c/EPA
United States
Environmental Protection
Age
Environmental Monitoring
and Support Laboratory
P.O. Box 15027
Las Vegas NV 89114
EPA-600 7-79-218
September 1979
Research and Development
Geothermal Environmental
Impact Assessment
Ground Water Monitoring
Guidelines for
Geothermal Development
Interagency
Energy-Environment
Research
and Development
Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad categories
were established to facilitate further development and application of environmental
technology. Elimination of traditional grouping was consciously planned to foster
technology transfer and a maximum interface in related fields. The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9, Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY—ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the effort
funded under the 17-agency Federal Energy/Environment Research and Development
Program. These studies relate to ERA'S mission to protect the public health and welfare
from adverse effects of pollutants associated with energy systems. The goal of the Pro-
gram is to assure the rapid development of domestic energy supplies in an environ-
mentally-compatible manner by providing the necessary environmental data and
control technology. Investigations include analyses of the transport of energy-related
pollutants and their health and ecological effects; assessments of, and development of,
control technologies for energy systems; and integrated assessments of a wide range
of energy-related environmental issues.
This document is available to the public through the National Technical Information
Service. Springfield, Virginia 22161
-------
EPA-600/7-79-218
September 1979
GEOTHERMAL ENVIRONMENTAL IMPACT ASSESSMENT
Ground Water Monitoring Guidelines
for Geothermal Development
by
Richard B. Weiss, Theodora O. Coffey,
Tamara L. Williams
Harding-Lawson Associates
San Rafael, California 94902
Contract No. 68-03-2668
Project Officers
Donald B. Gilmore
Environmental Monitoring and Support Laboratory
Las Vegas, Nevada 89114
and
Robert P. Hartley
Industrial Environmental Research Laboratory
Cincinnati, Ohio 54268
ENVIRONMENTAL MONITORING AND SUPPORT LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
LAS VEGAS, NEVADA 89114
-------
DISCLAIMER
This report has been reviewed by the Environmental Monitor-
ing and Support Laboratory, Las Vegas, U.S. Environmental
Protection Agency, and approved for publication. Approval does
not signify that the contents necessarily reflect the views and
policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute
endorsement or recommendation for use.
-------
FOREWORD
Protection of the environment requires effective regulatory actions
that are based on sound technical and scientific information. This
information must include the quantitative description and linking of
pollutant sources, transport mechanisms, interactions and resulting
effects on man and his environment. Because of the complexities in-
volved, assessment of specific pollutants in the environment requires
a total systems approach that transcends the media of air, water, and
land. The Environmental Monitoring and Support Laboratory, Las Vegas,
contributes to the formation and enhancement of a sound monitoring data
base for exposure assessment through programs designed to:
• develop and optimize systems and strategies
for monitoring pollutants and their impact
on the environment
• demonstrate new monitoring systems and tech-
nologies by applying them to fulfil special
monitoring needs of the Agency's operating
programs
This report discusses all the aspects of potential groundwater
pollution from geothermal resource development, conversion, and waste
disposal and provides guidelines for developing a groundwater monitoring
strategy for any such area. It is written for all levels of industry
and government where responsibility lies in the areas of groundwater
management.
For further information on this subject contact the Monitoring
Systems Research and Development Division, Environmental Monitoring
and Support Laboratory, Las Vegas, Nevada.
+ ' Geqrge B. Moreen
DJ.rector^
Environmental Monitoring and Support Laboratory
Las Vegas
ill
-------
ABSTRACT
The proposed ground water monitoring methodology for
geothermal development specifies a six-step planning and evalua-
tion procedure as follows: (1) define baseline conditions and
project development; (2) forecast aquifer conditions; (3) define
limits of detection; (4) evaluate ground water and disposal
facility monitoring techniques; (5) design monitoring plan and
alternatives; (6) implement monitoring plan. A philosophy of
prevention is emphasized as an integral, natural and essential
component of the monitoring methodology.
Natural geothermal processes and reservoir development are
discussed from the point of view of their effects on ground
water systems. A liquid-dominated field will produce a far
greater volume of effluent than a vapor-dominated field. Most
geothermal reservoirs exhibit fracture porosity. They may
consist of many separate pockets of fluids at various depths
and with varying degrees of hydraulic connection. The rela-
tionship between shallow ground water, deep ground water and
the geothermal system must be known in order to assess potential
degradation.
Borehole logging techniques provide continuous and detailed
vertical profiles of rock and fluid properties. These can be
used for ground water monitoring in geothermal areas to define
baseline conditions, construct wells and monitor observation
and injection wells. Well logging techniques cannot be directly
applied to the fractured, crystalline environment typical of
geothermal reservoirs since the standard calibration and
interpretation methods have been developed for the intergranular,
sedimentary petroleum environment.
Currently, the most effective disposal method for liquid
waste is injecting it into subsurface strata. Injection is also
beneficial in that it helps prevent subsidence and facilitates
reservoir recharge. Before injection of liquid wastes is
initiated, geologic, hydrologic and reservoir engineering evalua-
tions must be conducted. Potential problems may be chemical
(e.g., scaling, formation plugging, corrosion) or physical
(.induced seismicity, hydrofracturing, thermal stress) . Actual
injection experience is summarized, along with costs and areas
of the technology needing research and development.
IV
-------
CONTENTS
Disclaimer ii
Foreword iii
Abstract iv
Figures ix
Tables xii
SUMMARY 1
SECTION 1 - INTRODUCTION 7
1.1 Report Organization and Preparation 8
1.2 Related Studies 9
1.3 Ground Water Monitoring Philosophies 11
1.4 Terminology 14
References 17
SECTION 2 - GEOTHERMAL PROCESSES 18
2.1 Hot Water Systems 19
2.2 Vapor-Dominated Systems 20
2.3 Geothermal System Models 20
2.3.1 Basic Model 21
2.3.2 Conceptual Models for Two Types of
Geothermal Reservoirs 25
2.3.3 Salton Sea Geothermal Field Model 28
2.4 Geothermal Reservoir Engineering 30
2.4.1 Hypothetical Reservoir Models 32
2.5 The Character of the Geothermal Effluent . 38
2.5.1 Chemistry 38
2.5.2 Temperature 40
2.5.3 Volume of Fluid 43
-------
2.6 Resource Recovery 43
References 45
SECTION 3 - THE MONITORING METHODOLOGY 48
3.1 Define Baseline Conditions and Projected Development . 49
3.1.1 Chemical Characteristics of Waters 50
3.1.2 Geology and Hydrology 66
3.1.3 Well and Fluid Discharge Data 69
3.1.4 The Geothermal System 71
3.1.5 Projected Development 74
3.2 Forecast Aquifer Conditions 75
3.2.1 System Models 76
3.2.2 Potential Pollutant Mechanisms and Pathways . . 77
3.2.3 Chemical Reactions in the Aquifer 82
3.2.4 Effects of Alternative Development Scenarios . 83
3.3 Define Limits of Detection 83
3.3.1 Chemical Detection Sensitivity 84
3.3.2 Temporal and Spatial Sensitivity 86
3.4 Evaluate Monitoring Techniques 86
3.4.1 Monitoring Wells 87
3.4.2 Fluid Sampling and Chemical Analysis 88
3.4.3 Well Logging 88
3.4.4 Injection-Well Monitoring Techniques 89
3.4.5 Other Monitoring Techniques 90
3.5 Design Monitoring Plan and Alternatives 93
3.5.1 Spatial Distribution of Sample Points
and Sampling Frequency 93
3.5.2 Applicable Monitoring Techniques 97
3.5.3 Chemical and Physical Parameters 98
3.5.4 Regulatory Specifications 99
3.5.5 Cost Versus Confidence 103
3.6 Implement Monitoring Plan 105
3.6.1 Data Collection 105
3.6.2 Data Synthesis, Display and Interpretation . . 106
3.6.3 Review and Modify Monitoring Plan 108
References 110
VI
-------
SECTION 4 - APPLICATION OF WELL LOGGING TO GROUND WATER
MONITORING IN GEOTHERMAL AREAS 115
4.1 Subsurface Data Collection 115
4.1.1 Well Logging Tools 116
4.2 Log Interpretation Philosophies 123
4.2.1 Log Interpretation Problems 123
4.2.2 Geothermal Reservoir Classification for
Well Log Interpretation 125
4.3 Well Logging and Ground Water Monitoring 130
4.3.1 Well Construction 130
4.3.2 Defining Baseline Conditions 136
4.3.3 Monitoring the Injection Well 148
4.3.4 Monitoring Observation Wells 151
4.4 Well Logging Costs 151
References 156
SECTION 5 - GEOTHERMAL INJECTION TECHNOLOGY 160
5.1 Development of Injection Technology 160
5.1.1 Injection in the Petroleum Industry 160
5.1.2 Industrial Waste Injection 161
5.2 Characteristics of Geothermal Injection 162
5.2.1 Chemical Characteristics 162
5.2.2 Temperature 162
5.2.3 Quantity of Fluid 164
5.2.4 Depth 166
5.2.5 Flow Dynamics 168
5.2.6 Reservoir Rock Type 168
5.3 Geologic, Hydrologic and Reservoir Evaluation .... 169
5.4 Chemical Problems 169
5.4.1 Scaling 170
5.4.2 Formation Plugging 176
5.4.3 Corrosion 178
5.4.4 Hydrogen Sulfide 182
5.5 Physical Problems 183
VII
-------
5.5.1 Land Surface Deformation 183
5.5.2 Induced Seismicity 184
5.5.3 Hydrofracturing of Confining Formations .... 186
5.5.4 Thermal Stress 187
5.6 Injection System Design 187
5.6.1 Pretreatment 187
5.6.2 Delivery Systems 188
5.6.3 Well Design 188
5.6.4 Monitoring Injection Well Operation 191
5.6.5 Cost Analysis 193
5.7 Case Histories 197
5.7.1 Wairakei, New Zealand 197
5.7.2 Otake, Japan 197
5.7.3 Cerro Prieto, Mexico 198
5.7.4 Valles Caldera, New Mexico 199
5.7.5 The Geysers, California 199
5.7.6 Larderello, Italy 200
5.7.7 Ahuachapan, El Salvador 200
5.7.8 Imperial Valley Fields, U.S.A 201
5.8 Research Needs 202
5.8.1 Chemical Aspects 202
5.8.2 Equipment Development Needs 203
5.8.3 Reservoir Engineering 203
5.8.4 Physical Problems 204
References 205
APPENDIXES
A EPA POSITION ON SUBSURFACE EMPLACEMENT OF FLUIDS ... 210
B ABBREVIATIONS 213
C U.S.-METRIC CONVERSION TABLE 215
Vlll
-------
FIGURES
Number Page
1.1 Schematic diagram of areal spread of
degraded water 12
1.2 Flow chart of the monitoring methodology 15
1.3 The interrelationship of detective/ predictive
and preventive aspects of monitoring and
the monitoring methodology 15
2.1 Schematic diagram of a hydrothermal reservoir
in an extinct volcanic caldera 22
2.2 Schematic diagram of a volcano-tectonic
hydrothermal reservoir in a sedimentary basin . 23
2.3 Static temperature profiles in geothermal wells
from the Salton Trough 24
2.4 Modified model for Salton Sea geothermal system
showing source of recharge and water quality
changes 29
2.5 Location of producing and injecting wells of a
hypothetical reservoir ....- 34
2.6 A. A production-injection doublet
B. A system of nine production and five
injection wells 37
2.7 Comparison of concentration ranges of
constituents in geothermal and potable waters . 42
3.1 Temperature gradient map showing Known
Geothermal Resource Area (KGRA) locations
in Imperial Valley, California 56
3.2 Modified Stiff diagrams for characteristic ^
Imperial Valley ground waters 57
3.3 Schematic surface of specific conductance
values for Imperial Valley ground water .... 60
ix
-------
3.4 Langelier-Ludwig diagram for ground water data
in Imperial Valley, California 62
3.5 Salinity section A-B from Langelier-Ludwig
diagram for ground water data in Imperial
Valley, California 64
3.6 Modified Stiff diagrams representing hypothetical
analyses of ground water resulting from
specified chemical changes in Colorado River
water 65
3.7 Preliminary survey of wells in and near East
Mesa KGRA 72
3.8 Escape of injected fluid through deteriorated
cement seal (A) and through hole in casing (B) . 79
3.9 Escape of injected fluid through abandoned
boreholes in area 79
3.10 Escape of injected fluid through fractures (A)
and faults (B) 80
3.11 Location of injected fluid front by high-
frequency electromagnetic probes 94
3.12 Three dimensional perspective of a Langelier-
Ludwig diagram showing surfaces of salinity
sections 107
4.1 Gamma-gamma logs used to interpret position
of grout behind casing, and caliper logs
used to select depth for grouting and to
estimate volume required 135
4.2 Stratigraphic correlation with gamma and neutron
logs, Upper Brazos River Basin, Texas 138
4.3 Structure and thickness of aquifers A to E based
on well logs, National Reactor Testing Station,
Idaho 139
4.4 Temperature logs, Yukon Services well, Cook
Inlet Field, Anchorage, Alaska 140
4.5 Correlation of fracture zones between rotary
hole T-5 and core hole C-l utilizing caliper
logs, Upper Brazos River Basin, Texas 141
4.6 A comparison of a neutron log with porosity of
core samples, Upper Brazos River Basin, Texas . 143
-------
4.7 injectivity profiles and their relation to a
caliper log 145
4.8 Location of brine-fresh water interface on
neutron and fluid conductivity logs, hole T-19,
Upper Brazos River Basin, Texas 147
4.9 Typical simultaneous electronic casing caliper
and casing inspection logs run in 7-inch
casing 150
4.10 Maximum temperature of water between depths of
140 to 200 m (459 to 656 ft) based on well logs
of monitoring wells near a disposal well,
National Reactor Testing Station, Idaho .... 152
4.11 Cross section through part of National Reactor
Testing Station, Idaho showing decrease in
temperature and increase in resistivity of
water with distance from the disposal well,
which is nearest Well 43 153
4.12 Well logging costs 155
5.1 Specific gravity of sodium chloride formation
waters versus total solids in ppm 163
5.2 Specific gravity of distilled water as a
function of temperature 163
5.3 Water viscosity as a function of temperature
and salinity (equivalent ppm NaCl) 165
5.4 Geothermal reservoir depth distribution ..... 167
5.5 Well completion for maximum protection during
hazardous waste injection 189
5.6 Comparison of three types of completions used in
hazardous waste injection wells ........ 192
5.7 Typical well drilling costs 194
XI
-------
TABLES
Number Page
2.1 Comparison of Two Conceptual Hydrothermal
Reservoir Models 26
2.2 Properties of the Hypothetical Reservoir 33
2.3 Downhole Pumping Exploitation Characteristics
of a Hypothetical Reservoir for a 50 MWe
Power Plant 36
2.4 Comparison of Inorganic Chemical Water Standards
with Geothermal and Sea Water Analyses 41
3.1 Summary of Desired Subsurface Baseline Data
Elements and Methods Available for Their
Evaluation 51
4.1 Commercial Well Logging Tools 117
4.2 Utility of Various Well Logs in Nonsedimentary
Lithology 126
4.3 Geothermal Reservoir Classification Schemes . . . 128
4.4 Typing of Well-Known Geothermal Reservoirs .... 129
4.5 Well Logging for Well Construction 131
4.6 Well Logging for Defining Baseline Conditions . . 132
4.7' Well Logging for Injection Well Monitoring .... 133
4.8 Well Logging for Observation Well Monitoring
in Geothermal Environments 134
5.1 Factors Affecting Scaling in Geothermal Plants . . 171
5.2 Typical Preinjection Treatment Techniques
to Control Scale Formation 172
5.3 Typical Scale Removal Techniques Applicable
to Injection Systems 173
Xll
-------
5.4 Factors Affecting Material Corrosion in
Geothermal Plants 179
5.5 Capital Cost of Injection Wells, Data From
The Literature 195
5.6 Capital Costs for Injection Systems for Four
Well Capacities 196
Kill
-------
SUMMARY
This report discusses all aspects of potential ground water
pollution from geothermal resource development, conversion, and
waste disposal and proposes guidelines for developing a ground
water monitoring plan for any such development. Geothermal
processes, borehole logging and injection well technology as
they relate to geothermal development and ground water monitoring
are also outlined.
Monitoring for geothermal development differs from other
types of ground water monitoring in several ways. In addition
to geologic and hydrologic factors, the effects that reservoir
production and liquid waste disposal will have on the geothermal
and cool ground water systems of the area must be considered.
The chemical make-up of geothermal fluids throughout the
world varies widely both in number of chemical species and their
concentrations. Total dissolved solids may range from about 50
to almost 400,000 parts per million and pH from 2 to 10 units.
The fluid characteristics may vary from one reservoir to another,
from one well to another in the same reservoir, and over time in
the same well. Escape of harmful fluid to usable aquifers or sur-
face water is a prime potential hazard of geothermal development.
Injection is by far the preferred method of geothermal
liquid waste disposal. Alternatives to injection are environ-
mentally unacceptable in the United States because of the quan-
tity and chemical composition of the waste fluid. Therefore,
the proposed methodology emphasizes monitoring potential ground
water degradation due to fluid injection.
GEOTHERMAL PROCESSES
Liquid- or vapor-dominated hydrothermal systems are most
likely to be developed. A liquid-dominated system will produce
a far greater volume of effluent than a vapor-dominated one.
For example, the Wairakei field produces about 200 times as
much effluent per unit of electricity as The Geysers.
A geothermal reservoir may consist of many separate pockets
of fluids at various depths and with varying degrees of hydraulic
connection between them. The relationship between shallow ground
water, deep ground water and geothermal water must be known in
-------
order to assess potential degradation. A conceptual model to
illustrate these relationships would have four main features:
a natural heat source, a water supply, a ground water reservoir
and a caprock.
For effective waste disposal, the reservoir must be able to
accept all of the liquid waste at essentially the same rate it
is produced. This rate depends upon porosity and pore geometry
of the reservoir rocks, the thickness of the reservoir, the
specific weight and viscosity of the fluid, the applied pressure
gradient, the number and construction of wells and the overall
reservoir flow dynamics. Most geothermal reservoirs exhibit
fracture porosity as opposed to intergranular, vesicular or
vuggy porosity., A sense of the flow dynamics in a producing
reservoir can be developed through analysis of hypothetical
reservoir models.
Injection may affect resource recovery by maintaining
reservoir pressure, replenishing the geothermal fluid supply,
recovering the latent heat content of the reservoir rocks and
possibly cooling the geothermal reservoir fluid. In most geo-
thermal systems, resource recovery can be improved by artificial
recharge through properly emplaced injection.
THE MONITORING METHODOLOGY
Each potential development area is unique in terms of its
activity. Thus, one "monitoring plan" cannot be designed that
would apply to all sites. Instead, a general methodology is
suggested that can be adapted to the specifics of each site. In
applying this methodology a thorough knowledge of geology,
hydrology and geochemistry is required for meaningful evaluation
and analysis at each site. The six steps in the methodology are
outlined below.
1. Define Baseline Conditions and Projected Development
Baseline conditions—those that exist prior to injection—
must be established in order to determine hydrologic changes that
may occur during injection. These conditions include:
A) chemical characteristics of nongeothermal ground
water, geothermal ground water, and surface
waters;
B) geology and hydrology of the area;
C) location and well completion data for all wells
around the geothermal site; and
D) mechanics and characteristics of the geothermal
system.
-------
In addition, it is necessary to estimate production and
injection fluid volumes, as well as the chemical and physical
changes that the fluid will undergo between production and
disposal. The reservoir development plan, established by a
reservoir engineer, will specify optimum production and injection
rates and well locations. It will provide a basis for estimating
how the natural geothermal system will be stressed and how that
stress may also influence overlying aquifers. The chemical
characteristics of the post-process geothermal fluids must be
compared with those of natural geothermal fluid and nongeothermal
ground water. Local geologic and hydrologic factors at the
site must be defined; e.g. landsliding, which could cause blow-
outs and consequent ground water pollution, has been a major con-
sideration in planning geothermal facilities at The Geysers.
2. Forecast Aquifer Conditions
Forecasting the interaction between geothermal and nongeo-
thermal aquifers may help avoid potential problems. This
analysis will consider: (A) models of the cool ground water and
geothermal systems; (B) mechanisms and pathways of potential
pollutants; (C) chemical reactions in the aquifers; and (D)
effects of alternative developments.
3. Define Limits of Detection
Chemical changes in the ground water will occur in a spatial
and temporal matrix. The necessary chemical, spatial and temporal
sensitivity of detection in the matrix must be specified for each
area. The required sensitivity is mainly a function of:
A) chemical contrast of geothermal and nongeothermal
fluids;
B) environmental sensitivity to particular con-
stituents;
C) natural variations in water characteristics;
D) available analytic techniques;
E) hydrologic factors;
F) the relative size of development;
G) characteristics of potential pollutant pathways;
and
H) water use and well distribution density in the
area.
-------
Analysis of these parameters will aid in determining
sampling frequency, distribution and density of sample points,
significant chemical and physical parameters, and sampling and
analysis methods to be utilized.
4. Evaluate Monitoring Techniques
Ground water monitoring will take place at the disposal
facility as well as in the surrounding area. Techniques that
may be used are fluid sampling and analysis, well logging,
tracers, surface geophysics, pressure, temperature and flow
measurements and other special or developmental techniques. All
the monitoring techniques, except surface geophysics, involve
the use of wells.
Sampling and analysis of the fluid is the most common and
usually the only technique used in ground water monitoring.
When properly used, applied and interpreted, this technique
provides the most direct evidence of chemical changes in the
ground water. It will be used in all ground water monitoring
plans. The remaining techniques mentioned above may be used in
special situations.
Well logging is discussed later in this summary. Under
certain conditions, surface geophysical methods may supply infor-
mation on subsurface structure and ground water flow patterns.
Radioactive chemical and dye tracers have been successfully
applied to ground water investigations to determine ground water
flow paths, aquifer parameters and the vertical and horizontal
movement of water within a borehole. The most widely used tracer
in ground water study is tritium, a naturally occurring isotope
of hydrogen.
5. Design Monitoring Plan and Alternatives
To efficiently detect chemical changes in ground water
within the specified chemical, temporal and spatial limits, the
monitoring plan must not be constant or static. Some areas will
require more frequent sampling; others will require a denser
array of sampling points and still others will require different
analyses. The greatest risk of waste fluid escape is through or
around the outside of the injection well rather than by leakage
through permeable confining beds, fractures or unplugged wells;
therefore, emphasis will be placed on monitoring in and around
injection wells.
The components to be evaluated in arriving at an adequate
monitoring plan are: (A) spatial distribution of sample points
and sampling frequency; (B) applicable monitoring techniques;
CO chemical and physical parameters; (D) regulatory specifica-
tions; and (Epcost versus confidence.
-------
6. Implement Monitoring Plan
Implementing the monitoring plan will involve data collec-
tion at specified frequency and locations, and synthesis,
interpretation and display of the data. Past data will be
reviewed and correlated with new data. As the plan is carried
out, the actual needs of the area will become clearer and the
plan can be modified for more judicious and efficient monitoring.
APPLICATION OF WELL LOGGING TO GROUND WATER MONITORING IN
GEOTHERMAL AREAS
Well logs can play a vital role in a ground water monitor-
ing plan. They can:
A) provide continuous and detailed vertical
profiles of rock and fluid properties;
B) help define the baseline conditions and
characteristics of potential injection zones,
nearby ground water systems and their inter-
relationship;
C) aid in the construction of injection and
observation wells;
D) monitor the condition of the production and
injection wells; and
E) aid in monitoring wastewater flow patterns
and possible degradation of fresh ground
water throughout the monitoring network.
The objectives, operation, analysis and interpretation of
well logs in and around geothermal systems are considerably
different than those in petroleum systems. Petroleum reservoirs
most often occur in relatively soft sedimentary rocks with
intergranular porosity at temperatures less than 150°C (300°F)
and water saturation less than 100 percent. Geothermal reservoirs
usually occur in hard, saturated, fractured, crystalline rocks
at relatively high temperatures. Hence, well logging techniques
and interpretation may be adequate for geothermal reservoir
systems that occur in sedimentary environments, but unfamiliar
lithology poses problems. Standard calibration and interpreta-
tion techniques are often inadequate for nonsedimentary geothermal
reservoirs.
GEOTHERMAL INJECTION TECHNOLOGY
Injection technology is an interdisciplinary field
involving geology, hydrology, reservoir engineering, chemistry,
material science, mechanical engineering, well drilling and
-------
completion technology. Injection wells have been widely used
in the petroleum industry for industrial waste injection and
in several geothermal fields around the world. Much of the
knowledge gained from this experience can be applied to future
geothermal development. However, the characteristics of geo-
thermal injection in terms of chemistry, temperature, fluid
quantity, depth, flow dynamics and reservoir rock type are often
unique.
To conduct a safe and effective geothermal injection
program, the physical aspects of the geothermal reservoir and
its surroundings must be known. Details of these data and their
collection are outlined in the first two steps of the monitoring
methodology.
Chemical problems that may be encountered in geothermal
injection include: scale deposition in the injection lines
and well bore, plugging of the formation around the well and
corrosion of pipes in the system. High temperatures and
pressures as well as injection of water mixed from different
production wells add to the complexity of the chemical problems.
Physical problems in production-injection systems include
potential land surface deformation, induced seismicity, hydro-
fracturing of confining formations and introduction of thermal
stress. Injection of spent geothermal fluid into the geothermal
reservoir is generally recommended to minimize potential
subsidence by maintaining the fluid balance in the reservoir.
The possibility of triggering earthquakes by injection can be
minimized by not exceeding the original pore pressure of the
fluids. Hydrofracturing generally should be avoided.
Reservoir engineering calculations and reservoir modeling
are integral preliminary phases of designing the physical injec-
tion system. Then the type of pretreatment, the fluid delivery
system, the well design and the monitoring operations must be
considered. Injection well completion design for hazardous indus-
trial waste disposal is regulated in many states. Applying these
regulations to geothermal injection wells will ensure maximum
protection of usable subsurface waters.
The cost of a geothermal injection system varies as widely
as the size, type, and chemistry of geothermal reservoirs. The
total capital cost may vary from approximately $300,000 to
$1,000,000.
-------
SECTION 1
INTRODUCTION
Ground water monitoring for geothermal development differs
in several ways from monitoring for other purposes. General
ground water monitoring would include evaluation of the water
chemistry, geology and hydrology of an area. Monitoring for
geothermal development would include all these plus knowledge
of the type of geothermal system, its flow dynamics under
natural conditions, the type and extent of reservoir development
and the effects of injecting spent fluids. Hence/ an under-
standing of geothermal processes and injection well technology
is required for a viable design and assessment of a geothermal
ground water monitoring plan.
The bulk of geothermal waste is hot water, either as spent
liquid or steam condensate. In most cases, discharge of this
liquid waste to surface or usable ground water bodies would cause
chemical and thermal pollution. The preferred and most effective
disposal method to alleviate these problems is injecting these
wastes into the geothermal aquifer. In addition to minimizing
potential water degradation at feasible cost, injection of the
waste fluid may minimize induced subsidence and may improve
resource recovery. The focus of this report is towards monitor-
ing ground water degradation resulting from injection of spent
geothermal fluids.
This report provides a systematic methodology for designing
a ground water monitoring plan to detect and predict ground water
degradation due to geothermal development. It incorporates a
philosophy of prevention as an integral, natural and essential
component of monitoring since ground water degradation is often
an essentially irreversible process. Ground water degradation
is much more difficult to detect and trace than surface water
degradation. When it is possible, mitigation of ground water
degradation is difficult and expensive.
Utilizing this proposed methodology prior to injection and
enforcing rigid specifications for injection well construction
will greatly minimize the chances for ground water degradation.
In applying the methodology, well logging can provide valuable
contributions to well construction, design, baseline data
acquisition, injection, well monitoring, and observation well
monitoring.
7
-------
Case histories of ground water contamination due to injec-
tion of waste show that many of them could have been prevented
by appropriate study and facility design beforehand. If the
initial study shows that the injected fluid may interfere with
any useful ground water aquifer, then the design/ location and
flow rates of the offending wells must be modified before
production and injection begin.
Recent state and Federal regulations regarding the intro-
duction of fluids into ground water systems specify extensive
planning and analysis. The Safe Drinking Water Act (SDWA) of
1974 authorizes the U.S. Environmental Protection Agency (EPA)
to protect ground water. Several states have enacted statutes
requiring reports of various parameters before permits are
issued for deep-well disposal systems. These statutes and
policies, along with numerous discussions in the literature,
provide a sound basis for the development of a comprehensive
and integrated approach to ground water monitoring applicable
to geothermal injection well systems. This report suggests
such an approach based on previous work and policies and proposes
application of borehole logging technology. The resulting
methodology will provide a procedure for detecting and prevent-
ing ground water degradation due to geothermal development.
1.1 REPORT ORGANIZATION AND PREPARATION
Section 1 of the report discusses related studies on ground
water monitoring and injection, ground water monitoring
philosophies, and terminology used in this report.
Section 2 provides an overview of geothermal occurrences
and processes, with emphasis on how they may relate to and
affect ground water flow. This overview is provided by
discussions of the types of geothermal systems, geothermal
system models, geothermal reservoir engineering and resource
recovery as it relates to injection.
Section 3 of the report presents the monitoring methodology
itself. The six steps of the procedure are: (1) define baseline
conditions and projected development, (2) forecast aquifer
conditions, (3) define limits of detection, (4) evaluate
monitoring techniques, (5) design monitoring plan and alterna-
tives, and C6) implement monitoring plan. Each of these steps
is detailed in the section.
Section 4 covers the application of well logging to ground
water monitoring in geothermal environments. It surveys the
available logging techniques, their application and limitations
in the geothermal environment, their cost and description of
logs that would*-be applicable to each phase of the methodology.
-------
Section 5 provides an overview of geothermal injection well
technology. It includes discussion of characteristics of geo-
thermal injection, injection technology development in other
industries, chemical and physical problems associated with
injection, injection system design, geothermal case histories
and research needs.
Those who participated in the preparation of this report
include:
Project Management: Frank C. Kresse
Richard B. Weiss
Sections 1, 2 and 3: Richard B. Weiss
Section 4: Theodora O. Coffey
Section 5: Tamara Williams
Consultant on Well Logging
and Injection Technology: Subir K. Sanyal
Typing: Jodie de Bartok
Drafting: Glennda Rayburn
Editing: Linda Encinas
1.2 RELATED STUDIES
Many of the specific elements, (e.g., establishing geologic
and hydrologic conditions, field sampling techniques or
mathematical modeling) are tasks that are outside the scope of
this study and are only discussed summarily here, presuming the
reader has this background knowledge or will refer to the cited
references:
Several studies have been published in areas directly
related to the present study. These include five EPA reports on
"Monitoring Ground Water Quality":
- "Monitoring Methodology" (Todd, et al. 1976) out-
lines a step-by-step general ground water monitor-
ing procedure and discusses ground water quality.
It emphasizes surface sources of ground water
deg r adat ion.
- "Methods and Costs" (Everett, et al. 1976) discusses
methods of estimating current costs for monitoring
at the land surface, in the vadose zone and in the
zone of saturation, and for sample analysis.
-------
- "Data Management" (Hampton, 1976} is concerned with
the types of information to be managed (such as
geologic, hydrologic, water quality, temporal
and spatial) and discusses the collection,
communication, storage, processing and retrieval
of this information, along with possible applica-
tions of existing data management systems.
- "Illustrative Examples" (Tinlin, 1976) covers ground
water pollution case histories and evaluation of
monitoring techniques for brine disposal in
Arkansas, plotting waste contamination in New
York, landfill leachate contamination in
Connecticut, an oxidation pond in Arizona, and
multiple source nitrate pollution in California.
It also applies the monitoring methodology of
Todd, et al. (1976) to examples for
agricultural return flow, a septic tank, a per-
colation pond and a solid waste landfill.
- "Economic Framework and Principles" (Crouch,
et al. 1976) covers the legal and institutional
aspects of ground water pollution. Economic
issues of ground water monitoring are discussed
within a framework of an actual hydrogeologic
example and the economic principles are analyzed.
Other relevant documents published by EPA include:
- "Polluted Ground Water - Some Causes, Effects,
Controls and Monitoring" (TEMPO, 1973) discusses:
ground water quality and pollution, institutional
and legal aspects, salt water intrusion, pollution
from diversion of flow and direct disposal of
pollutants from industrial injection and other
types of wells, surface ponding, septic systems,
spraying, stream beds, landfills, tank or pipeline
leakage, and percolation from surface waters.
The following documents are particularly relevant to this
report since the preferred waste disposal method for geothermal
developments is injection of spent fluids.
- "Monitoring Disposal Well Systems" (Warner, 1975)
includes discussion of the subsurface environment,
acquisition of subsurface data, prediction of
aquifer response and surveillance of operating
wells.
10
-------
- "Review and Assessment of Deep Well Injection
of Hazardous Waste" (Reeder, et al. 1977) discusses
geologic, engineering, chemical and microbiological
aspects of deep well injection of hazardous
industrial waste, as well as ground water use and
monitoring. Additional topics are characterization
of waste, injection well inventory and case histories,
deep well injection research projects, economics
of deep well systems and legal and regulatory
considerations.
- "An Introduction to the Technology of Subsurface
Wastewater Injection" (Warner and Lehr, 1977)
a general treatise on all aspects of injection.
Many other relevant references are cited throughout the
report.
1.3 GROUND WATER MONITORING PHILOSOPHIES
Ground water monitoring philosophies can run the gamut from
no monitoring to a detailed, complex and comprehensive program.
Most plans will fall between these two extremes.
Ground water monitoring is conceptually fairly straight-
forward; however, the application and interpretation may be quite
complicated. In a simplified example (Fig. 1.1), a fluid is
introduced into an aquifer at Point A and monitored at Points
1, 2, 3, 4 and 5. Degraded water is detected in Wells 1 and 4.
This evidence is combined with information about direction and
rate of natural ground water flow to estimate the location of
the degraded water front. Most field situations are not so
simple. Some problems which may arise include the erratic
concentration of fluid constituents in the observation wells;
differential make-up of the subsurface (lensing strata, buried
stream channels or faults may alter flow rates in certain
directions); unforeseen natural temporal and spatial changes in
water characteristics; interaction of the geothermal system
with cooler ground water flow; interaquifer degradation from
poorly constructed wells. Complications such as these make
rigorous analysis and planning necessary to mitigate potential
ground water degradation.
Monitoring was not prevalent before the current era of
environmental consciousness and regulation. This is still the
case in areas where regulatory constraints have not yet been
formulated or are not being enforced, or where potential ground
water degradation has not been considered. This "philosophy"
essentially disregards the effects of disposal on the ground
water system.
11
-------
\
NATURAL GROUNDWATER
FLOW DIRECTION
WATER WELL
OBSERVATION
WELL —-~^.
0°
(PRODUCTION WELL)
-DEGRADED FLUID FRONT
f
I
\ 4
V O
V
\
N
O, N
INJECTION WELL}
Figure 1.1 Schematic diagram of areal
spread of degraded water.
12
-------
Active monitoring can be divided into three aspects:
1) detection,
2) prediction, and
3) prevention.
The goal of detective monitoring is to discover ground water
degradation if it occurs. Predictive monitoring proposes to
predict degradation before it occurs, or to predict its advance
once it has started. Preventive monitoring proposes to prevent
degradation. These aspects are discussed in more detail below.
A detective monitoring plan includes an array of monitoring
points which are sampled at specified intervals after the
degraded water is introduced into the ground. The samples are
analyzed chemically to detect the areal or spatial pattern and
rate of spread of a degraded water front. The sensitivity of
such a system depends on the density and distribution of
monitoring points and the sampling intervals, and will increase
in proportion to the number of sample points, frequency of
sampling, and thoroughness of the vertical and horizontal
distribution of monitoring points.
Using inferences based on geologic and hydrologic data and
adequate planning, reliable detection limits may be provided.
This type of plan might use existing wells and perforated
intervals. Additional observation wells may be drilled or
existing wells might be reperforated where necessary. The
geologic and hydrologic information will permit the extrapolation
of ground water flow data to areas without sample points.
Estimated flow rates and paths will provide a basis for deciding
on a reasonable sample interval.
A predictive monitoring plan involves more detailed geologic,
hydrologic, and geochemical investigations. The analysis will
most likely employ a conceptual or mathematical model and the
results may be used to estimate the potential extent and rate
of dispersion of degraded water into the ground water system.
This investigation and prediction often occurs after some
degraded water has been detected in an aquifer. Then it is
considered imperative to understand how the degraded water got
there, where it will go next, and how long it will take to get
there.
A preventive monitoring plan is similar in many respects
to the predictive plan. However, the disposal facility will be
designed and constructed to prevent escape of degraded fluid
(.i.e., well design in Section 3.4 and surface facilities in
Section 3.1). Additionally, the data gathering, analysis,
13
-------
interpretation and prediction all take place prior to introduc-
tion of degraded water into the ground water system. If the
prediction is that the wastewater will be unacceptably dispersed
into the ground water system, the proposed disposal plan will be
modified to produce acceptable dispersion rates and patterns.
These three aspects of monitoring are not mutually
exclusive. An adequate monitoring plan will incorporate each of
them. The interrelationship between these aspects and the
monitoring methodology is illustrated in Figs. 1.2 and 1.3.
Past monitoring plans have tended to be primarily detective.
Current plans for significant projects also incorporate
predictive and preventive aspects to varying degrees. The ground
water monitoring methodology proposed here is based on this
concept of an integrated approach.
1.4 TERMINOLOGY
Certain technical terms used in this report have been used
with different nuances of definition. To minimize misunder-
standings due to terminology, these terms, as they are used in
this report, are defined below.
"Monitoring methodology"—describes the process, or set of
considerations, evaluations and steps, that is used to develop a
"monitoring plan". So, the "methodology" is the general approach.
"Monitoring plan"—the site-specific detailed program for
predicting, detecting and preventing ground water degradation at
a particular geothermal development.
"Degradation"—an increase in the concentration of dissolved
or suspended materials in any water. It is used in preference
to other terms such as "contamination" or "pollution" which imply
a judgment on usability. Water may be degraded to a point at
which it is not usable for some purposes and still usable for
others. For example, a water may contain some elements in
concentrations toxic to humans, but may be suitable for some
industrial process; or a water with more than one or two ppm
boron may not be used for irrigation, but may be potable.
"Injection"—emplacement of a fluid in the ground through an
injection well. The geothermal industry has used the term
"reinjection" for this process, but this implies that the fluid
is being injected again, which is not necessarily so. The term
disposal well is avoided since it implies that the only function
of the well is for disposal.
Injection well technology is an interdisciplinary field.
Consequently, the somewhat different connotations that
14
-------
3) DEFINE LIMITS
OF DETECTION
MA) EVALUATE GROUND
WATER MONITORING
TECHNIQUES
4B) EVALUATE DISPOSAL
FACILITY MONITORING
TECHNIQUES
5) DESIGN MONITORING
PLAN AND ALTERNATIVES
6) IMPLEMENT MONITORING PLAN
Figure 1.2
Flow chart of the monitoring
methodology.
DETECTION
-S —•
03:
Z T)
~
DESIGN
MONITORING
PLAN
DEFINE LIMITS
OF DETECTION
S EVALUATE
MONITORING
TECHNIQUES
DEFINE BASELINE
CONDITIONS R
PROJECTED
DEVELOPMENT
DEFINE PREVENTIVE MEASURES
Figure 1.3 The interrelationship of detec-
tive, predictive and preventive
aspects of monitoring and the
monitoring methodology.
-------
permeability terms have developed in these related fields may
lead to some confusion. Therefore, the terms used in this
report are defined below.
The intrinsic property of the formation that allows fluid to
flow through it is called "permeability". It is defined as:
k = Cd2
where C is a dimensionless constant and d is a representative
pore diameter. This term is commonly used in the petroleum
industry and is expressed in units of area (e.g. cm ) .
The "darcy", adopted as the practical unit of permeability, is
defined as:
•j 2
1 darcy = 1 centippi se x 1 cnr3 /sec/cm
____.__.
and is also expressed in units of area.
The term "coefficient of permeability" is used to express
the flow property of a specific fluid flowing through a specific
rock matrix at a specific temperature and pressure head. This
term, often used in ground water hydrology, is sometimes
inadvertently referred to simply as "permeability" ergo the
confusion. The coefficient of permeability (K) , or hydraulic
conductivity, originally defined in 1856 by Darcy, is:
K =
A(dh/dL)
Q is the flow rate through a cross- sectional area. A, and (dh/dL)
is the hydraulic gradient. It is expressed in units of velocity
(e.g. cm/sec) . To minimize the confusion between these two
terms, the term "hydraulic conductivity" will be used in
reference to "coefficient of permeability" in this report. The
relation converting K to units of area is:
K =
v
where K = hydraulic conductivity, k = (intrinsic) permeability,
g = specific gravity of the fluid, and v = viscosity of the fluid.
So "permeability" is a property of the rock matrix alone, while
"hydraulic conductivity" is a property of the rock matrix and the
fluid.
16
-------
REFERENCES
Crouch, R. L., R. D. Eckert, and D. D. Rugg. 1976. Monitoring
Groundwater Quality: Economic Framework and Principles.
U.S. EPA, Las Vegas, Nevada. 907 pp.
Everett, L. G., K. D. Schmidt, R. M. Tinlin, and D. K. Todd.
1976. Monitoring Groundwater Quality: Methods and Costs.
U.S. EPA #600/4-76-023, Las Vegas, Nevada. 152 pp.
Hampton, N. F. 1976. Monitoring Groundwater Quality: Data
Management. U.S. EPA #600/4-76-019, Las Vegas, Nevada.
70 pp.
Reeder, L. R., J. H. Cobbs, J. W. Field, Jr., W. O. Finley,
S. C. Vokurka, B. N. Rolfe. June 1977. Review and
Assessment of Deep Well Injection of Hazardous Waste,
EPA 600/2-77-029a, 168 pp. (first of four volumes).
TEMPO. July 1973. Polluted Ground Water: Some Causes, Effects,
Controls and Monitoring. U.S. EPA Report No. 600/4-73-OOlb,
edited by Charles F. Meyer. 282 pp.
Tinlin, R. M. (ed.). 1976. Monitoring Groundwater Quality:
Illustrative Examples. U.S. EPA, Las Vegas, Nevada. 92 pp.
Todd, D. K., R. M. Tinlin, K. D. Schmidt, and L. G. Everett.
June 1976. Monitoring Groundwater Quality: Monitoring
Methodology. U.S. EPA Report 600/4-76-026. 172 pp.
Warner, D. L. July 1975. Monitoring Disposal Well Systems.
U.S. E.P.A. Report No. EPA-68014-74-008. 109 pp.
Warner, D. L., and J. H. Lehr. 1977. An Introduction to the
Technology of Subsurface Wastewater Injection. EPA-600/2-
77-240. 355 pp.
17
-------
SECTION 2
GEOTHERMAL PROCESSES
Three general classes of geothermal resources are commonly
recognized: hydrothermal convection systems, hot igneous
systems, and conduction-dominated areas. The great majority of
currently utilized geothermal resources are hydrothermal con-
vection systems, which are discussed in detail below. Hot
igneous systems include partially molten magma masses and "hot
dry rocks". Conduction-dominated areas include all remaining
areas of the earth's crust, especially those with somewhat
higher than normal temperature gradients (White and Williams,
1975).
Problems in utilization of hot igneous and conduction-
dominated systems involve lack of technology to efficiently
extract the heat. An experiment is underway at Los Alamos
Scientific Laboratory (LASL), New Mexico, to extract energy from
the nearby hot dry rock system. Cool water is injected in one
well, heats as it flows through a hydrofractured hot igneous
mass, and is pumped to the surface through a second well. If
this heat recovery technology is successfully developed, it will
produce a highly controlled geothermal ground water flow system
with unique flow dynamics, boundary conditions and fluid
chemistry.
Because of the large volumes of rock involved, the conduc-
tion-dominated areas contain the largest portion of heat in the
earth's crust, but at economically accessible depths the
temperatures are comparatively low. The geopressured resource
of the Gulf Coast is of particular interest in this category
since three forms of potential energy may be recovered—heat
and mechanical energy from the hot, overpressured pore fluids,
and dissolved methane.
Hydrothermal convection systems are usually divided into
two classes: those dominated by hot water and those dominated
by vapor (White, et al. 1971). Both exist as "reservoirs," i.e.,
a body of stored fluid in the pore space of a subsurface rock
formation. A geothermal reservoir may consist of many separate
pockets of interstitial pore fluids at various depths with vary-
ing degrees of hydraulic connection between. These individual
pockets are traditionally considered reservoirs in the oil and
18
-------
gas industry. Most geothermal resources have not been
sufficiently developed to define these individual pockets and
most of the fluid produced is a mixture of fluid from several
pockets.
In a typical hot water system, liquid is encountered at all
depths. It will produce "wet steam", i.e. steam coexisting
with hot water. A vapor-dominated reservoir produces "dry"
steam with little or no boiling water at the surface. Varying
amounts of noncondensible gases, consisting of carbon dioxide,
hydrogen sulfide, methane, nitrogen and others, may be
associated with each type of system.
2.1 HOT WATER SYSTEMS
Hot water systems are apparently far more abundant than
vapor-dominated ones. Pressure caused by the buoyant force of
thermally heated water causes hot water to flow upwards and to
convect when its flow to the surface is impeded.
The temperature of known hot water systems varies greatly
depending upon the proximity of the water reservoir to the heat
source, the temperature and size of the heat source, the heat
transfer mechanism and other factors. Temperature generally
increases with depth, although not in a predictable manner.
The highest recorded temperature of liquid-dominated geothermal
systems is about 370°C (698°F) at the Salton Sea field in the
Imperial Valley, California and at Cerro Prieto, Mexico.
Temperatures as high as 340°C (644°F) have been recorded in a
well recently drilled in Puna, Hawaii. Most systems, however,
have lower temperatures. The Wairakei field in New Zealand
has a reservoir temperature of 240 to 260°C (464 to 500°F).
Currently 150°C (300°F) is considered the minimum feasible
temperature for economic electric power generation.
The fluid in liquid-dominated geothermal fields varies from
highly acid to alkaline. The chief salts produced are sodium
chloride, calcium chloride, magnesium chloride, calcium
carbonate, sulfates and silica. Salinity generally ranges from
0.1 to 1.5% solids. However, in very rare cases, such as the
wells of the Salton Sea field, California, levels of dissolved
solids may rise as high as 35%. The chemistry of geothermal
fluids is discussed in more detail in Section 2.5.1.
In a typical hot water system, only part of the fluid will
be produced as steam, the rest as hot water. About 20% of the
total mass will flash to steam in a well with a bottom hole
temperature of 250°C (482°F) and a wellhead pressure of 0.34
MPa (50 psig). The rest of the fluid will be separated off as
superheated water at a temperature of about 140°C (284°F).
These are typical conditions for the Wairakei and Cerro Prieto
wells.
19
-------
The steam is conducted to the power plant. The water,
which is slightly superheated (relative to atmospheric pressure),
may be allowed to cool to boiling at atmospheric pressure and
then disposed of. Alternatively, the water may be conducted
to another separator for a second flash cycle, during which the
lower pressure steam will separate from the boiling water.
This steam may then be conducted to a second stage of power
production that uses a lower-pressure turbogenerator.
2.2 VAPOR-DOMINATED SYSTEMS
Vapor-dominated systems are quite rare. The oldest known
one is in Larderello, Italy. The largest known system is The
Geysers field in northern California.
Practically all of the well output in a "dry steam" field
may be directly transmitted to the power plant after minor
cleaning of the steam. When the content of noncondensible gases
is high, their extraction before the steam is introduced into
turbines improves the efficiency of condensing-type turbines.
"Dry steam" reservoirs probably contain some hot water as
well as steam. High-temperature water may convert into steam
in the rocks when the ambient pressure is reduced by venting.
The initial steam production reduces pressure in the surrounding
fluid. Pressure reduction below the boiling point will allow
conversion of liquid into dry steam in the rock pore space.
Because of the thermodynamic properties of water, vapor-
dominated reservoirs characteristically exhibit a narrow
temperature range, typically about 200° to 240°C (390° to 460°F).
2.3 GEOTHERMAL SYSTEM MODELS
The relationship between shallow ground water, deep ground
water and geothermal water must be known in order to assess the
potential degradation of ground water due to geothermal develop-
ment. This includes comprehension of the mechanisms of recharge
and discharge, heat transfer, fluid flow direction and rate,
water chemistry, geometry of ground water reservoirs and aquifers
field boundaries and the rate and location of fluid extraction
and injection. Available data will generally be insufficient.
Therefore, to understand the geothermal aspects of the ground
water system, it is necessary to utilize a conceptual model
consistent with the data available. Such a model, from basic
characteristics to general types to a specific Imperial Valley
geothermal field, is described below. As data are collected
and analyzed, the model should be improved to better represent
conditions in a specific reservoir.
20
-------
2.3.1 Basic Mode1
A basic model of a geothermal system contains four fea-
tures. They are: natural heat source, water supply, ground
water reservoir and caprock (Facca, 1973, p. 62). The heat
source in most cases would be of magmatic origin. The water
supply would have to be sufficient to saturate and replenish
the permeable rock or aquifer. The caprock would seal the
system to allow convection in the aquifer or at least trap
the water long enough for it to be heated (Figs. 2.1 and 2.2).
In this general model, ground water in the permeable aquifer is
heated by conductive heat flow through the bedrock from the
magmatic source. Since heated water rises, it starts convective
currents within the aquifer beneath the confining caprock. This
heated water may escape through the caprock via fractures or
wells. Fig. 2.3 shows temperature versus depth plots which are
typical of hydrothermal convection systems. The sharp decrease
in temperature gradient as it enters the hydrothermal convective
cell beneath the caprock is diagnostic of such cells.
The rate of recharge is delicately balanced against the
pressure differential between the reservoir and its surrounding
area. Evidence from repeat gravity surveys across the Wairakei
geothermal field in New Zealand (Hunt, 1970) suggests that at
present production levels, the recharge of that liquid-dominated
system is at least one-third the rate of extraction. This
conclusion is based upon calculation of the expected change in
the gravitational attraction of the earth across the field if no
recharge took place, in comparison with the actually observed
change.
A very large quantity of water flows through a geothermal
system over its life. Ellis (1966) has estimated that the
geothermal fluids in the Wairakei system in New Zealand have
been replaced several tens of times. Meidav, et al. (1975)
have estimated that the amount of geothermal water that has
flowed through the East Mesa geothermal system in the Imperial
Valley is on the order of 9 x 1015 kg (1 x 1013 tons). Such a
quantity represents a flow of water hundreds of times greater in
volume than the total pore space in that reservoir. Even such
geothermal systems as that at Larderello or The Geysers, have a
large inflow of water, despite the effective seal that such
systems must possess. Oxygen and hydrogen isotope studies
suggest that the level of recharge of both systems must be about
10% or more of the current rate of production of steam. In
The Geysers case, it is quite likely that an increase in produc-
tion would result in an increase in recharge.
21
-------
I.
:•-
SHALLOW
HOT AQUIFERS
LAVA FLOWS 8
V fo /SEDIMENTS.
-*• J c- -^T\ o HOT SPRING
^tV-A { S OR GEYSER
ALLUVIUM,
SEDIMENTS
Figure 2.1 Schematic diagram of a hydro-thermal reservoir in an extinct volcanic
caldera. (modified from Meidav and Sanyal, 1976)
-------
RIVER
SHALLOW AQUIFERS
X
ALLUVIUM
ALLUVIUM
> vw _u
r ROCKS OF LOW PERMEABILITY _-_-_-_-_-_-_-_-y x A v
Z-Z-Z-Z-Z-Z-Z-Z-Z-Z-Z-Z-K 'CAP' ROCK ) -Z-Z-Z-Z-Z-Z-Z-Z-Z-Z-Z-Z-Z// r £«-,*> A
POROUS
SEDIMENTARY^ ROCK
N A > A
CRYSTALLINE ROCKS 1
( IMPERVIOUS)
CONVECTING MAGMA
v v r
Figure 2.2 Schematic diagram of a volcano-tectonic hydrothermal reservoir in a
sedimentary basin. (Meidav and Sanyal, 1976)
-------
100
1000-
2000-
3000-
g 4000-
Q_
LJ
Q 5000-
6000-
7000-
8000-
9OOO-
150
200
Temperature (°F)
250 300
350
400
A
B
C
D
E
F
G
DUNES (OWR No.I)
HEBER {J.D.JACKSON Jr. No. I)
SALTON SEA (ELMORE No. I)
HEBER (G.T.W. No.3)
E. MESA (6-I)
E. MESA (8-I)
CERRO PRIETO
2500
50
100 150
Temperature (°C)
200
Figure 2.3 Static temperature profiles in geothermal wells
from the Salton Trough. (Geonomics, 1978a)
24
-------
2.3.2 Conceptual Models for Two Types of Geothermal Reservoirs
Two of the most common types of hydrothermal reservoirs
are volcanic and volcano-tectonic (Figs. 2.1 and 2.2). Both
types occur in areas of thinner than normal crust with shallow,
recent volcanism nearby. In the entire area west of the Rocky
Mountains, the thickness is less than the average of 40 to 50
km (25 to 30 mi); for example, it is 13 to 22 km (8 to 14 mi)
thick in Imperial Valley, California.
In both types of reservoirs, heat is transferred by conduc-
tion from a hot igneous mass to an aquifer. Local faults often
improve this mechanism by providing paths for more efficient
heat transfer by fluid convection. These areas may be
associated with large, extruded lava fields, volcanic calderas
or intrusives. Examples of these are the lava fields of the
northwestern United States and Valles Caldera, New Mexico;
Yellowstone, Wyoming; and Long Valley, California. The regular
elliptical shape of the Salton Sea geothermal anomaly has
supported the hypothesis that the heat source here is an
intrusive at shallow depth.
Exploitable geothermal reservoirs occur where the heat
source is near the surface and/or where the faults provide
adequate channels for flow of hydrothermal solutions. The
temperature and pressure of the fluids drop as they ascend along
faults, fractures or pore spaces. This results in precipitation
of silica and other minerals, or hydrothermal alteration of the
rock causing kaolinization, thereby sealing these channels in
the rock mass. Local tectonic adjustments may provide movement
necessary to keep some of these channels open. If they do not
remain open, the self-sealing mechanism will migrate to an
equilibrium depth where the temperature and pressure are
sufficient to provide conditions for the minerals to remain in
solution or not otherwise undergo alteration.
All high temperature hydrothermal reservoirs must have a
caprock (Facca, 1973). This may be a primary feature, formed
by deposition of an impermeable layer; for example, the deltaic
clay in Imperial Valley, or, it may be a secondary feature
formed by the self-sealing mechanism described above. An
example is the sealing of the fracture permeability in the hard,
fractured Franciscan graywacke in The Geysers steam field.
The difference between the volcanic and volcano-tectonic
reservoirs lies in the subsurface structure, rock types and
tectonic framework. Table 2.1 provides a convenient comparison
of the distinctive attributes of the two conceptual hydrothermal
reservoir models. i
25
-------
TABLE 2.1 COMPARISON OF TWO CONCEPTUAL HYDROTHERMAL
RESERVOIR MODELS (modified from Meidav and Sanyal,
1976)
Attribute
Surface manifestations
Drilling cost
Formation evaluation
and application of bore-
hole logging
Porosity and permeability
Vertical Permeability
Storage and flow
capacities
Reservoir development
and performance
prediction
Potential subsidence
Volcano-Tectonic
(Sedimentary) Model
Limited
Relatively lower
Simple
Largely intergranular
Restricted due to
interbedded impermeable
layers
Often distributed
uniformly
Relatively simple
Likely unless spent
fluid is injected
Volcanic Model
Some
Relatively higher
More complex
Controlled by fractures
Similar to horizontal in places
Distributed non-uniformly
More difficult
Less significant
26
-------
Volcanic Model—
In the volcanic system the reservoir consists of a fairly
homogeneous faulted and fractured crystalline igneous or
metamorphic fluid saturated mass (Fig. 2.1). It may occur
underneath a volcanic caldera, with the collapse structure
filled with lava flows and sediments. The Long Valley,
California geothermal system is an example of this model.
A shallow magma chamber supplies heat to the reservoir by
conduction. Large scale thermal convection cells, modified by
faults and fractures, develop as the fluid in the fractures is
heated. The convecting hydrothermal solutions circulate within
the crystalline rock mass, depositing silica and/or carbonate
near the top of the convection cell, thereby sealing the system.
The reservoir is most likely bounded by faults, and may contain
faults within its boundaries. The boundary faults may conduct
meteoric, fluvial or lacustrine water to recharge the reservoir.
The interior faults may conduct some of the hydrothermal solu-
tions to the surface, thus producing surface geothermal mani-
festations. These may be warm or hot springs, geysers, steam,
hydrothermal rock alterations, silica and travertine sinter,
warm ground water temperatures, fumaroles and solfataras or
hot soils. • Shallow hot aquifers recharged by the underlying
reservoir may also be present. However, the lack of surface
manifestations does not preclude the existence of a subsurface
geothermal reservoir, and surface manifestations do not imply
the existence of an exploitable resource.
Volcano-Tectonic Model—
In the volcano-tectonic model (Fig. 2.2) sediments are
deposited in a regional structural depression as it is
progressively downdropped from surrounding areas. The reservoir
consists of porous, clastic sediments with intergranular
permeability. The porous layers may be interspersed with less
permeable clay or volcanic flows. The geothermal fluid is
confined between a low permeability caprock and an impermeable
basement rock. Boundary faults or areas of lower permeability
define the limits of the system.
In this system heat is conducted from the heat source
through the crystalline rocks of the thin crust up to the
sedimentary deposits containing ground water. This water is
heated, rises and circulates back down forming a convection
cell. Faults extending deep into the basement complex and
closer to the heat source may also conduct some fluid up to the
convection cell. The convecting current is limited in its
upward extent by the reservoir caprock of either an existing
impermeable layer such as a shale, or one produced by the
self-sealing mechanism of the convecting fluids. It is probably
limited in its lateral extent by structural features.
27
-------
More permeable sediments may overlie the caprock and
contain the shallow aquifers of a distinct ground water flow
system. Recharge to the geothermal system may be conducted by
bounding faults or by downward percolation from the overlying
aquifers. This may occur through parts of the low permeability
caprock some distance from the convection cell. Surface
manifestations are generally less common in this type of system.
The geothermal systems in Imperial Valley are examples of
such a model. It is a broad structural and topographic
depression that has been filled with over 6,200 m (20,000 ft)
of later Tertiary deltaic and lacustrine sands/ silts and
gravels overlain by alluvium and lake sediments (Biehler, 1964).
The underlying pre-Tertiary granitic and metamorphic complex is
intensely step-faulted down from the mountains on both sides of
the valley. A specific model for the ,Salton Sea geothermal
field is described below.
2.3.3 Salton Sea Geothermal Field Model
A ground water and geothermal system model for the Salton
Sea geothermal field in Imperial Valley must consider: the
fault systems in the valley, the great variation in salt con-
centration in geothermal brines, two different meteoric water
sources, up to 6,000 m (20,000 ft) of sands, silts and clays
overlying an igneous and metamorphic basement complex and an
apparently random distribution of geothermal anomalies in the
valley (Rex, 1970). Conceptual models for genesis of Salton
Sea geothermal fluids have been proposed by Craig (1966) , Berry
(1966), Helgeson (1968), White (1968) and Dutcher, et al. (1972),
The last of these is described below.
In this model (Fig. 2.4), a shale dominant layer overlies
a sandstone dominant fractured brine reservoir. The interface
between these two layers in Imperial Valley appears to be
generally between 600 and 900 m (2,000 and 3,000 ft); although
it might be only about 300 m (1,000 ft) at Heber (Meidav, et al.
1975). Recharge is through downward percolation from shallower
aquifers above the brine reservoir and near the basin margins.
Part of the driving force for this water is the higher hydro-
static pressure of the cooler water outside the convection cells
(Dutcher, et al., 1972, p. 30-32). This inflowing water has
low oxygen and deuterium isotope ratios, which imply that the
water has not yet been heated. The model suggests high calcium
sulfate and a TDS content of about 35,000 mg/1 for the recharge
water of the Salton Sea geothermal field. The downward
percolating water is heated and its temperature increases as it
flows towards the convection cell.
The convecting brine becomes concentrated by escape of
water, carbon dioxide, hydrogen sulfide and other vapors through
28
-------
Channel greatl y
exaggerated
Alittle brine escapes upward
ratios of metals to S, COj,
and H20 decrease in shale,\
metal sulfides precipitated
Ci rculptino.
water high in
SO18 relative to
|isand enri_ <
cried in SD by
evaporation
' Vapor escapes upward; most
H,p and H^S condenses
COz is enriched by release in
carbonate-silicate reactions
Inflowing water low inSO
relative toSO16, lowinSD;
moderately high in CaS04
dissolved solids about
35,000 mq/t.
-Vapor
(H20,C02,H2S)
V
Convecting brine
is high in metolS;
low in dissolved
sulfidei
IDS about
260,000 mg/1
Reservoir rocks high in SO
relative to SO16 and low in SD
Zone of CaCO,
Zone of heat flow
by conduction ind
inflowing water,
\temperature
\ increases toward
\ convection cell
and \
CaS04 precipitation, \
causing reduced per—'
meability.
Zone of SiOz solution
causing increased per-
meability
\
HEAT
FLOW
Figure 2.4
Modified model for Salton Sea geothermal system
showing source of recharge and water quality
changes. (Dutcher, et al. 1972)
29
-------
fractures in the reservoir. Most of the water and hydrogen
sulfide condenses in the upward escaping vapor. Carbon dioxide
is enriched by release in carbonate-silicate reactions with the
surrounding rock. The convecting brine becomes high in metals,
low in dissolved sulfide and may attain a maximum TDS content up
to 385,000 mg/1. Within and around the convection cell, hydro-
thermal fluids are precipitating calcium carbonate and calcium
sulfate, which reduces permeability in the reservoir; while the
same fluid is dissolving silica causing increased permeability.
Brine that escapes upward reacts with the shale, resulting in
lower ratios of metals to sulfur, carbon dioxide and water.
Metal sulfides are also precipitated. It is a dynamic system
and all these processes and chemical reactions occur contin-
uously and simultaneously.
Dutcher, et al. (1972) suggest that the difference in brine
concentration between the Salton Sea geothermal field and other
geothermal fields to the south is due to a difference in con-
centration of salts in the recharge water. That is, the
recharge water for the Heber area, for example, may contain
1,500 to 2,000 mg/1 TDS. If this is concentrated about ten
times by the convection cell it would result in a brine with
about 20,000 mg/1 concentration. A similar mechanism for the
Salton Sea geothermal field, assuming a recharge with 25,000
to 35,000 mg/1 TDS, will result in a brine concentrated to
hundreds of thousands of milligrams per liter TDS.
2.4 GEOTHERMAL RESERVOIR ENGINEERING
For effective waste disposal, the reservoir must be able
to accept all of the liquid waste at essentially the same rate
it is produced. The production rate is determined by the
permeability and thickness of the reservoir, the specific weight
and viscosity of the fluid, the applied pressure gradient,
the number and construction of wells, and the overall reservoir
flow dynamics.
Permeability depends on the effective porosity and pore
geometry of the reservoir rock. Porosity in geothermal
reservoirs can be classified into three groups according to pore
geometry: (1) fracture porosity; (2) sedimentary intergranular
porosity; and (3) vesicular or vuggy porosity. Fracture porosity
can be superimposed on intergranular, vesicular or vuggy
porosity, creating more complex flow regimes. In geothermal
reservoirs, fracture porosity is by far the most common,
followed by intergranular; vesicular and vuggy porosities are
rare. Most reservoirs in igneous rocks have only fracture
porosity. High permeability and relatively low porosity are
typical of fracture porosity. The producing zones at The
Geysers, for example, have permeabilities up to several darcies,
but porosities are on the order of only 2 to 5%. On the other
30
-------
hand, sedimentary reservoirs with intergranular pore geometry
usually have moderate to low permeability and high porosity.
Permeabilities of 200 millidarcies with 20% porosities/ are
typical for the sedimentary reservoirs in the Imperial Valley.
Fracture porosity may also be superimposed on intergranular,
vesicular or vuggy porosity thereby creating more complex flow
regimes.
Volcanic flows, especially basalts, sometimes have vesicular
pore geometry. In this type, permeability is not directly
related to porosity. A highly porous pumice may have very low
permeability due to the lack of interconnection between the
voids. Vuggy porosity, a term generally applied to reservoirs
in carbonate rock, also gives rise to highly variable
permeability, depending upon the continuity of the vugs (solu-
tion voids).
The variation of hydraulic conductivity with temperature
around production and injection wells in a geothermal reservoir
must be considered. Hydraulic conductivity, K, is a function of
permeability, k, specific gravity of the fluid, g, and viscosity
of the fluid, v, (see Section 1.4 on terminology), according to
the following equation:
K = Jsa
v
The high temperatures of geothermal systems cause fluid viscosity
to fall more rapidly than the specific gravity thereby increasing
hydraulic conductivity up to several hundred percent compared to
that for normal ground water temperatures.
The extraction and injection of fluids for geothermal
energy production will change the natural flow regime. Location
and depth of producing and injection wells, as well as the rate
of withdrawal and injection, will greatly affect the flow
dynamics, and consequently the productivity of the reservoir.
Mathematical modeling of various combinations of injection and
production rates, locations, and depths superimposed on the
natural reservoir dynamics will aid in optimizing energy
production.
In addition to the fluid flow regime, it is important to
define the heat transfer mechanisms in a geothermal reservoir
before designing the energy production-injection system. The
long-term effects of injecting relatively cool liquid waste into
a geothermal reservoir must be evaluated. These effects depend
on the permeability, density, thickness and heat capacity of the
reservoir rock, regional fluid flow rate, spacing of injection
and production wells, rates of production and injection and the
31
-------
temperature of the injected fluid. A conceptual basis for
understanding these interactions can be provided by the
hypothetical reservoir models described below.
2-4.1 Hypothetical Reservoir Models
Several hypothetical reservoir models are discussed in this
subsection so the reader may develop a sense of the flow
dynamics in a producing reservoir. The models presented show
streamlines and the advance of cold fronts from the injection
wells in three types of production-injection patterns.
The reservoir and fluid properties for the first hypotheti-
cal reservoir are presented in Table 2.2 and their derivation
is summarized below.
Data for typical values of depth, area, gross thickness and
temperature were synthesized from Renner, et al. (1975). The
porosity, permeability, salinity and net thickness values were
assumed to be reasonable average estimates. The density and
viscosity were taken from standard tables for the specified
temperature, pressure and salinity of the fluid. Depth versus
pressure data for several reservoirs were used to derive the
hydrostatic gradient. The remaining parameters were computed
from these assumptions.
A downhole pumping exploitation scheme for the reservoir is
generated using the data in Table 2.2 and representative flow
rate of 0.063 cu m/sec (1000 gpm) per well. Four assumptions
will help define the exploitation scheme:
1) The reservoir is cylindrical.
2) The hottest part is at depth in the center.
3) The temperature declines towards the edges.
4) All produced fluid will be injected.
Hence, the production wells will be placed in the center of
.the reservoir to exploit the hottest fluid. The cylindrical
shape suggests that the injection wells be placed in a circular
array in the peripheral cooler part of the reservoir. Fig. 2.5A
illustrates a pattern such as this with a symmetrical well
arrangement where the injection rate per well is double the
production rate; therefore, there are twice as many producing
wells as inaction wells. Fig. 2.5B illustrates an idealized
flow (stream line) pattern for a hypothetical production scheme
rate ll^L^Tl^^ W inJ"Cti°n well^^The produc^on
rate is the same as the injection rate for each well.
32
-------
TABLE 2.2 PROPERTIES OF THE HYPOTHETICAL RESERVOIR
(Meidav and Sanyal, 1976)
Depth
Subsurface area
Gross thickness
Net thickness (permeable zone)
Bulk volume
Reservoir temperature
(highest near the center,
cooler near the edges)
Gross stored heat
Gross electrical potential
Net electrical potential
Expected life of a 200 MWe plant
Porosity
Permeability
Brine salinity (IDS)
Brine density
Brine viscosity
Hydrostatic gradient
Reservoir pressure
Maximum flow rate per unit
pressure drawdown
(Maximum productivity index)
1800m
26 sq km
1200m
300 m
31.7 cu km
138°-163°C
1.09xl019 J
3440 MW* centuries
69 MW*centuries
35 years
15%
lOOmillidarcies
5000 ppm
0.93 gm/cu cm
0.2 centipoise
0.0009 MPa/m
IS.OMPa
144kg/sec/MPa
(6000 ft)
(lOsq mi)
(4000 ft)
(1000 ft)
(7.6 cu mi)
(280°-325°F)
(2.6x 1018cal)
(58 Ibs/cu ft)
(0.43psi/ftV
(2600 psia)
(7980 IbsAr/psi)
(17.4 gal/min/psi)
(588 barrels/day/psi)
(93.5 cu m/day/psi)
33
-------
A.
-------
Table 2.3 gives the downhole pumping exploitation charac-
teristics of the hypothetical reservoir for a 50 MWe power
plant. In this development a production rate of 0.063 cu m/sec
(1000 gpm) will result in a calculated pressure drawdown of .39
MPa (57 psi). The actual drawdown, however, will be greater
due to interference from nearby wells and poor completion
efficiency. The cones of depression of adjacent producing wells
will interfere with each other. The closer the wells, the
greater the interference and the greater the drawdown in each
well. Unavoidable limitations, such as the effect of well
casing perforations, formation damage and incomplete penetration,
all contribute to reduced well performance efficiency and
thereby greater drawdown than those calculated for ideal condi-
tions. Further details on these reservoir engineering aspects
are provided in Meidav and Sanyal (1976).
Some other examples of hypothetical reservoir production
and injection schemes are presented by Tsang, et al. (1977). A
semi-analytic method was used to compute the progression of the
stream lines and cold fronts. Two examples are shown on Fig.
2.6: one is a production-injection pair (doublet) and the
other is a scheme with nine production and five injection wells.
The flow lines, assumptions and specifications for each example
are shown on the figure. In the production-injection doublet
(Fig. 2.6A), the heating of the fluid in the rock matrix causes
the cold front to progress more slowly than the hydrodynamic
front from the injection well towards the production well. In
this case it will take 15 years for the first cold front to
reach the production well. Up to that time the production
temperature will not be affected by the injection. Then the
temperature will drop sharply, approaching the injection
temperature asymptotically.
In the 14 well scheme (Fig. 2.6B) it will take 106 years
for the first cold front to reach a production well. Several
of the wells have much longer breakthrough times due to the
interference of adjacent wells.
In modeling geopressured reservoirs, Pritchett, et al. (1977)
determined that with injection a long interval of constant
temperature reigns in the production wells. When the cold front
from the injection wells reaches the production wells, the
production temperature declines. In this model the production-
injection well array is symmetrically arranged in a rectangular
grid, with basically an equal number of production and injection
wells.
35
-------
TABLE 2.3 DOWNHOLE PUMPING EXPLOITATION CHARACTERISTICS OF
A HYPOTHETICAL RESERVOIR FOR A 50 MWe POWER PLANT
(modified from Meidav and Sanyal, 1976)
Parameter Val ue
Well Diameter (7-5/8 in OD casing) 17.5 cm (6.875 in ID)
Well Drainage Radius 150 m (500 ft)
Expected flow rate per well 0.063 cu m/sec (1000 gpm)
Number of production wells 36
Number of injection wells 18
Injection rate per well 0.126 cu m/sec (2000 gpm)
Flowing bottom hole pressure at I1.6MPa (1680 psia)
producing well
Flowing bottomhole pressure at l9.9MPa (2890 psia)
injection well
Flowing production wellhead 0.52-0.69 MPa (75-100 psia)
pressure
Shut-in production wellhead pressure 0.34 MPa (50 psia)
Production wellhead temperature 150°C (300°F)
(average over 30 years)
Injection wellhead temperature 66 C (150 F)
Injection wellhead pressure 0.69-1.38 MPa (100-200 psia)
36
-------
A. SPECIFICATIONS'-
B. SPECIFICATIONS
Distance between wells-300 m.
Thickness of Aquifer - 100 m.
Porosity = 10%
Flow rate in each well = 100 m3/hr
Thickness of Aquifer^ 300m.
Porosity = 20 %
F!ow rate in each well = 193 m^/hr
40O 6OO
1600 m
"• I Natural flow
• I I m/yr
EXPLANATION
Streamlines
Cold temperature fronts
31 Number of years for front to
get to this location
O Injection Well
• Production Well
^ASSUMPTIONS, (Both Cases)
0.5 col/cm /°C - Volumetric Heat Capacity
of Rock
0.006 cal /cm/sec/°C- Thermal Conductivity
of Rock
Figure 2.6 A.
B.
A production-injection doublet
A system of nine production and
five injection wells
CTsang, et al. 1977)
37
-------
2.5 THE CHARACTER OF THE GEOTHERMAL EFFLUENT
2.5.1 Chemistry
The chemical characteristics of geothermal fluids through-
out the world vary over a wide range, both in number of chemical
species and their concentrations. For example, TDS ranges from
about 50 to 388,000 parts per million (ppm) and pH from 2 to 10
units. Geothermal waters can vary from entirely benign and
potable to highly corrosive and saline. The fluid charac-
teristics may vary from one reservoir to another, from one well
to another in the same reservoir, and over time in the same
well. For example, at East Mesa, California, the TDS ranges
from less than 2,000 ppm in Mesa 5-1 to more than 30,000 ppm
in Mesa 6-1 (Littleton and Burnett, 1978). In the Salton Sea,
California, field the composition varies from about 100,000 to
387,500 ppm, and fluid composition from individual wells can
vary more than 100,000 ppm (Palmer, 1975).
Knowledge of the chemical composition of geothermal waters
can provide much useful information about the reservoir, since
the kinds and amounts of constituents depend on the reservoir
environment: formation lithology, rock-water interaction,
rock-mineral-chemical equilibria, pressure and temperature. A
geographical variation in chemical characteristics can be
attributed mainly to variation in the nature of the subsurface
rocks, temperature, and distance from the source of recharge.
Temporal variation in the chemistry of geothermal fluids at a
particular site can have a number of causes, the most important
being the variation in the rate of fluid recharge (natural or
artificial) into the reservoir.
Although it is difficult to make a meaningful comparison of
geothermal fluid chemistry from various parts of the world, some
general correlations can be made. Low TDS is usually associated
with relatively low concentrations of each constituent. Hotter
waters tend to contain higher concentrations of constituents.
More saline waters appear to have a lower pH.
Because of the chemical diversity in geothermal waters, it
is difficult to arrive at an average value for concentration of
each constituent. As more data become available, statistically
significant mean and median values of each constituent may be
determined, at least for certain geographical areas. The median
value of TDS for geothermal water will most likely be in the
5,000 to 10,000 ppm range. Most geothermal fluids appear to be
acidic; that is, with pH less than 7.
In addition to dissolved and suspended solids, geothermal
water and steam contain a variety of noncondensible gases, some
of which may be detrimental to the surface environment. Hydrogen
38
-------
sulfide, a noncondensible gas constituent of many geothermal
fluids, has drawn considerable attention at The Geysers,
California, because of its odor. The other major noxious
components of many geothermal vapors are ammonia, carbon
monoxide, sulfur dioxide and mercury. Usually, noncondensible
gases constitute between 0.3% and 5% of the volume of the flashed
steam from geothermal fluids (Wood, 1973).
Small amounts of various toxic constituents are often
present in geothermal fluids. These include arsenic, boron,
nickel, zinc, rubidium, strontium and barium.
Geothermal fluids usually contain certain radioactive
elements in low concentrations, mainly radon, radium and
isotopes of uranium and thorium. The most thoroughly studied
radioactive element in geothermal fluids is radon^22^ a radio-
active gas. A study of 136 natural geothermal springs showed a
range of 13 to 14,000 picocuries per liter (pCi/1), with a
median around 510 pCi/1 (O'Connell and Kaufman, 1976).
Effluent from most vapor-dominated systems contains
significant amounts of hydrogen sulfide, which has an offensive
odor and is highly toxic. Ammonia, carbon dioxide, methane and
boric acid are also present in steam at The Geysers, but have not
posed a problem. Mercury vapor is common in vapor-dominated
systems and could pose a health hazard if allowed to accumulate
in the environment. Radium in geothermal vapor may precipitate
from solution and the radioactivity could be hazardous if radium
accumulates in the scale in transmission pipes. Radon is found
in the noncondensible gas fraction of some geothermal steam, but
unless trapped by a long period of atmospheric inversion it would
not be a health or safety hazard. The boron content in the
liquid fraction of The Geysers system has made injection of
liquid waste imperative there.
Liquid-dominated systems have essentially the same vapor
chemistry as the vapor-dominated systems. The liquid fraction
generally has high concentrations of salt and trace metals, and
a variety of other contaminants. Corrosion and surface and
ground water contamination can be serious problems in developing
liquid-dominated systems.
Small releases of radioactive elements may result from rock
fracturing. However, these will probably not exceed nuclear
power industry standards. Air and water quality problems in
development of hot dry rock systems are not expected to be as
severe as in convection systems, since introduced fluids are
circulated through the hot dry rock in a closed cycle. Scaling
and clogging could occur if the fluids become saturated with
minerals in the fractured rock chamber.
39
-------
Fluids in geopressured systems have high methane concentra-
tions. The geothermal liquid probably is highly saline and
corrosive.
Because electricity generation requires hotter waters, the
geothermal effluent is often more saline and generally poorer
than that from nonelectric uses.
Ch.emicals, such as acids or bases used for pH adjustment,
may be added to geothermal fluids to minimize scaling and
corrosion or to remove certain constituents. Although these
chemicals are not expected to be highly detrimental in them-
selves, they may contribute to degradation of the geothermal
liquid.
Chemical Characteristics and Water Use—
Chemical composition data for geothermal fluids is
summarized and compared with the chemical composition of drink-
ing water, selected regulatory standards and sea water in Table
2.4 and Fig. 2.7.
Geothermal waters may contain a high concentration of trace
elements compared to meteoric and sea water. Geothermal waters
in most instances cannot be used as a domestic or irrigation
water resource, due to high salinity and high concentrations of
trace elements, especially arsenic, boron, barium, fluoride,
manganese and zinc. These waters also contain significant
quantities of lithium, an element so unusual in ordinary water
that it is used as a tracer in ground water studies.
Because the trace element concentrations in water from
various geothermal fields and even from individual geothermal
wells in the same field differ substantially, hazards due to
trace element concentrations in geothermal fluids must be
individually evaluated.
A good background discussion of water quality with respect
to potential use appears in the section titled "Relationship of
Quality of Water to Use" in Hem (1970) and in Appendix A,
"Identification Systems and Criteria Used in this Report" in
California Department of Water Resources (1970).
2.5.2 Temperature
The temperature of spent geothermal effluent depends upon
its original temperature and the type and efficiency of
power plants range from about 45 to 120°C (110 to 250°F).
Waste heat can be significant in electric power generation.
As much as 85% of the available heat is wasted because of the
40
-------
TABLE 2.4
COMPARISON OF INORGANIC CHEMICAL WATER STANDARDS WITH
GEOTHERMAL AND SEA WATER ANALYSES (Geonomics, 197«b)
1
Substance
Arsenic
Barium
Bicarbonate
Boron
Cadmium
Calcium
Chloride
Chromium
Copper
Fluoride
Hydrogen
Sulfide
Iron
Lead
Magnesium
Manganese
Mercury
Nitrate
Selenium
Silver
Sodium
Sulfate
Zinc
TDS
pH
Drinking Water3 (mg/1)
USPHS USPHS
Recommended Mandatory
0.01
.
-
_
_
^
250h
~ U
i.oh
1.7
0.3h
-
-
0.05
-
45
_
c
2S°h
5
500h
-
0.058
1.0*
-
_
0.01g
_
0.05*
-
2.2
0.05h
-
0.05g
-
h
0 002"
10S
0.01g
0.058
.
-
_
-
6.5-8.5
Irrigating Water
(Ppm)
Threshold Limiting
1.0
-
"
0.5
_
_
100
0.1
-
-
-
-
-
-
_
_
_
-
200
.
500
7.0-8.5
5.0
.
-
2
_
_
350
1.0
-
-
-
-
-
-
_
_
_
-
1,000
-
1,500
6.0-9.0
Livestock Feeding
Water*5 (ppm)
Threshold Limiting
1
-
500
_
5
500
1,500
-
1
-
-
250
-
-
200
_
-
1,000
500
,
2,500
6.0-8.5
.
-
500
_
_
1,000 0
3,000
-
-
6
-
-
500
-
-
400
_
-
2,000
1,000
-
5,000
5.6-9.0
Geothermal
Water0
(ppm)
Range
0- 12
0- 250
0- 10,000
0- 1,200
0- 1
0- 63,000
0- 240,000
.
0- 10
0- 35
0.2- 74
0- 4,200
0- 200
0- 39,000
0- 2,000
0- 10
0- f35
trace
0- 2
0- 80,000
0- 84,000
0- 970
47- 390,000
2- 10
Sea Water
mg/1
0.003
0.03
142
4.6f
trace
400
19,000
-
0.003
1.3
0.01
tracef
1,350
0.002
trace f
• 0.5
0.004
trace^
10,500
2,700
0.01
34,560
-
' USPHS, 1962; EPA,
Todd, 1970
, Tsai, et. al., in
Goldberg, 1963
e Includes NO,', NH
1976; EPA, 1977a
press
, and dissolved nitrogen gas
Trace = <0.001 ppm (or mg/1)
* Maximum contaminant level specified in National Interim Primary
. Drinking Water Regulations (EPA, 1976)
Maximum contaminant level specified in National 'Secondary Drinking
Water Regulations (EPA, 1977a)
-------
B
Br-
C,'+
HCOj
cor
Cl
F~
I
NO,
Rh+
SiO,
sol~
TDS
10 '
i+ltUltllUUUUIUUIIUIUUWl lllllllllUt HIIIH
tin/l//t/n
-------
inefficiency of low-temperature conversion. Depending on the
cooling method, this heat is dissipated to the air (as with
cooling towers), discharged to surface water (from once-through
cooling water), or disposed of with the geothermal fluid.
In a total flow system, such as that proposed by Lawrence
Livermore Laboratory for the hot corrosive brines of the Imperial
Valley, most of the thermal energy—both liquid and vapor—is
used for power production; the temperature of the spent fluid is
low. in contrast, a one-cycle flashed steam system uses only
the steam fraction to drive the turbine. The remaining
unflashed water, comprising approximately 2/3 of the total mass
of produced fluid is discarded at high temperatures. For
example, it is estimated that a fluid from a 223°C (434°F) well
will be discharged from the conversion facility at 167°C (338°F).
2.5.3 Volume of Fluid
Several factors affect the volume of spent fluid discharged
from a geothermal conversion facility. These factors include:
type of system, use of the energy, capacity of the facility,
design of the conversion system and type of cooling method.
Liquid waste from hot dry rock systems may be minimal.
Water used for recovery is generally recycled through the hot
reservoir.
Much more fluid by weight is produced by liquid-dominated
fields than by vapor-dominated fields. The Wairakei field, for
example, produces 11.3 cu m/sec (180,000 gpm) of combined waste-
water, steam and condenser effluent. This is approximately 200
times the fluid volume produced by The Geysers per unit of
electricity (Swanberg, 1976) .
Plants developed for direct use of geothermal energy are
generally of a smaller scale than electric power facilities and
do not generate as much spent fluid. Such uses are space
heating, industrial process heating and agricultural applications
such as aquaculture, greenhouse heating and lumber processing.
2.6 RESOURCE RECOVERY
Resource recovery refers to the extraction of heat from the
geothermal reservoir. Four aspects of injection related to
resource recovery are: (1) maintenance of reservoir pressure;
(2) replenishing the geothermal fluid supply; (3) recovering
the latent heat content of the reservoir rocks; and (4) possible
cooling of the geothermal reservoir fluid.
In most geothermal systems the duration of sustained fluid
production or recharge mechanisms is not well defined. Properly
43
-------
emplaced artificial recharge, by injection, may alleviate these
concerns. Natural recharge of the res.ervoir may not be
sufficient for the sustained flow rates necessary for electric
power production. Recharge by injection may make up the dif-
ference between the natural recharge and the fluid production
necessary for power production.
If the voidage created in the reservoir by fluid extraction
is not filled the hydrostatic pressure will decline. Declining
pressure will cause declining productivity at the wellhead. If
the pressure is allowed to drop sufficiently then steam will
form in the reservoir. While steam does extract heat from the
rock matrix, it is not as efficient as water as a heat transfer
medium (Nathenson and Muffler, 1975). Additionally, steam from
a low-pressure reservoir may not provide the temperature and
pressure necessary at the wellhead for electric power generation.
In addition to providing fluid for a discharge-recharge
balance, the injected fluid may provide additional recoverable
heat. The total heat content of a geothermal system consists of
heat in the rock matrix plus heat in the water. Generally, the
water contains a much smaller proportion of heat than the rock
matrix. Cooler fluid injected into the reservoir will provide
a medium for transfer of more of the heat energy contained in
the rock. Injection wells must be far enough from production
wells so that the fluid residence time is long enough to
sufficiently reheat the fluid. If fluid is injected too close
to the production wells, or in an otherwise inappropriate
location, it may cool the existing heated fluid and reduce power
production. Hence, proper location of injection wells is
critical.
44
-------
REFERENCES
Berry, F. A. F. 1966. Proposed Origin of Subsurface Thermal
Brines, Imperial Valley, California (abs.). Bull. Am.
Assoc. Pet. Geol. Vol. 50, No. 3. pp. 644-645.
Biehler, S. 1964. Geophysical Study of the Salton Trough of
Southern California. Ph.D. diss. California Institute of
Technology. unpublished. 139 pp.
California Department of Water Resources. 1970. Geothermal
Wastes and Water Resources of the Salton Sea Area.
Bulletin 143-7. 123 pp.
Craig, H. 1966. Isotopic Composition and Origin of the Red Sea
and Salton Sea Geothermal Brines. Science. Vol. 154.
pp. 1544-1548.
Dutcher, L. C., W. F. Hardt and W. R. Moyle, Jr. 1972.
Preliminary Appraisal of Ground Water in Storage with
Reference to Geothermal Resources in the Imperial Valley
Area. USGS Circular 649. 57 pp.
Ellis, A. J. 1966. Volcanic Hydrothermal Areas and the
Interpretation of Thermal Water Composition. Bull. Vole.
29. pp. 575.
Facca, G. 1973. The Structure and Behavior of Geothermal Fields.
In: Geothermal Energy (Earth Sciences 12), Paris, The UNESCO
Press. pp. 61-72.
Geonomics, Inc. 1978a. Baseline Geotechnical Data for Four
Geothermal Areas in the U.S., Report for U.S.E.P.A.,
Environmental Monitoring and Support Laboratory, Las
Vegas, Nevada. 338 pp.
Geonomics, Inc. 1978b. Subsurface Environmental Assessment
for Four Geothermal Systems. Report For U.S.E.P.A.,
Environmental Monitoring and Support Laboratory, Las
Vegas, Nevada. 240 pp.
Helgeson, H. C. 1968. Geologic and Thermodynamic Characteristics
of the Salton Sea Geothermal System. Am. Jour. Sci. V. 266.
pp. 129-166.
4.5
-------
Hem, J. D. 1970. Study and Interpretation of the Chemical
Characteristics of Natural Water. USGS Water Supply
Paper 1473. 363 pp.
Hunt, T. M. 1971. Net Mass Loss from the Wairakei Geothermal
Field, New Zealand. Proceedings of the U.N. Symposium on
the Development and Utilization of Geothermal Resources,
Pisa, 1970. pp. 487-491.
Littleton, R. T., and E. Burnett. June 1978. Chemical Profile
of the East Mesa Geothermal Field, Imperial Valley,
California. In: Proceedings of the Second Workshop on
Sampling Geothermal Effluents, February 15-17, 1977, Las
Vegas, Nevada. EPA-6QO/7-78-121. pp. 175-189.
Meidav, T., R. James and S. Sanyal. 1975. Utilization of
Gravimetric Data for Estimation of Hydrothermal Reservoir
Characteristics in the East Mesa Field, Imperial Valley,
California. Stanford Geothermal Program Workshop on
Geothermal Reservoir Engineering and Well Simulation,
Stanford University. pp. 52-61.
Meidav, T., and S. K. Sanyal. December 1976. A Comparison of
Hydrothermal Reservoirs of the Western U.S. Electric
Power Research Institute Topical Report ER-364. Project
580. Prepared by Geonomics, Inc. 170 pp.
Nathenson, M., and L. J. P. Muffler. 1975. Geothermal Resources
in Hydrothermal Convection Systems and Conduction Dominated
Areas. In: Assessment of Geothermal Resources of the
U.S.-1975. D. E. White and D. L. Williams, eds. USGS
Circular 726. pp. 104-121.
O'Connell, M. F., and R. F. Kaufmann. March 1976. Radioactivity
Associated With Geothermal Waters in the Western United
States. U.S. E.P.A., Office of Radiation Programs, Technical
Note ORP/LV-75-8A. 34 pp.
Palmer, T. D. 1975. Characteristics of Geothermal Wells Located
in the Salton Sea Geothermal Field, Imperial County,
California. Lawrence Livermore Laboratory UCRL-51976.
54 pp.
Pritchett, J. W., S. K. Garg and T. D. Riney. 1977. Numerical
Simulation of the Effects of Preinjection Upon the
Performance of a Geopressured Geothermal Reservoir.
Geothermal Resources Council. Transactions. Vol. 1,
May 1977. pp. 245-247.
Renner, J. L., D. E. White, D. L. Williams. 1975. Hydrothermal
Convection Systems. In: Assessment of Geothermal Resources*
U.S. Geological Survey Circular 726. pp. 5-57.
46
-------
Rex, R. W. 1970. Investigation of Geothermal Resources in the
Imperial Valley and Their Potential Value for Desalination
of Water and Electricity Production. University of
California, Riverside. 14 pp.
Swanberg, C. A. 1976. Physical Aspects of Pollution Related
to Geothermal Energy Development. Proceedings, Second
UN Symposium on the Development and Uses of Geothermal
Resources, pp. 1435-1443.
Tsai, p., S. Juprasert and S. K. Sanyal. 1978. A Review of
the Chemical Composition of Geothermal Effluents. In:
Proceedings, Second Workshop on Sampling and Analysis of
Geothernal Effluents, February 15-17, 1977, Las Vegas,
Nevada, sponsored by EPA. pp. 84-96.
/
Tsang, C. F., M. J. Lippmann and P. A. Witherspoon. 1977.
Production and Reinjection in Geothermal Reservoirs. In:
Geothermal State of the Art. Geothermal Resources Council
Transactions. Vol. 1. pp. 301-304.
White, D. E., L. J. P. Muffler and A. H. Truesdell. 1971.
Vapor Dominated Hydrothermal Systems Compared With Hot
Water Systems. Econ. Geology. Vol. 66, No. 1. pp. 75-97,
White, D. E. 1968. Environments of Generation of Some Base-
metal Ore Deposits. Econ. Geol. Vol. 63, No. 4. pp.
301-335.
White, D. E., and D. L. Williams, eds. 1975. Assessment of
Geothermal Resources of the U.S.-1975. USGS Circular 726.
155 pp.
Wood, B. 1973. Geothermal Power in Geothermal Energy Review
of Research and Development, ed. by H. C. H. Armstead.
UNESCO, Paris. 186 pp.
47
-------
SECTION 3
THE MONITORING METHODOLOGY
The uniqueness of the natural environment and human
activities in each potential development area prevents specifi-
cation of one "monitoring plan" (see Section 1.4 on terminology)
that would apply to all sites. The ground water "monitoring
methodology" for geothermal development described below provides
a set of guidelines to aid in designing a site-specific monitor-
ing plan.
The methodology is a synthesis of approaches described in
several publications, governmental regulations and policies as
well as the personal experience of the authors and consultants.
It provides a systematic and comprehensive step-by-step outline
to develop a site-specific ground water monitoring plan for
geothermal development. However, it is by no means a cookbook,
and each step requires evaluation and analysis for each site.
It is important that personnel applying this methodology have
a thorough knowledge of geology, hydrology and geochemistry.
The six steps of the methodology (Fig. 1.2) are:
1) Define geologic, hydrologic and geochemical
baseline conditions as well as projected
reservoir and plant development;
2) Forecast aquifer conditions in the geothermal
reservoir as well as in overlying aquifers;
3) Define temporal, spatial and chemical limits
of detection;
4) Evaluate monitoring techniques for both
injection and observation wells;
5) Design monitoring plan and alternatives;
6) Implement monitoring plan.
The methodology is applicable to release of geothermal
fluid on the ground surface as well as below the surface. The
discussions emphasize monitoring, the major continuing ground
48
-------
water degradation threat of fluid injection. An accidental
spill, blowout or other mishap may release geothermal fluid on
the ground surface. However, this is a temporary, unpredictable
and often a practically unavoidable situation not subject to the.
detection problems of a subsurface release. Whenever such an
incident occurs, the problem is likely to be detected and treated
immediately. Unless preventive measures are taken (see Section
3.1.5) the fluid may percolate into a near-surface aquifer or
discharge into surface water.
For a surface release, the study of baseline conditions, as
specified in the methodology, will allow estimation of the rate
and direction of percolation and the chemical reactions of an
effluent with the near-surface sediments. A monitoring program
could also be set up, along the guidelines outlined in the
methodology, to detect the actual progress of the near-surface
degraded fluid front. This type of potential degradation due to
percolation from the surface is treated in more detail in Todd,
et al. (1976).
The steps in the methodology would generally be applied in
the order shown in Fig. 1.2 and are discussed below.
3.1 DEFINE BASELINE CONDITIONS AND PROJECTED DEVELOPMENT
Baseline conditions are those that exist prior to develop-
ment. These conditions may be natural or already altered by
human activities. The critical determination at this point is
to determine what exists now. This is necessary to provide a
basis for later comparison—to compare the results of the chosen
monitoring plan with the recognized datum established by the
baseline data acquisition task.
It is unlikely that any area will initially have baseline
data sufficient to make a complete environmental assessment for
ground water monitoring. Therefore, during the first phase,
each data category should be evaluated for completeness. Some
questions which might be asked are: is the stratigraphy known
in sufficient detail? are the hydrologic characteristics
adequately defined? is the distribution of wells sufficient for
adequate monitoring, especially in terms of depth, distribution
and sample density? are any wells particularly precarious with
respect to becoming polluted or providing a potential pollutant
Pathway? is the temporal distribution of historic chemical data
sufficient to establish patterns of temporal variation or
consistency in water quality?
Determination of data completeness will include considera-
tion of current regulations by EPA and others, as well as
knowledge of current and potential ground water use in the area.
Given the level of available data, and the knowledge these data
49
-------
have supplied, the investigators are in a good position to
evaluate which parameters are necessary to supplement the exist-
ing data and which can be deferred. After the initial data
gathering, synthesis and interpretation, a program should be set
up to supply the necessary supplemental data. The limitations
of the existing data should be explicitly defined.
To monitor for changes in ground water characteristics that
may occur during geothermal development and injection,the base-
line conditions that must be established include:
a. chemical characteristics of nongeothermal
ground water, geothermal ground water, and
surface waters;
b. geology and hydrology of the area;
c. location, well use and well completion data
for all wells in and around the geothermal
site; and
d. mechanics and characteristics of the geothermal
system.
These are discussed in the following subsections. Many of
the individual parameters that must be synthesized in this
analysis are summarized in EPA Administrators Decision State-
ment No. 5 (Appendix A). Table 3.1 summarizes desired subsur-
face baseline data elements and methods applicable for their
determination. Applications of borehole logging to baseline
data acquisition are discussed in Section 4.
3.1.1 Chemical Characteristics of Waters
Adverse changes in the chemical characteristics of ground
water provide the prima facie evidence of degradation. Hence,
determination of the characteristics is one of the most
important data sets that will be developed in this phase of
study. Data must be collected for all waters in the area,
including geothermal and nongeothermal ground water, surface
water, and any other disposeu water. The goal of this data
collection is to establish for each water type (including
industrial, municipal and agricultural releases):
1) chemical characteristics;
2) three-dimensional (spatial) distribution;
3) natural temporal variations or cycles;
50
-------
TABLE 3.1 SUMMARY OF DESIRED SUBSURFACE BASELINE DATA ELEMENTS
AND METHODS AVAILABLE FOR THEIR EVALUATION
(modified from TEMPO, 1973)
Baseline Data Element
Porosity
Permeability
Fluid pressures in formations
Water quality
Thickness and character of geologic
formations
Mineral content of formations
Temperature of formations
Amount of flow into various horizons
Identification and extent of aquifers,
aquitards and structural features
Piezometric surfaces
Hydrologic budget
Methods Available for Evaluation
Core tests; electric, radioactive and sonic
logs
Core tests; pumping or infection tests;
injectivity profiles; electric, nuclear
and sonic logs
Drill stem tests; water-level measurements
Sampling and analysis; electric and
nuclear logs
Drill time logs; drilling samples; cores;
electric, radioactive, caliper and
televiewer logs
Analysis of drilling and core samples
Temperature log
Injectivity profile
Geologic mapping; drill cores and drillers
logs
Well canvass and ground water-level
measurements
Synthesis of precipitation, evaporation
and evapotranspiration, stream flow,
artificial inflow and outflow data
51
-------
4) in conjunction with the geologic data,
chemical reactions and changes as the water
flows through subsurface materials; and
5) mixing relationships, if any, of these
waters, and where mixing occurs.
The consistency and accuracy of chemical sampling plays a
critical role and is discussed below. The next subsection
describes a sample baseline water chemistry investigation for
Imperial Valley, California. It provides an illustration of an
approach to achieving the goals stated above.
Sample Collection and Chemical Analysis—
Sampling and analysis of waters and geothermal effluents
is a specialized field. Some of the more relevant references
are: Brown, et al. (1970); Wood (1976); Reed (1975); Ellis, et
al. (1968); Presser and Barnes (1974); EPA (1974); EPA (1976);
EPA (1978;; American Public Health Association (1977); and
Watson (1978). These works detail most of the step-by-step
procedures that should be followed in collecting and analyzing
water and geothermal effluent samples. The document by Shannon,
et al. (1978) is a standard methods manual specifically for
sampling and analysis of geothermal fluids.
The following factors can influence the consistency and
accuracy of the chemical analysis:
a) parameters determined at the time of sampling;
b) time interval between sample collection and
analysis;
c) type of sample containers;
c) method of sample preservation;
e) procedure for sample collection;
f) analytic techniques;
g) representativeness of the sample.
To get an accurate measure of certain volatile constituents
and physical properties it is important to determine their values
at the time of sampling. Parameters that should be measured for
all waters at time of collection are pH, temperature, specific
conductance, and content of carbonate, bicarbonate, Eh and
dissolved oxygen (Wood, 1976). For geothermal waters, aqueous
carbon dioxide,* aqueous hydrogen sulfide, aluminum and ammonia
52
-------
should also be determined (Reed, 1975; Presser and Barnes,
1974) .
For accurate representation of the original chemical
species, the time between sample collection and analysis should
be kept to a minimum. The longer the sample is stored the
greater the chance for chemical and physical reactions in the
sample bottle, such as oxidation, reduction, precipitation,
adsorption and ion exchange. Iron is a particularly troublesome
component in this respect because it tends to oxidize and
precipitate (Brown, et al. 1970, p. 14; Presser and Barnes, 1974,
P. 2).
The type of sample container and its preparation can influ-
ence the accuracy and stability of the sample. Current trends
are towards use of polyethylene, Teflon or other plastic
containers, although Presser and Barnes (1974) note that plastic
bottles are permeable to oxygen. In collecting geothermal
samples, care should be taken not to use soft-glass bottles
(citrate of magnesia type) when collecting silica samples. In
general, the nature of the solutions will influence the rate
and types of materials dissolved from the container (Brown, et
al. 1970). Each sample container must be cleaned properly
before use.
Samples require certain types of preparation to ensure
minimal chemical changes during shipping and storage. Brown,
et al. (1970, pp. 16-17) recommend collecting four samples from
each location: one that is filtered immediately upon collec-
tion, one filtered and acidified, one unfiltered and settled,
and one unfiltered and well mixed. Each would be analyzed for
different groups of constituents. The filtered, untreated
sample would be tested for TDS, halogens, alkali metals, boron,
nitrogen (nitrate and nitrite), phosphorus, selenium, silica
and sulfate. The second sample would be analyzed for alkaline
earth metals, alkali metals, transition metals, aluminum, arsenic
and lead. The third sample would undergo tests for acidity,
alkalinity, calculated carbon dioxide, color, and laboratory pH.
Analysis of the fourth sample would be for ammonia, organic
nitrogen, chemical oxygen demand, cyanide, phosphorus,
suspended solids, total volatile solids and turbidity. If the
ammonia determination is made in the field it may not be nec-
essary to collect this fourth sample for geothermal fluids.
Additional specially treated samples may be necessary for
specific determinations or analysis. One of these is a diluted
sample for silica analysis. The geothermal fluid would be
diluted (e.g., 10:1) with distilled water to minimize the
Possibility of silica precipitation upon cooling. This type of
sample would be very important for silica geothermometer deter-
minations in the exploration phase of a geothermal development
and for anticipation of scaling in the development phase.
53
-------
Sample collection procedures must be specified and
consistently applied. These are fairly straightforv;ard for most
nongeothermal surface and ground water and are well outlined in
several of the previously cited references. With geothermal
fluids, a problem arises in collecting a representative sample.
To obtain a representative sample from a superheated geothermal
source, most investigators recommend collecting both liquid and
vapor samples. Techniques used for such sampling are discussed
in detail in Ellis, et al. (1968); Truesdale and Pering (1974);
Giggenbach (1976); Hill and Morris (1975); EPA (1976); EPA
(1978); Finlayson (1970); and Watson (1978).
In sampling hot springs, geysers, etc. use of a small hand
or battery-operated pump with a long tube that can be inserted
in the hot water is often convenient. This procedure will allow
the sampler better access to the part of the manifestation that
would provide the most appropriate sample.
Although the methods used in water analyses are fairly well
standardized the accuracy of analytic results should not be taken
for granted. Studies have shown significant inconsistencies in
analytic results from reputable laboratories (Ellis, 1976;
Watson, 1978). Hence, it is always advisable in any sampling
program to submit replicates, spikes, blanks and splits.
Consistent results from these analyses give confidence that the
laboratory's results are accurate and reproducible. Sometimes
it may be advisable to submit these samples to a referee
laboratory to resolve discrepancies.
A sample is only a small fraction of the water body being
sampled. Hence, it is important to ensure that this sample is a
true representation of the entire water body being sampled. For
a pumping well it should be collected close to the wellhead.
For a surface water sample the condition and activity of dif-
ferent parts of the water body must be considered before sampling.
A sample collected closer to the wellhead will have less
opportunity for contamination than one collected farther away.
For example, in most domestic wells the water is stored in metal
storage tanks and flows through metal pipes before it gets to
the tap. This storage and conveyance can introduce elements to
the water that are not representative of aquifer conditions.
A sample from a pumped or flowing well will be more
representative of the aquifer fluid than a sample that is simply
bailed from a well. Hence, it is always preferred to pump an
unused well before sampling. If the well does not have a pump,
a portable submersible pump can be used. if aeration of the
sample is not a critical factor, a shallow well can be pumped
using an air stream from a small portable gas-driven paint
sprayer compressor.
54
-------
A surface water body such as a spring should be sampled as
close to its source as possible. For some other water bodies it
may be necessary to mix samples from different parts to produce
a representative sample.
For most of the historical data used in the baseline data
acquisition it will be impossible to determine many of the
previously discussed factors that affect the resulting chemical
data. Therefore, even though the data can be used in attempting
to decipher temporal and spatial patterns, it would be desirable
to start collecting baseline chemical data as soon as possible
in a consistent, prescribed, reproducible manner. This would
allow direct comparisons between consistent sets of chemical
data collected before development begins and after development
has commenced.
Water Chemistry Baseline Data Acquisition and Synthesis,
Imperial Valley, California—
An example of how a baseline data acquisition program may
approach the five goals mentioned previously is described in
Geonomics (1978a and 1978b). These two reports comprise a
baseline data acquisition and subsurface environmental assess-
ment based on historical data for the Imperial Valley, California.
A summary of selected aspects of the program follows. It is
included to provide one suggested approach.
Imperial Valley contains six Known Geothermal Resource
Areas (KGRAs) (Fig. 3.1). Imperial Valley comprises the regional
ground water regime for East Mesa KGRA and water chemistry data
have been collected for the entire valley. The regional picture
provides the framework for understanding local hydrologic
patterns.
The first step in this program was collection of all
available historical water chemistry data. Next, several graphic
techniques were used to synthesize and interpret the hundreds
of chemical analyses. These were Stiff diagrams, Langelier-Ludwig
Plots and single chemical parameter contour surfaces. These
techniques are described in Section 3.6.2 on data synthesis. The
results of the study are summarized below.
Chemical characteristics of Imperial Valley .waters—Five
chemically distinct types of water were identified (.Fig. 3.2)
from plotting Stiff diagrams of the historical data on maps of
Imperial Valley. They are:
1) Simple sodium chloride water. This water has
a TDS range from somewhat over 700 mg/1 in the
southeastern portion of the valley to over
13,000 mg/1 in the central portion. One oil
55
-------
TEMPERATURE GRADIENT MAP
OF THE IMPERIAL VALLEY CALIFORNIA
ON
STATE o* CAU^OHNI*
DIVISION OF OIL a GAS
K> 15 (Km.)
TEMPERATURE GRADIENTS
. . GREATER THAN IO" F
I J PER 100 FEET IN DEPT
LEGEND
AREAS WITH HIGH
TEMPERATURE GRADIENTS
FAULTS WITH REPORTED SURFACE
RUPTURE DURING HISTORIC TIME
SINCE (769
8° TO 10" F PER IOO FEET
IN DEPTH
FAULTS WHICH APPEAR TO DISPLACE
QUATERNARY ROCKS OR DEPOSITS
6°TO B°F PER IOO FEET
IN DEPTH
4" TO 6' F PER
IN DEPTH
CALUHTHIl X.
2°T04°FPER IOOFEET
IN DEPTH
LESS THAN 2°F PER IOO
FEET IN DEPTH
TEMPERATURE GRADIENT DAT*
COM PI LED A INTERPRETED BY
JIM COMBS, U.C. RIVERSIDE
SEPT. I 971
Figure 3.1 Temperature gradient map showing Known Geothermal Resource Area (KGRA)
locations in Imperial Valley, California. (Palmer, 1975)
-------
A. Typical Simple
Sodium Chloride Water
B. Typical Sodium Chloride
Water with High Sulfate
and/or Magnesium
9540
15200
C. Typical Sodium Chloride
Water with High Calcium
D. Typical High Sulfate
Water
503
15,430
E. Typical Sodium Bicarbonate
Water
790
950
1760
850
EXPLANATION
MODIFIED STIF
IN PERCENT F
SAMPLE _ Ca-?<
DEPTH ^-201-r
INTERVAL Mg2^J£
""X/20!
F DIAGRAMS
REACTANCE
XWELL NUMBER
r-HCOsf CO,
iW*
0 Cl
f.\.,....'P<>%
DOT ^ l"7 T
SIGNIFIES I \
g?VEKNVALUE TOTAL DISSOLVED
*IV 1 SOLIDS (mg/i)
SAMPLE DEPTH CODE:
D SHALLOW INTERVAL(80-300ft)
^ INTERMEDIATE 1 NTERVAL(300-l500ft)
ODEEP INTERVAL (1500ft)
Figure 3.2 Modified Stiff diagrams for characteristic
Imperial Valley ground waters.
(Geonomics, 1978a)
57
-------
test well, more than 4000 m (13f000 ft) deep,
contained a sodium chloride water with almost
53,000 mg/1.
2) Sodium chloride water with high sulfate and/or
magnesium. This water is typical of Salton
Sea water and ranges from 1,490 mg/1 TDS in
the shallow depth zone just south of the Salton
Sea to over 37,000 rag/1 in the sea itself.
3) Sodium chloride water with high calcium. These
are all geothermal waters with TDS contents
from about 12,000 to over 300,000 mg/1.
4) High sulfate water. This water is typical of
the Colorado River. It has TDS contents
under 1,000 mg/1.
5) Sodium bicarbonate water. This water is typical
of the intermediate depth artesian aquifer. It
contains from less than 300 mg/1 TDS in fresh
waters flowing off the Peninsular Range into
the shallow aquifer in the southwestern corner
of the valley, to over 2,600 mg/1 in the
intermediate depth zone under the northern part
of East Mesa.
Spatial distribution of water types—The ground water
regime in Imperial Valley was divided into three depth zones
for this study. The near-surface upper 24 m (80 ft) was
excluded from the natural baseline data because of possible
contamination from percolating agricultural return waters.
Hence, starting at 24 m (80 ft) these depth zones are defined as:
Shallow - from 24 to 91 m (80 to 300 ft),
Intermediate - from 91 to 457 m (300 to 1,500 ft), and
Deep - deeper than 457 m (1,500 ft).
To determine the distribution of ground water in these
subsurface zones, Stiff diagrams were plotted on maps of
Imperial Valley for each depth zone.
The simple sodium chloride water (Fig. 3.2[A]) occurs in all
three depth zones, but there is a statistically significant
correlation of this water with the shallow depth zone. Water of
this type is fairly well distributed throughout the area east of
the New River within the shallow depth zone. The sodium chloride
water occurs very sparsely in the intermediate and deep aquifers.
58
-------
The sodium chloride water with high sulfate and/or magnesium
(Fig. 3.2[B1) is present in only the shallow and intermediate
depth zones. In the shallow zone it occurs near inlets in the
southern portion of the Salton Sea and is typical of Salton Sea
water. Similar type waters with high sodium chloride and notable
sulfate occur in the shallow and intermediate depth zones in
southeastern Imperial Valley. A few samples of similar waters
occur in the shallow depth zone to the west and northwest of El
Centro and in the northernmost area of the intermediate depth
zone samples.
The sodium chloride water with high calcium (Fig. 3.2[C])
occurs only in the deep aquifer. It occurs in the Cerro Prieto,
Heber and Salton Sea geothermal fields and there appears to be a
distinct north-south trend between these occurrences. It is
Possible that there is a hydraulic connection between these
waters but most likely they are all undergoing a similar chemical
reaction.
The high sulfate waters (Fig. 3.2[D]) are typified by
average Colorado River water. This water occurs mainly in the
shallow aquifer and has a statistically significant correlation
with it. It occurs mainly beneath and very close to canals in
the southeastern portion of Imperial Valley, obviously a result
of leakage of river water from the canals. One sample of this
water occurs in the intermediate depth zone directly beneath the
All-American Canal and one occurs in the deep depth zone, in the
Dunes geothermal field, just to the northwest of the Coachella
Canal.
Although the sodium bicarbonate waters (Fig. 3.2[E]) in
Imperial Valley occur in all three depth zones there is a marked
statistically significant correlation of this type of water with
the intermediate depth zone. Almost 80% of the samples of this
type of water occur in the intermediate depth zone. It is
largely confined to the artesian aquifer area, between the Alamo
River and the East Highline Canal. Four of the samples of this
water type also occur in the shallow depth zone Dust west of the
Elsinore fault trace, in the southwestern corner of Imperial
Valley.
Two distinct trends can be noted in the shallow depth zone.
First! ground water Increases in salinity from about 800 mg/1
in thi Southeastern corner of the valley to over 15,000 mg/1
approaching the center of the valley and the Salton Sea. The
salinity Sen decreases to the west. This trend is well
illustrated by the schematic surface of specific conductance
values (Fig. 3.3).
Secondly, water chemistry changes from * ^ ™' hjf
sulfate water in the southeastern corner of the valley, to a
59
-------
.33.50
N
(Ti
O
14,000 micromhos
Conlour interval
= 1000 micromhos
Figure 3.3 Schematic surface of specific conductance values
for Imperial Valley ground water. (Geonomics, 1978a)
32.5<
-------
simple sodium chloride water in the East Mesa area, to a sodium
chloride water with high sulfate and/or magnesium and high TDS
approaching the Salton Sea.
Temporal variations—Temporal variations in surface and
ground water quality must be considered. Seasonal variations
will be found in surface water quality and will probably be
reflected in ground water quality. Unfortunately, historic data
for this type of correlation are often quite limited.
Colorado River water at Yuma has varied in TDS from less
than 500 mg/1 to a current level of 700 to 800 mg/1. This trend
of increasing salinity will continue as more water is diverted
from the river above Yuma for irrigation.
The salinity of the Salton Sea has varied greatly through
its history, depending on variations of sea volume, mineral
content and quantity of inflowing waters and precipitation of
saturated ions. It has ranged from about 3,600 mg/1 in 1907
to 40,000 mg/1 in 1925 and it is currently around 36,000 mg/1.
The highest salinity was measured in 1936 at 213,000 mg/1.
A few wells in Imperial Valley produced historic chemical
data, but not enough for meaningful analysis. The most
important trend noted was anomalous analytical results between
two ends of a sampling period. Well No. 393, a shallow well in
the extreme southwestern Imperial Valley is an example. The
samples for 1967 and November 1972 are quite similar sodium
carbonate chloride waters with a somewhat smaller calcium content
in November 1972 than in 1967. The TDS content was 360 mg/1 in
!967 and 341 mg/1 in November 1972. However, a sample taken in
July 1972 exhibited distinctly higher sulfate and a TDS content
of 455 mg/1. This example again points out the necessity to
remain aware of possible undetected changes in water characteris-
tics between sampling periods and to specify adequate sample
intervals for each site.
Mixing relationships—Langelier-Ludwig (L-L) diagrams are
very useful for detecting the origin of waters and mixing rela-
tionships. In the Imperial Valley study distinct groupings
occur. Although there is some scatter, a definite concentration
°f shallow interval samples occurs in a zone with greater than
90% reactance* (S04+C1). Between approximately 55 and 85%
reactance (Na+K) (Fig. 3.4) of intermediate depth samples occur
°r
Percent reactance =
species (in milliequivalents/liter [meq/1])
sum of all anion species(in meq/1)
species (in meq/1)
" "" ', mk ' / * "~ „ ' / T \
sum of cations(in meq/1)
61
-------
S04 + CI (Percent Reactance)
1C
100
I
500(H),29r
SEA WATER
SALTON -"*
SEA WATg
(5 814
,!
,1. .
(3 31 1
(S 1813, 90S
^ ^
CD
U
£
jj
O.
£T
"c
1
OJ
n_
1 '
— '-
+
0
~Z.
T
0 uu
trTs 7' •L'r^V '^l?7Cl^jHf7 **« OEDD
1
X#^^* WW^"7 *" £2V !*' '"'A *"" ^™ ON SECTION A-D
^65 ^701 9Q ff5
!P MIB B357
a 142 &718 A7M
8 228 420
Q4 503JH) 369 B B
1*06
^79 *^H) |<( B258
03& ^^* B 355
p J53 «4' H320
T{8]
^404, 227
•
• 145 ^ B^86 B?4e
^^ B3BO
•91
^"B"?
OM "364
5 79
^B374
67 363
_ • 0J93
• 604ID)
p9« 133 134 378
ei2(s:
QCOLORADO RIVEH WATER
a144
0%,
Gift ^
2 (^ t^
mi2 ^^
^^
• H"" ^ O
zp ^%
To
mS
EXPLANATION
399 Ground water analysis and well number for
shallow sample (sample depth 80-3OO ft)
70I Ground water analysis and well number for
intermediate sample (sample depth 3OO- 1,500ft)
Ground water analysis, well number and KGRA
8«'EI code for qeothermal fluids (sample depths
greater than 1,500 ft)
KGRA code
(B) Brawley
(C) Cerro Prieto
(D) Dunes Thermal Anomaly
(E) East Mesa
( H ) Heber
(S) Salton Sea
Q Surface water analysis ( sources noted )
Area which includes samples projected onto
respective salinity sections
B C
i i i i 1 1— 1 ' '
0
, „
CD
O
O
"o
0
a>
Cd
"c
.
-------
in a zone with greater than 90% reactance CNa+K) and between
approximately 25 to 95% reactance (S04+C1) . Distinct clusters
in this zone occur between 25 and 30 and between 40 and 52%
reactance (SO4+C1) .
Most of the geothermal waters plot above 98% reactance
(S04+C1) or (Na+K) (Fig. 3.4). The Salton Sea wells occur at
100% reactance (SO4+C1) and between 64 and 69% reactance (Na+K) .
Three of the East Mesa wells plot at 98% reactance (Na+K) and
between 57 and 87% reactance (S04+C1) .
Concentration by evaporation can be seen in the lower
portion of an L-L diagram salinity section (Fig. 3.5) where the
Salton Sea geothermal fluid shows a relatively constant ratio of
chemical constituents steadily increasing in salinity. Salinity
section A-D (not shown) shows a very dense cluster of points in
a triangular pattern. This is suggestive of complete mixing of
the three waters having compositions represented by the apexes
of the triangular patterns. These are: a fairly pure water of
low salinity, a sodium bicarbonate water of about 60
milliequivalents per liter (meq/1) salinity, and a sodium
chloride water of about 110 meq/1 salinity.
Chemical reactions with subsurface materials — Some general
chemical changes were previously noted for the evolution of
Imperial Valley water in the "Spatial Distribution of Water
Types" subsection. In addition, to aid in tracing the possible
genesis of ground waters in Imperial Valley, chemical reactions
that could alter Colorado River source water were considered
Olmsted, et al. (1973).
In this approach recent Colorado River water is assumed to
undergo certain common geochemical reactions in combination with
various degrees of evaporative concentration. The results
of these reactions are computed, providing a set of known
chemical causes and effects on Colorado River water. By
comparing these known changes with analyses of actual waters it
is possible to get a suggestion of the processes the water has
undergone in the ground. A sample of some of these changes, as
shown by modified Stiff patterns, is presented in Fig. 3.6.
Since many people may be accustomed to seeing Stiff patterns
Plotted in milliequivalents per liter, this figure also compares
modified Stiff patterns plotted in percent reactance with those
Plotted in milliequivalents per liter to illustrate the charac-
teristics of these different representations.
The first chemical change shown, that of simple evapora-
tion, shows the percent reactance modified Stiff pattern remain-
ing the same size and shape with increasing concentration (Fig.
3*6). The modified Stiff pattern plotted in milliequivalents
63
-------
100
SECTION A-B
Q SURFACE WATERS
•E GEOTHERMAL FLUID - KGRA CODE
6 INTERMEDIATE SAMPLE
D SHALLOW SAMPLE
No * K ( Percent Reactance )
100
No + K (Percent Reactance)
Figure 3.5 Salinity section A-B from Langelier-Ludwig diagram
for ground water data in Imperial Valley, California.
(Geonomics, 1978a)
64
-------
U1
IX
T3OJ
730
COLORADO
RIVER
WATER
IX
2X 3X
CONCENTRATION BY EVAPORATION
4X
1460
EVAPORATED a SOFTENED
2920
I«SO ,
1450
1260
1140
1130
1130
mo
1110
2X
2920
' 2920
EVAPORATED & PRECIPITATED
EXPLANATION
MODIFIED STIFF DIAGRAMS PLOTTED IN MILLI EQUIVALENT PER
LITER AND PERCENT REACTANCE FOR COMPARISON OF
CORRESPONDING PATTERN CHANGES
MILLIEOUIVALENTS
PER LITER
HCOj
No
PERCENT REACTANCE
Co HCO3
Mg ) \ S0«
\
Hn
Cl
2010
SUM Of DISSOLVED SOLIDS. IN MILLIGRAMS PER LITER
5X
2060
2060
8X
3230
IOX
EVAPORATED, PRECIPITATED, 25 % SULFATE REDUCTION
3040
EVAPORATED, PRECIPITATED, 40% SULFATE REDUCTION
3570
1570
"570 ?350
SAME AS ABOVE, PLUS HARDENING EQUAL TO NET INCREASE IN SULFATE
2900
1550
1550 2320
SAME AS ABOVE, EXCEPTHARDENING UNTIL CI=Na IN MILLI EQUIVALENTS PER LITER
"""•
3X
2690
1520
4X
Figure 3.6 Modified Stiff diagrams representing hypothetical analyses of ground
water resulting from specified chemical changes in Colorado River
water. (modified from Olmsted, et al. 1973).
-------
per liter maintains a similar shape but elongates markedly with
increasing concentration due to evaporation.
Other chemical changes that may be expected in ground water
in this area are softening, carbonate precipitation, sulfate
reduction, hardening, resolution of precipitated salts, oxida-
tion of dissolved organic substances and mixing of waters of
different chemical composition. The last three processes would
be difficult to represent and calculate meaningfully so they are
not presented in Fig. 3.6.
Softening is the replacement of calcium or magnesium, the
hardness-causing constituents, by sodium. This generally occurs
by cation exchange with clay minerals. This change is illus-
trated in the second line of diagrams in Fig. 3.6.
When water with high bicarbonate and calcium or magnesium
content is sufficiently evaporated, precipitation of ca~lcium
or magnesium bicarbonate occurs. To distinguish between loss
of calcium and magnesium from softening and loss from carbonate
precipitation one can look for a reduction in bicarbonate, which
occurs with carbonate precipitation but not with softening. This
process is labeled "evaporated and precipitated" on the third
line of diagrams in Fig. 3.6.
Sulfate reduction is believed to be a major process occur-
ring in Yuma area ground water, an area very similar to Imperial
Valley. The process of sulfate reduction is poorly understood
but is known to be organic. The last four lines of diagrams in
Fig. 3.6 involve sulfate reduction.
Hardening, the reverse of softening, is a base-exchange
process where sodium is replaced by calcium or magnesium. Like
softening, hardening results from reactions with clay minerals.
The last two lines of diagrams in Fig. 3.6 involve hardening.
Comparison of the different chemical reactions shown in
Fig. 3.6 shows that the TDS content may be a poor measure of
the amount of evaporative concentration of the sample. The
chemical reactions the sample has undergone also significantly
influence the TDS. This can be seen in Fig. 3.6 by comparing
the TDS content (2,920 rag/1) of the sample with 4X "concentra-
tion by evaporation" with that (2,900 mg/1) of the sample with
10X concentration by "evaporated, precipitated, 40% sulfate
reduction".
3.1.2 Geology and Hydrology
The goals of the geologic and hydrologic investigation are
to determine:
66
-------
1) distribution of aquifers and aquitards,
including their lithology and mineralogy;
2) ground water flow rates and patterns, water
levels, transmissivities, recharge and dis-
charge areas and hydrologic budget; and
3) the location and extent of structural or
stratigraphic features that may provide
potential pollutant pathways and boundaries.
A fairly complete hydrologic and geologic investigation will
necessary to make these determinations. Particular emphasis
t be placed on determining the properties of the injection
horizon and its confining strata. In most cases the geothermal
reservoir will be the injection horizon. Again, this task
should start with existing data from the literature, private or
governmental agency files or reports, personal contacts,
Researchers in the area, etc. These data should then be analyzed
ln the context of the three goals mentioned above. One approach
^0 achieving these goals is outlined below.
Distribution of Aquifers and Aquitards—
Ideally, the areal extent and depth of aquifers and
ards as well as lithology and mineralogy can be directly
Derived from a closely spaced array of geologic and geophysical
logs. However, aquifer and aquitard units often must be
indirectly from limited well log data, with substantia-
from geologic maps and cross sections.
Extrapolations from only geologic data can lead to
Complications. For example, in attempting to correlate a geologic
j;ormation with a hydrologic unit (aquifer or aquitard) one may
tind that a geologic formation, as mapped, does not coincide with
a hydrologic unit. In addition, the geologic data may not
Provide sufficient subsurface detail for the hydrologic inves-
t;i9ation, especially for units as deep as geothermal reservoirs.
Sometimes the nature of the formations makes definition of
specific hydraulic units difficult or impossible. For example,
j-11 Imperial Valley, the units are of fluvial, alluvial and
^^custrine origin, consisting mainly of interbedded and lensing
sandstones and shales. All are cut by myriad fault traces.
Iri this environment, the lack of vertical or horizontal
Continuity causes difficulty in identifying continuous hydraulic
units, and only major discontinuities can be defined.
Detailed vertical lithologic profiles can be derived from
geophysical well logs described in Section 4 and from cores.
samples, ideally from cores, are necessary for detailed
67
-------
mineralogic and petrographic determinations. Before cores are
available, analyses of rocks from surface exposures of mapped
units will have to suffice. An important aspect of this task is
determination of the mineralogy of the injection horizon to
evaluate possible chemical reactions of the injected fluid with
the formation.
Ground Water Flow Rates and Patterns—
This will make up the largest part of the geologic and
hydrologic investigation. The determinations to be made in this
category include water tables and piezometric surfaces,
permeabilities and transmissivities, ground water flow direc-
tions and velocities, recharge and discharge areas and hydrologic
budget.
Available water table and piezometric surface data should be
supplemented by the data that is collected in the water well
canvass (see Section 3.1.3). These include maps of ground water
elevations and depths for each aquifer in the area. Ground
water flow directions can be seen on the ground water elevation
maps and hydraulic head can be computed. Implications about
permeability and recharge and discharge areas can also be
derived, but these extrapolations must be corroborated by further
data.
Permeabilities and transmissivities can often be extrapolated
from geologic data for sedimentary deposits with intergranular
permeability. However, most geothermal areas occur in igneous
or metamorphic terranes with fracture permeability, which is
much more difficult to estimate. It would be desirable to
conduct pump tests for these determinations.
Once the hydraulic head is determined from the water level
elevation map and the permeability is found or estimated, the
ground water flow velocity can be calculated.
The drainage basin determinations of recharge, discharge
and hydrologic budget can be derived from data on precipitation,
stream flow, evapotranspiration, soils, land and water use,
water level and quantities of water imported. Some of these
data will be derived from the water well canvass (Section 3.1.3).
Location of Structural and Stratigraphic Features—
Structural and Stratigraphic features can act as potential
pollutant pathways and boundaries. Faults play an extremely
important role in the hydrologic and geothermal systems. Faults
may provide structural control for the location of geothermal
reservoirs; they may provide conduits for the lateral and
vertical flow of*geothermal fluids and fresh ground water; and
68
-------
they can act as aquitards and aquicludes in the hydrologic
system. Stratigraphic features such as buried stream channels
can also influence ground water flow. Special attention must be
Paid to the detail, completeness and accuracy of fault locations
and other relevant structural and Stratigraphic features.
In many areas, the number of located faults is directly
Proportional to the detail of mapping done in the area, hence
the absence of faults on a published map is not necessarily
an accurate representation of the actual situation. This is
Specially true in an area like the Imperial Valley where most
fault traces are concealed. Here the detail of the fault traces
mapped are in direct proportion to the detail of geophysical
surveys conducted (Geonomics, 1978a).
3»1.3 Well and Fluid Discharge Data
Regardless of the amount and comprehensiveness of current
and historical v/ell and fluid discharge data the investigators
should conduct their own canvass and sampling program of wells
and fluid discharge sources (such as individual or municipal
yaste disposal sites). There are five main reasons for conduct-
lng this survey:
1) to determine or supplement hydraulic head
distribution data;
2) to establish or supplement ground water and
aquifer characteristics in the area with the
known procedures for water sampling used in
the monitoring plan;
3) to gain firsthand experience in locating,
inspecting and sampling the wells in order
to judge their potential usefulness as
monitoring wells;
4) to learn about abandoned wells in the area and
their potential for use in monitoring or
providing pollutant pathways; and
5) to identify other industrial, municipal or
domestic existing or potential sources of
ground water degradation.
The well and fluid discharge canvass should begin with
^tailed study of well and fluid discharge locations on
topographic maps. This would be followed by field verification
and investigation of the sites identified on the maps, and a
search for additional sites.
69
-------
Data collected in the field investigation for wells would
include as much of the following as possible:
1) well location and altitude;
2) owner;
3) water level;
4) cased depth, well depth;
5) construction, including year drilled, casing
diameter, casing material, grouting, packers,
perforation types and intervals;
6) drilling logs or other well logs;
7) previous chemical analyses;
8) water and well use (including amount);
9) previous water levels and dates;
10) type of pump;
11) yield of well;
12) water samples for analysis;
13) other relevant observations and impressions.
Preparation of a form outlining all these parameters would
help in reducing inconsistencies and omissions in the survey.
The goal of this phase is to collect as much information as
possible about the wells and their use.
Data similar to that for the water wells should be collected
for other fluid discharge sites, including data on waste fluid
disposal methods and quantity of wastes disposed. Later, if
any of- the monitoring wells show water degradation, knowledge of
the characteristics of the other potential pollutant sources
will help determine if the degradation is caused by the geo-
thermal fluid or another source.
In the next phase maps will be prepared showing the dis-
tribution of geothermal wells, nongeothermal wells, and fluid
discharges. These maps will show the relationships of ground
water use and fluid discharge to the geothermal well locations.
Of particular significance is the number, use and depth of
wells near and downgradient from the geothermal wells. A sample
map for a preliminary water well survey for the East Mesa KGRA
70
-------
(Fig. 3.7) shows several water wells around and downgradient
from the geothermal wells. However, almost none of these wells
is used or yields significant quantities of water.
Wells in poor condition that may provide conduits for
inter-aquifer pollution may be checked using cement bond, caliper,
borehole televiewer and pipe inspection logs CSee Section 4).
Particular attention should be paid to wells penetrating the
deeper aquifers.
3.1.4 The Geothermal System
To consider the interaction of the geothermal system with
the nongeothermal ground water regime, both systems must be
defined as accurately as possible. Undoubtedly less data will
be available for the geothermal system and much of the
definition will be based on models and extrapolation.
Aspects of the geothermal system that need to be defined
are:
1) the size, shape and boundaries of the reservoir;
2) the temperature and pressure of the reservoir
fluid;
3) the type of system;
4) the chemistry of the reservoir fluid;
5) the lithology and mineralogy of the reservoir; and
6) the recharge source for the reservoir.
This information will provide an idea of how the geothermal
system and nongeothermal system may be related in terms of
fluid chemistry, geologic structure and rock type, degree of
hydraulic connection, and common recharge sources. This infor-
Nation will allow estimation of the resource potential, which
will affect the projected development. These parameters are
discussed in the following paragraphs.
The size, shape and boundaries of the reservoir may be
aPproximated by geologic, surface geophysical and borehole log
data. Various types of resistivity, telluric, magnetotelluric
and microseismic techniques have been used in exploration for
and definition of geothermal reservoirs. Geologic structural
data is often used to define reservoir boundaries. For example,
it is hypothesized that the reservoir at Roosevelt Hot Springs,
Utah, is bounded by four faults, including the Dome fault to the
West, the Negro-Mag fault to the north, a mountain front fault
71
-------
EXPLANATION^
O GEOTHERMAL WELL
• DOMESTIC OR INDUSTRIAL WELL
• OTHER TYPE OF WELL
9 DOMESTIC/INDUSTRIAL AND OTHER TYPE OF WELL
(|) GEOTHERMAL AND OTHER TYPE OF WELL
NOTE: |N ORIGINAL REFERENCE EACH WELL WAS NU MBERED
AND CROSS REFERENCED TO WELL DESCRIPTION AND
CHEMICAL ANALYSIS TABLES.
5 10 Miles
HblTVILLE r
* —-
Figure 3 . 7
Preliminary survey of wells in and near East Mesa
KGRA (modified from Geonomics,
72
-------
to the east, and several smaller faults to the south. However,
accurate definition of the reservoir requires many drill holes!
Even in The Geysers, California, where over 120 wells have been
Drilled, the extent of the reservoir remains ambiguous.
The temperature of the reservoir fluid can be estimated by
various types of geothermometers applied to geothermal fluids
emanating at the surface and to fluids produced from wells.
Those based on silica solubility and on sodium-potassium-calcium
ratios are the most commonly used (Fournier, et al. 1974;
Pournier and Truesdell, 1974). Application of either of'these
techniques is based on many assumptions about the reservoir'and
genesis of the fluid. Since the accuracy of the assumptions
cannot be verified without detailed drill hole and geologic
data, the accuracy of the resulting estimates is questionable.
Results from the silica geothermometer are particularly subject
to suspicion due to the inability to identify the dissolved
silica species. Each silica mineral has a different solubility
and the assumption of the dissolved species will determine the
temperature computed.
Even temperatures measured in flowing wells will represent
°nly one location and depth for the reservoir. Experience has
snown that bottomhole temperatures vary considerably for
different wells in liquid-dominated reservoirs. For example
Production wells drilled in the Salton Sea field have '
Bottomhole temperatures ranging from less than 200°C (390°F)
to over 350°C (660°F) (Palmer, 1975). The variation may be
^-ess in steam-dominated reservoirs.
Dri11 holes are necessary to determine reservoir pressures
and to verify the type of system. Although surface geophysical
urveys (e.g. gravity) can provide indications of whether the
Ystem is liquid- or vapor-dominated, drill hole verification
•^s often necessary.
The chemistry of geothermal fluids in general is discussed
n Section 2.5.1 and sampling and analysis are discussed in
3.1.1.
Lithology and mineralogy of the reservoir are discussed in
ection 3.1.2 on Geology and Hydrology and in Section 4 on bore-
u°le logging.
Isotope and tracer studies can aid in determining recharge
ources. An isotope study of recharge sources for the Imperial
«J.iey geothermal systems determined that geothermal fluid in
-LJ- the systems, except the Salton Sea field, is derived from
oiorado River water (Coplen, 1972). Much water from the hot
1ft ""1**° t0 the south and east of the Salton Sea is derived from
3°cai precipitation. Tracer techniques are discussed in Section
'•4.5.
73
-------
3*1.5 Projected Development
Knowledge of the size and type of projected development
is necessary to estimate:
1) the volume of fluid to be extracted from the
reservoir and disposed of by injection or
other means? and
2) the chemical and physical changes that the
fluid will undergo from production through
disposal.
This information is important to predict the effects
geothermal development may have on the ground water system. To
accomplish this goal the following parameters must be defined:
1) the reservoir development plan; and
2) the proposed heat extraction and power produc-
tion processes.
The reservoir development plan will specify optimum
production and injection rates and well locations. This plan
will produce necessary information on how the natural geothermal
system will be stressed and if that stress may also influence
overlying aquifers.
Ideally, the geothermal reservoir under natural conditions
is largely isolated from overlying aquifers. In some areas,
however, such as Roosevelt Hot Springs, Utah, the geothermal
fluid naturally mixes with some of the overlying ground water,
probably as the fluid rises along faults that intersect the
overlying aquifers. Stressing the natural reservoir conditions
by pumping and injection could increase or decrease the effects
of natural mixing or could induce a hydraulic connection between
overlying aquifers and a reservoir that was formerly isolated.
Evaluation of these effects will be discussed in Section 3.2.
The type of energy use, conversion facility, and waste
disposal method will determine the change in temperature and
chemistry of the geothermal fluid and the type of monitoring
program to implement. In most energy conversion processes the
fluid will come out more saline than it started. The actual
change depends on the individual process and could vary from
minimal to significant. This change must be specified by the
process engineers. The chemical characteristics of the post-
process geothermal fluid must be compared with those of natural
geothermal fluid and nongeothermal ground water. In addition
to increased salinity, certain constituents which were in
innocuous concentrations prior to energy extraction may reach
potentially hazardous levels.
74
-------
Chemical reactions in the injection well and formation will
be influenced by temperature drops between production and
injection. Pretreatment of the waste fluid before injection
may be necessary to reduce potential scaling, corrosion and
formation plugging problems. A detailed overview of geothermal
injection technology, including descriptions of the chemical and
Physical problems that may be encountered, is presented in
Section 5.
Spent fluid disposal will most often be by injection and the
injection horizon will be specified by the reservoir engineer.
Jf a surface disposal method is chosen, the monitoring plan
Will have a different focus than if injection is used. This
difference will manifest itself in such ways as the depth and
spatial distribution of monitoring wells, determination of
Potential degraded fluid plumes and percolation patterns, etc.
Monitoring plans for surface sources are more fully described*
in Todd, et al. 1976.
Regardless of the disposal method chosen, the "preventive
^onitoring" philosophy can be applied in the design phases of
Development. Proper well design, construction, and monitoring
^111 vastly reduce the possibility of leakage through the
disposal well (see Section 3.5.4). Preventive measures for
surface disposal methods, as well as surface conveyances and
facilities used with injection, can also be planned. Possible
schniques include pressure monitors with alarms and automatic
snut-off valves, small levees around the facility, and an
impermeable material placed on or beneath the ground surface.
One additional aspect of projected development is detailed
valuation of the specific geologic and hydrologic conditions
*• the site, e.g., wells that have been drilled at The Geysers.
nen landslides started moving, some well casings were sheared
Busing uncontrolled blowouts. Hence, landslide potential is
°w a major consideration in planning geothermal facilities at
^ne Geysers. Another example of a site-specific geologic
onsideration is in Imperial Valley, where subsidence potential
Jjst be carefully evaluated because of the widespread network
r gravity-fed irrigation canals.
3'2 FORECAST AQUIFER CONDITIONS
Forecasting the interaction between geothermal and non-
jSothermal aquifers may help avoid potential problems. Base-
ine data acquisition should provide the data necessary for this
ask. The goal of this task is to determine the effects of the
woposed production and injection plan on the existing ground
«ter system. This analysis will consider the following factors
75
-------
1. models of the ground water and geothermal
systems;
2. potential pollutant mechanisms and pathways;
3. chemical reactions in the aquifers; and
4. effects of alternative development scenarios.
3.2.1 System Models
Models of ground water and geothermal ground water systems
can be fairly simple or very complex. They may involve only a
qualitative conceptual framework or a detailed, three-dimensional/
multivariable computerized mathematical representation. In a
ground water monitoring plan, the goal of any model is to predict
the movement of a particular fluid front. No model can be as
detailed as a natural system, but the more accurately the model
represents the actual field situation, the better it will predict
the migration of fluid fronts.
Additionally, all models are based on many assumptions,
simplifications and boundary conditions as well as the physical
and chemical parameters of the actual system. These factors and
limitations must be kept in mind when analyzing the results.
Ideally, the model will predict the effects of geothermal
production and injection on the existing ground water system.
To accomplish this, the model must incorporate much informa-
tion about the hydrologic, physical and chemical properties
of the system. Because some of these parameters may not be
available at the planning stages, the model may later have to
be refined. Selection of the most appropriate model will be
based on the level of information on subsurface conditions.
In the initial stages of the monitoring investigation, data
may be sufficient only for a simple, qualitative conceptual model
of the system. Eventually, a model that can evaluate the three-
dimensional combination of physical and chemical properties is
desirable. This model will include quantification of the spatial
and temporal distribution of:
1) hydraulic head;
2) temperature;
3) density;
4} chemical constituents; and
5) effects of geothermal fluid production and
injection on the existing ground water system.
76
-------
With this type of model, various sets of production and
injection plans could be put in to evaluate the effect of alter-
nate development scenarios on the ground water system.
Intercom?, Inc. (1976) has developed a transient three-
dimensional, subsurface, waste disposal, ground water model that
includes evaluation of single-phase fluid flow, convective and
conductive heat transfer and mixing of aquifer and injected
fluids. This is currently the most comprehensive, publicly
available model capable of evaluating a geothermal injection
situation. Input data requirements for such a model are quite
extensive and require a fairly detailed definition of the system
and the physical properties of each component.
A review of developments in ground water modeling over the
Past decade is presented in Narasimhan and Witherspoon (1977) .
A survey of ground water modeling in the USGS is presented in*
Appel and Bredehoeft (1976). This source lists all USGS ground
water modeling computer programs and their status of develop-
^ent. Modeling categories range from relatively simple two-
^imensional problems of ground water flow to complex three-
dimensional models involving fluid mixing and heat gradients.
3»2.2 Potential Pollutant Mechanisms and Pathways
Chemical and thermal pollution of ground water aquifers
Curing injection of waste can result from the following:
1. improperly constructed or deteriorated injec-
tion well;
2. improperly constructed, deteriorated or
ineffectively abandoned wells nearby;
3. escape of injected fluid from the receiving
formation through structural or stratigraphic
pathways;
4. hydrofracturing of confining formations with
high-pressure injection;
5. accidental spills at the ground surface;
6. percolation from storage ponds (enhanced by
higher temperatures);
7. percolation from discharge of mineralized fluids
through leaks in surface conveyances which are
part of the injection system;
8. chemical migration through confining beds due
to osmotic forces.
77
-------
These potential pathways and mechanisms are discussed below.
Improper construction, deterioration or failure of well
seals would allow fluids to flow vertically up or down the well
bore, depending on where the failure occurred (Fig. 3.8[A]}.
Casing failure could occur by corrosion (Fig. 3.8[B]). This
mechanism can occur in the injection well or other wells in the
area. They may be abandoned, used or infrequently used wells.
At The Geysers production wells drilled on landslides have blown
out when landslides were reactivated and the downslope movement
sheared the well casing.
Fig. 3.9 shows a hypothetical example of a potential escape
path for injected fluid through an abandoned well and another
well penetrating an aquifer overlying the confining bed of the
injection aquifer. This case illustrates an example of an
improperly plugged abandoned well, where the cement plug is
placed far above the perforated interval of the well. This has
allowed fluid to flow upwards in the well bore, through the
confining bed, into an overlying aquifer that is penetrated by
another well. This pathway could also exist by virtue of
deteriorated well seals around the casing or corroded casing.
To identify locations where this mechanism may occur it is
important to survey, where possible, the condition of all wells
in the area that penetrate deeper aquifers. In unplugged wells
this may be done by running cement bond logs.
Structural and stratigraphic pathways, such as faults,
fractures, ineffective caprock or buried stream channels may
allow fluid to travel along pathways not previously recognized
(Fig. 3.10). Hydrofracturing of confining formations due to
high-pressure injection may also create structural pathways in
the form of micro-fractures or joints. Hydrofracturing is
discussed in Section 5.5.3.
Accidental spills at the surface, percolation from holding
ponds, or leakage from surface conveyances would entail similar
pathways. The fluids would percolate from the surface downward
directly into the nearer surface aquifers. A spill, if not
contained, may also discharge fluid directly to surface streams,
lakes or canals.
Osmotic forces can cause slow migration of chemical con-
stituents of the waste fluid to a ground water aquifer through
an intervening caprock, which may act as an osmotic membrane.
However, pollution due to this effect is anticipated to be
minor and insignificant.
Although escape of fluids by any of these mechanisms is
of concern, the greatest risk of fluid escape is through the
injection well itself (Fig. 3.8) (Talbot, 1972). Currently
78
-------
:••*•••:••.••••••••-: J
'&••/&.• gravel .-^•fr
:•:•'.:.' potable water ]'•:
injected fluid
-cement seal
'^Mii^ii^^
^impermeable shale-^j-
limestone
n
I!
CO
rn
^impermeable shale -j
;'salt-water"':; v//
'. sandstone;/ 'V:;
Figure 3.8 Escape of injected fluid through
deteriorated cement seal (A) and
through hole in casing (B).
(modified from U.S. EPA, 1977}
pressure
monitor-
jjpotable 'water -V:
-£jmpermeable -
. shale-E
ffi
limestone
£
—— impermeable '-^r.
•. salt- wafer
;.' sandstone
Figure 3.9
Escape of injected fluid through
abandoned boreholes in area.
(modified from U.S. EPA, 1977)
79
-------
pressure
monitor
wffiA^ttv^v^B
potable water
:.••.•. sandstone.
impermeable shale_-_
impermeable shale —
••!:' 'permeable salt-:.'/-.
:.;•.:••>;.water sandstone v|'.';.'
—_—__— impermeable shal_e^-£ -£T
Figure 3.10 Escape of injected fluid through
fractures (A) and faults (B).
(modified from U.S. EPAf L977)
80
-------
Prescribed well construction practices and the large vertical
distances between the injection zones and usable aquifers,
reduce the probability of contamination of usable aquifers.
Evidence for this conclusion is the scarcity of reports of"
direct contamination from this type of source. (Tempo, 1973,
Pp. 2-9.) Rigorous planning and monitoring programs are
necessary to maintain this record.
In the special case of injection of cold water into a hot
ary rock formation for heat recovery, the resulting temperature
gradient in the rock should be considered. The differential
thermal stress fields created by the cold water may result in
excessive fracturing. Research has been done on this problem bv
the Futures Group (1975). *
Baseline geologic, hydrologic, and well data, as well as
input from system models, should be evaluated to predict poten-
tial pollutant mechanisms and pathways. The most obvious areas
that must be analyzed are the role each fault may play and the
Possibility of other wells intersecting the injection zone and
Providing a pathway for interaquifer leakage.
To evaluate the role of faults in an area with concealed
*ault traces, the investigators may need to conduct some surface
Geophysical surveys. An electrical resistivity survey was
conducted in the Salton Sea field which located at least four
Previously unidentified faults and confirmed the locations of
two others (Meidav, et al. 1976). It is hypothesized that these
^ bound the geothermal reservoir.
Geophysical well logging can be used to evaluate the subsur-
conditions of wells that may provide channels for inter-
Aquifer pollution. Descriptions of the logs to run for this
Purpose are presented in Section 4.
_ Unknown structural or stratigraphic pathways may be
iscovered by conducting tracer studies. A naturally occurring
racer is preferable since it would be in the system for a long
period and therefore would reveal any pathways that may take a
°ng time to show up. Lithium, a unique constituent of some
|eothermal fluids, may serve as an ideal natural tracer. A
imitation of this approach is that it will only detect "pathways
ccurring under natural pre-production conditions. Those that
tay be induced by production activity will only show up at that
t The baseline water chemistry study will also provide
^formation on potential pollutant pathways. In comparing the
istry of geothermal water with other wells and springs,
ng relationshiPs maY be derived. Determination of where
how this mixing occurs will provide clues to where geothermal
81
-------
waters mixing with ground water may increase or decrease with
production and injection.
Some questions to ask are:
- Do any ground water barriers show on the water
level contour maps? (Do they coincide with
mapped fault locations?)
- Does the construction of any used or abandoned
well indicate that it may provide a potential
pollutant pathway?
- Does the water chemistry data indicate anomalies,
discontinuities or spatial trends?
- How will the superposition of the production and
injection flow dynamics affect the confinement
of the reservoir? How will it affect flow in
overlying aquifers?
- Which faults are acting as conduits and which as
barriers to ground water flow?
- Which geologic units will confine injected fluids
and which will protect near surface aquifers?
- Are there any man-made subsurface structures
or excavations such as mines or tunnels that
may present a hazard?
3.2.3 Chemical Reactions in the Aquifer
The main chemical reactions that may be expected in the
injection zone are those that cause formation plugging around
the injection well. Silicates and carbonates cause most
plugging in geothermal systems. Other reactions should be
minimal since the composition of the injected fluid will be very
similar to that of the resident fluid in the injection zone.
The zone of reaction around geothermal injection wells is
probably similar to that around other types of industrial waste
disposal wells. A gradational boundary (zone of reaction)
develops between the injected and the formation fluids. In this
zone, some minerals will be dissolved or altered, metastable
sols and gels will form and new compounds will precipitate. The
extent of this zone will depend upon the length of time the fluid
has been injected and its flow rate, the mixing rate of the
injected fluid and formation water, and chemical reaction rates.
Although chemical reactions may be predicted by laboratory
mixing of the two fluids, the physical effect on the system is
82
-------
difficult to predict. A precipitate, for example, may immedi-
ately plug the formation, may pass through without plugging it
(especially if fracture permeability exists), or may slowly
accumulate and retard injection. Formation plugging can also be
caused by swelling clays due to formation water incompatibility.
However, this problem is restricted to sedimentary reservoirs.
Bacterial growth, which causes plugging around some nongeo-
thermal injection wells, should not be a problem in geothermal
injection wells. Few bacteria thrive in such high temperatures.
If the fluid is not filtered prior to injection, particulate
and colloidal matter may plug the formation.
3.2.4 Effects of Alternative Development Scenarios
This evaluation stresses the concept that there may be more
than one way to approach each development situation. If the
initial proposal looks like it may produce a hazardous condition
in usable aquifers, then the components producing that situation
should be altered. Situations that may develop are:
1) indications that the planned injection pressure
or expected increased pressure to maintain
the injection rate may hydrofracture the
confining formations of the reservoir;
2) indication from modeling that the planned
production and injection rates may have a
deleterious effect on the reservoir or over-
lying formations, e.g., changes in formation
pressure, subsidence, change in natural
recharge to shallower used-aquifers, etc.;
3) the planned location of an injection well is near
a hydraulically conductive fault;
4) indications that the temperature of the injected
fluid may contribute to plugging of the
injection zone;
5) geophysical well logs indicating that casing or
cement completion is not adequate.
If any of these, or other potentially hazardous situations,
recognized, an alternative plan should be developed before
disposal begins.
3'3 DEFINE LIMITS OF DETECTION
The monitoring plan should be designed to detect chemical
in the ground water. These changes should be perceived
83
-------
as occurring in a spatial and temporal matrix. The necessary
chemical, spatial and temporal sensitivity of detection in the
matrix must be specified for each area. The chemical sensitivity
required is mainly a function of:
a. chemical contrast of geothermal and nongeo-
thermal fluids;
b. environmental sensitivity to particular
constituents;
c. natural variations in water characteristics;
and
d. available analytic techniques.
The temporal and spatial sensitivity required is mainly a function
of:
a. hydrologic factors;
b. the relative size of the development;
c. characteristics of potential pollutant path-
ways; and
d. water use and well distribution density in the
area.
These parameters will be analyzed, based on pertinent
regulations by EPA and other agencies, to help determine sampling
frequency, distribution and density of sample points, significant
chemical and physical parameters, and methods of sampling and
analysis. Where ground water is used extensively, rapid detec-
tion of minute changes in quality would be more desirable
whereas in areas where ground water is poor and not used, larger
variations and less frequent sampling may be acceptable. Con-
sideration of the required limits of detection will ensure that
the monitoring plan is not over- or under-sensitive. The
chemical, temporal and spatial sensitivity factors are discussed
in detail below.
3.3.1 Chemical Detection Sensitivity
The chemical contrast of geothermal and nongeothermal fluids
will be defined in terms of 1) the magnitude of the chemical
difference between them, 2) identifiable characteristics of each,
and 3) their degree of natural mixing in different parts of the
study area. If the difference between the fluids is not
significant, then no ground water degradation will occur. The
monitoring program will then consist only of periodic sampling
84
-------
and analyses of production, injection and nongeothermal fluids
to ensure their continuing integrity.
The two fluids may be similar if the quality of geothermal
fluid is quite good or if nongeothermal ground water is of low
quality. For example, in the vapor-dominated Matsukawa geo-
thermal field, steam condensate has almost the same composi-
tion as that of water in a nearby river. In fact, this
condensate has been used with no adverse effect to irrigate
local rice paddies. On the other hand, some of the nongeothermal
9round water in the central part of the Imperial Valley has
salinity levels of over 15,000 ppm higher than some of the
geothermal fluids produced in the area.
It would be desirable if either the geothermal or nongeo-
thermal fluid had a unique chemical characteristic or "signature",
This would be an element, compound or isotope that: 1) occurred
only in one type of fluid in the area; and 2) remained
identifiable throughout the chemical reactions of the fluid.
An isotope study in Imperial Valley has been cited in Section
3.1.4. One possible "signature" could be an element such as
lithium which occurs in Imperial Valley geothermal fluids and is
quite rare in other ground waters. Lithium is especially
aPPropriate for this purpose because soils apparently absorb it
less than most other common ion exchange minerals.
The degree of natural mixing, if any, of the geothermal and
nongeothermal ground water will influence the required limits
°f detection. It will be necessary to define the temporal
Variation or trends of natural mixing at each location. These
Variations or trends can be seasonal, cyclic, increasing,
Decreasing, increasing ratios or decreasing ratios. Increasing
Ratios of geothermal fluid to nongeothermal fluid may result
from channels opening due to tectonic movements. Decreasing
Ratios may result from silicification of existing channels. It
is believed that at Roosevelt Hot Springs, Utah, the thermal
Waters naturally leak westward from the springs into the Milford
Valley cool ground water system (Parry, et al. 1976). Some
Researchers believe that production of the reservoir fluid may
lower the hydrostatic head and inhibit leakage. This would
Decrease the ratio of geothermal to cool ground water and
^hereby improve the quality of the latter.
Baseline chemical changes of samples over time must also be
considered in defining significant limits of detection. If such
changes have occurred under natural conditions, then detection
°f them after development would probably be unrelated to the
Development. The variation would have to be greater in quantity
c^ different in character than the naturally occurring difference
to be considered significant. For example, the varying quality
°£ agricultural, industrial or municipal wastewater would cause
85
-------
changes in ground water chemistry not attributable to geothermal
development.
Although the procedures for sample collection, handling and
analysis should always be rigorously followed, some situations
may not require the most sensitive analytic techniques but may
permit those which are faster or more economical. For example,
in some situations where only a rough check on a water of known
composition is necessary, a specific conductance measurement may
suffice.
3*3*2 Temporal and Spatial Sensitivity
The main hydrologic factors of concern are ground water
flow velocity and areal and depth distribution of aquifers.
Generally, the frequency of sampling should be proportional to
the fluid flow velocity. All other things being equal, the
faster the fluid is traveling the more frequently samples should
be taken. Unless tracer studies are conducted (see Section
3.4.5), flow velocity determination may be deceptive in most
geothermal reservoirs with fracture porosity. In contrast to
the more predictable flow in a homogeneous reservoir with
intergranular permeability, the flow in a reservoir with frac-
ture porosity may vary considerably from one location to
another.
Certain areas will require greater detection sensitivity
than others; these areas are near the injection well, the
aquifer immediately overlying the confining strata of the
reservoir, wells penetrating the injection zone, other deep
wells, and water use areas. This greater sensitivity will
require a denser array of sample points and more frequent
sampling.
3.4 EVALUATE MONITORING TECHNIQUES
Ground water monitoring should take place at the disposal
facility as well as in the surrounding area. All of the
monitoring techniques, except surface geophysics, involve the
use of wells. Techniques that may be used in the wells are
fluid sampling and analysis, well logging, tracers, pressure,
temperature and flow measurements, and other special or
developmental techniques. Each of these techniques and each
type of monitoring well must be evaluated for its applicability
to each geothermal development. This section discusses the
techniques, except for well logging, and types of monitoring
wells. The application of well logging to ground water monitor-
ing is discussed in Section 4.
86
-------
3.4.1 Monitoring Wells
Four types of monitoring wells may be used. The first
three listed have been defined by Warner (1975):
1) nondischarging wells constructed into the
receiving aquifer;
2) nondischarging wells constructed into or just
above the confining unit;
3) wells constructed above the confining unit;
and
4) the injection well.
Geologic data will be derived in drilling any of these wells,
but each has a distinct additional monitoring objective.
The nondischarging wells in the receiving aquifer can be
used to monitor the migration of the degraded fluid front.
In geothermal situations this migration can be monitored by
using temperature measurements in place of or in addition to
chemical analyses. The rate, location, and direction of
the injected fluid front will also provide information for the
reservoir engineers to confirm their computations on the travel
time of this front from the injection well to the production
well. The wells in the receiving aquifer can also be used to
monitor pressure or to detect chemical changes in the injected
fluid due to reactions with the reservoir rock matrix. Pressure
Monitoring may be useful in detecting leaks to nearby structural
features, abandoned wells or breaches in the confining layer.
In discussing nongeothermal injection wells, Warner (1975)
believes that monitoring wells constructed in the receiving
aquifer provide very little useful information that is important
to regulation. He states further that they are costly and may
Provide a conduit for potential degradation of overlying aquifers.
Wells constructed in or just above the confining unit may
detect leakage through the confining unit itself. These wells
should be nondischarging. If they are pumped they may induce
leakage that would not normally occur through the confining unit.
v
Wells constructed above the confining unit will detect
leakage into the unit they penetrate. These wells may be dis-
charging or nondischarging. A discharging well will represent
fluid from the volume encompassed by its cone of depression
while a nondischarging well will represent only fluid in the
well bore.
87
-------
When monitoring for the effects of geothermal injection, the
injection well itself may also be used as a monitoring well.
In monitoring programs for deep well injection, the injection
well is often the only monitoring well. Monitoring a properly
constructed injection well will provide information on the
volume of the injected fluid; its chemical and physical
properties; well head, annulus and bottom hole fluid pressures;
and condition of surface and subsurface facilities. Injection
well monitoring is discussed in more detail in Section 3.4.4.
Well Samples—
The fluid in an aquifer will vary in composition vertically
and horizontally. A sample taken from a nonflowing well will
represent only the fluid for a small portion of the aquifer.
A sample taken from a pumped well is a mixture of fluid
encompassing the cone of depression, and characterizes a much
larger volume of the aquifer fluid. Thus, if representative
samples of the aquifer fluid are desired, it is necessary to
pump the well.
Knowledge of the well construction is also necessary to
determine which part of which aquifer is tapped. This is
critical in order to specifically identify which strata may be
degraded.
Samples from some wells, especially nonflowing, shallow
ones, may be affected by infiltration from surface water through
a poor well seal. The resulting water sample may be of a better
quality than is actually in the aquifer.
3.4.2 Fluid Sampling and Chemical Analysis
Fluid sampling and analysis is the common and in most cases
the only technique used in ground water monitoring. When
properly used, this technique provides the most direct and
reliable evidence of chemical changes in the ground water. It
will be used in all ground water monitoring plans in addition
to any other techniques chosen to supplement it. An overview
of water sampling, collection and preservation is presented in
Section 3.1.1.
3.4.3 Well Logging
The potential applications of well logging to ground water
monitoring for geothermal development was of particular interest
to this study. Hence, this monitoring technique is treated in
detail in Section 4.
88
-------
3.4.4 Injection Well Monitoring Techniques
Injection well monitoring will be an extremely important
component of a geothermal ground water monitoring plan; in most
existing plans for deep well monitoring, it is the only method
used. The techniques used and scope for this type of monitor-
ing is different from that used for other monitoring wells.
Injection well monitoring includes determination of:
1) volume of injected fluid;
2) chemical and physical properties of fluid;
3) well head and annulus fluid pressures; and
4) condition of surface and subsurface facilities.
The following discussion of these components is extracted
Primarily from Warner (1975) .
The fluid flow rate, injection wellhead pressure and annulus
fluid pressure in the injection well must be monitored con-
tinuously to provide the necessary data for reservoir management,
yell maintenance and pollution control. Chemistry of the
injected fluid, annulus fluid and condition of subsurface
facilities should also be checked regularly.
The volume of injected fluid is monitored for several
reasons. First, it will enable computation of fluid flow
Distance in the injection zone. Second, it will aid in
interpretation of well behavior. Third, a record of the total
of fluid emplaced is required by regulatory agencies and
compute unit costs.
The chemical and physical properties of the injected fluid
monitored to ensure that the fluid remains within design
sPecifications. These properties may be monitored periodically
7r continuously. Parameters that have been monitored continuously
ln hazardous waste disposal wells are: suspended solids, pH,
specific conductance, temperature, density, dissolved oxygen
^nd chlorine residual. In addition, periodic complete chemical
Analyses should be conducted. The periodic biological analyses
Practiced in other types of disposal wells are not necessary in
tne geothermal environment.
Wellhead fluid injection pressure is monitored to keep a
c°ntinuous record of reservoir injectivity. Increasing pressure
^°uld indicate formation plugging. At all times pressures must
Ge kept below the level that would cause hydrofracturing of the
Reservoir or confining formation, or that would damage the well
In addition, regulations require this pressure
89
-------
The injection well is designed and constructed according to
rigorous specifications, which are outlined in Section 3.5.4.
In addition to other features, these specifications require a
sealed fluid-filled annulus. The pressure and chemistry of
the annular fluid are monitored to detect leakage in the
system. Depending on the composition of the fluid, adequate
chemical monitoring may be accomplished by placing conductivity
probes in the annulus, or by analyzing return flow for contamina-
tion in continuous cycling annulus fluid.
Corrosion rate can be determined by placing sample strips
of the tubing and casing material in the well, and checking
them periodically for weight loss.
When injecting chemically active fluid, it is important
that the well be shut down periodically for inspection and
testing. Inspection methods for casing, tubing, cement and well
bore include: (1) pulling the tubing and inspecting it visually
or instrumentally; (2) electromagnetic caliper or televiewer
logging of tubing or casing in the hole; (.3) pressure testing
of casing; (4) bond logging of casing cement; and (5) inspection
of casing cement or well bore with injectivity or temperature
profiles (Warner, 1975).
3.4.5 Other Monitoring Techniques
Other monitoring techniques that may be used in ground
water monitoring include surface geophysics, tracers and electro-
magnetic probes. These techniques would generally be used as
auxiliary methods for special situations. The principles behind
these methods and their application are briefly described below.
Surface Geophysics--
Surface geophysical methods can supply information on
subsurface structure and ground water flow patterns of hydrologi^
systems. These methods have been used successfully to delineate
geothermal reservoirs and ground water conduits, such as faults,
fractures or buried stream channels. Electrical techniques are
used primarily, but seismic, gravimetric and magnetic methods
can also supply information.
Electric methods--Electric methods have been used to detect
shallow zones of saline and/or thermal water pollution, which
can result from surface spills during well production or leakage
from geothermal holding ponds.
The direct current resistivity methods utilize a direct
or low frequency current introduced into the ground through a
pair of metal electrodes. The resulting potential difference
between the electrodes is a function of the resistivity of the
90
-------
subsurface rocks and fluids. The resistivity is, in turn, a
function of the degree of subsurface fluid saturation and
salinity, rock porosity, tortuosity of pore spaces, and
temperature. The depth of observation is determined by the
electrode spacing. Many different electrode configurations, or
arrays, are commonly used. Resistivity surveys have been used
successfully to determine the location of subsurface fluid
conduits, the location of impervious strata, and the location of
orine-fresh water interfaces. In areas where surface pollution
trom geothermal effluents is suspected, a resistivity survey
could be used to map the extent of the pollutant mass.
Spontaneous potential and streaming potential methods
involve the measurement of naturally occurring electric potentials
•Developed locally in the crust by electrochemical and/or
electrofiltration- activity. The potential generated by water
Coving through a porous medium, or streaming potential is of
special interest to hydrologic problems. Like resistivity
surveys, measurement of this potential has been used to delineate
^ock and fluids of varying resistivity to locate leakage from
reservoirs.
Electromagnetic or induction methods use a time-varying
J|agnetic field as an energy source. This generally high-frequency
ltagnetic field induces eddy currents in the presence of
Conductive materials Which in turn create their own magnetic
fields. The resultant magnetic field is measured in terms of
ne voltage induced in the receiver. Electromagnetic soundings
an be made on the ground or from a low flying aircraft. In
general, these methods lack the resolution and depth of penetra-
lon of direct-current resistivity; however, they are a more
jpid and less expensive reconnaissance method. Electromagnetic
wobmg is widely used in mineral exploration but has been little
sea m hydrologic surveys. However, they have been used
stectively to map buried stream channels where the channel
i-lling material has a significant resistivity contrast with
Lne surrounding rock (Collett, 1967).
The induced polarization method relies on measuring the
paying time of artificially induced electric potentials.
«e origin of the induced electric polarization is complex and
°t well understood, primarily because it is the result of
physiochemical conditions (Zodhy, et al. 1974). This
is useful in identifying stratigraphy in some areas
nere continuity of layers is poorly defined by resistivity.
Other surface geophysical methods—Other surface geophysical
ds have limited uses in defining subsurface hydrologic
m°?ultions useful for monitoring plans. The seismic refraction
hod is commonly used to determine the thickness of saturated
unents, depth to water and identification of buried stream
91
-------
channels. Gravity surveys supply a rapid and inexpensive means
of determining the gross configuration of aquifers, if there is
an adequate density contrast between the aquifer and the under-
lying bedrock. Magnetic surveys have been applied with limited
success to the study of magnetic aquifers, mainly basalt, and
to the configuration of basement rocks underlying water-bearing
sediments. Theory and applications of these methods are discussed
in detail by Zohdy, et al. (.1974) .
Tracers—
A tracer is a substance that is introduced at a known
location in a water flow system. The goal is to define flow
paths and parameters by detecting the tracer downgradient.
Radioactive, chemical and dye tracers have been successfully
applied to ground water investigations to determine ground water
flow paths, aquifer parameters, and the vertical and horizontal
movement of water within a borehole. The use of fluorescent
dyes has not been widespread in ground water monitoring programs.
Radioactive tracers offer the advantages of being detectable in
small quantities, of having short half-lives and, when handled
properly, of having relatively little adverse environmental
effects compared to some chemical tracers (Smith, 1976) .
The most widely used tracer in ground water studies is
tritium, a naturally occurring isotope of hydrogen, with
nuclear mass of 3. It is incorporated into the water molecule
to form tritiated water (HTO) and in this form can follow the
natural water flow. This unique feature makes it an extremely
good tracer for ground water monitoring investigations. The
disadvantages of tritium are that it requires sophisticated
apparatus for detection and concentration levels cannot be
measured in the field.
Tritium tracers have been used to monitor flow paths and
effects of injected steam condensate at The Geysers (Gulati,
et al. 1978). In this study, tritium was used primarily to
determine the effect of injection on producing wells and reservoir
characteristics.
This tracer has also been used to define geothermal
reservoir properties. Vetter (1977) describes tracing tritium
from injection wells to determine regional flow patterns within
the reservoir, delineate fractures (their number, orientation and
conductivity), detect permeable zones within the reservoir and
leakage across impermeable layers or behind well casings.
High-Frequency Electromagnetic Probes—
Theoretical and experimental studies have been conducted
using high-frequency electromagnetic probing for ground water
monitoring (Lytle, et al., 1976). When injected and resident
92
-------
fluids have dissimilar electrical properties, a three-dimensional
tluid flow profile can be determined. in this method, a
transmitter is placed in the injection well and receivers are
Placed in nearby observation wells. The time variation of the
attenuation and phase shift of the signal between the transmitter
and receivers is monitored (Fig. 3.11). For a constant
transmitting power, the receiver signals vary as the fluid
Progresses. The change in the signal is dependent on the
Percentage of the region the injected fluid occupies between
the transmitter and receivers.
Instrumentation required for this method is currently very
expensive, and interpretation techniques are still experimental.
3-5 DESIGN MONITORING PLAN AND ALTERNATIVES
Design of the monitoring plan is the culmination of all the
data collection and synthesis to this point. The goal of the
Plan is to have the capability of detecting chemical changes in
ground water characteristics within the specified chemical
temporal and spatial limits of detection. To efficiently '
achieve this capability the monitoring plan must not be constant
°r static. Some areas will require more frequent sampling
others will require a denser array of sampling points and still
on?efu Wili re(3uire different analyses. As the plan is carried
«ut the actual needs of the area will become clearer and the
P-Lan can be adjusted for more judicious and efficient monitoring.
The components to be evaluated in arriving at an adequate
monitoring plan are:
1) spatial distribution of sample points and
sampling frequency;
2) applicable monitoring techniques;
3) chemical and physical parameters;
4) regulatory specifications; and
5) cost versus confidence.
•'•hese aspects are discussed below.
3'5'1 Spatial Distribution of Sample Points and Sampling Frequency
One of the most critical determinations in designing the
.oring plan is the adequacy of the spatial distribution (area
wen pth) °f samPle Points. These points include monitoring
dS iu' surface water sampling locations, hot or cold springs and
^othermal manifestations. One method for determining an adequate
93
-------
surface
v///////////
Receiving
antenna
locations
X
Drill
hole 1
Advancing
fluid
front
Receiving
antenna
locations
Point of
fluid injection
and location
of transmitting
antenna
Drill
hole 2
\
Drill
hole 3
Figure 3.11
Location of injected fluid front by high-
frequency electromagnetic probes.
(Lytle, et al. 1976)
94
-------
spatial distribution of sample points is outlined below. Other
investigators may have different approaches; each must be
evaluated in its application at a specific site.
The choice of potential monitoring sites involves several
steps:
1) plot the areal extent of each aquifer unit on
a separate map;
2) plot the locations of potential degradation
sources or pathways for each aquifer penetrated,
including injection and production wells and
wells of known or suspected deteriorated con-
dition;
3) plot existing wells with potential for
monitoring in each of these units;
4) plot the locations of all other wells in use
or in potentially usable aquifers; and
5) if possible, estimate the radius of the cones
of depression for each discharging well.
In all cases the usable or potentially usable aquifers
will lie above the geothermal production and injection zone.
Ideally, there should be a monitoring well in each unit above
this zone, and near and somewhat downgradient from each
Potential pollutant source well or structural pathway. This
Would enable detection of possible leaks through each of these
Mechanisms.
Ideally, monitoring wells should also be located on the
vectors between the injection well and each discharging well
area. These monitoring wells should be as close to the injection
Well as practical.
The type of monitoring well distribution outlined above would
provide a considerable degree of detective protection. It would
cover each foreseeable pollutant pathway. Areas that have a
sufficient distribution of existing wells to use for monitoring
in such a pattern are rare. All existing wells suitable for
Monitoring in such locations should be incorporated into the
Monitoring plan.
Providing this extensive degree of protection in most areas
would require the effort and expense of drilling many monitoring
Wells. However, adequate protection may also be derived with a
system that is not quite as encompassing. Geologic, hydrologic
well log evidence may all but rule out leakage through some
95
-------
of the potential degradation pathways. For example, if well
logs indicate that the casing and cement in a well is completely
intact, then it would be unlikely to act as a conduit for
interaquifer flow, and drilling a monitoring well nearby would
be unnecessary.
If there are monitoring wells on each vector between the
injection and discharging well areas, one well near the
injection well may be close to several vector paths and hence
could satisfy the criteria for monitoring each flow path.
In deciding on locations for monitoring wells, the hydraulic
conductivity and gradient of each aquifer must be considered.
Areas of higher conductivity within an aquifer should be
specified with priority given to monitoring such locations,
especially those that occur between the injection well and
well discharge areas.
To monitor the advance of the degraded fluid front in the
receiving aquifer, nondischarging observation wells may be
constructed in appropriate locations. These wells will provide
information to verify or modify the predictions of degraded
fluid advance and perhaps reservoir performance.
Sampling Frequency for Monitoring Wells—
Initial sampling frequency for a monitoring plan is based on
experience with plans in similar localities. It will be
modified by successive approximations to the optimum frequency
for each well in the area. The sampling will be more frequent
and somewhat conservative until sufficient data have been
collected to justify less frequent sampling. There are several
relationships that should be considered in determining sample
frequency for a site-specific monitoring plan. All other things
being equal, sampling frequency would be greater
1) in an aquifer that has high fluid velocity
than in one with low fluid velocity;
2) closer to and downgradient from potential
pollutant sources, such as injection wells,
structural pathways; etc.
3) in the injection well.
Because of the much slower flow of ground water, its
chemistry fluctuates much more gradually than that of surface
water. Under natural conditions, deeper ground water, will
generally flow more slowly than shallower ground water and
fluctuations in their chemistry will be proportional. Super-
position of geothermal production and injection in a deep
96
-------
aquifer will change this natural relationship. An estimate of
an initial sampling frequency can be obtained by computing the
time it would take for the fluid to flow from an injection well
to a ^ nearby monitoring well. The sample frequency can be
specified as a fraction of this travel time.
Until the properties of the chemical changes and fluid flow
are known for each aquifer, the initial sample intervals would
generally be specified for one to several days, weeks or months,
depending on the monitoring locations. For example, a monitoring
well very close to and downgradient from the injection well in
the same aquifer or one immediately adjacent could be sampled
nourly or daily from the start of injection. As data are
collected to show that the system is behaving as predicted,
the intervals can be increased. They can be reduced or extended
as the monitoring program proceeds and the flow properties are
Better defined. The same interval adjustments need not apply to
ail the wells. in fact, some wells might need more frequent
sampling. Generally, the monitoring plan would not specify
sample intervals greater than one year for most locations.
Applicable Monitoring Techniques
The techniques chosen for the monitoring plan will most
J-ikely be a combination of chemical sampling and analysis, well
Bogging and injection well monitoring. other techniques will be
used only in special circumstances.
The importance of chemical sampling and analysis (discussed
in Section 3.1.1) is essential to any fluid quality monitoring
iJJ-an. in most cases it would provide the mainstay of the
Qetective monitoring plan.
Well logs would not be absolutely necessary but as
outlined in section 4, would provide information that would
yreatly add to the reliability, accuracy and confidence of the
Extremely important applications that should be included
1} continuing surveys to detect leakage from
injection and production wells; and
2) periodic surveys to detect leaks from other
wells penetrating the injection zone, or that
may provide conduits for interaquifer leakage.
detailed vertical profiles of rock and fluid properties
l?ed by wel1 logs would greatly enhance any monitoring plan
should be used at least in the baseline data acquisitiori
97
-------
injection well monitoring will be -an essential component
of any monitoring plan. This component is discussed in Section
3.4.4.
Tracers may be used in situations where it is desirable
to detect suspected pollutant pathways in certain areas. Surface
geophysical techniques may be used to detect the extent of
infiltration of a surface spill, or the location of a shallow
buried stream channel deposit.
3.5.3 Chemical and Physical Parameters
Standard chemical analyses should initially be recommended
for all sample points. These should include determination of:
silica
calcium
magnesium
sodium
potassium
carbonate and bicarbonate
sulfate
chloride
nitrate
fluoride
total dissolved solids
PH
gross radioactivity
temperature
The Geothermal Environmental Advisory Panel (GEAP, 1977)
adds dissolved oxygen and total phosphorus to this list.
Constituents that have local significance should also be added.
Some monitoring wells may have fluids entirely or partly
of geothermal or suspected geothermal origin. Analysis of
fluids from these wells may be conducted for the constituents
listed in the GEAP Guidelines as follows:
a. all parameters specified for ground water
b. gases: C02, H2S, SO2/ NH3 and Rn-222
c. aqueous species: As, Ag, B, Ba, Cd, Cr, Cu,
Fe, Hg, Mn, Mo, NH4, Pb, Se, Sr and Zn
However, the gases and aqueous species they list for analysis
do not include all the components of geothermal fluids. Other
aqueous species that have been found in significant quantities
(maximum reported concentration greater than 100 mg/1) are
98
-------
aluminum, bromide, iodine, lithium and rubidium. Most other
elements have also been found in minute quantities in some
geothermal fluids (Tsai, et al. 1978).
To determine which elements occur in a particular geothermal
fluid, some type of comprehensive spectrographic analysis may be
Performed at least in the beginning of the study (e.g. neutron
activation, spark source mass spectrometry) . This will provide a
survey of what elements are present. Certain elements that are
not detectable in these types of analyses may be accounted for
with other analytic techniques. This complete analysis should
Determine the complete suite of trace elements in the fluid.
Those that may involve: (1) significant potential environmental
effects; (2) chemical reactions in the geothermal system; or
(3) diagnostic tracer species may be specified for continued
monitoring.
Chemical analyses for the produced and injected fluid are
similar to those described for geothermal monitoring wells.
Additional physical measurements of fluid volume, pressure,
temperature and well conditions would be measured for the injec-
tion well, as described in Section 3.4.4 on injection well
Monitoring techniques.
Complete or standard chemical analyses may not be required
3t each sample interval. Where the specific conductance measured
in the field does not vary by more than 10% from the initial
Measurements, further analysis may be omitted.
Physical parameters to monitor are:
water level
temperature
specific conductance
well discharge volume and rate
well condition
The water level measurements will provide important informa-
tion on possible changes in hydraulic head, hence changes in
flow rate and direction, that may be caused by geothermal
Production and injection. Significant changes in well discharge
7^ta may have a similar effect. The temperature may provide
^formation on increasing or decreasing contributions from
Geothermal sources. The specific conductance is directly
Proportional to the salinity of the fluid and changes in it
changes in the fluid chemistry.
3«5.4 Regulatory Specifications
Federal and state regulations will require certain features
111 injection well planning, construction and monitoring. The
99
-------
Safe Drinking Water Act (SDWA) of 1974 gives the EPA authority
to protect ground water by regulating waste disposal wells and
to protect aquifers designated by the EPA as "sole sources" of
drinking water. It is the main Federal law regulating subsur-
face disposal. It contains guidelines for permission and
operation of waste injection wells. The states adopt programs,
subject to EPA approval, which will require permits for the
underground injection of wastes. Currently, the EPA has a
position statement (Administrator's Decision Statement No. 5,
Appendix A) including planning and monitoring requirements.
The USGS has issued Geothermal Resources Operational Orders
(GRO's) and guidelines for acquiring environmental baseline
data on Federal geothermal leases (GEAP, 1977). As of 1974,
several states that allow injection also had explicit or
implicit monitoring requirements (Reeder, et al. 1977, pp. 1
and 157). Most state regulation of disposal wells would be
administered by the State Engineer. Specific monitoring regula-
tions that may be required by a state must be included in the
overall monitoring plan. Many of these regulations also include
specifications for preventive measures in injection well design
and construction, and surface facilities for containment of
accidental spills.
The EPA statement has detailed data acquisition require-
ments for evaluation of the injection plan. This includes
evaluation of the geology, hydrology, chemistry, injection
interval, volume, rate and chemistry, bounding beds, well engi-
neering data. The portion relating to monitoring specifies,
"plans for monitoring including a multipoint fluid pressure
monitoring system constructed to monitor pressures above as well
as within the injection zones; description of annular fluid;
and plans for maintaining a complete operational history of the
well".
Safe Drinking Water Act—
Proposed regulations under the SDWA would govern State
Underground Injection Control Programs (SUICP's). Such programs
are intended to protect underground drinking water sources from
contamination by injected effluents. Subpart C of the SDWA,
"Protection of Underground Sources of Drinking Water," deals
with injection wells and requirements include the following:
1) all existing injection operations are reviewed,
and those which appear to endanger drinking
water sources are suspended for investigation;
2) proposed injection wells which intersect
aquifers with TDS less than 3,000 mg/1 must
use a double-pipe design (injection tubing
inside a casing) and a packer preventing
effluent from entering the annulus;
100
-------
3) all existing and proposed injection sites must
obtain an underground injection control permit,
the requirements of which are outlined in this
program;
4) public notice of permit applications must be
given, with provisions for a public hearing;
and
5) a specified monitoring program and reporting
procedure must be followed
The applicability of these requirements to geothermal wells
has not yet been established.
The EPA August 31, 1976, draft of the SUICP is presently
being revised due to intense public pressure. Options to the
initially proposed SUICP that have been proposed include:
1) Section 1421(b)(2) of the SDWA prohibits state
programs from interfering with or impeding injec-
tion connected with secondary or tertiary recovery
of oil or natural gas, unless underground drinking
water will be endangered by such injection. It
has been proposed that the geothermal industry
be treated similarly to the oil and gas industries.
2) Proposed regulations would be applied to
industrial waste injection, but less stringent
controls would be placed on "resource recovery
wells" (the name under which the proponents
categorize geothermal and petroleum industry
brine injection wells).
3) Non-retroactivity of proposed regulations.
4) Existing wells that are used only for brine
disposal be granted special permits or eliminated
within 10 years.
USGS Regulations—
GRO's are issued under the Geothermal Steam Act of 1970.
Pollution, waste disposal and water quality are discussed in
GRO Order No. 4 which covers general environmental protection
tequirements.
GRO Order No. 4 states that harmful liquid effluent must
disposed of in a manner conforming to Federal, state and
ional standards or be injected into the geothermal reservoir
Ot other formation approved by the Area Geothermal Supervisor.
101
-------
A plan of injection, injection reports, inspection criteria,
well drilling and conversion specifications is required.
The primary responsibility for water quality is delegated to
states with EPA approved standards.
Under GRO Order No. 5, a development operation plan is
required to include:
1) a proposed well spacing map;
2) geologic and geophysical map;
3} representative drilling program;
4) proposed utilization of geothermal resource;
5) surface equipment installations; and the
6) proposed liquid disposal program.
The plan of injection is required to include:
1) a map of wells and facilities;
2) injection fluid characteristics;
3) proposed disposal zone characteristics;
4) subsurface maps and cross sections;
5) logs and histories of disposal wells;
6) representative injection well drilling program;
7) proposed downhole and surface equipment;
8) proposed injectivity surveys and monitoring; and
9) the hydrology of the area.
The GEAP guidelines recommendations apply to baseline data
collection. Those relating to ground water include specifica-
tions for:
1) chemical analysis standards;
2) sources to be sampled, including surface and
ground water and geothermal fluid;
3) frequency and duration of sampling for surface,
ground and geothermal waters;
102
-------
4) physical parameters to be measured including
discharge, temperature, pH, specific conductance
and turbidity; and
5) chemical parameters to be measured.
3.5.5 Cost Versus Confidence
Assuming competent planning and execution of a monitoring
Plan there is a direct, relationship between the cost of the plan
and the degree of confidence obtained. This relationship is
generally perceived as exponential, and beyond a certain point
returns diminish per unit increase in cost. Although this
Point is rather subjective, it is hoped that a viable monitoring
Plan can be achieved before this point of diminishing returns is
Beached. To do that the degree of confidence desired must be
reconciled with the amount of money to be spent. In most cases
the study outlined in this methodology will allow this
reconciliation. Monitoring plan costs consist of costs for:
1. baseline study;
2. monitoring planning;
3. construction of new wells or modification of
existing ones;
4. sampling frequency;
5. chemical analyses of samples;
6. monitoring techniques used; and
7. data synthesis and interpretation.
The first two items are one-time costs, the third may be
one-time or periodic, the last four are continuing. If the
first two components are considered to be relatively fixed by
Regulatory constraints, precedents or environmental concerns,
components 3 through 7 are more discretionary. These aspects
discussed below.
Spatial Distribution —
Proper understanding of the geologic, hydrologic, well
a^d reservoir conditions will allow confident determination of
Probable flow paths. A few strategically located monitoring
^elis can function more effectively than many arbitrarily placed
Wells. If possible, existing wells should be used as monitoring
^slls. The well placement should be designed in a progressive
way. That is, a few wells could be constructed first in
103
-------
critical areas, e.g. close to and directly downgradient from
injection wells in each aquifer. These would be placed to
ensure detection of any escaping fluid from these sources. In
this type of plan, it is imperative that the first line critical
coverage be comprehensive. Additional wells would have to be
constructed if and when degraded fluid is detected in these first
line warning wells. Appropriate preventive measures would be
taken at the source of the leak and a second line of wells should
be constructed only in the affected aquifer. This type of
approach may be the most efficient since wells are only placed
in areas that have been shown to be suspect. It would provide
a good degree of confidence without the initial cost of an
extensive and widespread monitoring network.
Initially the only wells that may be conduits for interaquif&
pollution due to injection would be those that penetrate the
injection zone. In most areas only geothermal wells would be
in this category. Any wells which penetrate the injection zone
and are suspected of allowing vertical flow should be checked
with the well logs described in Section 4. If significant
quantities of fluid are escaping or could escape under produc-
tion and injection conditions, the well should be repaired or
plugged.
Any new wells should be designed to provide maximum safety
for potential pollution. Casing material should be chosen with
consideration of conducting well log surveys in the hole (i.e.,
perhaps case with fiberglass).
Sample Frequency—
Sampling frequency must be determined for each case. The
data must be evaluated as it is collected. Appropriate modifica-
tions outlined in Section 3.5.1 may then be made.
Chemical Analyses—
One way to reduce chemical analysis costs for the monitoring
wells is to analyze fewer parameters. Rigorous determination of
the parameters that are necessary for monitoring at a specific
site might reveal some constituents that may be safely deleted
from the original list. Another approach would use the field-
specific conductance determination after the initial complete .
analyses have been conducted. If this measurement is within
specified limits of the initial measurement, then another
complete chemical analysis would not be necessary. This may
maintain an acceptable degree of confidence in the water quality
monitoring and significantly reduce the operating costs.
104
-------
Other Monitoring Techniques—
Well logging and injection well measurements will probably
be the only other monitoring techniques used. Since well logging
is quite costly, it must be used judiciously. These costs are
outlined in Section 4.3.5.
Injection well measurements are necessary and fairly well
defined, so there are no significant cost savings considera-
tions.
Costs of other monitoring techniques are site and
sensitivity dependent. It would be necessary to evaluate each
°f these independently.
Data Synthesis and Interpretation—
One way to reduce the cost of data synthesis and interpreta-
tion is by automation. A computerized information storage,
Retrieval, mapping, interpretation and data management system
could be set up. Since computer systems are more economical
with repeated applications it would be advantageous to have one
system that all geothermal developers could use. The system
could be devised to allow application to all ground and perhaps
eyen surface water monitoring systems. Such a general applica-
tion would distribute the development cost over a larger body
of potential users and thereby reduce the unit cost. Develop-
^ent of such a data system is discussed further in Section 5.
3«6 IMPLEMENT MONITORING PLAN
Implementation of the monitoring plan will consist of three
Parts:
1} data collection and analysis;
2) data synthesis, display and interpretation;
3) review and modification of monitoring plan.
Data Collection
The data collection will be conducted at the specified
uency and locations. In most cases a well logging company
be contracted to run the well logs. The fluid sampling
analyses should be conducted according to the procedures
specified in the baseline data acquisition and monitoring plan
ign phases of the monitoring methodology. Some phases of
action well monitoring should be continuous and others should
periodic.
105
-------
3.6.2 Data Synthesis, Display and Interpretation
Data will be acquired regularly, according to the monitor-
ing plan, for fluid chemistry, well logging and injection well
monitoring. Previous data should be reviewed and correlated with
the new data. Synthesis of fluid chemistry data is discussed
below. Well logs are discussed in Section 4. The results of
injection well monitoring may require the attention of a
reservoir engineer.
Several graphical techniques for water chemistry data
synthesis are described by Hem (1910). The Stiff diagrams which
are described by Hem (1970) may have to be modified for applica-
tion to geothermal situations. In geothermal areas where the
TDS of the fluid varies by more than one order of magnitude,
percent reactance may be used instead of milliequivalents per
liter for the horizontal scale (Geonomics, 1978a). Although
this will eliminate the graphic depiction of concentration, it
will maintain the consistent symbol shape representing similar
proportions of constituents.
Langelier-Ludwig Diagrams—
The Langelier-Ludwig (L-L) diagram (Langelier and Ludwig,
1942) (Figs. 3.12, 3.4 and 3.5) is similar to the rhombohedral
section of the Piper diagram (Piper, 1944). It is a square plot
of percent reactance of alkalic cations (Na+K) ascending from 0
to 100 on the left hand vertical axis and hardness cations
(Ca+Mg) descending from 100 to 0 on the right hand vertical
axis. The horizontal axes plot percent reactance of carbonate
anions (HCO3+C03) and noncarbonate anions (SO4+C1), with each
axis reciprocating the scale of the opposite axis. This diagram
provides a method for "segregating analytic data for critical
study with respect to sources of the dissolved constituents in
waters, modifications in the character of water as it passes
through an area and related geochemical problems" (Piper, 1944).
It allows for investigation of compositional relations among
samples and statistical populations of samples in the form of
clusters of points.
Salinity sections (Figs. 3.12 and 3.5) can be drawn at any
orientation on the Langelier-Ludwig diagram to depict changes in
concentration. These sections are constructed by projecting all
the data points to be included in the section onto a straight
line extending from one L-L diagram axis to an opposite axis.
A triangle is formed by extending two lines from above at an
angle of about 90° to intersect the ends of the L-L diagram
section line. This section can be visualized as one side of a
four-sided pyramid with the L-L diagram as the base. The apex
of the pyramid would represent zero salinity. Lines of constant
chemical constituent ratio and decreasing salinity would connect
106
-------
100
PERCENT REACTANCE
ci
MID -LINES
OF SECTIONS
C HCO* + CO'
\ SURFACES OF
\ SALINITY
\ SECTIONS
\
SURFACE OF
LANGELIER-
LUDWIG
DIAGRAM
PERCENT REACTANCE
figure 3.12 Three-dimensional perspective of a Langelier-
Ludwig diagram showing surfaces of salinity
sections. (Geonomics, 1978a)
107
-------
points on the L-L diagram, at the pyramid base, to the pyramid's
apex. Then an appropriate milliequivalents-per-liter scale is
chosen as planes parallel to the base of the pyramid. The data
point is plotted on the salinity section where the salinity .
value plane intersects the chemical constituent ratio line.
Similar to the L-L diagram, the salinity sections also allow
for investigation of compositional relations and statistical
populations of samples.
Single Chemical Parameter Plots—
Three-dimensional surface plots of single chemical parameter
contours (Fig. 3.3) can be used as an additional technique for
correlation of ground water chemistry with aquifers, geologic
structure or geothermal anomalies. The ground water quality data
must be digitized, and computer contouring and surface drawing
routines are necessary. With minimal effort and cost this
technique could eventually allow generation of many vertical and
horizontal cross sections and surfaces, as well as trial
interpretations using different depth intervals. The essence of
the technique is to define subsurface structure and lithology
as depicted by changes in chemical parameters in the ground water
system. These features could be shown in the appropriate
combination of contoured cross sections and three-dimensional
surface plots.
A computerized information storage, retrieval, mapping,
interpretation system to handle and expedite assimilation of the
monitoring data is described in Section 5.
3.6.3 Review and Modify Monitoring Plan
Implementation of the monitoring plan will provide data on
the hydrodynamics and geochemistry of the ground water systems
that were not available during the initial design of the
monitoring plan. The additional chemical and physical parameters
provided by the monitoring plan will provide additional input
to any mathematical/computer model of the ground water system.
Quantification of these parameters may allow better predictions
for the behavior of the degraded fluid fronts. These predic-
tions may also provide information for modifications to the
monitoring plan. These additional data may reveal situations
such as:
1) faster or slower velocity than expected for a
degraded fluid front;
2) detection of degraded fluid where it was not
anticipated;
108
-------
3) indications of unanticipated advance in a
particular direction or strata;
4) detection of unanticipated chemical chang.es
or constituents; and
5) detection of structural, stratigraphic or
well-bore pollutant pathways, etc.
Evaluation of these data will lead to necessary and/or
beneficial modifications to the plan. This evaluation will
utilize considerations similar to those incorporated in the
initial design of the plan. Modification may include:
1) adding or deleting sample points;
2) increasing or decreasing sampling frequency
at particular sample locations;
3) adding or deleting certain chemical analyses
at particular locations;
4) specifying different analytic methods;
5) changing sample collection procedures;
6) changing data synthesis and interpretation
procedures; and
7) changing review and modification procedures.
109
-------
REFERENCES
American Public Health Association. 1977. Standard Methods for
the Examination of Water and Wastewater. 14th edition.
American Public Health Association, Washington, D.C.
Appel, C. A., and J. D. Bredehoeft. 1976. Status of Ground
Water Modeling in the USGS. USGS Circular 737. 9 pp.
Brown, E., M. W. Skougstad and M. J. Fishman. 1970. Methods
for Collection and Analysis of Water Samples for Dissolved
Minerals and Gases. U.S. Geological Survey Techniques of
Water Resources Investigations. Book 5, Chapter A-l.
160 pp.
Collett, L. S. 1967. Resistivity Mapping by Electromagnetic
Methods. In: Mining and Groundwater Geophysics/1967.
Geol. Survey of Canada. Economic Geology Report 26.
pp. 615-624.
Coplen, T. R. 1972. Origin of Geothermal Waters in Imperial
Valley of Southern California. in: Cooperative Investiga-
tions of Geothermal Resources in the Imperial Valley Area
and Their Potential Value for Desalting of Water and
Other Purposes. R. W. Rex, Principal Investigator.
University of California, Riverside. pp. E1-E33.
Ellis, A. J. 1976. IAGC Interlaboratory Water Analysis
Comparison Programme. Geochemia et Cosmochemica Acta.
Vol. 40. pp. 1359-1374.
Ellis, A. J., W. A. J. Mahon and J. A. Ritchie. 1968. Methods
of Collection and Analysis of Geothermal Fluids. 2nd
edition. Chemistry Division, New Zealand Department of
Science and Industry Research Report CD 2013. 51 pp.
Finlayson, J. B. 1970. The Collection and Analysis of Volcanic
and Hydrothermal Gases. Geothermics. Special Issue 2,
Vol. 2, Pt. 2. pp. 1344-1354.
Fournier, R. O., and A. H. Truesdell. 1974. Geochemical
Indicators of Subsurface Temperature - Part 2. Estimation
of Temperature and Fraction of Hot Water Mixed With Cold
Water. USGS Journal of Research. Vol. 2, No. 3. pp.
263-269.
110
-------
Fournier, R. 0., D. E. White and A. H. Truesdell. 1974.
Geochemical Indicators of Subsurface Temperature - Part 1.
Basic Assumptions. USGS Journal of Research. Vol. 2,
No. 3. pp. 259-262.
Futures Group. 1975. A Technology Assessment of Geothermal
Energy Resource Development. Prepared for National Science
Foundation. Contract C-836, Research Applications
Directorate. 554 pp.
Geothermal Environmental Advisory Panel. January 1977. Guide-
lines for Acquiring Environmental Baseline Data on Federal
Geothermal Leases. USGS, Menlo Park. 26 pp.
Geonomics. 1978a. Baseline Geotechnical Data for Four
Geothermal Areas in the United States. Report for EPA/EMSL,
Las Vegas, Nevada.
Geonomics. 1978b. Subsurface Environmental Assessment for Four
Geothermal Systems. Report for EPA/EMSL, Las Vegas, Nevada.
Giggenbach, W. F. 1976. A Simple Method for the Collection and
Analysis of Volcanic Gas Samples. Bulletin Volcanologique.
Vol. 39, No. 1. pp. 132-145.
Gulati, M. S., S. C. Lipman and C. J. Strobel. 1978. Tritium
Tracer Survey at The Geysers. In: Geothermal Energy: A
Novelty Becomes Resource, Transactions Geothermal Resource
Council. Vol. 2. pp. 237-240.
Hem, j. D. 1970. Study and Interpretation of the Chemical
Characteristics of Natural Water. USGS Water Supply Paper
1473. 363 pp.
Hill, J. H., and C. J. Morris. December 1975. Sampling a Two-
Phase Geothermal Brine Flow for Chemical Analysis.
Lawrence Livermore Laboratory, UCRL-77544. 27 pp.
^tercomp, Inc. 1976. A Model for Calculating Effects of Liquid
Waste Disposal in Deep Saline Aquifer. Prepared for USGS.
NTIS No. PB-256 903. 263 pp.
Langelier, W. F., and H. F. Ludwig. 1942. Graphical Methods
for indicating the Mineral Character of Natural Waters.
Journal of American Water Works Association. Vol. 34.
pp. 335-352.
, J., D. L. Lager, E. F. Laine and J. D. Salisbury. 1976.
Monitoring Fluid Flow by Using High Frequency Electro-
magnetic Probing. Lawrence Livermore Laboratory Report
No. UCRL-51979. 50 pp.
Ill
-------
Meidav, T., R. West, Am Katzenstein and Y. Ralstein. May 1976.
An Electrical Resistivity Survey of the Salton Sea
Geotheranal Field, Imperial Valley, California. Report for
Lawrence Livermore Laboratory. 131 pp.
Narisimhan, T. -N., and P. A. Witherspoon. May 1977. Recent
Developments in Modeling Ground Water Systems. Lawrence
Berkeley Laboratory. LBL-5209. 35 pp.
Olmsted, F. H., O. J. Loelty and B. Irelan. 1973. Geohydrology
of the Yuma Area, Arizona and California. USGS Professional
Paper No. 486-14. 227 pp.
Palmer, T. D. 1975. Characteristics of Geothermal Wells Located
in the Salton Sea Geothermal Field,.Imperial County,
California. Lawrence Livermore Laboratory, UCRL-51976.
54 pp.
Parry, W. T., N. L. Benson and C. D. Miller. 1976. Geochemistry
and Hydrothermal Alteration of Selected Utah Hot Springs.
National Science Foundation. Contract No. GI 4374,
University of Utah. Vol. 3. 131 pp.
Piper, A. M. 1944. A Graphic Procedure in the Geochemical
Interpretation of Water Analyses. Transactions, American
Geophysical Union. Hydrology Papers. pp. 914-923.
Presser, T.S., and I. Barnes. 1974. Special Techniques for
Determining Chemical Properties of Geothermal Water.
U.S. Geological Survey Water Resources Investigations 22-74.
11 pp.
Reed, M. J. 1975. The Collection of Geothermal Fluid Samples
for Chemical Analysis. In: Geothermal Professional Papers.
July 1975. California Division of Oil and Gas Report No.
TR14. pp. 3-4.
Reeder, L. R., J. H. Cobbs, J. W. Field, Jr., W. 0. Finley,
S. C. Vokurka, B. N. Rolfe. June 1977. Review and
Assessment of Deep Well injection of Hazardous Waste,
EPA 600/2-77-029a, 168 pp. (first of four volumes).
Smith, D. B. 1976. Nuclear Methods. In: Facets of Hydrology.
J. C. Rodda, ed. John Wiley & Sons, New York. pp. 61-84.
Talbot, J. S. 1972. Requirements for Monitoring of Industrial
Deep Well Disposal Systems. In: Underground Waste Manage-
ment and Environmental Implications. T. D. Cook, ed.
American Association of Petroleum Geologists. Mem. 18.
pp. 85-92.
112
-------
TEMPO. July 1973. Polluted Ground Water: Some Causes, Effects,
Controls and Monitoring. U.S. EPA Report No. 600/4-73-OOlb,
edited by Charles F. Meyer. 282 pp.
Todd, D. K., R. M. Tinlin, K. D. Schmidt, and L. G. Everett.
1976. Monitoring Ground Water Quality: Monitoring
Methodology. General Electric Company-TEMPO. U.S. EPA
No. 600/4-76-026. 172 pp.
Truesdell, A. H., and K. L. Pering. 1974. Geothermal Gas
Sampling Methods. USGS Open File Report 74-36-1. 6 pp.
Tsai, F., S. Juprasert and S. K. Sanyal. 1978. A Review of
Chemical Composition of Geothermal Effluents. In:
Proceedings, Second Workshop on Sampling and Analysis of
Geothermal Effluents. February 15-17, 1977. Las Vegas,
Nevada, sponsored by EPA, Las Vegas, Cincinnati and
Washington. pp. 84-96.
U.S. EPA. 1974. Methods for Chemical Analysis of Water and
Wastes. Environmental Monitoring and Support Laboratory,
Cincinnati, Ohio. EPA-625/6-74-003a. 298 pp.
U.S. EPA. 1976. Proceedings of the First Workshop on Sampling
Geothermal Effluents, Environmental Monitoring and Support
Laboratory, Las Vegas, Nevada. EPA-600/9-76-011. 234 pp.
U«S. EPA. July 1977. Pollution Control Guidance Document for
Geothermal Development, Power Technology and Conservation
Branch, Energy Systems Environmental Control Division,
Office of Energy, Minerals and Industry (unpublished draft).
112 pp.
U.S. EPA, 1978. Proceedings of the Second Workshop on Sampling
and Analysis of Geothermal Effluents. Prepared by Geonomics,
Inc. for EPA, Las Vegas, Nevada. 256 pp.
Better, 0. J. 1977. Tritium Tracer as a Means for Reservoir
Verification in Geothermal Reservoirs. Unpublished Report.
7 pp.
, D. L. July 1975. Monitoring Disposal Well Systems.
U.S. EPA Report No. EPA-680/4-74-008. 109 pp.
, J. C. 1978. Sampling and Analysis Methods for Geo-
thermal Fluids and Gases. Batelle Pacific Northwest
Laboratories, Richland, Washington. PWL-MA-572, UC-66d.
522 pp.
113
-------
Wood, W. W. 1976. Guidelines for Collection and Field Analysis
of Groundwater Samples for Selected Unstable Constituents.
In: Techniques of Water Resources Investigations of the USGS«
Book 1, Chapter D2. 24 pp.
Zohdy, A. A. R. , G. P. Eaton and D. R. Mabey. 1974. Applica-
tions of Surface Geophysics to Ground Water Investigations.
Techniques of Water Resources Investigations of the USGS.
Book 2, Chapter Dl. 116 pp.
114
-------
SECTION 4
APPLICATION OF WELL LOGGING
TO GROUND WATER MONITORING IN GEOTHERMAL AREAS
The application of well logging to ground water monitoring
geothermal areas is described in this section. Subsurface
a collection, well logging tools, log interpretation
Philosophies in geothermal environments, and logging costs are
Discussed. Well logs can aid in defining the baseline condi-
tions of a subsurface hydrologic regime by supplying data on
Aquifers, regional ground water flow paths, water quality and
the relationship of wastewater injection zones to nearby fresh
ground water systems. Well logs can also detect leakage from
injection and production wells and aid in monitoring water
Duality throughout the monitoring network.
4-l SUBSURFACE DATA COLLECTION
To design a ground water monitoring program, the geohydro-
c conditions in an area must be defined. The subsurface
necessary to define these conditions can be collected
rectly from drill holes by sampling fluids and rock, by flow
tests, and by well logging. Each of these techniques will eval-
uate different aspects of rock and fluid properties. For
e*ample, rock cores provide the best and most direct lithologic
<*nd mineralogic determinations, as well as some clues to in situ
Permeability. On the other hand, flow tests provide the best
aKta f°r in situ PermeabilitY and only very indirect evidence
about rock types. Data on water quality can be gained from
complete chemical analyses on fluid samples from a drill hole.
Well logs, with proper interpretation, can provide much
lnformation on rock and fluid properties such as lithology,
Porosity, permeability, density, relative water quality, fluid
m°vement and borehole temperature. However, logging cannot
entirely replace sampling. For correct interpretation, some
Camples, properly taken and analyzed, are essential to the
Interpretation of logs in each new geologic environment. Data
Collected by well logging must always be correlated to the
Barest wells where core or sampling data are available, and
Qne well, adequately sampled and logged, can serve as a guide
°r other wells in the same terrane.
115
-------
Under ideal conditions, correctly interpreted well logs can
provide continuous and detailed vertical profiles of rock and
water properties at an acceptable cost. The subsurface volume
explored by logs can be up to 100 times that investigated by
rock or fluid sampling in the well bore (Keys and MacCary,
1971). Logs also provide data collection capabilities in cased
holes. Applications of well logs in water quality monitoring,
drilling guidance, subsurface extrapolation of data and data
representation, storage, and manipulation are described below.
Sampling and analysis of well effluents is an important
means of detecting ground water pollution. However, as
described by Sanyal and Weiss (in press), this technique has
some drawbacks. Water sampled from an uncased well represents
a mixture of water'from various permeable horizons rather than
from an individual horizon. This situation is undesirable if
water quality varies significantly with depth within the bore-
hole. A sample from a cased well represents only the water at
the perforated interval. One way of obtaining a continuous
vertical profile of water quality with depth is to combine
borehole logging with well effluent sampling and analysis in
both open and cased holes.
Well logs can reduce drilling expenses by guiding the loca-
tion and construction of drill holes. They also provide means
of vertical and horizontal extrapolation of data derived from
these holes. Because the data are immediately accessible, their
quality can be assessed during the logging operation and addi-
tional data can be obtained at important depth intervals.
Periodic logging can permit observation of temporal changes in
the ground water quality and flow paths, rock characteristics
and the well condition caused by injection or some other
mechanism.
The graphic form of a well log permits quick visual
interpretation, data comparison and decision making at the site.
Data can be digitized in the field or office which permits
transportation by radio or telephone and storage on magnetic
tape. In this form, data are readily available and can be
correlated with data from other wells. This permanent recording
also allows for the subsequent investigation cf some geologic
or hydrologic factor not considered while the hole was being
drilled.
4.1.1 Well Logging Tools
Well logging tools can be divided into four general
categories: electric, nuclear, acoustic and specialized tools.
Those which have been used widely in the petroleum industry and
are readily obta-inable are listed in Table 4.1. The table also
gives descriptions of the parameters measured by the log and
116
-------
TABLE 4.1 COMMERCIAL WELL LOGGING TOOLS
TYPE OF
EQUIPMENT
ELECTRICAL TOOLS
Spontaneous potent Jol log
PROPERTIES MEASURED
Records natural elective potential d
between borehole fluid and rocfc moterialsr
APPLICATFOI^IS
Geologic correlation, bed thickness determination,
permeability, formation wnrer quality, porp
geometry, thickness of procuring aqyifer^
COMMENTS
All electrjcnf toe-Is
ttwstollic cwed hof*
Con oply he njn in liolft* fill*** with « conducting
fluid, pooi result* \n highly resistive formations,
not offered by wosltaut holes or fJffpp or variable
Resistivity tegs
-Single point devices
-Normal devices
-lateral devices
Records electrical resistance of a known or
volume of earth materials under (he e surrounding rode
Short normals—nwawwes apporent resistivity of
invaded zone.
Long normals—measures apparent resistivity
beyond the invaded
on r«*lsfivity beyo-nd invnded
Pom*ity, petftieobility, fonmjlion wafer quality,
pore geometry, Utholorjy, Ihpimol conductivity,
heat capocfty, thickness of producing aquifer-*
LUhotogy, geologic correlation, fracture location
normol device) vertical lilnologic detail
(Long normal device) Form alt on wof«r quality in
clastic TOC!
-------
Table 4.1 (continued)
00
TYPF OF
EQUIPMENT
NUCLEAR TOOLS
Noturol gamma roy log
Gamma speclrometry log
Gornrno-gommo density tog
PROPERTIES MEASURED
A FPL 1C AT IONS
COMMENTS
Neutron log
Pulsed neutron log
Neutron activotionl
ACOUSTIC TOOLS
Borehole televiewer
Sonic logs
-Acoustic velocity
Record? tlie armunl of natural gamma radiation
emitted by formation (U , Th ' K plus
daughter products)
Records energy distribution of natural gommo
rnHiotion7
Records the Intensity of gamma radiation fTOm a
source in the probe after it hm backscottered
ond allemialed within tl>e borehole and surround-
ing rocks'
Counts number of neutrons present of different
energy levels or the number ol^pcmmn photons
by neutron reoctions
Meosures thermal neutron decay lime
Measures nnd identifies oclivation- gammn rodio-
tion from decoy of neutron-activated elements
Takes an oriented acoustic "phologroph" of ^"^~
hole wall in the form of a continuous well log
1
Records P-wave attenuation
Records congressional pod shear wave
velocities
Records compressional and shear wove
amplitudes
Lilhology, straligraphic correlator., porosity,
permeability"*
Lilhology, slrotigrapliic correlation, identifica-
tion of artificial rodioisotopes in water/ useful ^
for Identifying fracture zones in crystalline rocks
Lilhology, butk density, total jjoiosily, locations
of cavities behind well casino/ thickness of
producing aquifer, thermal conductivity, heal
capacity, sonic velocities,5 location ol casing
collars and position of cement behind cosing,
water level, significant changes in fluid density
Moisture content (above water table), total
porosity (below water table),2 lilhology; hydrogen,
chlorine or silicon density; permeability, forma-
tion water quality, pore geometry, thickness of
producing aquifer^
Chlorine dctv.ity, porosity, water saturation,'
detection of channeling behind casing, formation
wciler quality (determination of boron, lithium,
silica, sails)
Elemental density (carbon, oxygen, silica,
aluminum, iron, calcium, magnesium)," lilhology,
formation voter quality In known lithologies
Fracture" and vug location (Iheir number, sire and
orientation), formation dip, lithology; diagnosis of
drilling, testing onH production problems
Homogeneity of cement,3 cement bond to cosing
and borehole wall, detection of microannulus or
channeling behind casing and location of cement
top*
Porosity/ (of inlergranular formation!), fraclu-e
orientation, mechanical properties of rock,
pressure, formation wafer quality, pore geometry,
thermal conductivity, heat Cryocily, thickness of
producing aquifer, lilhology
Cement bond quality, location of fractures,2
pore geometry, lilhology
All radioactive tools may be used in open or cased
holes
Best use in moderate-sized holes with no washout
Especially useful for Illltolngic identification of
crystalline rocks'
Used with sonic logs, eon determine fracture
porosity,1* especially useful for lilliologic
identification of crystalline rocks'
Con be used in fluid-filled or empty holes; used
with sonic logs, con determine Iroelure porosity,
large radius of Investigation
formation water salinity should he high nrel
porosity greater llroii 15%
Most effective technique of delecting minute
quantities of vorious elements, however, log cannot
distinguish if element contained In rock or fluid
All acoustic tools must he used in liquid-filled holes
Provides a substitute for coring in highly fractured
foimalions, resolution of fractures lo 0.8 mm wide
Data can be questionable in casings 40 cm in
diameter or larger
Cannot be used In empty holes; have been
employed In cased hales with up lo 50% cement
bond, "3 borehole should be lo gauge,3 useful in
low resistivity formations used with gamma-gommo
density or neutron logs can estimate fracture
porosity in known lilhologies
-------
Table 4.1 (continued)
TYPE OF
EQUIPMENT
Acoustic eoliper
SPECIALIZED TOOLS
PROPERTIES MEASURED
APPLICATIONS
Temperature log
Flowmeter log
Pipmeter log
Gravity meter log
Drilling time log
Mud log
Fluid pressure leg
Records wove reflection
Produces a profile of borehole size variation In open
holes, casing st« variation In cased holes
Records temperature of well bore as a function of
depth
Records fluid velocity versus depth
Resistivity of a thin section near the borehole to
produce 3 to 4 correlation curves on different
sides of the hole4
Records gravity differential between any two
sections in the borehole'
Records of times drilling starts ond stops, depths
and times connections ore made, drilling rate
and ambient ond borehole temperature!
Identification of mud and drilling cuttings
Measures fluid pressure as a function of depth
Accurate hole diameter In several direction;,
orientation of fractures, location of vugs and
bed boundaries in open holes and cosing inspec-
tion in cased holes
Location of fracture coving and cavity location*,
lilhology ond estimates of required cement
volumes in open holes, interior rasing deteriora-
tion in cased holes, hole diameter correction
factors for other log interpretations
Temperature, permeability, thermal conductivity,
heat capacity, geothermal gradient, location and
thickness of zones of fluid entry or discharge in
borehole, location of cement top, detection of
fluid flow behind pipe and casing, correction
factors for other log interpretations
Determines contribution of each zone to total
production or injection. Indicates changes in
flow patterns4
Formation dip relative to borehole,4 orientation
of fractures'^
Density, porosity, overage built density
lilhofogy, strotigraphlc correlation, formation
depth, correction factors (brother logs
Lilhology, stratigraphlc correlation, formation
depths, drilling control
Well control, borehole fluid movement
COMMENTS
Useful In large boreholes'
5 lo 20 cm Me Hiofiwter Mmllnlion with 1 5 mm
occtficicy
Accuracy to 0.3°C/I
Inflatable packer-type flowmeter useful for low
flow rotes, continuous-type flowmettr useful
for high flow rotes4
Electrically conductive mud must be used in
borehole ond the borehole must be relatively
free of covings'
Very large radius of Investigation'
-------
Table 4.1 (continued)
TYPE or
EQUIPMENT
Inactivity profile
Directional' survey
Rodiooctive or chemicol
trocer techniques
Borchole audio trocer
survey
Grodiomanometer*
Formation Interval Tester*
Formation sampler
Fluid sampler
PROPERTIES MEASURED
APPLICATIONS
COMMENTS
Under injection conditions, plots water loss a! a
function of depth'
Records angle borehole makes with vertical as o
Function of depth and drift azimuth'
Seorchs for and records the level of an inlroduced
tracer
Measures the level of sound generated by fluid
flow behind the cosing'
Measures the overage pressure gradient over
60 cm intervols
Perforates cosing and recovers simple of formation
fluid sealed under pressure, measures flowing and
shut-in pressures*
Side wall coring
Col I eels borehole fluid sample and measures
In situ formation pressures and temperatures
Determines relative magnitudes o£ permeability
under an imposed hydraulic stress'
.7
Bottom hole location with reference to suiface
location' and orientation of hole in ground
Boiehole fluid veloctfy.deteclion of Mow behind
the cosing,11 ground water flow patterns between
wells, permeability, identification of production
or injection horizons
Locales and estimates quantity of fluid flow
behind casing, flow rote"
Specific gravity profile, depth of pressure gradient
changes, location of fluid contacts, two-phase
fluid flow, slippage velocities ond volumetric (low
rote from each zone'
Formation water quality
Pmosify, permeability, mechanical properties, heot
conductivity, electrical resistivity
Borehole water quality
Radioactive tracer techniques best for measuring
very low flow rates
Temperature ond pressure changes must be
corrected, run with continuous flow meter for
mulli-phose fluid flow Interpretation, dynamic
range of 0 to 1.6 g/*l «" with 0.03 g/sq em
accuracy'
Con seal perforations mode*
• Sclilumberger Trademark
1 Boksr, et at. (1975)
2 Keys and MaeCary (1971)
3Wyllie (1963)
*Schl»mberger(l969, 1970,
197J, 1975, 1977)
5 S. K. Sanyal (pets, comm., 1978)
*Calvert, et al. (1977)
7 West, et al. (1975)
Ferronsky, et ol. (1968)
8
10
12
Keys ond Brown (1973)
Zemanelc, et at. (1970)
Britt (1976)
Peltitt (1975)
13 Muirand Zoeller (1967)
-------
those which may be derived from log analysis and interpreta-
tion; comments concerning the tool's limitations and advantages;
and, where available or applicable, its precision.
These tools, when used within their design limits, provide
direct measurements of:
1. borehole size and depth;
2. temperature, pressure and fluid flow within
the borehole; and
3. resistivity, sonic wave velocity and amplitude,
bulk density, gamma-ray intensity, neutron-
capture cross section, and self-potential of
penetrated formations.
Other parameters, which must be calculated, derived or
inferred from the log record include:
1. lithology, porosity, and permeability of
penetrated formations;
2. formation dips, pressure, density and
mechanical properties;
3. heat flow in the borehole;
4. borehole fluid characteristics;
5. casing condition;
6. cement bond quality;
7. borehole orientation; and
8. fracture detection and delineation.
A complete discussion of the principles and interpretation
^chniques of well logging methods is given by Keys and MacCary
(1971), Schlumberger (1970, 1974), Wyllie (1963) and several
°ther sources. A brief survey of these methods is presented
The electrical logging tools developed by the Schlumberger
Bothers in the late 1920s were the first to be extensively
by the petroleum industry. Downhole voltage and resistance
measured. Electrical tools are limited to use in uncased
fiberglass-cased holes. The logs can be used to define the
for stratigraphic correlation or .as indicators of
quality; they are especially useful in ground water inves-
121
-------
Radioactive logging is based on radiation scattering or
radioactive decay of unstable nuclei and the detection of the
emitted radiation. The radioactivity may be either natural or
induced, and can result from the injection of an isotope used as
a tracer. Because certain types of radiation are very
penetrating, these logs may be used in cased or uncased holes
and in the presence or absence of borehole fluids. Radioactive
logging techniques are useful for defining rock and fluid
properties in both cased and uncased holes; monitoring changes
in water quality in cased holes; detecting effluent leakage from
older wells; and determining ground water flow patterns between
wells.
Acoustic logging was developed in response to the need to
measure in situ sonic velocities of subsurface formations. These
tools measure compressional (P) and/or shear (S) wave velocities
and signal attenuation which are used to determine fractures,
porosity, cement bond quality, borehole size and condition, and
mechanical properties of the rock such as bulk modulus and
Young's modulus. Acoustic logs are run in uncased, fluid-filled
holes; however, some acoustic porosity logs have been run
successfully in cased holes (Muir and Zoeller, 1967).
The specialized tools have been developed to measure a
variety of parameters in the borehole, such as temperature;
fluid velocity, pressure and density; formation dip; and bore-
hole shape, size, condition and orientation.
Interpretation of rock and fluid properties from well logs
is enhanced by composite interpretation of several logs for the
same parameter. As stated by Keys and MacCary (1971):
As a general rule, the more types of geophysical
logs that are available for a single well and the
more wells that are logged within a given geohydro-
logic environment, the greater the benefits that
can be expected from logging. The synergistic
character of logs is due to the fact that each
type of log actually measures a different parameter,
and when several are analyzed together, each will
tend to support or contradict conclusions drawn
from the others. Similarly, a large number of
wells logged in one ground water environment will
provide a statistically meaningful sample of the
environment and reduce the chance of interpretive
errors.
122
-------
4.2 LOG INTERPRETATION PHILOSOPHIES
Well logging was developed specifically to service the
Petroleum industry in its particular environments for its
special problems. The objectives, operation, analysis and
interpretation of well logs in and around geothermal systems are
considerably different than those in petroleum systems. The
fundamental differences arise from the diverse physical charac-
teristics of the two types of reservoirs. Petroleum reservoirs
ittost often occur in relatively soft, sedimentary rocks, with
intergranular porosity, at temperatures less than 150°C (300°F)
and water saturation less than 100%. Geothermal reservoirs
Usually occur in hard, saturated, fractured igneous and
Netamorphic rocks at relatively high temperatures. Except for
Imperial Valley, California, very few hydrothermal systems of
commercial interest are known to occur in sedimentary rocks.
Because of these differences, direct transfer of petroleum
fogging technology to geothermal environments is impossible.
problems involved in developing the technology for geo-
environments are discussed in detail by Sanyal and
(1977) and summarized below.
4.2.1 Log Interpretation Problems
The most important objectives of petroleum well logging are
^termination of reservoir lithology, porosity, permeability and
^ydrocarbon saturation. For this reason, the bulk of research
J-or commercial logging tools has been confined to developing
newer and better tools and interpretation techniques for
Measuring these parameters. In geothermal well logging, the most
important objectives are the detection of large-scale faults and
fractures, and the determination of porosity (usually fracture-
type) , equilibrium formation temperature, and the thermal and
e-Lastic properties of reservoir rocks. These along with the
Assessment of water quality, necessary for ground water inves-
tions, have drawn little attention in the petroleum industry,
are relatively undeveloped.
General Problems —
im. C°re data from 9eothermal wells, necessary for accurate log
Interpretation, are scarce. Only a few dozen geothermal well
•J°gs and core analysis reports are publicly available. The
™ajor reason for the lack of geothermal cores, to date, is their
xpense. Presently the cost of coring is about $1,500 per meter
ompared to about $300 per meter for routine drilling (Los
^ Scientific Laboratory, 1978) .
Although it is advisable to run as many logs as economically
p°ssible, a full log suite may not be run in all geothermal wells
123
-------
because of cost consideration or operational problems. This
makes lithology identification and porosity evaluation diffi-
cult.
Another problem is that the present state of knowledge of
the effects of high temperature, pressure and geothermal fluid
chemistry on reservoir rocks is very primitive and the use of
core data to normalize well log data may not be justifiable in
many geothermal situations.
Assessing Log Quality—
The first task in geothermal well log interpretation is
assessing the log quality. Careful checking, crosschecking and
analysis are required to pinpoint any problems in the log. It
is important to have accurate drillers' logs, including records
of times at which drilling stopped, circulation stopped and
various logging tools reached the bottom; and the corresponding
temperature measurements. Unless these data are recorded, it
is difficult to assess the formation temperature at the time of
logging and other factors which affect rock and fluid properties
and log responses.
Sanyal and Meidav (1977) list several problems associated
with log quality in geothermal reservoirs which are also known
in the petroleum industry. These can usually be resolved when
they are detected. However, other types of problems, described
below, cannot be easily resolved.
Unfamiliar Lithology—
The state-of-the-art of well logging techniques and inter-
pretation is adequate for geothermal reservoir systems in
sedimentary environments, although some modification is needed
in analysis to account for possible contact metamorphism,
hydrothermal alteration or the presence of intrusives. Con-
ventional interpretation methods have been used with success in
shallow sedimentary ground water basins. However, an unfamiliar
lithology poses several problems in log interpretation. Standard
calibration and interpretation charts, log interpretation
equations and overlay and crossplot concepts for well log
analysis were designed for typical petroleum reservoir lithology:
sandstone, shale, limestone, dolomite and anhydrite. These
methods are often inadequate for nonsedimentary geothermal
reservoirs where crystalline igneous and metamorphic rocks,
glassy or crystalline volcanic rocks, and welded volcanic rock
material are often encountered. Empirical correlations so often
used for well log interpretation in sedimentary environments,
are yet to be developed for geothermal well log interpretation.
124
-------
For example, in the Raft River geothermal field, the points
on some multiple porosity crossplots fall outside the range of
standard charts. Lithology and porosity values can also change
abruptly with depth in a sedimentary section because of zones of
hydrothermal alteration or igneous intrusions which are often
encountered in geothermal environments. It is often difficult
to develop a useful empirical porosity-permeability correlation
from core data because of fractures and abrupt changes in
lithology.
Lack of calibration data for such lithologies also presents
a problem in interpretation. When an "unusual" rock type is
encountered, the log response may appear strange and some
correction may be necessary. For most unfamiliar lithologies/
such corrections cannot be estimated. For example, in the Hot
Dry Rock Project Well No. 2, some zones increase in bulk density
without corresponding decrease in the sonic travel time. This
Response was apparently a result of the presence of mafic rocks
encountered in these zones, having higher electron densities
than the medium used to calibrate the density logging tool (West,
et al. 1975). Calibration data were not available to resolve
the problem. Such "strange" responses can be expected from
Geothermal well logs run in unfamiliar lithology.
Conventional complex lithology analysis aided by computer
Processing, crossplots and histograms can be applied to igneous
and metamorphic rock formations. However, matrix properties,
such as bulk density, sonic travel time, matrix neutron porosity
and neutron cross section of these "unfamiliar" lithologies are
°ften unavailable or available only for a few of the minerals
which form these rocks (S. K. Sanyal, pers. comm., 1978). In
Edition, the mineral composition of these rocks may vary
Considerably from zone to zone. For sedimentary lithology,
these logs will be as usable as in the petroleum industry.
sanyal and Meidav (1977) summarize the utility of some conven-
tional well logs for geothermal reservoirs of the nonsedimentary
Ethology (Table 4.2).
4»2.2 Geothermal Reservoir Classification for Well Log
Interpretation ~~ """"""
The Department of Energy, in conjunction with the Los Alamos
Scientific Laboratory, has initiated a program to develop log
interpretation techniques for geothermal environments. As part
°£ this effort, a proposed classification scheme for geothermal
Reservoirs is being developed (S. K. Sanyal, pers. comm., 1978).
Reservoirs are grouped into a small number of classes to serve as
basis for the development of distinct sets of log responses
typical log analyses problems for each class. This system
be extended to well logging interpretation techniques for
125
-------
TABLE 4.2
UTILITY OF VARIOUS WELL LOGS
NON-SEDIMENTARY LITHOLOGY
(Sanyal and Meidav, 1977}
IN
LOG TYPE
1. Self Potential
COMMENTS
Often poorly developed; Unreliable -n massive igneous
or metamorphic sections; Works in seme 'ractured i^ieoi.5
and metamorphic sections; Usually not useful! for corre-
lation.
2. Electrical Resistivity
Mostly off scale in massive igneous and metamorphic
sections; Can be helpful in locating fractures; Short
spacing resistivity may be affected by the short-
circuiting effect of borehole fluid; Focussed devices
may be helpful for very resistive formations.
3. Acoustic Devices
BHC sonic log works well unless there is significant
hole enlargement; !^ay be useful for correlation;
Matrix velocity values nay not be known; Useful for
overlay and crossplot techniques; Full-wave sonic
log valuable in locating fractures; Borehole televiewer
valuable for locating fractures and delineating its
orientation.
4. Neutron Devices
Responses from unfamiliar lithology may be difficult
to assess; May be useful for correlation.
5. Density
Useful in estimating porosity and lithology; Caving
may oe a problem, Log response from unfamiliar litno-
logy difficult to assess; Useful for correlation.
6. Gamma Ray
Useful for correlation; Soectral gamma ray useful for
lithology identification.
7. Drilling Log
Drilling rate log useful for locating 'racture zones and
changes in lithology; Drill cuttings log useful for iitrto-
logy identification; Useful for correlation.
3, Caliper
Useful in resolving many log quality problems; 6-am
caliper can be used in determining dip of fracture-
zone washout ; Use in judging hole condition.
9. Dip Meter
Not always useful because of lack of laminated structure;
Dip of fractures may be unreliable.
10. Temperature
Useful for assessing temperature gradient, ana other
standard uses of temperature log.
*Schlumberger Registered Trademark
126
-------
defining baseline conditions of hydrologic systems in geothermal
environments. In this system, reservoirs are defined according
to:
I) Fluid type and temperature. Liquid- and vapor-
dominated systems have different log responses
as do oil and gas reservoirs. Temperature
affects most rock and fluid properties: in
geothermal reservoirs, well temperatures often
exceed the tolerance of many logging tools.
However, injection and monitoring wells are
expected to be within the tolerance range of
most tools (175°C [350°F]).
II) Lithology. Lithology is the most important
factor affecting log response, as the major
part of reservoir bulk density is the rock
matrix.
Ill) Overall geologic factors. Each geologic
province has a rock and fluid assemblage
particular to that area and associated with
certain characteristic log responses. This
type of classification is similar to the
use of names such as "Gulf Coast, Rocky
Mountains, California, North Sea," etc. in
the petroleum industry.
IV) Pore geometry. The nature and geometry of
pore spaces have strong influence on sonic,
electric and nuclear log responses from a
formation. In geothermal reservoirs, frac-
ture, or vesicular porosity often exists and
effective and total porosity values for these
reservoirs will differ.
V) Fluid chemistry. More diverse than most oil
field waters, geothermal waters vary both in
TDS concentration and chemical composition.
Log interpretation methods may need correc-
tion if major constituents other than sodium
chloride are present.
'This classification scheme is detailed in Table 4.3.
Some examples of well known geothermal areas typed by this
System are given in Table 4.4. With the increasing use of well
!ogs in the geothermal industry and the accumulation of core
from geothermal wells, interpretation of log response
nonsedimentary environments will develop. Well log data
127
-------
TABLE 4.3
GEOTHERMAL RESERVOIR CLASSIFICATION SCHEMES
(S. K. Sanyal, pers. comm., 1978)
I. According to Fluid Type & Temperature;
A. Steam
B. High-Temperature Water (>400°F)
C. Moderate-Temperature Water (300-400°F)
D. Low-Temperature Water (<300°F)
E. Dry
II. According to Lithologic Type;
A. Sedimentary
B. Metamorphic
C. Igneous (crystalline & glassy)
D. Volcanic Ash and Associated Sediments, Tuff
E. Breccia
F. Hydrothermally Altered
III. According to the Geologic Province;
A. Basin & Range
a. Wasatch Front
b. Central
c. Western Margin
B. Northwest Volcanic
a. Snake River
b. Cascade Range
c. Other
C. Salton Trough
D. Northern California Coast Range
E. Rio Grande Rift & Colorado Plateau Borderland
F. Hawaii
G. Alaska
IV. According to Pore Geometry;
A. Sedimentary Intergranular Porosity
B. Fracture
C- Vesicular or vuggy porosity
V. According to Salinity and Fluid Chemistry:
A. Low salinity (<5,000 ppm)
B. Moderate salinity (5-35,000 ppm)
C. High salinity (35-100,000 ppm)
D. Hyper saline (>100,000 ppm)
E. Dry
128
-------
TABLE 4.4
TYPING OF WELL-KNOWN GEOTHERMAL RESERVOIRS
(according to Classification Schemes in Table 4.3)
(S. K. Sanyal, pers. comm., 1978}
Reservoir
Location
Salton Sea
Brawley
Heber
East Mesa
Dunes
Cerro Prieto
The Geysers
Coso Hot Springs
t- Mono-Long Valley
£j Kilauea
Raft River
Mountain Home
Yellowstone
Roosevelt Hot Springs
Valles Caldera
Fenton Hill
Brady's Hot Springs
Steamboat Springs
Beowawe
Klamath Falls
Imperial County, California
Imperial County, California
Imperial County, California
Imperial County, California
Imperial County, California
Baja California, Mexico
Sonoma County, California
Inyo County, California
Mono County, California
Hawaii
Cassia County, Idaho
Elmore County, Idaho
Park County, Wyoming
Beaver County, Utah
Sandoval County, New Mexico
Sandoval County, New Mexico
Lyon/Churchill Co., Nevada
Washoe County, Nevada
Lander/Eureka Co., Nevada
Klamath County, Oregon
Type
IB, IIA/F, IIIC, IVA/B, VD
IB, IIA, IIIC, IVA, VC
1C, IIA, IIIC, IVA, VB
1C, IIA, IIIC, IVA, VB
ID, IIA, IIIC, IVA, VB
IB, IIA, IIIC, IVA, VB
IA, IIB, HID, IVB, VA
1C, IIB/C, IIIAc, IVB, VB
1C, IIC/D, IIIAc, IVB, VA
IB, IIC/F, IIIF, IVB/C, VA
ID, IIA/C/D, IIIBa, IVA, VA
1C, IIC/D, IIBa, IVB/C, VA
IA/B, IIC/D/F, IIIBc, IVB/C, VA
IB, IIC, IIIAa, IVB, VB
IB, IIC/F, HIE, IVB, VB
IE, IIC/F, HIE, IVB, VE
IB, IIA/C/D/F, IIIAc, IVA/B, VA
1C, IIC/D/F, IIIAc, IVB/C, VA
IB, IIC/D/F, IIIAb, IVB, VA
ID, IID/E, IIIBa/Ac, IVB, VB
-------
will provide valuable information for exploration and production
of the resource, for understanding subsurface injection environ-
ments, and for monitoring ground water degradation.
4.3 WELL LOGGING AND GROUND WATER MONITORING
Well logging can play a vital role in a ground water
monitoring plan. It can:
1) aid in the construction of injection and
observation wells;
2) help define the baseline conditions of the
subsurface injection environment, nearby
ground water systems and their inter-
relationship;
3) monitor the condition of the injection
we11; and
4) aid in monitoring wastewater flow patterns
and possible degradation of fresh ground
water throughout the monitoring network
These applications of well logs are discussed below and are
summarized in Tables 4.5, 4.6, 4.7 and 4.8.
Well design and construction factors such as hole diameter/
drilling method, casing, mudcake and degree of mud invasion
affect log quality (Keys and Brown, 1973). Where logging is to
play an important role in the monitoring plan, wells should be
designed for the maximum efficiency of logging.
4.3.1 Well Construction
Wells must be carefully constructed to intersect a deter-
mined horizon for injection and monitoring operations. Direc-
tional surveys are used to guide the drilling direction of the
well, determine the position of the hole in the ground, and
accurately locate the bottom of the hole. Caliper logs reveal
the actual shape and condition of the borehole, and allow
estimates of the volume of cement needed for casing installa-
tion. After casing installation, temperature and gamma-gamma
logs can be used to establish the position of the cement behind
the casing (Fig. 4.1; Keys and MacCary, 1971). Cement bond logs
show the degree of bonding between the casing and cement and
the cement and formation. The information from both of these
logs can determine whether the cement in the annulus provides
an adequate seal between the casing and borehole. If this seal
is not adequate, vertical migration of fluids may occur and
contaminate surrounding aquifers.
130
-------
TABLE 4.5 WELL LOGGING FOR WELL CONSTRUCTION
OBJECTIVE
LOGGING TOOL
Orientation of borehole and bottom-
hole location
Shape and condition of borehole (loca-
tion of caving or washout zones)
Volume of cement needed for casing
installation
Position of grout behind casing
Cement bond integrity
Top of cement
Depths of casing, tubing, screens and
perforations
Condition of casing
Casing leaks and/or plugged screens
Leaks in annular cement
Guide to screen setting or perforations
Directional survey
Caliper log
Caliper log
Temperature and gamma-gamma logs
Cement bond and temperature logs
Cement bond and temperature logs
Casing-collar locator
Caliper, borehole televiewer, casing
inspection logs
Tracer techniques, flowmeter log
Tracer techniques, temperature and
cement bond logs and gamma-gamma
log in older wells
All logs which provide data on
lithology, porosity and permeability
of units (see Table 4.6), flowmeter
and temperature logs
131
-------
TABLE 4.6 WELL LOGGING FOR DEFINING BASELINE CONDITIONS
(modified from Keys and MacCary, 1971)
OBJECTIVE
LOGGING TOOL
Lithology and strati graphic correlation
of aquifers and associated rocks
Total porosity or bulk density
Effective porosity
Secondary porosity and fracture
identification
Permeability
Location of water level or saturated
zones
Moisture content above the water
table
Direction, velocity and path of
ground water flow
Source and movement of water in a
well
Chemical and physical characteris-
tics of water (water quality,
salinity, temperature, density,
viscosity)
Electric, sonic, nuclear and caliper
logs in open holes, nuclear logs in
cased holes
Calibrated sonic logs in open holes,
calibrated gamma-gamma or neutron
logs in cased holes
Resistivity and nuclear magnetism
logs in open holes
Injectivity profiles and caliper,
resistivity, acoustic, borehole
televiewer and dipmeter logs in
open holes
No direct measurement made by
logging. May be related to porosity
and sonic, flowmeterand temperature
logs; injectivity profiles or tracer
techniques
Electric and temperature logs in open
holes, temperature, neutron and
gamma-gamma logs in cased holes
Neutron logs in open and cased holes
Single well and multi-well tracer
techniques, temperature and flow-
meter logs in open and cased holes
Injectivity profile, tracer techniques
or flowmeter logs during pumping
or injection, temperature logs
Electric logs in open holes, tempera-
ture, neutron, pulsed neutron and
neutron activation logs and Formation
Interval Tester* in cased holes
Electric logs mayolso be run in fiberglass cased holes.
* Schlumberger Trademark
132
-------
TABLE 4.7 WELL LOGGING FOR INJECTION WELL MONITORING
OBJECTIVE
LOGGINGTOOL
Condition of injection well tubing,
casing and cement
Injection pressure and flow rates
Location and direction of waste
dispersion, dilution and movement
from well
Change in relative formation water
quality and in brine/fresh water
interface
Leakage from the well
Aquifer solution or plugging
Casing inspection, cement bond and
caliper and borehole televiewer
Fluid pressure and flowmeter logs
Temperature, flowmeter and fluid
conductivity logs, tracer techniques
All logs providing data on the chemical
and physical characteristics of water
(see Table 4.6)
Temperature, nuclear, cement bond
logs, radioactive tracer techniques
Nuclear logs
133
-------
TABLE 4.8 WELL LOGGING FOR OBSERVATION WELL MONITORING
IN GEOTHERMAL ENVIRONMENTS (modified from U.S. EPA, 1975)
WELL TYPE
OBJECTIVE
LOGGING TOOL
Constructed in receiving
aquifer
Constructed in or just
above confining unit
Constructed in a fresh
water aquifer above or
downgradient from
receiving aquifer
Pressure in receiving
aqui fer
Rate and direction of
wastewater movement
Geochemical changes
in injected waste-
water
Detect shifts in fresh
water-saline water
i nterfaces
Detect leakage
through confining
unit
Detect evidence of
fresh water contam-
ination
Pressure log
Temperature, flowmeter and
fluid resistivity logs, radio-
active tracer techniques
All logs providing data on
the chemical and physical
characteristics of water (see
Table 4.6)
Electric and nuclear logs
Temperature, nuclear and
electric logs, radioactive
tracer techniques
Temperature, electric,
nuclear and all logs
providing data on water
quality (see Table 4.6)
134
-------
CMIPfRlOO GAMMA GiMMA 100
fU CEMENT POSTCEMfNT
GAMMA IOC
PHI CEMENT
Figure 4.1 Gamma-gamma logs used to interpret
position of grout behind casing, and
caliper logs used to select depth
for grouting and to estimate volume
required. (Keys and MacCary, 1971)
135
-------
The exact depths of casings, tubings, screens and perfora-
tions can be found with the casing-collar locator and the top
of cement with the temperature or cement bond logs. Well logs
can also be used to locate the most permeable units to set
perforations and screens. Most of these logs can be analyzed
at the well site to permit rapid decision making.
4.3.2 Defining Baseline Conditions
Well logs can provide essential data on subsurface geologic
and hydrologic parameters and thus aid in defining baseline
conditions and characteristics for a potential injection zone
and for the nearby ground water systems. Well logs furnish
continuous downhole, measurements of rock and fluid properties
which are essential to forecast aquifer conditions and predict
ground water flow patterns. Well logs can also give data to
predict the quantity and flow rate of wastewater acceptable to
che injection horizon. These parameters are summarized on
Table 4.6 and discussed below.
Lithologic Parameters—
The identification and correlation of lithologic units has
been one of the prime functions of well logging in the petroleum
industry. As discussed by Keys and Brown (1973), the primary
task for waste injection monitoring is the prediction of waste
movement. Such predicting requires an understanding of subsur-
face lithology, geometry and uniformity of subsurface units.
The dispersion coefficient, necessary to calculate underground
dispersion, is related to the geometry and uniformity of aquifers
and confining beds. Continuous downhole in situ measurements
of lithologic parameters by well logging also allow more
accurate predictions of well and injection performance since the
extrapolation of laboratory data from cores to the in situ
conditions may be erroneous. As discussed above, coring is also
very expensive in most geothermal environments because the forma-
tions are highly consolidated. Furthermore, in highly fractured
formations, information from cores on porosity and permeability
may be meaningless.
Lithology—Lithology may be determined from electric, sonic/
nuclear and other logs in open holes and by nuclear logs in
cased holes (Table 4.6). These logs measure differences in rock
characteristics due to density, mineral composition, porosity
and relative strength of units which can be interpreted to
decipher lithology. For maximum effect, several different logs
which measure different aspects of lithology should be run.
Since a log response is not unique, background information and
knowledge of regional geology in a new area is essential for
correct log interpretation.
136
-------
Well logs also provide a basis for delineating stratigraphy,
Keys and MacCary (1973) use gamma -ray and neutron logs to
correlate sedimentary units in a saline water basin (Fig. 4.2) .
The difference in character of both gamma-ray and neutron logs
in Hole T-14 above and below 457 m (1500 ft) is apparent and
marks the contact between two formations. Above this contact,
the holes penetrated fine-grained clastic rocks with a few beds
of anhydrite. Below 457 m (1,500 ft) lithology is dominated by
mudstone with numerous anhydrite beds. Anhydrite beds show up
as strong deflections to the left on the gamma logs.
Well log data can be essential to identifying aquifers and
in constructing contour maps showing aquifer thickness and
structure in unknown complex systems. Jones (1961a, 1961b)
describes the use of natural gamma-ray and caliper logs to
define the complex basalt aquifer system at the National Reactor
Testing Station, Idaho. Prior to this investigation, definition
of the geometry of basalt flows and interflow sediments of the
Snake River Plain had not been possible. From well log data,
Jones (1961b) correlated interflow sediments, identified the
ttost permeable units and was able to construct cross sections of
this very complex system (Fig. 4.3).
Since ground water temperature is rarely the same in
Different aquifers, temperature logs can be used to identify
Aquifers or perforated intervals contributing water to a flowing
well. Keys and MacCary (1971) show a series of temperature logs
from an unsuccessful plugged oil well in which perforated inter-
vals contributing water to the well are clearly distinguished
temperature logs (Fig. 4.4). Log A was made before the well
perforated and shows the normal temperature gradient profile
the field. Log B was made after perforations were made
the perforated intervals between 150 to 130 m (500 to 600
ft) are clearly distinguished. Log C shows that initially most
°f the water moving up the casing was coming from the 175 m (575
ft) aquifer. Log D shows that, after pumping for 24 hours, the
^•57 m (515 ft) aquifer began to contribute some water. Tempera-
ture logs B through D all show warmer water moving up the well
the casing was perforated.
Fractures — The delineation of large-scale fractures and
faults is also an important factor in defining baseline condi-
Fractures intersecting the borehole may be located
caliper, resistivity, acoustic, borehole televiewer and
logs. Fig. 4.5 shows the correlation of fracture zones
two wells by caliper logs. The repeatability of the
made on different dates in Hole C-l is evident. Small
found by the caliper at a depth of 58 m (190 ft) in
**ole c-1 can be correlated with similar fractures found at 55 m
ft) in Hole T-5, 5 km (3 mi) distant from Hole C-l. The
137
-------
RADIATION. IN PULSES PER SECOND
....
CO
2000'
1900
1800
1700
1600
1500
140C
1300
1200
4O 5060 70 100
I I I I L_
150
200
_J
RADIATION. IN PULSES PER SECOND
50 70 90 50 150 250
I i i J
RADIATION. IN PULSES
PER SECOND
bO 70 250 350
_BOTTOM LOG
RADIATION. IN PULSES PER SECOND
20 50 80 100 200
I L J L
GAMMA TLJ- NEUTRON-N
DICKENS CO
KENT CO
GAMMA J NEUTRON-N
40OO BOOO FEET
GAMMA T_5 NEUTRON-N
DATUM IS MEAN SE> LEVEL
Fic 2 Stratigraphic correlation with gamma and neutron logs, Upper Brazos
River Basin, Texas. (Keys and MacCary, 1973)
-------
1360
1350
1340 -
1330
1320
1310
1300 -
0 200 400 800 Feet
i H r1—r 1 '
0 100 200 Meters
4480
4460
4440
4420
4400
4380
4360
£ -J4340
I |4320
3
S I 4300
!T J 4280
<
- - 4260
- 4240
Figure 4.3 Structure and thickness of aquifers A to E based on well logs, National
Reactor Testing Station, Idaho. (Jones, 1961b)
-------
DEPTH
VIETERS
a.rc
Cosing |«t perforated
and then welt surged
August 6 through 9
Will pumping at 5 2°Cl Well pumping at
appro.imalely 50 flpm \ approximately K< gpm
after 24 hours of pomprng
Well italic
after perforation
Well static
before perforation
Upward flo~
brrne Iroce/ector log
Approximately M gpm
Casing interval
perforated
600
Figure 4.4 Temperature logs, Yukon Services well, Cook
Inlet field, Anchorage, Alaska.
(Keys and MacCary, 1971)
- 183
140
-------
SINGLE-POINT RESISTANCE LOGS
HOLE C-l — 3 MILES -HOLE T-5
CALIPER LOGS
HOLE T-5 Logged 11-26-64 HOLE C-l Logged 2-27-65
~\ i ii r~ ~~T
678 786 7 8 9 10 II 12
.15) (18) (20) (18) (20)05) (18) (20) (23) (25) (28) (30)
HOLE DIAMETER, IN INCHES (cm)
Figure 4.5 Correlation of fracture zones between rotary hole T-5
and core hole C-l utilizing caliper logs, Upper
Brazos River Basin, Texas. (Keys and MacCary, 1973)
-------
single-point resistance logs indicate that the fracture zones
in C-l and T-5 are in the same stratigraphic position (Keys and
MacCary, 1973).
Caliper and dipmeter logs may also give information on the
orientation and direction of fracture propagation from the
borehole. Landt, et al. (1977) describe several innovative,
theoretical techniques for fracture mapping using well logging
techniques.
Porosity—The distribution of pore spaces, or porosity, and
permeab i 1ity, which can have a local relationship to porosity,
are the most significant parameters for aquifer identification
and injection control. Porosity may be derived from electric,
nuclear, acoustic and specialized logs. An estimation of
porosity is essential for predicting the path of wastewater
dispersion and velocity and in calculating reservoir volumes.
In deriving values for porosity, it is necessary to note
the type of porosity present. Total and effective porosity are
nearly the same in sedimentary rocks; however, fractured igneous
and metamorphic rocks and volcanic rocks commonly have isolated
void spaces, making their total porosity greater than effective
porosity. Furthermore, in fractured rock, porosity values
obtained around a well bore may not be characteristic of the
reservoir.
Neutron logs indicate moisture content above the water
table and total porosity below. Chemically bound water cannot
be distinguished from free water on neutron logs and erroneous
values may be obtained if mineral hydrates are encountered.
Gamma-gamma logs are the most useful for determining total
porosity. Effective porosity can be estimated by nuclear
magnetism logs. Fig. 4.6 shows porosities measured from rock
cores in the laboratory and by neutron logs in an adjacent
hole. Once calibrated against a core, the neutron logs can be
used to estimate porosity values in the same lithologic units,
provided that hole diameter and borehole fluid composition are
known (Keys and MacCary, 1973) and the porosity is relatively
constant throughout the formations, as is common in sedimentary
rocks. Cross plots of at least two types of logs, each evalu-
ating a different aspect of the formation porosity, yield the
most accurate values.
Zones of fracture, or secondary, porosity, common to most
geothermal reservoirs, may be delineated by electric, acoustic,
caliper and dipmeter logs in uncased holes. Effective fracture
porosity in highly resistive plutonic and metamorphic saturated
rock may be identified from resistivity logs, since the water
filled fractures are relatively conductive. Single-point
resistivity logs can give accurate values for effective fracture
142
-------
CO I0°
£ (30)
I
h I
i .
<
200 r
300
DO
ILl
UJ
- 400
x (122)
h-
a
LJ
Q
/
100 200 300
RADIATION. IN PULSES PES SECOND
NEUTRON-N LOG IN
CASED HOLE
r-n
.
- — - — -.""-," -T .=1 J
CORE POROSITY MEASURED
IN THE LABORATORY
C-l
Figure 4.6 A comparison of a
neutron log with porosity
core samples, Upper Brazos
Texas. (Keys and MacCary,
River
1973)
Basin,
143
-------
porosity because this log detects nearly closed fractures
intersecting the borehole which are not delineated by other logs
(Keys and MacCary, 1971).
Siple (.1964) describes the use of acoustic velocity and
caliper logs to determine the distribution of fractures in
crystalline rock. Water transmitting properties of these
fractures were determined between wells by tracers (Marine,
1966) .
Permeability—Although permeability is the most important
parameter in the petroleum industry, no well logging tool
directly measures permeability. Permeability can be indirectly
measured from electrical (Ershaghi, et al. 1978), neutron,
and specialized logs. In sedimentary environments, an empirical
relationship between natural gamma log intensity and permeability
exists in specific areas (Rabe, 1957; Gaur and Singh, 1965).
However, derived empirical relations can only be used in
consistent geohydrologic environments and only after their rela-
tionship has been clearly established.
Keys (1967) uses injectivity profiling with radioactive
tracers to measure relative magnitudes of permeability by
plotting the water lost within each depth interval under the
imposed hydraulic stress. Fig. 4.7 is an example of such an
injectivity profile and its relationship to a caliper log.
Flowmeter logs used during pumping or injection indicate
most permeable zones based on water acceptance rate. Tempera-
ture logs can also delineate zones of water acceptance (Keys and
Brown, 1973). Velocity of ground water flow is an important
property related to permeability. Logging can provide rate
and direction of ground water flow if tracers are injected
into one well and searched for in another.
Fluid Parameters—
In addition to lithologic parameters, fluid parameters can
be obtained from well log data. These include formation and
borehole water quality, location of brine-fresh water interface,
fluid temperature, regional ground water flow patterns, and
fluid movement within the borehole.
Wat_er_ Quality—At any given temperature, the TDS content
of waters with predominantly sodium chloride character is directly
proportional to fluid resistance or conductivity. If other ions
are present, in some instances multiplying factors may be used
for conversion to electrically equivalent sodium chloride
concentrations. Because of this relationship, electric logs
provide data on water quality in uncased or fiberglass cased
holes.
144
-------
DEPTH
FEET
(METERS)
1250
(381)
1300
(396)
1400
(427)
1500
(457)
1600
(488)
—
—
i
-
1
|
._L
1
L
1
i
i i i
i
40 ,50 ,60 ,70
80
1
)
0
1 "1
L
J
i1
i
2 ,3
'
j
!5°
L
r-
^
L
c
[•
r
? i«°
GALLONS PER DAY LOSS
GALLONS PER DAY
PER FOOT
CALIPER LOG
INJECTIVITY PROFILES
Figure 4.7 Injectivity profiles and their relation to a
caliper log. (Keys and MacCary, 1971)
145
-------
In cased and uncased holes, neutron logs give chlorine
density and correspondingly relative water quality. Pulsed
neutron logs are also sensitive to quantities of boron, lithium
and silica in formation waters. Neutron activation logs give a
qualitative technique for identifying several elements easily
activated by thermal-neutron capture such as chloride, aluminum,
silica, magnesium, potassium, manganese and sodium (Keys and
McCary, 1971) and can indicate relative water quality in known
lithologics.
Fluid conductivity logs are a continuous record of the
conductivity of fluid in the borehole. Olmsted (1962) describes
the use of fluid conductivity and fluid temperature logs to
establish ground water flow patterns and to map the distribution
of ground water types and injected wastes from the National
Reactor Testing Station, Idaho.
Keys and MacCary (1973) demonstrate the use of several logs
to locate changes in TDS concentration of formation water with
depth. The right-hand log in Fig. 4.8 shows a large deflection
in the fluid conductivity log caused by changes in the chemical
quality of the borehole fluid, which is the location of the
brine-fresh water interface in the well. Because the fluid
conductivity tool only responds to the borehole fluid, the
interface in the rocks might be at some other depth. In this
situation, however, the shift on the neutron log on the left
side of Fig. 4.8 indicates that the interface in the borehole
corresponds with the actual one in the surrounding rocks.
Fluid Movement—Horizontal and vertical fluid flow within
a we 11 can be de»fTned by using flowmeter and temperature logs
and by various systems of injecting and detecting radioactive
or chemical tracers. These logs can also indicate the horizons
of water acceptance during injection.
Vertical flow in wells complicates sampling for water
quality. Without knowledge of flow, water samples may be
taken from stagnant zones or at depth intervals that do not
represent water quality in adjacent aquifers.
Techniques for measuring natural or induced vertical move-
ment of borehole fluids have been widely used in both the
petroleum and ground water wells. Flowmeter logs provide
continuous records of fluid velocities. Temperature and flow-
meter logs can establish entrance and exit zones of water in
the well. Repeated in several wells, areal patterns of ground
water flow can be established.
Single well tracer techniques employ various tracers
injected into a column of water and their movement to a detected
is timed. investigations with radioactive tracer techniques
146
-------
RESISTIVITY. IN OHM METERS
012345
L_ 1 _L_ L 1 I
NEUTRON-N
RADIATION, IN PULSES PER SECOND
TEMPERATURE
IN DEGREES
FAHRENHEIT
FLUID RESISTIVITY
I 1 1 1 '—] 1
0 0.1 0.2 0.3 0.4 05 0.6
RESISTIVITY, IN OHM METERS
Figure 4.8 Location of brine-fresh water interface on neutron and fluid
-------
showed upward and downward flow in the same monitoring well,
the changes of whole flow pattern being induced from injection
and pumping rates in nearby wells (Barraclough, et al. 1965).
Several radioactive tracer techniques including point dilution,
single-well pulse, and use of directional detector are discussed
by Keys and MacCary, 1971. These logs can record depth and flow
rate of horizontal flow into an aquifer and are used when flow
rates are too low to be measured by flow meter logs.
4-3.3 Mcmitgring the Injection Well
Injection well monitoring is the foremost method of waste
injection monitoring, as the greatest risk of waste fluid
escape is through pr around the outside of the injection well
rather than by leakage through permeable confining beds, frac-
tures or unplugged wells (Talbot, 1972) . Fluid movement takes
place behind the casing when the seal between the casing and
formation fails. Cement failure can result from a poor cement
job, fracturing of the cement sheath at the time of perforation,
or cracking of cement due to earth movements. The applications
of well logs to injection well monitoring are summarized in
Table 4.7 and discussed below.
At the injection well, well logs can monitor such factors
as:
1) condition of the injection tubing, well casing
and cement bond;
2) pressure and flow rate of injection;
3) location and direction of waste fluid movement
from the well;
4) relative changes in formation water quality and
temperature profiles along the well;
5) changes in the receiving aquifer porosity due to
aquifer solution or plugging; and
6) well efficiency, which is related to the pressure
and flow rates in the well and changes in porosity
of the receiving aquifer.
A program of injection well monitoring by well logs involves
continuous monitoring of pressure and flow rates and periodic
measurements of well condition and changes in other parameters
listed abuve. Direction of flow from the well and the location
of movement from the well can be determined by radioactive tracer
techniques, flow meter and temperature logs. Casing inspectionr
sonic, and to a lesser extent caliper, and borehole televiewer
148
-------
logs can determine the condition of tubing and casing, noting
areas of corrosion. Cement bond logs are essential to determine
the condition of the cement bond in the well. Fig. 4.9 shows a
hole in the casing, located by casing caliper and casing inspec-
tion logs.
Leakage from the well cannot always be noted by injection
pressure drops (increased well efficiency) or evidence from
cement bond or casing inspection logs. Radioactive tracer
techniques and temperature logs can be used to locate zones of
fluid flow behind the casing and may be the only way of detecting
whether annular leakage is or is not occurring from the -well
(Keys and Brown, 1973). Leakage from the well or movement of
formation waters from one zone to another over a period of time
can cause precipitation of radioactive minerals resulting in
an interval of high radioactivity behind the casing. This
anomaly can be picked up by comparison of recent and old gamma-
ray logs. Relative changes in formation water quality, such as
an increase in chloride, silica, boron or lithium content may
also be evidence of leakage from the injection well. Relative
changes in pulsed neutron, neutron activation, or electric
logs can indicate these possible changes in water quality.
Formation tester tools can also be used to determine water
quality by recovering samples of formation water through the
casing which are then chemically analyzed.
Changes in injection pressures can also imply solution or
plugging in the receiving aquifer. Gamma-gamma logs can
identify zones of hole enlargement behind the casing and pulsed
neutron and neutron activation logs may indicate areas of silica
Precipitation. Neutron logs and other porosity indicators can
reveal plugging of the aquifer.
Knowledge of post-injection fracturing is important to
Monitor the response of the aquifer to injection. Periodic
inspections with sonic logs can indicate changes in the elastic
Properties of reservoir rocks, and indicate conditions that could
°ause fracturing.
At the Rocky Mountain Arsenal well, thermal differential
stress is postulated to be a triggering mechanism contributing
to microfracturing of the rock matrix and induced seismic
Activity (Hoover and Dietrich, 1969). Temperature control of
injected geothermal wastewater in El Salvador also controlled
silica precipitation problems. Hence, temperature logs are
^n important tool in maintaining favorable conditions in the
injection well.
149
-------
INSIDE DIAMETER, Inches (cm) WALL THICKNESS, Inches (cm)
6.IU5.5) 6.2(15.7) 6.3p6Q) .328(-83) .408(1.04) .488(1.24)
6400
(1951)
HOLE
6450
(1966)
6500
(1982)
DEPTH, FT,(m)
Figure 4.9
Typical simultaneous electronic casing caliper
and casing inspection logs run in 7-inch casing*
(U.S. EPA, 1977)
150
-------
4.3.4 Monitoring Observation Wells
U.S. EPA (1975, 1977) proposes at least three different
types of monitoring wells and pertinent data to be collected
from each. The objectives and useful well logging tools for
monitoring at these three types of observation wells are
summarized in Table 4.8 and discussed below.
Observation wells are constructed where wastewater is
expected to appear first if there is leakage from the injection
horizon. Temperature and resistivity logs can be useful in
mapping wastewater fronts if the injected water is of a dif-
ferent temperature and TDS concentration than the regional ground
water system. At the National Reactor Testing Station, tempera-
ture and fluid resistivity logs were used to map the distribu-
tion from a deep disposal well (Fig. 4.10 and 4.11). Tracer
techniques may also be employed to map wastewater movement if
the tracer is introduced into the injection well and monitored
at observation wells.
Observation wells constructed in or just above confining
units of the injection horizon may be used to detect leakage
through the unit by temperature and water quality logs and tracer
techniques.
Observation wells constructed in the injection zone, can
Monitor pressure in the receiving unit and, depending on well
location, direction of wastewater flow may be deduced.
In observation well construction, it is possible to leave
the bottom few meters of the hole uncased or cased with plastic
°r fiberglass. In this way, electric well logging tools may be
Used, as well as pulsed neutron and neutron activation logs to
Monitor changes in regional water quality and map wastewater
fronts or plumes.
4.4 WELL LOGGING COSTS
Certain well logs must be run by a well service company
and others can be run by the geotherraal developer. For collec-
tion of baseline data, running as many logs as economically
Possible ensures the most accurate interpretation of rock and
fluid properties. In this instance, the service of a well
logging company will be required. However, for routine
Monitoring, the logging tools may be purchased, thereby reducing
expenses. These tools would consist of those used most
frequently for monitoring, such as those for temperature, flow
Meter and pressure logs. Periodic monitoring of the condition
°f the injection well, which requires more sophisticated and
varied equipment, usually will be performed by well service
Companies.
151
-------
Figure 4.10
Maximum temperature of water between
depths of 140 to 200 m (459 to 656 ft)
based on well logs of monitoring wells
near a disposal well, National Reactor
Testing Station, Idaho. (Jones, 1961b)
l 52
-------
WELL
43
WELL
4 7
WELL
49
WELL
59
UJ
-
fl
01
CO
C4400
«
a
u
>
c
0)
4300
w
•a
4200
WATER WATER
RESISTIVITY TEMPERATURE
OHV-METERS DEGREES F
18 . a_2 S3 36 60
WATER WATER AAr£R
R£SiSTi\,iTv TEWPERATJRE KESiSTiviTY ^EMPER AT ^R
DEGREES f CHM-VETERS DEGREES <=
2 >6 2.0 _ 5i 60 _ i4 IB 52 5 C 60
WATER WATER
RESISTIVITY TEMPERATURE
OHM-METER? DEGREES F
,4 IB _ 52 _5.6
EXPLANATION
O CLAY
[~j SILT 8 SAND
{^3 SCORIA
5UJ SILT a CINDERS
M DENSE BASALT
400
800 FEET
c
O
I
E
a,
O
D
BD
a>
E
^
-
•a
3
C
100
200 METERS
1360
1340
320
1300
1280
Figure 4.11
Cross section through part of National Reactor Testing Station,
Idaho, showing decrease in temperature and increase in resistivity
of water with distance from the disposal well, which is nearest
Well 43. (Jones, 1961b)
-------
The cost of well logging by well service companies depends
on: 1} the depth of the well, 2) the borehole condition (e.g.
unusual or high temperature conditions), 3) the well site
province as defined by the service company, and 4) the roundtrip
mileage from the company base. Fig. 4.12 shows the cost of
running individual logs versus the depth of the wells. These
costs are taken from Schlumberger Price List for California
Lands, 1978. Costs for individual logs are generally reduced
when logs are run concurrently. Combination services available
and prices are described in well service company brochures.
Charges must also be added for mobilization and for wells with
high bottomhole temperatures (over 150°C [300°F]). These costs
must be determined individually for each site.
154
-------
Figure 4.12 Well Logging Costs (Schlumberger, 1978)
155
-------
REFERENCES
Baker, L. E.r A. B. Campbell and R. L. Hughen. 1975. Well
Logging Technology and Geothermal Applications: A Survey
and Assessment with Recommendations. Sandia Laboratories
Report for ERDA, No. 75-0275. 67 pp.
Barraclough, J. T., W. E. Teasdale and R. G. Jensen. 1965.
Hydrology of the National Reactor Testing Station, Idaho.
U.S. Atomic Energy Commission Div. Tech. Inf. Rept. IDO-
22048. 107 pp.
Britt, E. L. 1976. Theory and Applications of the Borehole
Audio Tracer Survey. Transactions of the SPWLA
Seventeenth Annual Logging Symposium, June 9-12, Denver.
Society of Petroleum Engineers of AIME Paper No. 6552.
35 pp.
Calvert, T. J., R. N. Ran and L. E. Wells. 1977. Electro-
magnetic Propagation ... A New Dimension in Logging.
Society of Petroleum Engineers of AIME. Paper No. SPE
6542. 15 pp.
Ershaghi, I., E. L. Dougherty, D. Herzberg, and H. Ucok. 1978.
Permeability Determination in Liquid-Dominated Geothermal
Reservoirs Using the Dual Induction Laterolog. Paper
presented at the Society of Professional Well Log Analysts
Nineteenth Annual Logging Symposium, El Paso, Texas,
June 14-16. 20 pp.
Ferronsky, V. I., A. I. Danillin, V. T. Dubinchuk, V. S.
Goncharov, V. A. Polyakov, Yu. B. Seletskiy, and N. Ya.
Flekser. 1968. Radioisotope Investigative Methods in
Engineering Geology and Hydrogeology (Radioizopnye
methody issledovanija v inzhenernoy geologii i gidro-
geologii), Moscow, U.S.S.R., Atomizdat.
Gaur, R. S., and I. Singh. 1965. Relation Between Permeability
and Gamma-Ray Intensity for the Oligocene Sand of an
Indian Field, India (Republic). Oil and Natural Gas
Comm. Bulletin 2. pp. 74-77.
156
-------
Hoover, D. B., and J. A. Dietrich. 1969. Seismic Activity
During the 1978 Test Pumping at the Rocky Mountain Arsenal
Disposal Well. USGS Circular 613. 35 pp.
Jones, P. H. 1961a. Hydrology of Radioactive-Waste Disposal at
the Idaho Chemical Processing Plant, National Reactor
Testing Station, Idaho. In: Geological Survey Research,
1961. USGS Professional Paper 424-D. pp. D374-D376.
. 1961b. Hydrology of Waste Disposal, National
Reactor Testing Station, Idaho. An Interim Report.
U.S. Atomic Energy Commission Div. Tech. Inf. Rept.
IDO-22042. 82 pp.
Keys, W. S. 1967. The Application of Radiation Logs to Ground-
water Hydrology. In: Isotopes in Hydrology. Vienna,
Austria. International Atomic Energy Agency Symposium.
November 14-18, 1966. Proceedings. pp. 477-488.
Keys, W. S., and R. F. Brown. 1973. Role of Borehole Geophysics
in Underground Waste Storage and Artificial Recharge. In:
Underground Waste Management and Artificial Recharge.
Proceedings of the Second International Symposium on
Underground Waste Management and Underground Recharge,
Louisiana. Sponsored by the American Association of
Petroleum Geologists and the USGS. Vol. 1. pp. 147-191.
Keys, W. S., and L. M. MacCary. 1973. Location and Character
of the Interface Between Brine and Fresh Water from
Geophysical Logs of Boreholes in the Upper Brazos River
Basin, Texas. USGS Professional Paper 809-B. 23 pp.
Keys, W. S., and L. M. MacCary. 1971. Applications of Borehole
Geophysics to Water-Resources Investigations. In: Tech-
niques of Water Resources Investigations of the USGS. Book
2. Chapter EL. 126 pp.
Landt, J. A., J. C. Rowley, J. W. Neudecker and A. R. Koelle.
1977. A magnetic Induction Technique for Mapping Vertical
Conductive Fractures. Status Report. Los Alamos
Scientific Laboratory. Report LA-7049-SR. 9 pp.
^os Alamos Scientific Laboratory. 1978. Geothermal Log
Interpretation Newsletter. LASL 78-4, No. 2. M. Mathews,
Project Manager.
Marine, I. W. 1966. Hydraulic Correlation of Fracture Zones in
Buried Crystalline Rock at the Savannah River Plant, near
Aiken, South Carolina. USGS Professional Paper 550-D.
pp. D223-D227.
157
-------
Muir, D. M., and W. A. Zoeller. 1967. New Acoustic Tool Logs
Cased Holes. Oil and Gas Journal. October 23. pp.
106-112.
Olmsted, F. H. 1962. Chemical and Physical Character of Ground
Water in the National Reactor Testing Station, Idaho.
U.S. Atomic Energy Commission Div. Tech. Inf. Kept. IDO-
22043. 21 pp.
Pettitt, R. A. 1975. Testing, Drilling and Logging of
Geothermal Test Hole GT-2, Phase III. Los Alamos
Scientific Laboratory Report LA-5965-PR. 13 pp.
Rabe, C. L. 1957. A Relation Between Gamma Radiation and
Permeability, Denver-Julesburg Basin. American Institute
of Mining, Metallurgical and Petroleum Engineers
Transactions. Vol. 210. pp. 358-360.
Sanyal, S. K. 1978. pers. comm. Manager, Stanford University
Petroleum Research Institute, Stanford, California.
Sanyal, S. K., and H. T. Meidav. 1977. Important Considerations
in Geothermal Well Log Analysis. Society of Petroleum
Engineers, American Institute of Mining, Metallurgical and
Petroleum Engineers, Inc. Paper No. SPE 6535. 6 pp.
Sanyal, S. X. , and R. B. Weiss. In press. Borehole Geophysical
Logging as Complement to Well Effluent Sampling. in:
Proceedings of the Second Workshop on Sampling and Analysis
of Geothermal Effluents. Las Vegas, U.S. Environmental
Protection Agency. pp. 211-216.
Schlumberger Well Services. 1978. Price Schedule, General Terms
and Conditions, California Lands.
. 1977. Log Interpretation Charts. Schlumberger
Limited, Houston, Texas. 83 pp.
. 1975. Cased Hole Applications. Schlumberger
Houston, Texas. 124 pp.
. 1974. Log Interpretation. Volume II-Applications.
Schlumberger Limited, Houston, Texas. 116 pp.
. 1970. Log Interpretation Principles - Volume I -
Principles. Schlumberger Limited, Houston, Texas. 110 pp»
. 1969. Production Log Interpretation. Schlumberger
Limited, Houston, Texas. 115 pp.
158
-------
Siple, G. E. 1964. Geohydrology of Storage of Radioactive
Waste in Crystalline Rocks at the AEC Savannah River Plant,
South Carolina. In: Geological Survey Research, 1964.
USGS Professional Paper 501-C. pp. C180-C184.
Talbot, J. S. 1972. Requirements for Monitoring of
Industrial Deep Well Disposal Systems. In: Underground
Waste Management and Environmental Implications. T. D.
Cook, cd. American Association of Petroleum Geologists
Memoir 18. pp. 85-92.
U.S. EPA. 1977. An Introduction to the Technology of Subsurface
Wastewater Injection. D. L. Warner and J. H. Lehr,
authors. EPA-600/2-77-240. 355 pp.
1975. Monitoring Disposal Well Systems. D. L.
Warner, author. EPA 680/4-74-008. 109 pp.
West, F. G., P. R. Kintziner and A. W. Laughlin. 1975.
Geophysical Logging in Los Alamos Scientific Laboratory
Geothermal Test Hole No. 2. LASL Scientific Report.
10 pp.
Wyllie, M. R. J. 1963. The Fundamentals of Well Log
Interpretation. Academic Press, New York. 238 pp.
Zemanek, J., E. E. Glen, L. J. Norton, and R. L. Caldwell.
1970. Formation Evaluation by Inspection with the Borehole
Televiewer. Geophysics. Vol. 35, No. 2. pp. 254-269.
159
-------
SECTION 5
GEOTHERMAL INJECTION TECHNOLOGY
Injection technology is an interdisciplinary field involving
geology/ hydrology, reservoir engineering, chemistry, materials
science, mechanical engineering, well drilling, and completion
technology. The state-of-the-art is a result of field experience
as well as research and development in each of these areas of
expertise. Although there is interdisciplinary cooperation,
most studies for improving the technology concentrate on one
area. However, in designing an injection system, it is important
to combine applicable knowledge and experience from all areas.
Other methods of geothermal waste disposal have been
conceived and some have been tried. Examples include: surface
ponding at Cerro Prieto, Mexico; discharge to surface waters at
Wairakei, New Zealand; ocean disposal in El Salvador. In most
cases the quantity and chemical composition of the waste fluid
make these alternatives environmentally unacceptable in the
United States without prior treatment to remove the pollutants.
The cost of such treatment may often be prohibitive.
5.1 DEVELOPMENT OF INJECTION TECHNOLOGY
Injection is a common secondary recovery technique in the
petroleum industry. It is also used to dispose of industrial,
domestic, municipal and nuclear waste. About 90,000 secondary
injection wells and 30,000 disposal wells are believed to exist
in the U.S. (National Water Well Association, 1978). Much of
this nongeothermal injection technology can be applied to the
geothermal industry including pretreatment methods, delivery
system design, well design, control of chemical problems,
redevelopment techniques, monitoring methods, and reservoir
models. The petroleum and industrial waste disposal industries
have provided the most significant contributions to geothermal
injection technology. These are summarized below.
5.1.1 Injection in the Petroleum Industry
For the last 50 or 60 years, oil field brines have been
injected into subsurface formations and today petroleum waste
disposal is almost exclusively by injection. Brine injection
wells are deep to avoid contaminating fresh water aquifers, and
to increase the gravity head thereby decreasing pumping costs.
160
-------
The waste may be injected into formations above or below
the oil zone. Many fields inject into the oil-producing forma-
tion to displace the crude oil and increase recovery and economic
gain. This secondary recovery method is called water flooding
and the waste is generally supplemented with brine from forma-
tions other than the producing zone. Water flooding may increase
oil recovery by 20 to 50% (Sanyal, pers. comm., 1978).
Injection volumes vary from field to field. In the largest
known oil field in the U.S., the East Texas field, about 80,000
cu m/day (500,000 bbl/day) was being injected into 84 wells 'in
1972 (McWilliams, 1972). Injection volumes for other fields
or wells are not readily available.
The chemistry of oil field brines is highly variable.
Concentrations of total dissolved solids range from less than
100 to more than 100,000 ppm. Sodium or calcium chlorides are
the major constituents of most brines, but magnesium, bicarbonate
and sulfate predominate in some. Many other minor constituents
and trace elements can be present, complicating the chemical
properties of the fluids.
The most common chemical problems in petroleum field injec-
tion are precipitation, scale deposition, corrosion, and forma-
tion plugging due to chemical precipitation and/or bacterial
growth. Techniques to control these problems include coagulation,
chemical precipitation and corrosion control; pH fixation;
chemical removal of silica, iron and manganese; mechanical
sedimentation; filtration; aeration; addition of bactericide;
and well stimulation. Reports by Collins (1975), Donaldson, et
al. (1972) and Taylor, et al. (1939) describe some of these
Methods.
5.1.2 Industrial Waste Injection
Because of the increased attention paid to surface water
Pollution control since the early 1960s, injection of industrial
^aste has become widely used. Prior to 1960, only 22 industrial
waste disposal wells had been constructed. As of 1974, the
total was 322, of which 209 were operating.
Of the wells constructed by 1973, 94% were in sand, sand-
stone and/or carbonate rocks. About 20% utilized only gravity
for injection pressure; the rest were assisted by pumping
wellhead pressures up to 10.3 MPa (1500 psi). Most
ndustrial injection wells are between 300 and 1800 m (1000 and
6000 ft) deep.
161
-------
As of 1975, a total of 114,000 cu m (30 million gal) of
industrial waste was being injected per day into disposal wells.
This is about 0.08% of the total industrial liquid waste dis-
charge of the U.S. (Reeder, et al. 1977).
5.2 CHARACTERISTICS OF GEOTHERMAL INJECTION
Subsurface injection of geothermal effluent is a special
case of deep well injection. Unique characteristics of geo-
thermal effluent injection are the chemistry, temperature and
quantity of the fluid being injected, the injection depth,
reservoir rock type and the flow dynamics of the system. These
five characteristics are discussed below.
5.2.1 Chemical Characteristics
Two aspects of the chemical composition of the fluid are
particularly meaningful to the subject of injection. First, the
high salinity and other trace minerals may degrade the fluid in
usable aquifers, and second, certain constituents may contribute
to scaling or corrosion of the injection well, or plugging of
the receiving formation. Substances found in geothermal fluids
that tend to form scale are carbonate, silica, calcium, metals
and sulfate. Corrosive agents found in geothermal fluids are
chloride, oxygen, and the acidic nature of some fluids.
A third aspect, not quite as important as the first two,
is the variation of the specific gravity of the fluid with TDS
(Fig. 5.1). Since the composition of the fluid varies within
the reservoir and the well and with time, these density dif-
ferences may induce a density current component on the flow
system.
Knowledge of the chemical composition of geothermal fluids
can provide much useful information about the reservoir, since
the kinds and amounts of constituents depend on the reservoir
environment: formation lithology, rock-water interaction, rock-
mineral-chemical equilibria, pressure and temperature. The
geographical variation in chemical characteristics may be
attributed mainly to variation in the nature of the subsurface
rocks, temperature, and distance from the source of recharge.
Temporal variation in the chemistry of geothermal fluids at a
particular site can have a number of causes, the most important
being the variation in the rate of fluid recharge (natural or
artificial) into the reservoir.
5.2.2 Temperature
Geothermal fluid temperature in deep well disposal involves
the temperatures of two distinct fluids, the injected fluid and
the resident reservoir fluid. Depending on the power conversion
162
-------
1.15
o
o
i
1
O
1.10
1.05
1.00
I I I—L
i I i i
50.0OO 100,000 150.0OO
TOTAL SOLIDS (PPM)
2OO.OOO 250tOOO
Figure 5.1 Specific gravity of sodium chloride formation
waters versus total solids in ppm (Warner
1975,, from Pirson, 1963. D. 39)
TEMPERATURE" r°C)
49 71 93 116
38
160
160 20O 240
TEMPERATURE ,»F
260
320
Figure 5.2 Specific gravity of distilled water as a
:unction of temperature (Warner and Lehr,
1917, from Pirson, 1963, p. 39).
163
-------
and disposal well facilities either of these fluids could be
hotter or colder than the other. For example, if fluid is
cooled before injection and it is injected into a hot part of
the reservoir, it will be cooler than the fluid or the
receiving formation. If the fluid is not cooled appreciably and
it is injected into a cooler part of the reservoir the injectate
may be hotter than the resident formation fluid.
Temperature affects the rate of chemical reactions, the
solubility of materials, the fluid density and volume, and the
viscosity of the injected fluid in the well bore, casing, and
receiving formation. Generally higher temperatures imply faster
chemical reactions, increased solubility (except for calcium
carbonate, which is less soluble at higher temperatures),
decreased density, increased volume and decreased viscosity.
The temperature contrast between the receiving formation
and the injected fluid is important. Depending on the relative
temperatures and chemistries, the injected fluid may form
precipitates or may dissolve constituents from the formation or
formation fluid. These chemical reactions must be critically
evaluated in planning the disposal well system. In addition,
the temperature contrast will generate density currents which
will affect the flow patterns. Fig. 5.2 shows the change of
specific gravity with temperature in distilled water.
Changes in viscosity will affect the hydraulic conductivity
of the receiving formation. Higher temperatures will lower the
viscosity thereby increasing the hydraulic conductivity; lower
temperatures will have the opposite effect. Fig. 5.3 shows water
viscosity as a function of temperature and salinity. A 100°C
(212°F) temperature differential can change the hydraulic
conductivity by 300% or more. This will have a significant
effect on the receptivity of the aquifer.
5.2.3 Quantity of Fluid
For most geothermal operations the quantity of fluid
processed and injected will be enormous. For example, in opera'
tion of a 10 MW plant in the Otake, Japan fields 1,230,000 kg
(2,700,000 Ib) of fluid per hour are injected into three wells
(Kubota and Aosaki, 1976) at rates of 4.69, 6.06 and 9.08 cu m/
min (1,240, 1,600 and 2,400 gpm) in each well. At the steam-
dominated Geysers field, California, injection into six wells
at an average rate of 2.1 cu m/min (550 gpm) per well is taking
place in the production of 502 MWe (Chasteen, 1976, p. 1335).
Large commercial levels of production in a hydrothermal resource
would probably require significantly greater injection.
For example, in a feasibility study for a 25 to 50 MWe geotherma1
power plant, a hypothetical production scheme outlines utilizing
an array of 36 wells producing 0.063 cu m/sec (1,000 gpm) and 18
164
-------
RESERVOIR TEMPERATURE (°C)
66 93 121 149
68 100 150 200 250 300
RESERVOIR TEMPERATURE (»F)
350
Figure 5.3
Water viscosity as a function of
temperature and salinity (equivalent
ppm Nad) {Warner & Lehr, 1977, from
Pirson, 1963, p. 40)
165
-------
wells injecting 0.13 cu m/sec (2,000 gpm) each. (Meidav and
Sanyal, 1976).
For comparison, the average injection rate for 209 operating
wells disposing hazardous wastes in the U.S. was 0.063 cu m/sec
(100 gpm) although in some cases injection rates to 0.12 cu m/sec
(1,900 gpm) have been noted (Reeder, et al. 1977, p. 106 and
1050). Another study (Warner, 1972) showed that in 1967 one
percent of all industrial injection wells injected at a rate
greater than 0.05 cu m/sec (800 gpm). The petroleum industry has
commonly injected brines into petroleum reservoirs for secondary
recovery of petroleum. For example, in the Hastings oil field,
Brazoria and Galveston Counties, Texas, 25 wells are injecting
brine at an average rate of 0.004 cu m/sec (64 gpm) per well for
a total injection into the field of 7,900 to 9,500 cu m/day
(2.1 million to 2.5 million gal/day). In the East Texas oil
field, Upshun, Gregg, Rusk and Smith Counties, Texas, 84 wells
are injecting at an average rate of 0.011 cu m/sec (174 gpm) per
well (McWilliams, 1972). The total injection into the field is
80,000 cu m/day (21,000,000 gal/day). The quantities to be
injected from full-scale hydrothermal power production will be
significantly greater. These quantities of fluid production and
injection will stress the existing ground water regimes to a much
greater extent than most existing injection well systems.
5.2.4 Depth
Injection will generally take place within the geothermal
reservoir or at an equivalent depth. Although the tops of some
reservoirs may be as shallow as 0.3 km (1,000 ft), as in
Steamboat Springs, Nevada, most reservoirs upper boundaries
are deeper than 1 km (3,000 ft) (Renner, et al. 1975). However,
the top of the reservoir would establish only the upper limit
of injection depth and in most cases geothermal injection wells
would be deeper than 1 km (3,000 ft). Juul-Dam and Dunlap (1976)
report a mean depth to the reservoir of slightly over 1,520 m
(5,000 ft) and a median depth under 1,830 m (6,000 ft) (Fig. 5.4)-
Ideally, injection will take place below one or several
impermeable layers that will confine the fluid to the injection
depths and prevent it from migrating upward. In sedimentary
reservoirs, such as in the Imperial Valley, these confining
consist of impermeable clays and shales, perhaps with the self"
sealing features of silica or carbonate deposition forming a
reservoir caprock. In fractured igneous reservoirs this con-
fining layer would most probably consist of silica caprock
deposited by the self-sealing mechanism of the geothermal system*
166
-------
Mode
Median
o
OJ
3
CT
0>
5%
\
^
2,000
(610)
y
> i
4,000 6,000
(1220) 0629)
\
\^
i i
8,000
$439)
iir
i -^
I i *
7o
10.000 12,000
(36-19) (3659)
Geothermal Reservoir Depth, ft (m)
Figure 5.4
Geotherraal reservoir depth distribution.
CJuul-Dam and Dunlap, 1976)
-------
5.2.5 Flow
The flow dynamics of a geothermal injection system will
be different than those of most deep well disposal systems.
This difference is manifested by the combination of fluid
injection and extraction in the geothermal reservoir. In most
disposal well systems fluid is merely injected and left to
migrate solely under the pressure head of the injected fluid.
The pumping of the geothermal production wells will provide
additional stimulation to the flow network.
Injection to stimulate petroleum production — as opposed to
injection for waste disposal — is similar to the geothermal waste
injection. Also similar to geothermal extraction- injection are
the ground water barriers formed by doublet injection-extraction
well pairs. An example of this process has been described in a
study done by Stanford University for the Santa Clara Water
District CSheahan, 1977) .
One must acknowledge that extraction and injection of fluid
for geothermal energy production will superimpose additional
forces on an existing dynamic ground water flow system. Although
hydraulic conductivity between the geothermal reservoir and
overlying ground water aquifers will be poor, the additional
forces of production and injection may have some effect on the
gradients of overlying aquifers or may cause fluid to escape
from a previously confined reservoir. In planning an injection
well system, these possible effects must be recognized.
Mathematical models may help establish the effects of different
combinations of injection and production rates at different
locations and depths on the overall ground water flow.
5.2.6 Reservoir Rock Type
Most hazardous waste injection wells in the United States
are completed in sedimentary strata, largely sands, sandstones
and carbonates (Warner and Lehr , 1977, p. 6). The numerous
brine injection wells used in the petroleum industry are also
completed in sedimentary strata, mostly with intergranular
porosity. Because geothermal reservoirs tend to occur in
igneous or metamorphic rock masses with fracture permeability,
the flow mechanism and well performance characteristics for
geothermal injection and production wells differ from those for
most nongeothermal wells. The extent of these differences will
depend on the type and degree of fracturing in the aquifer. A
rock mass with a dense network of small open fractures may behave
like one with intergranular permeability. On the other hand, an
aquifer which carries the greatest part of its water through a
few major fractures will have quite different characteristics.
168
-------
5.3 GEOLOGIC, HYDROLOGIC AND RESERVOIR EVALUATION
The physical aspects of the geothermal reservoir and its
geologic and hydrologic surroundings must be known in order to
conduct a safe and effective geothermal injection program.
The necessary geologic and hydrologic data would include: 1)
descriptions of the geologic structure and stratigraphy (including
maps and cross sections), 2) locations, thicknesses, depths and
areal extents of aquifers, 3) transmissivities of aquifers, 4)
water table depths and elevations, 5) piezometric surfaces, 6)
recharge and discharge areas, and 7) rate and direction of
natural ground water flow. Details of these data and their
collection are outlined in the first two steps of the monitoring
methodology (Sections 3.1 and 3.2).
Reservoir development data must be superimposed on the
existing flow regime. Some hypothetical examples of this are
outlined in Section 2.4 on geothermal reservoir engineering.
Since geothermal fluid is injected not only as a means of
waste disposal but also to maintain reservoir pressure and reduce
subsidence potential, the fluid may be injected into the produc-
tion zone. Hydraulic confinement to this interval must be
maintained to prevent upward migration of the fluid and possible
contamination of overlying aquifers. In sedimentary reservoirs,
such as in the Imperial Valley, the confining layers are
impervious clays and shales. In fractured igneous reservoirs,
the confining layer may be formed by the self-sealing mechanism
of geothermal systems. This mechanism involves deposition of a
silica caprock at a depth where the temperature and pressure of
the rising silica-saturated geothermal fluid are low enough for
the silica to precipitate in the fractures and seal off the
reservoir (Facca, 1973). In some systems the mechanism may be
somewhat different and involve deposition of carbonate or other
Minerals.
5.4 CHEMICAL PROBLEMS
Certain chemical characteristics of geothermal production-
injection systems may create problems such as scale deposition
in the injection lines and well bore, plugging of the formation
around the well, and corrosion of pipes in the system. High
temperatures and pressures, as well as injecting water mixed
from different production, wells, add to the complexity of the
chemical problems. These problems and their solutions are
Discussed in this section.
169
-------
5.4.1 Scaling
Injectivity can be severely decreased by scale and
precipitates which accumulate in the well bore, in and around
the slotted liner, and in the injection formation. Plant process
conditions, such as temperature changes, pressure reduction,
the type of materials in contact with the fluid, and the contact
duration, affect scale deposition rates. The reactivity of
different surfaces may vary with flow rate and temperature.
Hence, different types and thicknesses of scale will result from
flow over surfaces of different shapes and materials (Wahl and
Yen, 1976). Table 5.1 lists some factors which affect scaling.
The following chemical mechanisms can deposit scale in
injection systems:
1) polymerization and precipitation of silica and
silicates;
2) precipitation of insoluble carbonates, sulfates
and hydroxides of alkaline earths;
3) precipitation of heavy metal sulfides; and
4) precipitation of redox products.
Deposition of silica and calcium carbonate (calcite) are the
most common scaling problems. Scale can be controlled by
chemical or physical methods. Table 5.2 presents some scale
prevention techniques and Table 5.3 presents scale removal
methods. Silica and calcite scale are discussed below.
Silica Scale—
Since nearly all geothermal waters contain appreciable
quantities of silica, the complexity of silicate chemistry has
presented great problems in production technology as well as
waste control technology. Neither the mechanism nor the
kinetics of silica polymerization are adequately understood
(Axtmann, 1976).
The silica concentration of extracted geothermal fluids is
equal to the solubility of quartz at the reservoir temperature.
Therefore, the silica concentration may greatly exceed the
solubility of amorphous silica (become supersaturated) at the
lower injection temperatures leading to silica precipitation
(silication). At Ahuachapan, for example, the reservoir
temperature is 235 to 245°C (455 to 473°F) and the fluid cools
to 98°C (208°F) at the surface. This temperature drop to the
local boiling point results in silica supersaturation. Resultaflt
precipitation and scaling are serious problems (Cuellar, 1976).
170
-------
TABLE 5.1 FACTORS AFFECTING SCALING IN GEOTHERMAL PLANTS
(Modified from Phillips, et al. 1976)
Factor
Fluid phase (steam or water)
Brine composition
Temperature and temperature changes
Pressure changes, including partial
pressure change in CO2/ ^S,
Velocity and turbulence
Residence time in each part of plant
Surface effects and surface to volume
ratio effects
Geometry of plant components
Effect on Scaling
More scale will be deposited from water
than from steam; flashing can induce
scale deposition.
Determines the type of scale.
Variable effects depending on chemistry;
lowering temperature induces silica
scaling, rising temperature induces
calcium carbonate scaling.
Causes change in solubility of materials.
Scale will tend to accumulate at places
where fluid velocity changes; e.g.
elbows, constrictions, valves.
Longer residence time tends to increase
degree of scaling.
The higher the surface to volume ratio/
the less scaling per unit area; scale
will tend to accumulate on irregulari-
ties of the surfaces.
Controls location of scaling, see
"velocity and turbulence" comment
above.
171
-------
TABLE 5.2 TYPICAL PREINJECTION TREATMENT TECHNIQUES TO CONTROL
SCALE FORMATION (Modified from Phillips, et at. 1976)
Scale Type
Silica and calcite
Silica
Silica
Silica and arsenic
Silica
Mixture of metal
sulfides, carbonates
and silicates
Control Technique
pH adjustment (acid
injection)
Injection of base (NH3
or NaOH)
Dilution of the unflashed
geothermal fluid
Sedimentation and co-
agulation (addition of
slaked lime, hydro-
chlorite, and flocculant)
Plain sedimentation;
retention tank
Application of electrical
potential
Location
Magmamax No. 1 well,
Niland, California
Sinclair wells, California
Namafjall, Iceland
Wai rake? and Broad lands,
New Zealand
Otake, Japan, and
Ahuachapan, El Salvador
Sinclair Well No. 4,
California
172
-------
TABLE 5.3 TYPICAL SCALE REMOVAL TECHNIQUES APPLICABLE TO
INJECTION SYSTEMS (Modified from Phillips, et ol. 1976)
Scale Type
'n
borehole
Calcite in bore
Removal Technique
Pump inhibited HC1 into
the well
Reaming or redrilling
Calcite in well casings Wash with inhibited HCI
Silica in flow control
equipment and heat
exchangers
Silica in borehole
Mixed scales in injection
and brine drain lines
Wash with ammonium
bi fluoride
Pump NaOH solution
into the well
Cavitation descaling
Location
East Mesa Well 5-1 and
Otake, Japan
New Zealand, Hungary,
and Mexico
Hungary and KWerau,
New Zealand
Hveragerdi, Iceland
Matsukawa, Japan
Nil and Geothermal Test
Facility, California
173
-------
Polymerization is a process where large molecules are formed
by the joining together of small molecules, e.g. formation of
Si02 chains. Polymerization of silica apparently does not occur
under extreme acidic or alkaline conditions. Dissolved monomeric
silica is relatively stable between pH 2 and 3. Maximum
polymerization has been reported at various pH's between 5.85
and 10. Silica becomes stable against polymerization at very
high pH's, presumably due to the increased solubility of the
silicate ion at these pH's (Axtmann, 1976). Increasing tempera-
ture and decreasing silica concentration lengthen the
polymerization induction period of silica. The polymerization
rate increases with temperature up to about 50°C (122°F) , then
decreases, perhaps due to mass transfer rates (Axtmann, 1976).
Various ions and the pH of the solution can inhibit or catalyze
silica polymerization (Her, 1973).
Chemical control of silica scale—Chemical methods of
silica scale control can be categorized into three types: 1)
prevention or postponing of scaling by inhibiting polymerization;
2) deliberate precipitation of silica prior to injection; 3)
chemical removal of the scale once it forms in the injection
system.
Silica polymerization can be inhibited by:
1. keeping the temperature high enough to prevent
silica saturation;
2. reducing turbulence, thereby avoiding increments
in the velocity gradients and collision of particles
which may increase pH;
3. lowering the pH of the solution below 6.5,
which substantially decreases polymerization.
Silica-laden discharge waters have been successfully treated
with slaked lime to precipitate silica as well as arsenic. This
process has been tested at the Wairakei and Broadlands fields in
New Zealand. Discharge water is contained in holding ponds to
allow silica polymerization, then slaked lime is added. Hydrated
calcium silicate gel precipitates and is separated in settling
tanks and dried. The calcium silicate precipitate can be used
in wallboards, insulants and perhaps in cement and ceramics. By
controlling lime addition rates the calcium silicate can be made
silica-rich or calcium-rich. (Rothbaum and Anderton, 1976). The
discharge water depleted in silica can then be disposed of without
silica deposition.
The water from the Otake geothermal field in Japan is ponded
for about one hour, while formation of nonadhesive colloidal
silica takes place. Once polymerization ceases the water can
174
-------
be handled without a serious silica scaling problem, (Yanagase,
et al. 1970)
Application of a negative electrical potential has been
found to reduce silica scaling for the Salton Sea geothermal
brines (Phillips, et al. 1976).
While prevention of scaling can be achieved by proper
treatment of the waste, it is also possible to chemically remove
the scale once it is formed. Though insoluble in water and
acid, silica turns to soluble silicate when treated with alkali
or carbonate. In a production well plugged with silica scale at
Matsukawa, Japan, 125 kg of NaOH dissolved in 300 liters of
water was placed in the wellhead for 8 minutes, the wellhead was
flushed with pure water for 15 minutes, and the process was
repeated. Scale removal was complete. Siliceous scale (90%
silica) has been successfully removed by treatment with a
sodium hydroxide (NaOH) solution under high pressure and
temperature. Use of NaOH has the disadvantage of precipitating
metal hydroxides which can cause formation plugging (Ozawa and
Fujii, 1970).
Of the scale prevention measures discussed above, deliberate
precipitation and removal of silica prior to injection is probably
the most certain means of control, as it decreases the poorly
understood problem of silica plugging in the injected forma-
tion. Preventing precipitation under surface conditions may
only delay deposition.
Calcite Scale—
Solubility of calcite is a function of CQ2 gas partial
pressure, temperature, pH and other species in solution.
Calcite precipitation may be controlled by controlling one or
more of these factors. A decrease in pressure as geothermal
fluid flows from the wellhead releases CO2 which in turn favors
formation of calcite. Monitoring a CO2 backpressure could
prevent this precipitation; however, the backpressure would also
decrease the flow rate.
Various scale inhibitors (polyelectrolytes, ester of
phosphoric acid, phosphorates, etc.) have been used to slow
down the precipitation rate of calcium carbonate. A glassy
phosphate called Calgon has been used to prevent scaling as
well as to control corrosion.
To remove carbonate scale, acid solutions can be used.
Calcite (CaCOa) dissolves in hydrochloric acid according to
the following reaction:
CaCO + 2HC1 -> Ca++ + H 0 + CO t + 2Cl~
3 22
175
-------
Acidizing has been used to remove scale from well casings
in the Kawerau Field, New Zealand; Otake, Japan; and East Mesa,
California. Caution must be used in dissolving calcite with
acid since highly acid solutions may create ^predicted
problematic chemical reactions when they enter the injected zone.
Also, as the injected fluid moves into the receiving formation,
dilution of the acid will decrease the solubility of CaCO3,
possibly resulting in formation plugging from CaCO^ deposition.
Physical Removal of Scale—
Scale, particularly carbonate, can be removed by physical
methods such as scraping or scratching the deposits from the
inside of well walls, casings and pipelines. Water jets are
also a potential method for removing scale.
In geothermal wells in New Zealand, Hungary and Mexico,
calcite has been removed with scratchers or reamers. The rotary
transverse motion of a reamer in the hole will remove deposits.
This method is fairly expensive as it requires a drill rig and
may shut the plant down for several days. Since only the bore
is cleaned, deposits in casing perforations and in the formation
remain, and cuttings from the reaming may get forced into
casing perforations and block flow.
Scrapers can remove scale from pipelines. There are
various types of scrapers—steel balls, chained rubber balls,
plugs and wire brushes, "go-devils" and spiral-brush "pigs".
They are inserted at inlet traps in a system and removed at
outlet traps, scouring the pipe between. A problem with
scrapers is that they can damage plastic lining of the pipes
(Phillips, et al. 1976).
In cavitation descaling, pulsating high-pressure water jets
are directed against scaled surfaces. The alternating pressures
cause bubbles to form and collapse on the scaled surface.
Collapse, or implosion, of these bubbles creates shock pressures
up to several hundred atmospheres, which break the scale from
the metal surface. At the Niland geothermal test facility,
cavitation successfully removed scale from about 366 m (1,200 ft)
of injection line.
5.4.2 Formation Plugging
When the injected fluid is mixed with formation water and
reacts with minerals, precipitation can occur, plugging the
formation.
176
-------
Reaction Between Injected Fluid and Receiving Formation—
The reactions between injected acidic industrial wastes and
receiving waters and minerals have been studied (Grubbs, 1972).
The zone of reaction around geothermal injection wells is
probably similar to that around industrial waste disposal wells;
however, in geothermal projects, the injected fluid and forma-
tion fluid may differ much less chemically. A gradational
boundary (zone of reaction) develops between the injected and
the formation fluids. In this zone, some minerals will be
dissolved or altered, metastable sols and gels will form and
new compounds will precipitate. The extent of this zone will
depend upon the length of time the fluid has been injected and
its flow rate, the mixing rate of the injected fluid and forma-
tion water, and chemical reaction rates. Although chemical
reactions may be predicted by laboratory mixing of the two
fluids, the physical effect on the system is difficult to
predict. A precipitate, for example, may cause immediate
plugging of the formation, be flushed through the receiving
formation without plugging it (especially if fracture
permeability exists), or may slowly accumulate and gradually
retard injection.
Silicates and carbonates are the most common cause of
plugging in geothermal systems. Their deposition is by the same
mechanisms as described for scaling, except that they are
deposited in the voids of the formation rather than in the well
casing.
Formation plugging can also be caused by swelling clays due
to water formation incompatibility. This problem is restricted
to sedimentary reservoirs.
Bacterial growth, which causes plugging around some nongeo-
thermal injection wells should not be a problem in most geothermal
injection wells. Few bacteria thrive in the high geothermal
temperatures. Cool injected fluid from open systems might have
bacterial contamination which could cause plugging.
If the fluid is not filtered prior to injection, particulate
and colloidal matter can plug the formation.
Control of Plugging—
Plugging of the formation can be minimized by settling,
filtering, and removing the solids and/or by adding inhibitors
to prevent precipitation of the solids. Even then, downhole
Plugging is possible. Well stimulation techniques include
acidizing and hydraulic fracturing. Shock treatment has also
been used with some success. It involves subjecting the forma-
tion to an almost instantaneous applied pressure differential
177
-------
(implosion) and sustaining this pressure for a period, thereby
loosening the obstructing material.
5.4.3 Corrosion
The complex chemistry of geothermal fluids impedes injection
technology development. Each field, sometimes each well, has its
own combination of corrosion mechanisms which must be defined
and controlled. Experience with seawater corrosion provides a
base for studying materials behavior in the similar, though
generally much more corrosive, geothermal fluid. Marine
experience includes studies of electrochemical potential of
various metals, as well as the effects of dissolved oxygen,
temperature and fluid velocities on corrosion mechanisms (La Que/
1975).
In geothermal fluids, the chloride ion (Cl~) is generally
considered the main corrosive, with hydrogen sulfide (H2S)
probably next in importance. Brine chemistry, pH, dissolved
gases (such as oxygen, carbon dioxide, and ammonia), temperature*
flow rate, well depth and pressure all affect corrosion, either
individually or in combination. The types of metals used also
influence the types and degree of corrosion in the system. Table
5.4 lists some of these factors.
Dissolved oxygen is important to corrosion. The ionization
of oxygen,
O + 2H 0 + 4e~ -*• 40H~
is believed to be the principal cathodic process in the electro-
chemical action of geothermal fluids. It is the main cause of
dissolution of metals such as steel and copper. On the other
hand, oxygen is required to produce the oxide films which make
many metals passive. Titanium, aluminum and stainless steel are
corrosion-resistant only after the surface is oxidized. Other
aspects of oxygen's role in corrosion remain unresolved.
The most common ways in which metals used in geothermal
installations corrode are (1) uniform attack or general corrosiofl'
(2) pitting and crevice corrosion, and (3) stress corrosion
cracking. Corrosion fatigue and erosion cavitation are also
observed in pumps. These types of corrosion are explained below*
Uniform attack or general corrosion refers to a regular,
nondiscriminating chemical reaction on an entire surface.
Examples of this type include rusting and other forms of oxida-
tion.
Pitting and crevice attack can affect materials that are
resistant to general corrosion. These local phenomena are
178
-------
TABLE 5.4 FACTORS AFFECTING MATERIAL CORROSION IN GEOTHERMAL
PLANTS (Modified from Phillips, et al. 1976)
Factor
Fluid phase (steam or water)
Brine composition
pH of fluid
Temperature & temperature changes
Partial pressures of CO2/ r^S,
NH3/ H2
Atmospheric O2 leakage into plant
system (e.g., piping, condenser,
heat exchanger)
Power plant material
Stress levels in materials and
especially cyclic stresses
Crevices
Presence of scale deposits
Velocity of fluid
Suspended solids content
Galvanic coupling of dissimilar
metals
Effeet on Corrosion
Affects the concentration of corrosive
constituents.
Affects the type & rate of corrosion;
more concentrated brines tend to be
more corrosive.
Extreme pH generally increases corrosion.
Higher temperatures generally accelerate
corrosion.
Increased gas pressure can increase
corrosion.
Increased oxygen increases corrosion.
(Should be selected to be resistant to
the fluid at the site).
High and/or repeated stress increases
corrosion.
Increases corrosion.
Can protect metal; also may create site
for local corrosion.
Increased velocity generally decreases
corrosion.
Can abrade surface and increase corrosion.
Increases corrosion.
179
-------
complex processes that depend on the physical, chemical and
metallurgical conditions at the solid-liquid interface. For
this reason, the use of high alloy metals is necessary with
corrosive geothermal fluids.
For stress corrosion cracking to occur, certain conditions
must be present. The material must be in an electrolyte con-
taining a certain amount of a cathodic depolarizer (oxygen, for
example); the material must be under stress; and there must be
enough contact time for an electromechanical action. Resulting
crack propagation may be intergranular or transgranular, and may
take minutes or years.
Sulfide cracking is a special case of stress corrosion
cracking that can develop in geothermal injection facilities.
This process is temperature and pH sensitive, and depends on
the interaction between chloride and sulfide in solution.
Because of the wide variation in chloride and sulfide concentra-
tions in geothermal fluids, the sulfide cracking tendency of
materials must be studied for each site.
Corrosion fatigue is related to stress corrosion cracking.
This condition results from stress alteration (tension to
compression) which reduces the service life of the material.
Dissolved oxygen and the presence of H^S in a salt solution
lower the fatigue limits for some materials. Purnps are the
primary target for corrosion fatigue in geothermal injection
systems.
Erosion and cavitation in corrosive media are not completely
understood. Erosion is created by abrupt momentum changes of
the fluid or by impact of particulate matter in the fluid.
Cavitation is associated with local hydrodynamic pressure
reductions which cause bubble formation and collapse against a
surface. Testing of full-scale equipment with the geothermal
fluid in question is essential to determine if the equipment will
resist erosion and cavitation.
Control of Corrosion—
In general, corrosion can be prevented or reduced in many
ways, such as the following:
1) Changing the physical, chemical, or dynamic
properties of the medium. Lowering the temperature
usually causes a pronounced decrease in corrosion
rate. Fluid flow generally increases corrosion,
although there are exceptions; stainless steel, for
example, has better corrosion resistance to a flow-
ing medium than to a stagnant solution. Very high
velocities should always be avoided, however,
180
-------
because of erosion-corrosion effects. Removing
oxygen by using an oxidizer is an old corrosion-
control technique. In modern practice, oxygen
removal can be accomplished by vacuum treatments,
inert gas sparging, or addition of oxygen
scavengers. Decreasing the corrosive concentra-
tion is also usually effective. Inhibitors may
be used. They can be considered as retarding
catalysts and can be classified according to
mechanism of operation and composition: a)
adsorption-type inhibitors (organic compounds
which adsorb onto the metal surface); b)
hydrogen-evolution poisons which retard the
hydrogen evolution reaction; c) scavengers such
as sodium sulfite and hydrazine which remove
dissolved oxygen from aqueous solutions; d)
oxidizers; and e) vapor-phase inhibitors.
2) Selection of proper material. Susceptibility
to electrochemical attack and other types of
corrosion must be evaluated. Use of alloys
does not stop the corrosion process. When an alloy
or other noble metal is used, some other less noble
metal almost always is coupled to it and sacrificed
to corrosion (cathodic protection). The use of
stainless steel with chrome content higher than
10% is generally recommended. in any case, mate-
rials being considered should be tested with the
geothermal fluid under the physical conditions
anticipated during operation.
Protective coating. The protection any coating
affords -is directly proportional to its
continuity. The most common coatings are
cement, plastic, tar and epoxy, including Teflons
Controlled sodium silicate, calcium carbonate
scale, and glassy phosphates have also been
used.
An interesting case of natural corrosion protec-
tion exists at the Salton Sea geothermal field.
Corrosion is generally not a problem above
depths of 300 to 600 m (1,000 to 2,000 ft)
Above 300 m (1,000 ft), silica scale apparently
protects the casing from corrosion. In several
downhole tests in production wells, only minimal
corrosion was observed because of the develop-
ment of a so-called "hard scale," a friable
glassy (amorphous) material which formed as a
thin film on the production lines. The "hard
scale" was primarily a silica and iron oxide
mixture with some sulfide.
181
3)
-------
4) Neutralizing pH of fluid.
5) Reducing the salt content of liquid waste may
reduce corrosion in injection wells.
Desalination has been studied at East Mesa
Test Site, California. However, desalination
to inject clean water is likely to be impractical
in terms of cost and generation of solid waste
or more concentrated brine. Additionally,
altering the chemistry of the geothermal fluid
prior to injection may create incompatibilities
between the injected fluid, the formation and
its fluid (Harding-Lawson, in press).
5.4.4 Hydrogen Sulfide
Significant amounts of dissolved H2S are often present in
geothermal steam condensate. The H2S may lead to the creation
of heavy metal sulfides, scale and to corrosion. HUS can be
removed by direct oxidation methods, including the following
which are similar to those used to abate air pollution from
cooling tower emissions:
- Direct injection of S02 in the waters to oxidize
the H2S to SO by the Glaus reaction.
- Simultaneous injection of a catalyst such as
sulfur dioxide or ferric ion, oxygen and air, or
oxygen.
- Addition of a metal catalyst, such as iron,
to the waters to promote direct oxidation by
the oxygen dissolved in the circulating waters
in the cooling towers.
- Addition of Cataban, a chelated iron compound,
to catalyze the direct oxidation of H2S.
Methods used for gas ejector emissions would not be as effective
on condensate.
Oxidation in the cooling tower may result in the formation of
insoluble metal sulfates (such as CaSO4) and elemental sulfur,
leading to pipe corrosion and plugging. Filtration or dispersion
could be applied to remove the solids from the condensates prior
to injection (Phillips, et al. 1976).
Eliminating silica deposits may prevent sulfide deposition
by inhibiting nucleation and growth on substrates. Also, since
sulfide solubility increases with decreasing pH, acidification
may prevent sulfide deposition (Owen, 1977).
182
-------
5.5 PHYSICAL PROBLEMS
Physical problems in a production-injection system involve
potential land surface deformation and induced seismicity,
hydrofracturing of confining formations and introduction of
thermal stress. These problems and their solutions are discussed
in this section.
5.5.1 Land Surface Deformation
Injection of spent geothermal fluid into the geothermal
reservoir is generally recommended to minimize potential
subsidence by maintaining the fluid balance in the reservoir.
However, inhomogeneities in the subsurface materials will
generally lead to localized areas of over-pressured and under-
pressured rocks in the vicinity of the geothermal injection
and producing wells, respectively. Lateral differences in
hydrostatic pressure can lead to horizontal movements as well
as the more obvious vertical ones. Changes in subsurface
temperature associated with production and injection wells can
cause additional movements in subsurface materials due to
differential thermal expansion. However, these thermally
induced movements will generally oppose the movements caused by
fluid pressure changes (Harding-Lawson, in press).
Land Surface Monitoring and Modeling—
The first step in controlling land surface deformation is
determining accurately the location, magnitude and direction
of these movements. Monitoring these movements will allow them
to be correlated with the rates and locations of geothermal
production and injection, and the properties and inhomogeneities
of the reservoir and overlying formations. If disruptive ground
motion does occur, then these types of correlation, incorporated
in analytic or mathematical models, will aid in determining the
modifications of the production-injection scheme and possible
prevention of further disruption.
Existing subsidence monitoring networks can be divided into
three categories: 1) horizontal control networks, 2) vertical
control networks, and 3) other measurement programs. For example,
in Imperial Valley, California, networks of both regional and
local extent have been established. The vertical network consists
of first- and second-order level lines, allowing maximum vertical
errors of 4.0 mm and 8.4 mm, respectively, per kilometer of
distance surveyed. The regional horizontal network is capable
of accuracies of 0.1 mm per kilometer, while the local networks
are capable of accuracies of 2 ram per kilometer. In addition,
developers of geothermal wells in the Imperial Valley are
required by state and county ordinances to install several bench
marks near each well and to periodically resurvey and tie them
183
-------
into the first- or second-order level lines, in order to detect
local subsidence that may be related to geothermal production.
Tiltmeters and extensometers can aid in precisely defining
the mechanism of the ground motion associated with geothermal
subsidence. This data can be used to distinguish between
subsidence due to geothermal fluid withdrawal and injection,
and subsidence due to other mechanisms. Extensometers can be
installed at various depths in wells where shallow ground water
is being pumped close to geothermal developments; these can be
used to monitor changes in water levels and. compaction at dif-
ferent depths in boreholes located between the geothermal
reservoir development and nearby farmlands. Data obtained can
be used to differentiate between deep compaction caused by
geothermal production and shallow compaction due to shallow
ground water withdrawal.
Tiltmeters help determine whether the ground deforms as a
stressed beam, develops vertical shear planes, or perhaps deforms
with some combination of the two mechanisms.
Ground movements can also be estimated or predicted using
computer models based on the theoretical relationships between
the rate and duration of removal and injection and the hydrologic
physical properties of the reservoir and overlying rocks. These
calculations suffer both from inadequacies in the theory and from
uncertainties in parameters (elastic constants, porosity,
permeability, density, degree of homogeneity of the rock, and
distribution of pre-existing in situ stresses). Nevertheless,
the computer models are useful for calculating deformations over
a wide range of possible parameters and source-sink locations.
The more accurate the empirical measurements of these reservoir
parameters, the greater the confidence in the estimated
predictions. Ideally, an iterative combination of theoretical
model studies with empirical data from a specific geothermal area
will provide the best estimates of ground movement potential for
a specific array of production and reinjection wells.
Additional information on the relationship between ground
subsidence and the extraction of fluids for nongeothermal purposes
is provided in papers by Poland (1973) and Poland and Davis
(1969). Systems Control (1976) has published a comprehensive
study of "The Analysis of Subsidence Associated with Geothermal
Development". A report entitled "Geothermal Subsidence Research
Program Plan" has been prepared for the U.S. Department of
Energy as a means of providing a unified approach to guide future
investigations of subsidence (Lawrence Berkeley Laboratory, 197?)•
5.5.2 Induced Seismicity
Three relationships between seismic activity and geothermal
occurrences have been noted: 1) geothermal anomalies are
184
-------
concentrated in seismically active regions of the world; 2}
microearthquake activity has been correlated with geothermal
anomalies; and 3) fluid withdrawal, with or without injection,
may trigger local seismic activity.
Since most geothermal fields are in tectonically active
regions, the possibility of triggering earthquakes by geothermal
production and injection is of some concern. Existing injection
well data have yielded clues to the ability of fluid injection
to trigger earthquakes. Of the thousands of existing oil field
and waste injection wells, only two instances of earthquakes
triggered by fluid injection have been cited (Warner, 1977).
One of these occurred at the Rocky Mountain Arsenal waste
disposal well near Denver, Colorado, and another occurred at the
Rangely Oil Field in northwestern Colorado (Raleigh, et al.
1976) where the largest event registered was a magnitude 6
earthquake. These earthquakes were believed to be caused by an
increase in pore pressure that reduced the effective stress
across fracture surfaces and resulted in shear failure. At the
Rocky Mountain Arsenal chemical wastes were being injected under
high pressure into deep wells in relatively dry formations.
Seismic events up to magnitude 5 were recorded.
Because geothermal reservoirs are generally saturated, have
high permeability and natural convective patterns, production
and injection of fluids in these reservoirs are not analogous to
the conditions at Rocky Mountain Arsenal and Rangely. The
Geysers, California and Wairakei, New Zealand are both in areas
of high natural seismicity. High microseismicity has been
recorded since geothermal production and injection began at The
Geysers. It is not known if this is higher than the natural
microseismicity rate since there was no detailed pre-production
microseismic monitoring. At Wairakei, no association between
Production/injection and increasing seismic activity has been
noted. Injection into hot dry rock reservoirs might have a
greater potential for generating earthquakes (Ridley and Taylor,
1976) . Since the formation is not initially saturated, the
introduction of fluid will significantly lower the effective
stress in the rocks, thereby lowering the force necessary to
cause displacement in fractures.
Control of Induced Seismicity—
The possibility of triggering earthquakes can be minimized
not exceeding the original pore pressure of the fluids.
Reservoir pressure reduction resulting from production of
Isothermal fluid will generally be transmitted through the
Reservoir to injection well locations. Pressure increases due
^o injection will be greatest close to the injection well, but
these increases will be exceeded by the reservoir pressure
Deductions within a moderate distance. This can be a very
185
-------
delicate situation but, generally, injection-induced seismicity
may be a problem only if a fault is located within a higher
pressure area close to an injection well.
Avoiding changes in pore pressure in overlying fresh water
aquifers is another reason to isolate the injection well from
those aquifers. The injection well or any abandoned well near
an injection well may provide a pathway for upward migration of
liquid wastes into overlying aquifers (see Section 3.2.2,
Potential Pollutant Mechanisms and Pathways). Entry of liquid
could increase pore pressure, thereby increasing seismicity.
Withdrawal of geothermal fluids may alter deep ground water
flow patterns, and perhaps even affect flow rates in overlying
aquifers. The effect of these alterations on the tectonic
stress regime is unknown. Several years of continuous monitor-
ing will be required to understand the effects of fluid withdrawal
and injection.
Measures to mitigate induced seismicity would involve
careful regulation of fluid production and injection rates.
Although this mechanism is generally understood, its applica-
tion, which would include computer analyses of reservoir
properties, existing and induced stress fields, and ground water
flow pattern, remains speculative.
Two criteria can be considered useful in differentiating
induced earthquakes from naturally occurring ones: changes in
the frequency-magnitude (recurrence) statistic in the area of
the geothermal field; and changes in depth and location of
events from those occurring prior to production activity
(Phelps and Anspaugh, 1976). In order to apply these criteria/
studies are normally made both prior to development and during
utilization of a geothermal resource to:
1) determine the "baseline" level of naturally
occurring earthquakes, including location and
depth; and
2) monitor the actual magnitude and distribu-
tion of earthquakes during utilization of the
reservoir.
5.5.3 Hydrofracturing of Confining Formations
Hydraulic fracturing (hydrofracturing) is a technique used
in nongeothermal injection to increase formation injectivity.
It is done by injecting fluid at pressures higher than the
fracture pressure of the system, thereby causing the reservoir
rock to fracture (Sun, 1973; Warner and Lehr, 1977).
186
-------
In geothermal injection, hydrofracturing should be avoided
unless the fracture strength of the confining formations or cap-
rock is significantly greater than that of the reservoir rock.
Induced fracturing of the caprock or confining formations may
disrupt the natural hydrologic regime. This could be detrimental
to reservoir productivity by releasing reservoir pressure. Addi-
tionally, geothermal fluid escaping through the induced fractures
could degrade water in overlying aquifers.
In reservoirs where the rock strength cannot be measured,
the original hydrostatic pressure of the reservoir must be deter-
mined. The injection pressure should be carefully monitored to
avoid exceeding the original reservoir pressure. Where the
fracture pressure of the rock is known, it must not be exceeded
during injection.
5.5.4 Thermal Stress
In the special case of injection of cold water into a hot
dry rock formation for heat recovery, the resulting temperature
gradient in the rock needs to be considered. The differential
thermal stress fields created by the cold water may result in
more extensive fracturing than that desired for heat recovery.
Research is being done on this problem by the Futures Group
(1975).
5.6 INJECTION SYSTEM DESIGN
Reservoir engineering calculations and reservoir modeling
are integral preliminary phases of designing the physical
injection system. These preliminary studies will provide infor-
mation on the optimum production and injection zones and rates.
This will thereby influence decisions on optimum locations,
depths and capacities of wells and whether intervals are
perforated or open. Some example results of these types of
calculations are included in Section 2.4
After the reservoir engineering and modeling phase, four
Aspects remain in the physical design of a geothermal injection
system: type of pretreatment, fluid delivery system, well
design and monitoring of operation. These aspects are discussed
below.
5.6.1 Pretreatment
The type of pretreatment depends on the downhole behavior of
the liquid waste. Most geothermal fluid will require some
Removal of dissolved and suspended solids prior to injection.
Settling ponds, tanks or filtration can be used to remove solids.
liquid waste may be chemically treated to reduce corrosivity
scaling. At the expense of heat loss, fluid from liquid-
187
-------
dominated reservoirs may be treated prior to entering the energy
conversion system. This would reduce chemical problems through-
out the system.
Various preinjection treatment methods for all types of
wastewater are described in detail by Warner (1977). Those
methods applicable to scaling control in geothermal wells are
summarized in Table 5.2.
5.6.2 Delivery Systems
Although gravity-is normally a sufficient force for geo-
thermal waste injection, it may be necessary in some instances to
pump the waste down the hole. Often, for example, plugging of
the formation will occur after injection has begun. To maintain
an adequate injection rate, the pressure must be increased by
pumping.
The type and size of pump may best be selected after the
well is installed and pumping tests are made to determine the
wellhead pressures needed for injection. Pump selection is made
on the basis of wellhead pressure, volumes to be injected, and
injection rate variability. Corrosion resistance is also a major
consideration in pump selection and design.
Centrifugal pumps are the most common type used for
injection. For low-pressure injection (less than 350 kPa [50
psi]), single-stage centrifugal pumps are adequate. Higher
wellhead pressures will require multiplex piston-type or
multi-stage centrifugal pumps (Warner, 1977).
5.6.3 Well Design
Injection well design is one of the most critical aspects of
injection well technology. For adequate pollution protection
rigorous safety specifications must be designated and enforced.
This includes specification of tubing placement within the casing'
monitoring fluid pressures, complete grouting and
welding, placement of packers, and selection of corrosion-
resistant casing and tubing.
Recommendations have been made for injection well comple-
tion design for hazardous waste disposal (Reeder, et al. 1977).
Following these recommendations for geothermal injection well
design and construction will ensure maximum protection of usabl6
subsurface waters and economic minerals. Fig. 5.5 is a detailed
diagram of the well completion recommended for maximum protec-
tion during injection of hazardous waste. The first casing
string (largest diameter) extends at least 15 m (50 ft) below
the lowest fresh water zone penetrated by the well, and is
grouted from the bottom to the surface. This and subsequent
188
-------
FINAL CASING
INTERMEDIATE CASING
SURFACE CASING
- Vo
LOWEST FRESH WATE_R
SOFT. MINIMUM
JL_
50 FT. MINIMUM
'
NOTE. A CASING STRING SHALL BE CEMENTED
THAT CEMENT SHAL1 BE AT LEAST 50 F f.
INTO THE NEXT LAR3ER STRING.
30FT. MAXIMUM^
A
.'1
_
;
'M
ANNULUS PRESSURE GAUGE
GROUND LEVEL
CEMENT
INJECTION TUBIN'3
-CEMENT
- PACKER
EXTERNAL CASING PACKER
INJECTION ZONE
Figure 5.5 Well completion for maximum protection
during hazardous waste injection.
CReeder, et al. 1977)
189
-------
smaller diameter strings have 15 m (50 ft) of cemented annulus
overlap with each other. All the casing is mechanically centered
for a consistent grout seal thickness. Packers are installed at
the top of the injection zone between the casing and the bore-
hole wall, as well as in the injection tube-casing annulus.
The external packer helps prevent attack of the cement and
casing by the injected fluid. The interior packer seals the
annulus so that it can be filled with noncorrosive fluid and
monitored for pressure and chemical changes which might indicate
leaks in the system.
Normally, geothermal injection wells are designed much like
production wells. The production-injection zone in an injection
well is sometimes cased with a slotted liner to prevent slough-
ing, whereas production wells may not require casing in that
zone. Fig. 5.6 shows a typical injection well design.
If the reservoir rock is very competent, an open hole
completion can be used. This has the advantage of maximizing
the receiving surface in the well. It also eliminates casing
and screen corrosion problems in the injection zone. Careful
completion of an injection well above the injection zone is
critical in preventing contamination of ground water. Casing
must be selected for corrosion resistance as well as internal
and external pressure. Welds must be inspected carefully. The
casing may require coating to prevent corrosion. The annulus
between the well bore and the casing must be cemented
completely. No pockets in the grout are tolerated, as they
provide pollutant pathways.
To protect the well casing from corrosive fluid, injection
tubing should be placed within the casing. Grade and weight of
tubing are selected on the basis of internal and external
pressures and on axial loads. The tubing size selected depends
on the planned injection rate. Tubing material selection, like
selection of all materials exposed to geothermal fluid, is
based on the chemistry of the fluid. This tubing provides
extra pollution protection in two ways. First, if the tubing
fails, the injected fluid will still be contained by the casing
and packers. Second, when the annulus between the tubing and
casing is filled with noncorrosive fluid it can be continuously
monitored for pressure changes which may indicate a leak in the
system. Unlike casing, this tubing can be replaced if it fails.
Packers should be installed in the annular space between the
tubing and casing to seal off sections. This sealing isolates
the injection zone and prevents circulation of the injected fluid
inside the casing. Since geothermal injection wells will be
constructed under current regulations they should have all the
features of this recommended design and construction.
190
-------
Fig. 5.6 shows three types of open-hole completions that
have been used in hazardous waste disposal wells. The one on
the right has all the features of the recommended design.
Although each of the remaining completions provide more protection
than injection through a simple cased hole, they do not provide
as much protection as the recommended completion. The one on
the left does not isolate the fluid filling the annulus between
the injection tubing and the casing. In this situation the
injected fluid can flow into the annulus thereby causing
casing corrosion. The absence of a packer to contain the annular
fluid under a constant pressure also prevents detection of leaks
through the tubing. The center diagram illustrates similar
deficiencies although pumping water through the annulus lessens
the possibility of corrosion from injected fluid flowing into
the annulus. Additional drawbacks of this design are: 1) the
amount of acceptable quality water that is required to place in
the well; and 2) the reduction in the amount of waste fluid
the formation can accept due to the emplacement of the additional
water injected through the annulus.
5.6.4 Monitoring Injection Well Operation
The flow rate, injection wellhead pressure and annulus
fluid pressure must be monitored continuously to provide the
necessary data for reservoir management, well maintenance and
pollution control. Chemistry of the injected fluid and annulus
fluid should also be monitored regularly.
Annulus fluid pressures and chemistry are monitored to
detect leakage in the system. Depending on the composition of
the fluid, adequate chemical monitoring may be accomplished by
placing conductivity probes in the annulus, or by analyzing
return flow for contamination in continuous cycling annulus
fluid.
Corrosion rate can be determined by placing sample strips
of the tubing and casing material in the well, and checking them
periodically for weight loss.
Where injecting chemically active fluid, it is important
that the well be shut down periodically for inspection and
testing. inspection methods for casing, tubing, cement and well
bore include: (1) pulling the tubing and inspecting it visually
or instrumentally; (2) electromagnetic caliper or televiewer
logging of tubing or casing in the hole; (3) pressure testing
of casing; (4) bond logging of casing cement; and (5) inspection
of casing cement or well bore with injectivity or temperature
profiles (Warner, 1975). Downhole geophysical methods are
described in detail by Harding-Lawson (in press).
191
-------
DESIGN:
LEAK DETECTION
CAPABILITY:
CASING PROTECTION:
Fluid Filled Annulus,
Pressure Monitor,
No Packer in Annulus,
Tubing Hanging Free
Minimal
Limited
Water Injected into Annulus,
Pressure Monitor
Minimal
Limited
Fluid Filled Annulus with
Pressure Monitor and Packer
Positive
Excellent
ro
injected
fluid
Figure 5.6 Comparison of three types of completions used in hazardous waste injection
viells. (.modified from Reeder, et al. 1977).
-------
5.6.5 Cost Analysis
The cost of a geothermal injection system varies as widely
as the size, type, and chemistry of geothermal reservoirs. The
following is a list of variable factors which control the cost
of a system:
Installation costs:
- Well drilling - depends on the number and depth
of wells, and the type of material penetrated.
Fig. 5.7 gives estimated average drilling costs.
- Well completion - depends on the number, depth,
material penetrated, and on the fluid chemistry.
Highly corrosive fluids may require expensive
resistant materials for grouting, tubing, casing
and other well components.
- Wellhead apparatus - higher injection pressures
will generally necessitate more expensive
injection pumps.
- Treatment facilities - the chemistry and volume
of the fluids are determining factors in
selecting types and size of preinjection treat-
ment facilities.
- Surface conveyances - production and injection,
well spacing (which, in turn, depends on the
reservoir dynamics), length of surface conveyance
from production to injection wells. The size
of the conveyances depends on the volume of
fluids injected. The type, and therefore the
cost, of the material used for the conveyances
is dependent upon the chemistry of the water.
A literature survey revealed examples of actual capital
costs for some geothermal well systems. These data are presented
in Table 5.5 showing, where possible, the breakdown according to
size and depth of the well and costs of construction, pumps
and pipeline. Table 5.6 presents the capital costs for injec-
tion systems for four well capacities.
As in any economic analysis, the location and time of the
development must be considered. Prices vary from one geographic
area to another and, in general, construction costs increase
with distance from a major city where supplies and equipment
can be obtained and manufactured. Engineering News-Record
periodically publishes construction cost indexes which enable
cost adjustments based on geographic area (Engineering News-
Record, 1978).
193
-------
DEPTH (KM)
1.0
Note: Data from 1971-1974
escalated to 1975 at
20%/year
/
AVERAGE
'VOLCANIC LITHOLOGY
/ AVERAGE
/""SEDIMENTARY LITHOLOGY
/ /
/AVERAGE FOR
^ OIL WELLS
0.5 —
V)
k» ™
ro
"o
» -
(0
c
0
5
+•*
(0
O
0 -
/— / — i .' r
X /
/ / / /
/ / ./ //
. / /
/ / x
>^ JJImperlal /
^ XX' Valley /
X ^
^ ^
^S^
-------
TABLE 5.5 CAPITAL COST OF INJECTION WELLS, DATA FROM THE LITERATURE
ID
Ln
(Sung, et al. 1977)
Reference
Date Published
Type Well
Location
Diameter (in.)
Total Vertical
Depth (ft.)
Straight (S) or
Directional (D)
Construction Cost
- Drilling Contractor Cost
- Mud Expense
- Casing & Tubing
- Cementing
- Logging
- Perforation
- Wellhead Equipment
- Engineering Supervision
- Control Sys.
Injection Pump
Injection Pipeline
Total Capital Cost
Pearson
Oct. 76
Injection
East Mesa, Ca.
$403,077
22,831
54,200
224,000
Pearson
Oct. 76
Injection
TRW
Dec. 74
Steam
Production
9-5/8
7,000
Glass
May 77
Steam
Production
The Geysers, Ca,
7,000 to 8,000
Sung, et al.
1977
Injection
AVERAGE
D (48°)
$535,448
D (30°)
$169,000
20,000
10,000
25,000
23,000
40,000
$658,000
30,000
124,500
50,000
33,000
20,000
28,000°
$1,003,500b
$170,000
20,000
70,000
40,000
20,000
50,000
40,000
10,000
30,000
50,000
Oc
500, 000^
a Includes overhead
° Includes miscellaneous costs amounting to $60,000
c No pipeline is required if directional drilling is used. This component will cost several hundred thousand dollars if
required
The total capital cost may vary from approximately $300,000 to $1 million. $500,000 is considered to be a reasonable
average based on costs of a number of existing geothermal and industrial waste disposal wells.
-------
TABLE 5.6 CAPITAL COSTS FOR INJECTION SYSTEMS FOR FOUR
WELL CAPACITIES (Hartley, 1978)
Flow
(liters/
minute)
10
100
500
1000
4000
5000
10,000
15,000
30,000
50,000
100,000
350,000
Well Capacity: 200 Ipr.i/woll Well Capacity:
Annual i?.ed
Wells Initial Cost Per
Required Capital Unit of
Cost Flowb
(S106) ($/1000i)
1 0.5 10.60
0.5 1.06
2 1.0 0.42
5 2.5 0.53
20 10. 0.53
25a 12/5 0.53
Wells Tail
Required Capi
Cos
(S/10002) ($10
1000 Ipm/woll Well Capacity:
Annual ized
ial Ciist For
tal Unit of
t Flowb
6) ($/1000Z)
1 0.5 1,06
1 0.5 0.212
1 0.5 0.106
U 2.0 0.106
5 2.5
0.106
10 5.0 0.106
15 7.5 0.106
303 15.
0.106
Wells Init
Required Capi
Cob
(SIC
^000 Ij.r./wcll Well Capacity: 8000 lv>r./wvll
Annual! zed
ial Cost Per
tal Unit of
t Flowb
6) (S/10001)
1 0.5 0.106
1 0.5 0.0264
2 1.0 0.0422
3 1.5 0.0316
4 2.0 0.0281
8 4.0 0.0281
13 6.5 0.0274
25a 12.5 0.0264
Aar>.ual i:red
Wells Initial Cost Per
Required Capital bV.it of
Cost Flowb
($106) ($/1000i)
1 0.5 0.0264
1 0.5 0.0211
2 1.0 0.0211
2 1.0 0.0141
4 2.0 0.0141
7 3.5 O.OI-.S
13 6.5 0.0137
44 22. 0.0133
f-1
VO
01
a. Arbitrary limit
b. Total cost is annuaiizod based on C * P(CSF), where P - Local.cose
interest and 30-year period. The demand factor for the well is> as
and CM' is capital recovery factor at 8Z
suracd to be 80%.
-------
Operation and Maintenance Costs:
- Injection pumping - depends on the volumes and
the injection pressure.
- Pretreatment - depends on the chemistry of the
geothermal fluid.
- Surface conveyance - depends on the size and
configuration of the system, the volumes
transported, and whether pumping is required.
- Removal of scale from injection system components -
depends on the chemistry and volumes of the fluid
and the types of scale control methods employed in
the system.
- Replacement of components - depends upon the
corrosivity of the fluids and the resistance of
the component material to chemical attack.
- Monitoring costs - the extent of monitoring
systems will depend on regulatory specifications,
the size, depth and dynamics of the reservoir
and the desired degree of aquifer protection.
5.7 CASE HISTORIES
5.7.1 Wairakei, New Zealand
The water-dominated field at Wairakei has been in production
since 1951. Injection has not been practiced here and up to 4.5
m (14.8 ft) of ground subsidence occurred between 1964 and 1974
(Stilwell, et al. 1976). This subsidence correlates with the
decrease in aquifer pressure and the total mass output of the
field (Hatton, 1970).
Injection might help control subsidence, but for fear of
contributing to the declining temperature of the reservoir
(from 250°C [482°F] in 1951 to 238°C [460°F] in 1970) injection
is not presently done regularly. The high silica content of the
water might create scaling and plugging in injection wells
(Rothbaum and Anderton, 1976).
5.7.2 Otake, Japan
The Otake geothermal field produces wet steam (80% liquid)
from a 300 m (1/000 ft) thick, highly permeable, fractured tuff
breccia. The reservoir rock is overlain and underlain by rela-
tively impermeable volcanics (Yamasaki, et al. 1970).
197
-------
The 10 MW power plant at Otake is solely for experimenta-
tion. Four wells provide 109,000 kg/hour (240,000 Ib/hr) of
steam to the turbine, with more than 400,000 kg/hr (880,000 Ib/hr)
of hot water separated prior to power generation. All the
geothermal water is injected into three wells completed in the
production zone (Kubota and Aosaki, 1976).
Injection testing began at Otake in 1968 and tests were
conducted from 1968 to 1971 using two converted production wells.
Drilling of the three injection wells now in use began in 1972.
Wells were drilled between 150 and 500 m (500 and 1,640 ft) from
the nearest production well to enable determination of minimum
spacing between future production and injection wells. Waste
fluid is injected below the producing level to prevent thermal
degradation of the reservoir or chemical degradation of overlying
aquifers.
Injection wells at Otake are cased essentially the same as
production wells, with casing to the bottom of the caprock and
cementing of the entire annular space. No injection tubing is
used. The wellhead valve is equipped with a return pipe for
shutoff if a blowout occurs. A flow meter on the wellhead
measures both velocity and volume.
Injection capacities of the three wells have dropped
significantly. The capacity of one well, for example, decreased
from 310,000 kg/hr to 120,000 kg/hr (680,000 to 260,000 Ib/hr)
during the first three years of operation. The ground water
level rose 30 m (100 ft) and 25 mm (1 in) of silica scale had
accumulated on the well wall at the wellhead. However, chemical
monitoring of surface waters indicates that the injected water
is not leaking to the surface (Kubota and Aosaki, 1976).
Prior to injection, production was declining by about 6% per
year. (In four years the power output went from 11 MW to 8.7
MW.) After injection began, the power station output rose to 10
MW without additional production wells.
Injection has apparently caused no temperature drop in the
reservoir or surrounding formation. The enthalpy of the
produced fluid has remained essentially constant.
Seismicity has been monitored at four stations near the
wells since injection began. No anomalous earthquakes have been
recorded (Kubota and Aosaki, 1976).
5*7'3 Cerro Prieto, Mexico
Liquid waste is currently stored in an evaporation and
sedimentation pond on clayey saline soils showing thermal
manifestations. Although no adverse effects have been
associated with this storage pond final disposal of the waste
198
-------
fluid will be by surface discharge or injection. Two waste
canal discharge schemes have been proposed: one to the Sea of
Cortez and one to the Laguna Salada. The 75 MW Cerro Prieto
geothermal plant produces wastewater with high concentrations
of SiO2f NaCl, KC1, and other salts. Injecting into a known
highly permeable area west of the fields might be feasible for
waste control, as well as reservoir recharge (Mercado, 1976).
5.7.4 Valles Caldera, New Mexico
During 1973 and 1974, injection tests were made in two
wells in the Valles Caldera liquid-dominated reservoir. Water
from a separator flowed into holding ponds and was then injected.
A total of about 380,000 cu m (100 million gallons) was injected
during that time without impaired injectivity (Chasteen, 1976).
5.7.5 The Geysers, California
Injection at The Geysers geothermal area began in 1969,
nine years after production began. By 1975, over 15 million cu m
(about 4 billion gal) had been injected. In 1967, the total
injection rate was about 0.21 cu m/sec (55 gpm) into six wells.
About 100 production wells have been drilled into the metamorphosed
highly fractured shale and sandstone reservoir. Wells are about
1,000 m (3,300 ft) deep on the average, with some deep wells up
to about 3,000 m (10,000 ft) (Chasteen, 1976).
The original injection wells were unsuccessful production
wells, into which slotted liners were placed to keep the holes
open for injection. The lower zone in the first injection well
became plugged, forcing the waste into a higher horizon and
ultimately cooling a nearby production well. Injection wells
are now drilled deeper than producing wells and as far as
possible from producing regions.
Preinjection settling basins allow solids to precipitate,
thereby helping to prevent contamination and plugging of the
injection wells. Oxidation and corrosion are controlled by
deaerating the injection system. No pumping is required for
injection at The Geysers.
Elemental sulfur in the injected water clogs the formation
around the wells, thereby decreasing injection rates. By
shutting the well off and allowing it to reach the 250°C (482°F)
reservoir temperature, the sulfur melts (at 114°C [237°F]). it
can then be flushed away from the hole by resuming injection.
Subsidence and microseismicity are being monitored at
The Geysers by the U.S. Geological Survey. The area exhibits
considerable microseismic activity. However, no subsidence or
anomalous microseismic activity has been observed which can be
correlated to geothermal development.
199
-------
5.7.6 Larderello, Italy
Some injection tests have been done in and around the
producing regions of this dry steam field, although the
condensate is not currently disposed of by injection.
In 1973 in the Viterbo region, fluid from one well was
injected into another, 4 km (2.5 mi) away. The 62°C (.144°F)
fluid was injected under gravity for nine days at rates ranging
from 0.0035 to 0.0355 cu m/sec (55 to 567 gpm) (Calamai, et al.
1973) . A seismic control program involving five microseismic
stations was operated for 40 days before, during and after the
injection test. No microshocks resulted from the injection test
(Cameli and Carabelli, 1976).
5.7.7 Ahuachapan, El Salvador
The Ahuachapan field produces high temperature, high salinity
wet steam from a fractured, high permeability reservoir 500 to
1200 m (1,600 to 3,900 ft) below the surface. Temperatures of
the reservoir exceed 200°C (392°F) and have 8000 ppm total
dissolved salts and 650 ppm SiC>2. Silica scaling is a serious
problem, and has rendered surface disposal uneconomical due to
clogging of pipes and ditches.
Injection experiments were done between 1970 and 1975.
The first injection well was drilled outside the reservoir to
avoid cooling the geothermal fluid, but permeabilities were too
low for effective injection. This forced a major revision in
test plans.
Several factors led to the decision to test injection
directly into the reservoir. Based on heat exchange studies,
it was determined that cooling of the reservoir by injected
water would be within acceptable limits (Bodvarsson, 1972).
Because of the high permeability of the reservoir, injection
into it would not require as much energy as the initial test
into low permeability material. Recycling the residual heat of
the wastewater might prolong the productive life of the reservoir*
A well was drilled in the production area for the injection
experiments, with the intent of converting it to a producer. It
was completed with a retractable liner which was extended from
the production casing to the bottom of the hole at 952 m (3,123
ft) below the surface. This was done to inject into permeable
horizons below the production zone.
Study of the chemical equilibrium at Ahuachapan indicated
that silica deposition could be avoided by separating the steam
from the water and maintaining the wastewater at a temperature
above 150°C (302°F) . Also, the temperature of the injected fluid
200
-------
should not differ from that of the reservoir rock by more than
50°C (122°F). By maintaining separator and wastewater temperature
between 152°C and 153°C (306°F and 307°F.) , no scaling or mineral
deposition occurred during the 244 days of injection experiments.
Temperature, pressure and chemical changes were monitored
in surface springs and production wells during the injection
tests. No changes were noted. Tritium tracers injected into the
injection zone showed up in small concentrations in production
wells within 500 m (.1,600 ft) of the injection well. However,
none of the tracer has appeared in surface water or fresh water
aquifers in the area.
Following the injection tests, the well was put into produc-
tion. Original production temperatures were readied in about
the same length of time that injection was performed (Einarsson,
et al. 1976).
5'7'8 Imperial Valley Fields, U.S.A.
The Imperial Valley is a large geothermal region in the
southwestern U.S. Reservoirs in this region are in sedimentary
rock; some have intergranular permeability and some have
fracture permeability. The fields are generally liquid-dominated
and though the temperatures are high, severe corrosion and
scaling problems have inhibited development of the resource.
Several test facilities are being designed and operated in the
Imperial Valley. The following fields are among those which are
being used for injection testing.
East Mesa—
The East Mesa geothermal field is a liquid-dominated
reservoir in fractured sandstone. High-temperature gradients,
relatively low salinity fluids (less than 30,000 ppm) and a
thick production zone make East Mesa an attractive area for
development. The Bureau of Reclamation has established East
Mesa as a national facility for testing materials and equipment
in a geothermal environment.
In 1975, testing began on experimental injection well
Mesa 5-1. Geothermal brine from a holding pond was injected at
rates between 0.003 and 0.015 cu m/sec (55 and 235 gpm)
(Mathias, 1976). Intermittent booster pumping was done as the
injection rate decreased with total injection volumes. High
corrosion of the instrument used for downhole pressure monitor-
ing was probably caused by high dissolved oxygen content.
Mixing of fluids from two wells caused great amounts of
calcium carbonate to precipitate in the injection pipeline
system. The precipitate was washed out with water but it
201
-------
has demonstrated the problem of mixing waters for injection
(U.S. Dept. of the Interior, 1977).
Niland—
Union Oil Company tested injection in this liquid-dominated
reservoir in 1964 and 1965. After overcoming the 1.4 MPa (200
psig) wellhead pressure, the testers found that the relatively
heavy, cold injected fluid effectively created a vacuum on the
injection system. This resulted in an injection rate of about
0.04 cu m/sec (600 gpm), and a total of 480,000 cu m
(126,000,000 gal) of fluid was injected. Injectivity loss or
reservoir response due to the injection test was not observed
(Chasteen, 1976).
Heber, California—
The liquid-dominated Heber field is the site of an experi-
mental geothermal facility. Plans for the Heber geothermal
demonstration plant involve installing 50 to 60 production wells
and 20 to 30 injection wells to serve a 200 MW plant. Nearly
all the produced geothermal fluid will be injected, with
injection wells located on the perimeter of the reservoir, about
3.2 km (2 mi) from the production area. Wells in the field 5
produce between 2.2 x 10b and 3.0 x 105 kg/hr (4.8 and 6.6 x 10
Ib/hr) and each injection well is expected to dispose of about
twice the volume that each production well produces. Scaling
is not expected to be a problem with this saline geothermal
fluid.
5.8 RESEARCH NEEDS
The area of liquid waste injection technology poses complex
problems. The most important of these are outlined below.
5.8.1 Chemical Aspects
1) Well scaling and formation plugging mechanisms
need to be better understood in order to control
these processes and to determine the longevity
of injection wells. Chemical aspects of scaling
and plugging that require further research
include:
a) silication chemistry;
b) chemical inhibitors;
c) effects of process variables;
d) effects of mixing fluids from different
wells; and
202
-------
e) injection/receiving fluid and injection
fluid/receiving rormation interactions.
2} Corrosion mechanisms and prevention measures
deserve attention. The following need study:
a) removal and disposal of corrosive components
of geothermal fluids; and
b) site-specific studies of corrosion-resistant
materials.
5.8.2 Equipment Development Needs
Specialized equipment is essential for a better understand-
ing of the subsurface injection horizons and for system
maintenance. Some candidates for research and development
include:
1) well logging tools and interpretation techniques
for use in high-temperature environments and
nonsedimentary lithologies;
2) instruments such as permeability indicators
and electrochemical monitoring tools; and
3) injection emergency backup systems.
5.8.3 Reservoir Engineering
The following aspects of geothermal reservoir engineering
need further study before geothermal reservoirs can be properly
managed. These include:
1) the understanding of pore geometry; particularly
the hydraulics of fracture permeability,
2) thermal stress in production zones,
3) advanced methods and interpretation for injec-
tion tests, and
4) computer modeling of reservoir response with
special attention to:
- pore geometry,
- temperature gradients—vertical and hori-
zontal, both natural and imposed,
203
-------
- existing hydraulic gradients and flow dynamics
including fluid density, pressure and TDS
concentration
- changes in fluid dynamics imposed by produc-
tion and injection wells, and
- decrease in injectivity due to plugging.
5.8.4 Physical Problems
Physical problems in injection procedures which need further
research are:
1) earthquake stimulation—the potential for
induced seismicity,
2) hydrofracturing—to increase injectivity while
preventing damage to reservoir caprock, and-
3) the effect of reservoir pressure changes on
surface buckling.
204
-------
REFERENCES
Axtmann, R. C. 1976. Chemical Aspects of the Environmental
Impacts of Geothermal Power. Proceedings, Second UN
Symposium on the Development and Use of Geothermal
Resources. pp. 1323-1327.
Bodvarsson, G. June 1972. Thermal Problems in the Siting of
Reinjection Wells. Geothermics. International Journal of
Geothermal Research. International Institute for Geo-
thermal Research, Pisa, Italy. Vol. 1, No. 2. pp. 63-66.
Calamai, A., P. Ceron, G. Ferrara, and G. Monetti. 1973. A
Reinjection Experiment in the Vico 1 Well. Geothermics.
Vol. 2, Nos. 3-4. pp. 117-118.
Cameli, G. M., and E. Carabelli. 1976. Seismic Control During
a Reinjection Experiment in the Viterbo Region (Central
Italy), Proceedings, Second UN Symposium on the Development
and Use of Geothermal Resources, pp. 1329-1334.
Chasteen, A. J. 1976. Geothermal Steam Condensate Reinjection.
Proceedings, Second UN Symposium on the Development and Use
of Geothermal Resources. pp. 1335-1336.
Collins, A. G. 1975. Geochemistry of Oil Field Waters.
Bartlesville Energy Research Center. Elsevier Scientific
Publishing Company, Amsterdam. 496 pp.
Cuellar, G. 1976. BehavJ.cr of Silica in Geothermal Waste
Waters. Proceedings, Second UN Symposium on the Develop-
ment and Use of Geothermal Resources, pp. 1343-1347.
Donaldson, E. C., R. D. Thomas, and K. H. Johnston. 1974. Sub-
surface Waste Injection in the United States—Fifteen Case
Histories. U.S. Bureau of Mines Information Circular 8636.
72 pp.
Engineering News-Record. May 11, 1978. Materials Prices, Vol.
200, No. 19. pp. 38-39.
Einarsson, S. S., A. Vides, G. Cuellar. 1976. Disposal of
Geothermal Waste Water by Reinjection. Proceedings, Second
UN Symposium on the Development and Use of Geothermal
Resources. pp. 1349-1363.
205
-------
Facca, G. 1973. The Structure and Behavior of Geothermal
Fields. Geothermal Energy. (Earth Sciences 12). UNESCO.
pp. 61-72.
Futures Group. 1975. A Technology Assessment of Geothermal
Energy Resource Development. Prepared for National Science
Foundation. Contract C-836, Research Applications
Directorate. 554 pp.
Glass, W. A. 1977. Drilling Methods and Costs at The Geysers.
In: Transactions, GRC Annual Meeting, May 9-11, 1977.
Vol. 1. pp. 103-106.
Grubbs, D. M., C. D. Haynes, T. H. Hughes, and S. H. Stow.
June 1972. Compatability of Subsurface Reservoirs with
Injected Liquid Wastes. Natural Resources Center, The
University of Alabama. 128 pp.
Hatton, J. W. 1970. Ground Subsidence of a Geothermal Field
During Exploitation. Proceedings, Second UN Symposium on
the Development and Use of Geothermal Resources, pp.
1294-1296.
Hartley, R. P. June 1978. Pollution Control Guidance for
Geothermal Energy Development. U.S. EPA, Industrial
Environmental Research Laboratory, Cincinnati. EPA-600/
7-78-101. 146 pp.
Her, R. K. 1973. Surface and Colloid Science. New York,
John Wiley and Sons. Vol. 6, Ch. 1. Matejevic, E., ed.
311 pp.
Juul-Dam, T. and H. F. Dunlap. 1976. Economic Analysis of a
Geothermal Exploration and Production Venture.
Proceedings, 2nd UN Symposium on the Development and Use
of Geothermal Resources. Vol. 3. pp. 2315-2324.
Kubota, K., and K. Aosaki. 1976. Reinjection of Geothermal Hot
Water at the Otake Geothermal Field. Proceedings, Second
UN Symposium on the Development and Use of Geothermal
Resources. pp. 1379-1383.
LaQue, F. L. 1975. Marine Corrosion: Causes and Prevention.
John Wiley, New York. 332 pp.
Lawrence Berkeley Laboratory. April 1977. Geothermal Subsidence
Research Program Plan. Energy and Environment Division.
LBL-5983, UC 66E. Ill pp.
Mathias, A. E. 1976. The Mesa Geothermal Field—A Preliminary
Evaluation of Five Geothermal Wells. Proceedings, Second UN
Symposium on the Development and Use of Geothermal
Resources. Vol. 3. pp. 1741-1747.
206
-------
McWilliams, J. 1972. Large Saltwater-Disposal Systems at East
Texas and Hastings Oil Fields, Texas, in Underground Waste
Management and Environmental Implications. T. D. Cook, ed.
American Association of Petroleum Geologists. Memoir 18.
pp. 331-340.
Meidav, T., and S. K. Sanyal. December 1976. A Comparison of
Hydrothermal Reservoirs of the Western U.S. Electric
Power Research Institute Topical Report ER-364. Project
580. Prepared by Geonomics, Inc. 17Q pp.
Mercado, S. 1976. Cerro Prieto Geothermoelectric Project:
Pollution and Basic Protection. Proceedings, Second UN
Symposium on the Development and Use of Geothermal
Resources. pp. 1394-1398.
National Water Well Association. 1978. The Well Log. Vol. 8,
No. 10. October 1978. pp. 1-2.
Owen, L. B. May 1977. Properties of Siliceous Scale from the
Salton Sea Geothermal Field. Geothermal Resources Council.
Transactions. Vol. 1. pp. 245-247.
Ozawa, T., and Y. Fujii. 1970. A Phenomenon of Scaling in
Production Wells and the Geothermal Power Plant in the
Matsukawa Area. Proceedings of the UN Symposium on the
Development and Utilization of Geothermal Resources.
pp. 1613-1618.
Pearson, R. O. October 1976. Planning and Design of Additional
East Mesa Geothermal Test Facilities. (Phase IB). Vol. 1.
Final Report by TRW, Inc. 91 pp.
Phelps, P. L., and L. R. Anspaugh. 1976. Imperial Valley
Environmental Project. Progress Report. Lawrence Livermore
Laboratory. UCRL-50044-76-1. 214 pp.
Phillips, S. L., A. K. Mathur, R. E. Deobler. 1976. A Survey of
Treatment Methods for Geothermal Fluids. Paper No. SPE
6606. Am. Inst. of Mining Engineers, Dallas, Texas. 8 pp.
Pirson, S. J. 1963. Handbook of Well Log Analysis. Prentice-
Hall, Inc., Englewood Cliffs:, New Jersey. 326 pp.
Poland, J. F. 1973. Subsidence in United States Due to Ground
Water Overdraft—A Review. Am. Soc. Civil Engineers.
Proceedings, Specialty Conference, Fort Collins, Colorado.
11 PP-
207
-------
Poland, J. F., and G. H. Davis. 1969. Land Subsidence Due to
Withdrawal of Fluids. In: Varnes, D. J., and G. Kiersch,
eds. Reviews in Engineering Geology. Geol. Soc. Am.
pp. 187-269.
Raleigh, C. B., J. H. Healy, and J. D. Bredehoeft. 1972. An
Experiment in Earthquake Control at Rangely, Colorado.
Science. Vol. 191. pp. 1230-1236.
Reeder, L. R. , J. H. Cobbs, J. VI. Field, Jr., W. D. Finley,
S. C. Vokurka, and B. N. Rolfe. June 1977. Review and
Assessment of Deep We.ll Injection of Hazardous Waste.
EPA-60Q/2-77-029a. 215 pp.
Renner, J. L., D. E. White, D. L. Williams. 1975. Hydrothermal
Convection Systems. In: Assessment of Geothermal Resources.
U.S. Geological Survey Circular 726. pp. 5-57.
Ridley, A. P. and C. L. Taylor. 1976. Earthquake-Related
Geologic and Seismic Safety of Geothermal Developments.
Second UN Symposium on the Development and Use of Geothermal
Resources. pp. 1411-1415.
Rothbaum, H. P., and B. H. Anderton. 1976. Removal of Silica
and Arsenic from Geothermal Discharge Waters by Precipita-
tion of Useful Calcium Silicates. Proceedings, Second UN
Symposium on the Development and Use of Geothermal Resources.
pp. 1417-1425.
Sanyal, S. K. June 1978. Personal communication.
Sheahan, N. T. January-February 1977. Injection/Extraction
Well System - A Unique Seawater Intrusion Barrier. Ground
Water, Vol. 15. No. 1. pp. 32-50.
Stilwell, W. B., W. K. Hall, and J. Tawhai. 1976. Ground
Movement in New Zealand Geothermal Fields. Proceedings,
Second UN Symposium on the Development and Use of Geothermal
Resources. pp. 1427-1434.
Sun, R. J. 1973. Hydraulic Fracturing as a Tool for Disposal
of Wastes in Shale. In: Underground Waste Management and
Artificial Recharge. Am. Assoc. of Petroleum Geologists.
pp. 219-272.
Sung, R., G. Hauser, G. Richard, J. Coffer, R. Walker and E.
Pulaski. December 1977. Preliminary Cost Estimate of
Pollution Control Technologies for Geothermal Development.
Preliminary Draft Report for U.S. EPA, IERL, Cincinnati,
Ohio. Prepared by TRW, Inc.
208
-------
Systems Control, Inc. 1976. The Analysis of Subsidence
Associated with Geothermal Development. Vol. 1, Handbook.
NSF/RAND 5139-1.
Taylor, S. S., C. J. Wilhelm and w. C. Holleman. January 1939.
Typical Oil Field Brine-Conditioning Systems: Preparing
Brine for Subsurface Injection. R.I. 3434, U.S. Dept.
of the Interior, Bureau of Mines.
TRW, Inc. December 31, 1974. Experimental Geothermal
Research Facilities Studies, Phase D. Vol. 1. Final
Report Prepared for NSF.
U.S. Dept. of the Interior, Bureau of Reclamation. 1977.
Geothermal Resource Investigations, East Mesa Test Site,
Imperial Valley, California. Status Report. April 1977.
99 pp.
Wahl, E., and I, yen. 1976. Scale Deposition and Control
Research, for Geothermal Utilization. Proceedings, Second
UN Symposium on the Development and Use of Geothermal
Resources. pp. 1855-1864.
Warner, D. L. 1975. Monitoring Disposal Well Systems. U.S. EPA
Report No. 680/4-75-008, Las Vegas. 109 pp.
Warner, D. L. 1972. Survey of Industrial Waste Injection Wells,
Final Report. U.S. Geological Survey Contract No. 14-08-
0010-12280. University of Missouri, Rolla. 3 vols.
Warner, D. L., and J. H. Lehr. 1977. An Introduction to the
Technology of Subsurface Wastewater Injection. EPA-600/2-
77-240. 355 pp.
Yamasaki, T. , Y. Matsumoto, and M. Hayashi. 197.0. The Geology
and Hydrothermal Alterations of Otake Geothermal Area,
Kujyo Volcano Group, Kyushu, Japan. UN Symposium on the
Development and Utilization of Geothermal Resources, Pisa,
Italy. Vol. 2, Part 1. pp. 197-207.'
Yanagase, T., Y. Suginohara, and K. Yanagase. 1970. The
Properties of Scales and Methods to Prevent Them. UN
Symposium on the Development and Utilization of Geothermal
Resources, Pisa, Italy. Vol. 2, Part 2. pp. 1619-1623.
209
-------
APPENDIX A
EPA POSITION ON SUBSURFACE EMPLACEMENT OF FLUIDS
(from Federal Register, April 9, 1974, p. 12922-12923)
INTRODUCTORY COMMENTS
The Environmental Protection Agency,
In concert with the objectives of the Fed-
eral Water Pollution Control Act, as
amended <33 U.S.C. 1251 et seq,; 86 Stat,
816 et seq.; Pub. L. 92-500) "... to
restore and maintain the chemical, phy-
sical, and biological Integrity of the Na-
tion's water" has established an EPA
policy on Subsurface Emplacement of
Fluids by Well Injection" which was Is-
sued Internally as Administrator's Deci-
sion Statement No. 5. The purpose of the
policy Is to establish the Agency's con-
cern with this technique for use in fluid
storage and disposal and its position of
considering such fluid emplacement only
where it is demonstrated to be the most
environmentally acceptable available
method ot handling fluid storage or dis-
posal. Publication of the Policy as In-
formation establishes the Agency's posi-
tion and provides guidance to other Fed-
eral Agencies, the States, and other In-
terested parties.
Accompanying the policy statement
are "Recommended Data Requirements
for Environmental Evaluation of Sub-
surface Emplacement of Fluids by Well
Injection wen system; and to Insure
ments Is to provide guidance for potential
Injectors and regulatory agencies con-
cerning the kinds of Information re-
quired to evaluate the prospective
injections well system; and to insure
protection of the environment. The
Recommended Data Requirements re-
Quire sufficient information to evaluate
complex Injection operations for haz-
ardous materials, but may be modified
In scope by a regulatory agency for
other types of injection operations.
The EPA recognizes that for certain
Industries and in certain locations the
disposal of wastes and the storage of
fluids in the subsurface by use of well
injection may be the most environmen-
tally acceptable practice available. How-
ever, adherence to the policy requires
the potential injector to clearly demon-
strate acceptability by the provision of
technical analyses and data Justifying
the proposal. Such demonstration re-
quires conventional engineering and
other analyses which Indicate beyond
a reasonable doubt the efficacy of the
proposed injection well operation.
Several issues within the policy should
be highlighted and explained to avoid
confusion. One of the goals of the pol-
icy is to protect the Integrity of the
subsurface environment. In the context
of the policy statement. Integrity means
the prevention of unplanned fracturing
or other physical impairment of the geo-
logic formations and . the avoidance of
undesirable changes in aquifers, mineral
deposits or other resources. It is recog-
nized that fluid emplacement by well
injection may cause some change in
the environment and, to some extent,
may preempt other uses.
Emplacement Is intended to Include
both disposal and storage. The differ-
ence between the two terms Is that stor-
age implies the existence of a plan for
recovery of the material within a rea-
sonable time whereas disposal implies
that no recovery of the material is
planned at a given site. Either opera-
tion would require essentially the same
type of information prior to injection.
However, the attitude of the appropriate
regulatory agency toward evaluation of
the proposals would be different for each
type operation. The EPA policy recog-
nizes the need for injection wells in cer-
tain oil and mineral extraction and fluid
storage operations but requires sufficient
environmental safeguards to protect
other uses of the subsurface, both dur-
ing the actual injection operation and
after the Injection has ceased.
The policy considers waste disposal by
well injection to be a temporary means of
disposal in the sense that it Is approved
only for the life of an Issued, permit.
Should more environmentally acceptable
disposal technology become available, a
change to such technology would be re-
quired. The term "temporary" Is not In-
tended to imply subsequent recovery of
210
-------
Injected waste for processing by another
technology.
Paragraph 5 of the policy and program
guidance provides thai EPA will apply the
policy to the extent of Its authorities la
conducting all EPA program activities.
The applicability of the policy to partici-
pation by the several States In the
NPDES permit program under section
402 of the Federal Water Pollution Con-
trol Act as amended has been established
previously by { 124.80(d) of Part 124 en-
titled "State Program Elements Neces-
sary for Participation In the National
Pollutant Discharge Elimination Sys-
tem." 37 FR 28390 (December 22, 1972).
These guidelines provide that each EPA
Regional Administrator must distribute
the policy to the Director of a State water
discharge permit issuing agency, and
must utilize the policy In his own review
of any permits for disposal of pollutants
Into wells that are proposed to be Issued
by States participating In the NPDES.
Dated: April 2.1974.
JOHN QUAKLES,
Acting Administrator.
ADMINISTRATOR'S DECISION
STATEMENT NO. 5 EPA POLICY
ON SUBSURFACE EMPLACEMENT
OF FLUIDS BY WELL INJECTION
This ADS records the EPA'g position on In-
jection wells and rubsurface emplacement of
nulda fay well Injection, and supersedes the
Federal Water Quality Administration'! order
COM 6040.10 Of October 1$. 1870.
Coals. The EPA Policy on Subsurface Em-
placement of nulda by Well Injection U
designed to:
(1) Protect the subsurface from pollu-
tion or other environmental hazard* Attrib-
utable to Improper Injection or Ul-cited in-
jection wells.
(2) Ensure that engineering and geological
safeguards adequate to protect the Integrity
of the subsurface environment are adhered
to In the preliminary Investigation, design.
construction, operation, monitoring and
abandonment phase* of Injection well proj-
ects.
(3) Encourage development of alternative
means of disposal which afford greater en-
vironmental protection.
Principal finding* on* policy rational.
The available evidence, concerning Injection
wells and subsurface emplacement of fluids
Indicates that:
(1) The emplacement of fluids by subsur-
face injection often Is considered by govern-
ment and private agencies as an attractive
mechanism for final disposal or storage owing
to: (•) the diminishing capabilities of sur-
face waters to receive effluents without vio-
lation of quality standards, and (b) the
apparent lower costs of this method of dis-
posal or storage over conventional and ad-
vanced wast* management techniques.
Subsurface storage capacity Is a natural re-
source of considerable value and like any
other natural resource Its u»» must be con-
served for maximal benefits to all people.
(2) Improper Injection of municipal or
Industrial wastes or Injection of other fluids
for storage or disposal to the subsurface en-
vironment could result In serious pollution
of water supplies or other environmental
hazards.
(3) The effects of subsurface Injection and
the fate of Injected materials are uncertain
with today's knowledge and could result In
serious pollution or environmental damage
requiring complex and costly -solutions on a
long-term basis.
Policy and program yuidancf. To ensure
accomplishment of the subsurface protection
goals established above It 1s the policy of the
Environmental Protection Agency that:
(1) The EPA will oppose emplacement of
materials by subsurface Injection without
strict controls and a clear demonstration
that such emplacement will not Interfere
with present or potential use of the subsur-
face environment, contaminate ground water
resources or otherwise damage the environ-
ment.
<2) All proposals for subsurface Injection
should be critically evaluated to determine
that:
(a) All reasonable alternative measures
have been explored and found less satisfac-
tory In terms of environmental protection;
(b) Adequate prelnjectlon tests have been
made for predicting the fate of materials
Injected;
(c) There Is conclusive technical evidence
to demonstrate that such Injection will not
Interfere with present or potential use of
water resources nor result In other environ-
mental hazards;
(d) The subsurface Injection system has
been designed and constructed to provide
maximal environmental protection;
(e) Provisions have been made for moni-
toring both the Injection operation and the
resulting effects on the environment;
(f) Contingency plans that will obviate
any environmental degradation have been
prepared to cope with all well shut-lus or any
well failures;
(B) Provision will be made for supervised
plugging of Injection wells when abandoned
and for monitoring to ensure continuing en-
vironmental protection.
(3) Where subsurface Injection Is practiced
for waste disposal. It will be recognized as »
temporary means of disposal until new tech-
nology becomes available enabling more as-
sured environmental protection.
(4) Where subsurface Injection Is practiced
for underground storage or for recycling of
natural fluids. It wUl be recognized that such
practice will cease or be modified when a
hazard to natural resources or the environ-
ment appears Imminent.
(3) The EPA will apply this policy to the
extent of Its authorities In conducting all
program activities, Including regulatory ac-
tivities, research and development, technical
as*lstance to the States, and the administra-
tion of the construction grants. State pro-
gram grants, and basin planning grants pro-
grams and control of pollution at Federal
facilities In accordance with Executive Order
I 1752.
WILLIAM D. RucittLBHAtrs.
Administrator.
6, 1973.
RECOMMENDED DATA REQUIREMENTS
FOR ENVIRONMENTAL EVALUATION
OF SUBSURFACE EMPLACEMENT OF
FLUIDS BY WELL INJECTION
The Administrator's Decision Statement
No. 6 on subsurface employment of fluids by
211
-------
well Injection has been prepared to establish
the Agency's pewit Ion on the us* of this dis-
posal and storage technique. To aid in Im-
plementation of the policy a recommended
data base for environmental evaluation ha*
been developed.
The following parameters describe the In-
formation which should be provided by the
Injector and are designed to provide regula-
tory agencies sufficient Information to evalu-
ate the environmental acceptability of any
proposed well Injection. A potential Injector
abould Initially contact the regulatory au-
thority to determine the preliminary Investi-
gative and data requirements for a particular
Injection well aa these may vary for different
kinds of Injection operations. The appropriate
regulatory authority will specify the exact
data requirements on a case by case basis.
(a) An accurate plat showing location and
surface elevation of proposed Injection well
site, surface- features, property boundaries.
and surface and mineral ownership at an
approved scale.
(b) Maps Indicating location of water wella
and all other wells, mines or artificial pene-
trations. Including but not limited to oil and
gas wells and eiploratory or test wells. show-
Ing depths, elevations and the deepest forma-
tion penetrated within twice the calculated
rone of Influence of the proposed project.
Plugging and abandonment records for all oil
and gas tests, and water wells should accom-
pany the map.
(c) Maps Indicating vertical and lateral
limits of potable water supplies'which would
Include both short- and long-term variations
In surface water supplies and subsurface
aquifers containing water with less than
10.000 mg '1 total dissolved solids. Available
amounts and present and potential uses of
these waters, as well as projections of public
water supply requirements must be consid-
ered.
(d) Descriptions of mineral resources pres-
ent or believed to be present In area of
project and the effect of this project on
present or potential mineral resources In the
area.
(e) Maps and cross sections at approved
scales Illustrating detailed geologic structure
and a stratlgraphlc section (Including for-
mations, llthology, and physical characteris-
tics) for the local area, and generalized maps
and cross sections Illustrating the- regional
geologic setting of the project.
(f) Description of chemical, physical, and
biological properties and characteristics of
the fluids to be Injected.
(g) PotenUometrtc maps at approved
scales and Isopleth Intervals of the pro-
posed Injection horizon and of those aquifers
Immediately above and below the Injection
horizon, with copies of alt drOl-stem test
charts, extrapolations, and data used In com-
piling such maps.
(b) Description of the location and nature
of present or potentially useable minerals
from the zone of Influence.
(1) Volume, rate, and Injection pressure
of the fluid.
(j) The following geological and physical
characteristics of the Injection Interval and
the overlying and underlying confining beds
should be determined and submitted:
(1) Thickness;
(2) area! extent;
(3) Uthology:
(4) grain mineralogy;
(6) type and mineralogy of matrix;
(8) clay content;
(7) clay mineralogy,
(8) effective porosity (Including an expla-
nation of how determined);
(B) permeability (Including an explana-
tion of how determined);
(10) coefficient of aquifer storage;
(11) amount and extent of natural frac-
turing;
(13) location, extant, and effects of known
or suspected faulting Indicating whether
faults are sealed, or fractured avenue* (or
fluid movement:
(13) extant and effetcs of natural solution
channels;
(14) degree of fluid saturation;
(15) formation fluid chemistry (Including
local and regional variations);
(10) temperature of formation (Including
an explanation of how determined);
(17) formation and fluid pressure (Includ-
ing original and modifications resulting from
fluid withdrawal or Injection);
(IB) fracturing gradients;
(19) diffusion and dispersion characteris-
tics of the waste and the formation fluid In-
cluding effect "f gravity segregation;
(30) compatibility of Injected waste with
the physical, chemical and biological char-
acteristics! of the reservoir; and
(31) tnj«ctlvlty profiles.
(k) The following engineering data should
be supplied:
(1) Diameter at hole and total depth of
well;
(3) type, size, weight, and strength, of all
surface, Intermediate, and injection casing
strings;
(3) specifications and proposed Installa-
tion of tubing and packers;
(4) proposed
-------
APPENDIX B
ABBREVIATIONS
ADA — anthroquinone disulfonic acid
acre-ft — acre-feet
atm — atmosphere
bbl — barrels
BLM — U.S. Bureau of Land Management
BTU — British Thermal Unit
.°C — degrees Celsius
cal — calorie
cm — centimeter
cu — cubic
db — decibel
DOE — Department of Energy (formerly ERDA)
ERDA — U.S. Energy Research and Development Administration
(presently DOE)
EPA — U.S. Environmental Protection Agency
°F — degrees Fahrenheit
ft — feet
gal — gallons
GEAP — Geothermal Environmental Advisory Panel (part of
USGS)
gpm — gallons per minute
gr/gal — grains per gallon
GRO — Geothermal Resources Operational Orders
(issued by USGS)
ha — hectare
hr — hour
in — inch
J — joule
kg — kilogram
KGRA — Known Geothermal Resource Area
km — kilometer
kPa — kilopascal (SI unit of pressure)
kW — kilowatt
kwh — kilowatt-hour
LASL — Los Alamos Scientific Laboratory
L-L — Langelier-Ludwig (type of geochemical diagram)
LLL — Lawrence Livermore Laboratory
1 — liter
lb — pound
1pm — liter per minute
213
-------
m — meter
mg/1 — milligram per liter
mi — mile
mm — millimeter
MPa — megapascal (SI unit of pressure)
mR — milliroentgen
MW — megawatt
MWe — megawatt (electricity)
pCi/1 — picocurie per liter
ppm — parts per million
psi — pounds per square inch
SDWA — Safe Drinking Water Act
sec — second
SUICP — State Underground Injection Control Program
sq — square
TDS — total dissolved solids
USGS — U.S. Geological Survey
USPHS — U.S. Public Health Service
214
-------
APPENDIX C
U.S.-METRIC CONVERSION TABLE
U.S. CUSTOMARY
inch (in)
foot (ft)
yard (yd)
mile (mi)
U.S. EQUIVALENT
Length
0.083 ft
0.33 yd, 12 in
3 ft, 36 in
5,280 ft, 1,760 yd
METRIC EQUIVALENT
25.4 millimeters (mm)
0.3048 meters (m)
0.9144 m
1.609 kilometers (km)
square foot (sq ft)
square yard (sq yd)
acre
square mile (sq mi)
Area
144 sq in
1,294 sq in, 9 sq ft
43,560 sq ft,
4 ,840 sq yd
640 acres
0.0929 square meters (sq m)
0.836 sq m
4,047 sq m,
0.404 hectare (ha)
2.59 square kilometers (sq km)
gallon
cubic yard (cu yd)
cubic mile (cu mi)
Volume
4 quarts
27 cu ft
3.785 liters <1)
7,645 cubic meters (cu m)
4.1655 cubic kilometers (cu km)
gallons per minute (gpm)
18.2 gpd/sq ft
(for water at 60"F)
pounds per hour
cu ft per sec (cfs)
Flow Rate
darcy
3.785 liters per minute (1pm),
6.309 x 10~5 cu m/sec
9.66 x 10"4 cm/sec
(for water at 20°C)
1.260 x 10~4 kg/sec
28.32 Ips, 0.02831 cu m/sec
Miscellaneous
•F « 1.8°C + 32
pounds per square inch (psi)
British Thermal Unit (BTU)
BTU/lb
pound (Ib)
ton
MWe-centuries
3.33 MWe-30 yrs
•C «= (°F - 32)5/9
6.889 kilopascals (kPa)
1,055 joules (J) ,
2S2 (gram) calories
2,325.84 J/kg
0.4536 kilogram (kg)
0.907 netric ton
215
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/7-79-218
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
GEOTHERMAL ENVIRONMENTAL IMPACT ASSESSMENT : Ground
Water Monitoring Guidelines for Geothermal Development
5. REPORT DATE
September 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Richard B. Weiss
Theodora 0. Coffee
Tamata L. Williams
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Harding-Lawson Associates
San Rafael, CA 94902
10. PROGRAM ELEMENT NO.
1NE827
11. CONTRACT/GRANT NO.
68-03-2668
12. SPONSORING AGENCY NAME AND AD.DRESS
U.S. Environmental Protection Agency—Las Vegas
Environmental Monitoring and Support Laboratory
Office of Research and Development
Las Vegas, Nevada 89114
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/07
15. SUPPLEMENTARY NOTES
Donald B. GiImore - Project Officer
Phone:
commercial (702)736-2969
FTS 595-2969
16. ABSTRACT
A proposed groundwater monitoring methodology for geothermal development specifies a
six-step planning and evaluation procedure. Natural geothermal processes and
reservoir development are discussed from the point of view of their effects on
groundwater. Borehole logging techniques are discussed. Potential problems from
chemical and physical plugging and geologic, hydrologic, and reservoir engineering
evaluations are listed and discussed.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
C. COSATI Field/Group__
Environmental monitoring
Groundwater
Water pollution control
Underground disposal
Geothermal Development
Geothermal Monitoring
Guidelines
Groundwater Monitoring
Borehole Logging Techniqu
Waste Water Injection
Technology
13B
s 14A.D
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)
21. NO. OF PAGES
230
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (R«v. 4-77) PREVIOUS EDITION is OBSOLETE
U.S. GOVERNMENT PRINTING OFFICE:
------- |