with metabisulfite
Reduction with ferrous sulfate
Chromium
(trivalent)
Precipitation (see above)
Ion Exchange
Copper
Precipitation with lime
Ion Exchange
Evaporative Recovery
Electrolytic Recovery
Cementation
Reverse Osmosis
Cyanide
Alkaline Chlorination
Electrolysis
Ozonation
Evaporation
Fluoride
Precipitation by lime addition
Precipitation by magnesium addition
Precipitation by alum addition
Adsorption on hydroxlapatite beds
Adsorption on alumina contact beds
-173-
-------
Table V.16: (continued)
Inorganic
Treatment Method
Iron
Oxidation-Precipitation
by
aeration, sand filtration
Oxidation-Precipitation
by
aeration, lime, sand fil-
tration
Oxidation-Precipitation
by
aeration, coke bed fil-
tration, sedimentation,
sand filtration
Oxidation-Precipitation
by
lime, aeration, diatomite
filtration
Oxidation-Precipitation
by
chlorination, alum-lime-
sodium silicate precipi-
tation, sand filtration
(deep well disposal)
Lead Ion Exchange
Precipitation by lime + sedimentation
Precipitation by caustic + sedimentation
Precipitation by ammonium hydroxide
Precipitation by dolmite + sedimentation
Precipitation by sodium carbonate + filtration
Precipitation by sodium phosphate + filtration
Precipitation by ferric sulfate + sedimentation
Precipitation by ferrous sulfate + sedimentation
Manganese Aeration
Ion Exchange
Catalysis
Oxidation-Precipitation by chlorine dioxide addition
Oxidation-Precipitation by manganese dioxide addition
Oxidation-Precipitation by potassium permanganate
addition
Mercury Precipitation by sodium sulfide addition
Precipitation by sodium hydrosulfide addition
Precipitation by magnesium sulfide addition
Precipitation by sulfide addition
Ion Exchange
Coagulation with alum
Coagulation with iron
Activated Carbon
-174-
-------
Table V.16: (continued)
Inorganic
Treatment
Method
Reduction
Reduction
Reduction
to Metallic
to Metallic
to Metallic
Form by
Form by
Form by
zinc
stannous chloride
sodium borohydride
Nickel Precipitation by lime
Precipitation by sulfide
Precipitation by alum
Ion Exchange
Reverse Osmosis
Evaporative Recovery
Selenium Coagulation with lime
Coagulation with ferric sulfate
Coagulation with alum
Activate Carbon + Cation Exchange + Anion Exchange
Silver Precipitation with ferric chloride
Ion Exchange
Reductive Exchange with zinc or iron
Electrolytic Recovery
Total Dissolved
Solids Reverse Osmosis
Electrodialysis
Distillation
Ion Exchange
Zinc Precipitation by lime addition
Precipitation by caustic addition
Ion Exchange
Evaporation
-175-
-------
proposed treatment technologies; removal efficiencies of various
technologies; and costs of various treatment technologies.
TREATMENT TRAINS
Due to the complex composition of most ground waters, no one unit
operation is capable of removing all of the contaminants present. It
may be necessary to combine several unit operations into one treatment
process to effectively remove the contaminants required. In order to
simplify, and make visible, the selection of the applicable treatment
trains, Table V.17 is presented showing a number of unit operations and
the waste types for which they are effective. Results presented by
Kincannon and Stover (1983) illustrate the levels of treatment achieved
by a treatment train, with these results given in Table V.18.
SELECTED REFERENCES
Brantner, Karl A. and Edward J. Cichon, "Heavy Metals Removal:
Comparison of Alternative Precipitation Processes", Industrial Waste-
Proceedings of the Thirteenth Mid-Atlantic Conference, 1981.
Dysken, John E., et al., "The Use of Aeration to Remove Volatile
Organics from Ground Water", Presented at 1982 Annual Conference of the
American Water Works Association held May 16-20, 1982, Miami Beach,
Florida.
Kaufmann, Henry G., "Granular Carbon Treatment of Contaminated
Supplies", Proceedings of the Second National Symposium on Aquifer
Restoration and Ground Water Monitoring, May 26-28, 1982, The Fawcett
Center, Columbus, Ohio.
Kincannon, D.F., and E.L. Stover, "Biological Treatability of Specific
Organic Compounds Found in Chemical Industry Wastewaters", Presented at
the 36th Purdue Industrial Waste Conference, West Lafayette, Indiana,
1981a.
Kincannon, D.F. and E.L. Stover, "Fate of Organic Compounds During
Biological Treatment", Presented at the 1981 National Conference on
Environmental Engineering, ASCE Environmental Engineering Division,
Atlanta, Georgia, 1981b.
-176-
-------
Table V.17: Summary of Suitability of Treatment Processes
Air
Stripping
Steam
Stripping
Carbon
Adsorption
Volatile
Organics
Suitable for
most cases
Effective
Concentrated
Technique
Inadequate
Remova1
Biological
pH Adjustment
Precipitation
Effective
Removal
Technique
Not
Applicable
Electrodialysis
Not
Applicable
Ion Exchange
Not
Applicable
Non-Volatile
Organics
Not
Suitable
Inorganics
Not
Suitable
Not
Suitable
Not
Suitable
Effective
Removal
Technique
Effective
Removal
Technique
Not
Applicable
Not
Applicable
Not
Suitable
Not
Suitable
Metals Toxic
Effective
Removal
Technology
Inefficient
Operation/
Inadequate
Removal
Not
Applicable
Inappropriate
Technology -
Difficult
Operation
-177-
-------
Table V.18: Ground Water Treatment Results From a Selected Treatment
Train
Parameter's Value After Treatment
Parameter
Raw Water
Activated Carbon Adsorption
Metals Steam Biological
Treatment Stripping 6000 mg/t 24,000 mg,'t Treatment
pH-units 6.0
Specific conductance-ug/fc 2800
TOC-mg/H 4000
COD-mg/fc 11750
BODj-mg/I 7025
Phenols-mg/I 26
SS-mg/l 360
VSS-mg/i 142
TDS-mg/i 3050
TVS-mg/i 1485
Arsenic-mg/lt. 0.12
Barium-mg/e. 0.25
Cadmium-mg/l 0.003
Chromiuin-mg/j, 0.09
Lead-mg/H 0.03
Mercury-mg/l <0.001
Selenium - mg/£ <0.001
Silver-mg/f, 0.001
Copper-mg/e 0.10
Iron-mg/e 352
Manganese-mg/j, 90
Nickel-mg/i 1.95
Zlnc-mg/i 0.69
Tetrahydrofuran- jjg/i 22000
1,1,1-Trichloroethane- yg/i 150000
Benzene- yg/e 68750
10.0 to 6.5
3685
0.03
0.15
<0.001
0.006
0.006
<0.001
<0.001
<0.001
<0.001
0.07
0
1.58
0.09
1700
3600
6,
2450
1575
2400
0.85
0.02
0.09
<0.001
0.006
0.006
<0.001
<0.001
<0.001
<0.001
0.07
0
0.89
0.09
6.6
2450
1130
1950
0.12
0.02
0.03
<0.001
0.006
0.006
<0.001
<0.001
<0.001
<0.001
0.07
0
0.28
0.09
7.3
34
125
5.5
0.12
1897**
(continued)
-178-
-------
Table V.18: (continued)
Parameter's Value
After Treatment
Activated Carbon Adsorption
Metals
Steam
Biological
Parameter
Raw Water
T reatment
Stri pping
6000 mg/fc
24,000 mg.'i Treatment
Trlchloroethylene- ug/J
338000
*
Methyl Isobutyl ketone-ug/i
76400
*
1098
Xylenes- ug/J.
*
*
+
Toluene-yg/J.
92000
*
25
Ethyl benzene- ug/l
23500
~
*
1,4-Dlchlorobenzene- ug/l
35
35
17
*
* *
1,2-D1chlorobenzene-ug/t
5
5
*
*
* *
Naphthalene- ug/1
i 1
£l
i1
Si
* *
2-Chl orophenol -ug/1,
540
540
40
il
<1 *
2-N1trophenol-gg/l
15
*
6
*
* ~
Phenol- ug/i
370
*
20
~
* *
2,4-Olmethy1 phenol-ug/I
20
*
+
*
* *
o-Cresol-ug/l
80
80
25
*
* *
m-Cresol-ug/J.
220
220
*
Si
* *
Benzoic acid-ug/l
1230
1230
*
~
* *
Pentachl orophenol-vig/l
40
40
40
8
2 *
* No peaks on chromatograms
** Benzene and trfchloroethylene peaks combined Into one peak
-179-
-------
Kincannon, D.F. and E.L. Stover, "Stripping Characteristics of Priority
Pollutants During Biological Treatment", Presented at the 74th Annual
AIChE Meeting, New Orleans, Louisiana (November 1981), 1981c.
Kincannon, D.F. and E.L. Stover, Final Report EPA Cooperative Agreement
CR 806843-01-02, "Determination of Activated Sludge Biokinetic Constants
for Chemical and Plastic Industrial Wastewaters".
Kincannon, D.F., and E.L. Stover, "Contaminated Groundwater Treatability
- A Case Study", Journ. American Water Works Association, Vol. 75, June
1983.
Kincannon, D.F., E.L. Stover, and Y.P. Chung, "Biological Treatment of
Organic Compounds Found in Industrial Aqueous Effluents", Presented at
the ACS National Meeting, Atlanta, Georgia (March 1981).
Kincannon, D.F., et al., "Removal Mechanisms for Biodegradable and Non-
Biodegradable Toxic Priority Pollutants in Industrial Wastewaters",
Presented at the 54th Annual Water Pollution Control Federation
Conference, Detroit, Michigan (October 1981).
Kincannon, D.F., et al., "Variability Analysis During Biological
Treatability of Complex Industrial Wastewaters for Design", Presented at
the 37th Purdue Industrial Waste Conference, West Lafayette, Indiana,
1982a
Kincannon, D.F., et al., "Predicting Treatability of Multiple Organic
Priority Pollutant Wastewaters from Single Pollutant Treatability
Studies", Presented at the 37th Purdue Industrial Waste Conference, West
Lafayette, Indiana, 1982b.
Krause, Terry L. and Enos L. Stover, "Evaluating Water Treatment
Techniques for Barium Removal", Journal American Water Works
Association, September 1982.
Lamarre, Bruce L., Frederick J. McGarry, and Enos L. Stover, "Design,
Operation, and Results of a Pilot Plant for Removal of Contaminants from
Ground Water", Third National Symposium and Exposition on Aquifer
Restoration and Ground Water Monitoring, The Fawcett Center, Columbus,
Ohio, May 25-26, 1983.
McDougall, W. Joseph, Richard A. Fusco, and Robert P. O'Brien,
"Containment and Treatment of the Love Canal Landfill Leachate", Journal
WPCF, Vol. 52, No. 12, Dec. 1980.
McKinnon, Ronald J. and John E. Dyksen, "Aeration Plus Carbon Adsorption
Remove Organics from Rockaway Township (New Jersey) Ground Water
Supply", Presented at the American Society of Civil Engineers 1982
Annual Convention held October 25-27, 1982, New Orleans, Louisiana.
-180-
-------
O'Brien, Robert P. and J.L. Fisher, "There is an Answer to Ground-water
Contamination", Water Engineering and Management, Vol. 130, No. 5, May
1983.
Patterson, J.W., Wastewater Treatment Technology, 3rd ed., Ann Arbor
Science, 1978.
Pekin, Tarik and Alan Moore, "Air Stripping of Trace Volatile Organics
from Wastewater", Proceedings of the 37th Industrial Waste Conference,
Purdue University, May 11, 12 and 13, 1982, West Lafayette, Indiana.
Stover, Enos L., "Removal of Volatile Organics from Contaminated Ground
Water", Proceedings of the Second National Symposium on Aquifer
Restoration and Ground Water Monitoring, May 26-28, 1982, The Fawcett
Center, Columbus, Ohio.
Tabak, H.H., et al., "Biodegrad-ability Studies with Organic Priority
Pollutant Compounds", JWPCF, 53(10), 1503, 1981.
Thomas, H.C., "Chromatogrpahy: A Problem in Kinetics", Annals of the
New York Academy of Science, 49:161, 1948.
-181-
-------
SECTION VI
IN-SITU TECHNOLOGIES
This section addresses aquifer restoration measures involving
treatment in-place. Included is brief information on in-situ chemical
treatment along with more extensive information on the use of both
natural and enhanced microbiological processes. In-situ treatment can
be used singly or in combination with other aquifer restoration
strategies.
IN-SITU CHEMICAL TREATMENT
In-situ chemical treatment of contaminated ground water can be
considered only in cases where contaminants are known, and the levels
and extent of contaminated water in the aquifer are defined. In-situ
treatment generally involves the installation of a bank of injection
wells at the head of, or within the plume of contaminated ground water.
A treatment agent is then pumped into the aquifer. The treatment agent
must be specific for different classes of contamination. For example,
heavy metals may be made insoluble and rendered immobile with alkalies
or sulfides, cyanides can be destroyed with oxidizing agents, cations
may be precipitated with various anions or by aeration, and hexavalent
chromium could be made insoluble with reducing agents.
Description of Process
The process of in-situ treatment can be accomplished by many means.
Two examples will be described herein. The first is a process to
remove iron and manganese called the Vyredox Method (Hallberg and
-182-
-------
Martinelli, 1976). This method was developed in Finland and is now used
in several European countries. The aim of the Vyredox Method is to
achieve a high degree of oxidation in the strata around a well. In
other words, the Eh and pH are kept so high that the iron and manganese
are precipitated and retained in the strata. The Vyredox concept is
illustrated in Figure VI.1. The water that enters the well is thus free
of iron and manganese. Iron is precipitated first in the zone furthest
from the withdrawal (supply) well. The number of living bacteria
increases as the withdrawal well is approached, and so does the number
of dead bacteria. The organic matter contained in the dead cells
becomes a carbon source for the bacteria that preferentially oxidize
manganese. This process takes place nearer the well, where the Eh is
higher. Thus, the Vyredox Method provides favorable conditions for
removing iron first, and then manganese.
Figure VI.2 illustrates a Vyredox plant for treating contaminated
ground water. It consists of one or more wells for supplying water.
Each well has a pipe fitted with a screen. A number of injection wells
(aeration wells) are situated in a ring around each supply well. The
number depends on the hydrogeological and geochemical conditions.
Conduits connect each well to a nearby building. The water forced into
the injection wells must first be degassed and enriched with oxygen.
This is done in a special aerator, the oxygenator.
A procedure of containment of a contaminant by polymerizing the
contaminant in the ground has been reported by Williams (1982). The
procedure has been used to contain approximately 4,200 gallons of
aerylate monomer which had leaked from a corroded underground pipeline.
-183-
-------
CD
¦C*
too HIV
60RFACE
)> LJ/f -.r/fjJu.'-l'it*'
smemei
•oo mv
200 mv
_ PURIFIED WATER
•~to consumer
DEGASSED AND
" AERATED WATER
WATER TABLE
. Mn PRECIPITATION ZONE
•F« PRECIPITATION ZONE
Figure VI.1: Vyredox Concept
-------
LAYOUT OF VYREDOX PLANT
0-0 SUPPLY WELLS AND AERATION WELLS
0 AERATOR. THE OXYGENATOR
0 DEGASSING TANK
PUMP FOR AERATED WATER
Figure VI.2: Vyredox Plant with Two Supply Wells Complete with Aeration
Wells and Oxygenator Building
-------
The acrylate monomer is a colorless mobile liquid less heavy than and
only moderately soluble in water. Vapors from the monomer are odorous
at 5 ppm, and in high concentrations can result in skin irritation, as
well as explosive potential. Under favorable conditions, the monomer
will polymerize to a soft rubbery texture.
Because imposed limitations, both by the hydrogeologic conditions
and by the practical characteristics of the site, there were few viable
options for recovering or containing the monomer. Therefore, a system
of four exfiltration galleries was designed and installed. Two-inch
diameter perforated PVC casings, in 40-foot lengths, were buried in
narrow trenches two feet below ground surface across the shallow
contaminated zone. A riser pipe and manifold header connected each
gallery to the solution tanks. Four thousand gallons of catalyst and
activator solution per treatment were used. Two separate treatments
four days apart were used. It is estimated overall, that 85 to 90
percent of the liquid monomer contaminant was effectively converted to a
solidified polymer.
Advantages and Disadvantages
In-situ chemical treatment is viable only under particular
hydrogeological and geocheraical conditions. Other aquifer restoration
measures, such as withdrawal and treatment, may be more appropriate for
consideration in meeting a given need.
IN-SITU BIOLOGICAL STABILIZATION
Biological treatment can be used to clean up contaminated aquifers.
Microbes can degrade most organic compounds; however, many man-made
-186-
-------
compounds are relatively refractive (Kobayoshi and Rittman, 1982).
Compounds which possess amine, methoxy, and sulfonate groups, ether
linkages, halogens, branched carbon chains, and substitutions at the
meta position of benzene rings are generally persistent. Environmental
factors such as dissolved oxygen level, pH, .temperature, oxidation-
reduction potential, availability of nutrients, salinity, and the
concentration of the compounds often control the biodegradation of the
compounds. The number and type of organisms present also play an
important role.
Microbial activity is likely to exist in most subsurface regions
where ground water is important (McNabb and Dunlap, 1975). Bacterial
levels typically around 10^ organisms/g dry soil have been found for
several shallow water table aquifers which have been investigated
(Wilson, et al., 1983b). The potential for biodegradation of a variety
of compounds in the subsurface has been demonstrated. The degradation
of petroleum products has been extensively studied. Litchfield and
Clark (1973) analyzed ground water samples from aquifers throughout the
United States that were contaminated with hydrocarbons, and found
hydrocarbon-utilizing bacteria in all the samples at levels up to 10^
per ml. After a gasoline spill in Southern California, McKee, Laverty
and Hertel (1972) found 50,000 gasoline-utilizing bacteria/ml or higher
in samples from wells which had traces of free gasoline while a non-
contaminated well had only 200 organisras/ml. Jamison, Raymond, and
Hudson, Jr. (1975) determined that the microbial population in an
aquifer contaminated by a gasoline pipeline break was limited by
inadequate levels of nitrogen, phosphorus, and oxygen. Experiments with
-187-
-------
microbial isolates from the ground water showed that the different
isolates were able to utilize certain components of the gasoline and
suggested that co-oxidation played an important role in the
biodegradation of the other compounds (Jamison, Raymond, and
Hudson, Jr., 1976). Complete conversion of hydrocarbons leads to carbon
dioxide, water, and new cell biomass; various intermediates are also
formed (Vanloocke, et al., 1975). In their column studies, Kappler and
Wuhrman (1978a, b) noted a lag period of 1 to 5 days (depending on
temperature) before the ground water microbial flora was able to
measurably degrade dissolved hydrocarbons. Addition of nitrogen
increased degradation. Stimulation of the microbial population by the
addition of nitrogen, phosphorus, and dissolved oxygen has been shown to
be effective in restoring hydrocarbon-contaminated aquifers (American
Petroleum Institute, 1980). Yang and Bye (1979) suggested that the
levels of trace elements are usually sufficient to support microbial
growth.
Many organic compounds which have contaminated ground water are
subject to microbial action. Wilson, et al. (1981, 1983 a and b)
studied the degradation of several halogenated volatile organics
commonly found as ground water contaminants and determined that some of
the compounds were degraded, although often at very slow rates.
Differences were noted between the activities of the microbial
populations from two site9 in degrading the synthetic organics. The
degradation of naphthalene has been studied in several systems.
Naphthalene has a number of uses and is considered to be hazardous
(Windholz, 1976). It should be a representative model for other
-188-
-------
compounds. Slavnia (1965) and Naumova (1960) found naphthalene-
utilizing organisms in the ground waters of the U.S.S.R. near oil and
gas beds. Ogawa, Junk, and Svec (1981) and Lee and Ward (1984) reported
that naphthalene was degraded rapidly in aerobic ground waters
contaminated by polynuclear aromatic hydrocarbons. However, Erlich, et
al. (1982) concluded that there was no evidence of anaerobic degradation
of naphthalene in ground water samples from an area contaminated by wood
creosoting products, although it was disappearing at a faster rate in
the aquifer than if only dilution were occurring. Naphthalene was
slightly sorbed onto sediments. It was biodegraded in an aquifer
recharged with reclaimed water from wastewater treatment after an
initial lag (Roberts, et al., 1980). The importance of acclimation to
the contaminant was demonstrated by these results.
Many ground water systems are deficient in oxygen and consequently
anaerobic degradation is important. Erlich, et al. (1982) and Rees and
King (1980) reported evidence of anaerobic degradation of phenolics
under aquifer conditions. Bouwer, Rittman, and McCarty (1981) and
Bouwer and McCarty (1983a, b) found degradation of several halogenated
aliphatic organic compounds under raethanogenic and denitrifying
anaerobic conditions that were not degraded under aerobic conditions.
Wood, et al. (1980) reported that anaerobic metabolism of some of these
compounds led to the formation of similar contaminants such as the
production of vinyl chloride, vinylidene chloride, cis and trans 1,2-
dichloroethene from trichloroethylene and/or tetrachloroethylene. None
of the aromatic compounds tested by Bouwer and McCarty (1983b) were
significantly utilized under anaerobic conditions. DiTommaso and Elkan
-189-
-------
(1973) and Leenheer, Malcolm, and White (1976) disclosed that methane
was produced by the degradation of a high organic content wastewater
after deep well injection. This provides evidence for anaerobic
degradation in the deep subsurface.
The potential for microbial degradation of contaminants has been
utilized to restore polluted environments. Atlas (1977) outlined the
following approaches that can be used to enhance the degradation of
petroleum: the microbial population can be altered by seeding; the
environment can be modified to enhance microbial activity; or the
petroleum can be modified to make it more susceptible to biodegradation.
Seed microorganisms would probably have to be a mixture of
microorganisms that collectively can degrade all the components of the
contaminant. Different mixtures of seed organisms will have to be used
for different environments and contaminant problems. The environment
can be modified to enhance microbial activity by the addition of
dissolved oxygen, nitrogen, phosphorus, sulfur, iron, magnesium,
calcium, sodium, and water which are not present in sufficient
quantities. Modification of the contaminant could involve increasing
the surface area available for biodegradation by dispersing the
contaminant or by emulsification. However, the emulsifying agent or
dispersant may prove toxic to the microbial population (Atlas, 1977).
Biological restoration of contaminated aquifers has been
accomplished using several techniques. Suntech pioneered the use of
"Bioreclamation" to restore petroleum contaminated aquifers (Raymond,
1974, 1978; Raymond, Hudson, and Jamison, 1976; and Jamison, Raymond and
Hudson, 1975). This process enhances the natural population by
-190-
-------
providing dissolved oxygen and nutrients to accelerate biodegradation.
This process has been modified for use at hazardous waste sites (J.R.B.
Associates, 1982; and Jhaveri and Mazzacca, 1982). Another technique
which has been employed is the addition of acclimated microbes and
nutrients to stimulate degradation (Quince and Gardner, 1982). Yet
another option is withdrawal and biological treatment by conventional
wastewater treatment processes, including activated sludge, lagoons
(facultative, anaerobic, or aerobic), rotating biological disks, and
trickling filters (J.R.B. Associates, 1982). Combinations of these
biological techniques with physical-chemical treatment have also been
used effectively (Shuckrow and Pajak, 1981).
Enhancement of the Indigenous Population
Stimulation of the native microbial population by the addition of
nutrients has long been suggested as a means of increasing the
degradation of hydrocarbons in aquatic and soil environments. In marine
environments, hydrocarbon degraders have been found in higher
concentrations in regions chronically polluted with petroleum (Atlas,
1977). Nitrogen and phosphorus levels are often inadequate and must be
supplemented. Use of an oleophilic fertilizer that concentrates the
nutrients at the water surface has been shown to increase the
degradation of oil slicks (Atlas and Bartha, 1973b). Levels of iron may
also be deficient (Dibble and Bartha, 1976). The temperature of the sea
water is a controlling factor in microbial degradation (Mulkin-Phillips
and Stewart, 1974b), especially in colder areas like the Arctic (Atlas
and Schofield, 1975). Seasonal and climatic variability, the types of
-191-
-------
hydrocarbons involved, and the nature of the site are also important
(Colwell and Walker, 1977). Other factors to be considered include the
dispersion of the hydrocarbon, turbulence, the concentration of the
hydrocarbon, and microbial predation (Zobell, 1973). The effectiveness
of nutrient supplementation has also been demonstrated in freshwater
systems; nutrient addition increased the degradation of gasoline in a
lake from 90 (control) to 96 percent (nutrient supplemented) after five
weeks (Horowitz and Atlas, 1977).
Addition of nutrients to hydrocarbon-contaminated surface soil has
proven useful in increasing microbial degradation. The total numbers of
microbes increase greatly after a petroleum spill; Odu (1972) noted an
increase from 10^ organisms/g to 10® organisms/g after an oil well
blowout. Application of fertilizer stimulated greater microbial growth
and utilization of some components of oil (Jobson, et al., 1974). Other
components of oil are not attacked by the microbes and remain in the
soil. Saturated fractions are highly degraded while asphaltenes and
aroraatics are often resistant to microbial attack (Jobson, Cook, and
Westlake, 1972). Lehtomaki and Niemela (1975) found that the addition
of yeast cells also served as a nutrient source. The soil must be
tilled to distribute the hydrocarbons through the soil and provide
oxygen (Raymond, Hudson, and Jamison, 1976). Wenstel, et al. (1981)
attempted to enhance the biodegradation of monochlorobenzene and Ethion
by the addition of nutrients (Difco nutrient broth) and aeration of the
soil. No evidence for significant microbial degradation of
monochlorobenzene or Ethion was observed, thus showing that stimulation
of the native microbial population will be ineffective against compounds
-192-
-------
which the microbes cannot readily attack, such as the water insoluble
Ethion or highly volatile raonochlorobenzene.
Enhancement of the natural microflora has been used in a number of
cases to degrade hydrocarbons in surface soils. Dibble and Bartha
(1979c) documented the cleanup of a large kerosene spill (1.9 million
liters) in New Jersey. Much of the kerosene was recovered by physical
means and by removing 200 nP of contaminated soil. Following
stimulation of microbial degradation by liming, fertilization, and
tillage, phytotoxicity was reduced. Microbial degradation was slowed by
the lower winter temperatures. Guidin and Syratt (1975) suggested that
a black covering be placed over the soil to increase the soil
temperature during the winter to overcome this problem. Landfarming has
been tested as a method of disposing of waste oil and oil sludges.
Landfarming relies on the soil organisms to degrade the waste applied to
the soil. Brown, et al. (1981) investigated the factors which
influenced biodegradation rates of oil sludges at a land farm site.
Only under excessively wet or dry conditions did the moisture content
become a dominant factor. The nitrogen level was critical since no
additional increase in the biodegradation rate occurred when potassium
and phosphorus were also added. The optimal carbon to nitrogen ratio
ranged from 150:1 to 10:1 depending upon the nature of the sludge.
Dibble and Bartha (1979a) conducted similar experiments on landfarming
waste oil and determined that carbon to nitrogen and carbon to
phosphorus ratios of 60:1 and 800:1 were optimal under the conditions
they used. Addition of micronutrients and organic supplements in the
form of yeast extract and trace elements or domestic sewage did not
-193-
-------
prove beneficial. Other experiments showed that urea formaldehyde was
the most satisfactory nitrogen source tested since it effectively
stimulated biodegradation and did not leach nitrogen which could
contaminate the ground water (Dibble and Bartha, 1979b). Weldon (1979)
found no adverse impact on ground water quality from landfarming
refinery oil sludges, although heavy metals were concentrated in the
soil and could be leached away. Another problem was runoff water from
the site could contain high amounts of oil and fertilizer (Kincannon,
1972).
These examples illustrate the effectiveness of nutrient
supplementation in aquatic and soil systems. Many of the same
procedures can be used to restore contaminated aquifers. Among the
first to suggest such actions were Williams and Wilder (1971) and McKee,
Laverty, and Hertel (1972) who investigated a large gasoline spill
(estimated to be 250,000 gallons) in Glendale, California, that had
contaminated an irrigation well. An investigation into the microbial
degradation of the gasoline showed that several bacterial species of the
genera Pseudomonas and Arthrobacter could utilize the gasoline when
supplied with trace nutrients and adequate dissolved oxygen (McKee,
Laverty, and Hertel, 1972). Bacterial degradation of trapped gasoline
was more rapid in the zone of aeration above the water table than in the
water-saturated zone. The levels of gasoline-utilizing bacteria in
contaminated wells were 50,000/ml or higher and gradually fell to the
background level of 200/ml as the gasoline disappeared. The authors
thought that biodegradation could eventually restore the aquifer to
service. At roughly the same time, Davis, et al. (1972) recommended
-194-
-------
that the addition of nutrients to stimulate degradation of hydrocarbons
in ground water might be a feasible solution to ground water pollution
problems.
Raymond, Jamison and their coworkers at Suntech were probably the
first to put these suggestions into operation. Raymond (1974) received
a patent entitled "Reclamation of Hydrocarbon Contaminated Ground
Waters" on a process to eliminate hydrocarbon contaminants in aquifers
by providing nutrients and oxygen to the hydrocarbon-utilizing
microorganisms present in the ground water. The nutrients and oxygen
are to be introduced through wells and circulated through the
contaminated zone by pumping one or more producing wells.
Supplementation of the nutrient and oxygen levels and recirculation of
the water would allow the normal microbial flora to decompose the
hydrocarbons more rapidly than under natural conditions. The nutrient
solution would contain sources of nitrogen and phosphorus and other
inorganic salts, if necessary, at concentrations of 0.005 to 0.02
percent by weight for each of the nutrients. Oxygen would be supplied
by sparging air into the ground water. The process was expected to be
largely complete within six months. Return to the normal levels of
bacteria would occur after nutrient addition was stopped, and since no
organisms were added, the normal flora would be maintained.
Suntech's process has been largely used to clean up gasoline
contaminated aquifers. The following steps are involved. Physical
methods are employed to recover as much of the gasoline as possible
(Suntech, 1977). If practical, it may be wise to continue to pump
contaminated wells to contain the gasoline (Raymond, Jamison, and
-195-
-------
Hudson, 1976). Then investigations of the hydrogeology and the extent
of contamination should be made (Suntech, 1977). A laboratory study
must then be conducted to determine if the native microbial population
can degrade the components of the spill and what kinds and amounts of
inorganic salts are required to stimulate degradation (Raymond, 1978).
The laboratory study determines what combination of nutrients gives the
maximal cell growth on gasoline in 96 hr at the ambient temperature of
the ground water. Considerable variation in the nutrient requirements
has been found between aquifers. One system required only the addition
of nitrogen and phosphorus sources (Raymond, Jamison, and Hudson, 1976),
while the growth of microbes in another aquifer was stimulated best by
the addition of ammonium sulfate, mono- and disodium phosphate,
magnesium sulfate, sodium carbonate, calcium chloride, manganese
sulfate, and ferrous sulfate (Raymond, et al., 1978). The form of the
nutrient which must be added also varies; ammonium sulfate gave much
greater growth than ammonium nitrate in one aquifer system. Chemical
analyses of the ground water provides little information as to the
nutrient requirements for the system. After the microbial investigation
has established the optimal growth conditions, the system for injecting
the nutrients and oxygen and producing water to circulate them in the
formation must be designed and built (Raymond, 1978). This work should
be under the direction of a competent ground water geologist since
controlling the ground water flow is critical to the success of the
operation. Placement of the injection and production wells such that
ground water flow goes through the contaminated zone is required.
Recycling the contaminated water from the producing wells is suggested,
-196-
-------
since it eliminates the problem of waste disposal and allows the
recirculation of unused nutrients. The screens for the wells should be
large enough to permit fluctuations of the ground water table due to
weather conditions or operation of the system.
Once actual operations are underway, nutrient addition can be by
batch or continuous feed (Raymond, 1978). Batch addition gives
satisfactory results and is more economical. For one system, the
nutrients were prepared as a 30 percent concentrate in a tank truck and
injected into the aquifer (Raymond, Jamison, and Hudson, 1976). This
may have resulted in osmotic shock to the microorganisms which come into
contact with the concentrate before dilution. Large amounts of
nutrients may be required; at one site, 16.65 tons of chemicals were
added (Minugh, et al., 1983), while a total of 87 tons of food grade
quality chemicals were purchased to clean up another site (Raymond,
Jamison, and Hudson, 1976). Oxygen can be supplied to the aquifer by
sparging air into wells with Carborundum diffusers powered by paint
sprayer-type compressors (Raymond, Jamison, and Hudson, 1976 ) or by
R
diffusers made from a short piece of DuPont Viaflo tubing for smaller
wells (Raymond, et al., 1978). The larger diffusers can provide up to
10 cubic feet of air per minute (SCFM) while the smaller tubing
diffusers can provide only 1 SCFM. Another approach was the use of
diffusers spaced along air lines buried in the injection trench (Minugh,
et al., 1983). The size of the compressor and the number of diffusers
is determined by the extent of contamination and the period allowed for
treatment (Raymond, 1978). The supply of dissolved oxygen may be the
limiting factor in the biostimulation process, especially in low
-197-
-------
permeability aquifers (Raymond, et al., 1978). As the levels of
contamination decrease, the biochemical oxygen demand lessens and the DO
level in the water rises. The system must be monitored to insure that
the levels of nutrients are at their optimal concentrations and are
being evenly distributed, and that the discharge water meets state or
Federal requirements (Raymond, 1978). Adequate supplies of nitrogen and
phosphate can be readily maintained once breakthrough has occurred
(Raymond, et al., 1978).
The bacterial population's response to nutrient addition should
also be monitored. The highest microbial population is likely to be in
the area of greatest contamination (up to 10? organisms/ml have been
reported); a map of the bacterial counts following a gasoline spill in
Ambler, Pennsylvania, resembled the contours of gasoline contamination
(Raymond, Jamison, and Hudson, 1975). During the course of the
biostimulation program at this site, 32 cultures thought to be the
predominant gasoline-utilizing bacteria were isolated from the ground
water. These cultures included 10 cultures assigned to Nocardia, two
Micrococcus, four cultures of Actinetobacter, eight Pseudomonas,
Flavobacterium devorars, and seven unidentified cultures. Nocardia
cultures were probably responsible for the major paraffinic hydrocarbon
degradation, and the Pseudomonas cultures were responsible for much of
the aromatic degradation (Jamison, Raymond, and Hudson, 1976). Co-
oxidation seemed to have played a major role in the degradation since
none of the isolates could grow on branched paraffins, olefins, or
cyclic alkanes as sole carbon sources. Bacteria capable of degrading
these compounds may also not have been isolated. Similar results from
-198-
-------
the microbiological studies were found during the biostimulation program
at Millville, New Jersey (Raymond, et al., 1978). The preliminary tests
before the bioreclamation operation began showed the presence of from
1C)2 to 10^ gasoline-utilizing organisms/ml in the ground water. The
microbial population responded to the addition of nutrients and oxygen
with a ten to thousand-fold increase in the numbers of gasoline-
utilizing and total bacteria in the vicinity of the spill, with levels
of hydrocarbon-utilizers in excess of lO^/ml in several wells. The
microbial response was an order of magnitude greater in the sand than
the ground water. Forty-one cultures were isolated from the soil and
ground water at this site, with 17 considered to be Pseudomonas, four
Flavobacterium, eleven Nocardia, and nine were not assigned to any
genus. Several of the Pseudomonas species were fluorescent in contrast
to the Ambler bacteria which did not include fluorescent pseudomonads.
Many of the cultures were composed of very small cells. After
biostimulation at a LaGrange, Oregon site, bacterial levels increased up
to six million times the initial levels.
Suntech's bioreclamation process has met with reasonable success
when applied to gasoline spills in the subsurface. After the
biostimulation program ended at Ambler, Pennsylvania, the gasoline-
utilizing bacterial levels declined from the high levels present during
the period of nutrient addition. This suggested a depletion of
nutrients and gasoline (Raymond, Jamison, and Hudson, 1975). The
concentration of gasoline in the produced water was not reduced during
the period of nutrient addition, but no gasoline was found in the
produced water 10 months after nutrient addition ceased. Estimates
-199-
-------
based on the amount of nitrogen and phosphate retained in the system
suggested that 744 to 944 barrels of gasoline were degraded. Residual
gasoline was found at the last sampling period at the Millville, New
Jersey, site, but no free hydrocarbon was observed in any of the wells
after the biostimulation program (Raymond, et al., 1978). The gasoline
concentrations in cores taken from the aquifer did not seem to change
substantially. Initially, the level of phenol in the ground water was a
problem, but it decreased to acceptable values after more aerobic
conditions were achieved in the aquifer. In the produced water which
had been circulated through the site, gasoline levels remained low and
fairly constant. The operation was successful in cleaning up the
aquifer to the point where no free gasoline was present and the site met
state approval. The operation was terminated and all well casings and
injection equipment removed. The nutrient supplementation program
succeeded in removing all the free product in the wells at the La
Grange, Oregon site (Minugh, et al., 1983); however, gasoline odors and
a cloudy sheen were detected in some of the pits dug after the cleanup
operation ended. Gasoline concentrations of 100 to 500 ppm were found
in the areas where the pits were dug with the average level of dissolved
organic carbon in the ground water at 20 ppm. Samples taken later
showed continued improvement. Vapor problems which had threatened two
restaurants were mitigated.
The advantages of the biostimulation process include:
(1) Useful for the removal of hydrocarbons and certain organic
compounds, especially water soluble pollutants and low levels
of other compounds which would be difficult to remove by any
other means.
-200-
-------
(2) Environmentally sound since no waste products are generated
and no ecological changes result because it utilizes the
indigenous microbial flora.
(3) Fast, safe, and generally economical.
(4) Treatment moves with the plume.
(5) Good for short term treatment of contaminated ground water
(Yang and Bye, 1979; J.R.B. Associates, 1982).
Its disadvantages are:
(1) Does not work with heavy metals and some organics.
(2) Bacteria can plug the soil and reduce circulation.
(3) Introduction of nutrients could adversely affect nearby
surface water.
(4) Residues may cause taste and odor problems.
(5) Could be expensive since equipment maintenance may be high and
long term injection of oxygen and nutrients may be necessary
to sustain high rates of degradation.
(6) Under certain conditions, such as high concentrations of
pollutants, it may be slower than physical recovery methods.
(7) Long term effects are unknown.
The bioreclamation process has been demonstrated to be useful in
the cleanup of hydrocarbon-contaminated aquifers of varying properties.
Concern was once expressed that the process might not be effective in
aquifers with low permeabilities. A laboratory study showed that
gasoline-utilizing bacteria could penetrate sand columns with effective
permeabilities ranging from 200 darcys to 3.5 darcys (sand packs of
coarse 20 mesh sand to very fine 80+ mesh sand) and consolidated
sandstone cores with effective permeabilities of 19 and 75 millidarcys
(Raymond, Jamison, and Hudson, 1975). The process has been used to
restore aquifers of dolomite (Raymond, Jamison, and Hudson, 1976), a
highly permeable sand (Raymond, et al., 1978), and alluvial fan deposits
-201-
-------
composed of sand, gravel, cobbles, and some clay and silt (Hinugh, et
al., 1983).
Alternative sources of oxygen have been suggested as a means to
increase the degradative activity in contaminated aquifers (Texas
Research Institute, 1982a). For example, hydrogen peroxide could be
injected into the contaminated soil above the water table in conjunction
with nutrients where it would decompose naturally or by enzymatic action
to increase the dissolved oxygen content. Hydrogen peroxide is
relatively inexpensive and does not present a persistent hazard in
ground waters (Texas Research Institute, 1982b). It is cytotoxic, but
many bacterial cells have enzymatic defenses (hydroperoxidases) against
it. Hydrogen peroxide was shown to be toxic to fresh bacterial cultures
at levels greater than 100 ppm, although mature cultures suffered less
and could function at levels as high as 10,000 ppm (Texas Research
Institute, 1982a). Another concern was that the hydrogen peroxide would
decompose before it reached the ground water; however, phosphate
buffered solutions at a pH of 7.0 and moderate flow rates were effective
for maintaining hydrogen peroxide levels (Texas Research Institute,
1982c). Subsequent experimentation with sand columns inoculated with
gasoline and gasoline-degrading bacteria showed that 1.0, 0.5, and 0.25
percent hydrogen peroxide solutions were toxic to the bacteria (Texas
Research Institute, 1983). Nagel, et al. (1982) documented the use of
ozone to treat petroleum contamination in Karlsruhe, Germany, that
threatened a drinking water supply. The polluted ground water was
withdrawn, treated with ozone, and infiltrated back into the system, via
three infiltration wells. About 1 g of ozone per g of dissolved organic
-202-
-------
carbon (DOC) was added to the ground water which increased the
biodegradability of the petroleum contaminants and added dissolved
oxygen; the dissolved oxygen levels increased in the ground water and
reached equilibrium at about 80 percent of the initial concentration
injected. The oxygen consumption peaked at about 40 kg/day during the
initial infiltration period. The DOC values in the wells fell from a
range of 2.5 to 5.5 g/m^ to a steady state of about 1.5 g/m^. Levels of
cyanide, a contaminant identified after the treatment began, also
decreased although biodegradation was not shown to be the cause. Total
bacterial counts in the ground water increased tenfold, but potentially
harmful bacteria did not increase. The drinking water from this aquifer
contained no trace of contaminants after l*s years of ozone treatment.
Jhaveria and Mazzacca (1982) of Groundwater Decontamination Systems
(GDS) adapted the biostimulation process to the cleanup of an aquifer
contaminated with compounds other than petroleum hydrocarbons; the
pollutants were solvents including methylene chloride, acetone, n-butyl
alcohol, and dimethylaniline. The contamination problem resulted from a
spill from a pharmaceutical company, Biocraft Laboratories, in Waldwick,
New Jersey where nearly 136,500 kg of the solvents were estimated to
have been lost. After consultation with Geraghty and Miller, Inc., the
initial consulting firm in Princeton-Aqua Science, and Suntech, Inc., it
was decided to contain the contaminated ground water on site and treat
it by stimulating the natural microbial population. The contaminated
ground water was pumped from dewatering trenches and wells to two
activating tanks where nutrients and oxygen were supplied. The water
was then pumped into two settling tanks and later injected into the
-203-
-------
subsurface enriched with nutrients, oxygen, and microorganisms. A
bacterial culture acclimated to the contaminants was established before
the biostimulation process began.
The biodegradation process was studied before the system was
designed (Jhaveria and Mazzacca, 1982). The wells had a population of
1C)3 to 10^ colonies per ml prior to the biostimulation programs, and the
addition of nitrogen and phosphate increased the growth of the organisms
as high as four times the control. A pilot study where nutrients and
oxygen were supplied to the aquifer showed that the microbial population
increased from lO^/ml to lO^/ml after seven days and remained constant
at that level. A batch process study using a ferraentor demonstrated
that after an acclimation period of eight days, a rapid decrease in the
COD of the ground water was noted; essentially all the compounds were
degraded. Additional air gave a quicker response. Two continuous
process studies were initiated to test the design of the plant. They
demonstrated that the addition of oxygen and nutrients to the acclimated
bacteria led to rapid removal rates of from 60 to 90 percent of the
methylene chloride, acetone, dimethylaniline, and butyl alcohol.
These successful tests led to the design and construction of the
system (jhaveria and Mazzacca, 1982). The above-ground portion
consisted of two activating tanks and two settling tanks. The
activating tanks were maintained at 20°C and supplied with air and the
following nutrients: ammonium chloride, monopotassium phosphate,
dipotassium phosphate, magnesium sulfate, sodium carbonate, calcium
chloride, manganese sulfate, and iron sulfate. The levels of the
nutrients were adjusted to provide the necessary nitrogen and
-204-
-------
phosphorus to stimulate in-situ degradation. Good results were achieved
by the system; removal of organics exceeded 95.9% in the above-ground
systems, and the levels of contamination in one of the producing wells
fell from 91 ppra of methylene chloride and 54 ppm of acetone, to less
than 1 ppm after a year. Similar degradation efficiencies were noted
for the monitoring wells. Sludge production was minimal since some of
the sludge was recycled from the settling tanks to the activating tanks
and some was allowed to pass to the recharge trenches to inoculate the
soil with acclimated microorganisms. An independent study by Professor
W.W. Umbreit of Rutgers University confirmed that methylene chloride was
oxidized by the GDS culture.
Enhancement of the microbial population ha3 also been reportedly
used to reduce levels of iron and manganese in the ground water
(Hallberg and Martinelli, 1976). The process known as the Vyrodex
method was developed in Finland and has been used in Sweden and other
locales where high levels of iron and manganese are found in the ground
water. Iron bacteria and manganese bacteria oxidize the soluble forms
of iron and manganese to insoluble forms; the bacteria use the electrons
adsorbed from the oxidation process as sources of energy. The Vyrodex
method works by adding dissolved oxygen to the ground water to stimulate
the iron and manganese to first remove the iron and later the manganese.
As the iron bacteria population builds up and begins to die, it supplies
the organic carbon necessary for the manganese bacteria. The efficiency
of the process increases with the number of aerations. Since no data
was presented that shows an increase in the number of iron and manganese
bacteria or in their activity, it is difficult to attribute the reduced
-205-
-------
level of iron and manganese in the ground water to microbial activity
instead of chemical oxidation from the introduction of dissolved oxygen.
Microbial activity in contaminated aquifers can also be enhanced by
altering the contaminant chemically or physically to make it more
degradable. One way to do that is to spread the contaminant over a
larger area by adding dispersants. Several dispersants were tested on
marine samples to determine their effects on the degradation of oil
(Atlas and Bartha, 1973a). They were found to increase the rate of
mineralization provided the sea water was amended with nitrate and
phosphate, but were not able to increase the extent of degradation.
Mulkin-Phillips and Stewart (1974a) tested four dispersants and found
that only one stimulated biodegradation. The four dispersants supported
microbial growth and were not toxic to the microbes, but did cause
population changes. Two oil herders tested by Atlas and Bartha (1973a)
were determined to increase the mineralization rate, but not the extent
of degradation. Application of surfactants has also been advocated to
clean-up contaminated aquifers, but they may be toxic and
nonbiodegradable (Texas Research Institute, 1979).
Addition of Acclimated Microorganisms
Another approach for restoring contaminated aquifers is the
addition of microbes to degrade the pollutants. Usually microbes that
have been acclimated to degrade the contaminants are used as "seed"; the
microorganisms may have been selected by enrichment culturing or genetic
manipulation. A mixed microbial population is often necessary to
degrade all the contaminants (Zajic and Daugulis, 1975). Microorganisms
-206-
-------
can become acclimated to the degradation of compounds by repeated
exposure to that substance. Felsot, Maddox and Bruce (1981) reported
that bacteria exposed to carbofuran yearly became so active in degrading
it, that it was no longer effective against the western corn rootworm.
Spain and Van Veld (1983) went so far as to suggest that the microbial
community near a pollutant spill might be primed with small amounts of
the pollutant to ensure a rapid microbial response. Seed organisms can
be isolated by enrichment culturing; the type of microorganisms which
are isolated are dependent on the source of the inoculum, the conditions
used for the enrichment, and the substrate (Atlas, 1977). Sequential
enrichment is a modification of enrichment culturing that allows
microorganisms that can more fully degrade the substrate to be isolated.
The substrate is inoculated with a microbial population and the
organisms which can degrade it are isolated. The undegraded substrate
is then used to isolate another set of organisms which are capable of
degrading the residual components. This process is continued until none
of the substrate remains or no new isolates are made. A potential
problem with such enrichments is that the organisms may interfere with
each other.
A new approach has recently been developed to induce acclimation by
genetically manipulating the microbes. Zitrides (1978) and McDowell,
Bourgeouis, and Zitrides (1980) used radiation to increase the genetic
variability of an adapted microbial population in hopes of producing
strains that could better degrade the contaminant. Selected strains of
bacteria chosen for their known ability to degrade similar compounds are
exposed to successively increasing concentrations of substrate. The
-207-
-------
fastest growing strains are irradiated. The genetic alterations should
increase the growth rate and fix the desired biochemical capability.
The adapted mutant will probably be at a disadvantage in the competition
with the native microbial population, and may only be able to
proliferate on the substrate upon which it was isolated.
Genetic engineering can be used to increase the degradation
capacity of microbes by improving the stability of the organisms,
enhancing their activity, providing them with multiple degradative
activities, and ensuring that they are safe both to the environment and
human health (Pierce, 1982). The ability to degrade some hydrocarbons
can be encoded on bits of extrachroraosomal DNA known as plasraids.
Plasmids can be transferred to an organism to increase the number of
compounds which can be attacked. The plasmids can also be fused
together to provide multiple degradative traits or to produce a novel or
previously unexpressed degradative pathway. Stable strains can be
engineered that will not pass on their plasmids or will be capable of
growth only under restricted conditions; this should limit their
potential for escape into the environment. Environmental conditions
such as temperature, pH, substrate concentration, oxygen tension, and
competition with the native microbial population may prevent the
genetically-engineered organism from reaching its degradative potential.
One successful application of genetic engineering is the
development of a bacterial strain capable of degrading 2,4,5-
trichlorophenoxyacetic acid (2,4,5-T). This strain was developed using
the technique of plasmid-assisted molecular breeding. This herbicide is
normally degraded very slowly and only by co-oxidative metabolism
-208-
-------
(Kilbane, et al., 1982). Kellogg, Chatterjee, and Chakrabarty (1981)
isolated an organism that could degrade 2,4,5-T as its sole carbon
source from a chemostat inoculated with microorganisms from a variety of
hazardous waste dumping sites and microorganisms harboring a variety of
pla8mids controlling the degradation of several hydrocarbons. The
chemostat was fed with low concentrations of 2,4,5-T and higher
concentrations of the plasmid substrates for 8 to 10 months. A 2,4,5-T
degrading population was isolated which could degrade up to 70 percent
of the 2,4,5-T supplied at 1.5 mg/ral. The strain was identified as
Pseudomonas cepacia AC 1100 (Chatterjee, Kilbane, and Chakrabarty,
1982). The ability of cepacia AC1100 to degrade 2,4,5-T in soil was
tested and it was shown to be able to degrade more than 95 percent of 1
mg/g of 2,4,5-T within a week. Soil samples contaminated with from
1000-20,000 Mg/g were inoculated with 5x10? AC1100 cells per g.
Complete degradation occurred in samples with 5000 Mg/g, and more than
90 percent of the 2,4,5-T was degraded after six weekly applications of
AC1100 for the 10,000 p g/g samples (Kilbane, Chatterjee, and
Chakrabarty, 1983). Other experiments were conducted to determine the
survival of AC1100 in the soil. When added to soil without 2,4,5-T the
levels of AC1100 dropped to negligible values after 12 weeks, but
following addition of 2,4,5-T the population increased until the
compound was exhausted. No appreciable effects on the number or types
of indigenous bacteria were noted when 10? AC1100 cells were added.
This indicates that no adverse ecological effects will occur when AC1100
is used as a seed to remove 2,4,5-T from soil. Until results from field
-209-
-------
tests are available, it will be difficult to determine how well the
organism will work under actual conditions.
There are problems with the use of genetically-engineered
microorganisms to degrade contaminants. Johnston and Robinson (1982)
could find no conclusive evidence that commercially available
genetically-engineered organisms were effective in establishing
themselves or significantly enhancing biodegradation of pollutants in
aeration basins or natural environments having an active native
microbial population. Most of the bacteria typically used in microbial
genetic work are eutrophs of the families Enterbacteriaceae and
Ps eudomonodo cea e which may not be able to attack substrates in the
parts-per-billion range that are often found in environmental samples.
Environmental stress such as unsuitable water quantities or the presence
of toxicants may also affect an introduced microorganism. However
genetic engineering holds great promise, especially in treatment
facilities where conditions can be controlled. These facilities will
probably be the first area where gene manipulation will be used.
Seeding microorganisms has been used in a number of different
environments to degrade organics. Gutnick and Rosenberg (1977) argued
that microbial seeding was ineffective in reducing oil contamination in
the sea, but that microbial seeding and nutrient supplementation would
work in contained environments like oil tankers. Miget (1972) found
that the effectiveness of microbial seeding in simulated marine
environments varied with the type and quantity of crude oil more than
with the inoculum density or nutrient salts concentration. Several
petroleum-degrading seed cultures have been developed for marine
-210-
-------
settings (Anonymous, 1970; Pacific Northwest Laboratories, 1970), but
the effectiveness of some of the agents is in doubt. Atlas and Bartha
(1973a) tested two bacterial inocula and found them to be totally
ineffective; the rates of crude oil mineralization in the inoculated and
control systems were identical and when the cultures were added to
sterile sea water, the rates of degradation were slower than in natural
sea water. A mutant bacterial culture used to degrade oil at a spill
contaminated beach was somewhat effective (McDowell, Bourgeois, and
Zitrides, 1980).
Schwendinger (1968) was one of the first to attempt to seed
microorganisms onto soil to effect the removal of oil. An inoculum of
Cellumonas sp. and nutrients was able to degrade the hydrocarbons more
effectively than just the addition of fertilizer alone. Other
researchers have met with varying success when seeding microorganisms
onto petroleum-contaminated soils. Lehtomaki and Niemela (1975) found
that the addition of 10® cells/g soil of two hydrocarbon-degrading
isolates did not significantly influence oil concentrations. The
application of 10*> oil-utilizing bacteria/cm^ resulted in a slight
additional degradation of the C20 to C25 group of n-saturated compounds
(Jobson, et al., 1974). Hunt, et al. (1973) advocated the use of
seeding to stimulate the degradation of oil in artic climates where
short summers and low soil temperatures may limit the activity and rate
of growth of the indigenous population. The addition of 10^
microorganisms per gram of dry soil was tested as a means of increasing
this activity. The inoculum was isolated from a soil taken around oil
seeps and enriched with Prudhoe Bay crude oil. The addition of nitrogen
-211-
-------
and phosphorus at 300 and 100 ppm, inoculation and adjustment of the pH
to 7 increased microbial activity over the controls by at least a factor
of 4 after 40 days. On the other hand, Westlake, Jobson and Cook (1978)
determined that the addition of oil-degrading bacteria to soil of the
boreal region of the artic resulted in no increase in the changes of the
recovered oil, but this may have been due to insufficient application.
Microbial seeding of soil environments has been applied to the
degradation of organics other than petroleum. Inoculation of cultures
acclimated to parathion could remove almost all the parathion within 11
days, while in the control, only 14 percent was removed (Daughton and
Hsieh, 1977). The acclimated culture could rapidly degrade the
parathion at levels up to 5000 ppm in non-flooded soils, and was able to
retain this capacity for 8 to 14 days after inoculation onto soils with
parathion. Edgehill and Finn (1983) isolated a strain of Arthrobacter
(ATCC3379) that could utilize pentachlorophenol as its sole carbon
source. In laboratory tests, inoculation of 10^ organisms per gram of
dry soil reduced the half life of the pesticide from 14 days to less
than one. Thorough mixing, larger inoculum sizes, and higher
temperatures increased the effectiveness of the treatment. A mixed
microbial population from primary sewage and another culture containing
primarily Pseudomonas were tested to determine their effects on the
degradation of formaldehyde and aniline in soil (Wentsel, et al., 1981).
This treatment was moderately successful in removing formaldehyde at
levels less than 2000 ppm. Aniline was not degraded by the primary
sewage inoculum, but was removed when a mixed population acclimated to
it was added along with nutrients and yeast extract. Application of
-212-
-------
microorganisms from sewage effluent was a low-cost, fairly effective
method for the removal of water soluble biodegradable organics. The use
of adapted mutant cultures to degrade chlordane and dinitrobenzene was
only partially effective with some loss of chlordane was noted. The
rates of dinitrobenzene loss were accelerated slightly in the chambers
inoculated with adapted microbes, but microbial activity was not shown
to be the cause.
The addition of adapted mutant microbes was not completely
successful in these cases, but has great potential. At least six
companies are involved in the production of microbial strains to be used
to treat abandoned hazardous waste sites and chemical spills (Anonymous,
1981 and Anonymous, 1982). They have had problems marketing their
bacterial products partly due to the number of irresponsible salesmen
who once sold enzymes as a potential cure-all. Polybac Corporation's
(Allentown, PA) products have been used to reduce formaldehyde levels in
the ground from 1000 ppm to under 50 ppm within 50 days, to clean up oil
spilled onto surface soils and a beach (McDowell, Bourgeois, and
Zitrides, 1980), to treat wastewater containing hazardous organics
(Wilkinson, Kelso, and Hopkins, 1978), to restore soils and ground water
contaminated by acrylonitrile (Polybac Corporation, 1983), and remove
orthochlorophenol which polluted soil and a pond (Anonymous, 1981).
O.H. Materials (Findley, OH) has used mutant bacteria to treat incidents
involving spills to soils and ground water of acrylonitrile; phenol and
its chlorinated derivatives (Walton and Dobbs, 1980); ethylene glycol,
propyl acetate and other compounds; dichlorobenzene, trichlorobenzene,
and methylene chloride (Quince and Gardner, 1982) and other compounds.
-213-
-------
Other companies identified by Anonymous (1981, 1982) which produce
mutant bacteria to detoxify soils contaminated with hazardous organics
include Sybron Biochemicals (Salem, VA), Flow Laboratories (Englewood,
NJ) and General Environmental Sciences (Cleveland, OH). Literature
written by the scientists at these companies indicate that their
products have been effective in restoring contaminated soils. An oil
pollution problem at a fuel transfer depot was treated with the addition
of 500 pounds of nutrients, 80 pounds of emulsifier, and rehydrated
mutant bacterial degraders along with tilling to provide aeration and
thorough mixing (McDowell, Bourgeois, and Zitrides, 1980). An
orthochlorophenol spill in Missouri contaminated soil and a pond
(Polybac Corporation, 1983). A spray/injection leachate system was
built using the pond as a treatment reactor. The pond was seeded with
Polybac's products and the concentration of orthochlorophenol fell from
15,000 ppm to less than 1 ppm within nine months. Polybac's products
were able to reduce acrylonitrile concentrations from 1000 to 1 ppm
within three months following a spill in Ohio. Surface foaming and a
spray/leachate recovery system with a portable bio-reactor were employed
in this cleanup.
Several factors must be considered before a cleanup system
employing acclimated bacteria can be implemented to restore contaminated
aquifers. The biodegradability of the contaminants must be assessed;
the source, quantity, and nature of the spilled material and the
environmental conditions of the site must be considered (McDowell,
Bourgeois, and Zitrides, 1980). A laboratory investigation of the
kinetics of biodegradation for the acclimated bacteria, the potential
-214-
-------
for inhibition under various conditions, the oxygen and nutrient
requirements, and effects of temperature should be done. The solubility
of the contaminant may need to be increased by adding emulsifiers to
allow microbial activity on the contaminants. The geology of the site
and the extent of contamination must be investigated. The stratigraphy
beneath the site and a description of the soils and bedrock should be
detailed (Quince and Gardner, 1982). The hydrogeologic data that is
needed is formation porosity, hydraulic gradient, depth to water,
permeability, ground water velocity and direction, and
recharge/discharge information. The quantity and character of the
contaminants and their location in the aquifer determines what
containment technique, recovery method, and treatment system should be
used. The systems which have used acclimated bacteria to restore
contaminated aquifers typically have relied on biological wastewater
treatment techniques such as activated sludge, aeration lagoons,
trickling filters, aerobic digestion, composting, and waste
stabilization. Many of the systems recharge the effluent from
biological treatment to the aquifer to create a closed loop of recovery,
treatment, and recharge. This flushes the contaminants out of the soil
rapidly and establishes hydrodynaraic control separating the contaminated
zone from the rest of the aquifer. Another benefit is that the
acclimated bacteria can be added to the aquifer and can act in situ to
degrade the contaminant. The recharge water can be adjusted to provide
optimal conditions for the growth of the acclimated bacteria and the
indigenous populations which may also act on the contaminants.
-215-
-------
Several examples illustrate the effectiveness of acclimated
bacteria to cleanup contaminated aquifers. An acrylonitrile spill was
handled by removing the ground water, treating it with mutant bacteria
in a small reactor, and recharging it (Polybac Corporation, 1983). The
levels of acrylonitrile fell from 1000 ppm to 1 ppm in three months.
For another acrylonitrile spill, the result of a tank car leak (Walton
and Dobbs, 1980), the primary treatment was air stripping, but after the
levels of acrylonitrile dropped enough to permit microbial growth,
mutant bacteria were seeded in the spill site. Following a period of
adaptation and growth, the bacteria were able to reduce the levels of
acrylonitrile to less than 2000 ppb. After treatment by clarification
and air stripping had reduced levels of the spilled organics (including
propyl acetate and ethylene glycol) at another site to concentrations
less than 200 ppm, bacteria, nutrients, and air were injected into the
subsurface (Quince and Gardner, 1982). This method reduced the levels
of the contaminants to below that required by the regulatory agencies,
with a reduction in the total organic carbon (TOC) from 40,000 to less
than 1 ppm. A similar treatment system was used to restore an aquifer
contaminated by organics lost from storage tanks. Air stripping and
inoculation with hydrocarbon degrading bacteria were able to remove 95
percent of the contamination.
At a site where 130,000 gallons of organic chemicals were spilled,
treatment was by clarification, adsorption onto granular activated
carbon (GAC), air stripping, and then recharge (Ohneck and Gardner,
1982). Once contamination levels fell below 1000 ppm, a biodegradation
program employing faculative hydrocarbon-degrading bacteria, nutrients,
-216-
-------
and oxygen was initiated. The goal for this portion of the treatment
waa to establish colonies of bacteria in the vadose zone to degrade the
contaminants trapped in the soil. The indigenous bacteria probably were
as important or more so for the degradation of the contaminants since
laboratory experiments showed that when they were supplied with
nutrients they were able to metabolize the contaminants at the same or
greater rates than the hydrocarbon-degrading bacteria inocula. Levels
of contamination decreased in one set of soil cores from about 25,000
mg/1 to about 2000 mg/1 in two months. Monitoring wells showed no
chemical contamination above the allowable criteria of 1 mg/1 at the end
of the treatment program. It was concluded that complete cleanup had
been achieved and had restored the quality of the aquifer to a level
equal to that existing prior to the spill. Shuckrow and Pajak (1981)
were less successful in their bench scale studies at a site in Muskegon,
Michigan, contaminated by several priority pollutants and at least 70
other organics. The priority pollutants included solvents, degreasers,
and vinyl chloride. Attempts to acclimate an activated sludge culture
to the new ground water were only minimally successful. A commercial
microbial culture was not effective either. However, coupling an
activated sludge process with treatment by granular activated carbon
proved beneficial since the activated sludge unit removed the organics
that passed through the GAC column. Up to 95 percent of the TOC was
removed so long as the GAC continued to function well. Microbial growth
on the activated carbon may also play a role in the removal of
contaminants (Werner, 1982).
-217-
-------
CONCLUSIONS
Of the three techniques discussed in this section, enhancement of
the native population and withdrawal and treatment seem to currently be
the most effective methods for biological restoration of aquifers
contaminated by organic compounds. Biostimulation by the addition of
oxygen, nitrogen, phosphorus, and other inorganic nutrients has been
chiefly used to reclaim aquifers contaminated by gasoline, and has been
effective in reducing the quantity of gasoline although not completely
eliminating it. Enhancing the indigenous microbial population was also
fairly effective in treating organic solvents which contaminated ground
water. Although alternative sources of oxygen such as ozone or hydrogen
peroxide increase the dissolved oxygen available for microbial activity,
they have the disadvantages of being toxic and being largely unproven.
Withdrawal and biological treatment of contaminated ground water has
been shown to be an effective method for restoration of aquifers.
Activated sludge treatment, often in conjunction with physical treatment
processes such as activated carbon adsorption or air stripping, has been
demonstrated to build up an acclimated microbial population that can
degrade the contaminants. While seeding of an acclimated or mutant
microbial population holds a great deal of potential since it reduces
the period needed for microbial acclimation to what are often
recalcitrant substances, results from previous attempts have not proven
it to be responsible for the removal of the contaminants. Further work
needs to be done to demonstrate that seeding microbes is a viable
technique for the restoration of contaminated aquifers.
-218-
-------
SELECTED REFERENCES
American Petroleum Institute, "Underground Spill Cleanup Manual",
American Petroleum Institute Publication No. 1628, 1980, Washington,
D.C.
Anonymous, "Microorganisms Consume Oil in Test Spills", Chemical and
Engineering News, Vol. 48, 1970, pp. 48-49.
Anonymous, "It's Hard to Find Jobs for Industrious Bugs", Chemical Week,
Vol. 128, 1981, pp. 40-41.
Anonymous, "Bugs Tame Hazardous Spills and Dumpsites", Chemical Week,
Vol. 130, 1982, p. 49.
Atlas, R.M., "Stimulated Petroleum Biodegradation", C.R.C. Critical
Reviews in Microbiology, Vol. 5, 1977, pp. 371-386.
Atlas, R.M. and Bartha, R. , "Effects of Some Commercial Oil Herders,
Dispersants, and Bacterial Inocula on Biodegradation of Oil in Sea
Water", In D.G. Ahearn, W.L. Cook and S.P. Meyers, eds., Proceedings,
The Microbial Degradation of Oil Pollutants, Atlanta, Georgia, December
4-6, 1972, Publication No. LSU-SG-73-01, 1973a, Louisiana State
University, Center for Wetland Resources, Baton Rouge, Louisiana, pp.
32-34.
Atlas, R.M. and Bartha, R., "Stimulated Biodegradation of Oil Slicks
Using Oleophilic Fertilizers", Environmental Science and Technology,
Vol. 7, 1973b, pp. 538-541.
Atlas, R.M. and Schofield, E.A., "Petroleum Biodegradation in the
Artie", In A.W. Bourquin, D.G. Ahearn and S.P. Meyers, eds.,
Proceedings, Impact of the Use of Microorganisms on the Aquatic
Environment, EPA 660/3-75-001, 1975, U.S. Environmental Protection
Agency, Corvallis, Oregon, pp. 183-198.
Bouwer, E.J., Rittman, B.E. and McCarty, P.L., "Anaerobic Degradation of
Halogenated 1- and 2- Carbon Organic Compounds", Environmental Science
and Technology, Vol. 15, 1981, pp. 596-599.
Bouwer, E.J. and McCarty, P.L. , "Transformation of 1- and 2- Carbon
Halogenated Aliphatic Organic Compounds Under Methanogenic Conditions",
Applied and Environmental Microbiology, Vol. 45, 1983a, pp. 1286-1294.
Bouwer, E.J. and McCarty, P.L., "Transformations of Halogenated Organics
Compounds Under Denitrification Conditions", Applied and Environmental
Microbiology, Vol. 45, 1983b, pp. 1295-1299.
Brown, K.W., et al., "Factors Influencing the Biodegradation of API
Separator Sludges Applied to Soil", In D.W. Shultz, ed., Proceedings,
Land Disposal: Hazardous Waste, Philadelphia, Pennsylvania, March 16-
-219-
-------
18, 1981, EPA-600/9-81-002b, U.S. Environmental Protection Agency,
Cincinnati, Ohio, pp. 19:88-199.
Chatterjee, D.K., Kilbane, J.J. and Chakrabarty, A.M., "Biodegradation
of 2,4,5-Trichloroacetic Acid in Soil by a Pure Culture of Pseudomonaa
cepacia", Applied and Environmental Microbiology, Vol. 44, 1982, pp. 514-
516.
Colwell, R.R. and Walker, J.D., "Ecological Aspects of Microbial
Degradation of Petroleum in the Marine Environment", C.R.C. Critical
Reviews in Microbiology, Vol. 5, 1977, pp. 423-445.
Daughton, C.G. and Hseih, D.P.H., "Accelerated Parathion Degradation in
Soil by Inoculation with Parathion-Utilizing Bacteria", Bulletin of
Environmental Contamination and Toxicology, Vol. 18, 1977, pp. 48-56.
Davis, J.B., et al., "The Migration of Petroleum Products in Soil and
Ground Water: Principles of Counterraeasures", American Petroleum
Institute Publication No. 4149, 1972, Washington, D.C.
Dibble, J.T. and Bartha, R. , "Effect of Iron on the Biodegradation of
Petroleum in Seawater", Applied and Environmental Microbiology, Vol. 31,
1976, pp. 544-550.
Dibble, J.T. and Bartha, R., "Effect of Environmental Parameters on the
Biodegradation of Oil Sludge", Applied and Environmental Microbiology,
Vol. 37, 1979a, pp. 729-739.
Dibble, J.T. and Bartha, R., "Leaching Aspects of Oil Sludge
Biodegradation in Soil", Soil Science, Vol. 127, 1979b, pp. 365-370.
Dibble, J.T. and Bartha, R., "Rehabilitation of Oil Inundated
Agricultural Land: A Case History", Soil Science, Vol. 128, 1979c, pp.
56-60.
DiTommaso, A. and Elkan, G.H., "Role of Bacteria in Decomposition of
Injected Liquid Waste at Wilmington, North Carolina", In J. Braunstein,
ed., Proceedings, Underground Water Management and Artificial Recharge,
Vol. 1, New Orleans, Louisiana, Sept. 1973, pp. 585-595.
Edgehill, R.U. and Finn, R.K., "Microbial Treatment of Soil to Remove
Pentachlorophenol", Applied and Environmental Microbiology, Vol. 45,
1983, pp. 1122-1125.
Erlich, G.G., et al., "Degradation of Phenolic Contaminants in Ground
Water by Anaerobic Bacteria, St. Louis Park, Minnesota", Ground Water,
Vol. 20, 1982, pp. 702-710.
Felsot, A., Maddox, J.V. and Bruce, W., "Enhanced Microbial Degradation
of Carbofuran in Soils with Histories of Furadan Use", Bulletin of
Environmental Contamination and Toxicology, Vol. 26, 1981, pp. 781-788.
-220-
-------
Guidin, C. and Syratt, W.J., "Biological Aspects of Land Rehabilitation
Following Hydrocarbon Contamination", Environmental Pollution, Vol. 8,
1975, pp. 107-112.
Gutnick, D.L. and Rosenberg, E., "Oil Tankers and Pollution: A
Microbiological Approach", Annual Review of Microbiology, Vol. 31, 1979,
pp. 379-396.
Hallberg, R.O. and Martinelli, R. , "Vyredox — In-Situ Purification of
Ground Water", Ground Water, Vol. 14, 1976, pp. 88-93.
Horowitz, A. and Atlas, R.M., "Response of Microorganisms to an
Accidental Gasoline Spillage in an Artie Freshwater Ecosystem", Applied
and Environmental Microbiology, Vol. 33, 1977, pp. 1252-1258.
Hunt, P.G., et al., "Terrestrial Oil Spills and Alaska: Environmental
Effects and Recovery", Proceedings, Joint Conference on Prevention and
Control of Oil Slicks, 1973, Washington, D.C., American Petroleum
Institute, pp. 773-740.
Jamison ,V.W., Raymond, R.L. and Hudson, Jr., J.O., "Biodegradation of
High-Octane Gasoline in Groundwater", Developments in Industrial
Microbiology, Vol. 16, 1975, pp. 305-311.
Jamison, V.W., Raymond, R.L. and Hudson, Jr., J.O., "Biodegradation of
High Octane Gasoline", In J.M. Sharpley and A.M. Kaplan, ed.,
Proceedings, 1976, Third International Biodegradation Symposium, Applied
Science Publishers, pp. 187-196.
Jhaveria, V. and Mazzacca, A.J,, "Bioreclamation of Ground and
Groundwater by the GDS Process", Groundwater Decontamination Systems,
Inc., 1982, Waldwick, New Jersey.
Jobson, A., Cook, F.D. and Westlake, D.W.S., "Microbial Utilization of
Crude Oil", Applied Microbiology, Vol. 23, 1972, pp. 1082-1089.
Jobson, A., et al., "Effects of Amendments on the Microbial Utilization
of Oil Applied to Soil", Applied Microbiology, Vol. 26, 1974, pp. 166—
171.
Johnston, J.B. and Robinson, S.G., "Opportunities for Development of New
Detoxification Processes through Genetic Engineering", In J.H. Exner,
ed., Detoxification of Hazardous Waste, Ann Arbor Science Publisher,
1982, Ann Arbor, Michigan, pp. 301-314.
J.R.B. Associates, "Handbook for Remedial Action at Waste Disposal
Sites", Final Report, EPA-625/6-82-006, 1982, U.S. Environmental
Protection Agency, Cincinnati, Ohio.
-221-
-------
Kappler, T. and Wuhrman, K., "Microbial Degradation of the Water-Soluble
Fraction of Gas-Oil, Part One", Water Research, Vol. 12, 1978a, pp. 327-
333.
Kappler, T. and Wuhrman, K., "Microbial Degradation of the Water Soluble
Fraction of Gas-Oil, Part Two", Water Research, Vol. 12, 1978b, pp. 335-
342.
Kellogg, S.J., Chatterjee, D.K. and Chakrabarty, A.M., "Plasmid-Assisted
Molecular Breeding: New Technique for Enhanced Biodegradation of
Persistent Toxic Chemical", Science, Vol. 214, 1981, pp. 1133-1135.
Kilbane, J.J., et al., "Biodegradation of 2,4,5-Trichlorophenoxy-Acetic
Acid by a Pure Culture of Pseudomonas cepacia", Applied and
Environmental Microbiology, Vol. 44, 1982, pp. 72-78.
Kilbane, J.J., Chatterjee, D.K. and Chakrabarty, A.M., "Detoxification
of 2,4,5-Trichlorophenoxy-Acetic Acid from Contaminated Soil by
Pseudomonas cepacia", Applied and Environmental Microbiology, Vol. 45,
1983, pp. 1697-1700.
Kincannon, C.B., "Oily Waste Disposal by Soil Cultivation Process", EPA-
R2-72-110, 1972, U.S. Environmental Protection Agency, Washington, D.C.
Kobayoshi, H. and Rittman, B.E., "Microbial Removal of Hazardous Organic
Compounds", Environmental Science and Technology, Vol. 16, 1982, pp.
170-183.
Lee, M.D. and Ward, C.H., "Microbial Degradation of Selected Aromatics
at a Hazardous Waste Site", Developments in Industrial Microbiology,
Vol. 25, 1984, (in press).
Leenheer, J.A., Malcolm, R.L. and White, W.R., "Investigation of the
Reactivity and Fate of Certain Organic Components of an Industrial Waste
after Deep-Well Injection", Environmental Science and Technology, Vol.
10, 1976, pp. 445-451.
Lehtoraaki, M. and Niemela, S., "Improving Microbial Degradation of Oil
in Soil", Ambio, Vol. 4, 1975, pp. 126-129.
Litchfield, J.H. and Clark, L.C., "Bacterial Activities in Ground Waters
Containing Petroleum Products", American Petroleum Institute, 1973,
Publication No. 4211.
McDowell, C.S., Bourgeois, Jr., H.J. and Zitrides, T.G., "Biological
Methods for the In-Situ Cleanup of Oil Spill Residues", Presented at:
Coastal and Off-shore Oil Pollution Conference, The French/American
Experience, Sept. 10-12, 1980, New Orleans, Louisiana.
-222-
-------
McKee, J.E., Laberty, F.B. and Hertel, R.M., "Gasoline in Ground Water",
Journal of Water Pollution Control Federation, Vol. 44, 1972, pp. 293-
302.
McNabb, J.E. and Dunlap, W.J., "Subsurface Biological Activity in
Relation to Ground Water Pollution", Ground Water, Vol. 13, 1975, pp. 33-
44.
Miget, R.J., "Bacterial Seeding to Enhance Biodegradation of Oil
Slicks", In D.G. Ahearn, W.L. Cook and S.P. Meyers, eds., Proceedings,
Microbial Degradation of Oil Pollutants, Atlanta, Georgia, Dec. 4-6,
1972, Publication No. LSU-SG-73-01, Louisiana State University, Center
for Wetland Resources, Baton Rouge, Louisiana, p. 32.
Minugh, E.M., et al., "A Case History: Cleanup of a Subsurface Leak of
Refined Product", In Proceedings, 1983 Oil Spill Conference:
Prevention, Behavior, Control and Cleanup, San Antonio, Texas, Feb. 28-
Mar. 3, 1983, pp. 397-403.
Mulkin-Phillips, G.J. and Stewart, J.E., "Effect of Four Dispersants on
Biodegradation and Growth of Bacteria on Crude Oil", Applied
Microbiology, Vol. 28, 1974a, pp. 547-552.
Mulkin-Phillips, G.J. and Stewart, J.E., "Effect of Environmental
Parameters on Bacterial Degradation of Bunker C Oil, Crude Oils, and
Hydrocarbons", Applied Microbiology, Vol. 28, 1974b, pp. 915-922.
Nagel, G., et al., "Sanitation of Ground Water by Infiltration of Ozone
Treated Water", GWF-Wasser/Abwasaer, Vol. 123, 1982, pp. 399-407.
Naumova, R.P., "A Comparative Study of Naphthalene-Oxidizing Organisms
in Underground Waters", Mikrobiologiyia, Vol. 29, 1960, pp. 415-418.
Odu, C.T.I., "Microbiology of Soils Contaminated with Petroleum
Hydrocarbons, Part One — Extent of Contamination and Some Soil and
Microbial Properties after Contamination", Journal of the Institute of
Petroleum, Vol. 58, 1972, pp. 201-208.
Ohneck, R.J. and Gardner, G.L., "Restoration of an Aquifer Contaminated
by an Accidental Spill of Organic Chemicals", Ground Water Monitoring
Review, Fall 1982.
Ogawa, I., Junk, G.A. and Svec, H.J., "Degradation of Aromatic Compounds
in Groundwater and Methods of Sample Presentation", Talanta, Vol. 28,
1981, pp. 725-729.
Pacific Northwest Laboratories, "Oil Spill Treating Agents — A
Compendium", American Petroleum Institute, Richland, Washington, pp.
170-178.
-223-
-------
Pierce, G.C., "Development of Genetically Engineered Microorganisms to
Degrade Hazardous Organic Compounds", In T.L. Sweeny, H.G. Bhatt, R.N.
Sykes and O.J. Sproul, eds., Hazardous Waste Management for the 1980's,
1982, Ann Arbor Science, Ann Arbor, Michigan, pp. 431-439.
Polybac Corporation, "Product Information Packet", 1983, Allentown,
Pennsylvania.
Quince, J.R. and Gardner, G.L., "Recovery and Treatment of Contaminated
Ground Water, Part I", Ground Water Monitoring Review, Fall 1982.
Raymond, R.L., "Reclamation of Hydrocarbon Contaminated Ground Waters",
U.S. Patent Office 3,846,290, Patented Nov. 5, 1974.
Raymond, R.L., "Environmental Bioreclaraation", For 1978 Mid-Continent
Conference on Exhibition on Control of Chemicals and Oil Spills, Sept.
1978.
Raymond, R.L., Hudson, J.O. and Jamison, V.W., "Oil Degradation in
Soil", Applied and Environmental Microbiology, Vol. 31, 1976, pp. 522-
535.
Raymond, R.L., Jamison, V.W. and Hudson, J.O., "Final Report on
Beneficial Stimulation of Bacterial Activity in Ground Waters Containing
Petroleum Products", Committee on Environmental Affairs, American
Petroleum Institute, 1975, Washington, D.C.
Raymond, R.L., Jamison, V.W. and Hudson, J.O., "Beneficial Stimulation
of Bacterial Activity in Ground Waters Containing Petroleum Products",
AIChE Symposium Series, Vol. 73, 1976, pp. 390-404.
Raymond, R.L., et al., "Final Report — Field Application of Subsurface
Biodegradation of Gasoline in a Sand Formation", American Petroleum
Institute, 1978, Project No. 307-77.
Rees, J.F. and King, J.W., "The Dynamics of Anaerobic Phenol
Biodegradation in Lower Greensand", Journal of Chemical Technology and
Biotechnology, Vol. 31, 1980, pp. 306-310.
Roberts, P.V., et al., "Organic Contaminant Behavior During Groundwater
Recharge", Journal of Water Pollution Control Federation, Vol. 52, 1980,
pp. 161-172.
Schwendinger, R.B., "Reclamation of Soil Contaminated with Oil", Journal
of Institute of Petroleum, Vol. 54, 1968, pp. 182-197.
Shuckrow, A.J. and Pajak, A.P., "Bench Scale Assessment of Concentration
Technologies for Hazardous Aqueous Waste Treatment", In D.W. Shultz,
ed., Proceedings, Land Disposal Hazardous Waste, Philadelphia,
Pennsylvania, Mar. 16-18, 1981, EPA-600/9-81-002b, U.S. Environmental
Protection Agency, Cincinnati, Ohio, pp. 341-351.
-224-
-------
Slavnia, G.P., "Naphthalene-Oxidizing Bacteria in Ground Water of Oil
Beds", Mikrobiologiya, Vol- 34, 1965, pp. 103-106.
Spain, J.C. and Van Veld, P.A., "Adaptation of Natural Microbial
Communities to Degradation of Xenobiotic Compounds: Effects of
Concentration, Exposure, Time, Inoculum, and Chemical Structure", Applied
and Environmental Microbiology, Vol. 45, 1983, pp. 428-435.
Suntech, Inc., "Environmental Bioreclamation Brochure", 1977.
Texas Research Institute, Inc., "Underground Movement of Gasoline on
Groundwater and Enhanced Recovery by Surfactants", Final Report,
American Petroleum Institute, 1979, Washington, D.C.
Texas Research Institute, Inc., "Enhancing the Microbial Degradation of
Underground Gasoline by Increasing Available Oxygen", Final Report,
American Petroleum Institute, 1982a, Washington, D.C.
Texas Research Institute, Inc., "Feasibility Studies on the Use of
Hydrogen Peroxide to Enhance Microbial Degradation of Gasoline",
American Petroleum Institute, 1982b, Washington, D.C.
Texas Research Institute, Inc., "Feasibility Studies on the Use of
Hydrogen Peroxide to Enhance Microbial Degradation of Gasoline",
Progress Report July 1982, American Petroleum Institute, 1982c,
Washington, D.C.
Texas Research Institute, Inc., "Progress Report: Biostimulation
Study", Feb. 1983.
Vanloocke, R., et al., "Soil and Groundwater Contamination by Oil
Spills: Problems and Remedies", International Journal of Environmental
Studies, Vol. 8, 1975, pp. 99-111.
Walton, G.C. and Dobbs, D., "Biodegradation of Hazardous Materials in
Spill Situations", Proceedings, 1980 National Conference on Control of
Hazardous Material Spills, Louisville, Kentucky, pp. 23-29.
Weldon, R.A., "Biodisposal Forming of Refinery Oil Wastes", Proceedings,
Oil Spill Conference, Mar. 19-22, 1979, Reprint.
Wentsel, R.S., et al. , "Restoring Hazardous Spill-Damaged Areas —
Technique Identification/Assessment", Final Report, EPA-600/2-7-81-1-
208, 1981, U.S. Environmental Protection Agency, Cincinnati, Ohio.
Werner, P., "Mikrobiologische Untersuchungen zur Chemish-Biologischen
Aufbereitung eines Humensaurehaltigen Grund-Wassers", Vom Wasser, Vol.
57, 1982, pp. 157.
Westlake, D.W.S., Jobson, A.M. and Cook, F.D., "In Situ Degradation of
Oil in a Soil of the Boreal Region of the Northwest Territories",
Canadian Journal of Microbiology, Vol. 24, 1978, pp. 254-260.
-225-
-------
Wilkinson, R.R., Kelso, G.L. and Hopkins, F.C., "State of the Art Report
— Pesticide Disposal Research", EPA-600/2-78-183, 1978, U.S.
Environmental Protection Agency, Cincinnati, Ohio.
Williams, D.E. and Wilder, D.G., "Gasoline Pollution of a Ground-Water
Reservoir — A Case History", Ground Water, Vol. 9, 1971, pp. 50-56.
Williams, E.B., "Contamination Containment by In-Situ Polymerization",
Aquifer Restoration and Ground Water Rehabilitation, Proceedings, Second
National Symposium on Aquifer Restoration and Ground Water Monitoring,
May 26-28, 1982.
Wilson, J.T., et al., "Transport and Fate of Selected Organic Pollutants
in a Sandy Soil", Journal of Environmental Quality, Vol. 10, 1981, pp.
501-506.
Wilson, J.T., et al., "Enumeration and Characterization of Bacteria
Indigenous to a Shallow Water Table Aquifer", Ground Water, Vol. 21,
1983a, pp. 134-142.
Wilson, J.T., et al., "Biotransformations of Selected Organic Pollutants
in Ground Water", Developments in Industrial Microbiology, Vol. 24,
1983b.
Windholz, M., ed., "The Merck Index", Merck and Co., Inc., 1976, Rahway,
New Jersey.
Wood, P.R., et al., "Introductory Study of the Biodegradation of the
Chlorinated Methane, Ethane, and Ethene Compounds", 1980.
Yang, J.T. and Bye, W.E., "Protection of Ground Water Resources from the
Effects of Accidental Spills of Hydrocarbons and Other Hazardous
Substances, (Guidance Document)", Technical Report, EPA-570/9-79-017,
1979, U.S. Environmental Protection Agency, Washington, D.C.
Zajik, J.E. and Daugulis, A.J., "Selective Enrichment Processes in
Resolving Hydrocarbon Pollution Problems", In A.W. Bourquin, D.G. Ahearn
and S.P. Meyers, eds., Proceedings, Impact of the Use of Microorganisms
on the Aquatic Environment, EPA-660/3-75-001, 1975, U.S. Environmental
Protection Agency, Corvallis, Oregon, pp. 169-182.
Zitrides, T.G., "Mutant Bacteria for the Disposal of Hazardous Organic
Wastewaters", Presented at Pesticide Disposal Research and Development
Symposium, Reston, Virginia, Sept. 6-7, 1978.
Zobell, C.E., "Microbial Degradation of Oil: Present Status, Problems,
and Perspectives", In D.G. Ahearn, F.D. Cook, and S.P. Meyers, eds.,
Proceedings, The Microbial Degradation of Oil Pollutants, Atlanta,
Georgia, Dec. 4-6, 1973, Publication No. LSU-SG-73-01, Louisiana State
University, Center for Wetlands Resources, Baton Rouge, Louisiana, pp.
3-16.
-226-
-------
SECTION VII
PROTOCOL FOR AQUIFER RESTORATION DECISION-MAKING
The general approach for developing aquifer restoration strategies
and selecting the most appropriate one for meeting a given need is, for
the most part, intuitively obvious. A logical first step is a
preliminary assessment of the nature of the problem. Based on the
preliminary assessment, a number of alternative strategies (remedial
measures) are developed. From the list of possible alternatives, an
optimum would be selected through systematically considering a series of
decision factors, environmental impact and cost-effectiveness analysis.
Implementation and construction of the chosen alternative would be next,
followed by monitoring of the effectiveness of the measure.
The objective of this section is to present a structured protocol
that can be followed to develop aquifer restoration strategies.
Emphasis is placed on the actual procedure for developing a list of
technical alternatives based on appropriate consideration of numerous
decision factors. The procedure is not intended to be a set of explicit
instructions; rather a general approach which, when modified could be
applied to a wide variety of ground water pollution problems. The
section begins with a review of pertinent literature; this is followed
by information on preliminary activities, and development and evaluation
of alternatives.
BACKGROUND INFORMATION
Most of the work in developing structured approaches or protocols
for addressing ground water pollution problems has been conducted in
-227-
-------
response to the Comprehensive Environmental Response, Compensation and
Liability Act, P.L. 96-510 (known as CERCLA or "Superfund").
Specifically, Superfund sites require the development of a "remedial
action master plan" (RAMP). The purpose of a RAMP is to identify the
type, scope, sequence and schedule of remedial projects which may be
appropriate in meeting an identified need (Kaschak and Nadeau, 1982).
Figures VII.1 and VII.2 are attempts to represent the RAMP process in
the form of flow charts. Additionally, Table VII.1 lists the phases of
a site contamination and liability audit. Two features are common to
all three of these outlines. First, they all have the same general
intuitive pattern. Second, they all have provisions for immediate
remedial actions to enable the conduction of studies to develop longer-
term solutions.
The RAMP is designed as an approach for developing an optimal
solution for meeting a given need. Incorporated within the RAMP is the
analysis of alternative remedial measures in order to decide on an
optimum strategy. The analysis involves consideration of three aspects:
(1) environmental impacts; 2) costs; and 3) risks. Risk assessment is
an area of study that is now receiving increased attention; Appendix F
contains some summary information on risk assessment as related to
aquifer restoration. The main problem faced by all risk assessment
techniques is that a large portion of the needed information, such as
risk pathways or acceptable concentrations, is unknown.
Because ground water cleanup activities are, in general, expensive,
considerable interest exists in analyzing costs. St. Clair, McCloskey
and Sherman (1982) discuss the advantages and disadvantages of risk
-228-
-------
/ request \
lor lundng y^i.
Lappfovedy
t$ Fund-\ Yes
Balancog y *
fesf mei7/
ScocMng
What type
No
No
' fh«r« ¦ \
threat 10 puttc \ ye»
health weftare /1 m
Source
Control
Measures
No
Acton
No
Action
action «
aooropmta?
data
Cooperative
Agreement
or Slate
Contract
Remedial investigation
No Fgn0-
tnanced
remedui acton
OM-SHe
Measures
NO
No
act
-------
ho
U)
0
1
CoAact *xj
•nriyti
data
lara or th»
No
RmwM
Figure VII.1: (continued)
Scrwninq
' Ara
r »aty to nftataniaty ^
mihgata trial to pubic
hee«h. watlara. or fta
IwWi: and not
•ufy to anpact
Rafino
nmmnfl
«*rrw*aantabi»y.
idaquicy and raf^
bdity. and potanM
uKaraa mpacta of
Mamaiwaa
YM
No
No
achon
Fewfy Study
-------
TIM IN
B
2
4
«
S
10
1!
14
It
It
20
22
24
MONTHS
NJ
Co
o Mocuntwtin e
LIWTTB ffUBIIU
INVESTIGATION/
tOSHILITT StUOl
B COOPERATIVE (
HCSmCKT
0 REVO
>tfpm
DESIGN
0 Rt¥P FROCUNtHENT 0
TKaTJCSTITTSr
REMOVAL OF
HAZARDOUS
WSTE MATERIALS
miw sequence
. PERMITS
NOTE: If FAST TRACl-REMOVAl OF MAZAROOUS MAST I
MATERIAL IS IWltWNTED. THE REMEDIAL INVESTIGATION
SHALL BE DELATED SO « TO COWEKCE AT THE COMPLETION
OF THE REMOVAL
PROCURE IC NT
REMEDIAL INVESTIGATION
FEASIIIIITT STUD*
-0 CONTRACTOR 0-
irwoiu—
INVESTIGATION
CONUNITV DEVELOP
RELATIONS FINN. HONK
PLAN 0 fur Q
<—. I WHTHfHT tWWHTr HUTIW5 KAMI —-»
Figure VII.2: Master Site Schedule (Kaschak and Nadeau, 1982).
-------
Table VII. 1: Site Contamination and Liability Audit Phased Structure
(Housman, Brandwein, and Unites, 1981).
Screening
Phases
Phase
Phase
1
2
Initial Property Inventory
Classification and Identification of
Potential Problem Properties
Phase
3
Preliminary Tield Screening
Phase
4
Prioritization of Problem Prooerties
Emergency
Action
Phase
Phase
5
Immediate Emergency Stop Action Response
Detailed Site
Investigation
and Remedial
Phases
Phase
Phase
6
7
Detailed Site ^ield Investigation
Definition of Remedial Strategies,
Risk and Financial Liability Assessment
and Remedial Cost Effectiveness
Phase
8
Selection of Preferred Remedial
Strategy
Phase
9
Implementation of Remedial Action
Phase
10
Certification of Performance and
Addressing Future Potential Liability
Issues
-232-
-------
assessment, cost/benefit analysis, cost-effectiveness analysis,
decision-tree analysis, trade-off matrices, and sensitivity analysis for
alternatives evaluation. Using certain elements of these techniques,
they developed a framework for evaluating the cost-effectiveness of
remedial actions. Evans, Benson, and Rizzo (1982) describe an
integrated, three-phased approach for cost-effective preliminary
assessments at hazardous waste sites. Dawson and Brown (1981) have
developed an integrated site restoration process, as outlined in Figure
VII.3, which includes cost-effectiveness considerations.
CONCEPTUAL FRAMEWORK FOR AQUIFER RESTORATION DECISION-MAKING
Depicted in Figure VII.4 is a flowchart for aquifer restoration
decision-making. Each of the steps in Figure VII.4 is discussed in
detail in the following sections. Although the general order of the
steps in Figure VII.4 is important (and to some degree dictated), it is
emphasized that none of the steps are completely independent of the
others. Additionally, the procedure is meant to be iterative, thus
allowing for refinement in both judgment and design.
Preliminary Activities
Preliminary activities associated with aquifer restoration
decision-making include the assemblage of a multi-disciplinary team and
problem definition and characterization through the conduction of a
preliminary study and evaluation of data and data needs.
Multi-Disciplinary Team
Ground water pollution is not strictly a hydrogeological problem.
-233-
-------
CHARACTERIZATION
ACTIVITIES
ASSESSMENT
ACTIVITIES
REMEDIATION
ACTIVITIES
f Are \
Data Sufficient
to Conduct Final
v Assessment? >
No
Ye
Are Risks
Acceptable?
No
Are
Residuals Risks
S. Acceptable?^
Collect New
Data
Reject
A1terna-
tive
Conduct Final
Assessment
Collect Existing
Data
Specify Minimum
Data Requirements
Select Recommended
Alternatives
Conduct Preliminary
Assessment
Design and Construct
Remedial Action
Implement Surveillance
and Monitoring Plan
Identify Feasible
Alternatives for
Remediation
Calculate Costs
Required for Each
Alternative
Rank. Alternatives
by Cost Effictive-
ness
Design Surveillance
and Monitoring
Plan
Calculate Risk
Reduction Potential
far Each Alternative
Calculate Risk After
Implementation of Each
Remediation Action
Alternative
Figure VII.3: Integrated Site Restoration Process (Dawson and Brown,
1981).
-234-
-------
Selection of
Multi-Disciplinary
Team
PRELIMINARY
ACTIVITIES
i
Problem Definition
and Characterization
1
Preliminary Study
Plume Delineation
Hydrogeologic Characterization
Site Characterization
Hater Use and Requirements
Human Health Costs/Risk Assessment
Land Use Patterns/Growth Projections
Regulations/Institutional Conatraints
Funding
Evaluate Data
Identify Data Needs
x
-* —» Goals Identification Matrix ~ Goal
Preliminary Feasible Alternatives « Technology-Decision Factor Matrix
I
Preliminary Screening
i
(Iteration)4—Scope Design —* Feasible Alternatives
DEVELOPMENT
OF
ALTERNATIVES
Economic Evaluation
Environmental Evaluation
Riak Assessment
EVALUATION
OF
ALTERNATIVES
Decision-Making Techniques (1)
(1) Discussed in Section VIII
Selective Alternative
SELECTION OF
AQUIFER RESTORATION
STRATEGY
Figure VI I.A: Flowchart for Aquifer Restoration Decision-Making.
-235-
-------
The solution to a ground water pollution problem will require
involvement from a number of entities and disciplines. Managerial
personnel will be involved in the overall planning and development of
the project. Technical personnel, both on-site and off-site, will aid
in the design of the remedial measure. Remedial-related personnel from
the construction industry will have involvement. Also, institutional
personnel from different levels of government will almost invariably
have involvement. Within each group, a number of different disciplines
might be required. For example, the technical personnel required could
include hydrogeologists, environmental engineers, soil scientists,
microbiologists, chemists, and toxicologists, to name a few.
Formulation and implementation of a solution for meeting a particular
need will require a multi-disciplinary approach involving a multi-
disciplinary team.
Problem Definition and Characterization
The obvious first step in actually dealing with a ground water
pollution concern is to define and characterize the problem. The
problem will need to be defined in terms of its temporal and areal
release patterns, and its urgency for formulating and implementing a
solution. The two temporal categories are "anticipated problems" and
"existing problems". Anticipated problems most usually will result from
planned facilities that have the potential to threaten ground water
supplies. The other type of problem is the existing problem. This is
the situation where a facility or activity with ground water threatening
potential is already in effect. Existing problems can be further sub-
-236-
-------
divided into those that have already degraded ground water, and those
that have not yet disturbed ground water but are expected to do so. The
approach needed in identifying potential solutions to either of the two
types of problems is similar, but because of their "temporal"
differences, the proposed solutions may differ significantly.
a. Anticipated Problems
Having identified an anticipated problem, the next step would be to
identify the areal release characteristics, and the potential duration
and urgency of the problem. The release characteristics of the problem
will be directly related to the source. The source, could be a point
source (disposal well), an area source (fertilizers in agriculture), a
line source (highway deicing salts), or a regional source (increasing
number of septic tanks in a region).
The duration of the problem can be classified as either acute or
chronic. Will the anticipated problem be short term such as lowering of
water levels due to construction dewatering, or will it be a long term
problem that will require long term solutions such as a series of
injection wells to prevent salt water intrusion?
The urgency of the anticipated problem will be a function of the
potential contaminant(s), local hydrogeological characteristics, and the
criticality of the threatened aquifer. Anticipated problems usually
will have no urgent need for a solution. Ground water pollution
potential can be minimized by incorporating certain controls in the
design stage.
-237-
-------
b. Existing Problems
With existing pollution problems the information needed for
characterization it is straightforward. Initial attention should be
given to source identification. The contaminant release characteristics
and duration of the problem can be characterized as outlined above. The
urgency or need for an immediate solution can sometimes be the critical
factor with an existing problem. If the polluted aquifer is a source of
drinking water, or the problem was discovered due to an adverse public
health reaction, there may be a need for an immediate (if only
temporary) solution.
A general description of the problem is necessary for defining the
scope and extent of further studies and ultimate remedial actions. For
example, if a hazardous substance is detected in a water supply well,
the immediate solution may be to provide an alternative supply of water.
This temporary measure may allow time for study and development of a
permanent solution.
Preliminary Study
After identifying an existing or proposed ground water threatening
activity or facility, the next effort should be toward further
information gathering and more detailed problem characterization. The
detail and duration of the preliminary study will be determined by the
urgency of the problem and funds available.
The information needs listed below are intended to be comprehensive
even though they are described in general terras. Not all information
needs will pertain to every problem. Probably the most useful function
-238-
-------
of these lists will be to aid in developing a list of needed information
for specific problems. The following information needs lists are not in
any particular order, although some of the needs are dependent on
others. An important issue that needs to be considered prior to
initiation of any detailed study would be that of funding. Once funding
possibilities have been determined, the scope of the analysis can be
more appropriately delineated. The other areas of information could
probably be approached in groups such as problem specific information
(plume delineation and hydrogeologic characteristics) site specific
information (site characterization, water use and requirements, and land
use patterns and growth projections), and others (human health costs and
risk assessment, and regulations and institutional constraints). Should
it be decided to go into more than just a cursory review of the problem,
sources of information may become a question. Some of the agencies
identified in considering regulations and institutional constraints most
probably will have information on other issues. Listed in Table VII. 2
are some other possible sources of information as related to the
different groups.
a. Plume Delineation
Plume delineation is the step in which the amount, nature, and
extent of the plume and the source of the ground water pollution is
characterized. The information obtained in this step will not only help
determine feasible aquifer restoration strategies but may also aid in
assessing the possibilities of pretreatment or economic recovery.
Information categories of interest are delineated as follows:
-239-
-------
Table VII.2: Potential Sources of Information
Group I Federal or state geological surveys, universities-
(problem specific) libraries, geology and engineering departments,
state health departments, property owner, county
records, well drillers.
Group II Weather bureaus, state water resources boards,
(site specific) census bureaus, soil and water conservation
districts, employment commissions, corporation
commissions, 208 studies, Department of
Agriculture, Forest Service.
Group III Medical libraries, state or Federal environmental
(other) protection agencies, state attorney generals
office.
-240-
-------
(1) Physical/Chemical Characterization of Pollutant(s) — A
complete physical/chemical characterization of the
pollutant(s) is essential. The physical/chemical
characterization will determine which surface treatment
technology will be required should the pollutant(s) be
removed. Some pollutants can be contained by subsurface
impermeable barriers, such as slurry walls, while others
(those with high ionic strengths) have been shown to actually
reduce the impermeability of such structures. A
physical/chemical characterization of the pollutant(s) can
also give information as to the possibilities for economic
recovery, i.e., will the pollutant(s) recovered be salvageable
or will it require treatment and disposal? In those cases
where a pollutant(s) is anticipated but not yet detected in
ground water, a physical/chemical characterization should be
done on the known or suspected source of pollution. A
physical/chemical characterization of a waste can give an idea
as to what pollutants to expect.
(2) Information on Transport and Fate of Pollutants(s) — Having
characterized the pollutant(s) it is then necessary to obtain
any and all information on the transport and fate of the
pollutant(s) in the subsurface environment. Information on
the attenuating capacity of the given soils for the
pollutant(s) is important. The behavior of the pollutant(s)
under the different pH environments of the subsurface is
important in that some pollutants can precipitate or
solubilize given the correct conditions. The final outcome of
this step should be information on the ability of the
particular pollutant(s) to actually migrate through the soil
structure and reach the ground water. Feasible mitigation
measures will be a function of this ability.
(3) Toxicity and Health Risks of Pollutant(s) — Information on
the toxicity and health risks of the pollutant(s) relative to
not only humans but livestock and vegetation should be
obtained. The ultimate use of the polluted ground water will
be a function of its toxicity and health risks. For example,
an aquifer that contains high concentrations of heavy metals
will have to be abandoned from further use, while one that has
elevated nitrate concentrations might be usable for irrigation
with minimal treatment. Specifically, information should be
obtained on toxic levels of the pollutant(s) perhaps in the
form of a standard. It is also desirable to obtain
information on the actual side effects of toxic levels of the
pollutant(s), i.e., does it cause temporary acute discomfort;
is it a potential carcinogen; are the effects long term or
life threatening?
(4) Areal Extent, Depth, Amount of Pollutant(s) — If possible, an
estimate of the areal extent, depth, and amount of
-241-
-------
pollutant(s) in the subsurface environment needs to be
obtained. The objective is to obtain an estimate of the
magnitude of the problem. The "size" of the problem to be
rectified, in itself, may limit the number of feasible
solutions. Ideally, data from existing monitoring wells,
well logs, etc., should be used to estimate the magnitude of
the problem. However, this information is often not
available. In most cases an "educated guess" of the magnitude
of the problem will have to be made by considering broad data
such as soil types, topography, climate, and duration of the
problem.
An important set of information pertinent to plume delineation is
the identification and characterization of the source of pollution. The
source will not always be readily identifiable and might require
extensive field surveys to be determined. Once the source has been
identified, information specific to it should be obtained. These
information categories are as follows:
(1) Physical/Chemical Characterization of Source — As mentioned
previously, a complete physical/chemical characterization of a
waste might be needed in order to predict potential
pollutants. However, this information can also be used for
reducing the problem at the source. A physical/chemical
characterization can aid in assessing the possibilities for
waste pretreatment, in-place stabilization of the waste,
recycling, or some other form of economic recovery.
(2) Variability of the Wastes — The variability of the wastes
must be considered in order that a pollutant specific
mitigation strategy is not implemented for a highly variable
waste. A highly variable waste source would include a
landfill that accepted both hazardous and non-hazardous
constituents. Other highly variable waste sources, which are
quite common today, are those situations in which records of
exactly what was disposed do not exist. When the materials
which have been disposed cannot be identified, it is difficult
to design anything but a general, all-inclusive type of
cleanup measure.
(3) Time Factors — Information on the time of existence of the
waste source is also needed. Data on the period a waste
source has been in existence can give insight to the magnitude
of the problem. Specifically, it is desirable to know how
long a given waste has been in-place, or how long a certain
activity has been operating.
-242-
-------
(4) Previous Waste Disposal Practices — It is important to know
historical information on the practice of waste disposal at
the source. Was the waste indiscriminantly dumped and
covered, or were there certain precautions taken? The
feasible solutions to the problem of leachates from an
abandoned open dump may differ from those of a sanitary
landfill with a leaking liner. Specifically, information is
needed on the exact disposal practices, the length of these
operations, any special ground water protection strategies
employed, and any or all daily records.
b. Hydrogeologic Characteristics
Assemblage of information on the hydrogeologic characteristics of a
site is essential for a successful aquifer restoration program. In
essence, this step involves characterization of the subsurface where
the problem exists. Subsurface characterization serves two main
purposes. First, a description of the hydrogeologic characteristics of
the site will allow for a better understanding of the magnitude of the
problem. A thorough hydrogeologic investigation will not totally
delineate the limits of the problem, but it will aid in estimating the
transport and fate of the pollutant in the subsurface. Second, a
thorough hydrogeologic investigation will aid in identifying and
designing potential aquifer restoration strategies. Information from
this step will also be valuable in the plume delineation step. Some
areas of needed information and reasons for their interest are as
follows:
(1) Geologic Setting and Generalized Soil Profiles
Determination of the types of soils is important for
determining both the capacity of the pollutant(s) to move
through the subsurface and the feasible restoration measures.
Certain soils will possess higher tendencies to attenuate the
pollutant(s) (through adsorption, precipitation, filtration,
etc.) than others. Likewise, certain soils will be amenable
to certain restoration strategies while others are not.
-243-
-------
(2) Soil Physical/Chemical Characteristics — Once a general soil
type has been identified, it is then necessary to characterize
this type both physically and chemically. Physical
characterization of the soil type will provide information on
the ability of the soil to filter the pollutant(s). The
physical characterization will also give an idea as to the
"workability" of the soil for different cleanup measures.
Chemical characterization will provide information on the
ability of the soil to chemically remove a given pollutant
through adsorption, precipitation, etc. In some cases, the
chemical composition of the soil will be needed to determine
the feasibility of a particular cleanup strategy, especially
in-situ technologies.
(3) Depth to Ground Water and Bedrock — The depth to ground water
in conjunction with the soil physical/chemical characteriza-
tion will give insight as to how long it will take a
pollutant(s) to actually reach the aquifer, if in fact it
will. If sufficient depth to the ground water exists in a
highly attenuating soil, minimal ground water pollution can be
expected. The depth to bedrock is needed to assess the
feasibility of some pollutant containment strategies such as
slurry walls, grout cutoffs, or sheet piles.
(4) Ground Water Flow Patterns and Volumes — The flow patterns
and volume of ground water threatened will play a vital role
in determining the feasible solutions to the problem.
Obviously the direction of flow will dictate the actual
physical placement of any proposed cleanup measures.
Similarly the volume of water affected will dictate the scope
of potential cleanup measures.
(5) Recharge Areas and Rates — Identifying recharge areas and
rates will play an important role in aquifer protection plans.
The use of institutional measures (such as zoning) to minimize
ground water threatening activities in aquifer recharge areas
is one such plan. Information on recharge areas and rates for
the case of an existing problem will be important for a number
of reasons. First, it should be determined whether or not the
source(s) is in a recharge area. Solutions to problems
located in recharge areas will most likely be more elaborate
than those not located in recharge areas, and should include
source removal if possible. Second, recharge rates will give
insight into the rate of pollutant(s) movement and
pollutant(s) dilution.
(6) Aquifer Characteristics — Identification of aquifer
characteristics will be essential for any analysis of ground
water flow and pollutant transport. This information becomes
extremely important if ground water modeling studies are to be
initiated. Listed in Table VII.3 are some of the
-244-
-------
Table VII. 3: Aquifer Characteristics
confined-unconfined
isotropic-anisotropic
homogeneous-non-homogeneous
hydraulic conductivity
dispersion coefficients
specific yield (storage coefficient)
-245-
-------
characteristics that need to be determined, either through the
use of extant data or the conduction of field studies.
(7) Existing Monitoring Well Locations and Procedures
Identification of existing monitoring well locations and the
parameters monitored can save both study time and costs.In
addition to providing immediate data, existing monitoring
wells can, in some cases, become permanent parts of a
monitoring network or converted to removal wells.
(8) Background Water Quality Data — Background water quality data
is important in determining the severity of the problem and
the appropriate remedial actions.
c. Site Characterization
A description of the general characteristics of the site of the
problem is also an important step. Because surface attributes of the
site will directly and indirectly affect the subsurface environment,
these attributes need to be identified. Information is needed on the
climatic factors of precipitation, temperature, and evapotranspiration.
Locational factors for which information is needed include topography,
accesibility, site size, proximity to surface water, and proximity to
population centers. Additional comments on each of these factors are as
follows:
(1)
(2)
3.
-246-
Precipitation — Precipitation in most cases will dictate
both the amount and rate at which pollutants from a site
"leach" to the ground water. This is especially true for
pollutants leaching from solid waste disposal areas.
Precipitation also affects the recharge rate of an
aquifer.
Temperature — Surface temperature will become an
important factor in determining the feasibility of
certain surface treatment strategies. For example, the
transfer of organics from ground water in an air-
stripping operation will be affected by temperature.
Evapotranspiration — The amount of water lost to the
atmosphere through transpiration and evaporation can be
important when considering the ultimate use of a site.
-------
For example, surface capping of a landfill relies in part
on evapotranspiration from the vegetative cover to reduce
the amount of water infiltrating to the solid waste.
(4) Topography — The general topography of the site will
affect the infiltration rate and the feasible solutions.
Areas of steep gradients will have little infiltration
and will be subject to high erosion rates. This could be
important when designing surface disposal sites.
(5) Accessibility — Related to topography is the
accessibility of the site. Areas of rugged terrain or
limited access will present not only construction
problems but could also hamper any subsequent operation
and maintenance activities.
(6) Site Size — The size of the site refers to the actual
surface areal extent of the problem. This factor will
affect the feasible solutions in that many of the
technologies are size specific. For example, soil
removal is probably not economically feasible for a
multi-acre contaminated site.
(7) Proximity to Surface Water — Locations of surface water
relative to the site are important for a number of
reasons. First, the solution to the ground water problem
should not create a surface water problem. In addition,
schemes that involve surface treatment may require
surface discharge. A location for this discharge needs
to be determined. Second, water may play an important
role in the construction (slurry walls) and/or operation
and maintenance (revegetation) of a site. A source of
water may be vital to the success of a given restoration
scheme.
(8) Proximity to Population Centers — The location of the
nearest population center relative to the site will help
determine both the magnitude of the problem and the
potential solutions. If the site is located near a large
population center supplied by a well field, the problem
may be more urgent than one located in a relatively
isolated area.
d. Water Use and Requirements
Determination of current ground water usage and future requirements
in the study area will aid in determining the criticality of the
-247-
-------
threatened or polluted ground water resource. Categories of information
needed and the reasons for their importance are as follows:
(1) Current Use — Information on the amount of water used for
domestic, agricultural and industrial purposes should be
obtained. Furthermore, this information should be divided
into those uses supplied by surface water and ground water.
Specifically, information on the amount of water withdrawn
from the aquifer in question and what it is used for, should
be obtained. This information could become critical should
abandonment of the aquifer be considered.
(2) Future Use — An estimate of predicted future water
requirements for domestic, agricultural and industrial
purposes should also be obtained. Those areas anticipating
large scale growth accompanied by high water demands may need
to cleanup existing ground water problems and institute
aquifer protection measures (institutional measures).
(3) Current/Future Water Quality Standards — Beneficial Use
Standards — Procurement of currently available or proposed
future beneficial use standards for local ground water could
play an important role in determining a feasible aquifer
restoration strategy. More specifically, beneficial use
standards may dictate the degree of treatment polluted ground
water must receive before subsequent use.
(4) Costs — An estimate of the current costs of supplying water
is needed, especially the costs of obtaining ground water from
the aquifer under study. The costs associated with changing
sources of supply should also be estimated. If changing to a
different source of water supply becomes prohibitively
expensive, it may be necessary to consider some more elaborate
treatment schemes for restoring the aquifer.
e. Human Health Costs and Risk Assessment
An assessment of risks and human health costs associated with the
polluted ground water is desirable. Having identified the toxicity and
health hazards of the given pollutant during the plume delineation step,
an assessment needs to be made of the actual potential for human health
impacts. This step is essential in evaluating the alternative of taking
-248-
-------
no action on aquifer restoration. Examples of needed information are as
follows:
(1) Potential Health Problems — From the plume delineation step
there should be information on the health problems associated
with the given pollutant(s). Specifically, information is
needed on whether the pollutant is a toxicant, carcinogen, or
threat to the general health of the population.
(2) Risk Assessment — Assessing the risks of illness is
desirable. Obviously, if a certain water is in violation of a
given health or water supply standard the risk is high. The
question of interest here is what are the risks of using a
potentially polluted yet acceptable (in accordance with the
standards or lack thereof) water supply? Ideally, it would
be desirable to have quantitative information on the number
and frequency of health incidents to be expected. A more
feasible solution would be to assign a subjective probability
of occurrence for different levels of contamination.
f. Land Use Patterns and Growth Projections
The objective of this information-gathering step is to insure that
potential restoration strategies do not become exercises in futility.
Specifically, it would be unwise to spend millions of dollars cleaning
up a particular problem that is just one of many currently or
potentially polluting a given aquifer. Useful information in this
regard includes current and future land use patterns and activity growth
in the study area. Examples of needed information are as follows:
a. Current Land Use Patterns — A determination of the existing
land use patterns will give considerable insight into the type
and degree of treatment a particular problem should be given.
The percentage of land devoted to urban, rural and industrial
uses in and around the site should be determined. A removal
and treatment system in a highly industrialized area could
lead to one group removing, treating and paying for pollution
it did not cause.
b. Growth Projections — A projection of growth in activities
that possess the potential for degrading ground water is
essential. If a certain ground water formation is already
degraded in quality, and the projections for future growth
-249-
-------
include increased activity in the area, then containment
rather than cleanup may be desirable. Listed in Table VII.4
are some of the more immediate ground water threatening
activities which should be explored.
g. Regulations and Institutional Constraints
A key step in any aquifer restoration strategy is to identify
institutional constraints and enforcement programs. A variety of
regulatory agencies for ground water quality control may exist within a
given state, and determining who is in charge may be somewhat of a
problem. Identification of all constraints and enforcement agencies is
necessary so that a proposed solution satisfactory to one agency does
not violate the requirements of another agency. Examples of necessary
information in this regard are as follows:
(1) Regulatory Agencies — As early as possible the regulatory
agencies in charge of ground water quality should be
identified. Areas of interest pertaining to ground water
would include monitoring of ground water, waste disposal and
construction permitting, and enforcement of standards.
(2) Standards and Guidelines — All pertinent water quality and
effluent standards should be obtained. In addition, some
states might have standards or guidelines for waste disposal
facilities, minimum construction standards, siting of specific
facilities, or required environmental studies.
(3) Legislation — There has been a good deal of recent
environmental legislation which could apply to a wide variety
of ground water pollution problems. Thorough knowledge of
pertinent legislation will not only provide goals and
guidelines for mitigation measures, but may also provide
insight as to possible sources of funding. For example,
consideration should be given to the requirements of the
Resource Conservation and Recovery Act (RCRA), the Surface
Mining Control and Reclamation Act (SMCRA), the "Superfund"
program (Comprehensive Environmental Response, Compensation
and Liability Act - CERCLA) for hazardous waste disposal
facilities, and the Safe drinking Water Act for "sole source"
aquifers. In addition, review of current state legislation
would be warranted.
-250-
-------
Table VII.4: Examples of Ground Water Threatening Activities
Industrial landfills and impoundments
Municipal landfills
Septic tanks
Mining, minerals extraction, petroleum production
Agricultural chemicals
-251-
-------
h. Funding
Probably the biggest obstacle to be overcome in any aquifer
restoration program will be funding. In simple terms, cleaning up
ground water can be a very expensive venture. Not only are the
technologies expensive to implement and operate, but they may be
difficult to design. If questions arise over who will pay for the
cleanup (as they often do), the pollution problem may end up in
litigation, thus adding to the expense and delaying the implementation
of aquifer restoration strategies. These delays exacerbate the problem
and escalate the costs. A preliminary review of the availability of
funds should moat probably be done immediately after identification of
the problem. Based on the availability of funds, the scope of the study
can be determined. If sufficient funds are not available, detailed
analysis and design will probably have to be foregone for a small-scale
aquifer management or policy analysis.
One of the key funding issues is related to responsibility for
cleanup costs. The pattern to date has been for the suspected polluters
to balk at accepting responsibility for a given case of ground water
pollution. The case then goes to litigation and the suspected polluter,
if found guilty, will often file bankruptcy rather than accept the
immense financial burden associated with ground water pollution cleanup.
This leaves the state or Federal government with the satisfaction of
having identified the polluter, but lacking in terms of a responsible
party for cleaning up. This type of arrangement is not the most
desirable for both sides and should be avoided if possible. The fact
-252-
-------
that legal recourse can be taken should be recognized, but an effort
towards joint cooperation is more desirable.
Data Evaluation
A critical aspect of the information gathering step will be
evaluating the quality of the information gathered. Kaschak and Nadeau
(1982) list three issues to be concerned with once available information
has been gathered as: (1) how good is the data today; (2) can an
engineering solution be properly developed and designed; and (3) will
the data be defensible in court? Consideration of the age of the
information, sampling and analysis protocols, and the chain of custody
of the information can aid in assessing whether or not the information
is accurate and/or useful.
Data Needs
Having completed the gathering and evaluation of available
information, the next step is to identify those areas where the search
for information was unsuccessful. Kaschak and Nadeau (1982) identify
this as the most difficult step, i.e., to decide what additional
information is necessary to be able to identify and evaluate alternative
restoration strategies without "studying the site to death".
Information voids should be identified and categorized. One
categorization scheme is as follows:
(1) Criticality — The relative importance needs to be considered
for each of the individual data gaps. It would probbly not be
justified to put as much effort into obtaining "regional
population growth patterns" as that of obtaining "toxicity"
information for the pollutants of concern. Some type of
priority ranking should be assigned to the missing
-253-
-------
information. Chapter VIII describes several techniques which
could aid in this ranking.
(2) Availability — For each of the information gaps consideration
should be given to whether or not the information is
obtainable. For example, records of disposal at a landfill
are desirable but are not always kept. If these records are
not available, there is no way of obtaining them. Conversely,
background water quality data may not be presently available
but is certainly obtainable through the drilling of monitoring
wells. In those cases where the information is not
obtainable, it could be excluded from the decision-making
process or a best estimate could be developed and utilized.
(3) Time — Certain information may be obtainable, but only after
extended periods of time. The background water quality data
mentioned previously is certainly obtainable, but monitoring
should occur over a period of time. Conversely, certain
aquifer characteristics can be obtained in just a few days.
If the problem being studied is of an emergency nature, then
data that requires long periods of time to gather should be
considered non-obtainable.
(4) Costs — For all the information voids, consideration must be
given to the cost of searching for and/or acquiring that
information. Obviously, an estimate of aquifer
characteristics from soils and geological maps will be less
expensive than extensive monitoring networks.
After the information gaps have been categorized under the above
headings, decisions can be made as to desirability and necessity of
filling the gaps. If the information-gathering process alone will
exceed the funding, the study may have to be abandoned or re-directed.
DEVELOPMENT OF ALTERNATIVES
The development of alternatives stage of the aquifer restoration
decision-making process shown in Figure VII.4 can best be described as
an interative process. The goals for addressing the established need
should be delineated, and potential technological approaches for meeting
the goals/need should be identified. These potential approaches
(strategies or alternatives) should be screened and scope designs
-254-
-------
developed for potentially feasible alternatives. This process may
require several iterations as additional information is developed.
Definition of Goals
The term "restoration" seems to imply that any ground water
remediation activity will attempt to return the aquifer to its original
condition. Although desirable, this is not always feasible. Therefore,
a subjective decision to be made in any aquifer restoration program is
related to the "goal" of the program. Specific strategies to achieve
the goal(s) will be dictated, for the most part, by physical, chemical
and monetary constraints.
This decision-making process includes four different goals:
prevention; abatement; cleanup; and restoration. Prevention, as the
name implies, means that pollution is not allowed to occur. The context
of "prevention" herein will be taken to mean "not allowing pollutants to
reach ground water". Abatement means "to put an end to". Hence,
abatement of ground water pollution will include the "cessation of
pollutants moving into the ground water, and the curtailment of the
movement of pollutants having already reached the ground water".
Cleanup refers to "elimination of pollutants through removal and
treatment or in-situ immobilization or treatment". Restoration will
include those measures that attempt to return the aquifer to its
original state. This most often will involve a cleanup strategy plus
recharge of treated or fresh water. It should be noted that these goals
are not totally independent of each other. More specifically, a truly
effective "cleanup" strategy may also include "prevention" and
-255-
-------
"abatement" steps. Table VII.5 contains a listing of various strategies
(detailed in Sections IV, V and VI) available for obtaining various
goals.
The definition of the goals of a ground water management strategy
does not represent a totally isolated evaluation. In some instances,
the goals may be limited by problem-specific conditions. Thorough
review of the results of the preliminary activities shown be undertaken
prior to establishment of goals.
Goals Identification Matrix
Table VII.6 is a goals identification matrix. The possible goals
are listed along with some decision factors and conditions under which
the goals might be appropriate. Review of the limiting factors for each
of the possible goals will enable a more rapid identification of
feasible and obtainable goals for an identified need. Not all possible
influencing factors have been listed in Table VII.6. Each identified
need will have its own influencing factors which should be considered in
the selection of the specific goal(s).
Technology-Decision Factor Matrix
Having identified a preliminary goal for an identified need, the
technology-decision factor matrix in Table VII.7 can be utilized. The
technology (technologies) that can be considerd for achieving various
goals are identified in the first two columns. After identifying the
feasible technology (technologies), the decision factors and conditions
under which the technology might be applied can be reviewed from the
remaining columns in Table VII.7. The end result should be a series of
-256-
-------
Table VII.5:Potential Aquifer Restoration Strategies for
Chronic Pollution Problems
Goal
Strategy Type
Action
Prevention
Institutional Measures
Source Control
1. Aquifer or effluent stan-
dards, effluent charges or
credits, land zoning.
1. Source reduction or removal
2. Optimum site selection
3. Man-made control options
a. Impermeable membranes
b. Impermeable materials
c. Surface capping
d. Collector drains
e. Interceptor trenches
Abatement
Waste Management
1. Modification of pumoing
2. Removal Wells
3. Pressure ridges
A. Subsurface barriers
Cleanup
Waste Treatment and
Disposal
1. Above-ground treatment
2. In-situ methods
Restoration
Cleanup Plus
Recharge
-257-
-------
Table VII.6: Goals Identification Matrix
Goals Temporal Funding Water Use Hydrogeological
Situation Costs Requirements Characteristics
Restoration Restoration Is a Restoration schemes Restoration should Restoration Is most
possible goal are the most expen- be Che primary feasible for shallow,
when considering slve management goal when dealing unconfined aquifers
existing problems, strategies. How- with ground water of originally high
ever, restoration that is or will be quality water,
will also have the the main source of
added benefits of wate" supply for a
returning the given area,
aquifer to a
beneficial use.
Cleanup
Cleanup is a
goal when
considering
existing
problems.
Cleanup strategies
are expensive.
Sometimes the
aquifer can be
returned to a
beneficial use
and there is also
the possibility
for economic
recovery from
certain pollutants.
Cleanup should be
considered when
the ground water
serves a benefi-
cial use other
than human con-
sumption such as
agricultural or
Industrial
purposes.
Cleanup measures
are most feasible
for shallow ground
water sources, but
do have some appli-
cability to deeper,
confined sources by
removal and treat-
ment.
Abatement
Prevention
Abatement can be
considered when
dealing with
existing problems
and can also be
Included as a
safeguard for
future facilities
or anticipated
problems.
Prevention mea-
sures are only
applicable to
future facili-
ties or antici-
pated problems
Abatement strategies Abatement measures
range from moderate are usually employ-
to very expensive
with little chance
for economic
recovery.
ed to save the
unpolluted por-
tions of an
aquifer while
conceding the
polluted portion
as lost.
Prevention strate-
gies range from low
to moderately expen-
sive with little
chance for economic
recovery.
Prevention mea-
sures should be
applied to protect
any currently used
or potentially
useable ground
Abatement measures
are most applicable
to shallow sources
with cessation
being the most
applicable abate-
ment measure for
deep resources.
No restrictions
-258-
-------
Table VII.6: (continued)
Goals
Areal
Extent
Human
Health Risks
Land Use
Patterns
Restoration
Cleanup
Abatement
As the size at the
problem Increases
the attractiveness
of the restoration
schemes decreases
due to Increased
costs. Problems
of regional nature
most probably will
be amenable Co
abatement measures.
Cleanup measures
are not economically
attractive for small,
localized problems
and do have economies
of scale for larger
p rob1cms. Howeve r,
if the problem
becomes extremely
large, such as
regional, these
economies of seals
disappear.
The abatement meas-
ures to be employed
are size specific
and cover the
whole range from
very small, local-
ized problems Co
large, regional
problems.
If human health
risks are high,
restoration or
cleanup should
be given top
priority if
technically
feasible. If
not technically
feasible,
abandonment and
development of
new source should
be considered.
See above.
Restoration and
cleanup measures
should not be
given high
priority in areas
that are or will
continue to be
subjected Co ground
water quality
stresses such as
industrial waste
disposal areas.
See above.
For low to moder-
ate health rlaks
abatement mea-
sures can be
employed to pre-
vent further
degradation of
an aquifer.
Abatement measures
are the most attrac-
tive alternatives
for ground water
pollution control
In areas subjected
to continued threats
if anything at all
1s Co be done.
Prevention Physical prevention No restrictions. No restrictions,
measures are size
specific. Some
institutional mea-
sures are applicable
on a regional level.
-258a-
-------
technologies or technology combinations that are potentially applicable
for meeting an identified need. Table VII.7 is not totally
comprehensive since there are many additional decision factors
associated with each of the listed technologies. However, Table VII. 7
does have general utility in that major decision factors are outlined.
Additionally, Table VII.7 should trigger information exchange among
members of the multi-disciplinary team. The team could develop an
expanded version of the technology-decision factor matrix specific to
the identified need under examination.
Preliminary Screening
Having developed a preliminary list of feasible technologies
(alternatives), a preliminary screening process can be used to narrow
the choices. The preliminary screening process will necessarily rely
heavily on professional judgment, however, listed below are some factors
that should be considered:
(1) Technical Feasibility — The technical feasibility of all
potential alternatives should be considered. Some
technologies have been widely used and their success
documented; others are still in the early stages of
development, or at least their application to ground water
pollution control has been limited. Another consideration
would be the technical capabilities of the personnel that are
to design and operate the proposed technology. Some
technologies are complex in design and require extensive
monitoring for successful operation and maintenance. If
trained personnel will not be available to handle these
responsibilities, perhaps a more passive system should be
considered.
(2) Public Acceptance — One important aspect of any aquifer
restoration scheme will be its acceptance by the public. If
the proposed solution will involve surface "eyesores" or
highly technical operational requirements, it might be
difficult to gain the acceptance of the general public.
-259-
-------
Table VII.7: Technology-Decision Factor Matrix
Technology
Management Strategy and
Complementary Technologies
Temporal Situation
Pollutant Type
Areal Extent
Institutional
Measures
Source Control
Strategies
Well Systems
Interceptor
Systems
Surface Water
Control
Implementation of insti-
tutional measures will
most probably be for
the prevention of
ground water pollution
but could also be used
as abatement measures.
Source control measures
are most applicable as
preventive techniques,
however, they are often
incorporated as part of
overall cleanup measures
at uncontrolled sites.
Well systems are appli-
cable to all management
strategies including
restoration. Well
systems in all cases
require complementary
technologies, usually
surface treatment and
discharge.
Interceptor systems can
be employed for cleanup,
abatement or prevention
of ground water pollu-
tion. However, in all
cases a complemeutary
technology is required,
usually in the form of
surface treatment and
discharge.
Surface water control
technologies are most
often used as preventive
measures but can also be
used aa part of abatement
schemes~
AppLicsbility to
existing problems is
probably minimal
except for orders to
cease a given activity.
Widest application is
for preservation of
current high quality
ground water.
Source control stra-
tegies are applicable
to existing problems
and should be included
in the design of all
future disposal
facilities.
Well systems are
applicable to both
existing and antici-
pated problems and
have been used to
control emergency
situations in certain
cases.
When dealing with
existing problems,
interceptor systems
will take the form
of interceptor
trenches, while
anticipated problems
are usually prevented
by employing collector
drain systems. There
is some applicability
of interceptor
trenches to emergency
spills involving
floating hydrocarbons.
No restrictions.
Well systems are appli-
cable to any pollutant
that flows with or on
top of the ground water.
Interceptor systems are
applicable to any pol-
lutant that flow* with
or on top of the ground
water.
No restrictions.
Well systems are prob-
ably not economically
feasible for small,
localized pollution
sources. Conversely,
for large sources of
pollution well systems
may become uneconomical
in that they by neces-
sity draw in both
polluted and non-
polluted water.
Interceptor systems are
feasible for small,
localised pollution
sources.
Surface water control Surface water coatrol is No restrictions.
can be applied to
existing problems as
part of an overall
abatement scheme and
should be considered
in the design of all
future land disposal
facilities.
applicable to most any
pollutant, however, the
possibility of gas pro-
duction should be con-
sidered when designing
"surface capping" mea-
sures for organic wastes.
Impermeable Liners are usually placed
Liners on e site prior to the
initiation of a given
activity, hence they are
most applicable as pre-
vention steps. Liners
themselves are not
totally effective and are
usually used in conjunc-
tion with surface water
control, source control,
or underdrein systems.
The applicability of
liners to existing
problems is limited
in that it usually
involves removal of
the waste or polluted
soi 1.
Some asphaltic and
treated soil liners
become more permeeble
upon exposure to lea-
chates. Some synthetic
liners show poor resis-
tance to hydrocarbons.
Because installation of
liners involves excava-
tion and specialized
construction, economies
of scale occur for
larger problems. How-
ever, if the size of
the problem is more than
5 to 10 acres, liners
may become uneconomical
due to material costs.
-260-
-------
Table VII.7: (continued)
Technology
Management Strategy and
Coaplementery Technologiea
Tempore1 Situation
Pollutant Type
Areal Extent
Sheet-piling
Sheet-piling follows the
attributes of grouting
end slurry systems in
that its aott practical
application* will be for
abatement and prevention
of ground weter pollu-
tion. However, the
nuaiber of caaee where
sheet-piling will be the
best technology is
minimal.
Sheet-piling ii
similar to grouting.
Probably the most
applicable situation
will be as emergency
•essurea for con-
tainment of spills.
Sheet-piles are gen-
erally corrosion
res istant.
As the aize of a pro-
ject increases, sheet
piling will become
uneconomical because
of high material and
shipping costs.
Slurry Walls
Surftce
Treatment
Io-Situ
Treatment
Grouting i« similar to
slurry systems in that
it will often rsquire
complementary tech-
nologies to accom-
plish restoration or
cleanup. Grouting
can be individually
applicable as an
abatement measure.
Slurry walls can only be
parts of overall restore
atioo or cleanup schemes.
Slurry wells are indivi-
dually applicable as
abatement messuree, how-
ever, some sore of % round
water removal system is
usually included. Slurry
walls are also iodivi-
duslly applicable aa
prevention measures, but
once again are usually
uaad in conjunction with
other measures.
Surface treatment tech*
nologiee cam be used in
systems designed to
prevent, abate, clean-
up or reetore polluted
aquifers. Treatment
technologies will
elweys require some
sort of pollutent re-
moval system.
la-aitu treatment tech-
nologies can be used in
systeaa designed to
abate, cleanup or re-
atore polluted aquifers,
la-aitu technologiee will
always require some sort
of nutrient/chemical
injectioo eystem.
Grouting is also a
versatile technology
in that it haa wide
application to both
existing and anti-
cipated pollution
problema. In addi-
tion grouting is
being promoted as
a procedure for con-
taining emergency
spills.
Slurry walls are very
versatile in that
they can be applied
to existing problems
aa abatement meesures
or they can be incor-
porated ia £f>« deaigo
of future fecilitise
for prevention of
ground watar pollu-
tion. Slurry walls
are probably not
applicable to emer-
gency eitustions
such se spills.
Surfece tra^twnt can
be used for acute
and chronic esses
of ground watsr pol-
lution. Treetment
cert be used at
existing sites and
at anticipated prob-
lems in e preventive
•ode.
ln-situ technologies
cen be applied to
both scute and chronic
cases of existing pol-
lution. the applica-
tion of in-situ tech-
nologies for treatment
of anticipated problems
ia not possible.
Grouts containing ben-
tonite clay are subject
to limitationa aimilar
to those for slurry
systems. Additionally,
chemical grouts are not
successful in highly
acidic or alkaline
environments¦
Slurry ayatems are sen-
sitive to specific
pollutants, especially
those of high ionic,
strength.
No restrictions.
In-situ treatment
is extremely pol-
lutant specific.
Biological methoda
are applicable
mainly to organics
and chemical methods
are applicable main-
ly to synthetic com*
pounds.
Costs for grout cutoff
systems are high* hence
they will be applicsble
to small localized cases
of pollution.
Slurry systems are more
epplicable to larger
problems. Slurry sys-
tema become more econ-
oaicel as the "length
to depth" ratio
increases.
Surface treetment hes
both upper and lower
size limitations.. Sur-
face treatment can be
too slaborate for small
sources yet uneconom-
ical for extremely
large areaa.
Same aa for Surface
Treetment.
-260a-
-------
Table VII.7: (continued)
Technology
Soils
Aquifer Characteristics Site Characteristics
Regulat ioaa
Institutional
Measure*
Source Control
Strategiea
No restrictions.
No restriction®.
Since most source con*
trol strategies will
involve surface activi-
ties, the site should
be readily accessible.
Well Systems Well systems require soils
that readily transmit
ground water and any
associated pollutants.
Well systems are amen-
able to most aquifer
types, however, they
may not be ideal for
extremely shallow
aqui Ears (< 5 m. ) .
Sites must be accessible
to drilling equipment.
Well systems usually
will require monitoring
and a receiving source
for their discharge. If
the removed water requires
treatment the site should
be accessible to or in the
proximity of surface treat-
ment facilities.
Interceptor
Systems
Interceptor syatems require Areas with high water
fairly loose, pervious
soils that readily transmit
ground water and any asso-
ciated pollutants.
tables or high imper-
vious soils or rock
layers are desirable.
Because the collected
water will need to be
removed, treated, and
discharged, the site
should be readily
accessible.
Surface Water
Control
Surface water control mea-
sures are applicable to
any soil. Surface cap-
ping becomes economical
when there exista native
clays readily available.
No restriction#.
degrading to maximize
runoff ia an important
aspect. Hence areaa
with extreme topographic
relief should probably
be considered less desir-
sble.
Impermeable Soils do not limit the
Liners application of liners,
except when native claye
are to be used as the
lioer. In this caae
the soils should be of
sufficient imperme-
ability and sufficient
thickness.
Very high weter tables
mey limit the use of
liners in that they may
cause the system to
"float".
The site should be
accessible to e variety
of construction vehicles,
including some special-
ised liner inatallation
machines.
Some states now have
regulationa governing
the in-place charac-
teristics which a
liner must possess.
-261-
-------
Table VII.7: (continued)
Technology
Soils
Aquifer Characteristics Site Characteristics
Regulations
Sheet-piling
Pile driviag requires a
relatively uniform,
loose, boulder-free eoil
for ease of construction.
Sheet-piling will be
limited to shallow
(<100 ft.) aquifers.
Similar to chose for
slurry systems except
the operation is not
water intensive.
Grouting Grouting is liaited to gran'
ular types of soils that
have a pore size Large enough
to accept grout fluids under
pressure yet small enough to
prevent significant pollu-
tant migration before imple-
mentation. Grouting in a
highly layered soil profile
may result in incomplete
formation of grout envelope.
Presence of high water
tables and rapidly
flowing ground water
limits groutability
due to dilution of
grouts and rapid trans-
port of contaminants.
Similar to chose for
slurry walls.
Some grouting proce-
dures are proprietary
and some chemicals
may be controlled
under the Underground
Injection Program.
Slurry Walls
Slurry systems are
applicable to most all
soil types. However,
in loose soils such as
sand or gravel there
can be significant
losses of slurry which
may make the system
uneconomical.
Slurry systems arc
generally applicable to
shallow (<150 ft.)
aqui fers.
Sites with general
accessibility, moderate
topographic rslief
and a ready source of
water are most desirable
Construction of slurry
systems is an equipment
intensive operation which
requires large amounts of
water.
Some construction
procedures are
patented and will re-
quire a license.
Surface
Treatment
No restrictions.
No reetrictions.
On-site surface treatment
vill require extremely
accessible sitee that cen
adequately handle construc-
tion treffic and equipment.
Once built, surface treat-
ment facilities will only
occupy a finite space, but
will have to remain access-
ible for monitoring ectivities.
In-Situ Host probebly applicable
Treetment only to permeable soils
which will readily accept
injected materials.
Shallow aquifers are
probebly more desirable.
Seme as for well systems.
-261a-
-------
(3) Physical Constraints — Some of the potential alternatives
might be eliminated by obvious physical constraints. For
example, placement of impermeable barriers might not be
feasible in a densely populated area.
Scope Design
Probably the best means of eliminating possible but non-feasible
alternatives would be through the use of a scope design. The scope
design is a small-scale analysis of the proposed alternatives for
preliminary estimates of size, costs, life expectancy, efficiency, etc.
The scope design will rely on estimates or rules-of-thumb figures for
costs, and other factors to generate an approximation for each of the
potential alternatives. The scope design will be less expensive, less
time consuming, and inherently less accurate than a complete engineering
and hydrogeological analysis. The objective of the scope design should
be to eliminate the obviously too expensive or too land-intensive
alternatives.
Iteration
With preliminary size and cost data from the scope design, an
iteration through the procedure outlined above would be appropriate.
Perhaps the data generated might indicate that all the possible
alternatives exceed the available funding. In this case, it might be
necessary to redefine the goals of the ground water management strategy,
and attempt to develop new possible technologies for achieving these
modified goals. The objective of iteration is to further narrow the
list of potential alternatives to those that are economically and
technically feasible.
-262-
-------
EVALUATION OF ALTERNATIVES AND SELECTION OF
AQUIFER RESTORATION STRATEGY
Selection of a single alternative from the list of feasible
alternatives should come from balanced consideration of technical,
economic and environmental factors. A philosophy often quoted and
generally accepted is that the "most inexpensive remedial action that
reduces risks to an acceptable level can be considered the most cost-
effective" (St. Clair, McCloskey and Sherman, 1982). This statement
raises a number of issues such as:
(1) What is an acceptable level?
(2) Acceptable to whom?
(3) Is the most cost-effective the optimum?
(4) Are there environmental implications in addition to economic
and risk implications?
(5) Are cost and risk equally important?
(6) Are confidence levels the same for cost and risk calculations?
The structure of this section will be to discuss methodologies for
and issues associated with evaluating alternatives based on their
economic, environmental and risk implications. Chapter VI presents some
decision-making techniques for integrating these evaluations and
developing an optimum strategy.
Economic Considerations
General statements concerning the costs of aquifer restoration are
difficult to make. One widely accepted statement is that the cost of
cleaning up a polluted aquifer is extremely expensive. This is true for
many but not all cases. Expense can be considered relative and not all
-263-
-------
ground water remediation activities are designed with cleanup as the
ultimate goal. Difficulties arise in any attempt to quantify or collate
information on the costs of aquifer restoration. First, minimal
detailed cost data is being or has been reported. This is due in part
to the confidentiality and litigation associated with many ground water
pollution problems. Additionally, the data that have been reported show
an extremely wide range, are highly problem specific, and sometimes are
estimates at best.
The fact that ground water cleanup costs exhibit a wide range can
be exemplified by citing some examples. Kaufmann (1982) has shown that
removing TCE and other organic compounds from the water supply of a
small New Jersey city can be accomplished for under $140,000 per year
using granular activated carbon. Neely, et al. (1981), on the other
hand, have estimated the cost of cleaning up ground water at a site
polluted by industrial solvents and acid sludge to be over $7 million.
Hittman Associates (1981) have preliminarily assessed the cleanup of
ground water polluted from abandoned lead and zinc mines in northeastern
Oklahoma to be in excess of $20 million.
The cost of a ground water remediation activity is second in
importance only to public health concerns. The technologies for
cleaning polluted ground water exist, but may not be economical. In a
recent study, Neely, et al. (1981) found that cost was the prime
determinant of the type of remedial technology employed at uncontrolled
hazardous waste sites. Since elaborate cleanup schemes are expensive,
companies may have opted for less expensive but environmentally-more
threatening prevention and abatement measures.
-264-
-------
Any ground water cleanup activity will require funding for
feasibility and engineering studies, construction, operation and
maintenance, and other costs. Any one of these areas can become a major
cost factor. For example, Roux Associates recently began a $1 million
study just to develop a list of recommendations for remedial actions at
Massachusetts toxic waste sites. In contrast, Kaufmann (1982) has shown
operating costs to be the major cost component of a granular activated
carbon cleanup system.
The generic assignment of costs to various aquifer restoration
technologies (strategies) is difficult. Individual technologies can
vary in cost depending on the type of pollutant, nature of the aquifer,
the ultimate goal of the technology, and many other factors. A good
deal of work has been done in assessing the costs of various components
of individual technologies. For example, extensive information has
become available on the cost of installation and operation of monitoring
wells. Monitoring wells should be an integral part of any aquifer
restoration technology. Minning (1982) provides a cost comparison for
five different drilling methods. Rehtlane and Patton (1982) examine the
economic tradeoffs of multiple port versus standpipe piezometers. Ward
(1982) examines the costs of in-situ leach wells for uranium extraction.
As noted in Section IV, Campbell and Lehr (1977) give an excellent
discussion of the costs of completing wells under a variety of
geological conditions.
There have been several recent studies to assign unit costs, or at
least comparative cost figures, for some of the physical measures for
minimizing, abating or controlling ground water pollution. Rishel, et
-265-
-------
al. (1983) have estimated costs for remedial response actions at two
types of uncontrolled and abandoned hazardous waste disposal sites .
Table VII.8 is a listing of costs for a landfill site, and Table VII. 9
summarizes costs for a surface impoundment site.
Tolman, et al. (1978) developed a summary of estimated costs and
characteristics of remedial methods for a hypothetical, single size (10-
acre) landfill; the summary is shown in Table VII.10. Additionally,
some qualitative comments can be made concerning some of the
technologies listed in Table VII.10. First, surface water control
technologies are usually only used as preventive and/or abatement
measures when dealing with ground water pollution. Hence, these
operations will require some complementary technology to cleanup an
aquifer. Of the four ground water control technologies listed in Table
VII.10, a bentonite slurry trench must be considered the most cost-
effective. Construction of grout curtains is slow, labor intensive and
expensive, and as such this process is probably only economically
feasible only small, localized cases of pollution. The plume management
strategies listed in Table VII.10 all will incur long-term operating and
maintenance costs. These costs are associated with manpower to oversee
operations, power supplies for pumps, and possibly accelerated
maintenance due to the nature of pollutants being pumped (Glover, 1982).
Also, the passive management systems (drains, interceptor trenches)
generally have lower operating costs than the active (pumped) systems.
In summary, most of the work to date in the area of costs of
aquifer restoration technologies has been in developing data for the
components of the technologies. Minimal information has been reported
-266-
-------
Table VII.8: Average U.S. Low and High Costs of Unit Operations for Medium-Sized
Landfill Sites (Rishel, et al., 1983)
Average U.S. Cost S Per Unit'
Initial Capital
Life Cycle Costs
Unit Operations
Unit
Low
High
Low
High
Total Units Used"
Contour grading and
Site area, ha
15.300
17.900
16.300
19.900
5.4 ha site area
surface water diversion
Bituminous concrete
Site area, ha
€7.300
92.700
67.300
92.700
5.4 ha site area
surface sealing
Revegetation
Site area, ha
3.450
16.500
14.300
18.100
5 4 ha site area
Bentonite slurry trench
Wall face area, m1
54 5
96 1
61 2
103
10.800 m2 wall face area
Grout curtain
Wall face area, m2
600
1.209
937
1.880
10.800 m2 wall face area
Sheet piling cutoff wall
Wall face area, m'
73
108
73
108
10.800 m2 wall face area
Grout bottom sealing
Site area, ha
5.282.000 10.209.000 5.296.0O0 10.224.000
5.4 ha site area
Drains
Pipe length, m
72 7
106
357
416
260 m pipe length
Well point system
Intercept tace area, m2
625
105
107
153
2.000 m2 intercept face
araa
Deep well system
Intercept face area, m2
116
18 3
28 6
37.2
01 CO
4.800 m2 intercept face
Bra a
Injection
Intercept face area, m2
77
90
1.760
1.785
01 CO
550 m2 intercept face area
Leachate recirculation
by subgrade irrigation
Site area, ha
5.270
8.360
19.700
24.000
5.4 ha site area
Chemical fixation
Site area, ha
69.100
130.000
82.500
145.000
5.4 ha site area
Chemical injection
Landfill volume, m3
167
328
2 16
381
150.000 m3 landfill volume
Excavation and reburial
Landfill volume, m3
116
120
116
120
596.000 m3 landfill volume
Ponding
Site area, ha
647
1.028
647
1.028
5.4 ha site area
Trench construction
Trench length, m
122
14 34
15 11
2032
930 m trench length
Perimeter gravel
trench vents
Trench length, m
992
144
100
146
935 m trench length
Treatment of contami-
nated ground water
Contaminated water.
L/d
1 52
257
2 52
4.38
440.740 L/d
contaminated water
Gas migration control -
passive
Site perimeter, m
161
241
168
256
935 m site perimeter
Gas migration control ¦
active
Site perimeter, m
113
173
167
279
935 m site perimeter
' For 5.4 ha site
are discounted at 114% to present value, capital costs are not amort tied
-------
Table VII.9: Average U.S. Low and High Costs of Unit Operations for Medium-Sized
Surface Impoundment Sites (Rishel, et al., 1983)
Average & S Cost $ Per Unit*
Initial Capital
Life Cycle Costs
Unit Operations
Unit
Low
High
Low
High
Total Units Used*'
Pond closure and contour
grading of surface
Site area, ha
26.900
35.100
35.900
53.500
0.47 ha site area
Bituminous concrete
surface
Site area, ha
48.500
70.700
48.500
70.700
0.47 ha site area
Revegetation
Site area, ha
2.540
3.820
3.970
5.450
0.47 ha site area
Slurry trench cutoff wall
Wall face area, m2
60 1
106
60 1
106
4.165 m2 \kall face area
Grout curtain
Wall face area, m2
326
631
343
649
4.104 m2 wall face area
Sheet piling cutoff wall
Walt face area, m2
768
115
94 6
135
4,100 m2 wall face area
Grout bottom seal
Site area, ha
868.000
1.621.000
1.024.000
1.792.000
0.47 ha site area
Toe and underdrains
Pipe Length, m
316
609
1,550
1.960
60 m pipe length
Well point system
Intercept lace area, m2
62 3
117
321
398
300 m2 intercept
face area
Deep well system
Intercept lace area, rri2
332
603
114.4
149
950 m2 intercept
face area
Well injection system
Intercept face area, m2
31 3
555
109
141
950 m2 intercept
face area
Leachate treatment
Contaminated water.
L/d
1.16
196
4 49
814
51.870 L/d
contaminated water
Berm reconstruction
Replaced berm. m1
2 98
3 80
4.00
585
410 m3 berm
Excavation and disposal
at secure landfill
Impoundment
volume, m3
260
268
260
268
5,000 impoundment volume
* Mid-1980 dollars, 10-year life cycle. O A M costs are discounted at 11.4% to present value, capita/ costs are not amortized.
" For 0.47 ha impoundment.
-------
Table VII.10: Summary of Estimated Costs and Characteristics of Remedial
Methods (Tolman, et al., 1978)
Method
Average
Estimated Costs *
($ in Thousands)
Characteristics/Remarks
Contour Grading
Surface Water Diversion
Surface Sealing
Clay (14-46 cm (6-18 in.))
Bituminous Concrete (4-13 cm
(1.5-5 in.))
Fly Ash (30-60 cm (12-24 in.))
PVC (30 mil)
Drainage Field (if required)
Revegetation on Slopes <12 percent
on Slopes >12 percent
Surface Water Control
184 Increases runoff, reduces infiltration.
20 Diverts surface water from fill.
234 If locally available, native clay is
economical means of retarding
infiltration.
315 Rapid coverage; can eliminate infiltration.
235 Material may leach metals; may be
available free.
482 Very impermeable; expensive seal; careful
subgrade preparation is necessary.
65 Carries infiltrated water off seal;
increases effectiveness of seal.
10 Stabilizes cover material; seasonally
19 increases transpiration; provides
aesthetic benefit.
Bentonite Slurry Trench
Grout Curtain
Sheet Piling
Bottom Sealing
Groundwater Control
670 Simple construction methods; retards
groundwater flow.
1,400 Very effective in permeable soils.
800 Widely used for shoring.
4,000 Leachate collection may be needed, acts as
a liner; difficult drilling through
refuse.
Drains
Well Point Dewatering
Deep Well Dewatering
Injection/Extraction Barrier
Spray Irrigation
At-grade Irrigation
Subgrade Irrigation
Chemical Fixation of Cover
Chemical Injection
Plume Management**
23 Effective in lowering water table a few
meters in unconsolidated materials; can
be used to collect shallow leachate.
185 Suction lift limits depth to 7-9m (20-30
ft); inexpensive installation; uses only
one pump; can be used to collect shallow
leachate.
183 Used in lowering deep water tables; one
pump needed per well; high maintenance
costs.
199 Creates a hydraulic barrier to stop
leachate movement; operation and
maintenance costs are high.
366 Spreads leachate over the landfill for
recvcllng; potential odor problem.
32 Gated pipe with ridge and furrow irriga-
tion; potential odor problem; recycles
leachate.
28 Large-scale drainage field; recycles
leachate.
Chemical lmmobllizat ion
145 Uses chemically fixed sludge to provide a
top seal; provides means of disposal for
sludge; helps stabilize landfill.
86 Immobilizes a single pollutant; In most
cases not feasible.
Excavation and Reburlal
4,570
Excavation and Reburlal
Very exnensive; difficult construction.
•Costs for hypothetical 4-hectare (10-acre) landfill. High and low estimates were averaged
to determine these costs.
**Costs include present worth of 20 years, operation, maintenance, and, where applicable,
power for 4-hectare (10-acre) landfill.
-269-
-------
in terms of costs of overall aquifer restoration strategies. Most
studies will require the analysis of individual cost components as
outlined in the following section.
Economic Evaluation
Economic evaluation of feasible alternatives can follow the
traditional procedure of itemizing all the costs for a given project,
ammortizing these costs for the life of the project, developing measures
of the benefits of the project, and comparing these data for each of the
alternatives. A number of unique concerns arise when considering an
economic evaluation of ground water contamination problems.
The first major concern is associated with the comparison of
alternatives on a common basis. Two alternatives may be projected to
achieve the same result through radically different methods. Relevant
issues under this heading are as follows:
(1) The first issue is the level of treatment or efficiency of the
system. One alternative might reduce the contaminant
concentration to parts per billion, another to parts per
million. Is the increased reduction in concentration needed
and/or economically justifiable?
(2) Some alternatives may be operation- and maintenance-intensive
while others, providing the same solution, might be less
intensive. Comparison of these types of alternatives is
difficult due to constantly changing interest rates.
(3) Alternatives providing similar solutions might have radically
different time frames or life expectancies. All projects
should be compared on a common project period. Bixler, Hanson
and Langner (1982) have identified this period to be the
lesser of: the period of potential exposure to the
contaminated materials in the absence of remedial action; or
20 years.
The second major concern is the cost of specific items associated
with alternatives. It is important that all cost items be identified.
-270-
-------
As an example of how extensive these lists can be, Table VII.11
summarizes the cost components for a well system (Lundy and Mahan,
1982). Bixler, Hanson and Langner (1982) note that cost items can be
monetary and non-monetary, direct or indirect. In addition to the
obvious direct monetary costs such as materials, equipment, etc., non-
monetary costs must be included such as relocation costs, loss of
revenues, decreased property values and others. It is important that
the list of cost items be complete and include all pertinent costs
(Thorsen, 1981).
Another concern associated with the economic evaluation of
alternatives is that of developing a measure of the benefits of an
alternative. Some benefits will be very difficult to assign dollar
values, for example, reduced health risks. Additionally, some benefits
may be hard to quantify, much less assign dollar values.
The final and most important issue associated with economic
analysis is the generic approach to be utilized. St. Clair, McCloskey
and Sherman (1982) analyzed the advantages and disadvantages of various
approaches, including risk assessment, cost-benefit analysis, cost-
effectiveness analysis, trade-off matrices, and sensitivity analysis.
Utilizing the positive attributes of these approaches, St. Clair,
McCloskey and Sherman (1982) have developed a framework for evaluating
cost-effectiveness of remedial actions at uncontrolled hazardous waste
sites. Table VII.12 displays an example trade-off utilizing the
developed framework.
In summary, there exists no ideal methodology for economic
evaluations of aquifer restoration strategies. The best approach is to
-271-
-------
Table VII. 11: Cost Components of Well Systems (Lundy and Mahan, 1982)
Cost Component
Cost Element
Principal Factors Affecting Cost
Plume Delineation (K)*
System Design (K)
Well System (K)
Surface Infrastructure (K)
Treatment Facility (K)
Well System (O&M)
Treatment Facility (O&M)
Soil borings
Monitor wells
Data analysis
Laboratory analysis
Reporting
Consulting fees
Computer time
Construction engineering
Well construction
Materials
Access roads
Power transmission
Piping
Construction engineering
Treatment system
Pump operation (power)
System maintenance
Chemicals
Labor
Power
Plume area and depth
Complexity of hydrogeology
Complexity of hydrogeology
Size of containment system
Plume size
Aquifer flux
Transmissivity
Plume size
Configuration of wells
System discharge
Composition/concentration of recovered water
Aquifer flux/pumping depth
Fluid corrosivity
System discharge
Composition/concentration of recovered water
Monitoring
Sampling
Analysis
Reporting
Complexity of system
*K denotes capital costs; 0 and M denote operation and maintenance costs.
-------
Table VII.12: Example Trade-Off Matrix for Cost-Effectiveness Assessment of Remedial
Actions at Uncontrolled Hazardous Waste Sites (St. Clair, McCloskey, and
Sherman, 1982).
i
ro
oj
I
Alternatives
Cost Ratings
••
9
0 ¦«? £
- c «> n
" 0 £ *¦
1 ;2 . ;
2 « e ® o
o a " £ O
(J O 2 O y\
Level ol Cleanup/
Isolation Achievable
Time to Achieve g)
Cleanup/Isolation •
•
Technology Status jj-
m
Usability ol Land 5
Alter Action
Capability ot Action to <
Minimize Community Impacts 3
During Implementation •
(A
Capability ol Action to _
Minimize Adverse Health & y
Environmental Impacts ¦»
During Implementation {!
•
Risk ol Failure
Impact ot Failure
Site-Specific Effectiveness
Measures
VI
0
— C
* £ 5 •
c 0 iS a
is.™ <•
CO »
¦5 < — E c 2
^ <= » 0 S - •
2°c itS i
£!< 0
- it) 5 £3 5 Ml £
- Q « <•> C
12 r Hj V) O u|
< s a 3 0 < H
1 Effectiveness Ratings
£ Cost Ratings
Weighting Faclois
Encompassing Slurry Walt
with Cap
Downgradienl Slurry Wall
with Extraction Wells.
Tieat wwilti Carbon
Extraction Wells.
Treat with Carbon
Extraction Wells.
Treat with BiOx
-------
develop a modified version of an existing methodology that is applicable
to the problem of concern. The methodology utilized should have the
following basic attributes: (1) it compares alternatives on a common
basis; (2) it possesses some measure of assessing benefits; (3) it is
comprehensive and considers all relevant factors; and (4) its results
are replicable and are not subject to bias.
Environmental Evaluation
Each of the proposed alternatives must also be analyzed for their
potential impact on the environment. There are a variety of approaches
that can be used to assess these impacts, including empirical assessment
methodologies, interaction matrices, network analyses, checklists, and
others (Canter, 1977). The purpose of this section is to identify the
need for environmental impact assessment in aquifer restoration
projects, and to identify some unique issues associated with it.
The first unique issue to be identified in association with ground
water cleanup activities is that some alternatives may represent "an
irretrievable committment of natural resources", i.e. some alternatives
may require that the aquifer remain permanently altered. This is
especially true of technologies such as impermeable barriers (slurry
walls, grouts). Secondly, because the unique aspects of ground water
remediation activities usually lie underground, the above ground
consequences of the remedial action should be relatively easy to assess.
More specifically, ground water remedial actions for the most part, do
not represent large, land intensive undertakings. Depending on the
technology employed, there will exist a potential to affect several
-274-
-------
environmental categories. Examples include air quality, surface water
quality, and noise. There should exist ample information on related
activities so that the assessment of the above-ground impacts of ground
water remedial actions is simple, straightforward and less time-
consuming than the assessment of subsurface effects.
The last issue associated with environmental evaluation is the
importance of public participation. Because of the recent attention
afforded ground water pollution episodes by the media, the public has
become ill-informed and frightened. It is desirable to avoid public
histeria in any project, and this is especially true for aquifer
restoration projects. Because ground water is a "hidden resource",
improving the public understanding may be especially difficult. Careful
planning and increased attention should be applied to the role of public
participation in any aquifer restoration studies. Freudenthal and
Calender (1981) note that public participation programs promote conflict
resolution by providing opportunities for individuals and opposing
groups to explore compromise solutions. Detailed information on public
participation in aquifer restoration decision-making is included in
Appendix E.
Risk Assessment
Cornaby, et al. (1982) stated:
"The paramount concern is that no known general methodology is
available for conducting overall environmental risk assessment,
risk assessment that includes both humans and especially non-human
or ecological receptors. In fact, the terminology in the
literature is not always clear, suggesting that our views and
knowledge of environmental risk assessment are still evolving."
-275-
-------
This statement should serve as adequate notice that assessing the risks
of aquifer restoration alternatives will not be easy. Berger (1982)
states that risk assessment has been defined as, "the identification of
hazards, the allocation of cause, the estimation of probability that
harm will result, and the balancing of harm with benefit". Moreover,
Berger (1982) states that selecting an effective remedial technique
involves the balancing of the need to contain contaminants within
acceptable levels against the costs associated with the cleanup
measures. St. Clair, McCloskey and Sherman (1982) state that "a risk
assessment involves the definition of the risks to the environment and
human health of continued pollution from a site" and "the most
inexpensive remedial action that reduces the risk to an acceptable level
could be considered the most cost-effective". Dawson and Sanning (1982)
note that the objective of a remedial action is to reduce associated
risks to an acceptable level.
The most important question associated with risk assessment is—
"how can the risks of aquifer restoration alternatives be assessed"?
There are two generally accepted approaches to risk assessment. The
first approach is to utilize criteria or standards for a pollutant and
work backwards, utilizing intrinsic properties of the pollutant and
aquifer, to develop a list of possible alternatives. The second
approach is to analyze the effectiveness of various alternatives and
compare their resultant concentrations with a given standard. No matter
which approach is used, a "criteria" or "standard" or "acceptable level"
is involved and these data, for the most part, are non-existent. Dawson
and Sanning (1982) outline a method for setting site restoration
-276-
-------
criteria by using air or water standards and working backwards with data
on dilution potential and distribution characteristics. This
methodology is dependent on existing criteria for other environmental
media and knowledge of transport mechanics, which may not be available.
Several risk assessment techniques have been developed for
assessing the risk of hazardous waste sites. With slight modifications,
some of these would be applicable to assessing different alternatives at
individual sites. Unterberg, Stone and Tafuri (1981) outline such a
procedure for assessing spill sites. Caldwell, Barrett and Chang (1981)
have developed a ranking system for the release of hazardous substances.
Nelson and Young (1981) discuss a location and prioritization scheme for
future investigation of abandoned dump sites. Unites, Possidento and
Housman (1980) describe the development of a site investigation manual
for risk assessment. Kufs, et al. (1980) have developed a methodology
for selecting sites for investigation based on their adverse
environmental impacts. Berger (1982) describes a general methodology
for assessing risks based on a similar approach as Kufs, et al. (1980).
This methodology considers four characteristics: receptors, pathways,
waste characteristics, and waste management practices. Additional
information on risk assessment and risk assessment techniques is in
Appendix F.
Nisbet (1982) discusses some uses and limitations of risk
assessments. Nisbet (1982) and Berger (1982) both point out that
without detailed analytical monitoring programs, risk assessments can
usually only produce a qualitative ranking of alternatives. Nisbet
(1982) also points out that risk assessments are difficult to conduct
-277-
-------
and give uncertain results, because: (1) exposure is usually variable
and poorly characterized; (2) toxicity information is hard to
extrapolate to humans from animals; and (3) ground water transport
mechanics are often unknown, thus making the population at risk
difficult to estimate.
In summary, risk assessment could be the most difficult aspect of
evaluating aquifer restoration alternatives. The problems to be
encountered stem from the fact that most of the needed information for
any comprehensive methodology will not exist. The use of estimations in
any methodology will limit its replicability.
The best approach to risk assessment will be to develop a
methodology suited to the identified need. Modification of existing
risk assessment techniques is one approach. The best risk assessment
technique will be one that best utilizes available information and
requires the least amount of estimated input.
SELECTED REFERENCES
Berger, I.S., "Determination of Risk for Uncontrolled Hazardous Waste
Sites", Management of Uncontrolled Hazardous Waste Sites, Hazardous
Materials Control Research Institute, 1982, pp. 23-26.
Bixler, B., Hanson, B. and Langner, G., "Planning Superfund Remedial
Actions", Management of Uncontrolled Hazardous Waste Sites, Hazardous
Materials Control Research Institute, 1982, pp. 141-145.
Caldwell, S., Barrett, K.W. and Chang, S.S., "Ranking System for
Releases of Hazardous Substances", Management of Uncontrolled Hazardous
Waste Sites, Hazardous Materials Control Research Institute, 1981, pp.
14-20.
Canter, L.W., Environmental Impact Assessment, 1st ed., McGraw-Hill,
1977.
Cornaby, B.W., et al., "Application of Environmental Risk Techniques to
Uncontrolled Hazardous Waste Sites", Management of Uncontrolled
-278-
-------
Hazardous Waste Sites, Hazardous Materials Control Research Institute,
1982, pp. 380-384.
Dawson, G.W. and Brown, S.M., "New Assessment Methods to Aid Site
Restoration", Management of Uncontrolled Hazardous Waste Sites,
Hazardous Materials Control Research Institute, 1981, pp. 79-83.
Dawson, G.W. and Sanning, D., "Exposure=Response Analysis for Setting
Site Restoration Criteria", Management of Uncontrolled Hazardous Waste
Sites, Hazardous Materials Control Research Institute, 1982, pp. 386-
389.
Evans, R.B., Benson, R.C. and Rizzo, J., "Systematic Hazardous Waste
Site Assessments", Management of Uncontrolled Hazardous Waste Sites,
Hazardous Materials Control Research Institute, 1982, pp. 17-22.
Freudenthal, H.D. and Calender, J.A., "Public Involvement in Resolving
Hazardous Waste Site Problems", Management of Uncontrolled Hazardous
Waste Sites, Hazardous Materials Control Research Institute, 1982, pp.
346-349.
Glover, E.W., "Containment of Contaminatd Ground Water: An Overview",
The Proceedings of the Second National Symposium on Aquifer Restoration
and Ground Water Monitoring, 1982, pp. 17-22.
Hittman Associates Inc., "Surface and Ground Water Contamination From
Abandoned Lead-Zinc Mines, Picher Mining District, Ottawa County,
Oklahoma", Oct. 1981.
Housman, J.J., Brandwein, D.I. and Unites, D.F., "Site Contamination and
Liability Audits in the Era of Superfund", Management of Uncontrolled
Hazardous Waste Sites, Hazardous Materials Control Research Institute,
1981, pp. 398-404.
Kaschak, W.M. and Nadeau, D.F., "Remedial Action Master Plans",
Management of Uncontrolled Hazardous Waste Sites, Hazardous Materials
Control Research Institute, 1982, pp. 124-127.
Kufs, C., et al., "Rating the Hazard Potential of Waste Disposal
Facilities", Management of Uncontrolled Hazardous Waste Sites, Hazardous
Materials Control Research Institute, 1980, pp. 30-41.
Lundy, D.A. and Mahan, J.S., "Conceptual Designs and Cost Sensitivities
of Fluid Recovery Systems for Containment of Plumes of Contaminated
Groundwater", Management of Uncontrolled Hazardous Waste Sites,
Hazardous Materials Control Research Institute, 1982, pp. 136-140.
Minning, R.C., "Monitoring Well Design and Installation", The
Proceedings of the Second National Symposium on Aquifer Restoration and
Ground Water Monitoring, 1982, pp. 194-197.
-279-
-------
Neely, N.S., et al., "Remedial Actions at Uncontrolled Hazardous Waste
Sites", EPA 430/9-81-05, Jan. 1981, U.S. Environmental Protection
Agency, Cincinnati, Ohio.
Nelson, A.B. and Young, R.A., "Location and Prioritization of Abandoned
Dump Site9 for Future Investigations", Management of Uncontrolled
Hazardous Waste Sites, Hazardous Materials Control Research Institute,
1981, pp. 52-62.
Nisbet, I.C.T., "Uses and Limitations of Risk Assessments in Decision-
Making on Hazardous Waste Sites", Management of Uncontrolled Hazardous
Waste Sites, Hazardous Materials Control Research Institute, 1982, pp.
406-407.
Rehtlane, E.A., and Patton, F.D., "Multiple Port Piezometers vs.
Standpipe Piezometers: An Economic Comparison" The Proceedings of the
Second National Symposium on Aquifer Restoration and Ground Water
Monitoring, 1982, pp. 287-295.
Rishel, H.L., et al., "Costs of Remedial Response Actions at
Uncontrolled Hazardous Waste Sites", EPA-600/52-82-035, March 1983, U.S.
Environmental Protection Agency, Cincinnati, Ohio.
St. Clair, A.E., McCloskey, M.H. and Sherman, J.S., "Development of a
Framework for Evaluating Cost-Effectiveness of Remedial Actions at
Uncontrolled Hazardous Waste Sites", Management of Uncontrolled
Hazardous Waste Sites, Hazardous Materials Control Research Institute,
1982, pp. 372-376.
Thorsen, J.W., "Technical and Financial Aspects of Closure and Post
Closure Care", Management of Uncontrolled Hazardous Waste Sites,
Hazardous Materials Control Research Institute, 1981, pp. 259-262.
Tolman, A.L., et al., "Guidance Manual for Minimizing Pollution From
Waste Disposal Sites", EPA-600/2-78-142, Aug. 1978, U.S. Environmental
Protection Agency, Cincinnati, Ohio.
Unites, D., Possidento, M. and Housman, J., "Preliminary Risk Evaluation
for Suspected Hazardous Waste Disposal Sites in Connecticut", Management
of Uncontrolled Hazardous Waste Sites, Hazardous Materials Control
Research Institute, 1980, pp. 25-29.
Unterberg, W., Stone, W.L. and Tafuri, A.N., "Rationale for Determining
Priority of Cleanup of Uncontrolled Hazardous Waste Sites", Management
of Uncontrolled Hazardous Waste Sites, Hazardous Materials Control
Research Institute, 1981, pp. 188-197.
Ward, J.R., "Well Design and Construction for In Situ Leach Uranium
Extraction", Proceedings of the Second National Symposium on Aquifer
Restoration and Ground Water Monitoring, 1982, National Water Well
Association, Columbus, Ohio, pp. 205-213.
-280-
-------
SECTION VIII
TECHNIQUES FOR DECISION-MAKING
Numerous decisions are required in the selection of an aquifer
restoration strategy for meeting a given need. Examples of required
decisions include those related to selection of containment options,
ground water treatment processes and design, and site closure options.
Alternatives exist for each required decision, and the selection of the
most appropriate one for meeting a given need can be a difficult task.
The alternative with the greatest potential for meeting a given need
will most likely not be the least environmentally damaging, least
costly, and least risky. Therefore, trade-off analysis typically
includes importance weighting of decision factors and the evaluation of
a set of alternatives relative to each factor. Although the
methodologies outlined in this section have found their greatest
application in environmental impact studies, they can be utilized for
decision-making related to aquifer restoration strategies.
Importance weighting is done in every decision related to the
selection of an aquifer restoration strategy. For example, decisions
are made on the relative importance of existing environmental resources,
the importance/significance of anticipated beneficial and detrimental
environmental impacts, monetary costs, and public acceptability. In
most cases, this weighting is done via informal, ad hoc approaches
without using available formalized techniques to provide a documented
record of the considerations and rationale. Professional judgment is
involved in both ad hoc and formalized approaches.
-281-
-------
The extent to which a set of alternative strategies and
combinations thereof are evaluated relative to a series of decision
factors depends upon the number of strategies being evaluated and the
nature and extent of the study. Thorough studies involving multiple
alternatives (strategies) will tend to give greater emphasis to the
comparisons of alternatives. Comparisons of alternatives are often made
on an ad hoc basis in the absence of quantitative information on the
alternatives, and without using available structured techniques to
provide a documented record of the considerations and rationale. Again,
as was the case for importance weighting, professional judgment is
involved in both ad hoc and structured approaches.
The primary focus of this section will be on the principles and
practice of importance weighting in decisions related to the selection
of an aquifer restoration strategy for meeting a given need. Briefer
information will be presented on techniques for evaluation of
alternatives relative to a series of decision factors. The reason for
this balance is that importance weighting requires more judgment and has
greater influence on the final selection than does the evaluation of
alternatives. This is not to say that the latter is unimportant, but
simply to indicate that more attention should be given to fundamental
importance weighting techniques. This section begins with a conceptual
framework for trade-off analysis and is followed by examples of both
informal and structured approaches for importance weighting. The
ranking/rating/scaling of alternatives are also briefly addressed
followed by information on the development of a decision matrix.
-282-
-------
CONCEPTUAL FRAMEWORK FOR TRADE-OFF ANALYSIS
Trade-off analysis typically involves the comparison of a set of
alternatives relative to a series of decision factors. For example, for
clean-up of a polluted aquifer there may be several treatment processes
which could be used, including both in-situ options and treatment
following pumpage to the surface. Factors related to these alternatives
can include the health risk, economic efficiency, social concerns
(public preference), and bio-physical, cultural and socio-economic
impacts (both beneficial and detrimental impacts should be considered).
Table VIII.1 displays a trade-off matrix for systematically comparing
specific alternatives relative to a series of decision factors. The
following approaches can be used to complete the trade-off matrix:
(1) qualitative approach in which descriptive information on each
alternative relative to each decision factor is presented in
the matrix;
(2) quantitative approach in which quantitative information on
each alternative relative to each decision factor is displayed
in the matrix;
(3) ranking, rating, or scaling approach in which the qualitative
or quantitative information on each alternative is summarized
via the assignment of a rank, or rating, or scale value
relative to each decision factor (the rank, or rating, or
scale value is presented in the matrix);
(4) weighting approach in which the importance weight of each
decision factor relative to each other decision factor is
considered, with the resultant discussion of the information
on each alternative (qualitative; or quantitative; or ranking,
rating, or scaling) being presented in view of the relative
importance of the decision factors; and
(5) weighting-ranking/rating/scaling approach in which the
importance weight for each decision factor is multiplied by
the ranking/rating/scale of each alternative, then the
resulting products for each alternative are summed to develop
an overall composite index or score for each alternative.
-283-
-------
Table VIII. 1: Trade-off Matrix
Alternatives
Decision Factor 1 2 3 A 5
Health Risk
Economic Efficiency
Social Concerns
(public preference)
Environmental Impacts
— Biophysical
— Cultural
— Socio-economic
-284-
-------
Decision-making which involves the comparison of a set of
alternatives relative to a series of decision factors is not unique in
terms of its application to environmentally-oriented problems such as
aquifer restoration. This approach represents a classic decision-making
problem which is often referred to as multi-attribute or multi-criteria
decision making. The conceptual framework for this type of decision-
making appeared in the early 1960's in the U.S. Department of Defense.
Since that time it has been applied to numerous situations requiring
decisions, including those which involve consideration of environmental
factors and impacts.
Two decision factors listed in Table VIII.1 are health risk and
social concerns (public preference). Public participation programs for
determining public preference are highlighted in Appendix E. One
approach addressing health risk involves risk assessment, and Appendix F
provides information on using risk assessment in decisions related to
remedial actions for polluted ground water.
IMPORTANCE WEIGHTING FOR DECISION FACTORS
If the qualitative or quantitative approach is used for completion
of the matrix as shown in Table VIII.1, information for this approach
relative to economic efficiency environmental impacts can be found in
Sections IV through VI. It should be noted that this information would
also be needed for the approaches involving importance weighting and/or
ranking/rating/scaling. If the importance weighting approach is used
the critical issue is the assignment of importance weights to the
individual decision factors, or at least the arrangement of them in a
-285-
-------
rank ordering of importance. Table VIII.2 summarizes some structured
importance weighting or ranking techniques which could be used to
achieve this step. These techniques have been used in numerous
environmental decision-making projects. Brief descriptions of several
reference sources and techniques from Table VIII.2 will be presented
later as illustrations of the techniques. In addition to the structured
techniques, less-formal approaches such as reliance on public
participation programs can also be used as the basis for importance
weighting.
SYSTEMATIC COMPARISON OF STRUCTURED IMPORTANCE WEIGHTING TECHNIQUES
As shown in Table VIII.2 there are a number of importance weighting
techniques which can be used in decision-making related to aquifer
restoration strategies. As would be expected, there are advantages and
limitations to specific structured importance weighting techniques. One
study to compare six techniques was conducted by Eckenrode (1965). To
insure that the research results would be useful in a variety of
situations where comparative judgments are required, three different
judgment situations were used, one in each of three experiments:
(1) A specific air defense system development, where the judges
were six persons who had personally conducted analytical
studies on the system in question and were thoroughly familiar
with it. They judged the relative importance of six carefully
defined system criteria frequently found in military air
defense system specifications: economy, early availability,
lethality, reliability, mobility, and troop safety.
(2) A general (hypothetical) air defense system development, where
the 12 judges used were familiar with the problems of air
defense system design and use in a general way from their
previous experience. The same criteria were judged as in the
specific system experiment.
-286-
-------
Table VII I.2: Importance Weighting Techniques
Reference Source
Importance Weighting Technique
Dalkey (1969)
De lphi
Dean and Nishry (1965)
Paired comparison
Dee, et al. (1972)
Ranked pairviae comparison
Eckenrode (1965)
Ranking
Rating
Paired comparison (3 types)
Successive compariaon
Edward* (1976)
tfcilti-attribute utility aeaaurement
Falk (1968)
Public preference
Finaterbuacb (1977)
Conaensal weights
Form la weights
Justified subjective weights
Subjective weights
Inferred subjective weights
Ranking
Equal weight
ftiltiple Methods
Gun, Roe fa and Kimball (1976)
Metfessel general allocation test
O'Connor (1972)
Haiti-attribute acaling
Ross (1976)
Paired comparison and checking
School of CEES and Oklahoma
Ranked pairwise compariaon
Biological Survey (1974)
Touaaaint (1975)
Delphi
Voelker (1977)
Nominal group proceas
-287-
-------
(3) A personnel selection problem, in which six judges who had
extensive experience in personnel subsystem management judged
the relative value of the following carefully defined
qualifications for a command and control personnel subsystem
manager: previous personnel subsystem experience, technical
competence, and demonstrated capabilities for fiscal planning,
work planning, and maintaining good client relations and good
staff relations.
It is realized that these three experiments do not address aquifer
restoration strategies; however, they can be used generically since they
involve importance weighting of decision factors (criteria). The six
importance weighting techniques which were compared for their
reliability and efficiency in collecting the judgment data were
(Eckenrode, 1965):
(1) Ranking: The judge (J) was asked to place a numerical rank
next to each criterion, indicating by 1 that criterion most
valuable in the situation, by 2, the next most valuable, etc.
(2) Rating: The criteria were presented next to a continuous
scale marked off in units from 0 to 10 (see Figure VIII.1). J
was asked to draw a line from each criterion to any
appropriate point on the value scale. He was permitted to
select points between numbers or to assign more than one
criterion to a single position on the scale.
(3) Partial Paired Comparisons I: The criteria were presented on
the ordinate and abscissa of a partial matrix (see Table
VIII. 3). J was asked to indicate in each block the number of
the more valuable of the pair of criteria which were the
coordinates of that block.
(4) Partial Paired Comparisons II: A list of criterion pairs was
presented in the form illustrated below, and J was asked to
circle the member of each pair which was more valuable for the
system in question.
Troop safety — Reliability
Each criterion was paired once with every other criterion.
(5) Complete Paired Comparisons: This method was the same as
partial paired comparisons II, except that the list was
doubled in length by requiring that each pair appear twice,
-288-
-------
RELIABILITY
LETHALITY
MOBILITY
EARLY AVAILABILITY
ECONOMY
TROOP SAFETY
-10 (MAXIMUM VALUE)
- 9
- 8
- 7
- 6
- 5
- 4
- 3
- 2
- 1
- 0(N0 VALUE)
Note: J draws a line from each criterion to the appropriate
point on the scale.
Figure VIII. 1: Rating Scale (Eckenrode, 1965).
-289-
-------
Table VIII. 3: Partial Paired Comparisons I (Eckerirode, 1965).
TROOP SAFETY
MOBILITY
ECONOMY
RELIABILITY
LETHALITY
Csl
CO
in
\o
1. EARLY AVAILABILITY
2. TROOP SAFETY
3. MOBILITY
4. ECONOMY
5. RELIABILITY
Note: J puts the number of the more valuable criterion of the pair
being considered in each block. Example: 2 in the "early
availability-troop safety" block means J judges the latter to
be more important for the system in question.
-290-
-------
once in the order A-B, the second time (elsewhere in the list)
in the order B-A, in order to eliminate any position error.
(6) Successive Comparisons: In this method the list of criteria
was presented to J who proceeded as follows:
(a) He ranked criteria in order of importance (as in the
ranking method above).
(b) He tentatively assigned the value (Vj) 1.0 to the most
important criterion, and other values (V^) between 0 and
1 to the other criteria in order of importance.
(c) He decided whether the criterion with value 1.0 was more
important than all other criteria combined.
If so, he increased Vj so that was greater than the
sum of all subsequent V's, i.e.,
Vl > E 2 ^
If not, he adjusted Vj (if necessary) so that V^ was less
than the sum of all subsequent V's, i.e.,
Vi < S 2 Vi
(d) He decided whether the second most important criterion
with value V2 was more important than all lower-valued
criteria; he then proceeded as in c above with V£.
(e) He continued until n - 1 criteria had been so evaluated.
All methods except successive comparison were used in all three
experiments. The successive comparisons method was used only in the
first experiment, in which the criteria were applied to a specific
hardware system design. All three experiments were conducted in the
same way. For each, a booklet was developed, the cover page of which
contained instructions to the judge, including a definition of each
criterion to be compared. Each succeeding page contained a form for
recording his judgments by one of the methods. The order of the forms
in the booklet was randomized and each judge was administered a booklet
individually in a quiet office, under instructions to complete it
-291-
-------
rapidly but without making mistakes. The experimenter recorded the time
required by the judge to complete each page of the booklet*
The consistency of importance weighting assignments among the six
techniques was very high in all three experiments. This means that
there is high reliability among the techniques in the importance weights
produced; that is, each method produced essentially the same ordering of
weights. It was also noted that each judge produced the same ordering
of weights regardless of the method he used. This high within-judge
reliability is to be expected when the set of items being judged is
small as in the present experiments and judges behave transitively.
This is the usual case in weighting multiple criteria, for sets of
criteria to be weighted frequently consist of between three and a dozen
items in practical situations. Taken together, this information
indicates that the techniques are reliable approaches for recording
judgments. Any of the six techniques could be used for importance
weighting of decision factors associated with the selection of an
aquifer restoration strategy. The individuals doing the weighting could
be technical specialists or public officials, or both.
RANKING TECHNIQUES FOR IMPORTANCE WEIGHTING
Ranking techniques for importance weighting basically involve the
rank ordering of decision factors in their relative order of importance.
If there are "n" decision factors, rank ordering would involve assigning
1 to the most important factor, 2 to the second-most important factor,
and so forth until "n" is assigned to the least important factor. It
should be noted that the rank order numbers could be reversed; that is,
-292-
-------
"n" could be assigned to the most important factor, "n-1" to the second
most important factor, and so forth until 1 is assigned to the least
important factor. Two illustrations of ranking, techniques will be
briefly described herein: (1) a public preference approach (Falk,
1968); and (2) the use of the Nominal Group Process technique (Voelker,
1977). Both of these techniques could be used for importance weighting
of decision factors associated with the selection of an aquifer
restoration strategy for meeting a given need.
Public Preference Approach
Falk (1968) described the use of public participation in the
assignment of relative importance weights to a series of factors. A
pilot study was described which involved determining the relative
importance of four tangible and five intangible factors, and identifying
a means of applying these measures of importance in selecting the most
acceptable one of three hypothetical roadway solutions. The approach
requires the assumption that frequency of citizen preference for one
factor over another is directly related to importance of that factor.
This approach could be used in the selection of an aquifer restoration
strategy for meeting a given need, and public participation techniques
are described in Appendix E.
Nominal Group Process Technique
The Nominal Group Process technique (NGP), an interactive group
technique, was developed in 1968. It was derived from social-
psychological studies of decision conferences, management-science
studies of aggregating group judgments, and social-work studies of
-293-
-------
problems surrounding citizen participation in program planning. The NGP
has gained wide acceptance in health, social service, education,
industry, and government organizations. For example, Voelker (1977)
described the application of NGP to rank decision factors important in
siting nuclear power plants. Siting decisions are routinely required
for injection wells for treated ground water. The following four steps
are involved in the use of NGP for importance weighting:
(1) Nominal (silent and independent) generation of ideas in
writing by a panel of participants.
(2) Round-robin listing of ideas generated by participants on a
flip chart in a serial discussion.
(3) Discussion of each recorded idea by the group for the purpose
of clarification and evaluations.
(4) Independent voting on priority ideas, with group decision
determined by mathematical rank-ordering.
RATING TECHNIQUES FOR IMPORTANCE WEIGHTING
Rating techniques for importance weighting basically involve the
assignment of importance numbers to a series of decision factors, and
possibly, although not always, their subsequent normalization via a
mathematical procedure. An example of a rating scale approach is shown
in Figure VIII.1 (Eckenrode, 1965). Four additional examples of rating
techniques will be presented herein: (1) the use of a pre-defined
importance scale (Linstone and Turoff, 1975); (2) the use of a group
approach for the assignment of scales of importance (Rau, 1980); (3) the
use of the multi-attribute utility measurement technique (Edwards,
1976); (4) and the use of a computerized version of a multi-attribute
utility measurement technique (Rugg and Feldman, 1982). These four
-294-
-------
racing techniques could be used for importance weighting of decision
factors associated with the selection of an aquifer restoration strategy
for meeting a given need. It should be noted that the public preference
approach (Falk, 1968) and the NGP technique (Voelker, 1977) described in
the previous sub-section could also be used for rating the importance of
decision factors.
Pre-defined Importance Scale
Decision factors can be assigned numerical values based on pre-
defined importance scales. Table VIII.4 delineates five scales with
definitions to be considered in the assignment of numerical values to
decision factors (Linstone and Turoff, 1975). Usage of the pre-defined
scales can aid in systematizing importance weight assignments.
Group Approach Using Scales of Importance
Rau (1980) suggested a group approach for importance weight
assignments, with the approach following a procedure that will produce
reliable results. Because these weights are essentially based on the
judgmental values or attitudes of those surveyed, the selected procedure
must be systematic and must be able to reduce all possible variation.
The group of persons ultimately selected for the weighting should
include a cross section of society such as individuals from governmental
agencies, politicians and decision makers, experts in the field of
hydrogeology and environmental evaluation, representatives from special
interest groups, and members of the general public. Groups of
individuals representing this cross section must be sampled a number of
times to obtain consistent estimates of the weights.
-295-
-------
Table VIII.4: Importance Scale (Linstone and Turoff, 1975).
Scale Reference Definitions
1. Very Important A most relevant point
First order priority
Has direct bearing on major issues
Must be resolved, dealt vith or treated
2o Important Is relevant to the issue
Second order priority
Significant impact but not until other
items are treated
Does not have to be fully resolved
3. Moderately Important May be relevant to the issue
Third order priority
May have Impact
May be a determining factor to major
issue
4. Unimportant Insignificantly relevant
Low priority
Has little impact
Not a determining factor to major issue
5. Most Unimportant
No priority
No relevance
No measurable effect
Should be dropped as
an item to consider
-296-
-------
The procedure suggested by Rau (1980) for determining the relative
importance of each decision factor (environmental impact area in this
example) consists of ranking and pair-wise comparisons. Each individual
is required to rank the impact areas and to compare in pair-wise fashion
the degree of importance of highest rank with the one immediately
following. If this procedure is followed in a systematic way, a weight
will be developed for each area. The procedure is repeated a number of
times for different groups in order to get the desired cross-sectional
population representation and the reliability needed for an importance
weighting. The basic steps for determining the importance weightings
are as follows (Rau, 1980):
(1) Step 1: Select a group of individuals for evaluation and
explain to them in detail the weighting concept and the use of
rankings and weightings.
(2) Step 2: Prepare a table with columns corresponding to the
range of values which can be assigned as a "score of
importance" to each impact area — for example, if five values
are possible, there would be five columns. The rows in the
table would correspond to the impact areas being ranked as to
importance. Table VIII.5 contains an illustration of five
columns for importance weighting (Rau, 1980).
(3) Step 3: Give a copy of the table developed in Step 2 to each
individual evaluator and repeat Steps 4-9 until no further
changes in the table entries are desired.
(4) Step 4: Ask each individual to place an "X", or other
signifying mark, in each column for each impact area. Thus, a
value of importance is assigned to each impact area.
(5) Step 5: Ask all individuals to compare the marked columns on
a pair-wise basis to insure that the impact areas are ordered
on the proper relative basis in their opinion. If not, they
should reassign their scores so as to have the desired
relative ordering of impact areas. (For example, on a scale
from 1 to 10, if a value of 10 has been assigned to impact
area A and it appears that A is twice as important as B,
impact area B should be assigned a value of 5.)
-297-
-------
Table VIII. 5: Example of the Development of Impact Area Importance
Weightings (Rau, 1980)
Impact area
Low Average
importance importance
A 1
12 3 4
High
importance
5 Total
Weighting
Park requirements
School age students
generated
Trips generated
Police protection
Fire protection
Public service costs
Total revenues
Employment (long-
term jobs)
Electricity
consumption
Natural gas
consumption
Solid waste
generated
Sewage discharge
Water consumption
2/43
3/4 3
2/43
4/43
4/43
5/43
5/43
4/43
3/43
3/4 3
2/43
3/4 3
3/4 3
43
1.0
-298-
-------
(6) Step 6: Ask each individual to add the value (or importance
score) selected for each of the impact areas to obtain a
total.
(7) Step 7: The individual should then divide the value selected
for each impact area by the total obtained in Step 6 to
determine the desired weighting for each area.
(8) Step 8: Collect the tables from each individual evaluator and
average the weightings determined for each impact area to
obtain a "group or composite average".
(9) Step 9: Present the averages obtained to the individual
evaluators and ask them to compare the group weightings with
those derived by each of them individually in Step 7.
(10) Step 10: If any one or more individuals desires to change the
assignment of scores based on what the group decided, go to
Step 4 and repeat the entire process. If none desire to
change their scores, stop the experiment, because the impact
area relative weightings of importance will have been derived.
As referred to earlier, Table VIII.5 contains an example of
importance weight assignments to 13 impact areas of interest based on 5
possible importance scores (Rau, 1980). By adding the scores
corresponding to each "X", a total of 43 points is obtained. Dividing
each score by 43, the relative importance weightings given by the last
column in Table VIII.5 are determined.
Multi-Attribute Utility Measurement Technique
Edwards (1976) described the multi-attribute utility measurement
(MAUM) technique for use in decision-making involving different publics.
The MAUM technique can be used to spell out explicitly what the values
of each participant (decision-maker, expert, pressure group, government,
etc.) are, show how much they differ, and in the process can frequently
reduce the extent of such differences. The basic assumption is that the
values of the participants are reflected in the importance weights
assigned to individual factors. It should also be noted that the usage
-299-
-------
of the MAUM technique can be by an individual, a small group of persons,
or multiple publics. The 10 basic steps in the MAUM technique are
(Edwards, 1976):
(1) Step 1: Identify the person or organization whose utilities
are to be maximized. If, as is often the case, several
organizations have stakes and voices in the decision, they
must all be identified (Utilities refer to general goals or
objectives in the terminology used herein).
(2) Step 2: Identify the issue or issues to which the utilities
needed are relevant. (issues refer to needs being addressed.)
(3) Step 3: Identify the entities to be evaluated. Formally,
they are outcomes of possible actions. But in a sense, the
distinction between an outcome and the opportunity for further
actions is usually fictitious. (Entities refer to
alternatives in the terminology used herein; outcomes would
reflect the evaluation of each entity relative to the decision
factors.)
(4) Step 4: Identify the relevant dimensions of value for
evaluation of the entities. As has often been noted, goals
ordinarily come in hierarchies. But it is often practical and
useful to ignore their hierarchical structure, and instead to
specify a simple list of goals that seem important for the
purpose at hand. It is important not to be too expansive at
this stage. The number of relevant dimensions of value should
be modest, for reasons that will be apparent shortly. (it
should be noted that dimensions of value refer to the decision
factors for evaluation of the alternatives.)
(5) Step 5: Rank the dimensions in order of importance. This
ranking job, like Step 4, can be performed either by an
individual or by representatives of conflicting values acting
separately or by those representatives acting as a group.
(6) Step 6: Rate dimensions in importance, preserving ratios. To
do this, start by assigning the least important dimension an
importance of 10. Now consider the next-least-important
dimension. How much more important (if at all) is it than the
least important? Assign it a number that reflects that ratio.
Continue up the list, checking each set of implied ratios as
each new judgment is made. Thus, if a dimension is assigned a
weight of 20, while another is assigned a weight of 80, it
means that the 20 dimension is 1/4 as important as the 80
dimension, and so on. By the time you get to the most
important dimensions, there will be many checks to perform;
-300-
-------
typically, respondents will want to review previous judgments
to make them consistent with present ones.
Step 7: Sum the importance weights, divide each by the sum,
and multiply by 100. This is a purely computational step
which converts importance weights into numbers that,
mathematically, are rather like probabilities. The choice of
a l-to-100 scale is, of course, completely arbitrary.
Step 8: Measure the location of each entity being evaluated
on each dimension. The word "measure" is used rather loosely
here. There are three classes of dimensions: purely
subjective, partly subjective, and purely objective. The
purely subjective dimensions are perhaps the easiest; you
simply get an appropriate expert to estimate the position of
the entity on that dimension on a 0-to-100 scale, where 0 is
defined as the minimum plausible value and 100 is defined as
the maximum plausible value. Note 'minimum and maximum
plausible" rather than "minimum and maximum possible". The
minimum plausible value often is not total absence of the
dimension. A partly subjective dimension is one in which the
units of measurement are objective, but the locations of the
entities must be subjectively estimated. A purely objective
dimension is one that can be measured non-judgmentally, in
objective units, before the decision. For partly or purely
objective dimensions, it is necessary to have the estimators
provide not only values for each entity to be evaluated, but
also minimum and maximum plausible values, in the natural
units of each dimension.
Step 9: Calculate utilities for entities. The equation is:
I
Ui = j wj Uij
where:
Ui ¦ aggregate utility for ith entity (overall evaluation
score, or index, for the ith alternative)
i 3 number of entities (number of alternatives)
wj = normalized importance weight of the jth dimension of
value (importance weight for jth decision factor);
the wj values are the output from Step 7.
j * number of dimensions of value (number of decision
factors).
u£j = scaled position of the ith entity on the jth
dimension (scaled position of ith alternative on jth
-301-
-------
decision factor); the u^j measures are the output
from Step 8.
(10) Step 10: Decide. If a single act is to be chosen, the rule
is simple: maximize u£. j If a subset of i is to be chosen,
then the subset for which i u^ is maximum at best.
Computerized Version of MAUM Technique
A simple computer program, called DECIDE, for use in importance
weighting and decision-making is available for the TRS-80 computer
system (Rugg and Feldman, 1982). The program can aid in decision-making
when the decision involves the selection of one alternative from several
choices. The following information describes how the program can be
used.
(1) The first thing the program does is ask you to categorize the
decision at hand into one of three categories: (1) choosing
an item (or thing), (2) choosing a course of action, or (3)
making a yes or no decision. You simply press 1, 2, or 3
followed by the ENTER key to indicate which type of decision
is facing you. If you are choosing an item, you will be asked
what type of item it is. To illustrate the mechanisms of the
computer program, Table VIII.6 contains a sample run (Rugg and
Feldman, 1982). This illustration could easily be adapted to
decisions related to the selection of an aquifer restoration
strategy for meeting a given need.
(2) If the decision is either of the first two types, you must
next enter a list of all the possibilities under
consideration. A question mark will prompt you for each one.
When the list is complete, type "END" in response to the last
question mark. You must, of course, enter at least two
possibilities. After the list is finished, it will be re-
displayed so that you can verify that it is correct. If not,
you must re-enter it.
(3) Now you must think of the different factors that are important
to you in making your decision. Each factor is to be entered
one at a time with the word "END" used to terminate the list.
When complete, the list will be re-displayed. You must now
decide which single factor is the most important and input its
number. (You can enter 0 if you wish to change the list of
factors.)
-302-
-------
Table VIII.6:Illustration of the Use of DECIDE (Rugg and Feldtnan, 1982)
I CAN HELP YOU MAKE A DECISION. ALL I NEED TO DO IS ASK SOME QUESTIONS
AND THEN ANALYZE YOUR RESPONSES.
WHICH OF THESE BEST DESCRIBES THE DECISION FACING YOU?
1) CHOOSING AN ITEM FROM VARIOUS ALTERNATIVES.
2) CHOOSING A COURSE OF ACTION FROM VARIOUS ALTERNATIVES.
3) DECIDING "YES" OR "NO".
WHICH ONE (1, 2, OR 3)? J.
WHAT TYPE OF ITEM IS IT
? VACATION
I NEED TO HAVE A LIST OF EACH VACATION UNDER CONSIDERATION.
INPUT THEM ONE AT A TIME IN RESPONSE TO EACH QUESTION MARK.
TYPE THE WORD "END" TO INDICATE THAT THE WHOLE LIST HAS BEEN ENTERED.
? CAMPING
? SAFARI
? TRIP TO D.C.
? END
OK. HERE'S YOUR LIST:
1) CAMPING
2) SAFARI
3) TRIP TO D.C.
IS THE LIST CORRECT (Y OR N)? Y
NOW, THINK OF THE FACTORS THAT ARE IMPORTANT IN CHOOSING THE BEST
VACATION.
INPUT THEM ONE AT A TIME IN RESPONSE TO EACH QUESTION MARK.
TYPE THE WORD "END" TO TERMINATE THE LIST.
? RELAXATION
? AFFORDABILITY
? CHANGE OF PACE
? END
-303-
-------
Table VIII.6: (continued)
HERE'S YOUR LIST OF FACTORS:
1) RELAXATION
2) AFFORDABILITY
3) CHANGE OF PACE
DECIDE WHICH FACTOR ON THE LIST IS THE MOST IMPORTANT AND INPUT ITS
NUMBER. (TYPE 0 IF THE LIST NEEDS CHANGING.)
? i
NOW LET'S SUPPOSE WE HAVE A SCALE OF IMPORTANCE RANGING FROM 0-10.
WE'LL GIVE AFFORDABILITY A VALUE OF 10 SINCE AFFORDABILITY WAS RATED THE
MOST IMPORTANT.
ON THIS SCALE, WHAT VALUE OF IMPORTANCE WOULD THE OTHER FACTORS HAVE?
RELAXATION
? 5^5
CHANGE OF PACE
? 9
EACH VACATION MUST NOW BE COMPARED WITH RESPECT TO EACH IMPORTANCE
FACTOR.
WE'LL CONSIDER EACH FACTOR SEPARATELY AND THEN RATE EACH VACATION IN
TERMS OF THAT FACTOR ONLY.
*** (HIT ANY KEY TO CONTINUE)
(A key is pressed)
LET'S GIVE CAMPING A VALUE OF 10 ON EVERY SCALE.
EVERY OTHER VACATION WILL BE ASSIGNED A VALUE HIGHER OR LOWER THAN 10.
THIS VALUE DEPENDS ON HOW MUCH YOU THINK IT IS BETTER OR WORSE THAN
CAMPING.
CONSIDERING ONLY RELAXATION AND ASSIGNING 10 TO CAMPING,
WHAT VALUE WOULD YOU ASSIGN TO
-304-
-------
Table VIII.6: (continued)
SAFARI? 3
TRIP TO D.C.? 9
CONSIDERING ONLY AFFORDABILITY AND ASSIGNING 10 TO CAMPING,
WHAT VALUE WOULD YOU ASSIGN TO
SAFARI? I
TRIP TO D.C.? 8
CONSIDERING ONLY CHANGE OF PACE AND ASSIGNING 10 TO CAMPING,
WHAT VALUE WOULD YOU ASSIGN TO
SAFARI? 60
TRIP TO D.C.? 25
TRIP TO D.C. IS BEST BUT IT'S VERY CLOSE.
HERE'S THE FINAL LIST IN ORDER. TRIP DO D.C. HAS BEEN GIVEN A VALUE OF
100 AND THE OTHERS RATED ACCORDINGLY.
HIT ANY KEY TO SEE THE LIST.
(A key is pressed)
100 TRIP TO D.C.
98.7 CAMPING
78.8 SAFARI
OK
-305-
-------
(4) The program now asks you to rate the importance of each of the
other factors relative to the most important one. This is
done by first assigning a value of 10 to the main factor.
Then you must assign a value from 0-10 to each of the other
factors. These numbers reflect your assessment of each
factor's relative importance as compared to the main one. A
value of 10 means it is just as important; lesser values
indicate how much less importance you place on it.
(5) Now you must rate the decision possibilities (alternatives)
with respect to each of the importance factors. Each
importance factor will be treated separately. Considering
only that importance factor, you must rate each decision
possibility. The program first assigns a value of 10 to one
of the decision possibilities. Then you must assign a
relative number (lower, higher, or equal to 10) to each of the
other decision possibilities.
(6) Armed with all this information, the program will now
determine which choice is best. The various possibilities are
listed in order of ranking. Alongside each one is a relative
rating with the best choice being normalized to a value of
100.
The DECIDE program listing is in Table VIII.7; the program can
currently accept up to ten alternatives and ten decision factors. If
more alternatives are to be evaluated, or if more decision factors are
involved, the value of MD in line 160 in the program listing should be
increased. The main routines in the DECIDE computer program are as
follows (Rugg and Feldman, 1982):
150- 190 Initializes and dimensions variables.
200- 360 Determines category of decision.
400- 490 Gets or sets T$.
500- 810 Gets list of possible alternatives from user.
900-1220 Gets list of importance factors from user.
1300-1490 User rates each importance factor.
1500-1900 User rates the decision alternatives with respect to each
importance factor.
2000-2110 Evaluates the various alternatives.
2200-2270 Sorts alternatives into their relative ranking.
2300-3000 Displays results.
5000-5020 Subroutine to clear screen and display header.
5200 Time wasting subroutine.
-306-
-------
100
110
150
160
170
180
190
200
210
220
230
240
250
260
270
280
290
300
310
320
330
340
350
360
400
410
420
430
440
450
460
470
480
490
500
510
520
530
540
550
560
570
580
590
Table VIII. 7: Program Listing for DECIDE (Rugg and Feldman, 1982)
REM: DECIDE - 16K
REM: (C) 1981, PHIL FELDMAN AND TOM RUGG
CLEAR 500
MD=10
DIM L$(MD),F$(MD),V(MD)
DIM C(MD,MD),D(MD),Z(MD)
E$=,IIEND"
GOSUB 5000
PRINT" I CAN HELP YOU MAKE A"
PRINT"DECISION. ALL I NEED TO DO IS"
PRINT"ASK SOME QUESTIONS AND THEN"
PRINT"ANALYZE YOUR RESPONSES."
FOR J=1 TO 30:PRINT"-";:NEXT
PRINT
PRINT"WHICH OF THESE BEST DESCRIBES"
PRINT"THE DECISION FACING YOU?"
PRINT" 1) CHOOSING AN ITEM FROM"
PRINT" VARIOUS ALTERNATIVES."
PRINT" 2) CHOOSING A COURSE OF ACTION"
PRINT" FROM VARIOUS ALTERNATIVES."
PRINT" 3) DECIDING 'YES' OR 'NO'."
PRINT
INPUT"WHICH ONE (1,2,OR 3)";T
IF T<1_ OR T>3 THEN 200
GOSUB 5000
ON T GOTO 420,440,460
PRINT"WHAT TYPE OF ITEM IS IT"
INPUT T$:GOTO 500
T$="COURSE OF ACTION"
GOTO 500
T$="1YES' OR 1 NO'":NI=2
L$(1)»"DECIDING YES"
L$(2)-"DECIDING NO"
GOTO 900
GOSUB 5000:NI=0
PRINT" I NEED TO HAVE A LIST OF EACH"
PRINT T$;" UNDER"
PRINT"CONSIDERATION.":PRINT
PRINT" INPUT THEM ONE AT A TIME IN"
PRINT"RESPONSE TO EACH QUESTION MARK."
PRINT
PRINT" TYPE THE WORD ,M;E$;,M TO"
PRINT"INDICATE THAT THE WHOLE LIST"
PRINT"HAS BEEN ENTERED.":PRINT
-307-
-------
Table VIII.7: (continued)
600 IF NI MD THEN 620
610 PRINT"-LIST FULL-":GOTO 650
620 NI=NI+lsINPUT L$(NI)
630 IF L$(Nl) E$ THEN 600
640 NI=NI-1
650 IF NI =2 THEN 700
660 PRINT
670 PRINT"YOU NEED AT LEAST 2 CHOICES!"
680 PRINT:PRINT"TRY AGAIN"
690 GOSUB 5200-.GOTO 500
700 GOSUB 5000
710 PRINT"OK. HERE'S YOUR LIST:"
720 PRINT:FOR J=1 TO NI
730 PRINT J;CHR$(8);") ";L$(J)
740 NEXT:PRINT
750 FOR J=1 TO 9:R$=INKEY$:NEXT
760 INPUT"IS THE LIST CORRECT (Y OR N)";R$
770 IF R$="Y" THEN 900
780 IF R$ <>"N" GOTO 700
790 PRINT
800 PRINT"THE LIST MUST BE RE-ENTERED"
810 GOSUB 5200:GOTO 500
900 GOSUB 5000:R$=INKEY$
910 PRINT" NOW, THINK OF THE FACTORS THAT"
920 IF T<3 THEN PRINT"ARE IMPORTANT IN CHOOSING THE"
930 IF T<3 THEN PRINT"BEST ";T$;".11
940 IF T=3 THEN PRINT"ARE IMPORTANT TO YOU IN"
950 IF T=3 THEN PRINT"DECIDING ";T$;"."
960 PRINT:PRINT" INPUT THEM ONE AT A TIME IN"
970 PRINT"RESPONSE TO EACH QUESTION MARK."
980 PRINT:PRINT" TYPE THE WORD ,";E$;"' TO"
990 PRINT"TERMINATE THE LIST."
1000 PRINT:NF=0
1010 IF NF>=MD THEN PRINT"— LIST FULL --"'.PRINT
1020 IF NF >=MD THEN GOTO 1060
1030 NF=NF+1:INPUT F$(NF)
1040 IF F$(NF)<>E$ THEN 1010
1050 NF=NF-1-.PRINT
1060 IF NF <1 THEN PRINT"YOU NEED AT LEAST 1 - REDO IT !"
1070 IF NF <1 THEN GOSUB 5200
1080 IF NF <1 THEN 900
1100 GOSUB 5000
1110 PRINT"HERE'S YOUR LIST OF FACTORS:"
1130 FOR J=1 TO NF
1140 PRINT J;CHR$(B);") ";F$(j)
-308-
-------
Table
1150
1160
1170
1180
1190
1200
1210
1220
1300
1310
1320
1330
1340
1350
1360
1370
1380
1390
1400
1410
1420
1430
1440
1450
1460
1470
1480
1490
1500
1510
1520
1530
1540
1550
1560
1570
1580
1590
1600
1610
1620
1630
1634
1638
VIII. 7: (continued)
NEXT
PRINT"DECIDE WHICH FACTOR ON THE"
PRINT"LIST IS THE MOST IMPORTANT AND"
PRINT"INPUT ITS NUMBER. (TYPE 0 IF"
PRINT"THE LIST NEEDS CHANGING.)"
INPUT A:A=INT(A)
IF A=0 THEN 900
IF A>NF OR A<0 THEN 1100
GOSUB 5000
IF NF=1 THEN 1500
PRINT" NOW LET'S SUPPOSE WE HAVE A"
PRINT"SCALE OF IMPORTANCE RANGING"
PRINT"FROM 0-10. WE'LL GIVE"
PRINT F$(A)A VALUE OF"
PRINT"10 SINCE ";F$(A)
PRINT"WAS RATED THE MOST IMPORTANT."
PRINT:PRINT" ON THIS SCALE, WHAT VALUE OF"
PRINT"IMPORTANCE WOULD THE OTHER"
PRINT"FACTORS HAVE?"
FOR J=1 TO NF
IF J=A THEN 1490
PRINT:PRINT F$(j)
INPUT V(J)
IF V(J)<0 THEN 1480
IF V(J)>10 THEN 1480
GOTO 1490
PRINT" IMPOSSIBLE VALUE - TRY AGAIN":G0T0 1430
NEXT
V(A)=10:Q=0:FOR J=»l TO NF
Q=Q+V( J) : NEXT: FOR J=>1 TO NF
V(J)=»V(J)/Q:NEXT:GOSUB 5000
IF T<>3 THEN PRINT" EACH ";T$;" MUST NOW"
IF T-3 THEN PRINT" DECIDING 'YES' OR DECIDING"
IF T-3 THEN PRINT"'NO' MUST NOW"
PRINT"BE COMPARED WITH RESPECT TO"
PRINT"EACH IMPORTANCE FACTOR."
PRINT"WE*LL CONSIDER EACH FACTOR"
PRINT"SEPARATELY AND THEN RATE"
IF T<>3 THEN PRINT"EACH ";T$;" IN TERMS"
IF T-3 THEN PRINT"DECIDING 'YES' OR DECIDING"
IF T-3 THEN PRINT"'NO' IN TERMS"
PRINT"OF THAT FACTOR ONLY."
PRINT:PRINT"xxx (HIT ANY KEY TO CONTINUE)"
R$=INKEY$:IF R$="" THEN1638
-309-
-------
Table VIII.7: (continued)
1640 PRINT:PRINT" LET'S GIVE ";L$(1)
1650 PRINT"A VALUE OF 10 ON EVERY SCALE."
1660 IF T<>3 THEN PRINT" EVERY OTHER ";T$
1670 IF T=3 THEN PRINT" THEN DECIDING 'NO'"
1680 PRINT"WILL BE ASSIGNED A VALUE HIGHER"
1690 PRINT"0R LOWER THAN 10. THIS VALUE"
1700 PRINT"DEPENDS ON HOW MUCH YOU THINK"
1710 PRINT"IT IS BETTER OR WORSE THAN"
1720 PRINT L$(1);
1800 FOR J=1 TO NF
1810 PRINT" "
1820 PRINT" CONSIDERING ONLY ";F$(j)
1830 PRINT"AND ASSIGNING 10 TO"
1835 PRINT L$(1)
1840 PRINT"WHAT VALUE WOULD YOU ASSIGN TO"
1850 FOR K=2 TO NI
1860 PRINT L$(K);:INPUT C(K,j)
1870 IF C(K,j)>=0 THEN 1900
1880 PRINT" ~ NEGATIVE VALUES ILLEGAL —"
1890 GOTO 1860
1900 NEXT:PRINT: C (1, J) =10:NEXT
2000 FOR J=1 TO NF:Q=0
2010 FOR K=1 TO NI
2020 Q=Q+C(K,J):NEXT
2030 FOR K=1 TO NI
2040 C(K,j)=C(K,j)/Q:NEXT:NEXT
2050 FOR K=1 TO NI:D(K)=0
2060 FOR J=1 TO NF
2070 D(K)»D(K)+C(K,J)*V(J):NEXT
2080 NEXT:MX=0:FOR K=1 TO NI
2090 IF D(K)>MX THEN MX=D(K)
2100 NEXT:FOR K=1 TO NI
2110 D(K)=D(K)*100/MX:NEXT
2200 FOR K=1 TO NI:Z(K)=K:NEXT
2210 NM=NI-l:FOR K=1 TO NI
2220 FOR J=1 TO NM:Nl-Z(j)
2230 N2=Z(J+l)
2240 IF D(Nl)>D(N2) THEN 2260
2250 Z(J+1)=N1:Z(J)=N2
2260 NEXT:NEXT:J1-Z(1):J2=Z(2)
2270 DF=D(J1)-D(J2):G0SUB 5000
2300 PRINT L$(Jl);" IS BEST"
2310 IF DF<5 THEN PRINT"BUT IT'S VERY CLOSE."
2320 IF DF<5 THEN 2380
2330 IF DF<10 THEN PRINT"BUT IT'S FAIRLY CLOSE."
-310-
-------
Table VIII.7: (continued)
2340 IF DF<10 THEN 2380
2350 IF DF<20 THEN PRINT"BY A FAIR AMOUNT."
2360 IF DF<20 THEN 2380
2370 PRINT"QUITE DECISIVELY."
2380 PRINT" HERE'S THE FINAL LIST IN"
2390 PRINT"ORDER. ";L$(J1)
2400 PRINT"HAS BEEN GIVEN A VALUE OF 100"
2410 PRINT"AND THE OTHERS RATED"
2420 PRINT"ACCORDINGLY."
2430 PRINT
2440 PRINT" HIT ANY KEY TO SEE THE LIST."
2450 R$=INKEY$
2460 IF R$="" THEN 2450
2470 PRINT
2400 FOR J=1 TO NI:Q=Z(j)
2490 PRINT D(Q),L$(Q):NEXT
3000 END
5000 FOR J=»l TO 500: NEXT
5010 CLS:PRINT@12,"DECIDE"
5020 PRINT:RETURN
5200 FOR J=1 TO 1500:NEXT:RETURN
-311-
-------
Definitions for the main variables in the DECIDE computer program
are as follows (Rugg and Feldman, 1982):
MD
Maximum number of decision alternatives.
NI
Number of decision alternatives.
NM
NI-1.
L$
String array of the decision alternatives.
NF
Number of importance factors.
F$
String array of the importance factors.
V
Array of the relative values of each importance factor.
A
Index number of most important factor.
C
Array of relative values of each alternative with respect to
each importance factor.
T
Decision category (l=item, 2=course of action, S^yes or no).
T$
String name of decision category.
E$
String to signal the end of an input data list.
J,K
Loop indicesc
R$
User reply string.
Q,N1,N2
Work variables.
D
Array of each alternative's value.
MX
Maximum value of all alternatives.
DF
Rating difference between best two alternatives.
Z
Array of the relative rankings of each alternative.
PAIRED COMPARISON TECHNIQUES FOR IMPORTANCE WEIGHTING
Paired comparison techniques for importance weighting basically
involve a series of comparisons between decision factors, and a
systematic tabulation of the numerical results of the comparisons.
These types of techniques have been extensively used in decision-making
efforts, including numerous examples related to environmental
improvement projects. Presented herein will be two examples of an
unranked paired comparison technique (Dean and Nishry, 1965; Canter,
1983; Ross, 1976; and Hyman, Moreau and Stiftel, 1982); and two examples
of a ranked pairwise comparison technique (Dee, et al., 1972; and School
of Civil Engineering and Environmental Science and Oklahoma Biological
Survey, 1974). Three examples of unranked paired comparison techniques
were discussed earlier in relation to a systematic comparison of
31.2-
-------
structured importance weighting techniques (Eckenrode, 1965). Both
unranked and ranked pairwise comparison techniques can be used for
weighting decision factors associated with the selection of an aquifer
restoration strategy for meeting a given need.
Unranked Paired Comparison Techniques
One of the most useful techniques for importance weighting of a
series of decision factors is the paired comparison technique developed
by Dean and Nishry (1965). This technique, which can be used by an
individual or group, involves the comparison of each decision factor to
each other decision factor in a systematic manner. Suppose that there
are four basic decision factors known as Fl through F4 (Fl could be
health risks, F2 could be economic efficiency, F3 could be social
concern, and F4 could be environmental impacts). The weighting
technique consists of considering each factor relative to every other
factor and assigning to the one of the pair considered to be the most
important a value of 1, and to the lesser important of the pair a value
of 0. The use of this paired comparison technique is shown in Table
VIII.8. It should be noted that the assignment of 0 to a member of a
pair does not denote zero importance; it simply mean that in the pair
considered it is of lesser importance.
A dummy factor, called F5, is also included in Table VIII.8. The
dummy factor is included so as to preclude the net assignment of a value
of 0 to any of the basic factors (Fl through F4) in the process of each
paired comparison. The dummy factor is defined as the least important
in each paired comparison within which it is included. If two factors
-313-
-------
Table VIII.8: Use of Paired-Comparison Technique for Importance
Weight Assignments
Factor Alignment of Weight* Sum F1C
F1
1
1
1
1
4
0.40
F2
0
1
0
1
2
0.20
F3
0
0
0
1
1
0.10
F4
0
1
1
1
3
0.30
F5 (dummy)
0
0
0
0
_0
0
10
1.00
-314-
-------
are considered to be of equal importance, then a value of 0.5 is
assigned to each factor within the pair (Canter, 1983).
Following the assignment of relative importance to each factor
relative to each other factor, with this process only being completed
following several iterations to be sure that each factor is being
considered in a consistent manner with each other factor, the rationale
for each decision should be documented. It cannot be over-stressed that
the most important aspect of using this technique is the careful
delineation of the rationale basic to each 1 and 0 assignment.
Following this documentation then the individual weight assignments are
summed, with the factor importance coefficient (FIC) being equal to the
sum value for an individual factor divided by the sum for all of the
factors. The total of the sum column should equal to (N)(N—1)/2 where N
is equal to the number of factors included in the assignment of weights.
In the example in Table VIII.8, five factors were included, hence the
sum column total should be equal to 10. The total of the FIC column in
Table VIII.8 should equal to 1.00.
The FIC column in Table VIII.8 indicates that Fl is more important
followed by F4, F2, and F3. Whether or not the actual FIC fractions are
used in a trade-off analysis, this paired-comparison approach has
enabled the rank ordering of the four decision factors from most
important to least important. In addition, importance weighting of sub-
factors can also be done by the same method. For example, in Table
VIII.1 environmental impacts also includes biophysical, cultural, and
socio-economic impacts.
-315-
-------
The paired-comparison technique based on Dean and Nishry (1965) has
been used in a number of environmentally-related studies. Examples
include selection of a sanitary landfill site (Morrison, 1974),
selection of a wastewater treatment system (Canter, 1976), presentation
of an environmental impact methodology for water resources projects
(Solomon, et al., 1977), and evaluation of the environmental impacts of
a wastewater treatment system (Canter and Reid, 1977).
Ross (1976) described a method to check for the consistency of
importance weight assignments made via the paired comparison technique.
The paired comparison technique and consistency procedure proposed by
Ross (1976) is as follows:
(1) When using this technique, an individual who has been
requested to assess a number (N) of decision factors is
presented with every possible combination of these factors,
and asked to make judgments as to which of each pair is more
important. His decisions are recorded in a paired-comparison
matrix as shown in Table VIII.9. An entry (Cjj) of "1" in
this matrix denotes that the row decision factor i (row stim-
ulus) was judged as being better, or more desirable, than the
column stimulus j. Once all possible pairs have been compar-
ed, and the decisions recorded in the matrix, the ranking of
stimuli can be readily ascertained by summing the rows of the
matrix. The stimuli are ranked in order of these row sums.
(2) For the information in Table VIII.9, the resultant ordering of
the stimuli is 5, 6, 1, 3, 2, and 4. When the paired-
comparison matrix is permuted according to the ranking
derived, a characteristic pattern appears in which the upper
right portion of the matrix is observed to be composed of l's,
and the lower left portion of O's (see Table VIII.10).
In the foregoing example, the individual making the comparisons,
hereafter termed the judge, has been perfectly consistent in his
judgments. It reveals that he has a clear idea of the stimuli, and that
he has a good decision rule to follow while making the individual paired
comparisons. Such is often not the case. Inconsistent judgments are
-316-
-------
Table VIII.9: Example Comparison Matrix (Ross, 1976)
Stimulus*
1
2
3
4
5
6
Row SlIBI
1
-
1
1
1
0
0
3
2
0
-
0
1
0
0
1
3
0
1
-
1
0
0
2
4
0
0
0
-
0
0
0
5
1
1
1
1
-
1
5
6
1
1
1
1
0
-
4
~Decision factors
-317-
-------
Table VIII.10: Permuted Example Comparison Matrix (Ross, 1976).
Stimulus 5 6 13 2 4 Row Sum
5 -11111 5
6 0-1111 4
1 00-111 3
3 000-11 2
2 0000-1 1
4 00000 - 0
-318-
-------
revealed by the presence of l's below the diagonal in the permuted
matrix. In Table VIII.11, for example, the preference of the judge for
4 over 5 is revealed. It is clearly a case of inconsistent judgment and
may indicate an unclear understanding of the stimuli, or a confused or
poor decision rule. It might also indicate that one attribute of
stimulus 4 was so far superior to that of stimulus 5 that it became the
sole determinant of the choice made in that particular comparison. The
paired-comparison technique permits and identifies inconsistencies that
would be lost in more traditional ranking approaches (Ross, 1976).
Hyman, Moreau and Stiftel (1982) have indicated that substantial
advances were made during the first decade of environmental impact
studies relative to eliciting and displaying expert judgments about the
relative importance of effects. In order to further these advances they
developed a new environmental assessment method known as SAGE. SAGE
stands for Social-judgment capturing — Adaptive — Goals-achievement —
Environmental assessment because it builds on the best characteristics
of several existing methods. The concepts of SAGE can be used in the
selection of an aquifer restoration strategy from a set of alternatives.
SAGE must be understood within an overall framework for assessing
the relative social worth of alternative projects or policies. The
general framework is that of the rational planning model, which is shown
in Figure VIII.2 as consisting of two phases: (1) a design phase; and
(2) an analytical phase. The design phase includes the tasks of (1)
identifying objectives, (2) establishing planning guides and criteria,
and (3) searching for and synthesizing alternative designs. The
analytical phase comprises four tasks: (1) identifying and predicting
-319-
-------
Table VIII.ll: Permuted Example Comparison Matrix (Incon-
sistent) (Ross, 1976).
Stimulus
5
6
1
3
2
4
Row Sun
5
-
1
1
1
1
0
4
6
0
-
1
1
1
1
4
1
0
0
-
1
1
1
3
3
0
0
0
-
1
1
2
2
0
0
0
0
-
1
1
4
1
0
0
0
0
-
1
-320-
-------
j_ , ! Planning Guides J J
-J^ Objectives •- - and Criteria j (
\ • /
^ * -A
• Design of Alternatives i
Basic
Studies
_ J
Pes ign
Phase
I Alternative 1 '
T
» Alternative 2 •
I I
• Alternative n •
r
Prediction of Physical
Outputs and Behavioral Responses
1
¦
Translation of Inputs, Outputs,
and Responses into Beneficial
and Adverse Effects on Objectives
Economic
Efficiency
Environmental
Quality
Other
Objectives
Scaling and Weighting of Effects within
Accounts to Derive Aggregate Indicators
Elicitation of Social Values Attached
to Incommensurate Objectives
Analyt ica I
Phase
Figure VIII.2: Design and Analytical Phases of the SAGE Method
(Hyman, Moreau, and Stiffel, 1982).
-321-
-------
operational measures of the physical, chemical, and biological
attributes of each alternative; (2) scaling or translating those
consequences into accounts of beneficial and adverse effects on each
objective including behavioral consequences; (3) eliciting relative
value weights that affected individuals or groups attach to the
contributions to each objective when they rank alternatives
preferentially; and (4) presenting the results in a form that is useful
to decision makers. This division of the rational planning model is
convenient because SAGE is concerned primarily with the analytical phase
(Hyman, Moreau and Stiftel, 1982).
The basic concept of the SAGE method is that a system of accounts
with characteristic attributes is used as the decision factors. A
relative importance coefficient (RIC) must be derived for each attribute
within each account. The RIC for each attribute is calculated using a
scoring procedure of attributes in an account involving pairwise
comparisons of all combinations (Dean and Nishry, 1965). The procedure
ensures that the sum of the RICs within each account is unity. An
aggregate index of the account score for an objective is obtained by
taking a linear combination of the scaled effects of the attributes
where each attribute is weighted by its RIC. This last step can be
summarized mathematically as:
Xij - £ (RICjk) (Gijk)
where (RlCj^) is the relative importance coefficient for attribute k
under the j*-*1 objective, is the effect of the i1-*1 alternative on
each attribute, and Xjj is the scaled score of the ith alternative on
the jth objective. As noted earlier, the importance weighting approach
-322-
-------
used by Hyman, Moreau and Stiftel (1982) was the same as that used by
Dean and Nishry (1965). Hyman, Moreau and Stiftel (1982) noted that the
approach has some limitations. For example, the largest RIC that can be
assigned to any attribute is (2/n), where n is the number of attributes.
Also, the entire importance weighting process relies heavily upon expert
judgment. Nonetheless, when accompanied by an explanatory text and a
full description of the actual effects, the method can be useful for
summarizing and comparing features of alternatives.
The unique aspect of the SAGE method is that there is weighting of
the accounts relative to each other. For example, Table VIII.1 lists
four major decision factors (health risk, economic efficiency, social
concerns, and environmental impacts) which would be called accounts in
the SAGE method. Attributes within an account listed in Table VIII.1
include biophysical, cultural, and socio-economic impacts. The SAGE
method elicits weights between accounts by a technique based on social
judgment theory (Hyman, Moreau, and Stiftel, 1982). The technique
involves presenting respondents with an identical set of cards. Each
card in the deck contains a verbal description of the effects of an
alternative on each of the decision factors, a9 well as the scaled
numerical scores for all of the accounts. In this Q-sort procedure, the
participants are asked to arrange the cards in order of their
preferences and then to score the alternatives on a scale ranging from 0
to 100.
Next, the analyst infers the weights from these rankings through
the use of regression analysis methods. To do so, it is necessary to
hypothesize a model that relates each participant's rankings of an
-323-
-------
alternative on an interval scale to the contribution of the alternative
to each of the objectives. In mathematical terms,
Yj = f(X£i, Xi2» Xi3» ^i4> ^£n)
where Yi is the score assigned to the i6*1 alternative and the variables
X-q to X£n refer to the contributions of the ifch alternative to the n
objectives. In estimating the model, an error term is added to reflect
inconsistencies in the participant's application of the model and also
failures of the model to replicate a participant's actual weighting
process. The simplest form of the regression model is linear:
Yi = a + b^i! + b2X£2 + . . . bnX£n + e^
where a is a constant, the b variables are weights attached to each of
the n objectives, and e£ is the error term. Given a set of n
alternatives and data on the X variables and for the set of
participants, the analyst can infer the weights and perform an analysis
of variance. If there is reason to believe that the rankings include
nonlinearities or interactions, a more complex, polynomial model can be
used instead.
Ranked Paired Comparison Techniques
The key feature of ranked pairwise comparison techniques relative
to unranked techniques is that an initial ranking of all decision
factors is required. Two examples of the use of ranked pairwise
comparisons in environmental impact studies will be cited. The first
deals with importance weighting for water resources projects (Dee, et
al., 1972), and the second with importance weighting for a waterway
-324-
-------
navigation project (School of Civil Engineering and Environmental
Science and Oklahoma Biological Survey, 1974). These examples could be
adapted for usage in selecting an aquifer restoration strategy.
The general methodology developed for water resources projects is
called the Environmental Evaluation System (EES). The relative
importance of the 78 parameters in the EES was expressed in commensurate
units, called parameter importance units (PIU), by quantifying several
individuals' subjective value judgments. The weighting technique used
by the method developers, Battelle-Columbus, was based on socio-
psychological scaling techniques and the Delphi procedure (Dee, et al.,
1972). The Delphi procedure will be discussed later. The importance
weighting technique in the EES is systematic, minimizes individual bias,
produces consistent comparisons, and aids in the convergence of
judgment.
The ranked pairwise comparison technique used by Battelle-Columbus
is a commonly-used socio-psychological scaling technique. In ranked
pairwise comparison, the list of decision factors (or parameters) to be
compared is ranked according to selected criteria and then successive
pairwise comparisons are made between contiguous parameters to select
for each parameter pair the degree of difference in importance. A
weighted list of the parameters is the output from this procedure. The
initial ranking was made based on considering the following three
criteria relative to each parameter (decision factor):
(1) Inclusiveness of parameter
(2) Reliability of parameter measurements
(3) Sensitivity of parameter to changes in the environment
-325
-------
It should be noted that the 78 parameters are basically
environmental parameters grouped into four categories (ecology,
environmental pollution, aesthetics, and human interest) and 17
quantitative components (species and populations, habitats and
communities, water pollution, air pollution, land pollution, noise
pollution, land, air, water, biota, manufactured objects, composition,
educational/scientific packages, historical packages, cultures,
mood/atmosphere, and life patterns) as shown in Figure VIII.3 (Dee, et
al., 1972). The following 10 steps were used in the importance
weighting technique:
(1) Step 1: Select a group of individuals for conducting the
evaluation and explain to them in detail the weighting concept
and the use of their rankings and weightings.
(2) Step 2: Rank the categories, components, or parameters that
are to be evaluated.
(3) Step 3: Assign a value of 1 to the first category on the
list. Then compare the second category with the first to
determine how much the second is worth compared to the first.
Express this value as a decimal (0 < x <_ 1).
(4) Step 4: Continue with these pairwise comparisons until all in
the list have been evaluated (Compare 3rd with 2nd, 4th with
3rd, etc.)
(5) Step 5: Multiply out percentages and express over a common
denominator, and average overall individuals in the
experiment.
(6) Step 6: In weighting the categories or components, adjust the
decimal values from Step 5 if unequal numbers of parameters
exist in the parameter groups being evaluated. Adjustment is
made by proportioning these decimal values in proportion to
the number of parameters included in that grouping.*
*The hierarchical system shown in Figure VIII.3 has an unequal number of
elements in each grouping. To be mathematically correct all levels of
the EES hierarchy should have an equal number of elements. However, at
the present time, there has not been sufficient knowledge in many of
-326-
-------
ENVIRONMENTAL IMPACTS
Ecology 240
Environmental Pollution 402
Species and Populations
Terrestrial
(14) Browsers and grazers
(14) Crops
(14) Natural vegetation
(141 Pest species
(14) Upland game birds
Aquatic
(14) Commercial fisheries
(14) Natural vegetation
(14) Pest species
(14) Sport fish
(14) Waterfowl
140
Habitats and Communities
Terrestrial
(12) Food web index
(12) Land use
(12) Rare and endangered
species
(14) Species diversity
Aquatic
(12) Food web index
(12) Rare and endangered
species
(12) River characteristics
(14) Species diversity
Ecosystems
Descriptive only
Legend
( ) Parameter Importance Units
~ Total
Water Pollution
(20)
Basin hydrologic loss
(25)
BOO
(31)
Dissolved oxygen
(18)
Fecal coliforms
(22)
Inorganic carbon
(25)
Inorganic nitrogen
(28)
Inorganic phosphate
(16)
Pesticides
(18)
pH
(28)
Stream flow variation
(28)
Temperature
(25)
Total dissolved solids
(141
Toxic substances
(20)
Turbidity
1318
Air Pollution
(5)
(5)
(10)
(12)
(5)
(10)
(5)
Carbon monoxide
Hydrocarbons
Nitrogen oxides
Particulate matter
Photochemical oxidants
Sulfur oxides
Other
52
Land Pollution
(14) Land use
(14) Soil erosion
I 28
Noise Pollution
(4) Noise
pr
Figure VIII.3: Environmental Evaluation System (Dee, et al., 1972)
-327-
-------
Esthetics
153
Human Interest
Land
(6)
Geologic surface material
(16)
Relief and topographic
character
(10)
Width and alignment
(32
Air
(3)
Odor and visual
(2)
Sounds
n~
Water
(10)
Appearance of water
(16)
Land and water interface
(61
Odor and floating
materials
(10)
Water surface area
(10)
Wooded and geologic
shoreline
|52~
Biota
(5)
Animals — domestic
(5)
Animals — wild
(9)
Diversity of vegetation types
(5)
Variety within vegetation
types
124
Man-made Objects
(10) Man-made objects
US
205
Educational/Scientific Packages
(13) Archeological
(13) Ecological
(11) Geological
(11) Hydrological
I 48
Historical Packages
(11) Architecture and styles
(11) Events
(11) Persons
(11) Religions and cultures
(11) "Western Frontier"
rsi"
Cultures
(14) Indians
(7) Other ethnic groups
(7) Religious groups
r
I 28
Mood/Atmosphere
(11) Awe-inspiration
(11) Isolation/solitude
(4) Mystery
(11) "Oneness" with nature
HE
Life Patterns
(13) Employment opportunities
(13) Housing
(11) Social interactions
03
Composition
(15) Composite effect
(15) Unique composition
fio"
Figure VIII.3: (continued)
-327a-
-------
(7) Step 7; Multiply these averages by the number of parameter
importance units to be distributed to the respective grouping.
(8) Step 8: Do Steps 2-7 for all categories, components, and
parameters in the EES*
(9) Step 9: Indicate to the individuals by controlled feedback
the group results of the weighting procedure.
(10) Step 10: Repeat the experiment with the same group of
individuals or another group to increase the reliability of
the results.
The following numerical example illustrates the use of the 10 steps
(Dee, et al., 1972). Consider three components (A, B, C) that have been
selected in Steps 1 and 2, with these components consisting of 8
parameters, 4 in A, 2 in B, and 2 in C.
Step 2 Ranking of component - B, C, A
Steps 3,4 Assigns weights
B = 1
C = 1/2 the importance of B
A = 1/2 the importance of C
Step 5 Multiply out percentages and express over common
denominator. Assume the average values of all
individuals are given below.
these areas to permit an equal number of elements at the same level of
detail. This difference in the number of elements from group to group
must be taken into consideration when assigning parameter importance
units in the ranking and weighting. Therefore, in the ranking and
weighting procedure (Steps 1-5), the researchers were asked to assume an
equal number of elements in the groupings being compared. These value
judgments were then "adjusted" in proportion to the number of elements
in each group. Because the purpose of the weighting procedure was to
assign weights to parameters, if an "adjustment" were not made the
individual parameters grouped under water pollution would not receive
sufficient weight as compared to those under noise because the total
number of units available under water pollution would have to be
distributed among 14 parameters whereas those for noise would all be
assigned to the single noise parameter. For this reason, comparisons
between components and categories should be based on average values of
the PIU for the grouping not the sum of the values.
-328-
-------
B = 1
C = .5
A = .25
1.75
B = 1/1.75 = .57
C - .5/1.75 = .29
A = .25/1.75 = .14
1.00
Step 6 Adjust for unequal number of parameters in each
component.
B = .57 x 1/4 = .14
C =¦ .29 x 1/4 = .07
A = .14 x 1/2 = ^07
.28
Using the new total, the components values are
B = .14/.28 = .50
C =¦ .07/.28 = .25
A » .07/.28 - .25
1.00
and the average values are
B = .50/2 = .25
C - .25/2 = .135
A = .25/4 = .0625
Step 7 Multiply these adjusted values by appropriate PIU, which
is assumed in this case to be 20.
20 x .5 =10
20 x .25 = 5
20 x .25 - 5
Step 8 Continue until reliable estimates are obtained.
In the EES, a total of 1000 PIU's are assigned to the parameters by
first distributing to the 4 categories, then to the 17 quantitative
components, and finally, to the 78 parameters. That is, the
participating group specifies, for example, which is more important,
aesthetics or environmental pollution, and then assigns appropriate
weights. The process is continued until all the units are distributed
-329-
-------
among all parameters. Dee, et al. (1972) noted that instead of using
the initial weight resulting from the importance weighting procedure, an
aggregate weight based on several iterations of the technique is
preferred. After each iteration, the participants are given selected
information about the group weights. This information can include the
group mean and variance, or other pertinent information. In the
weighting procedure employed in developing the EES, the participants'
mean value was given in the feedback stage. All of the weighting and
feedback was performed via formal feedback statements, thereby avoiding
undesirable direct interchange of judgments of the individuals in the
test.
The waterway navigation project included an evaluation of the
environmental impacts of 9 alternatives (8 waterway locational routes
and the no-action alternative) for extending waterway navigation from
Tulsa, Oklahoma to Wichita, Kansas (School of Civil Engineering and
Environmental Science and Oklahoma Biological Survey, 1974). Figure
VIII.4 presents a flow diagram of the environmental analysis approach
used in the study. The multidisciplinary team consisted of an
environmental engineer, botanist, zoologist, planner, and archaeologist.
Six basic environmental categories were selected, and a total of 102
specific parameters were identified and grouped into the categories.
The categories and parameters are listed in Table VIII.12 (School of
Civil Engineering and Environmental Science and Oklahoma Biological
Survey, 1974).
The research team discussed the six categories and agreed to
following rank order of decreasing importance: biology; physical
-330-
-------
ASSEMBLE MULTIDISCIPLINARY TEAM
DECIDE UPON BASIC CATEGORIES
OF ITEMS TO BE CONSIDERED
DECIDE UPON RELATIVE IMPORTANCE
OF EACH OF THE BASIC CATEGORIES
WITHIN EACH CATEGORY, DECIDE UPON
SPECIFIC PARAMETERS TO BE UTILIZED
FOR EACH PARAMETER WITHIN A CATEGORY
DETERMINE ITS RELATIVE IMPORTANCE
OBTAIN COMPARATIVE DATA FOR EACH PARAMETER
FROM ALL ALTERNATIVES
CONSTRUCT A PARAMETER FUNCTION GRAPH
FOR EACH PARAMETER
OBTAIN THE RESULTING ENVIRONMENTAL I
QUALITY SCORES FROM THE PARAMETER !
FUNCTION GRAPH j
Figure VIII.4: Flow Diagram of Environmental Analysis Procedure (School
of Civil Engineering and Environmental Science and
Oklahoma Biological Survey, 1974).
-331-
-------
MULTIPLY EACH PARAMETER ENVIRONMENTAL
QUALITY SCORE TIMES ITS APPROPRIATE
WEIGHT, LETTING THE LATTER VARY t 50%
SUM THE SCORES TIMES WEIGHTS
ITERATING 25 TIMES
OBTAIN A STATISTICAL DISTRIBUTION
AROUND EACH ROUTE SCORE
PRESENT A FINAL SEQUENCE OF
ROUTE SCORES (ALTERNATIVES) DEPICTING
WHICH ARE SIGNIFICANTLY DIFFERENT
Figure VIII.4: (Continued)
-332-
-------
Table VIII.12: Environmental Categories and Parameters for Waterway
Navigation Project (School of Civil Engineering and
Environmental Science and Oklahoma Biological Survey,
1974)
Category
Parameter
Biology
1.
Biological effect of water quality change
2.
Biological effect of salinization change
3.
Biological effect of eutrophication change
4.
Biological implications of siltation
5.
Browser populations (deer)
6.
Change from lotic to lentic habitat
7.
Changes in lowland feeding and breeding habitats
8.
Changes in lowland nesting habitat
9.
Expansion of population ranges
10.
Fishing pressure
11.
Food web index
12.
Grazers (cattle)
13.
Interruption of wildlife refuges
14.
Forest land removed
15.
Cropland removed
16.
Grassland removed
17.
Migration of waterfowl
18.
Miles of new shoreline (channel)
19.
Miles of new shoreline (lake)
20.
Number of oxbow lakes
21.
Pest species
22.
Potential areas for fish and wildlife management
23.
Potential for recreation areas
24.
Potential for sport and commercial fishing
25.
Rare and endangered species (plants and animals)
Physical/Chemical
26.
Recruitment from tributaries
27.
Size of oxbow lakes
28.
Terrestrial ecosystem stability or diversity
29.
Unique habitats
30.
Weedy aquatic vegetation
31.
Weedy terrestrial vegetation
32.
Air pollution from construction
33.
Alteration of waste assimilative capacity
34.
Blasting and drilling (noise and dust)
35.
Change in appearance of water
36.
Change in salt load carried by stream
37.
Change in sediment load carried by stream
38.
Commitment of resources
-333-
-------
Table VIII.12: (Continued)
39.
Degree of change in natural surface drainage
40.
Depth of cuts and effect on natural setting
41.
Flow modification and reduction in flood damage
42.
Increase in evaporation
43.
Miles of new stream
44.
Number of acres of land lost
45.
Number and type of new bridges
46.
Number of lakes created
47.
Number of locks and route distance (travel time
in route)
48.
Number of recreational areas developed and the
resulting pollution
49.
Noise from operation
50.
Other noise from construction
51.
Pollution from shipping
52.
Potential for affecting municipal pollution
53.
Potential for shipping accident
54.
Relocations (utilities, railroads, highways)
55.
Size of work force and effects of temporary camps
56.
Susceptibility of cuts to erosion
57.
Total quantity of excavation
58.
Water and air pollution resulting from industri-
alization
Regional
Compatibility
59.
Accessibility to first order transportation
(non-rail)
60.
Accessibility to railroad transportation
61.
Accessibility to second order transportation
62.
Conformance to existing and developing urban
patterns
63.
Construction-intermediate urban to site distance
(impact of construction phase)
64.
Construction-major urban to site distance (impact
of construction phase)
65.
Construction-minor urban to site distance (impact
of construction phase)
66.
Proximity to urban centers (intermediate)
67.
Proximity to urban centers (major)
68.
Proximity to urban centers (minor)
69.
Severance of surface transport routes
Archeology
70.
Age or occupational period of the site
71.
Concern by local population
72.
Cost of conducting on foot archaeological survey
73.
Depth of the occupational area
74.
Ecological setting
-334-
-------
Table VIII.12: (Continued)
75.
Eligibility of the site for state or national
register
76.
Estimated number of sites to be obtained by the
survey
77.
Importance in terms of local, state, and national
level
78.
Minimum salvage costs for estimated sites
79.
Nature of the site
80.
Number of known sites to be damaged
81.
Presence of single or multiple occupations
82.
Preservation of archaeological data
83.
Previous knowledge of the area
84.
Site frequency within the area concerned
85.
Site importance in terms of geographic area and
problems
86.
Site preservation or damages
87.
Size of occupational area
88.
Value of site for non-archaeological fields
Aesthetics
89.
Atmospheric turbidity
90.
Confinement of view
91.
Degree of alteration of natural setting
92.
Degree of urbanization
93.
Diversity of vegetation types
94.
Landform diversity
95.
Man-made visual obstructions
96.
River pattern (channelization)
97.
Turbidity
98.
Wildlife diversity
Climatology
99.
Increase in fog frequency
100.
Changes in microclimates (i 1 mile)
101.
Evaporative potential
102.
Inadvertent weather modification
-335-
-------
chemical, regional compatibility, archaeology, aesthetics, and
climatology. A total of 1000 importance points (or weights) were
distributed by the ranked pairwise comparison technique as used in the
EES (Dee, et al., 1972). In moat cases within a category, the
parameters were classified as to high, medium, or low importance, and
the allotted points for that particular category were distributed
accordingly. High importance parameters were allotted three times as
many points as low importance parameters, and medium importance
parameters were allotted twice as many points as the low importance
parameters.
Evaluation of the 9 alternatives in the waterway navigation project
relative to the 78 parameters was made via the use of parameter function
graphs yielding environmental quality (EQ) scores. Development and
usage of the graphs will be discussed later. The overall impact
evaluation was made via the development of Route Scores for each
alternative, with the Route Scores defined as follows (School of Civil
Engineering and Environmental Science and Oklahoma Biological Survey,
1974):
102
Route Scorej =» ^ (EQ)ij IW£
where:
EQij = environmental quality for jth alternative relative to ith
parameter.
IWj = importance weight of ith parameter.
Since the importance weights for the 6 categories and 102
parameters involved considerable subjective judgment, the weight term in
the above Route Score equation was allowed to vary randomly (+ 50Z) in
-336-
-------
calculating Route Scores. A computer program was used to calculate an
average route score following 25 iterations of the above equation.
DELPHI TECHNIQUE FOR IMPORTANCE WEIGHTING
The Delphi technique i9 a structured approach for achieving group
concensus about common issues of concern. It was used in a modified
form in the development of importance weights in the EES described
earlier (Dee, et al., 1972). The Delphi technique has been compared to
other group communication techniques as shown in Table VIII.13 (Linstone
and Turoff, 1975). Conventional Delphi is the method discussed herein,
while real time Delphi, that is, computer conferencing, refers to a
variation of conventional Delphi. The technique has broad versatility
and the results are easy to understand. Application of the method
involves a few days to as much as three months which makes the cost low
to medium compared to other methods. The expertise required to
administer the study ranges from very little to average and the
assistance required ranges from none to highly sophisticated computer
data analysis (Stanford Research Institute, 1975). The Delphi technique
could be used to weight decision factors associated with the selection
of an aquifer restoration strategy for meeting a given need.
Key elements in the successful conduction of a Delphi study are the
study director and the chosen panel of experts. The panel of experts
chosen may come from academic, governmental, consulting or industrial
backgrounds (Toussaint, 1975; Martino, 1972). There are three other
types of panelists: stakeholders (those directly affected), experts who
can supply experience and special application techniques, and
-337-
-------
Table VIII.13: Group Communication Techniques (Linstone and Turoff, 1975)
Conference
Telephone Call
Committee Meeting
Formal Conference
or Seminar
Conventional
Delphi
Real-Time
DelmM
Effective Croup
Site
Small
Small to medium
Small to large
Small to large
Small to large
Occurrence of
Interaction by
Individual
Coincident with
group
Coincident with
group
Coincident with
group
Random
Random
Length of
interaction
Short
Medium to long
Long
Short to medium
Short
Hunker of
Interaction*
Multiple, aa
required by group
Multiple, necessary Single
time delays
between
Multiple, necessary
time delays
between
MiltlplC, as
required by
Individual
Normal Mode
Range
Equality to
chairman control
(flexible)
Equality to
chairman control
(flexible)
Presentation
(directed)
Equality to
monitor control
(structured)
Equality to
monitor control
or group control
and no monitor
(structured)
Vriivclpsl Costa
Comunlcatlons
-Travel
-Individuals' time
-Travel
-Individuals' time
-Fees
-Monitor time
-Clerical
-Secretarial
-Communications
-Computer usage
Time-urgent
considerations
Forced delays
Forced delays
Time-urgent
considerations
Other Charac-
terlatlcs
-Equal flow of information to and from
all
-Can maximize psychological effects
-Efficient flow of
information from
few to many
-Equal flow of information to and
from all
-Can minimise psychological effects
-Can minimize time demanded of
respondents or conferees
-------
facilitators who can clarify, organize or stimulate (Linstone and
Turoff, 1975). The director nominates outside experts based on a
combination of their performance, years of experience, number of
publications, and status with peers (Helmer, 1966). Expertise can also
be assessed by contrasting it with public views of whomever is willing
to buy and pay for each individual opinion. Those selected may be asked
to nominate other panel members. The availability and willingness to
serve for the entire study is an important consideration. The number of
experts chosen is usually 7 to 10 for environmental issues, although
more panelists are used in forecasting social impacts (Toussaint, 1975).
Studies have shown that when employing a group to arrive at a decision,
the average group error decreases with an increase in the number of
individuals. As the panel increases in size there is an increase in
reliability up to 15 panel members (Gordon and Helmer, 1964).
Several rounds of questionnaires are used in the conventional
Delphi technique. The questionnaires may be prepared before or during
the selection of panel members. Once the questionnaires have been
designed and the panel selected, the next step is the administration of
the Delphi sequence which involves the mailing of questionnaires and
their return to the director. The director compiles and edits the
information received for successive rounds.
If the Delphi technique is being used for importance weighting, the
objective of round one is to get an initial estimate of the ranking and
relative importance weights of the decision factors. The objective
could be to identify potential decision factors given information about
the environmental setting and potential project. The information
-339-
-------
generated in the first round must be carefully reviewed by the director
and duplications eliminated. The director determines the median (point
of concensus) and the spread of opinion which is expressed by the
interquartile range (IQR) encompassing 25% of the responses above and
below the median. The panelists receive this written feedback through
the mail which eliminates the noise of an open discussion and the
dominance of certain individuals in the panel.
In round two panelists weigh the strength of their own convictions
against the group median and may even revise the ranking and relative
importance of their round one estimates. Experts who give responses
outside the IQR must write down their reasons. On the average, the
questionnaire and written material for the second round will be five to
ten times that of the first round (Linstone and Turoff, 1975). The
developments are reordered by using a combination of probability,
desirability, or feasibility scales. In many cases it may be desirable
to keep track of certain subgroups making up the respondent group as a
whole. This provides a mechanism to decide whether polarized views
reflect the affilitations or the background of the respondents. The
director compiles round two estimates and presents the new median, IQR,
and reasons for their distribution in round three.
Round three is conducted similar to round two and again experts
write down their reasons for disagreeing with estimates outside the new
IQR. The initial disagreera and new ones have a chance to convince
other panelists to change their estimates. The responses are summarized
by median, IQR, disagreement reasons and counter arguments. If it is
the last round, the director summarizes majority and minority opinions
-340-
-------
on projections for each development along with the above information
before a re-vote is taken. If possible, the re-vote should be put off
until a fourth round when everyone can see the additional remarks. In
the fourth round a re-vote is taken and the director compiles the new
median, IQR and reasons for consensus for each issue being addressed.
Delphi studies could utilize two to five rounds, but it has been found
that responses tend to stabilize after three rounds (Helmer, 1966).
Four rounds may be necessary to give minority opinions a chance at
shifting the median and IQR (Toussaint, 1975).
A number of major criticisms have been made against usage of the
Delphi technique. Helmer (1966) identified the following criticisms:
(1) There is instability in the panel memberships. Many studies
have shown that only a few of the panelists respond to
questionnaires, many drop out on succeeding rounds, and
convergence is impeded by too many panel substitutions.
(2) A big time lapse between succeeding rounds can cause shifting
of opinions.
(3) Ambiguous questions confuse panelists.
(4) Panelists' competence is questionable. Helmer (1966) suggests
asking experts questions only in their particular field and to
leave blanks when unsure of their own judgment.
On the positive side, there are several things which can be done to
improve the results of the use of the Delphi technique for importance
weighting. Martino (1972) recommended the following eight steps to
insure the smooth conduction and reliability of a Delphi study:
(1) Obtain agreement of experts to serve on a panel.
(2) Explain the Delphi procedure thoroughly.
(3) Avoid compound events in one analysis. If the event statement
contains one part the panelist agrees with, and another part
he disagrees with, it is difficult for him to know how to
-341-
-------
answer. Avoid compound questions such as: Capability A will
be achieved through the use of Technology B in the year .
The panelist may find two distinct parts for a single event,
and this is where feedback is valuable to help the director
improve his questions.
(4) Avoid ambiguous statements of events, and the careless use of
technical jargon. Avoid using the terms "everybody knows",
"common", "widely used", "normal", and "in general use".
(5) Make questions easy to answer such as filling in the blank, or
multiple choice. The arguments for and against each event
should be summarized so the panelists can connect them to
specific questions.
(6) Ask no more than 25 to 40 questions in any one Delphi round.
(7) The director should never inject his own opinion(s). If the
director becomes convinced that the panelists are overlooking
some significant element, he should recognize that he has
picked an unqualified panel and should repeat the work with
another panel.
(8) The director should be able to estimate the work load involved
for each panel member in a Delphi sequence. As a planning
factor, the director should allow two professional man-hours
per panelist per questionnaire. When using 100 or more
panelists it is advisable to use a computer to aid in the data
analysis.
Linstone and Turoff (1975) have made other procedural
recommendations which are stated below:
(1) Lay out the expected processing of the data throughout all the
rounds of the Delphi before you finalize the design. You may
be forced to later modify the procedure, but the process of
planning ahead will usually turn up any large problems in your
initial questionnaire design.
(2) Design the handling of data so that each response can be
processed as it comes in. By doing this you will not have a
frantic rush to analyze all the responses at once when the
last return comes in.
(3) At least two professionals should work on monitoring a Delphi
exercise, especially when the abstracting of comments is a
good portion of the exercise. With two individuals one can
always review what the other has done.
-342-
-------
(4) Pretest your questionnaire on a group before using it for the
rounds sequence.
(5) If you are covering a number of fields of expertise, make sure
each field is adequately represented in your group.
(6) The criteria for retaining an item for further evaluation
should be made clear at the beginning of the Delphi sequence.
(7) Interpersonal techniques such as interviews and seminars
should be interspersed with the rounds of questionnaires and
information feedback.
(8) When editing respondents comments, try to preserve the intent
of the originator. When editing from round to round, avoid
changing a statement so that it has one meaning in one round
and another meaning in another round.
(9) Standardized measures should be available to a respondent so
he can self-rate his competence or familiarity to specific
questions.
(10) If a multidisciplinary approach is desired respondents should
be encouraged to consider all items, but to make estimates
only on those items with which he feels comfortable.
Respondents may indicate their familiarity with a specialized
area or the importance of an item, without making probability
estimates.
(11) The source of a suggested item should be identified taking
care not to compromise the anonymity of specific inputs. Keep
track of how different subgroups in your respondent group vote
on specific items. This can be very useful in analyzing the
results and will produce situations where you want the
respondents to know the existence of polarizations or
differences based on particular backgrounds.
To serve as an example for an environmentally-related study,
Toussaint (1975) used the Delphi technique to develop importance weights
for 14 water pollution parameters related to the proposed Aubrey
Reservoir project in northern Texas. Eighteen environmental experts
from the region established regional values for the two judgment-based
decisions currently used in environmental impact assessment. These two
decisions were referred to as the weighting (determination of the
-343-
-------
relative emphasis or degree of importance each parameter is to be
assigned in the assessment) and scaling (determination of the magnitude
of effect resulting from a change in a parameter measure) process. The
Delphi procedure was used to reach a consensus of opinion.
SCALING/RATING/RANKING OF ALTERNATIVES
Scaling/rating/ranking of each alternative for each decision factor
is the second major aspect in the use of the multiple-criteria decision-
making approach. Several different techniques have been used for this
evaluation of alternatives in a decision. Examples of techniques which
have potential applicability to decisions related to the selection of an
aquifer restoration strategy for meeting a given need include the use of
(1) the alternative profile concept, (2) a reference alternative, (3)
linear scaling based on the maximum change, (4) letter or number
assignments designating impact categories, (5) evaluation guidelines,
(6) functional curves, or (7) the paired-comparison technique. Examples
of each of these techniques are as follows:
(1) Bishop, et al., (1970) contains information on the alternative
profile concept for impact scaling. This concept is
represented by a graphical presentation of the effects of each
alternative relative to each decision factor. Each profile
scale is expressed on a percentage basis ranging from a
negative to a positive 100%, with 100% being the maximum
absolute value of the impact measure adopted for each decision
factor. The impact measure represents the maximum change,
either plus or minus, associated with a given alternative
being evaluated. If the decision factors are displayed along
with the impact scale from +100% to -100%, a dotted line can
be used to connect the plotted points for each alternative and
thus describe its "profile". The alternative profile concept
is useful for visually displaying the relative impacts of a
series of alternatives.
(2) Salomon (1974) describes a scaling technique for evaluation of
cooling system alternatives for nuclear power plants. To
-344-
-------
determine scale values, a reference cooling system was used
and each alternative system compared to it. The following
scale values were assigned to the alternatives based on the
reference alternative: very superior (+8), superior (+4),
moderately superior (+2), marginally superior (+1), no
difference (0), marginally inferior (-1), moderately inferior
(-2), inferior (-4), and very inferior (-8).
(3) Odum, et al., (1971) utilized a scaling technique in which the
actual measures of the decision factor for each alternative
plan are normalized and expressed as a decimal of the largest
measure for that factor. This represents linear scaling based
on the maximum change.
(4) A letter scaling system is used in Voorhees and Associates
(1975). This methodology incorporates 80 environmental
factors oriented to the types of projects conducted by the
Department of Housing and Urban Development. The scaling
system consists of the assignment of a letter grade from A+ to
C- for the impacts, with A+ representing a major beneficial
impact and C- an undesirable detrimental change.
(5) Duke, et al., (1977) describe a scaling checklist for the EQ
account for water resources projects. Scaling is accomplished
following the establishment of an evaluation guideline for
each environmental factor. An evaluation guideline is defined
as the smallest change in the highest existing quality in the
region that would be considered significant. For example,
assuming that the highest existing quality for dissolved
oxygen in a region is 8 mg/1, if a reduction of 1.5 tng/1 is
considered as significant, then the evaluation guideline is
1.5 rag/1 irrespective of the existing quality in a given
regional stream. Scaling is accomplished by quantifying the
impact of each alternative relative to each environmental
factor, and if the net change is less than the evaluation
guideline it is insignificant. If the net change is greater
and moves the environmental factor toward its highest quality,
then it is considered to be a beneficial impact; the reverse
exists for those impacts that move the measure of the
environmental factor away from its highest existing quality.
(6) A paired-comparison technique can also be used for assigning
scale values to alternatives based on their impact on
environmental factors. The paired comparison technique for
accomplishing scaling is described in Dean and Nishry (1965).
(7) Functional curves can also be used to accomplish impact
scaling for environmental factors. The functional curve is
used to relate the objective evaluation of an environmental
factor to a subjective judgment regarding its quality based on
a range from high to low quality (Dee, et al., 1972).
-345-
-------
Paired Comparison Technique for Scaling/Rating/Ranking
One of the most useful techniques for 9caling/rating/ranking of
alternatives relative to each decision factor is the paired-comparison
technique described by Dean and Nishry (1965). This technique was also
described earlier relative to its use for importance weighting of
decision factors. Again, this technique can be used by an individual or
group for the scaling/rating/ranking of alternatives. Suppose that the
decision to be made involves four decision factors and that the
importance weights have been assigned as shown in Table VIII.8.
Furthermore, suppose that there are three alternatives (Al, A2, and A3)
to be evaluated relative to the four decision factors, and that Table
VIII.14 contains the relevant qualitative and quantitative information.
The scaling/rating/ranking technique consists of considering each
alternative relative to every other alternative and assigning to the one
of the pair considered to be the most desirable a value of 1, and to the
lesser desirable of the pair a value of 0. The use of this paired
comparison technique for the three basic alternatives and four basic
decision factors is shown in Tables VIII.15 through VIII.18,
respectively. It should be noted that the assignment of 0 to a member
of a pair does not denote zero desirability; it simply means that in the
pair considered it is of lesser desirability.
A dummy alternative, called A4, is also included in Tables VIII.15
through VIII.18. The dummy alternative is included so as to preclude
the net assignment of a value of 0 to any of the basic alternatives (Al
through A3) in the process of each paired comparison. The dummy
alternative is defined as the least desirable in each paired comparison
-346-
-------
Table VIII.14: Information for Trade-off Analysis
Alternatives
Decision
Factor
A1
A2
A3
F1
Has lowest
health risk
Has highest
health risk
Has medium
degree of
health risk
F2
Medium economic
efficiency
Low economic
efficiency
High economic
efficiency
F3
Undesirable social
impacts expected
No social
impacts expected
Beneficial social
impacts expected
F4
Decrease overall
environmental
quality by 20%
Increase overall
environmental
quality by 10%
Increase overall
environmental
quality by 10%
-347-
-------
Table VIII,15: Scaling/Rating/Ranking of Alternatives Relative to F1
Alternative
Aaeignment of Desirability
gum
ACC
Al
1 1 1
3
0.50
~2
0 0 1
1
0.17
A3
0 1 1
2
0.33
AA (dummy)
0 0 0
0
0
6
1.00
-348-
-------
Table VIII.16: Scaling/Rating/Ranking of Alternatives Relative
to F2
Alternative Assignment of Desirability Sum ACC
Al
1 0 1
2
0.33
A2
0 0 1
1
0.17
A3
1 1 1
3
0.50
A4 (dummy)
0 0 0
0
0
6
1.00
-349-
-------
Table VIII.17: Scaling/Ratlng/RAnking of Alternatives Relative
to F3
Alternative
Assignment
of Desirability
Sun
ACC
A1
0 0
1
1
0.17
A2
1
0
1
2
0.33
A3
1
1
1
3
0.50
A4 (dummy)
0
0 0
0
0
6
1.00
-350-
-------
Table VIII.18: Scaling/Rating/Ranking of Alternatives Relative
to FA
Alternative Assignment of Desirability Sum ACC
Al
0 0 1
1
0.17
A2
1 0.5 1
2.5
0.415
A3
1 0.5 1
2.5
0.415
A4 (dutnny)
0 0 0
0
6.0
0
1.00
-351-
-------
within which it is included. If two alternatives have the same
desirability relative to a decision factor (one is not more desirable
than another), then a value of 0.5 is assigned to each of the pair.
Following the assignment of the relative desirability of each
alternative relative to each other alternative, with this process based
on the qualitative and quantitative information in Table VIII.14, then
individual choice assignments are summed, with the alternative choice
coefficient (ACC) being equal to the sum value for an individual
alternative divided by the sum for all of the alternatives. The total
of the sum column in Tables VIII.15 through VIII.18 should equal to
(M)(M-l)/2 where M is equal to the number of alternatives included in
the assignments. In this example, four alternatives were included,
hence the sum column total in Tables VIII.15 through VIII.18 should be
equal to 6. The total of the ACC column should equal to 1.00.
The ACC column in Table VIII.15 indicates that alternative A1 is
the most desirable relative to decision factor Fl, and is followed by A3
and A2. Similar types of comments could be made for the ACC values in
Tables VIII.16 through VIII.18. Whether or not the actual ACC fractions
are used in a trade-off analysis, this paired-comparison approach has
enabled the rank ordering of the desirability of each alternative
relative to each decision factor. As noted earlier, the paired-
comparison technique based on Dean and Nishry (1965), including the
development of both FIC and ACC values, has been used in a number of
environmentally-related studies (Morrison, 1974; Canter, 1976; Solomon,
et al., 1977; and Canter and Reid, 1977). This technique could be
-352-
-------
easily applied to decision-making needs associated with the selection of
an aquifer restoration strategy for meeting a given need.
Functional Curves for Scaling/Rating/Ranking
Functional curves, also called value functions and parameter
function graphs, have been used in a number of environmental impact
studies for scaling/rating/ranking the impacts of alternatives relative
to a series of decision factors. They have potential for application in
aquifer restoration programs. Dee, et al. (1972) described the following
seven steps which could be used in developing a functional curve
(relationship) for an environmental parameter (decision factor):
(1) Step 1: Obtain scientific information when available on the
relationship between the parameter and the quality of the
environment. Also, obtain experts in the field to develop the
value functions.
(2) Step 2: Order the parameter scale so that the lowest value of
the parameter is zero and it increases in the positive
directions — no negative values.
(3) Step 3: Divide the quality scale (0-1) into equal intervals
and express the relationship between an interval and the
parameter. Continue this procedure until a curve exists.
(4) Step 4: Average the curves over all experts in the experiment
to obtain a group curve. (For parameters based solely on
judgment, value functions should be determined by a
representative population cross section.)
(5) Step 5: Indicate to the experts doing the value function
estimation the group curve and expected results of using the
curves in the EES. Decide if a modification is desired; if
needed go to Step 3, if not continue.
(6) Step 6: Do Steps 1-5 until a curve exists for all parameters.
(7) Step 7: Repeat experiment with the same group or another
group of persons to increase the reliability of the functions.
-353-
-------
The same basic approach as used by Dee, et al. (1972) was also
described by Rau (1980). Parameter function graphs were also developed
and used in the environmental impact study for the waterway navigation
project (School of Civil Engineering and Environmental Science and
Oklahoma Biological Survey, 1974). Examples of two functional curves
are shown in Figures VIII.5 and VIII.6 (Dee, et al., 1972). Usage of
the curves would involve entering the x-axis with extant or predicted
information and reading the resultant y-axis environmental quality
scale.
DEVELOPMENT OF DECISION MATRIX
The final step in multiple-criteria decision-making is to develop a
decision matrix displaying the products of the importance weights (or
ranks) and the alternative scales (or ranks). To complete the example
using the paired-comparison technique (Canter, 1983), Table VIII.19
summarizes the FIC values for the 4 decision factors (from Table
VIII.8), and the ACC values for the 3 alternatives (from Tables VIII.15
through VIII.18). The final product matrix is shown in Table VIII.20.
Summation of the products for each alternative indicates that
alternative A3 would be the best choice followed by Al and A2. The
bases for the differences in the three alternatives are indicated by the
fractions shown in Table VIII.20.
Rau (1980) provided an illustration of using importance weighting
and impact rating to develop a total impact evaluation for four
alternatives. As described earlier, importance weighting involved a
group approach using scales of importance. Table VIII.5 contains an
-354-
-------
IjO
OJ
t
1 a*
04
in
¦
10
a
4
No. vacm/1000 ind»««du»li
Figure VIII.5: Functional Curve for Species Diversity (Dee, et al., 1972).
-355-
-------
Figure VIII.6: Functional Curve for Dissolved Oxygen in
Water (Dee, et al., 1972).
-356-
-------
example of the use of this approach (Rau, 1980). Impact rating was used
based on pre-defined values as shown in Table VIII.21 (Rau, 1980). The
decision matrix is developed by multiplying each alternatives' impact
ratings by the importance weights from each of the decision factors.
In the Battelle EES an overall environmental impact index is
calculated based on the following two steps (Dee, et al., 1972):
(1) Obtain parameter data without the project for each of the 78
environmental factors. Convert these parameter data into EQ
scale values for each of the 78 parameters. Multiply these
scale values by the PIU for each of the individual parameters
to develop a composite score for the environment without the
project.
(2) For each alternative predict the change in the environmental
parameters. Utilizing predicted changes in the parameter
values, determine the environmental quality scale for each
parameter and each alternative. Multiply the environmental
quality values for each alternative by each PIU, and aggregate
the information for a total composite score.
The overall impact evaluations approach used in the waterway
navigation project was comprised of two parts (School of Civil
Engineering and Environmental Science and Oklahoma Biological Survey,
1974). The first part, called Optimum Pathway Matrix Analysis (OPMA),
uses a large number of data items from each route to derive a single
value for each route on an arbitrary scale of 0 (low environmental
quality) to 1000 (high environmental quality). The second part used is
a multidimensional ordination of routes based on principal components
analysis of parameter values (Jeffers, 1967). The ordination uses the
data to construct a multidimensional ordering of the routes based on
their similarity to each other rather than on some externally determined
scale.
-357-
-------
Table VIII.19: FIC and ACC Values for Example Decision Problems
ACC Value¦
Decision FIC
Factor* Values A1 A2 A3
F1 0.40 0.50 0.17 0.33
F2 0.20 0.33 0.17 0.50
F3 0.10 0.17 0.33 0.50
F4 0.30 0.17 0.415 0.415
-358-
-------
Table VIII.20: Product Matrix for Trade-off Analysis for Exam-
ple Decision Problem
FIC x ACC
Decision Factor
A1
A2
A3
F1
F2
F3
F4
0.200
0.066
0.017
0.051
0.068
0.034
0.033
0.124
0.132
0.100
0.050
0.124
0.334
0.259
0.406
-359-
-------
Table VIII. 21*. Basis for Impact Rating (Rau, L980) „
Impact on Present Condition Value
> 100% increase
+7
50-99.9% increase
+5
25-49.9% increase
+ 3
0-24.9% increase
+1
No change
0
0-24.9% decrease
-1
25-49.9% decrease
-3
50-99.9% decrease
-5
> 100% decrease
-7
-360-
-------
In the OPMA approach information from the importance weighting of
102 parameters was coupled with impact scaling information for the 102
parameters based on the use of parameter function graphs. As described
earlier, Route Scores were developed for the 9 alternatives. The OPMA
procedure was converted to a Fortran computer program which builds an n
X m matrix consisting of the scaled parameter values for n routes with m
components for each route. Each component vector is then multiplied by
its weight and each route vector is summed, giving a score for each
route. Since the weights were subjectively determined, each route score
was calculated 25 times with a different random error between 50%
introduced each time an importance weight was multiplied by a scaled
parameter value. The random error was modified by a parabolic
transformation of its frequency distribution such that the probability
of an error greater than 45% is 0.02, while the probability of an error
less than 5% is 0.2. Introduction of random errors and calculating each
route score 25 times allows the calculation of confidence intervals and
the use of statistical tests to determine whether differences between
routes are significant.
The computer output for OPMA consisted of a li9t of scores for each
route with the mean and confidence interal for that route, plus the
results of a Student's T evaluation of the differences between routes,
and a ranked list of the route scores. OPMA was done on each of the
major categories (Biology, Physical/Chemical, Regional Compatibility,
Archaeology, Aesthetics and Climatology) and on the total set of
parameter values. In addition, OPMA was performed separately using
parameters of high, medium and low importance. Finally, OPMA was done
-361-
-------
combining selected categories: Biology and Physical/Chemical; Biology,
Physical/Chemical, Regional Compatibility, and Archaeology; and Biology,
Physical/Chemical, Regional Compatibility, Archaeology and Aesthetics
(School of Civil Engineering and Environmental Science and Oklahoma
Biological Survey, 1974).
In the second part of the overall analysis of waterway navigation
routes, a non-polar principal components-based ordination was
constructed for the routes by the method of Jeffers (1967). In this
case (1) the data consisted of the matrix of scaled parameter values;
(2) the basic data were not transformed; (3) a correlation matrix was
calculated; and (4) parameters with scaled eigenvector values having
absolute values greater than 0.7 were used in the calculation of route
positions. A standard library program (FACTO) was used to calculate the
initial principal components analysis. Each eigenvector from the
principal components analysis was used to calculate route positions on
one axis of the ordination. Positions of the routes on an ordination
axis were calculated by (1) scaling the elements of the corresponding
eigenvector to a maximum value of +1.0; (2) deleting all component
vectors having scaled eigenvectors with absolute values less than 0.7;
(3) multiplying each of the remaining component vectors by its
eigenvector; and (4) summing the route vectors to produce the route
positions on the axis. The result of the ordination procedure is a
multidimensional graphical representation of similarity between the
routes (School of Civil Engineering and Environmental Science and
Oklahoma Biological Survey, 1974).
-362-
-------
The concepts of a weighting-scaling/rating/ranking approach as
demonstrated in these example studies have been used in a number of
other studies involving decisions which included environmental matters.
Table VIII.22 contains a summary listing of these examples along with
some other weighting-scaling checklists which have been used in
environmental decision-making and impact assessment studies (Canter,
1979). The general steps associated with this type of approach for
doing a trade-off analysis are outlined in Table VIII.23. These steps
could be used to aid in decision-making associated with the selection of
an aquifer restoration strategy for meeting a given need.
SELECTED REFERENCES
Bishop, A.B., et al., "Socio-Econoraic and Community Factors in Planning
Urban Freeways", Sept. 1970, Department of Civil Engineering, Stanford
University, Menlo Park, California.
Canter, L.W., "Supplement to Environmental Impact Assessment, Terrebonne
Regional Sewerage Facilities", Aug. 1976, Report submitted to GST
Engineers, Houma, Louisiana.
Canter, L.W. and Reid, G.W., "Environmental Factors Affecting Treatment
Process Selection", Paper presented at Oklahoma Water Pollution Control
Federation Annual Meetin, 1977, Stillwater, Oklahoma.
Canter, L.W., Water Resources Assessment - Methodology and Technology
Sourcebook, 1979, Ann Arbor Science, Ann Arbor, Michigan.
Canter, L.W., "Evaluation of Social and Environmental Impacts of
Emerging Technologies in Agricultural Production", Oct. 1983, University
of Oklahoma, Norman, Okalhoma.
Crawford, A.B., "Impact Analysis Using Differential Weighted Evaluation
Criteria", 1973, in J.L. Cochrane and M. Zeleny, editors, Multiple
Criteria Decision Making, University of South Carolina Press, Columbia,
South Carolina.
Dalkey, N.C., "The Delphi Method: An Experimental Study of Group
Opinion", Memorandum RM-5888-PR, June 1969, The Rand Corporation, Santa
Monica, California.
-363-
-------
Table VIII.22: Summary of Some Weighting-Scaling/Rating/Ranking
Methods for Trade-off Analyses
Author(s)
Salient Feature*
Canter (1976)
Canter and Re id (1977)
Crawford (1973)
Dee, et al. (1972)
Dee, et al. (1973)
Dunne (1977)
Gann (1975)
Lower Mississippi Valley
Division (1976)
Horrison (1974)
Oduo, et al. (1971)
Paul (1977)
Weighting-scaling checklist using weighted
rankings technique is described.
A weighting-scaling technique for evaluating
the environmental impact of wastewater
treatment process is described.
Weighting-scaling checklist for evaluation of
impacts of alternatives.
Weighting-scaling checklist with a good
listing of biological, physical-chemical,
aesthetic and cultural variables.
Weighting-scaling checklist based on relevant
matrices and networks. Ranges of scale
values. Concepts of "environmental assess-
ment trees" to account for interrelationships
among environmental factors.
Weighting-scaling checklist for sanitary
landfill site selection.
Weighting-sea ling checklist using paired
comparison technique is described.
Weighting-scaling checklist using habitat
approach is presented.
Weighting-scaling checklist using paired
comparison technique is described.
This weighting-scaling checklist includes an
error term to allow for mis judgment in the
assignment of importance weights. Computer-
isation of the methodology enables the
conduction of a sensitivity analysis.
Weighting-scaling checklist used to
prioritise potential projects.
-364-
-------
Table VIII.22: Continued
Author(a)
Salient Feature•
Raines (1972)
Weighting-scaling checklist; use of
functional curves to translate environmental
impact quantities into environmental cost
units.
Salomon (1974)
Weighting-scaling checklist; relative scaling
based on a reference alternative is used.
School of Civil Engineer-
ing and Environmental
Science and Oklahoma
Biological Survey (1974)
Smith (1974)
Toussaint (1975)
Wenger and Rhyner (1972)
Weighting-scaling checklist, which is similar
in concept to Dee, et al. (1972); an error
team is included to account for subjective
aisjudgments.
Weighting-scaling checklist for a rapid
transit system.
Weighting and scaling of 14 water pollution
parameters was established by the Delphi
procedure using two separate groups of 9
experts.
A stochastic computer procedure is used to
account for uncertainty in the weighting and
acaling checklist procedure for evaluation of
aolid waste system alternatives.
-365-
-------
Table VIII.23: Delineation of a Methodology for Trade-off Analysis
Involving Environmental Impacts
Elenent Delineation
A. Establish Interdis-
ciplinary Team
B. Select Decision
Factors and Assemble
Basic Information
1. Selection
(a) Select ¦embers of interdisciplinary
team.
(b) Designate team leader.
2. Review and Familiarization
(a) Review information on potential tech-
nologies.
(b) Visit locations vith technologies
being applied.
1. Selection
(a) Assemble preliminary list of decision
factors.
(b) Use technical questions and findings
from A.2, along with professional
judgment, to select additional re-
levant factors.
(c) Identify any resulting interactive or
cross-impact factors or categories.
2. Environmental Inventory
(a) Assemble extant baseline data for
•elected factors.
(b) Identify factors vith data de-
ficiencies, and plan data collection
effort.
(c) Conduct field studies or assemble
information on data-deficient factors.
-366-
-------
Table VIII.23: Continued
Element Delineation
C. Evaluate Alternativea 1. Prediction and Delineation
Relative to Decision
Factor* (a) Predict changes in each factor for
each alternative uaing available
techniques and/or professional judg-
ment .
(b) Delineate potential impacta of alter-
natives.
(c) Highlight significant impacts and "red
flag" any critical issues.
2. Weighting and Scaling
(a) Uae paired comparison technique, or
some other importance weighting
technique, to determine importance
coefficients for each factor (F1C).
(b) Scale/rate/rank predicted impacts
through development of alternative
choice coefficients, or use of some
other technique for evaluation of
alternatives relative to decision
factora (ACC).
3. Evaluation and Interpretation of Results
(a) Multiply FIC by ACC to obtain final
coefficient matrix. Sum coefficient
values for each alternative.
(b) Use values in final coefficient matrix
•s baaia for description of impacts of
alternatives and trade-offs between
alternat ives.
(c) Discuss any critical issues and
predicted impacts.
D. Document Results 1. Rationale
(a) Describe rationale for selection of
decision factors.
-3*7-
-------
Table VIII.23: Continued
Element Delineation
(b) Describe procedure for impact
identification and prediction, and
rationale for weighting, scaling/rat-
ing /ranking and interpreting results.
2. Provide Referencing of Sources of
Information
-368-
-------
Dean,B.V. and Nishry, J.J., "Scoring and Profitability Models for
Evaluating and Selecting Engineering Products", Journal Operations
Research Society of America, Vol. 13, No. 4, July-Aug. 1965, pp. 550-
569.
Dee, N., et al., "Environmental Evaluation System for Water Resources
Planning", Final Report, 1972, Battelle-Columbus Laboratories, Columbus,
Ohio.
Dee, N., et al., "Planning Methodology for Water Quality Management:
Environmental Evaluation System", July 1973, Battelle-Columbus
Laboratories, Columbus, Ohio.
Duke, K.M., et al., "Environmental Quality Assessment in Multi-objective
Planning", Nov. 1977, Final Report to U.S. Bureau of Reclamation,
Denver, Colorado.
Dunne, N.G., "Successful Sanitary Landfill Siting: County of San
Bernardino, California", SW-617, 1977, U.S. Environmental Protection
Agency, Cincinnati, Ohio.
Eckenrode, R.T., "Weighting Multiple Criteria", Management Science, Vol.
12, No. 3, Nov. 1965, pp. 180-192.
Edwards, W., "How to Use Multi-attribute Utility Measurement for Social
Decision Making", SSRI Research Report 76-3, Aug. 1976, Social Science
Research Institute, University of Southern California, Los Angeles,
California.
Falk, E.L., "Measurement of Community Values: The Spokane Experiment",
Highway Research Record, 1968, No. 229, pp. 53-64.
Finsterbusch, K., "Methods for Evaluating Non-Market Impacts in Policy
Decisions with Special Reference to Water Resources Development
Projects", IWR Contract Report 77-78, NOv. 1977, U.S. Army Engineer
Institute for Water Resources, Fort Belvoir, Virginia.
Gann, D.A., "Thermal Reduction of Municipal Solid Waste", Master's
Thesis, 1975, School of Civil Engineering and Environmental Science,
University of Oklahoma, Norman, Okalhoma.
Gordon, T.J. and Helmer, 0., "Report on a Long Range Forecasting Study",
P-2982, Sept. 1964, Rand Corporation, Santa Monica, California.
Gum, R.L., Roefs, T.G. and Kimball, D.B., "Quantifying Societal Goals:
Development of a Weighting Methodology", Water Resources Research, Vol.
12, No. 4, Aug. 1976, pp. 617-622.
Helmer, 0., Social Technology, 1966, Basic Books, New York, New York.
-369-
-------
Hyman, E.L., Moreau, D.H. and Stiftel, B., "SAGE: A New Participant-
Value Method for Environmental Assessment", Feb. 1982, Environment and
Policy Institute, East-West Center, Honolulu, Hawaii.
vJeffers, J., "Two Case Studies in the Application of Principal Component
Analysis", Journal of Applied Statistics, Vol. 16, 1967, pp. 225-236.
Lower Mississippi Valley Division, "A Tentative Habitat Evaluation
System (HES) for Water Resources Planning", Nov. 1976, U.S. Army Corps
of Engineering, Waterways Experiment Station, Vicksburg, Mississippi.
Martino, J.P., Technological Forecasting for Decision Making, 1972,
American Elsevier, New York, New York.
Morrison, T.H., "Sanitary Landfill Site Selection by the Weighted
Rankings Method", Master's Thesis, 1974, School of Civil Engineering and
Environmental Science, University of Oklahoma, Norman, Oklahoma.
O'Connor, M.F., "The Application of Multi-Attribute Scaling Procedures
to the Development of Indices of Value", June 1972, Engineering
Psychology Laboratory, University of Michigan, Ann Arbor, Michigan.
Odum, E.P., et al., "Optimum Pathway Matrix Analysis Approach to the
Environmental Decision Making Process — Test Case: ; Relative Impact of
Preposed Highway Alternates", 1971, Institute of Ecology, University of
Georgia, Athens, Georgia.
Paul, B.W., "Subjective Prioritization of Energy Development Proposals
Using Alternative Scenarios", (Paper presented at the Joint National
ORSA/TIMS Meeting, San Francisco, California, May 1977), Engineering and
Research Center, U.S. Bureau of Reclamation, Denver, Colorado.
Raines, G., "Environmental Impact Assessment of New Installations",
(Paper presented at International Pollution Engineering Congress,
Cleveland, Ohio, Dec. 4-6, 1972), Battelle Memorial Institute, Columbus,
Ohio.
Rau, J.G., "Summarization of Environmental Impact", in Environmental
Impact Analysis Handbook, Rau, J.G. and Wooten, D.C., editors, 1980,
McGraw-Hill Book Company, Inc., New York, New York, pp. 8-17 to 8-25.
Ross, J.H., "The Numeric Weighting of Environmental Interactions",
Occasional Paper No. 10, July 1976, Lands Directorate, Environment
Canada, Ottawa, Canada.
Rugg, T. and Feldman, P., "TRS-80 Color Computer Programs", 1982,
Dilithium Press, Beaverton, Oregon, pp. 25-36.
-370-
-------
Salomon, S.N., "Cost-Benefit Methodology for the Selection of a Nuclear
Power Plant Cooling System", Paper presented at the Energy Forum, 1974
Spring Meeting of the American Physical Society, Washington, D.C., Apr.
22, 1974.
School of Civil Engineering and Environmental Science and Oklahoma
Biological survey, "Mid-Arkansas River Basin Study — Effects Assessment
of Alternative Navigation Routes from Tulsa, Oklahoma to Vicinity of
Wichita, Kansas", June 1974, University of Oklahoma, Norman, Oklahoma.
Smith, M.A., "Field Test of an Environmental Impact Assessment
Methodology", Report ERC-1574, August 1974, Environraetnal Resources
Center, Georgia Institute of technology, Atlanta, Georgia.
Solomon, R.C., et al., "Water Resources Assessment Methodology (WRAM):
Impact Assessment and Alternatives Evaluation", Report 77-1, 1977, U.S.
Army Engineer Waterways Experiment Station, Vicksburg, Mississippi.
Stanford Research Institute, "Handbook of Forecasting Techniques", IWR
Contract Report 75-7, Dec. 1975, A report submitted to U.S. Army
Engineers Institute for Water Resources, Menlo Park, California.
Toussaint, C.R., "A Method for the Determination of Regional Values
Associated with the Assessment of Environmental Impacts", Ph.D.
Dissertation, 1975, School of Civil Engineering and Environmental
Science, University of Oklahoma, Norman, Oklahoma.
Voelker, A.H., "Power Plant Siting, An Application of the Nominal Group
Process Technique", ORNL/NUREG/TM-81, Feb. 1977, Oak Ridge National
Laboratory, Oak Ridge, Tennessee.
Voorhees, A.M. and Associates, "Interim Guide for Environmental
Assessment: HUD Field Office Edition", June 1975, Washington, D.C,
Wenger, R.B. and Rhyner, C.R., "Evaluation of Alternatives for Solid
Waste Systems", Journal of Environmental Systems, Vol. 2, No. 2, June
1972, pp. 89-108.
-371-
------- |