United States
Environmental Protection
Agency
Office of Emergency and Environmental
Remedial Response Response
Emergency Response Division Team
&EPA
Introduction to
Groundwater Investigations
Environmental Response
Training Program
-------
FOREWORD
This manual is for reference use of students enrolled in scheduled training courses of the U.S.
Environmental Protection Agency (EPA). While it will be useful to anyone who needs information
on the subjects covered, it will have its greatest value as an adjunct to classroom presentations
involving discussions among the students and the instructional staff.
This manual has been developed with a goal of providing the best available current information;
however, individual instructors may provide additional material to cover special aspects of their
presentations.
Because of the limited availability of the manual, it should not be cited in bibliographies or other
publications.
References to products and manufacturers are for illustration only; they do not imply endorsement
by EPA.
Constructive suggestions for improvement of the content and format of the manual are welcome.
-------
INTRODUCTION TO GROUNDWATER INVESTIGATIONS
(165.7)
3 Days
This introductory course is designed to provide participants with information concerning
hydrogeological processes and the necessary elements of a sound groundwater site investigation. It
is intended for personnel who are involved in groundwater contamination investigations but have
little prior hydrogeological experience. This course is not designed for geologists or
hydrogeologists.
Topics that are discussed include hydrogeological definitions and concepts; basic geology and
geochemistry; drilling, construction, and placement of monitoring wells; groundwater sampling
considerations; groundwater flow rates; and groundwater modeling.
Instructional methods include lectures, group discussions, case studies, arid class problem-solving
exercises.
After completing the course, participants will be able to:
• Identify the components of a groundwater system.
• List the primary hydrogeological factors to be considered in a site investigation.
• Construct a flow net and calculate the hydraulic gradient of a simple system.
• Discuss the primary advantages and disadvantages of the most common geophysical
survey methods.
• Identify the different types of pumping tests and the information that can be obtained
from each.
• Describe monitoring well drilling and sampling techniques.
U.S. Environmental Protection Agency
Office of Emergency and Remedial Response
Environmental Response Team
-------
CONTENTS
Acronyms and Abbreviations
Glossary
SECTION 1
SECTION 2
SECTION 3
SECTION 4
SECTION 5
SECTION 6
SECTION 7
SECTION 8
SECTION 9
SECTION 10
GEOLOGY
Article: Geometry of Sandstone Reservoir Bodies
HYDROGEOLOGY
THE HYDROGEOLOGICAL INVESTIGATION
Checklist for a Hydrogeological Investigation
GEOPHYSICAL METHODS
MONITORING THE VADOSE ZONE
WELL CONSTRUCTION
HYDROGEOCHEMISTRY
Article: Migration of Chlorophenolic Compounds at the Chemical Waste
Disposal Site at Alkali Lake, Oregon—1. Site Description and
Ground-Water Flow
Article: Migration of Chlorophenolic Compounds at the Chemical Waste
Disposal Site at Alkali Lake, Oregon—2. Contaminant
Distributions, Transport, and Retardation
Article: Using the Properties of Organic Compounds to Help Design a
Treatment System
GROUNDWATER FLOW RATES AND MODELING
PROBLEM EXERCISES
Problem 1—Flow Net Construction
Problem 2—Geologic Cross-Section Construction
Problem 3—Aquifer Tests
Problem 4—Groundwater Investigation
Problem 5—Nomograph
APPENDICES
Appendix A—Sampling Protocols
Appendix B—References
Appendix C—Sources of Information
9/93
Contents
-------
ACRONYMS AND ABBREVIATIONS
ACS American Chemical Society
AGI American Geological Institute
ARAR applicable or relevant and
appropriate requirement
ASTM American Society for Testing
and Materials
ATSDR Agency for Toxic Substances
and Disease Registry
atm atmosphere
BDAT best demonstrated available
technology
BM Bureau of Mines
BNA base/neutral/acid extractables
BOD biochemical oxygen demand
BTEX benzene, toluene, ethylbenzene,
and xylenes
CAA Clean Air Act
CDC Centers for Disease Control
CE current electrode
CERCLA Comprehensive Environmental
Response, Compensation and
Liability Act of 1980
CERCLIS CERCLA Information System
CERI Center for Environmental
Research Information
CFR Code of Federal Regulations
CLP Contract Laboratory Program
9/93
CFA continuous flight auger
COC chain of custody
COD chemical oxygen demand
COE U.S. Army Corps of Engineers
CWA Clean Water Act
DO dissolved oxygen
DOJ U.S. Department of Justice
DOT U.S. Department of
Transportation
DQO data quality objectives
DRI direct-reading instruments
DNAPL dense, nonaqueous phase liquid
Eh oxygen-reduction potential
EM electromagnetic
EMSL-LV Environmental Monitoring
Systems Laboratory - Las
Vegas
EP,
tox
toxicity-extraction procedure
toxicity
EPA U.S. Environmental Protection
Agency
EPIC Environmental Photographic
Interpretation Center
ER electrical resistivity
ERP Emergency Response Plan
Acronyms and Abbreviations
-------
ERT EPA Emergency Response
Team
ERTS Earth Resources Technology
Satellite
EROS Earth Resources Observation
Systems
ESB EPA Environmental Services
Branch
ESD Environmental Services
Division
eV electron volt
FIFRA Federal Insecticide, Fungicide,
and Rodenticide Act
FIT field investigation team
FRP fiberglass reinforced plastic
FS feasibility study
FSP field sampling plan
GAC granular activated carbon
GC gas chromatograph
GC/MS gas chromatography/mass
spectrometry
gpm gallons per minute
GPR ground-penetrating radar
GWA Ground Water Act of 1987
HASP health and safety plan (see also
site safety plan)
HAZMAT hazardous materials team
HRS
hazard ranking system
HSL hazardous substance list
(previous term for target
compound list)
HSA hollow-stem auger
HSO health and safety officer (see
also SSC, SSHO, and SSO)
HSWA Hazardous and Solid Waste
Amendments (to RCRA, 1984)
HWS hazardous waste site
ICS incident command system
IDL instrument detection limit
IDLH immediately dangerous to life
and health
IP ionization potential
IR infrared (spectroscopy)
K hydraulic conductivity
LEL lower explosive limit
LNAPL light, nonaqueous phase liquid
LUST leaking underground storage
tank
MCL maximum contaminant level
MCLG maximum contaminant level
goal
MDL method detection limit
MSL mean sea level
m/sec meters per second
MHz megahertz
Acronyms and Abbreviations
9/93
-------
MS/MS mass spectrometry/mass
spectrometry
NEAR nonbinding preliminary
allocation of responsibility
NCIC National Cartographic
Information Center
NCP National Oil and Hazardous
Substances Pollution
Contingency Plan
NEIC National Enforcement
Investigation Center
NIOSH National Institute of
Occupational Safety and Health
NIPDWR National Interim Primary
Drinking Water Regulations
NOAA National Oceanic and
Atmospheric Administration
n.o.s. not otherwise specified (used
in shipping hazardous material)
NPDES National Pollutant Discharge
Elimination System
NPL National Priorities List
NRC Nuclear Regulatory
Commission
NSF National Sanitation Foundation
NTIS National Technical Information
Service
NWS National Weather Service
OERR EPA Office of Emergency and
Remedial Response
OHMTADS Oil and Hazardous Materials
Technical Assistance Data
System
OSHA Occupational Safety and Health
Administration
OSWER EPA Office of Solid Waste and
Emergency Response
OVA organic vapor analyzer (onsite
organic vapor monitoring
device)
OWPE EPA Office of Waste Programs
Enforcement
PAC powdered activated carbon
PAH polycyclic aromatic
hydrocarbons
PCB polychlorinated biphenyls
PCDD polychlorinated dibenzo-p-
dioxin
PCDF polychlorinated dibenzofuran
PCP pentachlorophenol
PE potential electrode
PEL permissible exposure limit
PID photoionization detector
PO project officer (EPA)
POHC principle organic hazardous
constituent
POM polycyclic organic matter
POTWs publicly owned treatment
works
ppb parts per billion
9/93
Acronyms and Abbreviations
-------
PPE personal protective equipment
ppm parts per million
PRP potentially responsible party
psig pounds per square inch gauge
PVC polyvinyl chloride
QA quality assurance
QA/QC quality assurance and quality
control
QAMS quality assurance management
staff
QC quality control
RA remedial action
RAS routine analytical services
RCRA Resource Conservation and
Recovery Act of 1978
RD remedial design
REM remedial planning
REM/FIT remedial planning/field
investigation team
RI/FS remedial investigation and
feasibility study
\
ROD record of decision
RPM EPA remedial project manager
RQ reportable quantity
RSPO remedial site project officer
RSCC Regional Sample Control
Center
SARA Superfund Amendments and
Reauthorization Act of 1986
SAS special analytical services
SCBA self-contained breathing
apparatus
SCS Soil Conservation Service
SDL sample detection limit
SOW A Safe Drinking Water Act
SI site inspection
SITE Superfund Innovative
Technology Evaluation
SM site manager
SOP standard operating procedure
SP spontaneous potential
SQG small quantity generator
SSC site safety coordinator
SVOC semivolatile organic
compound
SWDA Solid Waste Disposal Act
TAT technical assistance team
TCLP toxicity characteristic leaching
procedure
TEGD Technical Enforcement
Guidance Document
TDS total dissolved solids
TLV threshold limit value
TOC total organic carbon
Acronyms and Abbreviations
9/93
-------
TOH total organic halogen
TOX total organic halides
TSCA Toxic Substances Control Act
TSDF treatment, storage, and disposal
facility
TUHC total unburned hydrocarbons
UEL upper explosive limit
UMTRCA Uranium Mill Tailing Radiation
Control Act
USCG United States Coast Guard
USCS Unified Soil Classification
System
USDI U.S. Department of the Interior
USGS U.S. Geological Survey
UST underground storage tank
UV ultraviolet
VOA volatile organic analysis
VOC volatile organic compound
9/93 5 Acronyms and Abbreviations
-------
GLOSSARY
acre-foot
adsorption
advection
alluvium
an iso tropic
aquifer
aquifer test
aquitard
artesian
artificial recharge
artesian aquifer
bedload
enough water to cover 1 acre to a depth of 1 foot; equal to 43,560
cubic feet or 325,851 gallons
the attraction and adhesion of a layer of ions from an aqueous solution
to the solid mineral surfaces with which it is in contact
the process by which solutes are transported by the bulk motion of the
flowing groundwater
a general term for clay, silt, sand, gravel, or similar unconsolidated
material deposited during comparatively recent geologic time by a
stream or other body of running water as a sorted or semisorted
sediment in the bed of the stream or on its floodplain or delta, or as
a cone or fan at the base of a mountain slope
hydraulic conductivity ("K"), differing with direction
a geologic formation, group of formations, or a part of a formation
that contains sufficient permeable material to yield significant
quantities of groundwater to wells and springs. Use of the term
should be restricted to classifying water bodies in accordance with
stratigraphy or rock types. In describing hydraulic characteristics such
as transmissivity and storage coefficient, be careful to refer those
parameters to the saturated part of the aquifer only.
a test involving the withdrawal of measured quantities of water from,
or the addition of water to, a well (or wells) and the measurement of
resulting changes in head (water level) in the aquifer both during and
after the period of discharge or addition
a saturated, but poorly permeable bed, formation, or group of
formations that does not yield water freely to a well or spring
confined; under pressure sufficient to raise the water level in a well
above the top of the aquifer
recharge at a rate greater than natural, resulting from deliberate or
incidental actions of man
see confined aquifer
the part of the total stream load that is moved on or immediately above
the stream bed, such as the larger or heavier particles (boulders,
pebbles, gravel) transported by traction or saltation along the bottom;
the part of the load that is not continuously in suspension or solution
9/93
1
Glossary
-------
capillary zone
capture
coefficient of storage
cone of depression
confined
confined aquifer
confining bed
diffusion
discharge area
discharge velocity
dispersion
drawdown
effective porosity
negative pressure zone just above the water table where water is drawn
up from saturated zone into soil pores due to cohesion of water
molecules and adhesion of these molecules to soil particles. Zone
thickness may be several inches to several feet depending on porosity
and pore size.
the decrease in water discharge naturally from a ground-water
reservoir plus any increase in water recharged to the reservoir
resulting from pumping
the volume of water an aquifer releases from or takes into storage per
unit surface area of the aquifer per unit change in head
depression of heads surrounding a well caused by withdrawal of water
(larger cone for confined aquifer than for unconfmed)
under pressure significantly greater than atmospheric throughout and
its upper limit is the bottom of a bed of distinctly lower hydraulic
conductivity than that of the material in which the confined water
occurs
geological formation capable of storing and transmitting water in
usable quantities overlain by a less permeable or .impermeable
formation (confining layer) placing the aquifer under pressure
a body of "impermeable" material stratigraphically adjacent to one or
more aquifers
the process whereby particles of liquids, gases, or solids intermingle
as a result of their spontaneous movement caused by thermal agitation
an area in which subsurface water, including both groundwater and
water in the unsaturated zone, is discharged to the land surface, to
surface water, or to the atmosphere
an apparent velocity, calculated from Darcy's law, which represents
the flow rate at which water would move through the aquifer if it were
an open conduit (also called specific discharge)
the spreading and mixing of chemical constituents in groundwater
caused by diffusion and by mixing due to microscopic variations in
velocities within and between pores
the vertical distance through which the water level in a well is lowered
by pumping from the well or a nearby well
the amount of interconnected pore space through which fluids can
pass, expressed as a percent of bulk volume. Part of the total porosity
Glossary
9/93
-------
evapotranspiration
flow line
fluid potential
gaining stream
groundwater
groundwater divide
groundwater model
groundwater reservoir
groundwater system
head
heterogeneous/geological
formation
homogeneous
will be occupied by static fluid being held to the mineral surface by
surface tension, so effective porosity will be less than total porosity.
the combined loss of water from direct evaporation and through the
use of water by vegetation (transpiration)
the path that a particle of water follows in its movement through
saturated, permeable rocks (synonym: streamline)
the mechanical energy per unit mass of water or other fluid at any
given point in space and time, with respect to an arbitrary state of
datum
.a stream or reach of a stream whose flow is being increased by inflow
of groundwater (also called an effluent stream)
water in the zone of saturation
a ridge in the water table or other potentiometric surface from which
groundwater moves away in both directions normal to the ridge line
simulated representation of a groundwater system to aid definition of
behavior and decision-making
all rocks in the zone of saturation (see also aquifer)
a groundwater reservoir and its contained water; includes hydraulic
and geochemical features
combination of elevation above datum and pressure energy imparted
to a column of water (velocity energy is ignored because of low
velocities of groundwater). Measured in length units (i.e., feet or
meters).
characteristics varying aerially or vertically in a given system
geology of the aquifer is consistent; not changing with direction or
depth
hydraulic conductivity "K" volume flow through a unit cross-section area per unit decline in head
(measured in velocity units and dependent on formation characteristics
and fluid characteristics)
9/93
Glossary
-------
hydraulic gradient
change of head values over a distance
H, - H2
hydrograph
impermeable
infiltration
interface
intrinsic permeability
where:
H = head
L = distance between head measurement points
graph that shows some property of groundwater or surface water as a
function of time
having a texture that does not permit water to move through it
perceptibly under the head difference that commonly occurs in nature
the flow or movement of water through the land surface into the
ground
in hydrology, the contact zone between two different fluids
pertaining to the relative ease with which a porous medium can
transmit a liquid under a hydrostatic or potential gradient. It is a
property of the porous medium and is independent of the nature of the
liquid or the potential field.
hydraulic conductivity ("K") is the same regardless of direction
low velocity flow with no mixing (i.e., no turbulence)
a stream or reach of a stream that is losing water to the subsurface
(also called an influent stream)
in reference to groundwater, withdrawals in excess of natural
replenishment and capture. Commonly applied to heavily pumped
areas in semiarid and arid regions, where opportunity for natural
replenishment and capture is small. The term is hydrologic and
excludes any connotation of unsatisfactory water-management practice
(see, however, overdraft).
nonsteady state-nonsteady (also called unsteady state-nonsteady shape) the condition when the
isotropic
laminar flow
losing stream
mining
shape
rate of flow through the aquifer is changing and water levels are
declining. It exists during the early stage of withdrawal when the
water level throughout the cone of depression is declining and the
shape of the cone is changing at a relatively rapid rate.
Glossary
9/93
-------
nonsteady state-steady
shape
optimum yield
overdraft
pellicular water
perched
permeability
permeameter
piezometer
porosity
potentiometric surface
recharge
recharge area
safe yield
saturated zone
is the condition that exists during the intermediate stage of withdrawals
when the water level is still declining but the shape of the central part
of the cone is essentially constant
the best use of groundwater that can be made under the circumstances;
a use dependent not only on hydrologic factors but also on legal,
social, and economic factors
withdrawals of groundwater at rates perceived to be excessive and,
therefore, an unsatisfactory water-management practice (see also
mining)
water adhering as films to the surfaces of openings and occurring as
wedge-shaped bodies at junctures of interstices in the zone of aeration
above the capillary fringe
unconfined groundwater separated from an underlying body of
groundwater by an unsaturated zone
the property of the aquifer allowing for transmission of fluid through
pores (i.e., connection of pores)
a laboratory device used to measure the intrinsic permeability and
hydraulic conductivity of a soil or rock sample
a nonpumping well, generally of small diameter, that is used to
measure the elevation of the water table or potentiometric surface. A
piezometer generally has a short well screen through which water can
enter.
the ratio of the volume of the interstices or voids in a rock or soil to
the total volume
imaginary saturated surface (potential head of confined aquifer); a
surface that represents the static head; the levels to which water will
rise in tightly cased wells
the processes of addition of water to the zone of saturation
an area in which water that is absorbed eventually reaches the zone of
saturation
magnitude of yield that can be relied upon over a long period of time
(similar to sustained yield)
zone in which all voids are filled with water (the water table is the
upper limit)
9/93
Glossary
-------
slug-test
specific capacity
specific yield
steady-state
storage
storage coefficient "S"
storativity
sustained yield
transmissivity
vadose zone
an aquifer test made by either pouring a small instantaneous charge of
water into a well or by withdrawing a slug of water from the well
(when a slug of water is removed from the well, it is also called a
bail-down test)
the rate of discharge from a well divided by the drawdown in it. The
rate varies slowly with the duration of pumping, which should be
stated when known.
ratio of volume of water released under gravity to total volume of
saturated rock
the condition when the rate of flow is steady and water levels have
ceased to decline. It exists in the final stage of withdrawals when
neither the water level nor the shape of the cone is changing.
in groundwater hydrology, refers to 1) water naturally detained in a
groundwater reservoir, 2) artificial impoundment of water in
groundwater reservoirs, and 3) the water so impounded
volume of water taken into or released from aquifer storage per unit
surface area per unit change in head (dimensionless) (for confined,
S = 0.0001 to 0.001; for unconfined, S = 0.2 to 0.3)
the volume of water an aquifer releases from or takes into storage per
unit surface area of the aquifer per unit change in head (also called
coefficient of storage)
continuous long-term groundwater production without progressive
storage depletion (see also safe yield)
the rate at which water is transmitted through a unit width of an
aquifer under a unit hydraulic gradient
the zone containing water under pressure less than that of the
atmosphere, including soil water, intermediate vadose water, and
capillary water. Some references include the capillary water in the
saturated zone. This zone is limited above by the land surface and
below by the surface of the zone of saturation (i.e., the water table).
Also called the unsaturated zone or zone of aeration. According to
Freeze and Cherry (1979):
1. It occurs above the water table and above the capillary fringe.
2. The soil pores are only partially filled with water; the moisture
content 6 is less than the porosity n.
3. The fluid pressure;? is less than atmospheric; the pressure head \f/
is less than zero.
4. The hydraulic head h must be measured with a tensiometer.
Glossary
9/93
-------
5. The hydraulic conductivity K and the moisture content 6 are both
functions of the pressure head ^.
water table surface of saturated zone area at atmospheric pressure; that surface in
an unconfmed water body at which the pressure is atmospheric.
Defined by the levels at which water stands in wells that penetrate the
water body just far enough to hold standing water.
9/93 l Glossary
-------
Section 1
-------
GEOLOGY
PERFORMANCE OBJECTIVES
At the end of this lesson, participants will be able to:
• Define the Doctrine of Uniformitarianism
• Describe the three basic rock types and evaluate each as
aquifers
• Describe the rock forming processes found on the Rock
Cycle diagram
• Identify the medium responsible for the erosion and transport
of sediments
• Describe the following depositional environments:
Alluvial fans
Braided streams
Meandering streams
Coastal (deltaic, interdeltaic, barrier island complexes)
Wind-blown deposits
Carbonates.
-------
GEOLOGY
NOTES
MW1
720 It
MW3
718.25 «
9/93
Geology
-------
NOTES
Doctrine
of
Uniformitarianism
"The Present is the
Key to the Past"
James Mutton, 1785
THE ROCK CYCLE
Deposition
Transport
t
Weathering
Igneous rocks
X
Melting
Unification
\
Sedimentary rocks
\
Metamorphism
S
Metamorphic rocks
s
Geology
9/93
-------
NOTES
SEDIMENTATION
• Erosion processes (weathering)
• Transport agents
• Deposition
• Lithification
EROSION PROCESSES
• Wind
• Ice
• Water
• Biology
• Gravity
TRANSPORT AGENTS
• Wind
• Ice
• Water
• Biology
• Gravity
9/93
Geology
-------
NOTES
DEPOSITION
• Wind
• Ice
• Water
• Gravity
LITHIFICATION
Cementation
Diagenesis
TYPES OF CEMENT
• Silica
• Iron oxides
• Kaolinite
• Montmorillonite
• Illite
• Calcite (aragonite)
Geology
9/93
-------
SEDIMENTARY ROCKS
Composed of particles of any rock type
- "Pores" form during deposition
Most aquifers are sedimentary rocks
SEDIMENTARY ROCKS
NOTES
Limestone
Shale
Sandstone
Coal
Dolomite
Siltstone
Conglomerate
Evaporite
METAMORPH1SM
• Recrystallization
• "Earth's sweat"
9/93
Geology
-------
NOTES
METAMORPHIC ROCKS
PRESSURE
METAMORPHIC ROCKS
Marble Slate
Quartzite Phyllite
Gneiss Schist
"EARTH"S SWEAT"
Geology
9/93
-------
MELTING/MAGMA
NOTES
IGNEOUS ROCKS
• Intrusive
e.g., granite
• Extrusive
e.g., basalt
IGNEOUS ROCKS
Gabbro Basalt
Granite Rhyolite
9/93
Geology
-------
NOTES
INTRUSIVE IGNEOUS
ROCK BODIES
Dikes'
•;:; '••• '•••• Batholith
Adapted from General Biology by Robert Foster. 1969. p 63.
SEDIMENTARY ROCKS
Composed of particles of any rock type
- "Pores" form during deposition
Most aquifers are sedimentary rocks
ROCK TYPE
ENVIRONMENT
Conglomerate Landslide, alluvial fan
Sandstone Rivers, streams, beaches,
deltas, dunes, sand bars
Clay/shale Lagoon, lake, flood plain,
deeper ocean
Limestone Coral reef, atoll,
deeper ocean
Geology
9/93
-------
CRITERIA TO DEFINE
DEPOSITIONS
ENVIRONMENTS
NOTES
LONGITUDINAL PROFILE
A Alluvial and landslide
B Braided stream
M Meandering stream
C Coastal
if Stream headwaters
L (length)
A'*
T
B
H (height)
I Mouth of
^ stream
0/0?
Geology
-------
NOTES
CRITERIA
• Longitudinal channel profile
• Median channel-grain size
• Sphericity/sorting
CRITERIA
• Penetration of stream
• Width-to-depth ratio
• Degree of sinuosity
LONGITUDINAL
CHANNEL PROFILE
Geology
10
9/93
-------
LONGITUDINAL PROFILE
A Alluvial and landslide
B Braided stream
M Meandering stream
C Coastal
• Stream headwaters
-L(length)-
!lP'o///e
t
I (height)
I Mouth of
TX sti
cean
STREAM GRADIENTS
NOTES
High«-
-> Low
MEDIAN
CHANNEL-GRAIN SIZE
9/93
11
Geology
-------
NOTES
MEDIAN CHANNEL-GRAIN SIZE
Small
RELATIONSHIP OF
1000
% 100
"E
o
> 10
o
0
^ 1.0
0.1
Size o.c
(mm) Cl
STREAM VELOCITY
<^p:
" Transpor
/
i /.
01 0.01
ay Silt
erosion \ '. \ '. '.~~ ' s/ /
'// / / /// / •
//:::::•::
tation /• '•'•'•'•'•'•'•'•'•'•'•'•
Xx ;: Deposition ;:
/S'. '.'.'.'.'.'.'.'':
/ /. . ....................
f /
S .............
: :i :::::: :i ::::: ::p: ::::::
0.1 1.0 10 100
Sand Gravel
SPHERICITY/SORTING
Geology
12
9/93
-------
SPHERICITY
Angular <-
Rounded
fc *
SORTING
Poor
fc ft
Well
a-.*.-
x.» -
PENETRATION OF
STREAM
NOTES
9/93
13
Geology
-------
NOTES
STREAM CHANNEL
Penetration
Shallow
Deep
WIDTH-TO-DEPTH RATIO
STREAM CHANNEL
Width-to-Depth Ratio
High *-
-» Low
Geology
14
-------
DEGREE OF SINUOSITY
STREAM CHANNEL
Sinuosity
Low <-
-> High
:::::::::::::::: M :::::: « .>« ucean
C ; ; ;
DEPOSITIONAL
ENVIRONMENTS
NOTES
9/93
15
Geology
-------
NOTES
DEPOSITIONS
ENVIRONMENTS
• Alluvial fan
• Braided stream
Meandering stream
Coastal deposits
Wind-blown deposits
ALLUVIAL FAN
Geology
16
9/93
-------
BRAIDED STREAM
NOTES
9/93
17
Geology
-------
NOTES
MEANDERING STREAM
COASTAL DEPOSITS
Geology
18
9/93
-------
NOTES
WIND-BLOWN DEPOSITS
9/93
19
Geology
-------
NOTES
CARBONATES
• Limestones
• Dolomites
EVAPORITES
• Carbonates
• Sulfates
• Chlorides
GLACIATION
Geology
20
9/93
-------
NOTES
GLACIERS/FREEZE-THAW
• Weathering and transport
• Large scale changes
• Poor to excellent sorting
(e.g., glacial till and outwash)
PROCESSES OF GLACIATION
• Erosion
• Transportation
• Deposition
9/93
21
Geology
-------
Geometry of Sandstone Reservoir Bodies1
Abstract Natural underground reiervoin capable of
containing water, petroleum, and goiej include sand-
ilonei, limeitonei, dolomilei. and fractured rocki of vari-
oui typel. Comprehensive research ond exploration el-
forti by the petroleum induilry hove revealed much aboul
Ihe character ond diilribution of carbonate rockj (lime-
ilonei ond dolomilei) ond londilonei. Porosity ond per-
meability of the depoiili ore criteria (or determining
their efficiency 01 re»ervoiri for fluids. Trends ol certain
tondilonei are predictable. Furthermore, sandstone res-
ervoin hove been letl effected than carbonate reiervoin
by postdepositionol cementation ond compaction. Frac-
ture porosity has received lest concentrated study; hence,
we know lest obout this type of reservoir. The discussion*
in thii poper ore confined lo sandstone reservoirs.
The principal sandstone-generating environmenli ore
(1) fluvial environments such 01 olluvial ions, braided
streams, ond meandering streams; (2) distributory-chonnel
ond dello-fronl environments of various types of deltas;.
(3) coastal barrier islands, tidal channels, ond chenier
plains; (4} desert ond coastal eolion plains; end (5)
deeper marine environments, where Ihe sands ore dis-
tributed by both normal ond density currents.
The alluvial-fan environment is characterized by flash
floods ond mudflows or debris flows which deposit the
coarsest ond most irregular sand bodies. Braided streams
hove numerous shallow channels separated by brood
sandbars; lateral channel migration results in Ihe deposi-
tion of Ihin, lenticular tond bodies. Meandering streams
migrate within belli 20 timei the channel width ond
deposit two very common types of sond bodies. The
processes ol bank-coving ond point-bar accretion result
in lateral channel migration and the formation of sand
bodies (point bars) within each meander loop. Natural
cul-oRi ond channel diversions result in the abandon-
ment of individual meanders ond long channel seg-
ments, respectively. Rapidly abandoned channels are
filled with some sond but predominantly with fine-grained
sediments (clay plugs), whereas gradually abandoned
channels ore filled mainly with sands ond silts.
The most common sandstone reservoirs ore of deltaic
origin. They ore laterally equivalent to fluvial sands ond
prodelto ond marine cloys, ond they consist of two types:
delta-front or fringe sands ond abandoned dislributory-
chonnel sands. Fringe sands are sheeltike, ond their rand-
word margins are abrupt (against organic cloys of the
deltaic plain). Seaward, these sands grade into the
finer prodeUo ond marine sediments. Distributary-channel
sandstone bodies are narrow, they hove abrupt bosol
contacts, and they decrease in groin size upward. They
cut into, or completely through, Ihe fringe sands, ond
olso connect with the upstream fluvial sands or braided
or meandering streams.
Some of the more porous ond permeable sandstone
reservoirs ore deposited in the cooslot tnlerdeltoic realm
ol sedimentation. They consist of well-sorted beoch and
shorefoce sands associated with barrier islands ond tidal
channels which occur between barriers. Harrier land
bodies orr long and narrow, are aligned parallel with
RUFUS J. I.BLANC1
Houston, Texas 77001
the coastline, and are characterized by on upward in-
crease in grain size. They ore flanked on the landward
side by lagoonol cloys and on the opposite side by
marine clays. Tidal-channel sond bodiei hove obrvpt
basal contacts and range in groin size from coarse ol
Ihe base lo fine ol the lop. laterally, they merge with
barrier sands ond grade into the finer sediments of
tidal deltas ond mud flats.
The molt porous and permeable sandstone reservoirs
ore products of wind activity in coastal ond desen re-
gions. Wind-laid (eolian) sands ore typically very well
sorted ond highly crossbedded, and they occur ai ex-
tensive sheets.
Marine sandstones are those associated with normal-
marine processes of the continental shelf, slope, and
deep and those due to density-current origin (turbidites).
An important type of normal-marine sond is formed
during marine transgressions. Although these sondi ore
extremely thin, they are very distinctive ond widespread,
hove sharp updip limits, ond grade seaward into marine
shales. Delta-fringe and borrier-shorefoce sonds ore two
other types of shallow-marine sands.
Turbidiles have been interpreted to be associated
with submarine canyons. These sonds are transported
from nearshore environments seaward through canyons
ond are deposited on submarine fans in deep marine
basins. Other turbidiles form as o result of slumping of
deltaic focies at shelf edges. Turbidite sands ore usually
associated with thick marine shales.
1 Manuscript received, March 17, 1972.
'Shell Oil Company. This paper is based on the
writer's 30 years of experience in studies of modem
and ancient clastic sediments—from 1941 to 1948, with
the Mississippi River Commission, under the guidance
of H. N. FJsJc, and, since August 1948, with Shell
Development Company and ShelJ Oil Company.
The writer is grateful lo Shell Oil Company for per-
mission to publish this paper, and he is deeply indebted
to Alan Thomson for his critical review of the manu-
script; he is also grateful to Nick W. Kusakis, John
Bush, Dave C. Fogt. Gil C. Flanagan, and George F.
Korenek for assistance in the preparation of illustra-
tions and reference material; to Aphrodite Mamoulides
and Bernice Melde for their library assistance; to Dar-
leen Vanderford for typing the manuscript, and to Judy
Breeding for her editorial assistance.
Numerous stimulating discussions of models of clas-
tic sedimentation and the relationship of sedimentary
sequences to deposition*] processes were held with
Hugh A. Bernard and Robert H. Nanz, Jr., during the
late 1940s and 1950s, when we were closely associated
with Shell's early •exploration research effort. The
writer is particularly indebted lo these two men for
their numerous contributions, many of which are in-
cluded in this paper.
The writer also wishes to tbanlc W. B. Bull, Univer-
sity of Arizona, for hij valuable suggestions concerning
the alluvial-fan model of clastic sedimentation.
133
Reproduced by Permission
-------
134
Rufus J. LeBlanc
INTRODUCTION
Important natural resources such as water,
oil, gas, and brines are found in underground
reservoirs which are composed principally of
the following types of rocks: (1) porous sands,
sandstones, and gravels; (2) porous limestones
and dolomites; and (3) fractured rocks of vari-
ous types. According to the 1971 American Pe-
troleum Institute report on reserves of crude oil
and natural gas, sandstones are the reservoirs
for about 75 percent of the recoverable oil and
65 percent of the recoverable gas in the United
States. It is also estimated that approximately
90 percent of our underground water supply
comes from sand and gravel (Walton, 1970).
Sandstone and carbonate (limestone and do-
lomite) reservoirs have been intensively studied
during the past 2 decades; consequently, the
general characteristics and subsurface distribu-
tion of these two important types of reservoirs
are relatively well known in numerous sedimen-
tary basins. The factors which control the ori-
gin and occurrence of fracture porosity have
received less attention; thus, our knowledge
and understanding of this type of reservoir are
more limited.
The detection of subsurface porosity trends
within sedimentary basins was recognized by the
petroleum industry as one of its most signifi-
cant problems, and for the past 2 decades it has
addressed itself to a solution through extensive
research. Largely as a result of this research,
which is summarized below, our ability to de-
termine trends of porous sedimentary rocks has
progressed noticeably, especially during the
past 10 yean.
The amount of porosity and permeability
present within sedimentary rocks and the ge-
ometry of porous rock bodies are controlled
mainly by two important factors: (1) the envi-
ronmental conditions under which the sedi-
ments were deposited and (2) the postdeposi-
tional changes within the rocks as a result of
burial, compaction, and cementation. Postde-
positional diagenetic processes have less effect
on the porosity and permeability of sands and
sandstones than they have on carbonate sedi-
ments; consequently, porosity trends are signifi-
cantly more predictable for sandstones than for
limestones and dolomites.
Organization of paper—The following two
parts of this paper give a brief historical sum-
mary of the early research on clastic sediments
and present a classification of environments of
deposition and models of clastic sedimentation.
A resuml of significant studies of modern clas-
tic sediments—mainly by the petroleum indus-
try, government agencies, and universities—
follows. The main part of the paper concerns
the sedimentary processes, sequences, and ge-
ometry of sand bodies which characterize each
of the following models of clastic sedimenta-
tion: alluvial fan, braided stream, meandering
stream, deltaic (birdfoot-lobate and cuspate-
arcuate), coastal interdeltaic (barrier island
and chenier plain), and marine (transgressive,
submarine canyon, and fan).
HISTORICAL SUMMARY OF EARLY RESEARCH ON
MODERN CLASTIC SEDIMENTS
Geologists are now capable of interpreting
the depositional environments of ancient sedi-
mentary facies and of predicting clastic poros-
ity trends with a reasonable degree of accuracy
(Peterson and Osmond, 1961; Potter, 1967;
Rigby and Hamblin, 1972; Shelton, 1972).
This capability stems from the extensive re-
search conducted on Holocene sediments by
several groups of geologists during the past 3
decades. Conditions which led to this research,
and the most significant studies of clastic sedi-
mentation which provided the models, criteria,
and concepts necessary to make environmental
interpretations, are summarized below.
During the late 1930s and early 1940s, pe-
troleum geologists became aware that improved
methods of stratigraphic interpretations were
badly needed, and that knowledge and geologic
tools necessary to explore for stratigraphic
traps were inadequate. A detailed study made
by the Research Committee of The American
Association of Petroleum Geologists on the re-
search needs of the industry ultimately led to
the establishment of geologic research depart-
ments by major oil companies. By 1948, explo-
ration research by the oil industry was iaHts
early stages, and expansion proceeded /rapidly
thereafter.
Meanwhile, some very significant develop-
ments were occurring at Louisiana State Uni-
versity. H. V. Howe and R. J. Russell, together
with their graduate students, had already pub-
lished several Louisiana Geological Survey bul-
letins summarizing their pioneer work on the
late Quaternary geology of southern Louisiana
(Howe and Moresi, 1931, 1933; Howe el al.,
1935; Russell, 1936). Their early work on the
Mississippi deltaic plain and the chenier plain
of southwestern Louisiana is considered to be
the beginning of the modern environmental ap-
proach to stratigraphy. Fisk became fascinated
-------
Geometry of Sandstone Reservoir Bodies
135
with the Howe and Russell approach, and he
applied results of their research to his study of
Tertiary sediments. The work of Fisk (1940)
IB central Louisiana, which included a study of
the lower Red River Valley and part of the
Mississippi Valley, attracted the attention of
General Max Tyler, president of the Mississippi
River Commission in Vicksburg. General Tyler
engaged Fisk as a consultant and provided him
with a staff of geologists to conduct a geologic
investigation of the lower Mississippi River al-
luvial valley.
The Fisk (1944) report on the Mississippi
Valley, which now has become a classic geo-
logic document, established the relations be-
tween alluvial environments, processes, and
character of sediments. The AAPG, recogniz-
ing the significance of this contribution, re-
tained Fisk as Distinguished Lecturer, and the
results and significance of his work became
widely known. One of his most significant con-
tributions came when, as the petroleum indus-
try was getting geologic research under way, he
was selected by a major oil company to direct
its geologic research effort in Houston.
By 1950, a few major oil companies were
deeply involved in studies of recent sediments.
However, the small companies did not have
staff and facilities to conduct this type of re-
search, and American Petroleum Institute Proj-
ect 51 was established for the purpose of con-
ducting research on recent sediments of the
Gulf Coast. Scripps Institution of Oceanogra-
phy was in charge of the project, which contin-
ued for 8 years. Results of this research were
available to all companies (Sbepard el al,
1960).
While the petroleum industry was conduct-
ing "in-house" research and supporting the API
project, some significant research was being
done by the U.S. Waterways Experiment Sta-
tion in Vicksburg, Mississippi, and by the new
Coastal Studies Institute at Louisiana State Uni-
versity under the direction of R. J. Russell.
These two groups conducted detailed studies of
recent sediments for many years, and results
were made available to the petroleum industry.
By 1955, a fairly good understanding of pro-
cesses of sedimentation and character of related
sediments in several depositional environments
had been acquired. Although the application of
this wealth of knowledge to operational prob-
lems was very difficult, some useful applica-
tions nevertheless had been made by the middle .
1950s, and it was generally agreed that the ini-
tial research effort was successful.
Since 1955, geologists all over the world
have become involved in studying recent sedi-
ments and applying the results to research on
older rocks. Geologists with the U.S. Geologi-
cal Survey and several universities have con-
ducted studies of alluvial fans, braided streams,
and eolian deposits; and the oceanographic in-
stitutions, such as Scripps, Woods Hole, and
Lamont, have investigated deep-marine • sedi-
ments on a worldwide basis. Publication of pa-
pers on clastic sedimentation has been increas-
ing rapidly. The first textbook on the geology
of recent sediments cites more than 700 refer-
ences, 75 percent of which have appeared since
1955 (Kukal, 1971). Many of these contribu-
tions, considered to be most significant to the
current understanding of clastic sediments, are
cited in this paper.
MODELS AND ENVIRONMENTS OF
CLASTIC SEDIMENTATION
The realm of clastic sedimentation can be di-
vided into several conceptual models, each of
which is characterized by certain depositional
environments, sedimentary processes, se-
quences, and patterns. What are considered to
be some cf the most common and basic models
and environments9 of clastic sedimentation, ar-
ranged in order from the periphery to the cen-
ter of a depositional basin, are listed below and
are shown on Figures 1—4.
Continental
Alluvial (fluvial) models
Alluvial fan
Braided stream
Meandering stream (includes flood basins be-
tween meander belts)
Eolian (can occur at various positions within con-
tinental and transitional models)
Transitional
Deltaic models
Birdfoot-lobate (fluvial dominated)
Cuspate-arcuate (wave and current dominated)
Estuarine (with strong tidal influence)
Coastal-interdellaic models
Barrier-island model (includes barrier islands,
lagoons behind barriers, tidal channels, and
tidal deltas)
Cbenier-plain model (includes mud flats and
cheniers)
Marine
(Note: Sediments deposited in shallow-marine en-
vironments, such as deltas and barrier islands, are
1 The classification of depositiona] environments pre-
sented herein was initially developed by the writer and
his colleague, Hugh A. Bernard, during the early 1950s
(LeBlanc and Bernard, 1954) and was recently modi-
fied (Bernard and LeBlanc, 1W5). For other classifi-
calions, refer to Laporte (1968), Selley (1970), Crosby
(1972), and Kukal (1971).
-------
136
Rufus J. LeBlonc
MINtAV
AUVVIAt 'AN
tIAIDIO IIIIAH
Ml ANOf *>NG ItltAM
IO1IAN
TION»l
DIIIAIC
COAtlAl
1MKIOIIIAIC
CMfNIII flAIW
lAlllll.tUANC
COM»lf •
Fio. 1—Some common models of clastic sedimentation. See Figures 2-4 for details.
included under the transitional group of environ-
ments.)
Transgressive-marine model
Submarine-canyon and submarine-fan model
RESUME OF SIGNIFICANT STUDIES OF
MODERN CLASTIC SEDIMENTATION
Alluvial Fans
Although much work has been done on allu-
vial fans, only a few papers discuss the relation
of sedimentary sequences to depositional pro-
cesses. Some of the more important contribu-
tions are by Rickmers (1913), Pack (1923),
Blackwelder (1928), Eckis (1928), Blissen-
bach (1954), McKee (1957), Beaty (1963),
Bull (1962, 1963, 1964, 1968, 1969, 1971),
Hoppe and Ekman (1964), Windir (1965),
Anstey (1965), Denny (1965, 1967), Legget
ei al. (1966), and Hooke (1967).
Braided Streams
Early papers on braided streams concerned
channel patterns, origin of braiding, and physi-
cal characteristics of braided streams. Signifi-
cant studies of this type were conducted by
Lane (1957), Leopold and Wolman (1957),
Cbein (1961), Krigstrom (1962), Fahnestock
(1963), andBrice (1964).
The relatively few papers on the relation of
braided-stream deposits to depositional pro-
cesses did not appear until tbe 1960s. Doeglas
(1962) discussed braided-stream sequences of
the Rhone River of France, and Ore (1963,
1965) presented some criteria for recognition
of braided-stream deposits, based on tbe study
of several braided streams in Wyoming, Colo-
rado, and Nebraska. Fahnestock (1963) de-
scribed braided streams associated with a gla-
cial outwash plain in Washington. More re-
cently, Williams and Rust (1969) discussed the
sedimentology of a degrading braided river in
the Yukon Territory, Canada. Coleman (1969)
presented results of a study of the processes
and sedimentary characteristics of one of the
largest braided rivers, the Brahmaputra of
Bangla Desh (formerly East Pakistan). N.
Smith (1970) studied tbe Platte River from
Denver, Colorado, to Omaha, Nebraska, and
used the Platte model to interpret Silurian
braided-stream deposits of the Appalachian re-
gion. Waechter (1970) has recently studied the
braided Red River in the Texas Panhandle, and
Kessler (1970, 1971) has-investigated the Ca-
nadian River in Texas. Boothroyd (1970) stud-
ied braided streams associated with glacial out-
wash plains in Alaska.
-------
ENVIRONMENTS
DEPOSITIONAL MODELS
AllUVIAl
(FlUVIAl)
Z
LU
Z
>-
Z
o
AUUVIAl
FANS
(APEX, MIDDLE
& BASE OF FAN)
BRAIDED
STREAMS
MEANDERING
STREAMS
(ALLUVIAL
VAllEY)
STREAM
FIOWS
VISCOUS
HOWS
MEANDER
BEITS
FIOODBASINS
CHANNELS
SHEETFIOOOS
•SIEVE DEPOSITS-
DEBRIS FLOWS
MUDFLOWS
CHANNELS
(VARYING SIZES)
LONGITUDINAL
TRANSVERSE
CHANNELS
NATURAL LEVEES
POINT BARS
STREAMS, LAKES
& SWAMPS
O
»
o
i*
a
3
a.
w
5"
3
n
30
It
2
o
OB
O
a.
EOLIAN
COASTAL DUNES
DUNES
DESERT DUNES
OTHER DUNES
TYPES:
TRANSVERSE
SEIF
(LONGITUDINAL)
BARCHAN
PARABOLIC
DOME-SHAPED
Flo. 1—Alluvial (fluvial) and eolian environments and models of clastic sedimentation.
-------
ENVIRONMENTS
DEPOSITIONAL MODELS
Z
o
OO
2
DELTAIC
UPPER
DEITAIC
PIAIN
LOWER
DEITAIC
PLAIN
FRINGE
DISTAL
MEANDER
BELTS
FIOODBASINS
DISTRIBUTARY
CHANNELS
INTER-
DISTRIBUTARY
AREAS
INNER
OUTER
CHANNELS
NATURAL LEVEES
POINT BARS
STREAMS,
LAKES
& SWAMPS
CHANNELS
NATURAL LEVEES
MARSH,
LAKES,
TIDAL CHANNELS
4 TIDAL FLATS
RIVER-MOUTH
BARS
BEACHES 8.
BEACH RIDGES
TIDAL FLATS
BIRDFOOT-LOBATE
DELTA
CUSPATE-ARCUATE
DELTA
CSTUARINE DELTA
•*•
O
Fie. 3—Deltaic environments and models of clastic sedimentation.
-------
Geometry of Sandstone Reservoir Bodies
139
Meandering Streams
H. N. Fiik's studies of the Mississippi allu-
vial valley, conducted for the Mississippi River
Commission during the period 1941—48, repre-
sent the first significant contribution on mean-
dering stream environments and deposits. This
pioneer effort provided geologists with knowl-
edge of the fundamental processes of alluvial-
valley sedimentation. Another study of a mean-
dering stream, the Connecticut River, and its
valley was made by Jahns (1947). Important
work on alluvial sediments deposited by mean-
dering streams was also done. by Sundborg
(1956) in Sweden, and by Frazier and Osanik
(1961), .Bernard and Major (1963), and
Harms et al. (1963) on the Mississippi, Brazos,
and Red River point bars, respectively. Thus,
by 1963 the general characteristics of point-bar
sequences, and the closely related abandoned-
channel and flood-basin sequences, were suffi-
ciently well established to permit geologists to
recognize this type of sedimentary deposit in
outcrops and in the subsurface.
Other important contributions were made by
Allen (1965a) on the origin and characteristics
of alluvial sediments, by Simons el al. (1965)
on the flow regime in alluvial channels, by Ber-
nard et al. (1970) on the relation of sedimen-
tary structures to bed form in the Brazos valley
deposits, and by McGowen and Gamer (1970)
on coarse-grained point-bar deposits.
Deltas
The early work by W. Johnson (1921, 1922)
on the Fraser delta, Russell (1936) on the Mis-
sissippi delta, Sykes (1937) on the Colorado
delta, and Fisk (1944) on the Mississippi delta
provided a firm basis for subsequent studies of
more than 25 modern deltas during the late
1950s and the 1960s.
Fisk continued bis studies of the Mississippi
delta for more than 20 years. His greatest con-
tributions were concerned with the delta frame-
work, the origin and character of delta-front
sheet sands, and the development of bar-finger
sands by seaward-migrating rivermouth bars.
Scruton's (1960) paper on delta building
and the deltaic sequence represents results of
API Project 51 on the Mississippi delta. Addi-
tional research on Mississippi delta sedimenta-
tion, sedimentary structures, and mudlumps
was reported by Welder (1959), Morgan
(1961), Morgan et al. (1968), Coleman et al.
(1964), Coleman (1966b), Coleman and Gagli-
ano (1964, 1965), and also by Kolb and Van
Lopik (1966). Coleman and Gagliano (1964)
also discussed and illustrated processes of
cyclic sedimentation. The most recent papers
on the Mississippi delta are by Frazier (1967),
Frazier and Osanik (1969), and Gould
(1970).
Studies of three small birdfoot deltas of
Texas—the Trinity, Colorado, and Guadalupe
—were made by McEwen (1969), Kanes
(1970), and Donaldson (1966), respectively.
In addition, Donaldson et al. (1970) presented
a summary paper on the Guadalupe delta.
These four contributions are valuable because
each one presents photographs and logs of
cores of complete deltaic sequences.
European geologists associated with the pe-
troleum industry and universities also have
made valuable contributions to our understand-
ing of deltas. Kruit (1955) and Lagaaij and
Kopstein (1964) discussed their research on
the Rhone delta of southern France, Allen
(1965c, 1970) summarized the geology of the
Niger delta of western Africa, and van Ande)
(1967) presented a resum6 of the work done on
the Orinoco delta of eastern Venezuela. More
recently, the Po delta of Italy was studied by B.
Nelson (1970) and the Rhone delta of southern
France by Oomkens (1970).
Other recent contributions on modern deltas
are by Coleman et al. (1970) on a Malaysian
delta, by R. Thompson (1968) on the Colo-
rado delta in Mexico, and by Bernard et al.
(1970) on the Brazos delta of Texas.
The deltaic model is probably the most com-
plex of the clastic models. Although additional
research is needed on this aspect of sedimenta-
tion, the studies listed have provided some
valuable concepts and criteria for recognition
of ancient deltaic facies.
Coastal-Interdeltaic Sediments
Valuable contributions to our knowledge of
this important type of sedimentation have been
made by several groups of geologists. In the
Gulf Coast region, the extensive Padre Island-
Laguna Madre complex was studied by Fisk
(1959), and the cbenier plain of southwestern
Louisiana was studied by Gould and McFarlan
(1959) and Byrne et al. (1959). The Galves-
ton barrier-island complex of the upper Texas
coast was investigated mainly by LeBlanc and
Hodgson (1959), Bernard et al. (1959, 1962),
and Bernard and LeBlanc (1965).
Among the impressive studies made by Euro-
peans during the past 15 years are those by van
Straaten (1954), who presented results of very
-------
Z
O
V—
10
Z
COASTAL
INTER-
DELTAIC
ENVIRONMENTS
COASTAL
PLAIN
(SUBAERIAl)
SUBAQUEOUS
BARRIER
ISLANDS
CHENIER
PLAINS
TIDAL
LAGOONS
TIDAL
CHANNELS
SMALL
ESTUARIES
BACK 8AR.
BARRIER.
BEACH.
BARRIER FACE.
SPITS 4 FLATS.
WASHOVER FANS
BEACH
& RIDGES
TIDAL FLATS
TIDAt FLATS
TIDAL DELTAS
SHOALS
& REEFS
DEPOSITIONAl MODELS
BARRIER IS.
COMPLEX
CHENIER
PLAIN
so
c
*
OB
Q
3
UJ
Z
SHALLOW
MARINE
DEEP
MARINE
INNER
SHELF.
(NERITIC)
MIDDLE
SHOALS
& BANKS
OUTER
CANYONS
f ANS (DELIAS)
SLOPE &
ABYSSAL
TRENCHES &
TROUGHS
SHALLOW
MARINE
DEEP
MARINE
FIG. 4—Coaslal-inlerdellaic and marine environments and models of clastic sedimentation.
-------
Geometry of Sandstone Reservoir Bodies
141
significant work on tidal flats, tidal channels,
and tidal deltas of the northern Dutch coast,
and by Horn (1965) and H. E. Reineck
(1967), who reported on the barrier islands and
tidal flats of northern Germany.
During the past several years, a group of ge-
ologists has conducted interesting research on
the coastal-interdeltaic complexes which char-
acterize much of the U.S. Atlantic Coast re-
gion. Hoyt and Henry (1965, 1967) published
several papers on barriers and related features
of Georgia. More recently, results of studies at
the University of Massachusetts on recent
coastal environments of New England were re-
ported by Daboll (1969) and by the Coastal
Research Group (1969).
In addition, Curray et al. (1969) describsd
sediments associated with a strand-plain barrier
in Mexico, and Potter (1967) summarized the
characteristics of barrier-island sand bodies.
Eolian Sand Dunes
Prior to the middle 1950s, eolian deposi-
tional environments were studied principally by
European geologists (Cooper, 1958). Since
that time, the coastal sand dunes of the Pacific,
Atlantic, and Gulf coasts of the United States,
as well as the desert dunes of the United States
and other countries, have been investigated by
university professors and by geologists with the
U.S. Geological Survey. Some of the most sig-
nificant contributions, especially those con-
cerned with dune stratification, are discussed in
the section on the eolian model of clastic sedi-
mentation.
Marine Sediments
Early work on modern marine sands, exclu-
sive of those deposited adjacent to and related
to interdeltaic and deltaic depositional environ-
ments, was conducted largely by scientists asso-
ciated with Scripps, Woods Hole, and Lament
oceanographic departments. Several aspects of
marine sediments were discussed by Trask et al.
(1955), and the recent sands of the Pacific
Ocean off California were studied by Revelle
and Shepard (1939), Emery et al. (1952), and
Emery (1960a). Stetson (1953) described the
northwestern Gulf of Mexico sediments, and
Ericson et al. (1952, 1955)-and Heezen et al.
(1959) investigated the Atlantic Ocean sedi-
ments. Later, Curray (1960), van Andel
(1960), and van Andel and Curray (1960) re-
ported results of the API project on the Gulf of
Mexico. A few years later, results of the API
project studies on the Gulf of California were
reported by van Andel (1964) and van Andel
and Shor (1964). Menard (1964) discussed
sediments of the Pacific Ocean. For a more
complete list of references to studies of recent
marine sands, the reader is referred to Kuenen
(1950), Guilcher (1958), Shepard et al.
(1963), and Kukal (1971).
Much of the early research on modern ma-
rine environments was devoted to submarine
canyons, fans, and basins considered by the in-
vestigators to be characterized mainly by tur-
bidity-current sedimentation. Several scientists
affiliated with Scripps and the University of
Southern California published numerous papers
on turbidites which occur in deep marine ba-
sins.
It is extremely difficult to observe the pro-
cesses of turbidity-current sedimentation under
natural conditions; consequsntly, the relations
between sedimentary sequences and processes
are still relatively poorly understood. Much of
the research dealing with turbidity currents has
been concerned with theory, laboratory models,
and cores of deep-water sediments deposited by
processes which have not been observed.
ALLUVIAL-FAN MODEL or CLASTIC
SEDIMENTATION
Occurrence and General Characteristics
Alluvial fans occur throughout the world,
adjacent to mountain ranges or high hills. Al-
though they form under practically all types of
climatic conditions, they are more common and
best developed along mountains of bold relief
in arid and semi-arid regions (Figs. 5, 6).
The alluvial-fan model has the following
characteristics: (1) sediment transport occurs
under some of the highest energy conditions
within the entire realm of clastic sedimentation,
(2) deposition of clastic sediment occurs di-
rectly adjacent to the areas of erosion which
provide the sediments, and (3) deposits are of
maximum possible range in size of clastic parti-
cles (from the largest boulders to clays) and
are commonly very poorly sorted compared
with other types of alluvial sediments (Fig. 5).
The size of individual alluvial fans is con-
trolled by drainage-basin area, slope, climate,
and character of rocks within the mountain
range. Individual fans range in radius from sev-
eral hundred feet to several tens of miles. Co-
alescing fans can occur in linear belts that are
hundreds of miles long. Fan deposits usually at-
tain their maximum thicknesses and grain size
near the mountain base (apex of fan) and
-------
142
Rufus J. LeBlonc
-•Aif 0« lOt
or FAN
HC1ION l-f
Fie. 5—Alluvial-fan model of clastic sedimentation.
gradually decrease in thickness away from the
apex.
The alluvial-fan environments commonly
grade downstream into braided-stream or
playa-lake environments. In some areas, where
mountains are adjacent to oceans or large in-
land lakes, alluvial fans are formed under both
subaerial and submerged conditions. Such fans
are now referred to as "Gilbert-type" deltas.
Alluvial-fan deposits form important reser-
voirs for groundwater in many areas, and adja-
cent groundwater basins are recharged through
the fan deposits which fringe these basins.
Source, Transportation, and Deposition
of Sediments
Tectonic activity and climate have a pro-
found influence on the source, transportation,
and deposition of alluvial-fan deposits. Uplift
of mountain ranges results in very intensive
erosion of rocks and development of a very
high-gradient drainage system. The rate of
weathering and production of clastic material is
controlled mainly by rock characteristics and
climate (temperature and rainfall).
Clastic materials are transported from source
areas in mountains or high hills to alluvial fans
-------
Geometry of Sandstone Reservoir Bodies
143
by several types of flows: stream flows and
sbeetfloods and debris flows or mudflows. Sedi-
ment transport by streams is usually character-
istic of large fans in regions of high to moder-
ate rainfall. Mudflows or debris flows are more
common on small fans in regions of low rain-
fall characterized by sudden and brief periods
of heavy downpours.
Stream deposits—Streams which drain rela-
tively small segments of steep mountain ranges
have steep gradients; they may erode deep can-
yons and transport very large quantities of
coarse debris. The typical overall stream gra-
dient is concave upward, and the lowest gra-
dient occurs at the toe of the fan (Fig. 5).
Hooke (1967) described a special type of
stream-flow deposit, which he called "sieve de-
posits," on fans which are deficient in fine sedi-
ments. These gravel deposits are formed when
water infiltrates completely into the fan. Bull
(1969) described three types of water-laid sedi-
ments on alluvial fans: channel, sheetflood, and
sieve deposits. Stream channels radiate outward
from the fan apex and commonly are braided.
The processes of channel migration, diversion,
abandonment and filling, and development of
new main channels and smaller distributary
channels on the lower part of the fan surface
are characteristic features. Most fan surfaces
are characterized by one or a few active chan-
nels and numerous abandoned channels. De-
posits on abandoned portions of gravelly and
weathered fan surfaces are referred to as
"pavement."
Alluvial-fan channel deposits have abrupt
basal contacts and channel geometry; they are
generally coarse. Bull (1972) described chan-
nel deposits as imbricated and massive or thick-
bedded.
Heavy rainfall in mountainous source areas
can result in floods on alluvial fans. The rela-
tively shallow and wide fan channels are not
capable of carrying the sudden influx of large
volumes of water; consequently, the streams
overtop their banks and flood part of the fan
surface. The result is the deposition of thin lay-
ers of clastic material between channels. Bull
(1969) reported sheetflood deposits to be finer
grained than channel deposits, cross-bedded,
and massive or thinly bedded.
Debris-flow deposits—Some workers refer to
both fine-grained and coarse-grained types of
plastic fiowage in stream channels as mudflows,
but others consider mudfiows to be fine-grained
debris flows. Examples of transportation and
deposition of clasUc sediments by mudflows
Flo. 6—Slratigraphic geometry of an alluvial fan.
After Bull (1972).
were first described by Rickmers (1913) and
Blackwelder (1928). The following conditions
favor the development of roudflows: presence
of unconsolidated material with enough clay.to
make it slippery when wet, steep gradients,
short periods of abundant water, and sparse
vegetation.
Pack (1923) discussed debris-flow deposi-
tion on alluvial-fan surfaces. Debris flows occur
as a result of very sudden, severe flooding of
short duration. Beaty (1963) described eye-
witness accounts of debris flows on the west
flank of the White Mountains of California and
Nevada. Debris flows follow channels, overtop
the channel banks, and form lobate tongues of
debris along channels. Debris-flow deposits are
very poorly sorted, fine- to coarse-grained, and
unstratified; they have abrupt margins. This
type of deposit is probably most common on
the upper parts of the fans between the apex
and midfan areas.
Summary: Character and Geometry of
Alluvial-Fan Deposits
Most of the alluvial-fan studies conducted
thus far have been concerned primarily with
the origin and general characteristics of fans
and the distribution of sediments on the sur-
faces of fans. An exception is Bull's excellent
summary paper (Bull, 1972), which contains
significant data on the geometry of channel,
sheetflood, debris-flow, mudflow, and sieve de-
posits. The abstract of Bull's paper is quoted
below:
-------
144
Rofuj J. LeBlanc
Alluvial fans commonly are thick, oxidized, erogenic
deposits whose geometry is influenced by the rate and
duration of uplift of the adjacent mountains and by
climatic factors.
Fans consist of water-laid sediments, debris-flow de-
posits, or both. Water-laid sediments occur u channel,
sheet flood, or sieve deposits. Entrenched stream chan-
nels commonly are backfilled with gravel that may be
imbricated, massive, or thick bedded. Braided sheets of
finer-grained sediments deposited downslope from the
channel may be cross-bedded, massive, laminated, or
thick bedded. Sieve deposits are overlapping lobes of
permeable gravel.
Debris-flow deposits generally consist of cobbles and
boulders in a poorly sorted matrix. Mudflows are fine-
grained debris flows. Fluid debris flows have graded
bedding and horizontal orientation of tabular particles.
Viscous flows have uniform particle distribution and
vertical preferred orientation that may be normal to
the flow direction.
Logarithmic plots of the coarsest one percenlile ver-
sus median particle size may make patterns distinctive
of deposilional environments. Sinuous patterns indicate
shallow ephemeral stream environments. Rectilinear
patterns indicate debris flow environments.
Fans consist of lenticular sheets of debris (length/
width ratio generally 5 to 20) and abundant channel
fills near the apex. Adjacent beds commonly vary
greatly in particle size, sorting, and thickness. Beds ex-
tend for long distances along radial sections and chan-
nel deposits are rare. Cross-fan sections reveal beds of
limited extent tbat are interrupted by cui-and-fill struc-
tures.
Three longitudinal shapes are common in cross sec-
lion. A fan may be lenticular, or a wedge that is either
thickest, or thinnest, near the mountains.
Ancient Alluvial-Fan Deposits
Some examples of ancient alluvial-fan depos-
its which have been reported from the United
States, Canada, Norway, and the British Isles
are summarized in Table 1, together with other
types of alluvial deposits.
BRAIDED-STREAM MODEL or CLASTIC
SEDIMENTATION
Occurrence and General Characteristics
Braided streams occur throughout the world
under a very wide range of physiographic and
climatic conditions. They are common features
on extensive alluvial plains which occupy a po-
sition in the clastic realm of sedimentation be-
tween the high-gradient alluvial-fan environ-
ment at the base of mountain ranges and the
low-gradient meandering-stream model of sedi-
mentation (downstream). In physiographic
provinces characterized by mountainous areas
adjacent to the sea, the braided-stream environ-
ment can extend directly to the coastline and
thus constitute the predominant environment of
alluvial deposition. In this type of situation,
meandering streams do not exist (Fig. 7). The
braided stream is also a common feature of gla-
cial outwash-plains associated with the fluvio-
glacial environment.
The braided-stream model is characterized
by extremely variable rates of sedimentation in
multiple-channel streams (Fig. 8), the patterns
of which vary widely compared with meander-
ing channels. Braided channels are usually wide
and shallow; they contain numerous bars, are
slightly sinuous or straight, and migrate at
. rapid rates. Stream gradients are high, are quite
variable, and are less than those of alluvial fans
but generally greater than those of meandering
streams. Large fluctuations in discharge occur-
ring over short periods of time are also com-
mon. The combination of steep gradients and
high discharge rates results in the transporta-
MARINE
At • AUUVIAl IAN
Ml t Mf ANDMIMO HI (AM
D • Dili*
Fic. 7—B raided-si/earn model of clastic sedimentation.
-------
Geometry of Sandstone Reservoir Bodies
145
tion and deposition of large amounts of coarse
material, ranging from boulders to sand.
Braided-stream deposits overall are finer than
those of alluvial fans, coarser than those of
meandering streams, and quite varied in stratifi-
cation.
Source, Transportation, and Deposition
of Sediments
Aggrading braided streams transport very
large quantities of clastic material derived from
a variety of sources, such as ourwash plains, al-
luvial fans, mountainous areas, and broad
plains. Unlike that of meandering streams, the
bulk of the sedimentary load of most braided
streams is transported as bed load. Rates of
sediment transport and deposition are ex-
tremely variable, the maximum rate occurring
during severe floods of short duration. High-
gradient upstream segments of braided streams
close to source areas are characterized by depo-
sition of poorly sorted clastic sediments which
Table 1. Examples of Ancient Alluvial-Fan, Braiderl-Stream, and Meandering-Stream Deposits
AHuvM fan
Arizona
California
CaUTornia
Colorado
Colorado
Colorado
Connecticut Valley
Massachusetts
Massachusetts
Montana
Montana
Montana
S.W. USA
Texas
Wyoming
Northeastern Canada
Northwest Territories
Wales and Scotland
Norway
Braldtd Slrtam Meandering Slrtam
Colorado
Illinois
Illinois
Kansas
IJano Estacado
Maryland Maryland
Michigan
Montana
Mississippi Mississippi
Montana
New York New York
New Jersey, New York
Pennsylvania
Pennsylvania
Pennsylvania
Texas
West Virginia
Wyoming
Wyoming
Nova Scotia
Northwest Territories Northwest Territories
England
South Wales
Scotland
Spain
Spitsbergen Spitsbergen
Nc» South W«lei
Composite
Arizona
California
California
Colorado
Colo. Plateau
Colo. Plateau
Kansas
Massachusetts
Montana
Nebraska
Nebraska
North Dak ota
Oklahoma
Rhode Island
Texas
Alberta
Quebec
Author
Melton. 1965
Crowell. 1954
Fiona 1. 1967
Galehouse. 1967
Boggs. 1966
Bolyard, 1959
Brady. 1969
Finch. 1939
Stokes. 1961
Howard, 1966
Hubert. I960
Klein. 1968
Hewitt el ol.. 1965
Shelton. 1972
Lins. 1930
Shelion. 1971
Bretz & Horberg. 1949
Hansen. 1969
Wessel. 1969
Stanley. 1968
Mutch. 1968
Shideler. 1969
Gwinn. 1964
Eerg & Cook. 1968
Gwinn & Mulch. 1965
Shelion. 1967
Wilson. 1967. 1970
Beaiy. 1961
Exum & Harms, 1968
Harms. 1966
Banner. 1968
Smith. 1970: Shelion, 1972
Royse. 1970
Visher, I965b
Eeulner rl ol.. 1967
Smilh. 1970
Ryan. 1965
Mutch. 1968
Bull. 1972
Fisher & McGowen, 1969
McGowen & Groat. 1971
McGowen & Garner. 1965 ; Shelion, 1971
Beeroower. 1964. 1969
Berg. I96S
Spearing. 1969
Byers. 1966
Dineley & Willinms, 1968
Klein. 1962
Way, 1968
Miall. 1970
Allen. 1964; Laming. 1966
Bluck. 1965. 1967
Kelling. 196S
Nilsen. 1969
Williams, 1966. 1969
Nagtcgaal. 1966
Moody. Slunn, 19(6
Conolly. 1965
-------
146
Rufui J. LeBlanc
FIG. 8—Types of brajded-stream channels and ban.
range in size from boulder to sand. Farther
downstream, there is a gradual decrease in
grain size and an increase in sorting.
The bed-load materials are transported under
varying bed-form conditions, depending upon
river stage. Coleman (1969) reported ripple
and dune migration in the Brahmaputra River
of Bangla Desh ranging from 100 ft to 2,000 ft
(30-610 m) per day. Chein (1961) reported
downstream movement of sandbars in the Yel-
low River of China to be as great as 180-360
ft (55—110 m) per day. (For comparison, the
rate of bed-load movement in the meandering
Mississippi is about 40 ft [12 m] per day.)
Process of channel division (braiding) by de-
velopment of bars—The exact causes of chan-
nel division which results in the development of
the braided pattern are not very well under-
stood. Two methods in which channel division
takes place have been described by Ore (1963)
as follows:
Leopold and Wolman (1957, p. 43-44), using re-
sults of bolh stream-table studies and observations of
natural braided streams, discuss in some detail bow
channel division may take place. At any time, the
stream is carrying coarser fractions along the channel
center than at the margins, and due to some local hy-
draulic condition, pan of the coarsest fraction is depos-
ited. Finer material is, in part, trapped by coarser par-
ticles, initiating a central ridge in the channel. Progres-
sive additions to the top and downstream end of the
incipient bar build the surface toward water level. As
progressively more water is forced into lateral channels
beside the growing bar, the channels become unstable
and widen. The bar may then emerge as an island due
to downcutting in lateral channels, and eventually may
become stabilized by vegetation. New bars m«y then
form by the same process in lateral channels. These
authors stress that braiding is not developed by the
stream's inability to move the total quanuty of sedi-
ment provided to it; as incapacity leads merely to ag-
gradation without braiding. The condition requisite to
braiding is that the stream cannot move certain sizes
provided; that is, the stream is incompetent to trans-
port the coarsest fraction furnished to a given reach.
Observations for the present study substantiate the
braiding process of Leopold and Wolman.
Many features of streams, ban, and braided reaches
result from changes in regimen (e.g., discharge, load,
gradient), to a large extent representing seasonal fluc-
tuations. Other features of ban result from normal
evolution, and represent no change in regimen.
The incipient longitudinal bar formed in a channel
commonly has an asymmetric, downstream-pointing,
crescentic shape. This coarse part a the "nucleus" of
the bar, is coarser than successive additions to the
downstream end, and largely retains its position and
configuration as long as any part of the bar remains.
During longitudinal bar evolution downstream of this
incipient bar the water and its sediment load com-
monly sweep from one lateral channel diagonally
across the downstream end of the bar, forming a
wedge of sediment with an advancing front at its
downstream edge. This wedge of sediment is higher at
its downstream edge, both on the longitudinal ban de-
scribed here, and where found as transvene ban to be
considered later. The latter build up the channel floor,
independent of longitudinal bar development, simply
by moving downstream.
After a certain evolutionary stage, bar height stops
increasing because insufficient water for sediment
transport is flowing over its surface, and deepening and
widening of lateral channels slowly lower water level.
From then on, the bar may be either stabilized by veg-
etation or dissected.
Widening of a reach after bar deposition is in some
cases associated with lateral dissection of the newly
formed bar. Most erosion, however, apparently occurs
on the outer channel margin* If water level remains
essentially constant for long periods of time, lateral
dissection may establish terraces along bar margins. A
compound terrace effect may be established during
falling water stages. The constant tendency of the
stream to establish a cross-sectional profile of equilib-
rium is the basic cause of lateral cutting by the stream.
Longitudinal ban which become awash during high-
water stages may be dissected by small streams flowing
transversely over their surfaces. In stream-table experi-
ments, sediment added to a system eroding transverse
channels on bar surfaces is first transported along lat-
eral channels beside the ban. Eventually, these chan-
nels fill to an extent that sediment starts moving trans-
versely over bar surfaces, and fills bar-top, transverse
channels. The addition of sufficient sediment to fill lat-
eral and bar-top channels often culminates in a trans-
verse bar covering the whole bar surface evenly.
Another process of braiding, in addition to that de-
scribed by Leopold and Wolman, takes place in well
sorted sediments, and involves dissection of transverse
ban. This is in opposition to construction of longitudi-
nal bars in poorly sorted sediment, the type of braiding
discussed above. Both types may occur together geo-
graphically and temporally. During extended periods of
high discharge, aggradation is by large tabular bodies
of sediment with laterally sinuous fronts at the angle
of repose migrating downstream,- Stabilization of dis-
charge or decrease in load after establishment of these
transverse ban results in their dissection by anastomos-
ing channels; ban in this case form as residual ele-
ments of the aggradational pattern.
The transient nature of braided stream deposiu'onal
surfaces is characteristic of the environment. Tbo
streams and deposiu'onal areas within the stream exhibit
profound lateral-migration tendencies, especially during
-------
Geometry of Sandstone Reservoir Bodies
147
periods of high discharge. Channel migration takes
place on several scales. Individual channels erode later-
ally, removing previously deposited ban. They divide
and coalesce, and several are usually flowing adjtcent
to one another concurrently within the main channel
system. The whole channel system, composed of sev-
eral flowing channels with bars between, also exhibits
migrating tendencies.
Braided-stream deposits—Our knowledge of
modern braided-stream deposits has increased
substantially during the past several years as a
result of studies of several rivers in Wyoming,
Colorado, and Nebraska by Ore (1963, 1965);
the Brahmaputra River of Bangla Desh (for-
merly East Pakistan) by Coleman (1969); the
Platte River of Colorado and Nebraska by N.
Smith (1970); the Red River of the Texas pan-
handle by Waechter (1970); the Canadian
River of northwest Texas by Kessler (1970,
1971); and the Copper River of Alaska by
Boothroyd (1970). These studies revealed that
braided-stream deposits are laid down princi-
pally in channels as longitudinal bars and trans-
verse bars. Abandoned-channel deposits (chan-
nel fills) have been reported by Doeglas
(1962) and Williams and Rust (1969).
According to Ore (1963, 1965), longirudi-
nal-bar deposits occur mainly in upstream
channel segments and transverse bars are more
common in downstream segments; however, in
some places these two types of bars occur to-
gether (Fig. 8). Longitudinal-bar deposits are
lens-shaped and elongated La the downstream
direction. Grain size decreases downstream
from coarse to fine in an individual bar; depos-
its are poorly sorted and mainly horizontally
stratified but laterally discontinuous. Trans-
verse-bar deposits occur as long thin wedges
and are highly dissected by channels. The
downstream edges of transverse bars migrate to
produce planar cross-stratification and some
festoon crossbedding. Sediments of transverse
bars are generally finer and better sorted than
those of longitudinal bars.
N. Smith (1970) described some very signifi-
cant relations between types of bars, stratifica-
tion, and grain size in the Platte River. In the
upstream segment in Colorado, the deposits
consist mainly of longitudinal bars character-
ized by low-relief stratification, generally hori-
zontally bedded but including some fesloon
crossbedding. The downstream channel seg-
ment in Nebraska is characterized by trans-
verse-bar deposits consisting of better sorted,
fine-grained sand with abundant tabular cross-
stratification and some festoon crossbedding.
The Red River braided-stream sediments of
West Texas consist of longitudinal-bar deposits
with low-angle or horizontal stratification; they
are deposited during waning flood stages
(Waechter, 1970). Low-river-stage deposits
consist mainly of migrating transverse-bar de-
posits (in channels) with tabular cross-stratifica-
tion and some festoon crossbedding. The migra-
tion of very shallow channels results in stratifica-
tion sets that are horizontal, tabular or lentic-
ular, and laterally discontinuous.
Kessler (1970) reported longitudinal-bar de-
posits consisting mainly of fine sand in up-
stream reaches of the Canadian River in West
Texas. Transverse-bar deposits are predominant
in the downstream part of the area studied.
Kessler (1971) also discussed individual flood
sequences of deposits which contain parallel
bedding and tabular and small-ripple cross-lam-
inations. These sequences are covered by clay
drapes and are laterally discontinuous.
Coleman (1969) presented the results of a
significant study of one of the largest braided
rivers of the world, the Brahmaputra in Bangla
Desh. This river is 2-6 mi (3-9.5 km) wide
and migrates laterally as much as 2,600 ft (790
m) per year; deposition of sediments in its
channels during a single flood occurs in a defi-
nite sequence of change, ranging from ripples
up to 5 ft (1.5 m) high that migrate down-
stream 400 ft (120 m) per day to sand waves
50 ft (15 m) high that migrate up to 2,000 ft
(610 m) per day.
Williams and Rust (1969) presented results
of a very detailed study of a 4-mi (6.5 km)
segment of a degrading braided stream, the
Donjek River of the Yukon Territory, Canada.
They divided the bar and channel deposits,
which range from coarse gravels to clays, into
seven facies. Ninety-five percent of the bar de-
posits are of the longitudinal type and consist
of gravel, sand, and some finer sediments.
Abandoned-channel deposits consist of grada-
tional sequences of gravels, sand, and clays that
become finer upward.
Summary: Braided-Stream Deposits
Most of the sediments of modern braided
streams studied during the past decade have
been referred to by authors as transverse- or
longitudinal-bar deposits. These sediments were
deposited within braided channels during vary-
ing discharge conditions ranging from low wa-
ter to flood stage. Thus, all longitudinal and
transverse bars should be considered as a spe-
cial type of bed form occurring within active
braided channels.
-------
148
Rufus J. LeBlanc
C - CMANNll
0 • OtlTA
IS • IIAIDID Stlf AM
MIW-MIAN IOW WATER
IM • iNfllMIDIAIE MOOD SIAOf
H • POINT IAI
MAI flOOD ItACI
71.0343-7
FIG. 9—Setting and general-characteristics of raeanderiag-stream model of clastic sedimentation.
Studies by Doeglas (1962) and Williams
and Rust (1969) are significant because they
describe abandoned-channel deposits. Doeglas
discussed the methods of channel abandonment
and described the channel-fill deposits as coarse
grained, with channel or festoon laminations, in
the upstream portions of abandoned channels,
and as fine grained, silty, and rippled in the
downstream portions of abandoned channels.
Ancient Braided-Stream Deposits
Some examples of ancient braided-stream de-
posits which have been reported from the
United States, Spitsbergen, and Spain are sum-
marized in Table 1.
MEANDERINC-STREAM MODEL OF
CLASTIC SEDIMENTATION
Occurrence and General Characteristics
Meandering streams generally occur in
coastal-plain areas updip from deltas and
downdip from the braided streams. The axis of
sedimentation is usually perpendicular to the
shoreline (Fig. 9).
This model is characterized by a single-chan-
nel stream which is deeper than the multichan-
nel braided stream. Meandering streams usually
have a wide range in discharge (cu ft/sec)
which varies from extended periods of low-wa-
ter flow to flood stages of shorter duration.
Flooding can occur one or more times per year
and major flooding once every several years.
The meandering channel is flanked by natural
levees and point bars, and it migrates within a
zone (meander belt) about 15 to 20 times the
channel width. Channel segments are aban-
doned and filled with fines as new channels de-
velop.
• Source, Transportation, and Deposition
of Sediments
Sediments are derived from whatever type of
deposit occurs in the drainage area. Clays and
fine silts are transported in suspension (sus-
pended load), and coarser sediments such as
sand, gravel, and pebbles are transported as bed
load. Sediment transport and deposition during
extended low-water stages are confined to the
channel and can be nil or very slow. Maximum
sediment transport occurs during rising flood
stage when the bed of the channel is scoured.
The maximum rate of sediment deposition
occurs during falling flood stages. Grain size
depends on the type of sediment available to .
the channel; the coarsest sediments are depos-
ited in the deepest part of the channel, and the
finest sediments accumulate in floodbasins and
in some parts of the abandoned channels.
Channel migration and deposition of point-
bar sediments—The most important processes
of sedimentation in the meandering-stream
model are related to channel migration which
occurs as a result of bank caving and point-bar
accretion (Fig. 10). The process of bank cav-
ing occurs most rapidly during falling flood
-------
Geometry of Sandstone Reservoir Bodies
149
I IAMK CAVING
\ FOINT-tAI ACCIItlON
Fic. 10—Areas of bank caving and point-bar accretion along a meandering channel.
stage, when currents of maximum velocities are
directed against the concave bank. Bank caving
occurs at maximum rates in beads where the
bed and bank materials are very sandy. Rates
are much slower in areas where banks are char-
acterized by clayey sediments (Fisk, 1947).
Deposition occurs on the convex bar (point
bar) simultaneously with bank caving on the
concave bank.
Bank caving and point-bar accretion result in
channel migration and the development of the
point-bar sequence of sediments (Fig. 11). The
point bar is probably the most common and
significant environment of sand deposition. The
thickness of this sequence is governed by chan-
nel depths. Point-bar sequences along tie Mis-
sissippi River attain thicknesses in excess of
150 ft (45 m). Medium-size rivers like the Bra-
zos of Texas produce point-bar sequences that
are 50 ft (15 m) thick (Bernard et al, 1970).
Channel diversions and filling of abandoned
channels—The process of channel diversion
and channel abandonment is another character-
istic feature of meandering streams. There are
two basic types of diversion and abandonment:
(1) the neck or chute cutoff of a single mean-
der loop and (2) the abandonment of a long
channel segment as a result of a major stream
diversion (Fisk, 1947).
Meander loops which are abandoned as a re-
sult of neck or chute cutoffs become filled with
sediment (Fig. 12A). The character of the
channel fill depends on the orientation of the
abandoned loop with respect to the direction of
flow in the new channel. Meanders oriented
with their cutoff ends pointing downstream
•(Fig. 12B).are filled predominantly with clays
(clay plugs); those oriented with the cutoff
ends pointing upstream are filled principally
with sands and silts.
A major channel diversion is one which re-
sults in the abandonment of a long channel seg-
ment or meander belt, as shown in Figure 13.
Channeling of flood water in a topographically
low place along the bank of the active channel
can rapidly erode unconsolidated sediments
and create a new channel. This process can
happen during a single flood or as a result of
several floods. The newly established channel
has a gradient advantage across the topographi-
cally lower floodbasio. A diversion can occur at
any point along the channel.
Fic. 11—Development of point-bar sequence of sediments.
-------
150
Rufus J. LeBlanc
CHUTE CUTOFF
TYPES OF CHANNEL FILLS
Fio. 12—Channel diversion, abandonment, and filling as a result of neck and chute cutoffs.
The character of the sediments which fill
long channel segments is governed by the man-
ner of channel diversion. Abrupt abandonment
(during a single flood or a few floods) results
in the very rapid filling of only the upstream
end of the old channel, thus creating a long sin-
uous lake. These long, abandoned channels
(lakes) fill very slowly with clays and silts
transported by flood waters (Fig. 14, left).
Gradual channel abandonment (over a long
period) results in very gradual channel deterio-
ration. Diminishing flow transports and depos-
its progressively smaller amounts of finer sands
and silts (Fig. 14, right).
Summary: Characteristics of Meander-Belt
and Floodbasin Deposits
The rneandering-stream model of sedimenta-
tion is characterized by four types of sedi-
ments: the point bar, abandoned channel, natu-
ral levee, and floodbasin. The nature of each of
these four types of sediments and their interre-
lations are summarized in Figure 15.
Only two main types of sand bodies are asso-
ciated with a meandering stream: the point-bar
^:<^
.-?:*-;j:^*O
•'-*A»IAN0I1 III!
-AIANDOMIO »IAWDtl lilt
MAtlMf MIAMMI lilt CCMIttVCIID IO*OC«AfNKAUT MIOM AUVVtAt ltf>CI
AMD 10NC IINUOUI COUIU.riOOO WAKIt (HVHUO IO IOW rlOOOIAIIH
DfVIIO' NIW CHAMMtt.
sands and the abandoned-channel fills. The for-
mer, which are much more abundant than the
latter, occur in the lower portion of the point-
bar sequence and constitute at least 75 percent
of the sand deposited by a meandering stream.
Coalescing point-bar sands can actually form a
"blanketlike" sand body of very large regional
extent. The continuity of sand is interrupted
only by the "clay plugs" which occur in aban-
doned meander loops or in the last channel po-
sition of meander belts which have been aban-
doned abruptly.
Examples of ancient alluvial deposits of
meandering-stream origin which have been re-
ported in the literature are summarized in
Table 1.
DELTAIC MODELS OF CLASTIC SEDIMENTATION
Occurrence and General Characteristics
Deltaic sedimentation occurs in the transi-
tional zone between continental and marine (or
inland seas and lakes) realms of deposition.
Deltas are formed under subaerial and suba-
queous conditions by a combination of fluvial
and marine processes which prevail in an area
where a fluvial system introduces land-derived
sediments into a standing body of water.
Fic. 13—Major channel diversion and abandonment
of a meander belt
Flo. 14—Variations in character of abandoned channel
fill typical of meander bells.
-------
3-
a.
a
cr
g.
e
ex
•a.
I-
B.
F
prEri^ i"7^ - ""^^•'•''*1
HiTvui ttvtr Mraim (VL)
injinnxT m* vtKt or *«mn , tun tm CUTI
^rwflTT» n ovtiww ruoe w*nn . winttiT itcno*
occuu ABJjetxT 10 ewntL. dtroim ntm »«*T rvx
OUPnrn. »•» CUM "IP" FlDOOlAtlit CtAYI .
mur»cc cowitm or mi* CTCIIC in«in or IICMT
eoLMtD. •o*tirwT*i.i.T.iiBett )*i*m inn (eirwmo
tuiiMc orMum rum) nr*MTte r«c* L*CN or*ti rt
io*t i on tfi»u vnin ocvttof iirwtcy rumo IUIOM.
CmMLl Itqumcc tl MPH ^t*l Wit *wo O->MI
im IAMM •« inn orw*«H. ovtiLiti rLoooMUK
CUTI « ouo poirr wt Mrotin. nictar rmnoi
or PMTVML Lrvrtf »it trxn-ro »T MWV CAVI«C rttxcji,
ru)oo»Aliv otroiin (ri)
tummni *ceuMtn>iioH or CUT) **> turr CUTI
CAtLi uvut ft^nw or m ruxw*
rum. emu. to ou»t* rvooermi* Btnvin tvc* *i
ikAnofn HL««MB irtT Dtw»in tn runoiAtm cun.
IRKIDKI oonitn or CTCLIC UTUU or CUTI im
mrr CUTI DCP«TiO" or MCAMIC tin itfitKzm nnira
1* eutu (tit mn CXXIUU.T owui 10 riPi VTNUKOC.
uvct p*it or oupffd.riiL (ActiTt otumL-riLU if
vtit itxiui n> TXI uvu ratnoM or mi MIVT M*
itauncT. ccmiiTtw or miitvt IAMB ovtiura n
ruTtm ctM>*itDOci IAN*, ix owiMtu MIIOI vni
AUKDOMrt CUBUAlLY T»11 UVO ICCT10H It OVtIUIV
IT cioii-BEMKt iA»n inn AM tun wtai itrti.
itxr Mrasitiov WIIK ritruL ruw in mi CMM*a_
Tttl ami MOST WIT CtMltlTl OP F
tTACI Of IVI OtAJnvL.
own (Acrivt oumnti-riLU 11 •IUTTVT OVTKLAII rr
A UUTIVTLT THICK ITCT1OT OP OKAF1C I1OI CUTI (CT
nu IUNDOMCP iiAci). nit QUIT n rtrr (Dtiut TO
ICMI r*in or T«t runoutt* w rot IT*.
5HAU-SCA1E CIOJS-HOOINC
UIGt-SCAU ClOSS-ltOOINC
MOOING
It^l UNO
CUT AND 5HT
MAI. fjOOO ttvtl
POIITT.VI otrain (P«)
auMi TO rm arVAU* IKQUCKI or snowr*
BtmtTtD OH TV! OOKVU WPK fMt) OP MOUTtPa
PcroilTloiuL rvartu AOKMI
MR (rtCM VIKPCflT rtMTTOJI OT
TO m maiift r»« OP tut KAI) ii
ACTtBIID Iff TMt rOUXVlPO PW» tOMMt
». natuT "IT or wt IIMUATBT rut or
••MI ruxjo eoN*tTiaM). u» POM eenitn mmv «r
»MIJ. tirrui.
l. rovrto* or i*t (CHIBKIJ oovoo rr WTO «niw
rtoo* op mKMM AM iBitwasun NAcvimci. M>
POM PUMMtMrTU n»U KIPPUS AK» NOtlCOVUL.
KPOIln Or TMII tOMt CDMIBTS OP CTO.IC Win OT
MxitovtAtir-MBoa *«• HMU »e*u CMKI-UBOCB ntn
tin IAHK Ilin. »ITW K»C CUTt.
I. P««TION OP Ml MOST PVKWDfm OOVIHU «f ttATtt
(rMM MX TO PLOOM1ACU). M» TOW OMIIT* Or UKI
tIPPLO OUUOTU). fUUOMT AU NAt*U BANM Vim
tOH* OUTTU TNAT AU PUTOOH CXMI-MDOU. VOTUim
•OH TMM VHin V •OKIIOWTALLT-M9MB AHD MILL KALI
CBMt-UMt* U«M.
I. KMT ION or oumvt ctimuLLT ALMTS eovtm n
MATTI. MM POKN POT OMfm. Bt>tMHm AU M&tltVt
AM* nn rooBLT »tmn». vnuu.* continue CUT MUJ.
AS HI eKvt n Aooinioi (orrvAp) tens I. j.
AM * AU tVPCVOtO On* (OMtS I. I. AM I USFtCIIVCLT.
Twit rwtcui iu«in u m MTaorMun or tm nne»t
poin us stqnNct IMMI on tw urr.
UTBU OP CLAT (>•»•» *« »««"ltt» (PMM WtPIMIff^
ovn PAIT or TM "I WIIMC v*n« PLOD* ITMO nm
TtLOCin M Ptl. TWC CtAT MAPO AU PUguDrrtT fU-
•n*v«. mr CAN ocon AmrMau l» nt trauira.
•VI tK MUM PU»» m IVTIMOUTt PAITV OP TM
ftAI (>«*LLO' PAIT OP rtOO* CHAIWIU ACTS A* A SPtCUt
TTPI OP »u raw. mmiavtauT.kOMB urn* A*I MMMITC*
ovn nut MIC* PAITI. »*p«t MUCH irtu ovu ni UK
ucu AU puvun* A* inu. o*u tiHiwe. VIOALLT vim
vtn «iw ttr«.
O
a
o
3
o
ex.
OB
O
o.
-------
152
Rufus J. LeBlonc
Fie. 16—Occurrence of deltaic models of
clastic sedimentation.
Large deltas usually are associated with ex-
tensive coastal plains; however, all coastal
plains do not include large deltas. The deltaic
environment occurs downstream from the
meandering-stream environment and is directly
adjacent to, and updip (landward) from, the
marine environment; it is flanked by the
coastal-interdeltaic environment. Most large
deltas occur on the margins of marine basins,
but smaller deltas also form in inland lakes,
seas, and coastal lagoons and estuaries (Fig.
16).
That portion of a delta which is constructed
under subaerial conditions is called the "deltaic
plain"; that portion which forms under water is
called the "delta front," "delta platform," and
"prodelta." The bulk of the deltaic mass is de-
posited under water.
Deltas are considered to be extremely impor-
tant because they are the sites of deposition of
sand much father downdip than the interdeltaic
environment, as well as being the sites where
clastic deposition occurred at maximum rates.
Source and Transportation
of Sediments
Sediments deposited in large deltas are de-
rived from extensive continental regions which
are usually composed of rock types of varied
compositions and geologic ages. Thus, the com-
position of deltaic sediments can be quite var-
ied.
The sediment load of rivers consists of two
parts: (1) the clays and fine silts transported in
suspension and (2) the coarser silts and sands,
and in some cases gravels, transported as bed
load. The ratio of suspended load to bed load
varies considerably, depending upon the rock
types and climatic conditions of the sediment-
source areas. The suspended load is generally
much greater than the bed load.
The transportation of sediment to a delta is
an intermittent process. Most rivers transport
the bulk of their sediments during flood stages.
During extended periods of low discharge, riv-
ers contribute very little sediment to their del-
tas.
The extent to which deltaic sediments are
dispersed into the marine environment is de-
pendent upon the magnitude of the marine pro-
cesses during the period that a river is in flood
stage. Maximum sediment dispersal occurs
when a river with a large suspended load
reaches flood stage at the time the marine envi-
ronment is most active (season of maximum
currents and wave action). Minimum dispersal
occurs when a river with a small suspended
load (high bed load) reaches flood stage at a
time when the marine environment is relatively
calm.
Size of deltas—There is an extremely wide
range in the size of deltas;* modern deltas
range in area from less than 1 sq mi (2.6 sq
km) to several hundreds of square miles. Some
large deltaic-plain complexes are several thou-
sand square miles in area. Delta size is depen-
dent upon several factors, bu^ the three most
important are the sediment load of the river;
the intensity of marine currents, waves, and
tides; and the rate of subsidence. For a given
rate of subsidence, the ideal condition for the
construction of a large delta is the sudden large
influx of sediments in a calm body of water
with a small tidal range. An equally large sedi-
ment influx into a highly disturbed body of wa-
ter with a high tidal range results in the forma-
tion of a smaller delta, because a large amount
of sediment is dispersed beyond the limits of
what can reasonably be recognized as a delta.
Rapid subsidence enhances the possibility for a
large fluvial system to construct a large delta.
Types of deltas—A study of modem deltas
of the world reveals numerous types. Bernard
' Published figure) on areal extent of deltas arc based
on size of the deltaic plain and do not include sub-
merged portions of the delta, which in many cases are
as large as or larger than the deltaic plains.
-------
Geometry of Sandstone Reservoir Bodies
153
(1965) summarized some of the factors which
control delta types as follows:
Deltas and deltaic sediments are produced by the
rapid deposition of stream-borne materials in relatively
still-standing bodies of water. Notwithstanding the
effects of subsidence and water level movements, mosl
deltaic sediments are deposited off the delta shoreline
in the proximity of the river's mouth. As these materi-
als buiJd upward to the level of the still-standing body
of water, the remainder of deltaic sediments are depos-
ited onshore, within the delta's flood plains, lakes,
bays, and channels.
Nearly 2,500 years ago, Herodotus, using the Nile as
an example, stated that the land area reclaimed from
the sea by deposition of river sediments is generally
deltoid in shape. The buildup and progradation of del-
taic sediments produces a distinct change in stream
gradient from the fluvial or alluvial plain to (he deltaic
plain. Near the point of gradient change the major
courses of rivers generally begin to transport much
finer materials, to bifurcate into major distributaries,
and to form subaerial deltaic plains. The boundaries of
the subaerial plain of an individual delta are the lat-
eral-most distributaries, including their related sedi-
ments, and the coast line. Successively smaller distribu-
taries form sub-deltas of progressively smaller magni-
tudes.
Deltas may be classified on the basis of the nature of
their associated water bodies, such as lake, bay, inland
sea, and marine deltas. Other classifications may be
based on the depth of the water bodies into which the;.'
prograde, or on basin structure.
Many delta types have been described previously.
Most of these have been related to the vicissitudes of
sedimentary processes by which they form. Names
were derived largely from the shapes of the delta
shorelines. The configuration of the delta shores and
many other deposilional forms expressed by differ en I
sedimentary facies appear to be directly proportional
to the relative relationship of the amount or rale of
river sediment influx with the nature and energy of the
coastal processes. The more common and better under-
stood types, listed in order of decreasing sediment in-
fiux and increasing energy of coastaj processes (waves,
currents, and tides), are: birdfool, lobate, cuspate, ar-
cuate, and estuarine. The subdeltas of the Colorado
River in Texas illustrate this relationship. During the
first part of this century, the river, transporting ap-
proximately the same yearly load, built a birdfoot-lo-
bate type delta in Matagorda Bay, a low-energy water
body, and began to form a cuspate delta in the Gulf of
Mexico, a comparatively high-energy water body.
Many deltas are compounded; their subdeltas may be
representative of two or more types of deltas, such as
birdfoot, lobate, and arcuate. Less-known deltas, such
as the Irrawaddy, Ganges, and Mekong, are probably
mature csluarine types. Others, located very near major
scarps, are referred 10 the "Gilbert type," which is sim-
ilar to an alluvial fan.
Additional studies of modem deltas are re-
quired before a more suitable classification of
delta types can be established. J. M. Coleman
(persona] commun.) and his associates, to-
gether with the Coastal Studies Institute at
Louisiana State University, are presently con-
ducting a comprehensive investigation of more
than 40 modern deltas. Results of their studies
undoubtedly will be a significant contribution
toward the solution to this problem.
Only three types of deltas will be considered
in this report: the birdfoot-lobate, the cuspate-
arcuate, and the estuarine.
Sedimentary Processes and Deposits of
the Birdfoot-Type Delta
The processes of sedimentation within a
delta are much more complex and variable
than those which occur in the meandering-
stream and coastal-inlerdeltaic environment of
sedimentation. It is impossible to discuss these
deltaic processes in detail in a short summary
paper such as this; therefore, only a brief sum-
mary of the following significant processes is
presented.
1. Dispersal of sediment in the submerged parts of
the delta (from river mouths seaward);
2. Formation of rivermouth bars, processes of chan-
nel bifurcation, and development of distributary chan-
nels;
3. Seaward progradation of delta, deposition of the
deltaic sequence of sediments, and abandonment and
filling of distributary channels; and
4. Major river diversions, abandonment of deltas,
and development of new deltas.
Dispersal and deposition of sediments—Rjv-
erborne sediments which are introduced ID a
standing body of water (a marine body or in-
land lakes and seas) are transported in suspen-
sion (clays and fine silts) and as bed load
(coarse silts, sands, and coarser sediments).
Most of the sands and coarse silts are deposited
in the immediate delta-front environment as
rivermouth bars and slightly beyond the bar-
front zone. The degree of sand dispersal is, of
course, controlled by the level of marine en-
ergy; however, in most birdfoot deltas, sands
are not transported beyond 50-ft (15 m) water
depths. Fisk (1955) referred to the sands de-
posited around the margins of the subaerial del-
taic plain as "delta-front sands," and they are
called "delta-fringe sands" herein.
The finer sediments (clays and fine silts),
which are transported in suspension, are dis-
persed over a much broader area than the
fringe sands and silts. The degree of dispersal is
governed by current intensity and behavior.
Accumulations of clays seaward of the delta-
fringe sands are referred to as "prodelta" or
"distal clays" (Fig. 17).
Channel bifurcation and development of dis-
tributary channels—Some of the most signifi-
cant deltaic processes are those which result in
the origin and development of distributary
-------
154
Rufus J. LeBlanc
IMI /_ _ /
/V /-.-/__
r"_r^TLT\ — —
_1 x /-_-
n—~ A" *~
FIG. 17—Distribution of distributary-channel and fringe sands in a birdfool-lobale delta.
channels. Welder (1959) conducted a detailed
study of these processes in a part of the Missis-
sippi delta, and Russell (1967a) summarized
the origin of branching channels, as follows:
The creation of branching channels is determined by
the fact that threads of maximum turbulence and tur-
bulent interchange (Auslausch; 1.2.3; 3.5) lie deep and
well toward the sides of channels, particularly if they
have flat beds (typical of clay and fine sediments in
many delta regions). These threads are associated with
maximum scour and from them, sediment is expelled
toward areas of less turbulence and Auslausch. Signifi-
cant load is propelled toward mid-channel, where
shoals are most likely to form.
Ultl SUM OF CKAMNU SUIOIVIilOM
OIICINAI WAMCMING O' » CXIIA CMAMNfl
Fio. 18—Stages in development of channel bifurcation.
After Russell (1967).
At its mouth, the current of a delta channel contin-
ues forward (as a result of momentum) and creates jet
flow into the lake or sea il enters. A-fter leaving the
confinement imposed by fixed banks, however, the cur-
rent flares marginally to some extent (widening the jet,
reducing its velocity, and eventually dissipating its Sow
energy). Near the termination of confining banks the
jet flow is concentrated and moves ahead into relatively
quiet water. With flaring of jet flow comes an increase
in spacing between threads of most intense turbulence
and exchange. There is a tendency toward scour below
each thread, but the exchange prpcess sends most of
the entrained material toward marginal quiet water on
both sides (Fig. 8 fFig. 18 of this paper]). Deposition
creates a submarine natural levee on the outer side of
each thread. Sediment is also attracted toward and de-
posited in the widening area of mid-channel water,
where it builds a shoal. The channel divides around the
shoal, creating two distributaries, each of which devel-
ops its own marginal threads of maximum turbulence,
perpetuating conditions for other divisions below each
new channel mouth. If not opposed by wave erosion
and longshore currents, the subdivision continues in
geometric progression (2, 4, 8, 16, etc.) as the delta
deposit grows forward. >'_
The marginal natural levees are submarine features
at first and fish may swim across their crests. Later
they grow upward, and for awhile become areas where
logs and other flotsam accumulate and where birds
walk with talons hardly submerged. Salt- or fresh-wa-
ter-toleranl grasses invade the shallow water and newly
created land, first along levee crests, later to widen as
the levees grow larger. Salicomia and other plants be-
come established pioneer trees such as willows, and
eventually in the plant succession comes the whole
complex characteristic of natural levees upstream. In
tropical areas mangroves are likely to become the
dominant trees.
-------
Geometry of Sandstone Reservoir Bodies
155
A similar conversion exists in mid-channel, where
the original iboal becomes land and either develops
into a lenticular or irregular island or becomes the
point of land at the bead of two branching distributar-
ies.
Progradation of delta and deposition of del-
taic sequence—Fisk's discussion of the process
of distributary-channel lengthening (prograda-
tion of delta seaward) is probably one of the
most significant of his many contributions on
deltaic sedimentation (Fisk, 1958). His de-
scription of this important aspect of delta de-
velopment is presented below. (Stages in the
development of a birdfoot-type delta are shown
in Figure 19.)
Each of the pre-modcrn Mississippi River courses
was initiated by an upstream diversion, similar to the
one presently affecting the Mississippi as the Atchafa-
laya River enlarges (Fisk, 1952). Stream capture was a
gradual process involving increasing flow through a
diversional arm which offered a gradient advantage to
the gulf. After capture was effected, each new course
lengthened seaward by building a shallow-water delta
and extending it gulfward. Successive stages in course
lengthening are shown diagrammalically on Figure 2
TFig. 19, this paper). The onshore portion of the delta
surface ... is composed of distributaries which are
flanked by low natural levees, and interdistributary
troughs holding near-sea-level marshes and shallow wa-
ter bodies. Channels of the principal distributaries ex-
tend for some distance across the gently sloping offshore
surface of the delta to the inner margin of the steeper
delta front where the distributary-mouth bars are situ-
ated. The offshore channels are bordered by submarine
levees which rise slightly above the offshore extensions
of the imerdisiributary troughs.
In the process of course lengthening, the river occu-
pies a succession of distributaries, each of which is fa-
vorably aligned to receive increasing flow from up-
stream. . . . The favored distributary gradually widens
and deepens to become the main stream . . . ; its natural
levees increase in height and width and adjacent inter-
distributary troughs fill, permitting marshland develop-
ment. Levees along the main channel are built largely
during floodstagc; along the distal cods of distributar-
ies, however, levee construction is facilitated by cre-
vasses . . . which breach the low levees and permit
water and sediment to be discharged into adjacent
troughs during intermediate river stages as well as dur-
ing fioodslage. Abnormally wide sections of the levee
and of adjacent mudflats and marshes are created in
this manner, and some of the crevasses continue to re-
main open and serve as minor distributaries while the
levees increase in height. Crevasses also occur along
the main stream during floodstagcs . . . and permit
tongues of sediment to extend into the swamps and
marshes for considerable distances beyond the normal
toe position of the levee.
Distributaries with less favorable alignment are
abandoned during the course-lengthening process, and
their channels arc filled with candy sediment. Aban-
doned distributaries associated with the development of
the present course below New Orleans vein the marsh-
lands. . . . Above the birdfoot delta, the pattern is simi-
lar to that of the older courses . . . ; numerous long,
branching distributaries diverge at a low angle.
Stream diversions, abandonment of deltas,
and development of new deltas—Deltas pro-
grade seaward but they do not migrate later-
ally, as a point bar does, for example. A delta
shifts position laterally if a major stream diver-
sion occurs upstream in the alluvia! environ-
ment or in the upper deltaic-plain region (Fig.
20). Channel diversions were discussed in the
section on the meandering-stream model.
Deltas, like meander belts, can be abandoned
abruptly or gradually, depending upon the lime
required for channel diversion to occur. Once a
delta is completely abandoned, all processes of
deltaic sedimentation cease to exist in thai par-
ticular delta. With a standing sea level, the sedi-
ments of the abandoned delta compact, and
subsidence probably continues. The net result is
the encroachment of the marine environment
over the abandoned delta. This process has er-
roneously been referred to by some authors
as "the destructive phase of deltaic sedi-
mentation." The author maintains that the
proper terminology for this process is "trans-
gressive marine sedimentation." The two pro-
cesses and their related sediments are signifi-
cantly different, as the discussion of the trans-
gressive marine model of sedimentation demon-
strates (see the succeeding section on this
model).
As the marine environment advances land-
ward over an abandoned subaqueous delta
front and the margins of the deltaic plain, the
upp:r portion of tie deltaic sequence of sedi-
ment is removed by wave action. The amount
of sediment removed depends on the inland ex-
tent qf the transgression and on the rate of sub-
sidence. The front of the transgression is usu-
ally characterized by deposition of thin marine
sand units. Seaward, sediments become finer
and grade into clays. Thus, local marine trans-
gressions which occur because of delta shifts
result in the deposition of a very distinctive
marine sedimentary sequence which is easily
distinguished from the underlying deltaic se-
quence.
Concurrent with marine transgression over
an abandoned delta, a new delta will develop on
the flanks of the abandoned delta. Sedimentary
processes in the new delta are similar to those
described under the discussion of progradation
of deltas.
Repeated occurrences of river diversions re-
sult in the deposition of several discrete deltaic
masses which are separated by thin transgres-
sive marine sequences (Fig. 20). Under ideal
conditions, deltaic fades can attain thicknesses
-------
I
•s
B-
g-
•o
B
8
o.
C.
D.
a.
n
I
o'
p
a
JL
In
O
-------
a.
C.
a'
8
s
•
ex
I
r»
CL
CL
r£_
E.
D.
C
o
IP
o
O
3
a.
jo
m
ft>
2
O
O
a.
tn
-------
158
Rufus J. LeBlanc
of several hundreds of feet in a large sedimen-
tary basin.
Sedimentary Processes and Deposits of the
Cuspate-Arcuate Type of Delta
The shape of a delta is controlled by the in-
fluence of marine processes which are active
against the delta front (Table 2). Russell
(1967a) presented the following excellent sum-
mary of the modification of deltas by marine
processes.
The deposiliooal processes characteristic of river
mouths are opposed by marine processes that work
toward removal of deposits. In a quiet sea or lake the
geometric increase in number of distributaries is most
closely approached. Below the most inland and earliest
forking of the river, the delta builds out as a fan-
shaped accumulation, with distributaries creating ribs
with natural levees separating basins that widen and
open toward the sea. The point deserving greatest em-
phasis is that the entire delta system originates under-
water and only later becomes features visible as land.
The ideal delta front is arcuate or has a bird-foot
shape as viewed from the air or indicated on a chart.
The latter pattern indicates a condition in which the
deposition of load is dominant over the efforts of ma-
rine processes. It results from the forward growth of
natural levees and the inability of longshore currents to
carry away sediment about as rapidly as it is brought
to the river mouth. The delta of the Mississippi is the
largest and most typically cited example. Some talons
of the foot extend out more than 20 miles and the
basins between natural levees flank V-sbaped marshes
and bays up to about 1.5 fathoms deep. Many smaller
bird-fool deltas occur in lakes and estuaries, where
there is relatively little distance for fetch to generate
high waves and where there are only feeble longshore
currents.
The arcuate-front deltas, such as those of the Nile
and Niger, indicate sufficient wave action and removal
of sediment by longshore currents to maintain rela-
tively stable, smooth fronts. In some cases the momen-
tum of jet flow is apparently sufficient to prevent much
flaring, and a single pair of natural levees advances
seaward to form a cuspaie delta front, localized along
a single channel. The Tiber, Italy, is the commonly
cited example. The Saknyra River, on the Black Sea
coast of Turkey bas such, a delta, but the reason is
dominance of wave action. Ahead of it is a large area
of shoal water with an extremely irregular system of
channels and natural levees (changing so rapidly that a
pilot keeps daily watcb over them in order to guide
boats back to the river mouth). Levees are prevented
from growing up to sea level because wave erosion
keeps them planed off to a depth of a few feet and
because longshore currents entrain and transport sedi-
ment away effectively enough to prevent seaward
growth of land area. This leaves but one channel
mouth in a central position as a gently protruding sin-
gle cusp.
The modern Brazos River delta of Texas
(constructed since 1929) is a good example of
a small modem arcuate delta which has been
strongly influenced by marine processes. Ber-
nard et al. (1970) discussed this small delta
and its vertical sequence of sediments. Stages in
the development of this type of delta are shown
in Figure 21.
The modern Niger delta of western Africa is
a classic example of a large arcuate-type delta
that is highly influenced by marine processes
and tidal currents. Allen (1965c, 1970) de-
scribed the environments, processes, and sedi-
mentary sequences of this interesting delta. On
the basis of data presented by Allen, it is obvi-
ous that, although there are many similarities
between the Niger delta and the birdfoot-type
Mississippi delta, there are certainly some sig-
nificant differences. For example, from the
standpoint of sand bodies, the characteristics
and geometry of the delta-fringe sands of the
Niger are considerably different from those of
the Mississippi. As indicated on Figure 22, a
very large quantity of the sand that is contrib-
uted to rivermouth bars by the Niger is trans-
ported landward and deposited on the front of
the deltaic plain as prominent beach ridges
(this is a special form of delta-fringe sand ac-
cording to the writer's deltaic classification).
This process results in the development of a
thick body of clean sands along the entire front
of the deltaic plain.
Another important difference between the
Niger and the Mississippi is the occurrence of a
very extensive tidal-marsh and swamp environ-
ment on the Niger deltaic plain behind (land-
ward of) the prominent beach ridges. This en-
vironment is characterized by a network of nu-
merous small channels which connect with the
main distributary channels. These channels,
which are influenced by a wide tidal range, mi-
grate rather freely and become abandoned to
produce extensive point-bar deposits and many
abandoned channel fills. In contrast, the Missis-
sippi River distributary channels migrate very
little and, hence, point-bar sands constitute
only a small percentage of the deltaic deposits.
In summary, the Niger arcuate delta is char-
acterized by prominent delta-fringe sands
(which include the beach-ridge sands) occur-
ring as a narrow belt along the entire front of
the deltaic plain. Point-bar sand bodies are very
common directly adjacent- to and landward of
the delta-fringe sands. The combination of
high-level marine energy and strong tidal cur-
rents results in development of a relatively
large quantity of distributary and point-bar
(migrating channels) sands.
-------
Tnble 2. Fnclors Which Innuence Characteristics of Deltaic Deposits
|.,U **t»UriM-r
tf1**r»l. CTMltl-t
o
9
O
5
o
If
a.
wi
-------
160
Rufus J. LeBlanc
IW I
• (••*(!•• ilt^tly •%•» ••— ••••l***4. tw*l*llBB •• B#^f Itrmt* i
kU tt*f* ltk**«TKLK^ »^!««H «^ |k* k*t ••••••• BlUCkW <• Ik* I
Tkv* (*• IMtlct VI*** •••U •** •»»•• I* Iwi* I ••*••« »"M •( ||H
FlC. 21—Stages in development of a cuspale della.
Sedimentary Processes and Deposits of
Estuarine-Type Delta
Large deltas such as the Ganges, Amazon,
and Colorado (in Gulf of California) are con-
sidered to be examples of estuarine-type deltas.
Although our knowledge of these deltas is ex-
tremely limited, it is now reasonably well estab-
lished that they are associated with extreme
tidal conditions (up to 25 ft [8 m] at the mouth
of the Colorado River). It is apparent that very
strong tidal currents have a profound influence
on the distribution of sediments. Sands are
known to be transported for great distances in
front of these deltas; however, the geometry of
these sand bodies is unknown. Additional stud-
ies of this type delta are badly needed.
-------
B
4
5
o.
C.
B.
8.
S.
o.
a.
s
T
if t%. h.*«h »i«f •••fi** •• tt» *.nt r*««. n>if i* f **fT *f»«i*t
lV*t it Ik. «*!«• (f'fiH.f ».**.r< Ifc. *^*t***v* »-«t
T-. -.'.-r1-.'.I EACH- IIOGES-.'.-LT'.-.'.'-r-
O
n
o
3
n
Q
3
Q_
vt
O
X>
(V
CD
O
CU
-------
162
Rufus J. LeBlanc
Table 3. Examples of Ancient Deltaic Deposits
CtograpMc Occurrtnce Author
California
lllinou
Indiana
Iowa ft lUinoii
Kanaai
Louisiana
Michigan
Mill-. La. It. Ala.
Montana
Nebraska
New Mexico
New York
New York & Ontario
New York
North Dajtela
Ohio
Oklahoma
Oregon
Pa.. W. Va., Ohio
South Dakota
Tcxu
W. Va, P».. Ohio
Wyoming
Wyoming A Colorado
Several lUlu, U.S.A.
N. Appalachian*
Central Appalachians
Centra! Appalachian!
Upper Miki. embaymem
& Illinois basin
Upper Mia. Valley
OkU, low*. Mo., Kanx,
111.. Ind.. Ky.
OkU. to Pens.
Ceotral Gulf Coait
AJberu. Canada
Ireland'
ScetUnd
Todd and Monroe. 1968
Llneback. 1968
Swum 11 ol., I9&5
Hrabarand Pollcr, 1969
Wierand Girdley. 1963
Laury, 1968
Brown. 1967
H.ian, 1965
Clark and ROUK, 1971
Curuv 1970
Asacez. 1969
Galloway. 1968
Sirni, 1967
Shelion. 1972
SchlcE and Mocnch, 1961
Friedman and Johnioa. 1966
Lumadcn and Pellelier, 1969
Manini. 1971
Wolff. 1967
Sbelion. 1972
Knighu 1969
Leneand Owen. 1969
Biuch. 1953. 1971
Shelion, 1972
Vijher el at., 1971
Doll, 1964, 1966
Suavely n ni. 1964
Becrbowcr. 1961
Ferro and Cavaroc, 1969
Pettyjohn, 1967
Brown, 1969
Fuhet and McGoweo. 1969
Gregory, 1966
UBUoc. 1971
Nam, 1954
Shannon and Dahl, 1971
Wermund and Jcnkiru, 1970
Sbelion, 1972
Oooaldioa. 1969
Barlow and Haun. 1966
Dondanvillc, 1963
Hale. 196!
Paul I. 1962
Weiroer. 1961b
Weimer, 1963
Fisher «;o/., 1969
Fenn. 1970
Horowitz, 1966
Dcnnuon. 1971
Pryor. 1960, 1961
Swaon, 1964
Manoi, 1967
Wanltis it ol.. 1970
Mann and Thomas, 1968
Carrigy. 1971
Sbawa, 1969
Shephcard and Hills. 1970
Thachuk, 1968
Allen, 1962
Taylor. 1963
Hubbmrd, 1967
Greerumiih. 1966
Summary: Deltaic Sand Bodies
There are three basic types of deltaic sand
bodies: delta-fringe, abandoned distributary-
channel, and point-bar sands. The relative
abundance and general characteristics of these
sand bodies in the three types of deltas consid-
ered herein are summarized below.
Birdfoot-rype delta—The most common
sands are those of the delta-fringe environment.
These sands occur as relatively thin, wide-
spread sheets, and they contain a substantial
amount of clays and silts.
Abandoned distributary channels contain
varied amounts of sand, probably composing
less than 20 percent of the total delta sand con-
tent. These sand bodies are long and narrow,
are only slightly sinuous, and are encased in the
delta-fringe sands or prodelta clays, depending
upon channel depths and the distance that the
delta has prograded seaward.
Cuspate-arcuate type of delta—Delta-fringe
sand complexes are wide (width of delta),
though individual sand bodies are relatively
narrow, and are generally much cleaner than
delta-fringe sands of the birdfoot-type delta.
Distributary-channel sands and point-bar
sands are much more common than in bird-
foot-type deltas and can constitute up to 50
percent of the total sand content of the delta.
These two types of sands are encased in delta-
fringe and prodelta sediments.
Estuarine-type delta—Delta-fringe sands
appear to be much more common than distrib-
utary and point-bar sands. They probably ex-
tend for great distances within the marine envi-
ronment in front of the delta; however, their
geometry remains unknown.
Ancient Deltaic Deposits
Deposits of deltaic origin have been reported
from more than 40 states and from several for-
eign countries. Some examples are summarized
in Table 3.
COASTAi-lNTERDELTAIC MODEL OF.
SEHIMENTATIOK
Setting and General Characteristics
This type of sedimentation occurs in long,
narrow belts parallel with the coast where
shoreline and nearshore processes of sedimen-
tation predominate. The ideal interdeltaic de-
posit, as the name implies, occurs along the
coast between deltas and comprises mud fiats
and cbeniers (abandoned beach ridges) of the
chenier-plain complex and the barrier-island—
lagoon-tidal-channel complex (Fig. 23). It can
also occur along the seaward edge of a coastal
plain which is drained by numerous small
streams and rivers but is devoid of any sizable
deltas at the marine shoreline.
-------
Geometry of Sandstone Reservoir Bodies
163
Fie. 23—General selling and characteristics of coastal-inlerdellaic model of clastic sedimentation.
Source and Transportation of Sediments
Most of the sediments deposited are derived
from land, but minor amounts come from the
marine environment. A portion of the sediment
transported to the marine shoreline by rivers
and smaller streams is dispersed laterally by
marine currents for great distances along the
coast. Clays and fine silts are carried in suspen-
sion, and sand is transported mainly as bed
load or by wave action in the beach and near-
shore zone. The suspended silt and clay load is
dispersed at a rapid rate and is most significant
in the development of the mud fiats of the che-
nier plain. Lateral movement of the sand bed
load occurs at a relatively slow rate and is most
significant in the development of the cheniers
and the barrier-island complex.
A minor amount of sediment can also be de-
rived from adjacent continental-shelf areas if
erosion occurs in the marine environment.
Sedimentary processes and deposits of che-
nier plain—Major floods result in the sudden
large influx of sediments at river mouths. Much
of the suspended load introduced to the coastal-
marine environment is rapidly dispersed later-
ally along the coast by the predominant long-
shore drift. A considerable portion of this sus-
pended load is deposited along the shoreline
(on the delta flank) as extensive mud flats.
This period of regressive sedimentation (pro-
gradation or offiap) occurs in a relatively short
period when rivers are at flood stages (Fig.
24).
During long periods when rivers are not
flooding, the supply of sediment to the coast is
reduced considerably or is nil. Coastal-marine
currents and wave action rework the seaward
edge of the newly formed mud flat, and a trans-
gressive situation develops. A slight increase in
Flo. 74—Stages in development of a chenier plain.
-------
164
Rufus J. LeBlanc
sand supply can result in a regressive situation,
and the initial transgressive beach accumulation
will grow seaward by regressive beach accretion
to form a long, narrow, well-defined chenier on
the seaward edge of the extensive mud flat.
Another period of river flooding develops an-
other mud flat on the seaward edge of the cbe-
nier. During the subsequent nonflood season,
the coastal-transgressive processes produce an-
other beach ridge. Thus, over a long period, a
chenier plain consisting of mud flat and beach
ridges is constructed.
The width of a mud flat is varied and is de-
pendent on the magnitude and duration of a
river flood. The size of the chenier (height and
width) is determined by two factors: duration
of the nonflood season (absence of muds) and
magnitude of coastal-marine processes, includ-
ing storm tides and waves.
Small streams which drain to the coastline
across a chenier plain contribute little sedi-
ment to the chenier-plain environment. The
mouths of these streams are generally deflected
in the direction of the littoral drift.
Sedimentary processes and deposits of bar-
rier-island complex—The typical barrier-island
complex comprises three different but related
depositional environments: the barrier island,
the lagoon behind the barrier, and the tidal
channel-tidal deltas between the barriers.
The seaward -face of a barrier island is pri-
marily an environment of sand deposition.
Coastal-marine energy (currents and wave ac-
tion) is usually much greater than in the che-
nier-mud-flat regions. Sediments are trans-
ported along the coast in the direction of the
predominant littoral drift. Coarser sands are
deposited mainly on the beach and upper
shoreface, and finer sands are deposited in the
lower shoreface areas. Silts and clays are de-
posited in the lower shoreface zones on the ad-
jacent shelf bottom—at depths greater than
40-50 ft (12-15 m). Storm tides and waves
usually construct beach ridges several feet
above sea level, depending on the intensity of
storms, and also transport sandy sediments
across the barrier from the beach zone to the
lagoon.
Under ideal conditions, a barrier grows sea-
ward by a beach-sboreface accretion process to
produce a typical barrier-island sequence of
sediments which grades upward from fine to
coarse (Figs. .25, 27). The various organisms
which live in the beach, shoreface, and adjacent
offshore areas usually have a significant influ-
ence on the character of sedimentary struc-
tures.
Dry beach sand can be transported inland by
the wind and redeposited as dune sand on the
barrier, in the lagoon, or on the mainland
across the lagoon.
Tidal channel-tidal delta—Tidal action
moves a large quantity of water in and out of
lagoons and estuaries through the tidal chan-
nels which exist between barrier islands. These
channels are relatively short and narrow and
vary considerably in depth. Maximum chan-
nel depths occur where the tidal flow is
confined between the ends of barriers. The
channel cross section is asymmetric: one side
of the channel merges with the tidal flats and
spit; the opposite side of the channel has abrupt
margins against the barrier (Fig. 26).
As marine waters enter the lagoon or estuary
system during rising tides, the inflow attains its
maximum velocity in the deepest part of the
confined channel. The tidal flow is dispersed as
it enters the lagoon, and current velocities are
greatly reduced. The result is the deposition of
sediment in the form of a tidal delta which con-
sists of a shallow distributary channel separated
by sand or silt shoals. Similar tidal deltas are
also formed on the marine side of the system
by similar processes associated with the falling
or outgoing tide.
The depth of tidal channels and the extent of
tidal deltas are dependent on the magnitude of
the tidal currents. The deepest channels and the
largest deltas are associated with large lagoons
and estuaries affected by extreme tidal ranges.
Tidal channels migrate laterally in the direc-
tion of littoral drift by eroding the barrier head
adjacent to the deep side of the channel and by
spit and tidal-flat accretion on the opposite side.
Flo. 25—Stages in development of a barrier island.
-------
Geometry of Sondstone Reservoir Bodies
165
Lateral migration of the tidal system results ia
the deposition of the tidal-channel and tidal-
delta sequences of sediments.
Summary: Characteristics of Coastal-
Interdeltaic Deposits
The coastal-interdeltajc model of sedimenta-
tion is characterized by six distinct but related
types of deposits: mud flat, cbenier, barrier is-
land, lagoon, tidal channel, and tidal delta.
Characteristics of these deposits are summa-
rized in Figure 27.
. Three main types of sand bodies are associ-
ated with this model: barrier island, chenier,
and tidal channel-tidal delta. The barrier-is-
land sand body, which is the largest and most
significant of the three, is long (usually tens of
miles )and narrow (2—6 mi or 3-10 km), is
oriented parallel with the coastline, and attains
maximum thicknesses of 50-60 ft (15-18 m).
The chenier sand bodies are very similar to
those of the barriers; however, they are gener-
ally only about a third as thick. Tidal-channel
sand bodies are oriented perpendicular to the
barrier sands, and their thickness can vary con-
siderably (less than, equal to, or greater than
that of the barrier sands), depending on the
depth of tidal channels.
Ancient Coastal-interdeltaic Deposits
Examples of ancient coastai-interdeltaic de-
posits reported from 13 states are summarized
in Table 4.
Table 4. Examples of Ancient Coastal-
Interdetuic Deposits
CfOgraphlc
Author
Colorado
Florid*
Georgia
Illinoii
Louisiana
Louisiana & Arkansas
MonUna
New Mexico
New York
Oklahoma It Kansas
Texas
Wyoming
Griffith. 1966
Gremillion rl al., 1964
Haili and Hoyi, 1969
MacNeil. I9JO
Rusnak. 1957
Sloane, 1958
Thomas and Mann, 1966
Berg and Da vies, 1968
Cannon. 1966
Da vies n el.. 1971
Shcllon. 1963
Sabins. 1963
McCavc. 1969
Bass n cl.. 1937
Boyd and Dyer, 1966
Dodge, 1963
Fisher and McGowco. 1969
Fisher n al.. 1970
Shcllon. 1972
Harms rl el.. 1963
Jaclu. 1963
Miller. 1962
Paull, 1962
Scrulon, 1961
Wtimcr. 1961a
Fic. 26—Relation of Lida! channels and udtl dclus to
barrier u
EOLIAN MODEL or SAND DEPOSITION
Occurrence and General Characteristics
A very common process of sedimentation is
transportation and deposition of sand by ihe
wind. Two basic conditions are necessary for
the formation of windblown sand deposits:, a
large supply of dry sand and a sufficient wind
velocity. These conditions are commonly pres-
ent along coastlines characterized by sandy
beaches and also in semiarid regions and de-
serts, whsre weathering and fluvial sedimenta-
tion produce a large quantitiy of sand (Fig.
28).
Under certain conditions, sands on the
downstream parts of alluvial fans and along
braided streams are transported and redepos-
ited by the wind (Glennie, 1970). Sands origi-
nally deposited on point bars of meandering
streams and along distributary channels of some
deltas are also picked up by the wind and rede-
posited locally as dune sand. Similarly, sands
deposited along beacbes of the coastal-inierdel-
taic environments are redcposiled by onshore
winds as sand dunes on barrier islands or on
the mainland. Thus, the eolian process of sand
deposition is likely to occur within all models
of clastic sedimentation discussed in the pre-
ceding sections.
Eolian Transport and Sedimentation
The complex processes of sand transport and
deposition by the wind were studied and de-
-------
:§•
Is
B °-
I!
B
•a
o
a.
2.
LAGOON
5.1.
[•••••••I HOmiONTAl
"' ItODING
CKOSS-
'/£< ItOOINC
r5^!
I vvl
nyyi O*OANIC
r"TtTT MATEDI.Al
SAND
CIAY 1
SHI
COASTAL
PLAIN
-=--- MA«INt
LAGOON
-BARRIER ISLAND
^-IM^UUN ^"BAR
^=-^?T7r~r~^^7^~^7
•' /•' /'/.:•' v /'/'/ / '-/
/'' Li^-^X^-A^-fW,/
B — C
LAGOON AND BARRIER ISLAND
V
. v ...
v.y.v
• • \i • •
t- — -""
v .....
• v •
.v ....
— V
IK3-C-C
SEA
LtVEl
r
v'.
. • . v.'
' V .' '
' • v
V
aBM1
TC
TD
«>-tltl -it •*••<• xutvt*lf
• / / ( *'
rfe^fe^™^^^^
TC-TIOAl CHANNEL
10-IIDAl DELIA
TIDAL CHANNEL AND DELTA
CH -CHENIER
MF -MUDFLAT
MF
CH
v : •
• v
PLAIN
MF
-------
Geometry of Sandstone Reservoir Bodies
167
I- BARCHAN
1- TRANSVERSE
I - LONGITUDINAL (SEIF)
P- PARABOLIC
WADI, ARROTO OR
MAIDED STREAM
COASTAL DUNES
DESERT DUNES
^
C.-=\-: SHEET-
/•"•' SAND,
'• 3->FIUING,
"~IAGOON -
cs;-ffi-;rv:-5^?;*:.v.
*~:>?vV^ v>'^: £A£-^K'V'&S
ff.'.',' ^a *vC*;
'I-OJ63-J7
Flo. 28—Occurrences of eoliao sands in coastal and desert regions.
-------
168
Rufus J. LeBlanc
MAC* MM MM(t
Low moundi end snail elongate ridge* a Dunci against vegetation coalesce to fore
lew feet high occurring adjacent and lonj. slightly sinuous ridge or series of
parallel to sand beach and shoreline, ridges parallel to coastline. Closely
usually partly stsbllUtd by vegetation, associated vlch beach accretion ridgei foraeJ
by wave action. Characteristic of barrier
islands and shorelines on flankt of deltai.
ConcentrU with steep slope on concave U-shaped with open end toward beach (wind-
(lecward) side facing away fro* beach. Horni ward) and steep side away fron beach.
extend downwind. Can occur as scattered Kesults froa sand blowouts. Middle part
Isolated dunes or several barchans can Join noves forward (downwind) with respect to
to fore tlnuou* ridge which resembles sides. Long ares usually anchored by
transverse dunes. vegetation.
Dune* or ridges occur parallel or slightly Elongated parallel Co wind dlrecclon and
oblique Co coastline and elongaced In usually oblique or perpendicular Co
dlrecclon perpendicular Co effecclve wind coaiclloe. Cross lection svmecrlcal .
dlrecclon. Generally lyoaecrtcal In cross Separated from each other by flat area*.
secclon. Leeward tide steep and windward Self dunef are cpeclal Cype of
tide hat very gentle tlope. longitudinal dunes.
FIG. 29—Some common types of coastal dunes which also occur in deserts.
scribed by Bagnold (1941). Recently, Glennie
(1970) summarized this type of sedimentation
as observed under desert conditions.
The most common method of sand deposi-
tion is in the form of sand dunes. Many types
of dunes have been recognized and described
by numerous authors (Fig. 29). H. Smith
(1954) presented the following classification
and description of coastal dunes which can oc-
cur either under active or stabilized conditions.
1. Foredune ridges, or elongate mounds of sand up
to a few lens of feet ID height, adjacent and parallel
with beaches.
2. U-shaped dunes, arcuate to hairpin-shaped saud
ridges with the open end loward the beach.
3. Barchans, or cresceniic dunes, with a steep lee
slope OD the concave side, which faces away from the
beach.
4. Transverse dune ridges, trending parallel with or
oblique to the shore, and elongated in a direction es-
sentially perpendicular to the dominant winds. These
dunes are asymmetric in cross profile, having a gentle
slope on the windward side and a sleep slope on the
leeward side.
5. Longitudinal dunes, elongated parallel with wind
direction and extending perpendicular or oblique to tbe
shoreline; cross profile is typically symmetric.
6. Blowouts, comprising a wide variety of pits,
troughs, channels, and chute-shaped forms cutting into
or across other types of dunes or sand hills. Toe larger
ones are marked by conspicuous heaps of sand on tbe
landward side, assuming the form of a fan. mound, or
ridge, commonly with a slope as steep as 32° facing
away from the shore.
1. Attached dunes, comprising accumulations of
sand trapped by various types of topographic obstacles.
McK.ee (1966) described an additional type,
the dome-shaped dune, from White Sands Na-
tional Monument, and Glennie (1970) de-
-------
Geometry of Sandstone Reservoir Bodies
169
scribed the self dune of Oman, which is a spe-
cial type of longitudinal dune. Many other
dune types have been described; however, the
above types appear to be the most common.
Studies of modern eolian sand bodies—
Cooper (1958) reviewed the early studies of
sand dunes, mainly.by Europeans, and summa-
rized the status of dune reseach in North
America. Additional sand-dune studies in the
United States since 1959 were made in Alaska
by Black (1961), on the Texas coast by Mc-
Bride and Hayes (1962), on the Georgia coast
by Land (1964), in the Imperial.Valley of Cal-
ifornia by -Norris (1966), in coastal California
by Cooper (1967), and in the San Luis Valley
of Colorado by R. Johnson (1967). Additional
studies outside the United States were made in
southern Peru by Finkel (1959), in Baja Cali-
fornia by Inman et al. (1966), in Libya by Mc-
Kee and Tibbitts (1964), in Russia by Zenko-
vich (1967), and in Australia by Folk (1971).
During the past several years, some very im-
portant studies on eolian sands, which included
detailed observations of internal dune structure
and stratification in deep trenches cut through
dunes, were made along the Texas coast by
McBride and Hayes (1962), in White Sands
National Monument by McKee (1966), along
the Dutch coast by Jelgersma et a!. (1970), and
in the deserts of the Middle East by Glennie
(1970). These authors presented photographs
and sketches of various types of sedimentary
structures exposed in trench walls and de-
scribed their relations to dune types, wind re-
gime, and grain-size distribution. These studies
have provided some badly needed criteria for
recognition of ancient eolian sands. The follow-
ing summary of the geometry and general char-
acteristics of modern eolian sand bodies was
prepared largely from the references cited
above.
Summary: Coastal Eolian Sand Bodies
Coastal eolian sand bodies, consisting of sev-
eral types of dunes, are very long and quite
narrow; they range in thickness from a few feet
to a few hundreds of feet, and are aligned par-
allel with or oblique 10 the coastline. Because
these sands are derived from beach deposits
and form in vegetaled areas, they commonly
contain fragments of both shells and plants.
They are characterized by high-angle crossbed-
ding and are usually well sorted. The adjacent
and laterally equivalent beach deposits are gen-
erally horizontally bedded and have some low-
angle crossbedding.
Summary: Eolian Sand Bodies of
Desert Regions
Desert eolian sand bodies differ from coastal
eolian sands mainly in their distribution. The
internal sedimentary structures and their rela-
tions to dune types are similar (Bigarella et al.,
1969). Self dunes are products of two wind di-
rections and appear to occur more commonly
in desert areas. These dunes are characterized
by high-angle crossbedding in two directions.
Ancient Eolian Deposits
Ancient eolian deposits have been reported
from the Colorado Plateau by Baars (1961)
and Stokes (1961, 1964, 1968), from the
southwestern United States by McKee (1934),
from England by Laming (1966), and from
Brazil and Uruguay by Bigarella and Salamuni
(1961). Criteria for recognition of eolian de-
posits have been summarized by Bigarella
(1972).
MARINE CLASTIC SEDIMENTATION
Transportation and deposition of sand in the
marine environment occur under a wide range
of geologic and hydrologic conditions, ranging
from those of the coastal shallow-marine envi-
ronments to the deeper water environments of
the outer continental shelves, the slopes, and
the abyssal plains (Fig. 30).
As indicated in the Introduction, sands de-
posited under regressive (progradational) con-
ditions within the coastal shallow-marine envi-
ronments are considered herein as products of
either the coastal-interdeltaic or the deltaic
model of sedimentation. Other important shal-
low-marine sand bodies are produced as a re-
sult of marine transgressions.
During the past several years, studies made
principally by the major oceanographic institu-
tions on the modern deep-marine environments
and research by petroleum geologists, univer-
sity professors, and graduate students on an-
cient clastic sediments of various geologic ages
have revealed that sand bodies of deep-marine
origin are rather common throughout the
world. Although most geologists now accept
the fact that some sands are of deep-marine or-
igin, our understanding of the various geologic
processes which produce these sand bodies is
relatively poor. The writer's personal experi-
ence with this type of clastic sedimentation is
limited; however, on the basis of familiarity
with the literature on marine sediments, it ap-
pears that most deep-marine sands are depos-
-------
170
Rufus J. LeBlanc
tUl-IIANlGKlltr
SC-1UIUAIINE CANTON
SI-Hll«illH( IAN
FIG. 30—Deposition of sand in marine environments.
ited under three principal types of environmen-
tal conditions: (1) on the outer shelf, the
slope, and the continental rise, as a result of
slumping, sliding, and tectonic activity such as
earthquakes; (2) in abyssal plains, by density
(turbidity) and bottom currents; and (3) in
submarine canyons, fan valleys, and fans, by
both bottom and density currents.
Only two of these several types of marine
sands are discussed in this paper: (1) the shal-
low-marine sands deposited as a result of trao-
gressive-marine sedimentation associated with
the shifting of deltas and (2) the deep-marine
sands deposited In submarine canyons, fan val-
leys, and fans.
Transgressive-Marine Model of
Clastic Sedimentation
Setting and general characteristics—Deposi-
tion of clastic sediments during periods of ma-
rine transgressions (onlap) is a common pro-
cess of sedimentation in most basins. There are
two basic types of marine transgression: that
which is associated with the shifting of deltas as
a result of major river diversions during a pe-
riod of standing sea level, and that which oc-
curs as a result of a relative rise in sea level
(due to subsidence of a coastal plain or eustatic
rise in sea level). The inland and lateral extents
of marine transgressions resulting from delta
shifts are limited in size, depending on the di-
-------
Geometry of Sandstone Reservoir Bodies
171
mensions of the abandoned deltaic plains. Ma-
rine transgressions resulting from relative
changes in sea level extend over much broader
regions and are commonly referred to as re-
gional transgressions. Their dimensions are
governed mainly by the topography of the
coastal plain being transgressed and by the
amount of relative rise in -sea level. Thus, trans-
gressive-marine deposition can occur locally
over abandoned deltas or regionally over eo-
lian, alluvial, interdeltaic, and deltaic deposits
of a large part of a coastal plain.
Modern marine transgressions resulting from
major changes in drainage and delta shifts have
been described by several authors: Russell,
1936; Russell and Russell, 1939; Kruit, 1955;
van Straaten, 1959; Scruton, 1960; Curray,
1964; Coleman and Gagliano, 1964; Rainwa-
ter, 1964; Coleman, 1966b; Scott and Fisher,
1969; L. Brown, 1969; and Oomkens, 1970.
Sources, transportation, and deposition of
sediments—After a delta is abandoned because
of upstream channel diversion, a very signifi-
cant change occurs in conditions of sedimenta-
tion. The abandoned deltaic plain and subaque-
ous delta front no longer receive sediment and
gradually subside owing to the compaction of
the deltaic deposits. The seaward edge of the
abandoned delta is attacked by marine wave
and current action and recedes landward at rel-
atively slow rates. As marine processes erode
the upper part of the deltaic sequence, the
sandy sediments within the sequence are win-
nowed and deposited along the advancing
shoreline as barrier islands, beaches, and shal-
low-marine sands; finer sediments are deposited
farther offshore. Thus, the transgressive-marine
depositional profile is characterized by sands
and shell material nearshore and by progres-
sively finer sediments offshore. Over a period of
time, as the transgression proceeds inland, the
thin veneer of shallow-marine sands which are
deposited over the underlying delta sediments is
in turn overlain by marine silts and clays.
Stages in the development of such a trangres-
sive-marine sand body are illustrated in Figure
31.
Character of sediments—This type of sedi-
mentation, although largely restricted in extent
to abandoned deltas and adjacent and laterally
>,£"< - '*/ f*~^ *. v-r
--•ii"-- ., «,/., V. ' PIOGSADING BEUA
"*-'"'- ' A^ -^~-
Of OEIIAIC SEQUENCE UNDE« «EG«ESSIVE CONDITIONS
' MNIR SANDS
AND SIITS
SAND AM
tHtui
'"",*
AtANDONMENI Of DELIA
MAtlNE ItANSGIESSION AND
B' DlvdOfMlNl OF NEW DEUA.
EAJIt SIAGf Of IIANSGIfSSIVE MAUN! SEDIMBNIA1 ION
ftAMSGMSllvr /
MAUN! SfOUtNCI \
At!A O' '
TIANSGftCSSIVt '. -'.>T
MAtINf SI DIMINUTION
~~".\ MAIINt ClATl
TIANSGtf SStVl MAtlNf SAND
'DCUAIC UOUFNCt
CONIWUID ItANSGKiJIVE MAtIN! SIOIHENIAIIOK AND OfVSIOfMfWI Of
ItiNSG8ESSIV[ SEQUENCE 0! UDIMiKIS
Flc. 31—TransgTessive-m»riDc Kdimenlation reiulling from delti shifts.
-------
172
Rufus J. LeBlanc
equivalent interdeltaic and offshore-marine en-
vironments, is significant because it produces
very diagnostic blanketlike layers of marine
sediments (thin shallow-marine sand overlain
by clays) which separate the individual deltaic
units (Fig. 31). These layers usually provide the
only good correlations within thick deltaic fa-
cies, and the marine shales act as impervious
seals between deltaic sand bodies. The trans-
gressive-marine sands containing calcareous
shell material usually become cemented and
thus do not form very efficient reservoirs.
Submarine Canyon-Fan Model of
Clastic Sedimentation
Occurrence and general characteristics—The
occurrence of modern and Pleistocene sands in
deep-marine environments of the world is well
documented as a result of numerous deep-sea
investigations by oceanographic institutions
during the past 20 years. Although there is
much controversy regarding the origin of these
sands, it is certain that such sands do exist. An
analysis of the literature reveals that some of
the most common deep-sea sands are those as-
sociated with submarine canyons and fans.
(For a discussion of types of submarine can-
yons, troughs, and valleys, the reader is re-
ferred to Shepard and Dill, 1966.)
Submarine canyons and fans are common
features associated with continental shelves,
slopes, and rises. The canyons and fans off the
Pacific Coast of the United States and Canada
have received the most attention. Significant
papers on these features off the coasts of Wash-
ington, Oregon, California, and Baja Califor-
nia, and off the Gulf and Atlantic coasts, are
listed in the selected references. Also included
are references to papers on canyons and fans in
the Mediterranean, the Atlantic Ocean off Af-
rica, and the Indian Ocean off Pakistan.
Characteristics and origin of submarine can-
yons have been discussed by numerous authors
(for summary, see Shepard and Dill, 1966).
Although it is still uncertain how some deep-sea
canyons and valleys originated, it is now rea-
sonably well established that a large number of
canyons and fans are related to rivers, and that
they were formed during stages of low sea level
of the Quaternary Period (Figs. 32, 33). For
AUUVIAl
VMttt
CONllNtNUI IISI
U'Ptl AIMSAI
(NllfNCHED
via IT
' JUIAfllAL ri
\ AND SKMI
I.'-::>SVIMAIINE
''•'•'•I CANTON
•-1§~"
'vkjtf^jji-fo'
FIG. 32—Stages in development of submarine canyon and fan.
Stage A: Standing sea-level situation. Development of alluvial valley and delta and deposition of marine
clays on shelf and slope. Base of aggrading river is well below sea level.
Stage B: Falling and low-sea-level situation. Development of entrenched-valley system on coastal plain and
of submarine canyon offshore. Bases of entrenched valley (near coast) and of canyon are well below sea level.
Rates of sedimentation are very high. Material removed by canyon-culling and sediments flowing through
canyon while it formed are deposited as extensive submarine fan.
Stage C: Rising and standing sea-level situation. AJIuvjalion of entrenched-valley system and partial filling
of cajiyon. Rates of sedimentation are greatly reduced after sea. level reaches a stand. Slight modification of
fan by normal-marine processes occurs.
-------
Geometry of Sandstone Reservoir Bodies
173
FJC. 33—Relation of submarine-fan deposits to submarine-canyon and entrenched-valley system.
example, the Mississippi Canyon off southern
Louisiana is a continuation of the late Pleisto-
cene Mississippi entrenched valley system
(Fisk, 1944; Osterboudt, 1946; Fish and Mc-
Farlan, 1955; and Bergantino, 1971). Also, the
Astoria canyon and fan off Oregon are related
to the Columbia River (Duncan and Kulm,
1970); the Newport Canyon is related to the
Santa Ana River of California (Felix and Gors-
line, 1971); the Congo Canyon connects with
the Congo River (Heezen et al, 1964); the
Monterey and Soquel canyons and fans occur
off the Great Valley of California (Martin and
Emery, 1967); the Bengal deep-sea fan and the
"Swatch-of-No-Ground" canyon occur off the
Ganges River delta (Curray and Moore,
1971); and the Inguri canyon is related to a
river flowing in the Caspian Sea (Trimonis and
Shimkus, 1970). The National Geographic
magazine maps of the Indian and Atlantic
Ocean floors (Heezen and Tharp, 1967, 1969)
show large fans off the Indus and Amazon Riv-
ers and also off the Laurentian Troueh, and the
Hudson Canyon is associated with the Hudson
River. Seismic reflection surveys between can-
yon heads (on shelves) and the coastline most
probably will reveal more examples of canyons
related to entrenched river valleys on land.
There is an extremely wide variation in the
size of submarine canyon-fan systems. Some of
the small ones off California studied by Gors-
line and Emery (1959) include short canyons
5-10 mi (8-16 km) long and fan areas of
about 50 sq mi (130 sq km). The largest can-
yon-fan systems studied thus far are those of
the Congo, Ganges, and Rhone Rivers. The
Bengal fan is 2,600 km long and 1,100 km
wide; the Congo fan is more than 520 km long
and 185 km wide; and one of the largest fans
off the Pacific coast of the United States, the
Delgade fan, is 300 km long and 330 km wide
(Normark, 1970). Menard (1960) discussed
the dimensions of several other fans.
Some very significant studies of deep-sea
sands associated with canyons and fans—bas:d
on core, seismic reflection, and bathymetric
data, and bottom observations and photography
by divers—have been made during the past 3
years (Winterer el al., 1968; Carlson and Nel-
son, 1969; Shepard el al.. 1969; Curray and
Moore, 1971; Normark, 1970; Nelson et al.
1970; Piper, 1970; Duncan and Kulm, 1970;
and Felix and Gorsline, 1971).
Physiographic features.—Detailed bathy-
metric surveys over several canyons and fans of
various sizes have revealed that these subma-
rine features are characterized by physio-
graphic features very similar to those of subaer-
ial alluvial fans. The canyons are V-shaped and
have steep walls and gradients. The surfaces of
the fans are characterized by lower gradient
distributary channels with natural levees and by
topographically low interchannel areas. Some
fans arc crossed by relatively large fan valleys
-------
174
Rufus J. LeBlanc
Flo. 34—Submarine-canyon and fan model of clastic sedimentation.
which also have natural levees. The principal
physiographic features of a typical canyon and
fan are illustrated in a generalized fashion in
Figure 34.
The overall shape of a submarine fan can ei-
ther be symmetrical or asymmetrical, depend-
ing on strong current directions and on the
presence of high topographic features on the
abyssal plains. Sizes of the distributary chan-
nels and natural levees are widely varied; the
larger channels usually have the highest and
broadest natural levees. The lower parts of fans
merge with the abyssal plains. Channels on fan
surfaces were probably formed by depositional
processes. Erosional channels that have been
reported probably represent an entrenchment
stage, as is the case with subaerial alluvial fans.
Many canyons were cut across continental
shelves and slopes, and the fans were con-
structed at the base of the slopes or on the con-
tinental rises. Some canyons presently do not
extend landward across the continental shelves
(e.g., the Mississippi Canyon) because they
have been filled with sediments. Seismic surveys
reveal that this type of canyon was once con-
nected with inland entrenched valley systems.
Longitudinal profiles of canyons and fans are
concave upward. The steepest gradients occur
in the upper (landward) portions of canyons,
and the lowest gradients occur on the outer or
lower portions of fans.
Depositional processes and character of sedi-
ments—It is absolutely certain that large quan-
tities of sediment, including a significant
amount of sand, have been transported through
submarine canyons and deposited as submarine
fans in deep-sea environments. The manner in
which these sediments were transported, espe-
cially the sands, is much less certain. Nearly 2
decades ago, some very strong statements were
made by oceanographers regarding the turbid-
ity-current origin of both the canyons and the
fan deposits. Although no one had actually
seen or measured a turbidity, current in a can-
yon or over a submarine fan, the turbidity-cur-
rent concept was very popular with most
oceanographers during the early 1950s. During
the past 20 years, numerous additional observa-
tions have been made, but no one has yet seen
a live turbidity current in a natural marine en-
vironment. On the basis of direct observations
of the ocean bottom and sedimentary structures
in cores, many oceanographers now believe
that some submarine-fan sand deposits were
transported mainly by normal bottom currents,
especially during low stages of sea level of the
Pleistocene. A typical example is the origin of
the sand associated with the Mississippi cone in
-------
Geometry of Sandstone Reservoir Bodies
175
ACllVf MltlllUUIT
•IAHOOXIO IMMilllll UH CK1NNII-IIU
IKIIItHAKMIt D(«1P1IIII«1 CIAWIOI MAT CONSIII
CIUNMII UNO!
Fio. 35—Generalized distribution of submarine-fan deposits.
the Gulf of Mexico off southern Louisiana.
Greenman and LeBlanc (1956) did not con-
sider these sands to be of turbidity-current ori-
gin, but Ewing et al. (1958) were certain that
the sands were transported and deposited by
turbidity currents. Twelve years later, Huang
and Goodell (1970) concluded, on the basis of
detailed studies of sedimentary structures ob-
served in numerous cores, that the sands are
not of turbidity-current origin, but that the
mechanisms of transport are bottom currents,
differential pelagic settling, and mass movement
by sliding and slumping. Walker and Massingill
(1970) reported that part of the Mississippi
cone sediments were recently involved in large-
scale slumps. They presented evidence that one
slump moved from near the mouth of the Mis-
sissippi Canyon southeastward for at least 160
n. mi. Thus, the origin of these deep-sea sands
and many others remains a problem.
Regardless of the mechanisms of sediment
transport through submarine canyons and of
deposition of fans, the general nature and dis-
tribution of fan deposits have been determined
for .several fans. The coarsest and most poorly
sorted sediments occur in canyons. Sands are
common in distributary channels and fan val-
leys and on the lower parts of the open fan.
Sandy sediments also occur on natural levees,
but the intercbannel areas are characterized by
fine-grained sediments (Fig. 35). Core data
from several fans indicate that sand bodies are
usually thin and very lenticular, and are inter-
bedded with fine-grained sediments.
For details concerning the sedimentary struc-
tures which characterize submarine-canyon and
fan deposits, the reader is referred to Carlson
and Nelson (1969); Sbepard et al. (1969);
Stanley (1969); Huang and Goodell (1970);
and Haner (1971).
Horn ct al. (1971) described the characteris-
tics of sediments related to submarine canyons,
fans, and adjacent abyssal plains of the north-
east Pacific Ocean off Alaska, Canada, Wash-
ington, Oregon, and northern California. They
interpreted sediments with a wide range in
layer thickness, with graded and nongraded lay-
ers, and with sand in the basal parts of graded
units to be proximal turbidites related to main
routes followed by turbidity currents (probably
channels). The finer grained sediments, mainly
graded silts and clays, were interpreted as distal
turbidites deposited beyond the main avenues
of turbidite flows.
It is the opinion of the writer that many of
the submarine canyons and related fans which
now are found off rivers are the products of en-
trenchment (canyons) and deposition (fans)
during stages of low sea level of the Pleisto-
cene. Oceanograpbers who have studied several
of these fan deposits have concluded that they
are of Miocene to Pleistocene age. The geo-
logic-age determinations were made on the ba-
sis of present sediment load of the related riv-
ers and known thickness of fan deposits. This
writer suggests that rates of sedimentation were
probably several times greater during Pleisto-
cene low-sea-level stages than at the present
time (period of higher and standing sea level)
and, consequently, that the fan deposits are
probably chiefiy of Pleistocene age.
Ancient examples of submarine canyon and
fan deposits—Some examples of ancient depos-
its of submarine canyon and fan origin have
been described from the Gulf coast by Oster-
houdt (1946), Bornhauser (1948, 1960), Hoyt
(1959), Paine (1966), and Sabate (1968);
from California by Sullwold (1960), Martin
(1963), Bartow (1966), Dickas and Payne
(1967), Normark and Piper (1969), Piper and
Normark (1971), Davis (1971), Fischer
(1971), and Shelton (1972); from Canada by
Hubert et al. (1970); from Europe by Walker
(1966), Stanley (1967, 1969), and Kelling and
Woollands (1969); and from Australia by Co-
nolly (1968).
-------
176
Rufus J. LeBlanc
SELECTED REFERENCES
Al-Habeeb, K. H., 1970, Sedimenlology of the flood-
plain sediments of the middle Euphrates River, Iraq:
Baghdad Univ., MS thesis.
Allen, J. R. L., 1962, Petrology, origin, and deposition
of the highest lower Old Red Sandstone of Shrop-
shire, England: Jour. Sed. Petrology, v. 32, p. 657-
697.
1964, Studies in fluvialile sedimentation: six cy-
clothems from the lower Old Red Sandstones, An-
glo-Welsh basin: Sedimentology, v. 3, p. 163-198.
1965a, A review of the origin and characteris-
tics of Recent alluvial sediments: Sedimentology, v.
5, p. 89-191.
1965b, Coastal geomorphology of eastern Nige-
ria: beach-ridge barrier-islands and vegetated tidal
flats: Ceologie en Mijnbouw, v. 44, p. 1-21.
1965c, Late Quaternary Niger delta and adja-
cent areas: sedimentary environments and lilhofa-
cies: Am. Assoc. Petroleum Geologists Bull., v. 49,
no. 5, p. 547-600.
1970, Sediments of the modem Niger delta, in
Deltaic sedimentation—modern and ancient: Soc.
Econ. Paleontologists and Mineralogists Spec. Pub.
15. p. 138-151.
American Petroleum Institute, 1971: Petroleum facts
and figures.
Anderson, G. S., and K. M. Hussey, 1962, Alluvial fan
development at Franklin Bluff, Alaska: Iowa Acad.
Sci. Proc., v. 69, p. 310-321
Andrews, P. B., 1970, Facies and genesis of a hurri-
cane washover fan, St. Joseph Island, central Texas:
Texas Univ. Bur. Econ. Geology Rept. Inv. 67, 147 p.
Anstey, R. L., 1965, Physical characteristics of alluvial
fans: U.S. Army Quartermaster Research and Eng.
Command, Earth Sci. Div. Tech. Rept. ES-20, 109 p.
Arnborg, L. E., 1948, The delta of the Angerman
River: Geog. Annaler, v. 30, p. 673-690.
Asseez, L. O., 1969, Paleogeography of Lower Missis-
sippian rocks of Michigan basin: Am. Assoc. Petro-
leum Geologists Bull., v. 53, no. 1, p. 127-135.
Attia, M. I., 1954, Deposits in the Nile Valley and the
delta: Cairo, Egypt Geol. Survey, 356 p.
Axelson, V., 1967, The Laiuure delta, a study of del-
taic morphology and processes: Geog. AnnaJcr, v.
49, p. 1-127.
Baars, D. L., 1961, Permian blanket sandstones of Col-
orado Plateau, in J. A. Peterson and J. C. Osmond,
eds., Geometry of sandstone bodies: Am. Assoc. Pe-
troleum Geologists, p. 179-219.
Bagchi, K., 1944, The Ganges delta: Calcutta, Calcutta
Univ. Press, 157 p.
Bagnold, R. A., 1941, The physics of blown sand and
desert dunes: New York, W. Morrow and Co., 265 p.
Bakker, W. T. J. N. P., and T. Edelman, 1965, The
coast line of river-deltas: 9th Conf. on Coastal En-
gineering Proc., p. 199-218.
Barlow, J. A., Jr., and J. D. Haun, 1966, Regional stra-
tigraphy of Frontier Formation and relation to Salt
Creek field, Wyoming: Am. Assoc. Petroleum Geol-
ogists Bull., Y. 50, p. 2185-2196.
Bartow, J. A., 1966, Deep submarine channel in upper
Miocene, Orange County, California: Jour. Sed. Pe-
trology, v. 36, no. 3, p. 700-705.
Bass, N. W., el al., 1937, Origin and distribution of
Bartlesvillc and Burbani shoestring oil sands in
parts of Oklahoma and Kansas: Am. Assoc. Petro-
leum Geologists Bull., v. 21, p. 30-66.
Bales, C. C., 1953, Rational theory of delta formation:
Am. Assoc. Petroleum Geologists Bull., v. 37, p.
2119-2162.
Beall, A. O., Jr., 1968, Sedimentary processes operative
along the western Louisiana shoreline: Jour. Sed. Pe-
trology, v. 38, p. 869-877.
-Beaty, C. B., 1961, Boulder deposit in Flint Creek Val-
ley, western Montana: Geol. Soc. America Bull., v.
72, p. 1015-1020.
1963, Origin of alluvial fans, White Mountains,
California and Nevada: Assoc. Am. Geographers
Annals, v. 53, p. 516-535.
Beerbower, J. R., 1961, Origin of cyclothems of the
Dunkard Group (Upper Pennsylvanian-Lower Per-
mian) in Pennsylvania, West Virginia, and Ohio:
Geol. Soc. America Bull., v. 72, p. 1029-1050.
1964, Cyclothems and cyclic depositional mech-
anisms .in alluvial plain sedimentation: Kansas Geol.
Survey Bull. 169, v. 1, p. 31-42.
1969, Interpretation of cyclic Permo-Carbonif-
erous deposition in alluvial plain sediments in West
Virginia: Geol. Soc. America Bull., v. 80, p. 1843-
1848.
Berg, R. R., 1968, Point bar origin of Fall River sand-
stone reservoirs, northeastern Wyoming: Am. Assoc.
Petroleum Geologists Bull., v. 52, no. 11, p. 2116-
2122.
and B. C. Cook, 1968, Petrography and origin
of lower Tuscaloosa sandstones, Mallalieu field, Lin-
coln County, Mississippi: Gulf Coast Assoc. Geol.
Socs. Trans., v. 18, p. 242-255.
and D. K. Davies, 1968, Origin of Lower Cre-
taceous Muddy Sandstone at Bell Creek field, Mon-
tana: Am. Assoc. Petroleum Geologists Bull., v. 52,
no. 10, p. 1888-1898.
Bergantino, R. N., 1971, Submarine regional geomor-
phology of the Gulf of Mexico: Geol. Soc. America
Bull., v. 82, p. 741-752.
Bernard, H. A., 1965, A resume of river delta types:
Am. Assoc. Petroleum Geologists Bull., v. 49, p.
334-335.
and R. J. LeBlanc, 1965, Resume of Quater-
nary geology of the northwestern Gulf of Mexico
province, in Quaternary of the United States, Review
volume for 7tb Cong., Imernat. Assoc. for Quater-
nary Research: Princeton, New Jersey, Princeton
Univ. Press, p. 137-185.
and C. F. Major, 1963, Recent meander belt
deposits of the Brazos River: an alluvial "sand"
model:. Am. Assoc. Petroleum Geologists Bull., v.
47, no. 2, p. 350.
R. J. LeBlanc, and C. F. Major, 1962, Recent
and Pleistocene geology of southeast Texas, in Geol-
ogy of the Gulf Coast and central Texas and guide-
book of excursions, Geol. Soc. America 1962 Ann
Mtg.: Houston Geol. Soc., p. 175-224.
C. F. Major, and B. S. Parrott, 1959, The GaJ-
veslon barrier island and environs—a model for pre-
dicting reservoir occurrence and trend: Gulf Coast
Assoc. Geol. Socs. Trans., v. 9, p. 221-224.
el al., 1970, Recent sediments of southeast
Texas; a field guide to the Brazos alluvial and del-
taic plains and the Galveston barrier island complex:
Texas Univ. Bur. Econ. Geology Guidebook No. 11.
Beulner, E. C., el al.. 1967, Bedding geometry in a
Pennsylvanian channel sandstone: Geol. Soc. Amer-
ica Bull., v. 78, no. 7, p. 911-916.
Bigarella,.J. J., 1972, Eolian deposits—their character-
istics, recognition, and importance, in J. K. Rigby
and W. K. Hamblin, eds.. Recognition of ancient
-------
Geometry of Sandstone Reservoir Bodies
177
sedimentary environments: Soc. Econ. Paleontolo-
gists and Mineralogists Spec. Pub. 16, 340 p.
and R. Salamuni, 1961, Early Mesozoic wind
patterns as suggested by dune bedding in the Botu-
catu Sandstone of Brazil and Uruguay: CeoL Soc.
America Bull., v. 72, p. 1089-1106.
tl al., 1969, Coastal dune structure from
Parana, Brazil: Marine Geology, v. 7, p. 5-55.
Bird, E. C. F., 1961, The coastal barriers of East
Gippsland, Australia: Jour. .Geology, v. 127, p. 460-
468.
1962, The river deltas of the Gippsland l»kes:
Royal Soc. Victoria Proc., v. 75, pt. 1, p. 65-74.
Black, R. F., 1961, Eolian deposits of Alaska: Arctic,
v. 4, no. 2, p. 89-111.
Blackwelder, E., 1928, Mudflow as a geologic agent in
semi-arid mountains: Geol. Soc. America Bull., v.
'39, p. 465-483.
1931, Desert plains: Jour. Geology, v. 39, p.
133-140.
Blissenbach, E., 1952, Relation of surface angle distri-
bution to particle size distribution on alluvial fans:
Jour. Sed. Petrology, v. 22, p. 25-28.
1954, Geology of alluvial fans in semi-arid re-
gions: Geol. Soc. America Bull., v. 65, p. 175-189.
Bluet, B. J., 1964, Sedimentation of an alluvial fan in
southern Nevada: Jour. Sed. Petrology, v. 34, no. 2,
p. 395-400.
1965, The sedimentary history of some Triassic
conglomerates in The Vale of Glamorgan, South
Wales: Sedimenlology, v. 4, p. 225-246.
1967, Deposition of some upper Old Red Sand-
stone conglomerates in the Clyde area: a study in
the significance of bedding: Scottish Jour. Geology,
Y. 3, no. 8, pt. 2, p. 139-167.
Boggs, S., Jr., 1966, Petrology of Minium Formation,
east-central Eagle Count)1, Colorado: Am. Assoc.
Petroleum Geologists Bull., v. 50, no. 7, p. 1399-
1422.
Bolyard, D. W., 1959, Pennsylvanian and Permian stra-
tigraphy in Sangre De Cristo Mountains between La
Vela Pass and Weslclifie, Colorado: Am. Assoc. Pe-
troleum Geologists Bull., v. 43, p. 1896-1939.
Bonham-Carier, G. F., and A. J. Sutherland, 1967, Dif-
fusion and settling of sediments at river mouths: a
computer simulation model: Gulf Coasl Assoc.
Geol. Socs. Trans., v. 17, p. 326-338.
Boothroyd, J. C., 1970, Recent braided stream sedi-
mentation, south-central Alaska (abs.): Am. Assoc.
Petroleum Geologists Bull., v. 54, no. 5, p. 836.
Bornhauser, M.. 1948, Possible ancient submarine can-
yon in southwestern Louisiana: Am. Assoc. Petro-
leum Geologists Bull., v. 32, p. 2287-2290.
1960, Depositional and structural history of
northwest Hartburg field. Newton County, Texas:
Am. Assoc. Petroleum Geologists Bull., v. 44, p.
458-470.
Bouma. A. H., 1965. Sedimentary characteristics of
samples collected from some submarine canyons:
Marine Geology, v. 3, no. 4, p. 291-320.
Boyd, D. R., and B. F. Dyer, 1966, Frio barrier bar
system of south Texas: Am. Assoc. Petroleum Geol-
ogists Bull., v. 50, p. 170-178.
Brady, L. L., 1969, Stratigraphy and petrology of the
Morrison Formation (Jurassic) of the Canon Cily,
Colorado area: Jour. Sed. Peirology, v. 39, no. 2, p.
632-648.
Brctz, J. H., and L. Horberg, 1949, The Ogallala For-
mation west of the Llano Eslacado: Jour. Geology,
v. 57, no. 5, p. 477-490.
Brice, J. C., 1964, Channel pattern and terraces of the
Loup Rivers in Nebraska: U.S. Geol. Survey Prof.
Paper 422-D, 41 p.
Brickman, R., and L. J. Pons, 1968, A pedo-geomor-
phological classification and map of the Holocene
sediments in the coastal plain of the three Guianas:
Wageningen, Netherlands Soil Survey Institute, 40 p.
Brown, L. F., Jr., 1969, North Texas (eastern shelf)
Pennsylvanian delta systems, in Delta systems in ex-
ploration for oil and gas: Texas Univ. Bur. Econ.
Geology, p. 40-53.
Brown, S. L., 1967, Stratigraphy and depositional envi-
ronment of the Elgin Sandstone (Pennsylvanian) in
south-central Kansas: Kansas Geol. Survey Bull.
187, pt. 3, 9 p.
Bufnngton, E. C., 1952, Submarine "natural levees":
Jour. Geology, v. 60, p. 473-479.
Bull, W. B., 1960, Types of deposition on alluvial fans
in western Fresno County, California (abs.): Geol.
Soc. America Bull., v. 71, p. 2052.
1962, Relation of textural (cm) patterns to de-
positional environment of alluvial fan deposits: Jour.
Sed. Petrology, v. 32, p. 211-216.
1963, Alluvial fan deposits in western Fresno
County, California: Jour. Geology, v. 71, p. 243-
1964, Geomorpbology of segmented alluvial
fans in western Fresno County, California: U.S.
Geol. Survey Prof. Paper 352-E, p. 59-129.
1968, Alluvial lans: Jour. Geol. Education, v.
16, p. 101-106.
1969, Recognition of alluvial-fan environments
in stratigraphic record (abs.): Am. Assoc. Petro-
leum Geologists Bull., v. 53, p. 710.
1972, Recognition of alluvial-fan deposits in
the slrau'graphic record, in J. K. Rigby and W. K.
Hamblin, eds., Recognition of ancient sedimentary
environments: Soc. Econ. Paleontologists and Miner-
alogists Spec. Pub. 16, 340 p.
Busch, D. A., 1953, The significance of deltas in sub-
surface exploration: Tulsa Geol. Soc. Digest, v. 21,
p. 71-80.
1971, Genetic units in delta prospecting: Am.
Assoc. Petroleum Geologists Bull., v. 55, p. 1137-
1154.
Buttner, P. J. R., 1968, Proximal continental rhythmic
sequences in the Genesee Group (lower Upper De-
vonian): Geol. Soc. America Spec. Paper 106, p.
109-126.
Byers, P. R., 1966, Mineralogy and origin of the East-
end and Whitemud Formations of south-central and
southwestern Saskatchewan and southeastern
Alberta: Kingston, Ontario, Queen's University, Mas-
ter's thesis.
Byrne, J. V., el oh. 1959, The chenier plain and its
stratigraphy, southwestern Louisiana: Gulf Coast
Assoc. Geol. Socs. Trans., v. 9, p. 237-260.
Cadwell, D. H., and J. H. Moss, 1969, Floodplain sedi-
ments and their movement during a high discharge,
near Manheim, Pennsylvania: Pennsylvania Acad.
Sci. Proc., v. 43, p. 185-186.
Cannon, J. L., '1966, Outcrop examinalion and inter-
pretation of paleocurrent paiterns of the Blackleaf
Formation near Great Falls, Montana: Billings Geol.
Soc. 17th Field Conf. Guidebook, p. 71-111.
Carlson, P. R.. and C. H. Nelson, 1969, Sediments and
sedimentary structures of the Astoria submarine can-
yon-fan system, northeast Pacific: Jour. Sed. Petrol-
ogy, v. 39, p. 1269-1282.
Carrigy, M. A., 1971, Deltaic sedimentation in Atha-
-------
178
Rufus J. LcBlanc
basca tar sands: Am. Assoc. Petroleum Geologists
Bull., v. 55. p. 1155-1169.
Casey, S. R., and R. B. Cantrell, 1941, Davis sand lens,
Hardin field. Liberty County, Texas, in A. I. Levor-
scn, ed., Straligraphjc type oil fields: Am. Assoc. Pe-
troleum Geologists, p. 564-599.
Cavaroc, V. V., 1969, Delineation of some outer del-
taic plain subenvironments based on sedimentary
properties in vertical section (abs.): Am, Assoc. Pe-
troleum Geologists Bull., v. 53, no. 3, p. 710-711.
Challinor, J., 1946, Two contrasted types of alluvial
deposits: Geol. Mag., v. 83, p. 162-164.
Cbamberlin, T. K., 1964, Mass transport of sediment
in the heads of Scripps submarine canyon, Califor-
nia, in R. L. Miller, ed., Papers in marine geology:
New York, The MacmUlan Co., p. 42-64.
Cbawner, W. D., 1935, Alluvial fan flooding: the
Montrose, California flood of 1934: Geog. Rev., v.
25. p. 255-263.
Chein, C., 1961, The braided river of the lower Yellow
River: Sci. Sinica, v. 10, no. 6, p. 734-754.
Clark, R. H., and J. T. Rouse, 1971, A.closed system
for generation and entrapment of hydrocarbons in
Cenozoic deltas, Louisiana Gulf Coast: Am. Assoc.
Petroleum Geologists Bull., v. 55, p. 1170-1178.
Coastal Research Group, 1969, Field trip guidebook,
coastal environments of northeastern Massachusetts
and New Hampshire: Massachusetts Univ. Geol.
Dept. Contr. 1-CRG, 462 p.
Coleman, J. M., 1966a, Ecological changes in the mas-
sive fresh-water clay sequence: Gulf Coast Assoc.
Geol. Socs. Trans., v. 16, p. 159-174.
1966b, Recent coastal sedimentation: central
Louisiana coast: Louisiana Slate Univ. Coastal Stud-
ies Inst. Tech. Rept. 29, p. 1-73.
1969, Brahmaputra River: channel processes
and sedimentation: Sed. Geology, v. 3, no. 2-3, p.
129-239.
and S. M. Gagliano, 1964, Cyclic sedimentation
in the Mississippi River deltaic plain: Louisiana
State Univ. Coastal Studies Inst. Tech. Rept. 16, pt.
G; also in Gulf Coast Assoc. Geol. Socs. Trans., v.
14.
and 1965, Sedimentary structures: Mis-
sissippi River deltaic plain, in Primary sedimentary
structures and their hydrodynamic interpretation:
Soc. Econ. Paleontologists and Mineralogists Spec.
Pub. 12. p. 133-148.
el al., 1964, Minor sedimentary structures in a
prograding distributary: Marine Geology, v. 1, p.
240-258.
el a!., 1970, Sedimentation in a Malaysian high
tide tropical delta, in Deltaic sedimentation—modem
and ancient: Soc. Econ. Paleontologists and Miner-
alogists Spec. Pub. 15, p. 185-197.
Collinson. J. D., 1968, Deltaic sedimentation units 'us
the Upper Carboniferous of northern England: Sedi-
mentology, v. 10, no. 4, p. 233-254.
1968, The sedimentology of the Grindslow
shales and the Kinderscant grit: a deltaic complex in
the Namurian of northern England: Jour. Sed. Pe-
trology, v. 39, no. 1, p. 194-221.
Conolly, J. R., 1965, Petrology and origin of the Her-
vey Group, Upper Devonian, central New South
Wales: Geology Soc. Australia Jour., v. 12, pt. 1, p.
123-166.
1968, Submarine canyons of the continental
margin, East Bais Strait (Australia): Marine Geol-
ogy, v. 6, p. 449-461.
and W. J. Cleary, 1971, Braided deep sea deltas
and the origin of turbidite sands (abs.): Am. Gco-
phys. Union Trans., v. 52, p. 244.
Cooper, W. S., 1958, Coastal sand dunes of Oregon
and Washington: Geol. Soc. America Mem. 72, p.
1-169.
1967, Coastal dunes of California: Geol. Soc.
America Mem. 104, 131 p.
Corbeillc, R, L., 1962, New Orleans barrier island.
New Orleans area: Gulf Coast Assoc. Geol. Socs.
Trans., v. 12, p. 223-229.
Credner, G. R., 1878, The deltas: Petermanns Geog.
Mitt., v. 56, 74 p. (in German).
Crcssey, C. B., 1928, The Indiana sand dunes and
shorelines of the Lake Michigan basin: Geol. Soc.
Chicago Bull. 8, p. 37.
Crickmay, C.. H., 1955, Delta formation: Am. Assoc.
Petroleum Geologists Bull., v. 39, p. 107-114.
1960, Lateral activity in a river of northwestern
Canada: Jour. Geology, v. 68, p. 377-391.
Croft, A., 1962, Some sedimentation phenomena along
the Wasatch Mountain front: Jour. Geophys. Re-
search, v. 67, p. 1511-1524.
Crosby, E, J., 1972, Classification of sedimentary envi-
ronments, in J. K. Rigby and W. K. Hamblin, eds.,
Recognition of sedimentary environments: Soc.
Econ. Paleontologists and Mineralogists Spec. Pub.
16, 340 p.
Crowell, J. C., 1952, Submarine canyons bordering
central and southern California: Jour. Geology, v.
60, p. 58-83.
1954, Geology of the Ridge Basin area: Cali-
fornia Div. Mines Bull. 170, Map Sheet No. 7.
Culbertson, J. K^ and D. R. Dawdy, 1964, A study of
fluvial characteristics and hydraulic variables, middle
Rio Grande, New Mexico: U.S. Geol. Survey Water-
Supply Paper 1498-F, 74 p.
Curray, J. R., 1960, Sediments and history of Holocene
transgression, continental shelf, northwest Gulf of
Mexico, in F. P. Shepard et al., eds., Recent sedi-
ments, northwest Gulf of Mexico: Am. Assoc. Petro-
leum Geologists, p. 221-266.
1964, Transgressions and regressions, in Papers
in marine geology: New York, The Macmillan Co.,
p. 175-203.
1969a, Estuaries, lagoons, tidal flats and deltas,
in New concepts of continental margin sedimenta-
tion: Am. Geol. Inst. Short Course Notes, Nov. 7-9,
p. II1-1 to 111-30.
1969b, Shore rone sand bodies: barriers, cbeni-
ers and beach ridges, in New concepts of continental
margin sedimentation: Am. Geol. list. Short Course
Notes, Nov. 7-9. p. II-l to 11-18.
and D. G. Moore, 1963L Fades delineation by
acoustic-reflection: northern Gulf of Mexico: Sed[-
mcDtology, v. 2, p. 130-148.
and 1964, Holocene regressive littoral
sand, Costa de Nayaril, Mexico, in Deltaic and shal-
low marine deposits: New York, Elsevier Publishing
Co., p. 176-182.
and 1971, Growth of the Bengal deep-
sea fan and denudation in the Himalayas: Geol. Soc.
America Bull, v. 82, p. 563-572.
el al., 1969, Holocenc. history of a strand plain,
lagoonal coast. Nay am, Mexico, in Coastal lagoons,
a symposium: Internal. Symposium on Coastal La-
goons Mem., Mexico, D. F., p. 63-100.
el al., 1970, Late Quaternary sea-level studies
in Micronesia: Carmarsc expedition: Geol. Soc.
America Bull., v. 81, p. 1865-1880.
Curtn, D. M., 1970, Miocene deltaic sedimentation,
-------
Geometry of Sandstone Reservoir Bodies
179
Louisiana Gulf Coast, in Deltaic sedimentation—
modern and ancient: Soc. Econ. Paleontologists and
Mineralogists Spec. Pub. 15, p. 293-308.
Daboll, J. M., 1969, Holocene sediments of the Parker
estuary, Massachusetts: Massachusetts Univ. Dept.
Geology Contr. No. 3-CRG, 138 p.
Damuth, J. E., and R. W. Fairbridgc, 1970. Equatorial
Atlantic deep-sea arkosic sands and ice-age aridity in
tropical South America: Geol. Soc. America Bull., v.
81, p. 189-206.
Davits, D. K.., 1966, Sedimentary structures and subfa-
cies of a Mississippi River point bar: Jour. Geology,
v. 74, no. 2, p. 234-239.
el al., 1971, Recognition of barrier environ-
ments: Am. Assoc. Petroleum Geologists Bull., v.
55, p. 550-565.
Davis, J. R., 1971, Sedimentation of Pliocene sand-
stones in Santa Barbara Channel, California (abs.):
Am. Assoc. 'Petroleum Geologists Bull., v. 55, p. 335.
Davis, W. M., 1938, Sbeetfloods and streamfioods:
Geol. Soc. America Bull., v. 49, p. 1337-1416.
Dennison, J. M., 1971, Petroleum related to Middle
and Upper Devonian deltaic facies in central Appa-
lachia: Am. Assoc. Petroleum Geologists Bull., v. 55,
p. 1179-1193.
Denny, C. S., 1965, Alluvial fans in the Death Valley
region, California,-Nevada: U.S. Geol. Survey Prof.
Paper 466, 62 p.
1967, Fans and pediments: Am. Jour. Sci., v.
265, p. 81-105.
Dickas, A. B., and J. L. Payne, 1967, Upper Paleocene
buried channel in Sacramento Valley, California:
Am. Assoc. Petroleum Geologists Bull, v. 51, no. 6,
p. 873-882.
Dickinson, K. A., el al., 1972, Criteria for recognizing
ancient barrier coastlines, in J. K. Rigby and W. K.
Hamblin, eds., Recognition of sedimentary environ-
ments: Soc. Econ. Paleontologists and Mineralogists
Spec. Pub. 16, 340 p.
Dietz, R. J., and H. J. Knebel, 1971, Thou Sans Fond
submarine canyon, Ivory Coast., Africa: Deep-Sea
Research, v. 18, p. 441-447.
Dill, R. F., el al., 1954. Deep-sea channels and deltas
of the Monterey submarine canyon: Geol. Soc.
America Bull., v. 65, p. 191-194.
Dillon, W. P., 1970, Submergence effects on a Rhode
Island barrier and lagoon and inferences on migra-
tion of barriers: Jour. Geology, v. 78, no. 1, p. 94-106.
Dineley, D. L., and B. P. J. Williams, 1968, Sedimenta-
tion and paleoecology of the Devonian Escuminac
Formation and related strata, Escuminac Bay, Que-
bec: Geol. Soc. America Spec. Paper 106, p. 241-
264.
Dobby. E. H. G., 1936. The Ebro delta: Geog. Jour.,
v. 87, p. 455-469.
Dodge, C. F., 1965, Genesis of an Upper Cretaceous
offshore bar near Arlington, Texas: Jour. Sed. Pe-
trology, v. 35, no. 1, p. 22-35.
Doeglas, D. J., 1962, The structure of sedimentary de-
posits of braided rivers: Sedimentology, v. 1, p.
167-190.
Donaldson, A. C., 1966, Deltaic sands and sandstones,
in Symposium on recently developed geologic princi-
ples and sedimentation of the Permo-Pennsylvanian
of the Rocky Mountains: Wyoming Geol. Assoc.
20th Ann. Field Conf. Guidebook, p. 31-62.
1969, Ancient deltaic sedimentation (Pennsyl-
vanian) and its control on the distribution, thickness
and quality of coals, in A. C. Donaldson, ed., Some
Appalachian coals and carbonates: models of an-
cient shallow-water deposition: West Virginia Geol.
mod Econ. Survey, p. 93-123.
et al, 1970, Holocene Guadalupe delta of
Texas Gulf Coast, in Deltaic sedimentation—modern
and ancient: Soc. Econ. Paleontologists and Mineral-
ogists Spec. Pub. 15, p. 107-137.
Dondanville, R. R., 1963, The Fall River Formation,
northwestern Black Hills, lithology and geologic his-
tory: Wyoming Geol. Assoc.-Btilings Geol. Soc.
Joint Field Conf. Guidebook, p. 87-99.
Don, R. H., Jr., 1964, Ancient deltaic sedimentation in
eugeosynclinal belts, in Deltaic and shallow marine
deposits: Amsterdam, Elsevier Pub. Co., p. 105-113.
1966, Eocene deltaic sedimentation at Coos
Bay, Oregon: Jour. Geology, v. 74, p. 373—420.
Duncan, J. R., and L. D. Kulm, 1970, Mineralogy,
provenance and dispersal history of late Quaternary
deep-sea sands in Cascadia basin and Blanco frac-
ture zone of Oregon: Jour. Sed. Petrology, v. 40, p.
874-887.
Eckis, R., 1928, Alluvial fans of the Cucamonga dis-
trict, southern California: Jour. Geology, v. 36, p.
224-247.
Emery, K. O., 1960a, Basin plains and aprons off
southern California: Jour. Geology, v. 68, p. 464-
479.
1960b, The sea off southern California: New
York, John Wiley and Sons, Inc.
ei al., 1952, Submarine geology off San Diego,
California: Jour. Geology, v. 60, no. 6, p. 511-548.
Ericson, D. B., M. Ewing, and B. C. Heezen, 1951,
Deep-sea sands and submarine canyons: Geol. Soc.
America Bull, v. 62, p. 961-965.
and 1952, Turbidity currents
and sediments in the North Atlantic: Am. Assoc.
Petroleum Geologists Bull., v. 36, p. 489-511.
et al., 1955, Sediment deposition in deep Atlan-
tic, p. 205-220, in Arie Poldervaan, ed., Crust of
the earth: Geol. Soc. America Spec. Paper 62, 762 p.
el al., 1961, Atlantic deep-sea cores: Geol. Soc.
America Bull., v. 72, p. 193-285.
Evans, G., 1965, Interlidal fiat sediments and their en-
vironments of deposition in the Wash: Geol. Soc.
London Quart. Jour., v. 121, p. 209-245.
Ewing, M., ei al., 1958, Sediments and topography of
the Gulf of Mexico, in L. C. Weeks, ed., Habitat of
oil: Am. Assoc. Petroleum Geologists, p. 995-1053.
Exum, F. A., and J. C. Harms, 1968, Comparison of
marine-bar with valley-fill slraligraphic traps, west-
em Nebraska: Am. Assoc. Petroleum Geologists
Bull., v. 52, no. 10, p. 1851-1868.
Eymann, J. L., 1953, A study of sand dunes in the
Colorado and Mojave Deserts: Los Angeles, Univ.
Southern California, MS thesis, 91 p.
Fahnestock, R. K., 1963, Morphology and hydrology
of a glacial stream—While River, Mount Rainier,
Washington: U.S. Geol. Survey Prof. Paper 422-A,
70 p.
Felix, D. W., and D. S. Gorsline, 1971, Newport sub-
marine canyon, California: an example of the effects
of shifting loci of sand supply upon canyon position:
Marine Geology, v. 10, p. 177-198.
Ferguson, J., 1863, Delta of the Ganges: Geol. Soc.
London Quart. Jour., v. 19, p. 321-354.
Perm, J. C., 1970, Allegheny deltaic deposits, in Del-
taic sedimentation—modem and ancient: Soc. Econ.
Paleontologists and Mineralogists Spec. Pub. 15, p.
246-255.
and V. V. Cavaroc, Jr., 1969, A field guide to
Allegheny deltaic deposits in the upper Ohio. Valley:
-------
180
Rufus J. LeBlanc
Ohio Geol. Soc. and Pittsburgh GeoL Soc.
Field. M. E., and O. H. Pilkey. 1971, Deposition of
deep-sea sands: comparison of two areas of the Car-
olina coQlinenlal rise: Jour. Sed. Petrology, v. 41, p.
526-536.
Finch, W. L, 1959, Geology of uranium deposits in
Triassic rocks of the Colorado Plateau region: U.S.
Geol. Survey Bull. 1074-D. 164 p.
Finkel, H. J., 1959, The barchans of southern Peru:
Jour. Geology, v. 67, p. 614-647.
Fischer, P. J., 1971, An ancient (upper PaJeocene)
submarine canyon and fan: the Meganos Channel,
Sacramento Valley, California (abs.): 67th Ann.
Geol. Soc. America Cordilleran Section Mtg. Prog.,
v. 3, no. 2, p. 120.'
Fisher, W. L., 1969, Gulf Coast Basin Tertiary delta
systems, in Delta systems in exploration for oil and
gas: Texas Univ. Bur. Econ. Geology, p. 30-39.
and L. F. Brown, Jr., 1969, Delta systems and
oil and gas occurrence, in Delta systems in explora-
tion for oil and gas: Texas Univ. Bur. Econ. Geol-
ogy, p. 54-66.
and J. H. McGowen, 1969, DeposiUonaJ sys-
tems in Wilcox Group (Eocene) of Texas and their
relationship to occurrence of oil and gas: Am. As-
soc. Petroleum Geologists Bull., v. 53, no. 1, p. 30-
54.
et al, 1969, Delta systems in the exploration
for oil and gas: Texas Univ. Bur. Econ. Geology,
78 p.
et al., 1970, Depositional systems in the Jack-
son Group of Texas—their relationship to o2, gas
and uranium: Texas Univ. Bur. Econ. Geology Geol.
Circ. 70-4.
Fisk, H. N., 1940, Geology of Avoyelles and Rapides
Parishes, Louisiana: Louisiana Geol. Survey GeoL
Bull. No. 18.
1944, Geological investigation of the alluvial
valley of the lower Mississippi River: Mississippi
River Commission, Vicksburg, Mississippi.
1947, Fine-grained alluvial deposits and their
effects on Mississippi River activity: Mississippi
River Commission, Vicksburg, Mississippi
1952, Geological investigation of the Atchafa-
laya basin and problems of Mississippi River diver-
sion: Mississippi River Commission, Vicksburg, Mis-
sissippi, p. 1-145.
1955, Sand facies of Recent Mississippi delta
deposits: 4th World Petroleum Cong. Proc., Sec. I/
C, p. 1-21.
1958, Receni Mississippi River sedimentation
and peat accumulation, in Ernest Van Aelst, ed., 4th
Congres pour 1'avancemeni des etudes de stratigra-
phie et de geologic du Carboniferc, Heerlen, 1958:
Comple Rendu. v. 1, p. 187-199.
1959, Padre Island and the Laguna Mad re fiats,
coastal south Texas, in 2d Coastal Geog. Conf.,
April 6-9: Nail. Acad. Sci.-NaU. Research Council
Pub., p. 103-151.
1961, Bar-finger sands of the Mississippi delta,
in J. A. PeiersoD and J. C. Osmond, eds.. Geometry
of sandstone bodies: Am. Assoc. Petroleum Geolo-
gists, p. 29-52.
and E. McFarlan, Jr., 1955, Laic Quaternary
dellaic deposits of the Mississippi River, in A. Pol-
dervaart, ed., The crust of the earth: Geol. Soc.
America Spec. Paper 62, p. 279-302,
el al., 1954, Sedimentary framework of the
modern Mississippi delia: Jour. Sed. Petrology v
24, no. 2, p. 76-99.
FlemaL. R. C., 1967, Sedimentology of the Sespe For-
mation, southwestern California: Princeton, New
Jersey, Princeton Univ., PhD thesis, 258 p.
Folk, R. L., 1971, Longitudinal dunes of the northwest-
em edge of the Simpson Desert, Northern Territory,
• Australia, Part 1. Geomorphology and grain size re-
lationships: Sedimentology, v. 16, p. 5-54.
and W. C. Ward, 1957, Brazos River bar—a
study of the significance of grain size parameters:
Jour. Sed. Petrology, v. 27, p. 3-26.
Franer, D. E., 1967, Recent deltaic deposits of the
Mississippi River: their development and chronol-
ogy: Gulf Coast Assoc. Geol. Socs. Trans., v. 17, p.
287-315.
— ; and A. Osanik, 1961, Point-bar deposits, Old
River locksile, Louisiana: Gulf Coast Assoc. Geol.
Socs. Trans., v. 11, p. 121-137.
and 1969, Receni peat deposits, Louisi-
ana coastal plain: GeoL Soc. America Spec. Paper
114, p. 63-85.
Freeman, J. C., 1949, Strand-line accumulation of pe-
troleum, Jim Hogg County, Texas: Am. Assoc. Pe-
troleum Geologists Bull., v. 33, no. 7, p. 1260-1270.
Friedkin, J. F., 1945, A laboratory study of the mean-
dering of alluvial rivers: U.S. Waterways Experiment
Sia., Vicksburg, Mississippi, May.
Friedman, G. M., and K. G. Johnson, 1966, The Devo-
nian Calskill deltaic complex of New York, type ex-
ample of a "teutonic delta complex," in Deltas in
their geologic framework: Houston Geol. Soc.
Galehouse, J. J., 1967, Provenance and paleocurrents
of the Paso Robles Formation, California: Geol.
Soc. America Bull., v. 78, p. 951-978.
Galloway, W. E., 196S, Depositional systems of lower
Wilcox Group, north-central Gulf Coast basin: Gulf
Coast Assoc. Geol. Socs. Trans., v. 18, p. 275-289.
Geijskes, D. C., 1952, On the structure and origin of
the sandy ridges in the coastal zone of Surinam:
Kouinkl Nederlandsch Aardrijksk. Genoot.
Tijdschr., v. 69, no. 2, p. 225-237.
Gellert, J. F., 1968, The Yangtze River mouth and
delta: Warsaw, Przeglad Geog., v. 40, p. 413-418
(in German).
Giddings, J. L., 1947, Mackenzie River delta chronol-
ogy: Tree Ring Bull., v. 13, p. 26-29.
Gilbert, G. K., 1914, The transportation of debris by
running water: U.S. Geol. Survey Prof. Paper 86.
Giles, R. T., and O. H. Pilkey, 1965, Atlantic beach
and dune sediments of the southern United Stales:
Jour. Sed. Petrology, v. 35, no. 4, p. 900-910.
Gilluly, J., 1927, Physiography of the Ajo region, Ari-
zona: Geol. Soc. America BulL, v. 48, p. 323-348.
Glenn, J. L., and A. R. Dahl, 1959, Characteristics
and distribution of some Missouri River deposits:
Iowa Acad. Sci. Proc., v. 66, p. 302-311.
Glennic, K. W., 1970, Desert sedimentary environments
—Developments in Sedimentology, no. 14: Amster-
dam, Elsevier Pub. Co., 222 p.
Gorsline, D. S., 1967, Sedimentologic studies of the
Colorado delta: Southern California Univ. Geol.
RepL 67-1, 121 p.
and K_ 0. Emery, 1959, Turbidity-current de-
posits in San Pedro and Santa Monica basins oft
southern California: Geol. Soc. America Bull., v. 70,
p. 279-290.
et al., 1967, Studies of submarine canyons and
fans off southern California: Southern Calif. Univ.
GeoL RcpL 67-3, 27 p.
Gould, H. R., 1970, The Mississippi delia complex, in
Deltaic sedimentation—modern and ancient: Soc.
-------
Geometry of Sandstone Reservoir Bodies
T81
Econ. Paleontologists and Mineralogists Spec. Pub.
15. p. 3-30.
and E. McFarlsn, Jr., 1959, Geologic history of
the chenier plain, southwestern Louisiana: Gull
Coast Assoc. Geol. Socs. Trans., v. 9, p. 261-270.
Gourou, P., 1967, The Zambezi River delta: Ada
Geog. .Lovanieosia, v.-5, p. 31-36 (in French).
Greenman. N. N., and R. J. LeBlanc, 1956, Recent ma-
rine sediments and environments of northwest Gulf
of Mexico: Am. Assoc. Petroleum Geologists Bull.,
v. 40, p. 813-847.
Greensrailh, J. T., 1966, Carboniferous deltaic sedi-
mentation in eastern Scotland, a review and reap-
praisal, in Deltas in their geologic framework: Hous-
ton Geol. Soc.
and E. V. Tucker, 1969, The origin of the Ho-
locene shell deposits in the chenier plain facies of
Essex (Great Britain): Marine Geology, v. 7, p.
403-425.
Gregory, J. L., 1966, A lower Oligocene delta in the
subsurface of southeast Texas: Gulf Coast Assoc.
Geol. Socs. Trans., v. 16, p. 227-241.
Gremillion, L. R., el a!., 1964, Barrier and lagoonal
sets on high terraces in the Florida panhandle:
Southeastern Geology, v. 6, p. 31-36.
Griffith, £. G., 1966, Geology of Saber bar, Logan and
Weld Counties,. Colorado: Am. Assoc. Petroleum
Geologists Bull., v. 50, p. 2112-2118.
Griggs, G. B., and L. D. Kulm, 1970, Sedimentation in
Cascadia deep-sea channel: Geol. Soc. America
Bull., v. 81, p. 1361-1384.
Guilcber, A., 1958, Coastal and submarine morphol-
ogy: London, Methuen & Co., Ltd.: New York.,
John Wiley and Sons, Inc., 274 p.
1959, Coastal sand ridges and marshes and
their continental environment near Grand Popo and
Ouidah, Dahomey: 2d Coastal Geog. Conf., U.S.
Office Naval Research-Natl. Acad. Sci, Washing-
ton, D.C., p. 189-212.
Guy, H. P., el a}., 1966, Summary of alluvial channel
data from flume experiments, 1956-1961: U.S. Geol.
Survey Prof. Paper 462-1, p. 1-96.
Gwinn, V. E., 1964, Deduction of flow regime from
bedding character in conglomerates and sandstones:
Jour. Sed. Petrology, v. 34, no. 3, p. 656-658.
and T. A. Mulch, 1965, Inlerlongued Upper
Cretaceous volcanic and nonvolcanic rocks, central-
western Montana: Geol. Soc. America Bull., v. 76,
p. 1125-1144.
Hack, J. T., 1941, Dunes of the western Navajo coun-
try: Geog. Rev., v. 31, p. 240-263.
Hails, J. R., 1964, The coastal deposilional features of
southeastern Queensland: Australia Geog., v. 9, p.
208-217.
and J. H. Hoyt, 1969, The significance and lim-
itations of statistical parameters for distinguishing
ancient and modern sedimentary environments of the
lower Georgia coastal plain: Jour. Sed. Petrology, v.
39, no. 2, p. 559-580.
Hale, L. A., 1961, Late Cretaceous (Monlanan) stra-
tigraphy, eastern Wasbakic basin, Carbon County,
Wyoming: Wyoming Geol. Assoc. 16th Field Conf.
Guidebook, p. 129-137.
Hamilton, E. L., 1967, Marine geology of the abyssal
plains in the Gulf of Alaska: Jour. Geophys. Re-
search, v. 72, p. 4189-4213.
Haner, B. E., 1971, Morphology and sediments of Re-
doodo submarine fan, southern California: Geol.
Soc. America BulJ., v. 82, p. 2413-2432.
Hansen, H. J., 1969, Depositions! environments of sub-
surf ice Potomac Group in southern Maryland: Am.
Assoc. Petroleum Geologists Bull., v. 53, DO. 9, p.
1923-1937.
Hantzschel, W., 1939, Tidal fiat deposits (Wat ten-
schlick), in P. D. Trask, ed., Recent marine sedi-
ments: Am. Assoc. Petroleum Geologists, p. 195-
206.
Happ, S. C, et al., 1940, Some aspects of accelerated
stream and valley sedimentation: U.S. Depl. Agricul-
ture Tech. Bull., v. 695, 134 p.
Harms, J. C., 1966, Stratigraphic traps in a valley fill,
western Nebraska: Am. Assoc. Petroleum Geologists
Bull., v. 50, p. 2119-2149.
and R. K. Fahnestock, 1965, Stratification bed
forms and flow phenomena (with an example from
the Rio Grande), in Primary sedimentary structures
and their bydrodynamic interpretation: Soc. Econ.
Paleontologists and Mineralogists Spec. Pub. 12, p.
84-115.
et al., 1963, Stratification in modern sands of
the Red River, Louisiana: Jour. Geology, v. 71, p.
566-580.
el al., 1965, Deposilional environment of the
Fox Hills sandstones near Rock Springs uplift, Wy-
oming: Wyoming Geol. Assoc. 19th Field Conf.
Guidebook, p. 113-130.
Hattin, D. E., 1965, Stratigraphy of the Graneros shale
(Upper Cretaceous) in central Kansas: Kansas
Geol. Survey Bull. 178.
Heezen, B. C., and Marie Tbarp, 1967, Indian Ocean
floor: Natl. Geog. Mag., October.
and 1969, Atlantic Ocean floor: Nail.
Geog. Mag., June.
et a!., 1959, The floors of the oceans: Geol.
Soc. America Spec. Paper 65, 122 p.
et al., 1964, Congo submarine canyon: Am. As-
soc. Petroleum Geologists Bull., v. 48, p. 1126-1149.
Hewitt, C. H., el al., 1965, The Frye in situ combuslion
test—reservoir characteristics: Soc. Petroleum Engi-
neers AIME Petroleum Trans., March, p. 337-342.
Holeroan, J. N., 1970, The sediment yield of major riv-
ers of the world: Water Resources Research, v. 4,
no. 4, p. 737-747.
Hookc, R. L., 1967, Processes of arid-regional alluvial
fans: Jour. Geology, v. 75, no. 4, p. 438-460.
Hoover, R. A., 1968, Physiography and surface sedi-
mcnt facies of a recent tidal delta, Harbor Island,
central Texas coast: Austin, Texas, Texas Univ.,
PhD thesis, 223 p.
Hoppe, G., and S. R. Ekman, 1964, A note on the
alluvia] fans of Ladljouagec, Swedish Lappland:
Geog. Annaler, v. 46, p. 338-342.
Horn, D., 1965, Zur Geologischen entwickjung der
sudlichen schleimundung in Holozan: Meyniana, v.
15, p. 41-58.
Horn, D. R., el al., 1971, Turbidites of the northeast
Pacific: Sedimcnlology, v. 16, p. 55-69.
Horowitz, D. H., 1966, Evidence for dcliaic origin of
an Upper Ordovician sequence in the central Appa-
lachians, in Deltas in their geologic framework:
Houston Geol. Soc., p. 159-169.
Houbolt, J. J. H. C., and J. B. M. Jonker, 1968, Recent
sediments in the eastern part of the Lake of Geneva
(Lac Leman): Geologic en Mjjnbouw, v. 47, p. 131-
148.
Howard, J. D., 1966, Patterns of sediment dispersal in
the Fountain Formation of Colorado: Mtn. Geolo-
gist, v. 3, p. 147-153.
Howe, H. V., and C. K_ Moresi, 1931, Geology of
-------
182
Rufus J. LeBlanc
Iberia Parish: Louisiana Ceol. Survey Geol. Bull.
No. I.
and 1933, Geology of Lafayette and St.
Martin Parishes: Louisiana Geol. Survey Geol. Bull.
No. 3, 238 p.
el al., 1935, Reports on the geology of Cam-
eron and Vermilion Parishes: Louisiana Geol. Sur-
vey Geol. Bull. No. 6, 742 p.
Hoyt, J. H., 1964, High angle beach stratification, Sa-
pelo Island, Georgia: Jour. Sed. Petrology, v. 32, p.
309-311.
1967, Barrier island formation: Geol. Soc.
America Bull., v. 78, p. 1125-1136.
1969, Chenier versus barrier, genetic and strati-
graphic distinction: Am. Assoc. Petroleum Geolo-
gists Bull., v. 53, p. 299-306.
and V. J. Henry, Jr., 1965, Significance of inlet
sedimentation in the recognition of ancient barrier
islands, in Sedimentation of Laic Cretaceous and
Tertiary outcrops, Rock Springs uplift: Wyoming
Geol. Assoc. 19th Field Coni Guidebook, p. 190-
194.
and 1967, Influence of island migration
on barrier island sedimentation: Geol. Soc. America
Bull., v. 78, p. 77-86.
and 1971, Origin of capes and shores
along the southeastern coast of the United States:
Geol. Soc. America Bull., v. 82, p. 59-66.
and R. J. Weimer, 1963, Comparison of mod-
ern and ancient beaches, central Georgia coast: Am.
Assoc. Petroleum Geologists Bull., v. 47, p. 529-531.
el al., 1964, Late Pleistocene and Recent sedi-
mentation on Georgia coast, in Deltaic and shallow
marine deposits: New York, Elsevier Publishing Co.,
p. 170-176.
Hoyt, YV. V., 1959, Erosional channel in the middle
Wilcox near Yoakum, Lavaca County, Texas: Gulf
'Coast Assoc. Geol. Socs. Trans., v. 9, p. 41-50.
Hrabar, S. V., and P. E Potter, 1969, Lower west Ba-
deu (Mississippiac) sandstone body of Owen and
Green Counties, Indiana: Am. Assoc. Petroleum Ge-
ologists Bull., v. 53, no. 10, p. 2150-2160.
Huang, T., and H. G. Goodell, 1970, Sediments and
sedimentary processes of eastern Mississippi cone,
Gulf of Mexico: Am. Assoc. Petroleum Geologists
Bull., v. 54, p. 2070-2100.
Hubbard, J. A. E. B., 1967, Facics patterns in the Car-
rowmoran Sandstone (Visean) of western County
Sligo, Ireland: GeoL Assoc. Proc., v. 77, pt. 2, p.
233-254.
Hubert, C., el al., 1970, Deep-sea sediments in the
lower Paleozoic Quebec supergroup: Geol. Assoc.
Canada Spec. Paper 7, p. 103-125.
Hubert, J. F., 1960, Petrology of the Fountain and
Lyons Formations, Front Range, Colorado: Colo-
rado School Mines Quart., v. 55, p. 1-242.
1964, Textural evidence for deposition of many
western Atlantic deep-sea sands by ocean-bottom
currents rather than turbidity currents: Jour. Geol-
ogy, v. 72, p. 757-785.
Inman, D. L., and R. A. Bagnold, 1963, Littoral pro-
cesses, in The sea, v. 2, chap. 2: New York, John
Wiley and Sons, p. 529-553.
ei al., 1966, Coastal sand dunes of Guerrero
Negro, Baja California, Mexico: Geol. Soc. America
Bull., v. 77, no. 8, p. 787-802.
Ives, R. L., 1936, Desert floods in Sonoyta Valley: Am.
Jour. Sci., v. 32, p. 349-360.
Jacka, A. D., 1965, Deposiu'onaJ dynamics of the Al-
mond Formation, Rock Springs uplift, Wyoming:
Wyoming Geol. Assoc. 19lh Field Conf. Guidebook,
p. 81-100.
Jahns, R. H., 1947, Geological features of the Connect-
icut Valley, Massachusetts as related to recent
floods: U.S. Geol. Survey Water-Supply Paper 996,
157 p.
Jelgersma, S., J. de Jong, and W. H. Zagwijn, 1970,
The coastal dunes of the western Netherlands; geol-
ogy, vegetational history and archeology: Geol.
Stichting Med., new ser., no. 21, p. 93-154.
Johnson, A., 1965, A model for debris flow: University
Park, Pennsylvania, Pennsylvania State Univ., PhD
thesis, 233 p.
Johnson. R. B., 1967, The great sand dunes of southern
Colorado: U.S. Geol. Survey Prof. Paper 575-C, p.
C177-C183.
Johnson, W. A., 1921, Sedimentation of the Fraser
River delta: Canada Geol. Survey Mem. 125, 46 p.
1922, The character of the stratification of the
sediments of the Recent delta of the Fraser River,
British Columbia, Canada: Jour. Geology, v. 30, p.
115-129.
Kanes, W., 1969, Tidal inlets, tidal deltas, and barrier
islands versus alluvial deposits: a discussion of sedi-
mentary criteria, in A. C. Donaldson, ed., Some Ap-
palachian coals and carbonates: models of ancient
shallow water deposition: West Virginia Geol. and
Econ. Survey, p. 259-263.
1970, Facies and development of the Colorado
River delta in Texas, in Deltaic sedimentation—mod-
ern and ancient: Soc. Econ. Paleontologists and
Mineralogists Spec. Pub. 15, p. 78-106.
Kclling, G., 1968, Patterns of sedimentation in
Rhondda Beds of South Wales: Am. Assoc. Petro-
leum Geologists Bull., v. 52, no. 12, p. 2369-2386.
and D. J. Stanley, 1970, Morphology and struc-
ture of Wilmington and Baltimore submarine can-
yons, eastern United States: Jour. Geology, v. 78, p.
637-660.
and M. A. Woollands, 1969, The stratigraphy
and sedimentation of the Llandoverian rocks of the
Rboyader district, in A. Woods, ed., The Precam-
brian and lower Paleozoic rocks of Wales: Univ.
Wales Press, p. 255-282.
Kessler, L. G., II, 1970, Bar development and aggrada-
tional sequences in the braided portions of the Cana-
dian River, Hutchinson, Roberts and Hemphill
Counties, Texas (abs.): 23d Ann. Rocky Mountain
Sec Geol. Soc. America Mtg. Prog., v. 2, no. 5, p.
338.
1971, Characteristics of the braided stream de-
posiuonal environment with examples from the
South Canadian River, Texas: Earth Sci. Bull., v. 4,
no. 1. p. 25-35.
King, C. A. M., 1959, Beaches and coasts: London,
Edward Arnold Publications, Ltd., 403 p.
Klein, G. deV., 1962, Triassic sedimentation. Maritime
Provinces, Canada: Geol. Soc. America Bull., v. 73,
p. 1127-1146.
1967, Comparison of Recent and ancient tidal
fiat and estuarine sediments, in Estuaries: Am. As-
soc. Adv. Sci. Pub. 83, p. 207-218.
1968, Sedimentology of Triassic rocks in the
lower Connecticut Valley, in Guidebook for field
trips in Connecticut, New England Internal. Colle-
giate Geol. Conf.: Connecticut Geol. and Nat. His-
tory Survey Guidebook 2, p. (c-1) 1-19.
1970, Depositjonal and dispersal dynamics of
intcnJdal sand bars: Jour. Sed. Petrology, v. 40, p.
1095-1127.
-------
Geometry of Sandstone Reservoir Bodies
183
1971, A sedimentary model for determining pa-
leotidal range: Geol. Soc. America Bull, v. 82, p.
2585-2592.
and J. E, Sanders, 1964, Comparison of sedi-
• and E. W. Eden, 1940, Sand waves in the lower
tneots from Bay of Fundy and Dutch Wadden Sea
tidal flats: Jour. Sed. Petrology, v. 34, no. 1, p. 18-
24.
Knight, W. V., 1969, Historical and economic geology
of Silurian Clinton Sandstone of northwestern Ohio:
Am. Assoc. Petroleum Geologists Bull, v. 53, no. 7,
p. 1421-1452.
Kolb, C. R., 1962, Distribution of soils bordering the
Mississippi River from Donaldsonville to Head of
Passes: U.S. Army Corps Engineers, Waterways
•Expt. Sta. Tech. RepL 3-601, p. 1-61.
1963, Sediments forming the bed and banks of
the lower Mississippi River and their efiect on river
migration: Sedimentology, v. 2, p. 227-234.
and R. 1. Kaufman, 1967, Prodelta days of
southeast Louisiana: Internal. Marine Geolech. Re-
search Conf. Proc., MonticeUo, Illinois, p. 3-21.
and J. R, Van Lopik, 1966, Depositional envi-
ronments of the Mississippi River deltaic plain,
southeastern Louisiana, In Deltas in their geological
framework: Houston Geol. Soc., p. 17-61.
el al., 1968, Geological investigation of the Ya-
zoo basin, lower Mississippi Valley: U.S. Army
Corps Engineers, Waterways Expt. Sia. Tech. Repl.
3-480.
Komar, P. D., 1969, The channelized flow of turbidity
currents with application to Monterey deep-sea
channel: Jour. Geopbys. Research, v. 74, p. 4544-
4558.
1970, The competence of turbidity current
flow: Geol. Soc. America Bull., v. 81. p. 1555-1562.
Krause, D. C., tl al., 1970, Turbidity currents and ca-
ble breaks in the western New Britain trench: Geol.
Soc. America Bull., v. 81, p. 2153-2160.
Krigstrom, A., 1962, Geomorphological studies of San-
dur plains and their braided river in Iceland: Geog.
Annals, v. 44, p. 328-346.
Kruit, C., 1955, Sediments of the Rh&nc delta; I, Grajn
size and microfauna: Koninkl. Nederlandse Geol.
Mijnb. Genoot. Verb., Geol. Scr., v. 15, p. 357-514.
Kuenen, Ph. H., 1950, Marine geology: New York,
John Wiley and Sons, Inc., 568 p.
Kukal, Z., 1964, Diagnostic features of Paleozoic del-
taic sediments of central Bohemia: Sedimenlology, v.
3, p. 109-113.
1971, Geology of Recenl sediments: New
York, Academic Press, 490 p.
Lagaaij, R., and F. P. H. W. Kopslein, 1964, Typical
features on a fiuvio-marine offlap sequence, in Devel-
opments in Sedimenlology, v. 1, Deliaic and shallow
marine deposits: Amsterdam, Elsevier Publishing
Co., p. 216-226.
Laming, D. J. C., 1966, Imbrication, paleocurrents, and
other sedimentary features in the lower New Red
Sandstone, Devonshire, England: Jour. Sed. Petrol-
ogy, v. 36, no. 4, p. 940-959.
Land, L. S.. 1964, Eolian cross beddine in the beach
dune environment, Sapelo Island, Georgia: Jour.
Sed. Petrology, v. 34, no. 2, p. 389-394.
Lane. E. W., 1957, A study of the shape of channels
formed by natural streams flowing in erodibic mate-
rial: U.S. Army Corps Engineers. Missouri River
Div., Omaha, Nebraska, Sediment Ser. 9, 106 p.
1963. Cross-stratification in San Bernard River,
Texas, point bar deposits: Jour. Sed. Penology, v.
33, p. 350-354.
Mississippi River: Western Soc. Engineers Jour., v.
45, no. 6, p. 281-291.
Lapone, L. F., 1968, Ancient environments: Engle-
wood Cliffs. New Jersey, Prentice-Hall, Inc., 116 p.
Laury, R. L., 1968. Sedimentology of the Pleasanrview
Sandstone, southern Iowa and western Illinois: Jour.
Sed. Petrology, v. 38, no. 2, p. 568-599.
Lawson, A. C., 1913, The petrographic description of
alluvial fan deposits: California Univ. Pubs. Geol.
Sci., v. 7, p. 325-334.
LeBlanc, R. J., and H. A. Bernard, 1954, Resume of
late Recent geological history of the Gulf Coast:
Geologic en Mijnbouw, new series, v. 16c, p. 185-194.
and W. D. Hodgson, 1959, Origin and develop-
ment of the Texas shoreline: Louisiana Slate Univ.
Coastal Studies Icsl., 2d Coastal Geog. Conf., p.
57-101; EPR Pub. 219, Shell EiP Research, Hous-
ton.
LeBlanc, R. J., Jr., 1971, Environments of deposition
of the Yegua Formation (Eocene), Brazos County,
Texas: Corpus Chrisli Geol. Soc. Bull., v. 11, no. 6,
70 p.
Leggel, R. F., R. J. E. Brown, and G. H. Johnston,
1966, Alluvial fan formation near Aklavik, North-
west Territories, Canada: GeoL Soc. America Bull.,
v. 77. p. 15-30.
Lent, G. W., and D. E. Owen, 1969, Grain orientation
in a Bcrca Sandstone channel at south Amhersl,
Ohio: Jour. Sed. Petrology, v. 39, no. 2. p. 737-743.
Leopold, L. B., and M. G. Wolman, 1957, River chan-
nel patterns: braided, meandering and straight: U.S.
Geol. Survey Prof. Paper 282-B, 85 p.
and 1960, River meanders: Geol. Soc.
America Bull., v. 71, p. 769-794.
el al., 1966, Channel and hillslope processes in
a semi-arid area, New Mexico: U.S. GeoL Survey
Prof. Paper 352-G.
Lineback, J. A., 1968, Turbidites and other sandstone
bodies in the Borden Sillslone (Mississippian) in Il-
linois: Illinois Geol. Survey Circ. 425, p. 1-29.
Lins, T. W., 1950, Origin and environment of the Ton-
ganoxie Sandstone in northeastern Kansas: Kansas
Geol. Survey Bull., v. 86, p. 107-140.
Liteanu, E., and A. Pricajan, 1963, The geological
structure of the Danube delta: Bucharest, Hidrobiol-
ogia, v. 4, p. 57—82 (in Rumanian).
Logvinenko, N. V., and I. N. Remizov, 1964, Sedimen-
lology of beaches on the north coast of the Sea of
Azov, in Deltaic and shallow marine deposits, devel-
opments in Sedimentology, v. 1: Amsterdam, Elsevier
Publishing Co., p. 244-252.
Lucke, J. B., 1934, A study of Barnegat Inlet, New
Jersey and related shoreline phenomena: Shore and
Beach, v. 2. p. 45-93.
Lum, D., and H. T. Stearns, 1970, Pleistocene stratig-
raphy and euslalic history based on cores at Waim-
anaio, Oahu, Hawaii: Geol. Soc. America Bull., v.
81, p. 1-16.
Lumsden, D. N., and B. R. Pelletier, 1969, Petrology
of the Grimsby Sandstone (Lower Silurian) of On-
lario and New York: Jour. Sed. Petrology, v. 39, no.
2, p. 521-530.
Luslig, L. K., 1963, Competence of transport in allu-
vial fans: U.S. Geol. Survey Prof. Paper 475-C, p.
126-129.
1965. Clastic sedimentation in the Deep Springs
Valley, California: U.S. Geol. Survey Prof. P«pei
352-F.
Maclnlyrc, I. G., and J. D. Milliman, 1970, Physio-
-------
184
Rufuj J. LeBlanc
graphic features on the outer shelf and upper slope,
Atlantic continental margin, southeastern United
Slates: Geol. Soc. America Bull., v. 81. p. 2577-
2589.
MacNeil, F. S., 1950, Pleistocene shore lines in Florida
and Georgia: U.S. Geol. Survey Prof. Paper 221-F.
Mammerickx, J., 1970, Morphology of the Aleutian
abyssal plain: Geol. Soc. America Bull., v. 81, p.
3457-3464.
Mann, C. J., and W. A. Thomas, 1968, Ancient Missis-
sippi River: Gulf Coast Assoc. Geol. Socs. Trans., v.
18. p. 187-204.
Manos, C., 1967, Depositional environment of Spar-
land cycle them (Pennsylvanian), Illinois and Forest
City basins: Am. Assoc. Petroleum Geologists Bull.,
v. 51, no. 9, p. 1843-1861.
Martin, B. D., 1963, Rosedale channel—evidence for
late Miocene submarine erosion in Great Valley of
California: Am. Assoc. Petroleum Geologists Bull.,
v. 47, p. 441-456.
and K. O. Emery, 1967, Geology of the Monte-
rey Canyon, California: Am. Assoc. Petroleum Geol-
ogists Bull., v. 51, p. 2281-2304.
Martini, I. P., 1971, Regional analysts of sedimentology
of Medina Formation (Silurian), Ontario and New
York: Am. Assoc. Petroleum Geologists Bull., v. 35,
p. 1249-1261.
Malhews, W. H., and F. P. Shepard, 1962, Sedimenta-
tion of Fraser River delta, British Columbia: Am.
Assoc. Petroleum Geologists Bull., v. 48, no. E, p.
1416-1444.
Matthews, J. L., 1966, Sedimentation of the coastal
dunes at Oceano, California: Univ. California at Los
Angeles, PhD thesis, 138 p.
McBride, E. F., and M. 0. Hayes, 1962, Dune cross
bedding on Mustang Island, Texas: Am. Assoc. Pe-
troleum Geologists Bull., v. 46, p. 546-552.
McCave, I. N., 1969, Correlation using a sedimcntolog-
ical model of pan of the Hamilton group (Middle
Devonian), New York State: Am. Jour. Sci., v. 267,
no. 5, p. 567-591.
McCloy, J. L., 1969, Morphologic characteristics of the
Blow River delta, Yukon Territory, Canada: Baton
Rouge, Louisiana, Louisiana State Univ., PhD thesis,
161 p.
McEwen, M. C., 1969, Sedimentary facics of the mod-
em Trinity delta, in Galveston Bay geology: Hous-
ton Geol. Soc., p. 53-77.
McGee, W. J., 1897, Sheetflood erosion: Geol. Soc.
America Bull., v. 8, p. 87-112.
McGowen, J. H., and L. E Garner, 1969, Comparison
of Recent and ancient coarse-grained point bars
(abs.): Am. Assoc. Petroleum Geologists Bull., v.
53, no. 3, p. 731-732.
and 1970, Physiographic features and
stratification types of coarse-grained point bars:
modern and ancient examples: Sedimentology, v. 14,
p. 77-111.
and C. G. Groat, 1971, Van Horn Sandstone,
West Texas—an alluvial fan model for mineral ex-
ploration: Texas Univ. Bur. Econ. Geology, 57 p.
McKee, E. D., 1934, The Coconino Sandslone—its his-
tory and origin: Carnegie Inst. Washingion Pub.
440, Conlr. Paleontology, p. 77-115.
1939, Some types of bedding in the Colorado
River delta: Jour. Geology, v. 47, p. 64-81.
1957, Primary structures of some Recenl sedi-
ments: Am. Assoc. Petroleum Geologists Bull., v. 41,
p. 1704-1747.
1966, Structures of dunes at White Sands Na-
tional Monument, New Mexico: Sedimenlology, v. 7,
nj. 1, 69 p.
and G. C. Tibbitts, 1964, Primary structures of
a seif dune and associated deposits of Libya: Jour.
Sed. Petrology, v. 34, no. 1, p. 5-17.
McKenzie, P., 1958, The development of sand beach
ridges: Australian Jour. Sci., v. 20, p. 213-214.
Melton, M. A., 1965, The geomorphic and paleocli-
malic significance of alluvial deposits in southern
Arizona: Jour. Geology, v. 73, p. 1-38.
Menard, H. W., 1955, Deep-sea channels, topography,
and sedimentation: Am. Assoc. Petroleum Geologists
Bull, v. 39. p. 236-255.
1960, Possible pre-Pleistocene deep-sea fans off
central California: Geol. Soc, America Bull., v. 71,
p. 1271-1278.
1964, Marine geology of the Pacific: New
York, McGraw-Hill Book Co., Inc., 271 p.
el al., 1965, The Rhone deep-sea fan, in Subma-
rine geology and geophysics: London, Butterwonh,
p. 271-286.
Miall, A. D., 1970, Continental marine transition in the
Devonian of Prince of Wales Island, Northwest Ter-
ritories: Canadian Jour. Earth Sci., v. 7, p. 125-144.
Miller, D. N., 1962, Patterns of barrier bar sedimenta-
tion and its similarity to Lower Cretaceous Fall
River stratigraphy, in Symposium on Early Creta-
ceous rocks: Wyoming Geol. Assoc. 17th Field
Conf. Guidebook, p. 232-247.
Milling, M. E., and E. W. Bebrens. 1966. Sedimentary
structures of beach and dune deposits: Mustang Is-
land, Texas: Inst. Marine Sci. Pub., v. 11, p. 135-
148.
Moody-Stuart, M., 1966, High and low sinuosity
stream deposits, with examples from the Devonian
of Spitsbergen: Jour. Sed. Petrology, v. 36, no. 4, p.
1102-1117.
Moore, D., 1966, Deltaic sedimentation: Earth Sci.
Rev., v. 1, p. 87-104, Amsterdam, Elsevier Publish-
ing Co.
Moore, D. G., 1970, Reflection profiling studies of the
California borderland: structure and Quaternary tur-
biditc basins: Geol. Soc. America Spec. Paper 107.
Moore, G. T., and D. 0. Asquith, 1971, Delta: term
and concept: Geol. Soc, America Bull., v. 82, p.
2563-2568.
Morgan, J. P., 1961, Mudlumps at the mouths of the
Mississippi River, in Genesis and paleontology of the
Mississippi River mudlumps: Louisiana Geol. Survey
Geol. Bull. No. 35.
1970, Depositional processes and products in
the deltaic environment, in Deltaic sedimentation—
modern and ancient: Soc.- Econ. Paleontologists and
Mineralogists Spec. Pub. 15, p. 31-47.
ei al., 1953, Occurrence and development of
mud fiats along the western Louisiana coast: Louisi-
ana State Univ. Coastal Studies InsL Tecb. RepL
No. 2.
tt al., 1968, Mudlumps: diapiric structures in
Mississippi delta sediments, in J. Braunslein and G.
D. O'Brien, eds., Diapirisrn and diapirs: Am. Assoc.
Pelroleum Geologists Mem. 8, p. 145-161.
Muehlbcrger, F. B., 1955, Pismo beach-Point Sal dune
field, California: Lawrence, Kansas, Univ. Kansas,
Master's thesis, 106 p.
Muller, G., 1966, The new Rhine delta in Lake Con-
siance, in M. L. Shirley, ed., Deltas in their geologic
framework: Housion Geol. Soc., p. 107-124.
Mulch, T. A., 1968, Pennsylvanian non-marine sedi-
ments of the Narragansetl Basin, Massachusetts and
-------
Geometry of Sandstone Reservoir Bodies
185
Rhode Island: Geol. Soc. America Spec. Paper 106,
p. 177-209.
Nagtegaal, P. J. C., 1966, Scour-and-fill structures from
a fluvial piedmont environment: Geologic en Mjjn-
bouw, v. 45. no. 10, p. 342-354.
Nanz, R. H., 1954, Genesis of Oligocene sandstone res-
ervoir, Seeligson field, Jim Wells and Kleberg Coun-
ties, Texas: Am. Assoc. Petroleum Geologists Bull.,
v. 38, p. 96-117.
Nelson, B. W., 1970, Hydrography, sediment dispersal,
and recent historical development of the Po River
delta, Italy, in Deltaic sedimentation—modem and
ancient: Soc. Econ. Paleontologists and Mineralo-
gists Spec. Pub. 15, p. 152-184.
Nelson, C. H., 1968, Marine geology of Astoria deep-
sea fan: CorvaJlis, Oregon, Oregon 'State Univ., PhD
thesis, 287 p.
el at., 1970, Development of the Astoria can-
yon-fan physiography and comparison with similar
systems: Marine Geology, v. 8, p. 259-291.
Nichols, M. M., 1964, Characteristics of sedimentary
environments in Moriches Bay, in Papers in marine
geology: New York, The Macmillan Co., p. 363-
383.
Nflscn, T. H., 1969, Old Red sedimentation in the Bue-
landet-Vaerlandet Devonian district, western Nor-
way: Sed. Geology, v. 3, no. 1.
Nonnarlc, W. R., 1970, Growth patterns of deep sea
fans: Am. Assoc. Petroleum Geologists Bull., v. 54,
p. 2170-2195.
and D. J. W. Piper, 1969, Deep-sea fan-valleys,
past and present: Geol. Soc. America Bull., v. 80, p.
1859-1866.
Nonis, R. M., 1966, Barcban dunes of Imperial Valley,
California: Jour. Geology, v. 74, p. 292-306.
Nossin, J. J., 1965, The geomorphic history of the
northern Padang delta: Jour. Tropical Geog., v. 20,
p. 54-64.
Nunnally, N. R., 1967, Definition and identification of
channel and overbank deposits and their respective
roles in flood plain formation: Prof. Geographer, v.
19, no. 1, p. 1-4.
Odem, W. 1., 1953, Subaerial growth of the delta of
the diverted Brazos River, Texas: Compass, v. 30, p.
172-178.
Oomkens, E., 1967, Depositions! sequences and sand
distribution in a deltaic complex: Geologic en
Mijnbouw, v. 46E, p. 265-278.
1970, Deposiu'onal sequences and sand distribu-
tion in the post-glacial Rhone delta complex, in Del-
laic sedimentation—modem and ancient: Soc. Econ.
Paleontologists and Mineralogists Spec. Pub. 15, p.
198-212.
Ore, H. T., 1963, Some criteria for recognition of
braided stream deposits: Contr. Geology, v. 3, DO. 1,
p. 1-14.
1965, Characteristic deposits of rapidly aggrad-
ing streams, in Sedimentation of Laic Cretaceous
and Tertiary outcrops, Rock Springs uplift: Wyo-
ming Geological Assoc. 19th Field Coaf. Guidebook,
p. 195-201.
Oslcrboudt, W. J., 1946, The seismograph discover)' of
an ancient Mississippi River channel (abs.): Geo-
physics, v. 11, p. 417.
Olvos, E. G., Jr., 1970, Development and migration of
barrier islands, northern Gulf of Mexico: Geol. Soc.
America Bull., v. 81, p. 241-246.
Pack, F. J., 1923, Torrential potential of desert waters:
Pan-Am. Geologist, v. 40, p. 349-356.
Paine, W. R., 1966, Stratigraphy and sedimentation of
Hackberry shale and associated beds of southwestern
Louisiana: Gulf Coast Assoc. Geol. Socs. Trans., v.
16.
Paull, R. A, 1962, Deposiu'onal history of the Muddy
Sandstone, Bighorn basin, Wyoming: Wyoming
Geol. Assoc. 17th Field Conf. Guidebook, p. 102-
117.
Peterson, J. A, and J. C. Osmond, eds., 1961, Geome-
try of sandstone bodies: Am. Assoc. Petroleum Ge-
ologists, 240 p.
Petrescu, G., 1948, Le delta maritime du Danube; son
evolution physico-geographiquc ct les problems qui
s'y posent: Rumania, Univ. lasi Ann. Sci., 2d sec.
(Sc. Nat.), v. 31, p. 254-303.
Pettyjohn, W. A., 1967, New members of Upper Creta-
ceous Fox Hills Formation in South Dakota, repre-
senting delta deposits: Am. Assoc. Petroleum Geolo-
gists Bull., v. 51, no. 7, p. 1361-1367.
Perzerta, J. R., 1968, The St. Clair River delta: Univ.
Michigan, PhD thesis, 168 p.
Phleger, F. B, 1965, Sedimemology of Guerrero Negro
Lagoon, Baja California, Mexico, in 17th Submarine
geology and geophysics—Colston Research Society
Symposium Proc.; Bristol, England, Paper 171, p.
205-237: London, Butterworths.
1969, Some general features of coastal lagoons,
in Coastal lagoons, a symposium: Internet. Sympo-
sium on Coastal Lagoons Mem., Mexico, D F., p. 2-
25.
and G. C. Ewing, 1962, Sedimemology and
oceanography of coastal lagoons in Baja California,
Mexico: Geol. Soc. America Bull., v. 73, p. 145-
182.
Pierce, J. W., 1970, Tidal inlets and washover fans:
Jour. Geology, v. 78, no. 2, p. 230-234.
and D. J. Colquhoun, 1970, Holocene evolution
of a portion of the North Carolina coast: Geol. Soc.
America Bull., v. 81, no. 12, p. 3697-3714.
Piper, D. J. W., 1970, Transport and deposition of Ho-
locene sediment of La Jolla deep sea fan, California:
Marine Geology, v. 8, p. 211-227.
and W. R. Norroark, 1971, Re-examination of
a Miocene deep-sea fan and fan valley, southern
California: Geol. Soc. America Bull., v. 82, p. 1823-
1830.
Potter, P. E., 1967, Sand bodies and sedimentary envi-
ronments, a review: Am. Assoc. Petroleum Geolo-
gists Bull., v. 51, p. 337-365.
Price, W. A., 1947, Equilibrium of form and forces in
tidal basins of coast of Texas and Louisiana: Am.
Assoc. Petroleum Geologists Bull., v. 31, no. 9, p.
1619-1663.
1954, Environment and formation of the cbe-
nier plain: Texas A&M Research Found. Proj. 63,
Dept. Oceanography, Texas A&M College.
1963, Patterns of flow and channeling in tidal
inlets: Jour. Sed. Petrology, v. 33, no. 2, p. 279-290.
Priddy, R. R., n al., 1955, Sediments of Mississippi
Sound and inshore waters: Mississippi Geol. Survey
Bull. 82.
Pryor, W. A., 1960, Cretaceous sedimentation in Upper
Mississippi embayment: Am. Assoc. Petroleum Geol-
ogists Bull., v. M. p. 1473-1504.
——— 1961, Sand trends and paleoslope in Illinois ba-
sin and Mississippi embayment, in J. A. Peterson
and J. C. Osmond, Geometry of sandstone bodies:
Am. Assoc. Petroleum Geologists, p. 119-133.
1967, Biogcnic directional features on several
point bars: Sed. Geology, v. 1, p. 235-245.
Psuty, N. P., 1965, The geomorphology of beach ridges
-------
186
Rufus J. LeBlanc
in Tabasco, Mexico: Louisiana State Univ. Coastal
Studies Inst. Tech. Kept. 30, p. 1-51.
Pugh, J. C., 1953, The Porto Novo-Badagri sand ridge
complex: Nigeria, Univ. College of 1 bad an Dept.
Geography, Research Notes, DO. 3, p. 1-14.
Rahn, P. H., 1967, Sheetflood, streamfloods, and the
formation of pediments: Assoc. Am. Geographers
Annals, v. 57, p. 593-604.
Rainwater, E. H., 1964, Transgressions and regressions
in the Gulf Coast Tertiary: Gulf Coast Assoc. Geol.
Socs. Trans., v. 14, p. 217-230.
Rapp, A., 1959, Avalanche boulder tongues in Lapp-
land: Geog. Annaler, v. 41, p. 34-48.
Reineck, H. E., 1963, Sedimenlgefuge im bereicb der
sudlichen Nordsee: Abhandl. Senckenberg: Naturf
Gesell. 505, p. 1-136.
1967, Layered sediments in tidal flats, beaches,
and shelf bottoms of the North Sea, in Estuaries:
Am. Assoc. Adv. Sci. Pub. 83, p. 191-206.
and I. B. Singh, 1967, Primary sedimentary
structures in the Recent sediments of the Jade,
North Sea: Marine Geology, v. 5, p. 227-235.
Reineck, H. R., 1972, Tidal fiats, in J. K. Rigby and
W. K. Hamblin, eds., Recognition of sedimentary •en-
vironments: Soc. Econ. Paleontologists and Mineral-
ogists Spec. Pub. 16, 340 p.
Revelle, R., and F. P. Shepard, 1939, Sediments off the
California coast, in F. P. Shepard el al., eds., Recent
marine sediments: Am. Assoc. Petroleum Geologists,
p. 245-282.
Rex. R. W., 1964, Arctic beaches. Barrow, Alaska, in
Papers in marine geology: New York, The Macmil-
lao Co., p. 384-400.
Rickmers, W. R., 1913, The Duab of Turkestan: New
York, Cambridge Univ. Press, 563 p.
Rigby, J. K., and W. K. Hamblin, eds., 1972, Recogni-
tion of sedimentary environments: Soc. Econ. Pale-
ontologists and Mineralogists Spec. Pub. 16, 340 p.
Rona, P. A., 1970, Submarine canyon origin on upper
continental slope off Cape Hatteras: Jour. Geology,
v. 78, p. 141-152.
Royse, C. F., Jr., 1970, A sedimentologic analysis of
the Tongue River-Sentinel Butle interval (Paleo-
cene) of the Williston basin, western North Dakota:
Sed. Geology, v. 4, p. 19-80.
Ruhc, R. U., 1964, Landscape morphology and alluvial
deposits in southern New Mexico: Assoc. Am, Ge-
ographers Annals, v. 54, p. 147-159.
Rusnak, G. A., 1957, A fabric and petrographic study
of the Pleasantview Sandstone: Jour. Sed. Petrology,
v. 27, no. 1, p. 41-55.
1960, Sediments of Laguna Madrc, Texas, in
Recent sediments, northwest Gulf of Mexico: Am.
Assoc. Petroleum Geologists, p. 153-196.
Russell. R. J., 1936, Physiography of lower Mississippi
River delta, in Lower Mississippi River delta: Louisi-
ana Geol. Survey Geol. Bull. 8, p. 1-199.
1942, Geomorphology of the Rh6ne delta:
Assoc. Am. "Geographers Annals, v. 32, p. 149-254.
1967a, RTvcr and delta morphology: Louisiana
Stale Univ. Coastal Studies Inst. Tech. Repl. 52, 49
P-
1967b, Aspects of coastal morphology: Geog.
Annaler, v. 49, ser. A, p. 299-309.
and H. V. Howe, 1935, Cheniers of southwest-
ern Louisiana: Geog. Rev., v. 25, no. 3, p. 449-461.
and R. D. Russell, 1939, Mississippi River delta
sedimentation, in P. D. Trask, ed., Recent marine
sediments: Am. Assoc. Petroleum Geologists D
153-177. '
Ryan, D. J., 1965, Cross-bedding formed by lateral ac-
cretion in the Catskill Formation near Jim Thorpe,
Pennsylvania: Pennsylvania Acad. ScL Proc., v. 38,
DO. 2, p. 154-156.
Ryder, J. M., 1971, The stratigraphy and morphology
of para-glacial alluvial fans in south-central British
Columbia: Canadian Jour. Earth Sci., v. 8, p. 279-
298.
Sabate, R. W., 1968, Pleistocene oil and gas in coastal
Louisiana: Gulf Coast Assoc. Geol. Socs. Trans., v.
18, p. 373-386.
Sabins, F. F., Jr., 1963, Anatomy of stratigraphic trap,
Bisti field, New Mexico: Am. Assoc. Petroleum Ge-
ologists Bull., v. 47, p. 193-228.
Saucier, R. T., 1964, Geological investigation of the St.
Francis basin: U.S. Anny Corps Engineers, Water-
ways Expt. Sta. Tech. Rept. 3-659.
1967, Geological investigation of the Boeuf-
Tensas basin, lower Mississippi Valley: U.S. Army
Corps Engineers, Waterways Expt. Sta. Tech. Rept.
3-757.
Schlee, J. S., and R. H. Moencb, 1961, Properties and
genesis of "Jackpile" Sandstone, Laguna, New Mex-
ico, in J. A. Peterson and J. C. Osmond, eds., Geom-
etry of sandstone bodies: Am. Assoc. Petroleum Ge-
ologists, p. 134-150.
Scboll, D. W., el al., 1970, The structure and origin of
tbe large submarine canyons of the Bering Sea: Ma-
rine Geology, v. 8, p. 187-210.
Schumm, S. A., 1963, Sinuosity of alluvial rivers on the
great plains: Geol. Soc. America Bull., v. 74, p.
1089-1100.
1971, Fluvial paleochanncls, in J. K. Rigby and
W. K. Hamblin, eds., Recognition of sedimentary en-
vironments: Soc. ECOD. Paleontologists and Mineral-
ogists Spec. Pub. 16, 340 p.
Schwartz, M. L., 1971, The multiple causality of bar-
rier islands: Jour. Geology, v. 79, no. 1, p. 91-94.
Scott, A. J., and W. L. Fisher, 1969, Delta systems and
deltaic deposition, in Delta systems in exploration
for oil and gas: Texas Univ. Bur. Econ. Geology, p.
10-39.
Scruton, P. C., 1960, Delta building and the deltaic
sequence, in F. P. Shepard el al.. eds., Recent sedi-
ments, northwest Gulf of Mexico: Am. Assoc. Petro-
leum Geologists, p. 82-102.
1961, Rocky Mountain Cretaceous stratigraphy
and regressive sandstones: Wyoming Geological As-
soc. 16th Field Conf. Guidebook, p. 241-249.
Selley, R. C., 1970, Ancient sedimentary environments:
Ithaca, New York, Cornell Univ. Press, 237 p.
Shannon. J. P., Jr., and A. R. Dahl, 1971, Deltaic
stratigraphic traps in West Tuscola field, Taylor
County, Texas: Am. Assoc. Petroleum Geologists
Bull., v. 55, p. 1194-1205.
Sharp, R. P., 1942, Mudflow levees: Jour. Georaor-
pbology, v. 5, p. 222-227.
and L. H. Nobles, 1953, Mudfiow in 1941 at
Wrightwood, southern California: Geol. Soc. Amer-
ica Bull., v. 64, p. 547-560.
Sbawa, M. S., 1969, Sedimentary history of tbe Gil-
wood Sandstone (Devonian), Utikuma Lake area,
Alberta, Canada: Bull. Canadian Petroleum Geol-
ogy, v. 17, p. 392-409.
Sbelton, J. W., 1965, Trend and genesis of lowermost
sandstone unit of Eagle Sandstone at Billings, Mon-
tana: Am, Assoc. Petroleum Geologists Bull., v. 49,
p. 1385-1397.
1967, Siratigraphic models and general criteria
for recognition of alluvial, barrier-bar, and turbidity-
-------
Geometry of Sandstone Reservoir Bodies
187
current uod deposits: Am. Assoc. Petroleum Geolo-
gists Bull., v. 51, no. 12, p. 2441-2461; also in Bill-
ings Geol. Soc. 17th Field Conf. Guidebook, p. 1-
17.
1972, Models of sand and sandstone .deposits:
Oklahoma Geol. Survey Bull, (in press).
Shcpard, F. P., 1960, Gulf Coast barriers, in F. P. She-
pard el ai, eds.. Recent sediments, northwest Gulf of
Mexico: Am. Assoc. Petroleum Geologists, p. 197-
220.
F. B Phleger,-and Tj. H. van Andel, 1960,
Recent sediments, northwest Gulf of Mexico: Am.
Assoc. Petroleum Geologists, 394 p.
1965, Importance of submarine valleys in fun-
neling sediments to the deep sea, in Progress in
.oceanography, v. 3: New York, The Macmillan Co.,
p. 321-332.
and E. C. Buffington, 1968, La Jolla submarine
fan-valley: Marine Geology, v. 6, p. 107-143.
and R. F. Dill, 1966, Submarine canyons and
other sea valleys: Chicago, Rand McNally, 381 p.
and D. G. Moore, 1955, Sediment zones bor-
dering the barrier islands of central Texas coast,
Finding ancient shorelines: Soc. ECOD. Paleontolo-
gists and Mineralogists Spec. Pub. 3, p. 78-96.
and 1960, Bays of central Texas coast,
in F. P. Sbepajd ti al., eds., Recent sediments,
northwest Gulf of Mexico: Am. Assoc. Petroleum
Geologists, p. 117-152.
el ai., eds., I960, Recent sediments, northwest
GuU of Mexico: Am. Assoc. Petroleum Geologists,
394 p.
el ai, 1963, Submarine geology: New York,
Harper and Row, 557 p.
el at., 1969, Physiography and sedimentary pro-
cess of La Jolla submarine fan and fan-valley, Cali-
fornia: Am. Assoc. Petroleum Geologists Bull., v.
53, p. 390-420.
Shepheard, W. W., and L. V. Hills, 1970, Depositional
environments, Bearpaw-Horsesboe Canyon (Upper
Cretaceous) transition zone, Drumbeller "Badlands,"
Alberta: Bull. Canadian Petroleum Geology, v. 18,
p. 166-215.
Shideler, G. L., 1969, Dispersal patterns of Pennsylva-
nian sandstones in the Michigan basin: Jour. Sed.
Petrology, v. 39, no. 3, p. 1229-1237.
Shirley, M. L., ed., 1966, Deltas: Houston Geol. Soc.,
251 p.
Shuiyak, B. A., and V. L. Boldyrev, 1966, The pro-
cesses of beach ridge formation: Oceanology (Engl.
ed.), v. 6, no. 1, p. 88-94.
Siebold, E., 1963, Geological investigation of ncarshore
sand transport—examples of methods and problems
from Baltic and North Seas, in Progress in oceanog-
raphy, v. 1: New York, The MacmiJJan Co., p. 3-
70.
Simons, D. B., and E. V. Richardson, 1961, Forms of
bed roughness in alluvial channels: Am. Soc. CiviJ
Engineers Proc., Jour. Hydraulics Div., v. 87 no.
HY 3, May, p. 87-105.
1962, The effect of bed roughness on depth-dis-
charge relations in alluvial channels: U.S. Geol. Sur-
vey Water-Supply Paper 149S-E.
1966, Resistance to flow in alluvial channels:
U.S. Geol. Survey Prof. Paper 422-J, 69 p.
el at., 1965, Sedimentary structures generated
by flow in alluvia) channels, in Primary sedimentary
structures and their bydrodynamic interpretation:
Soc. Econ. Paleontologists and Mineralogists Spec.
Pub. 12, p. 34-52.
Sims, J. D., 1967, Geology and sedimentology of the
Livingston Group, northern Crazy Mountains, Mon-
tana: Evanston, Illinois, Northwestern Univ., PhD
thesis, 109 p.
Sioli, H., 1966, General features of the delta of the
Amazon, in Scientific problems of the humid tropic
zone deltas and their implications: UNESCO, Dacca
Symposium Proc., p. 381-390.
Sloane, B. J., Jr., 1958, The subsurface Jurassic Bod-
caw sand in Louisiana: Louisiana Geol. Survey
Geol. Bull. 33.
Smith, A. E., Jr., 1971, Deltas of the world—modem
and ancient, bibliography: Houston Geol. Soc. Delta
Study Group, 42 p.
Smith, A. J., 1966, Modern deltas: comparison maps,
in Deltas in their geologic framework: Houston
Geol. Soc., p. 233-251.
Smith, H. T. U, 1954, Coastal dunes: Coastal Geog.
Conf., February, U.S. Office Naval Research, p. 51-
56.
Smith, N. D., 1970, The braided stream depositiona]
environment: comparison of the Plane River with
some Silurian clastic rocks, north central Appalach-
ians: Geol. Soc. America Bull., v. 81, p. 2993-3014.
Snavely, P. D., Jr., ei ol., 1964, Rhythmic-bedded eu-
geosynclinal deposits of the Tyee Formation, Oregon
Coast Range, in Symposium on cyclic sedimentation:
Kansas Geol. Survey Bull. 169. p. 461-480.
Spearing, D. R., 1969, Stratigraphy, sedimentation, tud
tectonic history of the Paleocene-Eocene Hoback
Formation of western Wyoming: Ann Arbor, Michi-
gan, Univ. Michigan, PhD thesis, 179 p.
Stanley, D. J., 1967, Comparing patterns of sedimenta-
tion in some modem and ancient submarine can-
yons: Earth and Planetary Sci. Letters, v. 3, p. 371-
380.
1968, Graded bedding-sole marking-graywacke
assemblage and related sedimentary structures in
some Carboniferous flood deposits, eastern Massa-
chusetts: Geol. Soc. America Spec. Paper 106, p.
211-239.
1969, Submarine channel deposits and their
fossil analogs ("fluxolurbidiles"), in New concepts
of continental margin sedimentation: Am. Geol.
Inst. Short Course Notes, Nov. 7-9, p. DJS-9-1-
DJS-9-17.
1970, Flyschoid sedimentation on the outer At-
lantic margin off northeast North America, in Flysch
sedimentology in North America: Geol. Assoc. Can-
ada Spec. Paper No. 7, p. 179-210.
1972, Submarine channel deposits, fluxoturbi-
diles and other indicators of slope and base-of-slope
environments in modern and ancient marine basins,
in J. K.. Rigby and W. K. Hamblin, eds., Recognition
of sedimentary environments: Soc. Econ. Paleontolo-
gists and Mineralogists Spec. Pub. 16, 340 p.
and N. SiJverberg. 1969, Recent slumping
on the continental slope off Sable Island bank,
southeast Canada: Earth and Planetary Sci. Letters,
v. 6, p. 123-133.
ei al.. 1971, Lower continental rise east of the
middle Atlantic states: predominant sediment disper-
sal perpendicular to isobaths: Geol. Soc. America
Bull., v. 82, p. 1831-1840.
Stetson, H. C., 1953. The sediments of the northwest-
ern Gulf of Mexico—Part 1. The continental terrace
of the western Gulf of Mexico: its surface sedi-
ments, origin and development, in Papers in physical
oceanography and meteorology: Massachusetts Inst.
Tech. Abs. Theses, v. J2, no. <, p. 5-«5.
-------
188
Rufus J. LeBlanc
Stokes, W. L., 1961, Fluvial and eolian sandstone bod-
ies in Colorado Plateau, in J. A. Peterson and J. C.
Osmond, eds., Geometry o( sandstone bodies: Am.
Assoc. Petroleum Geologists, p. 151-178.
1964, Eolian varving in the Colorado Plateau:
Jour. Sed. Petrology, v. 34, no. 2, p. 429-433.
1968, Multiple parallel-truncation bedding
planes—a feature of wind-deposited sandstone for-
mations: Jour. Sed. Petrology, v. 38, no. 2, p. 510-
515.
Sullwold, H. H., Jr., 1960, Tarzana fan, deep subma-
rine fan of late Miocene age, Los Angeles County,
California: Am. Assoc. Petroleum Geologists Bull,
v. 44, p. 433-457.
Sundborg, A., 1956, The river Klaralven. A study of
fluvial processes: Geog. Annaler, v. 38, p. 127-316.
Swann, D. H., 1964, Late Mississippiao rhythmic sedi-
ments of Mississippi Valley: Am. Assoc. Petroleum
Geologists Bull., v. 48, no. 5, p. 637-658.
el al., 1965, The Border Siltslone (Mississip-
pian) delta in southwestern Illinois: Illinois Geol.
Survey Circ. 386, 20 p.
Swift, D. J. P., 1969, Inner shelf sedimentation: pro-
cesses and products, in New concepts of continental
margin sedimentation: Am. Geol. InsL Short Course
Notes, Nov. 7-9. p. DS-4-1-DS-4-46.
and R. M. McMullen, 1968, Preliminary studies
of intertidal sand bodies in the Minas Basin, Bay of
Fundy, Nova Scotia: Canadian Jour. Earth Sci., v. 5,
no. 2, p. 175-183.
Sykes, G., 1937, The Colorado delta: Carnegie Inst.
Washington Pub. 460, 193 p.
Taylor, J. H., 1963, Sedimentary features of an ancient
deltaic complex: the Wealden roots of southeastern
England: Sedimenlology, v. 2, p. 2-28.
Tbachuk, N. M., 1968, Geological study of the Middle
Devonian Gil wood arkoses in the Nipisi area, Al-
berta: Petroleum Soc. of Canadian InsL Mining and
Metallurgy, 19th Ann. Tech. Mtg., Preprint No.
6828, 17 p.
Thorn, B. G., 1967, Mangrove ecology and deltaic geo-
morphology: Tabasco, Mexico: Jour. Ecology, v. 55,
p. 301-343.
Thomas, W. A., and C. J. Mann, 1966, Late Jurassic
deposiuonal environments, Louisiana and Arkansas:
Am. Assoc. Petroleum Geologists Bull., v. 50, p.
178-182.
Thompson, R. W., 1968, Tidal Sat sedimentation on
the Colorado River delta, northwestern Gulf of Cali-
fornia: Geol. Soc. America Mem. 107, 133 p.
Thompson, W. O., 1937, Original structures of beaches,
bars, and dunes: Geol. Soc. America Bull., v. 48, p.
723-752.
Todd, T. W., 1968, Dynamic diversion: influence of
longshore currenl-tidal Sow interaction on chenier
and barrier island plains: Jour. Sed. Petrology, v. 38,
p. 734-746.
and W. A. Monroe, 1968, Petrology of Doraen-
gine Formation (Eocene) at Potrero Hills and Rio
Vista, California: Jour. Sed. Petrology, v. 38, p.
1024-1039.
Trask, P. D., el al., 1955, Recent marine sediments—a
symposium: Soc. Econ. Paleontologists and Mineral-
ogists Spec. Pub. 4, 736 p.
Treadwell, R. C., 1955, Sedimentology and ecology of
southeast coastal Louisiana: Baton Rouge, Louisi-
ana, Louisiana Slate Univ., unpub. thesis, p. 1-176.
Trimonis, E. S., and K. M. Shimkus, 1970, Sedimenta-
tion at the head of a submarine canyon: Acad. Sci.
USSR, Oceanology (Engl. ed.), v. 10, p. 74-«5.
Trowbridge, A. C, 1911, The terrestrial deposits of Ow-
ens Valley, California: Jour. Geology, v. 19, p. 709-
747.
1930, Building of Mississippi delta: Am. Assoc
Petroleum Geologists Bull., v. 14, p. 867-901.
van Andel, Tj. H., 1960, Sources and dispersion of Ho-
loccne sediments, northern Gulf of Mexico, in F. P.
Shepard et al., Recent sediments, northwest Gulf of
Mexico: Am. Assoc. Petroleum Geologists, p. 34-
55.
1964, Recent marine sediments of Gulf of Cali-
fornia, in Tj. H. van Andel and G. C. Shor, Jr., eds..
Marine geology of the Gulf of California: Am, As-
soc. Petroleum Geologists Mem. 3, p. 216-310.
1967, The Orinoco delta: Jour. Sed. Petrology,
v. 37, no. 2, p. 297-310.
and J. R. Curray, 1960, Regional aspects of
modem sedimentation in northern Gulf of Mexico
and similar basins, and paleogeographic significance,
in F. P. Shepard et al., eds., Recent sediments of
northwest Gulf of Mexico: Am. Assoc. Petroleum
Geologists, p. 345-364.
and G. C. Shor, Jr., eds., 1964, Marine geology
of the Gulf of California: Ana. Assoc. Petroleum
Geologists Mem. 3, 408 p.
Van Lopik, J. R., 1955, Recent geology and geo-
morphjc history of central coastal Louisiana: Louisi-
ana State Univ. Coastal Studies Inst. Tech. RcpL 7,
p. 1-89.
Vann, J. H., 1959, The geomorphology of the Guiana
Coast: Louisiana Slate Univ. Coastal Studies Inst.,
2d Coastal Geog. Conf. Proc., p. 153-187.
van Straaten, 1_ M. J. U., 1951, Texture and genesis of
Dutch Wadden Sea sediments: 3d Internal. Cong.
Sedimentology Proc., Netherlands, p. 225-244.
1954a, Composition and structure'of Recent
marine sediments in the Netherlands: Leidscbe GeoL
Meded., pL 19, p. 1-110.
1954b, Sedimentology of Recent tidal flat de-
posits and the Psammites du Condroz (Devonian):
Geologic en Mijnbouw, v. 16, p. 25-47.
1959, Littoral and submarine morphology of
the Rhone delta: Louisiana State Univ. Coastal
Studies InsL, 2d Coastal Geog. Conf. Proc., p. 233-
264.
1961, Sedimentation in tidal flat areas: Alberta
Soc. Petroleum Geologists Jour., v. 9, p. 203-226.
1965, Coastal barrier deposits in south- and
north-Holland . . . : Netherlands, GeoL Sticht.
Meded, new series, no. 17, p. 41-75.
Venkataratbnam, K.., 1970, Formation of the barrier
spit and other sand ridges near Chilka Lake on the
east coast of India: Marine Geology, v. 9, no. 2, p.
101-116.
Visber, G. S., 1965a, Use of vertical profile in environ-
mental reconstruction: Am. Assoc Petroleum Geolo-
gists Bull., v. 49, no. 1, p. 41-61.
1965b, Fluvial processes as interpreted from
ancient and Recent fluvial deposits, in Primary sedi-
mentary structures and their hydrodynamic interpre-
tation: Soc. Econ. Paleontologists and Mineralogists
Spec. Pub. 12, p. 116-131..
1968, Depositional framework of the Blue-
jacket-Banlesville sandstone, in Guidebook to geol-
ogy of the Bluejacket-Banlesville sandstone, Okla-
homa: Oklahoma City Geol. Soc., p. 32-51.
1971, Physical characteristics of fluvial depos-
its, in J. K. Rigby and W. K. Hamblin, eds., Recog-
nition of sedimentary environments: Soc. Econ. Pa-
leontologists and Mineralogists Spec. Pub. 16, 340 p.
-------
Geometry of Sandstone Reservoir Bodies
189
el a!., 1971, Pennsylvanian delta patterns and
petroleum occurrences in eastern Oklahoma: Am.
Assoc. Petroleum Geologists Bull., v. 55, p. 1206-
1230.
Volker, A., 1966, The deltaic area of the Irrawaddy
River in Burma, in Scientific problems of the humid
tropic zone delta and their implications: UNESCO,
Dacca Symposium Proc., p. 373-379.
Waechier, N. D., 1970, Braided stream deposits of the
Red River, Texas -panhandle (abs.): Ceol. Soc.
America Abs. with Programs, v. 2, no. 7, p. 713.
Walker, J. R.. and J. V. Massingill, 1970, Slump fea-
tures on the Mississippi fan, northeastern Gulf of
Mexico: Geol. Soc. America Bull., v. 81, p. 3101-
3108.
Walker, K. R., 1962, Lithofacies maps of Lower Mis-
sissippian elastics of eastern and east-central United
Stales: Am. Assoc. Petroleum Geologists Bull., v. 46,
p. 105-111.
Walker, R. G., 1966, Deep channels in turbidite-bear-
ing formations: Am. Assoc. Petroleum Geologists
Bull., v. 50, p. 1899-1917.
Walton, W. C., 1970, Groundwater resource evalua-
tion: New York, McGraw-Hill Co., 664 p.
Wanlcss, H. R., el al, 1970, Late Paleozoic deltas in
the central and .eastern United States, in Deltaic sed-
imentation—modem and ancient: Soc. Econ. Pale-
ontologists and Mineralogists Spec. Pub. 15, p. 215-
245.
Wanne, J. E., 1966, Geology of Mugu Lagoon: Univ.
California al Los Angeles, PhD dissert.
1971, Paleoecological aspects of a modem
coastal lagoon: California Univ. Pubs. Geol. Soc., v.
87, 134 p.
Wa^, J. H., Jr., 1968, Bed thickness analysis of some
Carboniferous fiuvial sedimentary rocks near Jog-
gins, Nova Scoua: Jour. Sed. Petrology, v. 38, no. 2,
p. 424-433.
Weidic, A. E., 196JS, Bar and barrier-island sands: Gulf
Coast Assoc. Geol. Socs. Trans., v. IS, p. 405-415.
Weimer, R. J., 1961 a, Spatial dimensions of Upper
Cretaceous sandstones, Rocky Mountain area, in
J. A. Peterson and J. C. Osmond, eds., Geometry of
sandstone bodies: Am. Assoc. Petroleum Geologists.
1961 b, Upper Cretaceous delta on tectonic
foreland, northern Colorado and southern Wyoming
(abs.): Am. Assoc. Petroleum Geologists Bull., v.
45. p. 417.
1965, Late Cretaceous deltas. Rocky Mountain
region (abs.): Am. Assoc. Petroleum Geologists
Bull., v. 49, p. 363.
Welder, F. A., 1959, Processes of deltaic sedimentation
in the lower Mississippi River: Louisiana State Univ.
Coastal Studies Insl. Tech. Rept. No. 12, p. 1-90.
Wermund, E G., and W. A. Jenkins, Jr., 1970, Recog-
nition of deltas by fining trend surfaces to Upper
Pennsylvanian sandstones in north-central Texas, i»
Deltaic sedimentation—modem and ancient: Soc.
Econ. Paleontologists and Mineralogists Spec. Pub.
15, p. 156-263.
Wessei, J. M., 1969, Sedimentary history of Upper
Triassic alluvial fan complexes in north-central Mas-
sachusetts: Massachusetts Univ. Dept. Geology
Contr. No. 2, 157 p.
Wier, C. E., and W. A. Girdley, 1963. Distribution of
the Inglefield and Dicksburg Hills Sandstone mem-
bers in Posey and Vanderburgb Counties, Indiana:
Indiana Acad. Sci. Proc., v. 72, p. 212-217.
Wilde, P., 1965, Recent sediments of the Monterey
deep-sea fan: Cambridge, Massachusetts, Harvard
Univ., PhD dissert., 153 p.
Williams, G. E., 1966, Paleogeography of the Torrido-
nian Applecross Group: Nature, v. 209, no. 5030, p.
1303-1306.
1969, Characteristics and origin of a Pre-Cam-
brian pediment: Jour. Geology, v. 77, p. 183-207.
Williams, P. F., and B. R. Rust, 1969, The sedimentol-
ogy of a braided river: Jour. Sed. Petrology, v. 39,
no. 2, p. 649-679.
Wilson, M. D., 1967, The stratigraphy of the Beaver-
bead Group in the Lima area, southwestern
Montana: Evanston, Illinois, Northwestern Univ.,
PhD dissert., 183 p,
1970, Upper Cretaceous-Paleocene synorogenic
conglomerates of southwestern Montana: Am. As-
soc. Petroleum Geologists Bull., v. 54, p. 1843-1867.
Winder, C. G., 1965, Alluvial cone construction by al-
pine mudflow in a humid temperate region: Cana-
dian Jour. Earth Sci., v. 2, p. 270-277.
Winterer, E L., et al.. 1968, Geological history of the
pioneer fracture zone with the Delgada deep-sea fan
northeast.Pacific: Deep-Sea Research, v. 15, no. 5, p.
509-520.
Wolff, M. P., 1967, Deltaic sedimentation of the Mid-
dle Devonian Marcellus Formation in southeastern
New York: Ithaca, New York, Cornell Univ., PhD
dissert., 231 p.
Wolman, M. G., and L. M. Brush, 1961, Factors con-
trolling the size and shape of stream channels in
coarse noncohesive sands: U.S. Geol. Survey Prof.
Paper 282-G, p. 183-210.
and L. B. Leopold, 1957, River flood plains:
some observations on their formation: U.S. Geol.
Survey Prof. Paper 232-C, p. 87-107.
Wurster, P., 1964, Delta sedimentation in the German
Keuper Basin, in Deltaic and shallow marine deposi-
tion: Amsterdam, Elsevier Publishing Co., p. 436-
446.
Zenkovich, V. P., 1964, Formation and burial of accu-
mulative forms in littoral and nearshore marine envi-
ronments: Marine Geology, v. 1. p. 175-180.
1967, Processes of coastal development: New
York, Imerscience Publishers—Division of John Wi-
ley and Sons, Inc.
-------
190
Rufus J. LeBlonc
Discussion
EDWARD N. WILSON, Kentucky Geological Sur-
vey, Lexington, Kentucky
You remarked that some of the deltaic bod-
ies were rather thin and not very extensive. In
the central United States, the Pennsylvania Sys-
tem contains several of these deltaic sequences
and some of them are fairly thick. I should like
to ask if there is anything inherently disadvan-
tageous to these sandstone bodies for emplace-
ment of limited volumes of waste?
R. J. LEBLANC
I do not think so, for the following reasons.
Deltas of various sizes can prograde seaward
into a basin over long periods of time. Thus,
they can produce relatively thick deltaic sand
bodies over extensive areas which consist of
several individual genetic units stacked over
each other. It is true that some of the Pennsyl-
vanian sandstones are thick and occur over ex-
tensive areas. There is nothing wrong with
these sandstones from the standpoint of the em-
placement of limited amounts of waste into
them. It is. important to mention that deltaic
sands grade seaward into prodelta silts and
clays.
PAUL WITHERSPOON, University of California,
Berkeley, California
First, I want to compliment you on a very
excellent review of depositional conditions. I
wanted to ask if you have looked at conditions
such as the Mount Simon Sandstone of central
United States; 1,000-2,000 ft thick, it can be
traced all the way across Indiana, Illinois,
Ohio, and up to New York, where it is called
the "Potsdam," and to Minnesota and southern
Illinois. Would the mechanisms you have de-
scribed relate to accumulation of that thick
sand body over those hundreds of miles?
R. J. LEBLANC
I cannot answer that specific question be-
cause I am not familiar with the Mount Simon
Sandstone. However, I can comment on other
sandstones which occur over very extensive
areas. For example, the Castlegate Sandstone of
northwestern Utah extends for many miles
from west to east; but the Castlegate is not a
uniform sandstone deposited in one environ-
ment. Actually, it consists of alluvial-fan,
braided-stream, and deltaic sandstones. I be-
lieve that many other sandstones are similar to
the Castlegate in that they are extensive but of
multiple origin; therefore, the models I de-
scribed can explain their origin.
JIM HALLORAN, Montana Water Resources
Board, Helena, Montana
Can you give us some idea what this barrier-
island model will look like after marine trans-
gression or regression?
R. J. LEBLANC
One of the largest oil fields discovered in the
United States during the past several years is
the Bell Creek field of Montana. Two profes-
sors from Texas A&M University correctly in-
terpreted this reservoir as a barrier-island sand-
stone body. I refer you to Dr. R. R. Berg's1 ex- .
cellent paper on this barrier-bar sandstone, be-
cause time does not permit a detailed answer to
your question.
'Berg. R. R., and D. K. Davits, 1968, Origin ol
Lower Cretaceous Muddy Sandstone at Bell Creek
field, Montana: Am. Assoc. Petroleum Geologists
Bull., v. 52, no. 10, p. 1888-1898.
-------
Section 2
-------
HYDROGEOLOGY
PERFORMANCE OBJECTIVES
At the end of this lesson, participants will be able to:
• Explain the hydrologic cycle and its relationship to rocks and
groundwater
• Differentiate between recharge and discharge areas
• Describe the relationship between porosity and permeability
• Describe the effect of porosity and permeability on
groundwater flow
• Define aquifer, aquitard, and aquiclude
• Compare confined aquifers and unconfmed aquifers.
-------
NOTES
HYDROGEOLOGY
HYDROGEOLOGY
The study of the interrelationships of
geologic materials and processes
with water, especially groundwater
H H V~
s x I
O ^
• ^^ • o
o"'".
y/^\ *J f^
W 1
9/93
Hydrogeology
-------
NOTES
EARTH'S WATER RESOURCES
• 80% oceans
• 19% groundwater
• 1% ice
• 0.002% streams and lakes
« 0.0008% atmosphere
EARTH'S WATER USES
• Drinking
• Irrigation
• Fisheries
• Individual processes
• Transportation
• Waste disposal
HUMAN WATER CONTENT
45 - 75% by weight
65% average male
55% average female
Hydrogeology
9/93
-------
NOTES
Precipitation is a
beginning point for
the hydrologic cycle
HYDROLOGIC
CYCLE
Transpiration /
Precipitation
ivaporation
"Runoff
Water
table
DIRECT INFILTRATION
^Precipitation
Infiltration
9/93
Hydrogeology
-------
NOTES
CONTROLS ON INFILTRATION
• Soil moisture
• Compaction of soil
• Microstructures in the soil
• Vegetative cover
• Temperature
• Surface gradient
STREAM FLOW
Q = Av
GAINING STREAM
Discharge = 8 cfs
Discharge =10 cfs
Hydrogeology
9/93
-------
LOSING STREAM
Discharge = 10 cfs,
Discharge = 8 cfss
Ground surface
r
Vadose
zone
\^Pore spaces partially]
filled with water
Saturated
zone
i
#* from water taBle? -^r
"Groundwater
POROSITY
(n)
The volumetric ratio between the void
spaces (Vv) and total rock (V,):
n = Vv • n = S + S,,
NOTES
9/93
Hydrogeology
-------
NOTES
PRIMARY POROSITY
Refers to voids that were formed at
the same time the rock was formed
Void Space
Percent _
Porosity
Total Volume - Volume Soil Particles
Total Volume
Sand Grain
X 100
SECONDARY POROSITY
Refers to voids that were formed
after the rock was formed
Hydrogeology
9/93
-------
NOTES
SECONDARY POROSITY
PERMEABILITY
The ease with which water will
move through a porous medium
£>
'«
o „»
Percent Por
> A 0
> o c
-
c
:ia
POROSITY
L'D---D o":
y Sand Gravel Sandstone
9/93
Hydrogeology
-------
NOTES
CONDUCTIVITY
Clay Sand Gravel Sandstone
AQUIFER
A permeable geologic unit with the
ability to store, transmit, and
yield water in usable quantities
AQUITARD
A layer of low permeability that
can store and transmit groundwater
from one aquifer to another
Hydrogeology
9/93
-------
NOTES
AQUICLUDE
A confining layer with low
permeability that is essentially
impermeable
UNCONFINED AQUIFER
(Water Table)
A permeable geologic unit having the
ability to store, transmit, and yield
water in usable quantities
III
MCONFINED AQUIFER
T71
|
^
C
V
'
Vc
itE
ir
ta
b
e
Uhconfined aquifer '•
JXxxA^^V^
rXXXJ* Confmine
'S/Wv V V N/ V N^
AAX
: unit
V w
AA^vwyS(?S^S<
- aquitardO<><>
-------
NOTES
CONFINED AQUIFER
(Artesian)
An aquifer overlain by a confining layer
whose water is under sufficient pressure
to rise above the base of the confining
layer if it is perforated
(
3ONFINED AQUIFER
Confining
unit
- aqultard
''> 9*
&
£
';** o • -A *<% "»••
*
Potentlometric
surface
', •" * " ^ 4
, '". >'S ;'AV-!
-*^^^s-U^ ; "*{ ' ^ *
Confined' aquifer A •?
-.*.' "r- ,v ,,^«,vf
•i/. ,-"';v.'«.C;^
Confining unit • aquitard
Recharge Vadose zone
Water table
Confining layers
Hydrogeology
10
P/P3
-------
AQI
JIFERS AND AQUITARDS
^~ — '
$$99S'5x> s
UnconTineci.aciuiTer —
,* «.-,** =
Aquitard
^^^KiM^^^^ ^rfHp-
7>99999
= .Water /-
= table 'f.
•sz.
=
^
' Aquitard
^Confined aquifer^"' * •" n • f"-\
sz
?
ARTESIAN GROUNDWATER
SYSTEM
Recharge area
Flowing
artesian Recharge area
well
'.•/.'.sandstone'.'.'.'.'.'
^ ^ —
— — — ^.-
•^ _ Shale — _
POTENTIOMETRIC SURFACE
The elevation that water will rise to
in an opening (well) if the upper
confining layer of a confined
aquifer is perforated
NOTES
9/93
11
Hydrogeology
-------
NOTES
TOTAL HEAD
Combination of elevation (z) and
pressure head (hp)
ht • = z + hp
Total head is the energy imparted to a
column of water
SPECIFIC RETENTION
(SR)
The water in an aquifer that will not drain
by gravity and remains attached to the
aquifer media
SPECIFIC YIELD
(SyJ
The water in an aquifer that will
drain by gravity
Hydrogeology
12
9/93
-------
ROCK/WATER
RELATIONSHIPS
VOID SPACE
(Porosity)
SATURATION
NOTES
9/93
13
Hydrogeology
-------
NOTES
WATER RETAINED AFTER GRAVITY
DRAINAGE
(Specific Retention)
HYDRAULIC CONDUCTIVITY
(Kj __
The volume of flow through a unit
cross section of an aquifer per unit
decline of head
HOMOGENEOUS
Hydraulic conductivity is not
dependent on position within a
geologic formation
Hydrogeology
14
9/93
-------
HETEROGENEOUS
Hydraulic conductivity is dependent
on position within a geologic
formation
ISOTROPIC
Hydraulic conductivity is independent
of the direction of measurement at a
point in a geologic formation
ANISOTROPIC
Hydraulic conductivity varies with
the direction of measurement at a
point in a geologic formation
NOTES
9/93
15
Hydrogeology
-------
NOTES
ft
Pol«m'°m«nc Surface
V x"V-»»
AJ jEaj ^ •H3
"*• :M; • V?'lb$I
., •' "- • '-' '...-' /.....-.M ,-;A«$
DARCY'S LAW
• The flow rate through a porous material is
proproportional to the head loss and
inversely proportional to the length
of the flow path
• Valid for laminar flow
• Assume homogeneous and isotropic
conditions
tf
A
K
K
41 k ^\
L ' |
n ^ '
VJ ^ i 1
Gradient = H/L = 1, the energy required to
move water the distance L
Q = quant ty of flow(gpd)
A = cross-sectional area of flow(ft 3)
K = hydraulic conductivity = gpd/ft3
Hydrogeology
16
9/93
-------
DARCY'S LAW
Q = KIA
• Q = discharge
• K = hydraulic conductivity
• I = hydraulic gradient
• A = area
GROUNDWATER FLOW DIRECTION
Underflow
\
Mixed
NOTES
9/93
17
Hydrogeology
-------
NOTES
STREAM GRADIENTS
High
Low
Channel Gradients in Alluvial Systems
GW Flow Direction
I Underflow
D Mixed
Q Baseflow
SUnknown
MISSMOARK2SCI RQ HUMGMARK1 PT SP
Rivers
STREAM CHANNEL
Width-to-Depth Ratios
High
Low
Hydrogeology
18
9/93
-------
NOTES
Width-to-Depth Ratios in Alluvial Systems
250
GW Flow Direction
I Underflow
0 Mixed
Ell Baseflow
S Unknown
ARK2HUM MO SCI QM MISS RQ ARK1 PT SP
Rivers
STREAM CHANNEL
Width-to-Depth Ratios
High
-> Low
Stream Penetration in Alluvial Systems
c
o
0)
d>
0.
I
GW Flow Direction
• Underflow
Q Mixed
EH Baseflow
SP GM PT HUM ARK1 MO SCI MISS ARK2
Rivers
9/93
19
Hydrogeology
-------
NOTES
STREAM CHANNEL
SINUOSITY
Low <-
-> High
SINUOSITY OF RIVER CHANNELS
GW Flow Direction
I Underflow
n Mixed
D Baseflow
nknown
PT SP ARK1 QM HUM MO RQ ARK2 SCI MISS
Rivers
GREAT
MIAMI
IVER
Hydrogeology
20
9/93
-------
NOTES
ARKANSAS RIVER
BARRIER ISLAND
West Bay
Gulf of Mexico
9/93
21
Hydrogeology
-------
Section 3
-------
THE HYDROGEOLOGICAL
INVESTIGATION
PERFORMANCE OBJECTIVES
At the end of this lesson, participants will be able to:
• Describe three types of hydrogeological investigations
• Given a list of seven elements involved in conducting a
hydrogeological investigation, describe the processes
involved in accomplishing each element
• List three pertinent questions that must be addressed when
investigating groundwater contamination.
-------
THE HYDROGEOLOGICAL
INVESTIGATION
CONTAMINATION
•:= Leach ate
Groundwater
REGIONAL INVESTIGATIONS
• Cover large areas (10-100 square
miles)
• Are used to
- Locate potential sources
- Determine regional geology
- Determine regional hydrology
NOTES
9/93
The Hydrogeological Investigation
-------
NOTES
REGIONAL INVESTIGATIONS
• Include evaluation of
- General geology
- General hydrology
- Regional water quality
• Are not detailed studies
REGIONAL INVESTIGATION OF
COLBERT LANDFILL, SPOKANE, WA
Geology and hydrology
"...hydrogeologic conditions...
...dominated by...
...most recent glacial period..."
REGIONAL INVESTIGATION OF
COLBERT LANDFILL, SPOKANE, WA
Climate
"...values at site...are similar...
...to airport...
...because elevations are...
...similar."
The Hydrogeological Investigation
9/93
-------
LOCAL INVESTIGATIONS
• Cover few square miles
• Are used to
- Define geology, hydrology,
geochemistry, and water quality
- Locate sources
LOCAL INVESTIGATIONS
Include
- Topographic maps, geologic
maps, soils maps, well location
maps, and source locations
Are more detailed studies
LOCAL INVESTIGATION OF
COLBERT LANDFILL, SPOKANE, WA
Hydrogeology
"...system...defined as containing
three aquifers and three aquitards."
NOTES
9/93
The Hydrogeological Investigation
-------
NOTES
SITE INVESTIGATIONS
• Cover immediate area of site
• Are used to determine
- Site geology
- Site hydrology
- Contaminant migration controls
SITE INVESTIGATIONS
• Include
- Water level maps, geophysical
surveys, soil samples, water
samples, tank location maps,
tank inventories, and monitoring
wells
• Are most detailed (expensive) studies
CONDUCTING THE
INVESTIGATION
1. Establish objectives
2. Collect data
3. Conduct field investigation
4. Compile data
s. Interpret data
e. Develop conclusions
7. Present results
The Hydrogeological Investigation
9/93
-------
NOTES
ESTABLISH OBJECTIVES
• Detection
• Monitoring
• Site evaluation
• Selection of control methods
• Selection of treatment/remedial
methods
ESTABLISH THE OBJECTIVES
COLBERT LANDFILL, SPOKANE, WA
Focus
• Do affected residents have supplied
water?
• Is the landfill a continuing source?
« What are extent and processes for
offsite movement?
COLLECT DATA
Research Records
Maps (soil, geologic, topographic,
county, and state)
Aerial photographs
Location of pumping centers (wells)
and discharge rates
Stream discharge
9/93
The Hydrogeological Investigation
-------
NOTES
AERIAL PHOTOGRAPHY
Historical photography
(1920 - present)
Contract photography
(current site)
EPIC
Environmental Photographic
Interpretation Center
Western Region - EMSLVLas Vegas, Nevada
Eastern Region - EPA-EPIC/Warrenton, Virginia
USDA ASCS
(Agricultural Stabilization & Conservation Service)
Aerial Photography Field Office
Salt Lake City, Utah
(801) 525-5856
1945 - present
black & white
color infrared
The Hydrogeological Investigation
9/93
-------
NOTES
SANBORN FIRE
INSURANCE MAPS
• 1869 to 1950s
• Communities over 2000 population
• Updated periodically
• Locations of industries, pipelines,
storage vats, old dumps, and
wetlands
COLLECT DATA
Research Records
• Well logs
• Climatologica! data
• Chemical data
• Source or potential source location
COLLECT BACKGROUND DATA
COLBERT LANDFILL, SPOKANE, WA
Previous site use
"...electronics manufacturing...
...spent organic solvents...
...poured...down sides...
...hundred gallons a month."
9/93
The Hydrogeological Investigation
-------
NOTES
INFORMATION SOURCES
U.S. Geological Survey
State water commission
State geological survey
Departments of agriculture
INFORMATION SOURCES
• Soil Conservation Service
• Weather service
Site records
University research (theses, papers,
etc.)
CONDUCT FIELD INVESTIGATION
(After Record Search)
Note topography
Locate outcrops
Note joint patterns
Measure stream flow
Measure stream temperature
Note stream patterns
The Hydrogeological Investigation
9/93
-------
NOTES
CONDUCT FIELD INVESTIGATION
COLBERT LANDFILL, SPOKANE, WA
Monitoring wells
"...intended to compliment...
...wells installed during...
...previous investigations..."
CONDUCT FIELD INVESTIGATION
(After Record Search)
• Note flora
• Locate springs and seeps
• Conduct geophysical survey
• Note soil characteristics
• Note rock type
• Install monitoring wells
CONDUCT FIELD INVESTIGATION
(After Record Search)
Measure water levels
Conduct aquifer tests
Note aquifer thickness
Note confining layers
Determine hydrogeologic characteristics
Determine limits (hydrologic, geologic, and
climatic)
9/93
The Hydrogeological Investigation
-------
NOTES
CONDUCT FIELD INVESTIGATION
COLBERT LANDFILL, SPOKANE, WA
Soil characteristics
"...percolation tests indicate..
...surface runoff...
...occurs very infrequently..."
CONDUCT FIELD INVESTIGATION
(After Record Search)
• Conduct tracer studies
• Note hydraulic connections
• Collect samples
• Determine background water quality
• Identify types of contaminants
• Locate potential sources
• Determine types of aquifers involved
COMPILE DATA
Prepare water level maps
Prepare cross sections
Plot potentiometric surface
Determine contaminant mobility
characteristics
The Hydrogeological Investigation
10
9/93
-------
COMPILE DATA
COLBERT LANDFILL, SPOKANE, WA
Contaminant mobility
• Moving with gravity - DNAPLs
• Solubilized in groundwater flow
• Volatilized in vadose - molecular diffusion
INTERPRET DATA
Locate recharge areas
Locate discharge areas
Identify sources (responsible parties)
Predict contaminant impact and fate
INTERPRET DATA
COLBERT LANDFILL, SPOKANE, WA
Identify responsible parties
"...'secondary' sources...
...may be major source...
...continued contamination..."
NOTES
9/93
n
The Hydrogeological Investigation
-------
NOTES
DEVELOP CONCLUSIONS
• Is there a problem?
• How bad is it?
• Who is responsible?
• Can it be remediated?
• How?
PRESENT RESULTS
Review of reports by others
(responsible parties, consultants, etc.)
Formal report of investigation
Hearings
Public meetings
The Hydrogeological Investigation
12
9/93
-------
CHECKLIST FOR A HYDROGEOLOGICAL INVESTIGATION
HAZARDOUS WASTE SITES INFORMATION LIST
When evaluating activities at sites where hazardous wastes may be causing or contributing to
groundwater contamination, it is important to gather as much information as possible. The
development of as much site information as possible can often provide valuable insight about site
history, waste disposal practices, regional and local geology and the potential for impacts to the
environment in the site vicinity.
In order to make your information gathering efforts easier, what follows is a list of the types of
questions which may be helpful to a site investigation. While these questions are oriented more
towards field activities, the questions also may prove to be helpful to those people responsible for
evaluating the adequacy of other site assessment documents.
Sources
National Water Well Association. 1991. Groundwater and Unsaturated Zone
Monitoring and Sampling. 45 pp. In: Practical Handbook of Groundwater
Monitoring.
U.S. EPA. 1986. RCRA Ground Water Monitoring Technical Enforcement
Guidance Document. 208 pp.
Stropes, D.F. 1987. Unpublished Research: Technical Review of Hazardous
Wastes Disposal Sites. 25 pp.
I. SITE/FACILITY HISTORY
A. Waste disposal history of the site.
1. Is this a material spill or other emergency response activity not at a Toxic
Substances Storage and Disposal Facility (TSSDF) ?
2. What hazardous wastes are being manufactured, stored, treated or disposed
of at the site?
3. For active manufacturing operations, what industrial processes are being used
and what raw materials are used in the industrial processes?
4. Are the raw materials altered or transformed in any way during industrial
processes to result in waste materials which are different from the raw
materials?
5. How long has the facility been in operation?
9/93 13 The Hydro geological Investigation
-------
6. Have the types of hazardous wastes manufactured, stored, treated or disposed
of at the site changed during the history of the site?
7. Have the industrial processes used at the site changed over the history of the
site?
8. If the industrial processes are different, what previous industrial processes
were used in the past, how long, and what types of wastes were end products
of the processes.
9. What environmental media (air, land, water) have been or are being affected
by the facility/site activities?
10. What is the form of the site wastes (sludge, slurry, liquid, powder,
containerized, bulk storage)?
11. How much waste is generated or disposed of at the location daily?
12. What is the history of above and underground storage tank use at the site?
13. What types of regulated manufacturing or pollution control units exist at the
facility?
14. What governmental agencies are responsible for the regulated units?
15. Do any historical records about the site exist? If so, where are these records?
16. Has a check of any existing historical maps or aerial photos been done to
provide further insight as to past site activities?
17. Is there any history of groundwater contamination as a result of the site
activities?
B. Details of the site disposal activities.
1. Are the site disposal units currently in compliance with all applicable rules,
regulations and standards?
2. Are disposal areas isolated from the subsurface by the use of liners,
impermeable material, etc.?
3. What type of isolating material is in use?
4. Are multiple isolation systems in use?
5. Is a leachate/contaminant collection system in use?
The Hydrogeological Investigation 14 9/93
-------
6. Are any monitoring wells installed adjacent to the disposal/collection system
units?
7. Is there a surface water run-off control system?
8. Are any parts of the site/facility capped with an impermeable cover material?
9. What is the condition of the cap?
10. Are areas of previous wastes disposal well defined?
C. What is the nature of groundwater usage from aquifers beneath the site or in adjacent
areas?
1. Do any water supply wells exist in the aquifers beneath the site and adjacent
areas?
2. Are water supply wells used for potable water supplies or for industrial
process water?
3. Is the groundwater treated prior to use?
4. What are the water supply wells' pumpage rates? Daily? Monthly?
Annually?
5. What are the depths of the wells' screened intervals?
6. What other well drilling, well construction or well completion information is
available?
7. Do subsurface geologic well logs exist for the wells?
8. Are the wells up gradient, at, or down gradient of the site/facility?
9. Does pumping from these wells modify the regional groundwater table or
potentiometric surface?
II. HYDROGEOLOGIC CHARACTERIZATION
A. Has the purpose of the hydrogeologic investigation been clearly and adequately
defined?
1. Characterize the hydrogeologic system at the site.
9/93 15 The Hydro geological Investigation
-------
2. Determine whether there has been downgradient degradation of water quality
from a potential source of contamination.
3. Determine the upgradient source of contamination at a known downgradient
contamination receptor (well, spring, or surface water body).
B. Has the site location and all major site features been shown on a map?
1. Has the site been located on a state map?
2. Has the site been located on a USGS 7-1/2 minute topographic quadrangle
map published at a scale of 1:25,000?
3. Have coordinates for further site identification (latitude, longitude, degrees,
minutes, seconds, or a site specific grid system) been provided?
C. Has a base map of the site been prepared?
1. What is the map source?
2. Are air photos available?
3. Are all components of a map (north arrow, scale, map legend) shown and
defined on the base map?
4. Does the map show elevations and contours?
5. Is the scale of the map adequate to delineate on-site features' dimensions
adequately?
6. Does the map show features adjacent to the site which may be pertinent to the
hydrogeologic investigation?
7. Are all natural physical features (topography, surface waters, surface water
flow divides, etc.) shown on the map?
D. Has the subsurface geology been identified?
1. Is the geologic interpretation based on soil borings and well drilling logs?
2. Have any other reference materials been used?
3. Are aquifers present beneath the site?
4. Is the first aquifer encountered confined or unconfmed?
5. Are all aquifers and confining units continuous across the site?
The Hydro geological Investigation 15 9/93
-------
6. Have all geologic strata been described (thickness, rock type,
unconsolidated/consolidated materials, depth, etc.)?
7. Do multiple aquifers exist at the site?
8. Have any porous versus fractured flow media been described?
E. Do the driller's logs of the deepest borings at each well cluster show that soil
material samples were collected at five foot intervals? If not at what intervals were
samples taken?
1. Is there a stratigraphic log of the deepest boreholes?
2. Were the borings extended to a depth below any confining beds beneath the
shallowest aquifer?
3. Have enough borings of the area been done to adequately define the
continuity and thickness of any confining beds?
4. Have all logs been prepared by a qualified geologist, soil scientist or engineer
using a standardized classification system?
5. Were any laboratory tests conducted on the soil and soil material samples?
What types of tests were performed?
6. Were grain size distributions used to determine the size of the gravel pack or
sand filter placed in the annular space opposite the well screen?
F. Have field and/or laboratory permeability tests been performed to identify variations
in aquifer and confining bed properties?
1. What type(s) of tests were performed?
2. What was the range of hydraulic conductivity values found in the aquifer?
What was the arithmetic mean value?
3. What was the range of hydraulic conductivity values found in the confining
bed? What was the arithmetic mean value?
4. Where are the most permeable subsurface zones located relative to the waste
disposal facility?
5. Have geologic cross-sections been constructed?
G. Have field and/or laboratory tests been performed to determine the specific yield,
storativity or effective porosity of the aquifer?
9/93 17 The Hydro geological Investigation
-------
1. What type(s) of tests were performed?
2. What is the range of specific yield, storativity or effective porosity values?
3. What are the average values?
H. Has the horizontal groundwater flow direction been determined?
1. Have a minimum of three piezometers been installed to determine the
direction of flow in the aquifer?
2. Do any water-level readings show local variations of the water table caused
by mounds or sinks?
3. Do any identified mounds or sinks result in alterations of the regional or local
horizontal groundwater direction of flow?
4. Do any surface features which may have an effect on the horizontal flow
exist?
5. Have all piezometer installations in the uppermost aquifer been screened at
approximately the same depth below the water table?
6. Do any discernible seasonal variations in water levels exist?
7. Do any short term variations in water levels exist? If so what possible causes
may explain these variations?
I. Has the magnitude of the horizontal hydraulic gradient been determined at various
locations across the site?
1. What is the average horizontal hydraulic gradient at the site?
2. Where is the horizontal hydraulic gradient the steepest?
3. Does this location correlate to a known area of lower hydraulic conductivity
in the aquifer?
4. Does this location correlate to a known area of lower aquifer thickness?
5. Where is the horizontal hydraulic gradient the lowest (flat)?
6. Does this location correlate to a known area of greater hydraulic conductivity
in the aquifer?
7. Does this location correlate to a known area of greater aquifer thickness?
The Hydro geological Investigation ig 9/93
-------
J. If multiple aquifers exist, have wells been installed in each aquifer to determine the
vertical component of groundwater flow?
1. Have the wells in each aquifer been installed in a single borehole or in
separate boreholes?
2. If a single borehole was used, what tests were conducted to assure that no
leakage between the upper and lower aquifers exist?
3. If a single borehole was used, what well installation, construction and
development techniques were used to assure that no leakage between the
upper and lower aquifers exist?
4. Based on the difference in hydraulic head between upper and lower aquifers
can the site be described as:
a. predominantly a recharge area?
b. predominantly a discharge area?
c. predominantly an area of horizontal flow?
5. If recharge, discharge or horizontal flow areas exist, have these locations
been shown on a hydrogeologic map (including supporting cross-sections) of
the site?
K. Has the magnitude of the vertical hydraulic gradients been determined at various
locations across the site?
1. What is the average vertical hydraulic gradient at the site?
2. Where is the vertical hydraulic gradient the steepest?
3. Can this location be correlated to any known areas of lower hydraulic
conductivity?
4. Where is the vertical hydraulic gradient the flattest?
5. Based upon the vertical hydraulic gradient, what relationship exists between
the shallow and deeper aquifers?
6. Is there any regional or off-site vertical hydraulic gradient information which
may support or conflict with the site's vertical hydraulic gradient data?
L. Determination of seepage velocities and travel times.
1. What is the average seepage velocity of water moving from the waste facility
to the downgradient site boundary?
9/93 19 The Hydro geological Investigation
-------
2. What is the average travel time of water to move from the waste facility to
the nearest downgradient monitoring wells?
3. What is the basis for the seepage velocity and travel time determinations?
M. Have potentiometric maps, flow nets, geologic maps and cross-sections been prepared
for the purpose of showing the direction of groundwater flow at the site?
Horizontal Flow Components (plan view)
1. Do the contours and contour intervals between the equipotential lines
adequately describe the flow regime?
Suggested Contour Intervals:
a. 0.1 to 0.5 feet if the horizontal flow component is relatively flat.
b. 0.5 to 1.0 feet if the horizontal flow component is moderately steep.
c. 1.0 to 5.0 feet if the horizontal flow component is extremely steep.
2. Have the equipotential lines been accurately drawn?
a. With respect to the elevations of water levels in the wells or
piezometers?
b. With respect to nearby or on-site rivers, lakes, wells or other
boundary conditions?
c. With respect to other naturally occurring or manmade physical
features that might cause groundwater mounds or sinks in the area?
d. Do variations in the spacing of equipotential lines correspond to
known areas with relative transmissivity variations?
3. Do the constructed groundwater-flow lines cross equipotential lines at right
angles?
4. Can conclusions about the aquifer(s) relative homogeneity and isotropy be
made based on variations in the flow lines?
Vertical Flow Components (cross-sections)
1. Is the transect for the vertical flow component cross-section(s) laid out along
the line of a groundwater flow path as seen in the plan view?
2. Is the variation in land-surface topography accurately represented on the
cross-section(s)?
The Hydro geological Investigation 20 9/93
-------
3. Have both vertical and horizontal scales been provided?
4. What differences exist between the vertical and horizontal scales?
5. Are all monitoring wells, piezometers, and screened intervals accurately
shown?
N. What is the site water quality and geochemistry?
1. What are the upgradient-groundwater quality conditions?
2. What are the downgradient-groundwater quality conditions?
3. What water-quality parameters have been determined downhole?
4. What water-quality parameters have been determined at the well head?
5. Have all appropriate field water-quality determinations, equipment used and
procedures been followed?
6. Does an adequate QA/QC procedure exist?
7. What if any, relationship exists between the site water-quality conditions and
the past and/or present activities at the site?
DETECTION MONITORING SYSTEM
A. Are the facility upgradient and downgradient monitoring wells properly located to
detect any water-quality degradation from the waste source(s)?
Horizontal Flow
1. Will groundwater from the upgradient well locations flow through or under
the waste source in an unconfined aquifer?
2. Will groundwater from the upgradient well locations flow beneath the waste
source and under an overlying confining bed in a confined aquifer?
3. Will groundwater from the upgradient well locations flow beneath the waste
source in an unconfined aquifer separated from the waste source by an
impervious liner?
4. Will groundwater from the waste-source area flow toward downgradient
wells?
9/93 21 The Hydro geological Investigation
-------
Vertical Flow
1. Are the monitoring wells correctly screened to intercept a possible
contaminant plume from the waste source based on an accurate interpretation
of the vertical flow regime (recharge area, discharge area, or area of
horizontal flow)?
B. Are the monitoring wells located adequately to provide sufficient groundwater- flow
information?
1. Are regional water levels unaffected by local groundwater mounds or sinks?
2. Are additional monitoring wells located to provide water-level information
from local groundwater mounds or sinks?
3. Do upgradient and downgradient monitoring wells provide representative
samples?
C. Monitoring Well Construction
1. Were precautions taken during the drilling of the borehole and installation of
the well to prevent introduction of contaminants into the well?
2. Is the well casing and screen material inert to the probable major
contaminants of interest?
3. What type of well casing and screen material was used?
4. Does the casing and screen material manufacturer have any available
information about possible leaching of contaminants from the casing and
screen material?
5. How are the well casing and screen segments connected?
6. If cement or glue has been used what is the potential for contaminants to
leach into the groundwater?
7. Were all downhole well components steam cleaned prior to installation?
8. If another cleaning technique was used, what materials were used?
D. Are there as built drawings or details of each monitoring well nest or cluster showing
depth of well, screen intervals, type and size of screen, length of screen and riser,
filter packs, seals, protective casings, etc.?
1. Do the figures show design details of as built wells as opposed to details of
proposed wells?
The Hydro geological Investigation 22 9/93
-------
2. Are the well depth(s) and diameter(s) shown?
3. Are the screened intervals and type and size of screen openings shown?
4. Is the length of the screen shown?
5. Is the length of the riser pipe and stick-up above the land surface shown?
6. Is the filter pack around the screened interval shown?
7. Does the filter pack extend at least one foot above and below the screened
intervals?
8. What type of sealant was placed in the annulus above the filter pack?
9. What is the thickness of the seal?
10. How was the seal put in place?
11. Will a protective casing or reinforced posts be necessary to protect the
monitoring well from damage?
12. Is any manufacturer's information available to verify that all materials used
in the well construction do not represent potential sources of water
contamination?
13. Have samples of well construction materials used been kept for future
analysis to verify that the materials do not represent sources of water
contamination?
E. Are the screened intervals appropriate to the geologic setting and the sampling of a
potential problem?
1. Is the screen set opposite a stratigraphic layer with relatively high hydraulic
conductivity?
2. Is the screened interval set sufficiently below the water table so that water-
level measurements can be taken and water samples can be collected during
periods of low water level?
3. Is the screened interval placed in the aquifer(s) of concern?
4. If a single long screen was installed over the entire saturated thickness of the
aquifer, what effect will this have on analytical data from this monitoring
well?
5. Has the entire aquifer thickness been penetrated and screened?
9/93 23 The Hydro geological Investigation
-------
6. Have piezometers been installed to determine vertical and horizontal flow
directions?
7. Is the base of the waste disposal unit above the seasonal high water table?
8. What is the thickness of the unsaturated zone between the base of the waste
disposal unit and the seasonal high water table?
F. Has a professional survey been conducted to determine the elevation and location of
the measuring point at each well with reference to a common datum?
1. Is the survey at each monitoring well accurate to ±0.01 feet?
2. Is each surveyed measuring point located at the top of the well casing?
3. What benchmark was used as a starting point for the survey?
4. Are the elevations of the measuring point at each well referenced to mean sea
level and not some local datum?
G. Has an adequate sampling and analysis program been written for the site?
1. Are the major contaminants inorganic compounds?
2. Will field filtering and preservation be done in the field?
3. Are the major contaminants organic coumpounds?
4. Where would you expect to find the contaminants in the aquifer?
a. Floating at the top of the aquifer (LNAPLs)?
b. Dissolved in the groundwater and flowing with it?
c. Concentrated at the bottom of the aquifer (DNAPLs)?
5. What are the possible degradation end products of the original organic
contaminants?
6. Is the sampling method adequate to prevent any loss of volatile constituents?
7. Are field measurements such as pH, Eh, specific conductance, dissolved
oxygen and temperature taken in the field?
8. Does a generic sampling and analysis protocol exist?
9. Does the sampling and analysis protocol address sample preservation, storage,
transport, container identification and chain of custody procedures?
The Hydro geological Investigation 24 9/93
-------
10. Is the analysis laboratory certified by EPA for the analyses to be performed?
11. Did the laboratory provide input to the sampling and analysis program?
12. Is the sampling and analysis program written, clear, concise, understandable
and site specific?
H. Has a QA/QC plan been written for the groundwater monitoring program?
Water-Level Measurements
1. Have worksheets containing relevant fixed data, some of which are indicated
below, been prepared for use by the person taking water-level readings?
a. Well identification number?
b. Location of measuring point of each well?
c. Elevation of measuring point at each well relative to mean sea level?
d. Elevations of screened interval at each well?
e. Type of measuring instrument to be used?
2. Do the worksheets for use by the person taking water-level readings have
columns for computation of:
a. Depth to the water table?
b. Measuring point data to be added to or subtracted from readings of
measuring instrument?
c. Adjusted depth to water surface?
d. Conversion of depth to water surface?
3. Do the worksheets have a space for pertinent comments?
Sample Collection
1. Are well purging procedures prior to sampling described as written
procedures?
2. Is the method of purging specified?
3. Is the sample collection technique specified?
4. Is the sample storage vessel described?
9/93 25 The Hydro geological Investigation
-------
5. Is the sample volume specified?
6. Is the sample identification system described?
7. Are there provisions for:
a. Trip blanks?
b. Spiked samples?
c. Duplicate, replicate or blind samples?
8. Is the sampling frequency specified?
Sample Analysis
1. Does the laboratory have a QA/QC program for all samples?
2. Is the QA/QC program written?
3. Does the laboratory provide information on the accuracy and precision of the
analytical results?
4. Did the laboratory participate in the development of the sampling and analysis
plan for the groundwater monitoring program and the QA/QC plan for the
non-laboratory portion of the sampling program?
I. Is the QA/QC plan being followed during implementation of the sampling and
analysis program?
1. Is the same consultant who prepared the QA/QC plan responsible for its
implementation?
2. How many copies are there of the QA/QC plan?
3. Who has copies and where are they located?
4. Does the field person taking water-level measurements and collecting samples
have a copy?
5. Does the field person understand the importance of following the QA/QC
plan explicitly each time?
6. What safeguards and checks are there to assure there will be no deviation
from the QA/QC plan in the field and the laboratory?
J. If any more field work or data will be necessary to meet the objectives of the
hydrogeologic investigation, what types of additional field installations and data will
be needed?
The Hydro geological Investigation 26 9/93
-------
Section 4
-------
GEOPHYSICAL METHODS
PERFORMANCE OBJECTIVES
At the end of this lesson, participants will be able to:
• Describe the five geophysical techniques listed below:
1. Magnetics
2. Electromagnetics (EM)
3. Electrical resistivity
4. Seismic refraction
5. Ground-penetrating radar
• List the physical properties measured by the five geophysical
techniques
• Identify interferences for each the geophysical technique
• Explain the need for geological control when making
geophysical interpretations.
-------
NOTES
GEOPHYSICAL METHODS
GEOPHYSICS
• Nonintrusive, investigative tool
• Methods are site specific
• Data must be "ground truthed"
• Professional interpretation necessary
RELATIVE SITE COVERAGE
Volume of typical Volume of drilling
geophysical measurement or water sampling
9/93
Geophysical Methods
-------
NOTES
"GROUND TRUTHING"
Correlation of physical evidence
(i.e., rock cores) to geophysical
data
ANOMALY
Significant variation from background
GEOPHYSICAL TECHNIQUES
• Magnetics
• Electromagnetics (EM)
• Electrical resistivity
• Seismic refraction
• Ground-penetrating radar
• Borehole geophysics
Geophysical Methods
9/93
-------
MAGNETICS
Measurement of magnetic field strength
in units of gammas
Anomalies in magnetic field strength are
primarily caused by variations in
concentrations of ferromagnetic
materials in the vicinity of the sensor
MAGNETICS
Advantages
Relatively low cost (cost-effective)
Short time frame required
Little, if any, site preparation needed
Simple survey sufficient (Brunton)
MAGNETICS
Disadvantages
Cultural noise limitations
Difficulty in differentiating between steel
objects
NOTES
9/93
Geophysical Methods
-------
NOTES
ELECTROMAGNETICS
• Based on physical principles of inducing
and detecting electrical flow within
geologic strata
• Measures bulk conductivity (the inverse
of resistivity) of geologic materials
beneath the transmitter and receiver coils
• Currents are induced by application of
time-varying magnetic fields
ELECTROMAGNETICS
Advantages
Rapid data collection with minimum
personnel
Lightweight, portable equipment
Commonly used in groundwater pollution
investigations
ELECTROMAGNETICS
Disadvantages
Cultural noise limitations
Limitations in areas where geology varies
laterally (anomalies can be misinterpreted
as plumes)
Geophysical Methods
9/93
-------
NOTES
ELECTRICAL RESISTIVITY
Measures the bulk resistivity of the
subsurface in ohm-meters
Current is injected into the ground
through surface electrodes
ELECTRICAL RESISTIVITIES OF
GEOLOGIC MATERIALS
Function of:
« Porosity
• Permeability
• Water saturation
• Concentration of dissolved solids in
pore fluids
ELECTRICAL RESISTIVITY
Advantages
Qualitative modeling is possible
Models can be used to estimate depths,
thicknesses, and resistivities of
subsurface layers
Layer resistivities can be used to estimate
resistivity of saturating fluid
9/93
Geophysical Methods
-------
NOTES
ELECTRICAL RESISTIVITY
Disadvantages
• Cultural noise limitations
• Large area free from grounded metallic
structures required
• Labor intensive ( 2-3 person crew)
SEISMIC TECHNIQUES
• Seismic refraction
• Seismic reflection
SEISMIC REFRACTION
• Measures travel time of acoustic wave
refracted along an interface
• Most commonly used at sites where
bedrock is less than 500 feet below
ground surface
Geophysical Methods
9/93
-------
NOTES
SEISMIC REFRACTION
Advantages
• Can determine layer velocities
• Can calculate estimates of depths to
different interfaces
• Can obtain subsurface information
between boreholes
• Can determine depth to water table
SEISMIC REFRACTION
Assumptions
Velocities of layers increase with depth
Velocity contrast between layers is
sufficient to resolve interface
Geometry of geophones in relation to
refracting layers will permit detection of
thin layers
SEISMIC REFRACTION
Disadvantages
Assumptions must be made
Assumptions must be valid
9/93
Geophysical Methods
-------
NOTES
SEISMIC REFLECTION
• Measures travel time of acoustic wave
reflected along an interface
• Precise depth determination cannot be
made without other methods
• Magnitude of energy required is limiting
factor
GROUND-PENETRATING RADAR
A transmitter emits pulses of
high-frequency electromagnetic waves
into the subsurface which are scattered
back to the receiving antenna on the
surface and recorded as a function of
time
GROUND-PENETRATING RADAR
• Depth penetration is severely limited by
attenuation of electromagnetic waves into
the ground
• Attenuating factors
- Shallow water table
- Increase in clay content in the
subsurface
- Electrical resistivity less than
30 ohm-meters
Geophysical Methods
9/93
-------
NOTES
GROUND-PENETRATING RADAR
Advantages
• Continuous display of data
• High resolution data under favorable site
conditions
• Real-time site evaluation possible
GROUND-PENETRATING RADAR
Disadvantages
• Limitations of site-specific nature of
technique
• Site preparation necessary for survey
BOREHOLE GEOPHYSICS
9/93
Geophysical Methods
-------
NOTES
Spontaneous
potent al
Resistivity
— short
— long
Geologic
log
clay
sand
few clay
layers
(fresh water)
shale
dense rock
IMS
sandstone
SH layers
(brackish
water)
shale
few SS
layers
sandstone
(saline water)
(weathered)
dense rock
probably
granite
Gamma
ray
Neutron
Comparison of electric and
radioactive borehole logs
BOREHOLE GEOPHYSICS
• Spontaneous potential
• Normal resistivity
• Natural-gamma
• Gamma-gamma
Geophysical Methods
10
9/93
-------
NOTES
BOREHOLE GEOPHYSICS
• Neutron
• Caliper
• Acoustic
• Temperature
SPONTANEOUS POTENTIAL
Records natural potential between
borehole fluid and surrounding materials
Mainly used for geologic correlation,
determining bed thickness, and separating
nonporous from porous rocks
(i.e., shale-sandstone, shale-carbonate)
Can only be run in open, fluid-filled
boreholes
RESISTIVITY
Measures apparent resistivity of a volume
of rock/soil surrounding the borehole
Radius of investigation is generally equal to
the distance between the borehole current
and measuring electrodes
Can only be run in open, fluid-filled
boreholes
9/93
11
Geophysical Methods
-------
NOTES
GAMMA
• Measures the amount of natural-gamma
radiation emitted by rocks/soils
• Main use is for identification of lithology
and stratigraphic correlation
• Can be run in open or cased and fluid- or
air-filled boreholes
GAMMA-GAMMA
Measures the intensity of gamma
radiation from a source in the probe
after it is backscattered and attenuated
in the rock/soils surrounding the borehole
Main use is for identification of lithology
and measurement of bulk density and
porosity of rocks/soils
Can be run in open or cased and fluid- or
air-filled boreholes
NEUTRON
• Measures moisture content in the vadose
zone and total porosity in sediments and
rocks
• Neutron sources and detector are
arranged in logging device so that output
is mainly a function of water within the
borehole walls
• Can be run in open or cased and fluid- or
air-filled boreholes
Geophysical Methods
12
9/93
-------
NOTES
CALIPER
Records borehole diameter and provides
information on fracturing, bedding plane
partings, or openings that may affect
fluid transport
Can be run in open or cased and fluid- or
air-filled boreholes
ACOUSTIC
A record of the transit time of an
acoustic pulse emitted and received by
the logging tool
Response is indicative of porosity and
fracturing in sediments or rocks
Can be run in open or cased, fluid-filled
boreholes
TEMPERATURE
A continuous record of the temperature
of the environment immediately
surrounding the borehole
Information can be obtained on the
source and movement of water and the
thermal conductivity of rocks
Can be run in open or cased, fluid-filled
boreholes
9/93
13
Geophysical Methods
-------
Section 5
-------
MONITORING THE
VADOSE ZONE
PERFORMANCE OBJECTIVES
At the end of this lesson, participants will be able to:
• Describe the vadose zone
• List three reasons why the vadose zone is important in
groundwater investigations
• Describe the principles of soil gas wells
• Describe the operation of pressure vacuum lysimeters
• Characterize the limitations of soil gas wells
• Characterize the limitations of vacuum lysimeters.
-------
MONITORING THE
VADOSE ZONE
THE VADOSE ZONE
Consists of:
• Soils and particulate material
• Vapors in pore spaces
• Liquids on grain surfaces
Ground surface
r
Vadose
zone
_
"~\Pore spaces partially-^
filled with wate
Saturated
zone
. Groundwater.
NOTES
9/93
Monitoring the Vadose Zone
-------
NOTES
PHYSICAL PROPERTIES
Vadose Zone
• Organic matter
• Lithology/stratigraphy
• Thickness
• Grain size distribution
PHYSICAL PROPERTIES
Vadose Zone
• Water content
• Soil density
• Specific yield
• Specific retention
CHEMICAL CHARACTERISTICS
Vadose Zone
• Soil vapors/gases
• Pore water
Monitoring the Vadose Zone
9/93
-------
NOTES
WATER QUALITY
Common Parameters
• Temperature
• pH
• Conductivity
• Chemical analysis
THE VADOSE ZONE
Consists of:
• Soils and particulate material
• Vapors in pore spaces
• Liquids on grain surfaces
TENSIOMETER
Measures the capillary pressure in soil
Advantages
— Inexpensive
- Durable
- Easy to operate
9/93
Monitoring the Vadose Zone
-------
NOTES
TENSIOMETER
Disadvantages
- Ineffective under very dry
conditions because of air entry
- Sensitive to temperature changes
- Sensitive to atmospheric pressure
changes
- Sensitive to air bubbles in lines
- Requires a long time to achieve
equilibrium
ELECTRICAL
RESISTANCE BLOCKS
c
1
1
i
~~\ Current source
«-»
Water
«->
Water
content
^^ Field calibration
Resistance
ELECTRICAL
RESISTANCE BLOCK
Measures moisture content in soil
Advantages
- Suited for general use
- Inexpensive
- Can determine suction or moisture
content
- Requires little maintenance
Monitoring the Vadose Zone
9/93
-------
ELECTRICAL
RESISTANCE BLOCK
Disadvantages
- Ineffective under very dry
conditions
- Sensitive to temperature
- Calibration is time-consuming
- Affected by salinity
NEUTRON MOISTURE LOGGING
Am-Be
Detector
• Neutrons from source are slowed down by
hydrogen cloud
• Hydrogen sources are water and
contaminants
NEUTRON
MOISTURE LOGGING
Interacts with hydrogen in water
Advantages
- Readings directly related to soil
moisture
- Moisture content can be measured
regardless of physical state
NOTES
9/93
Monitoring the Vadose Zone
-------
NOTES
NEUTRON
MOISTURE LOGGING
Disadvantages
- Expensive
- No information on soil-water
pressure
- No information on changes in
density
- Not accurate for small changes
- Requires care in handling source
- Requires license to use instrument
GAMMA-RAY ATTENUATION
Cs
Detector
Changes in attenuation indicate
differences in moisture content
GAMMA-RAY ATTENUATION
Determines soil density
Advantages
- Can measure wetting front within
2cm
Monitoring the Vadose Zone
9/93
-------
NOTES
GAMMA-RAY ATTENUATION
• Disadvantages
- Expensive
- Radioactive source requires
special care
- Changes in bulk density affect
calibration (e.g., swelling and
frost heave)
PSYCHROMETER
Measures relative humidity of soil
water
Advantages
- Measures capillary pressure
under very dry conditions
PSYCHROMETER
Disadvantages
- Very sensitive to temperature
fluctuations
- Expensive
- Complex
- Performs poorly in wet media
9/93
Monitoring the Vadose Zone
-------
NOTES
FLOW RATES
• Infiltrometers (constant head)
• Test basins (falling head)
• Water budgets (hydrologic cycle)
• Tracer studies
- Dyes
- Radioactive isotopes
- Selected ions
• Problem:
Water is held under tension in
the vadose zone and will not
flow into wells
• Solution:
Create an area of lower potential
to induce flow into a sampling
device
LYSIMETER
A device for sampling interstitial
moisture in the unsaturated zone
Monitoring the Vadose Zone
9/93
-------
LYSIMETER INSTALLATION
Tubing
•4
Reservoir
pipe
Powdered silica sand
EVACUATION OF LYSIMETER
Closed valve
Mine
spoil
Suction:
pump
. Backfill
soil
— Ceramic tip
COLLECTION OF PORE WATER
Both valves closed
NOTES
9/93
Monitoring the Vadose Zone
-------
NOTES
TRANSFER TO SAMPLE BOTTLE
SAMPLING TIME vs. SAMPLE VOLUME
Time after evacuation (days)
SOIL GAS SURVEYS
SOURCE
Monitoring the Vadose Zone
10
9/93
-------
SOIL GAS WELL
Schematic
H
Vadose
zone
Seal
Well
Soil gas
SOIL GAS SURVEYS
xxxxxxxxxx
X _ X
II
SOURCE
——
XXXXXXX + +
SOIL GAS SURVEYS
xxxxxxxx
_mmm^^^^^^ x
SOURCE I +
1 + + +
X X X X X
NOTES
9/93
11
Monitoring the Vadose Zone
-------
NOTES
CROSS SECTION
I Source V Vapors
Vadose
II!
Leachate
•:::::::• Plume .::::::: :j:;::::,
••::;.: :_:•..-..-._.,..- •-• •• "
Saturated MH^ Regional now
PLAN VIEW
Regional Flow
Monitoring the Vadose Zone
12
9/93
-------
Section 6
-------
WELL CONSTRUCTION
PERFORMANCE OBJECTIVES
At the end of this lesson, participants will be able to:
• List six types of drilling methods
• Identify conditions under which the following types of
drilling methods would be used:
1. Air rotary
2. Mud rotary
3. Cable tool
4. Hollow-stem auger
• Describe three methods of well development
• Define the major components of a monitoring well from a
diagram
• List three uses for wells.
-------
NOTES
WELL CONSTRUCTION
USES FOR WELLS
• Monitoring
° Remediation
• Lithology
• "Ground truthing"
TYPES OF DRILLING METHODS
• Mud rotary
• Air rotary
• Cable tool
• Reverse circulation
• Solid-stem auger
• Hollow-stem auger
9/93
Well Construction
-------
NOTES
MUD ROTARY
Advantages
• Availability
• Satisfactory drilling in most formations
• Good depth capability
• Wide variety of formation logging
• Modest cost
• Good gravel pack and casing seal
MUD ROTARY
Disadvantages
Requires drilling fluid
- Difficult to remove
- May affect sample integrity
Circulates contaminants
Mobility may be limited
Poor rock or soil sample recovery
AIR ROTARY
Advantages
• No drilling fluid required
• Excellent drilling in hard rock
• Good depth capability
• Excellent delineation of water-bearing
zones
• Potential to evaluate hydraulic properties of
water-bearing zones
Well Construction
9/93
-------
NOTES
AIR ROTARY
Disadvantages
Casing may be required during drilling
Cross contamination of different formations
possible
Limited equipment availability/mobility
Difficult formation sampling
High cost of drilling
CABLE TOOL
Advantages
Good sample recovery
Good delineation of water-bearing zones
during drilling
Highly mobile
Good drilling in most formations
Inexpensive
CABLE TOOL
Disadvantages
Slow
Requires driving casing in unconsolidated
formations
May be necessary to double case hole for
good seal or gravel pack installation
9/93
Well Construction
-------
NOTES
REVERSE CIRCULATION
Advantages
Formation water is not contaminated by the
drilling water
Good sample recovery
No caving in unconsolidated formations
REVERSE CIRCULATION
Disadvantages
Not readily available
Expensive
Sealing of wells and placement of grout
may be difficult
SOLID-STEM AUGER
Advantages
Fast in shallow, unconsolidated formations
Inexpensive to operate
Highly mobile
Requires no drilling fluid
Well Construction
9/93
-------
NOTES
SOLID-STEM AUGER
Disadvantages
• Cannot be used in consolidated formations
• Limited depth capability (175-200 feet)
• Possible borehole collapse after auger is
removed
• Difficult sampling
HOLLOW-STEM AUGER
Advantages
Highly mobile
No drilling fluid required
Problems of hole caving minimized
Soil sampling relatively easy
HOLLOW-STEM AUGER
Disadvantages
Cannot be used in consolidated formations
Limited depth capability (175-200 feet)
Cross contamination of permeable zones is
possible
Limited casing diameter
9/93
Well Construction
-------
NOTES
MONITORING WELL
CONSTRUCTION
WELL CONSTRUCTION
MATERIALS
• Well screen/riser/well points
- Teflon®
- Stainless steel
- PVC
• Sand/gravel/filter pack
• Bentonite/grout/cement
MONITORING WELL
CONSTRUCTION
Unconfined aquifer
Confined aquifer
Well Construction
9/93
-------
NOTES
MONITORING WELL - UNCONFINED AQUIFER
A Steel cap
Well cap ^
Riser —
V
Well screen —
*c
p_
A
i
'•»•»
-
3
%<
£
<
i
_ -•*
^^. Grout
— Grout
— Bentonite
— Gravel pack
!§&5&&M*M^
MONITORING WELL - CONFINED AQUIFER
cap
Well screen
Plug.
• Bentonite
- Gravel pack
WELL AND AQUIFER
DEVELOPMENT
• Surge block
• Bailer
• Pulse pumping
• Air surging
9/93
Well Construction
-------
NOTES
POOR WELL DEVELOPMENT
WELL DEVELOPMENT - SURGE BLOCK
1
»>
WELL DEVELOPMENT - BAILER
v •/ -
Construction
9/93
-------
NOTES
WELL DEVELOPMENT - PULSE PUMPING
-*
fit;
Pul<
'*, 'V ' '*
»* "f-
Pulse pumping
3S<^g>(Sj>i®«S!i«*^^
WELL DEVELOPMENT - AIR SURGING
, , ^ k.
^Ff^l
^nK,l »-,-%. . . • •" ,„,
Sft^'f'v'"^
•>|i||o2, '->^ "
:::::::::::.::::::::4r
, ,x; . .
tr •, • •
-H~
fe_^
——;_____
_. .
Air
*::::::.:::::.::::::::::
. xxXxxxxxxxxxxxxxxxxxxxxxx xx xxxx x
SUMMARY OF DRILLING AND
MONITORING WELL CONSTRUCTION
Hydrogeologic environment
- Type of formation
- Depth of drilling
Type of pollutant
Drilling location
Monitoring well design
Drilling equipment availability
Cost
9/93
Well Construction
-------
Section 7
-------
HYDROGEOCHEMISTRY
PERFORMANCE OBJECTIVES
At the end of this lesson, participants will be able to:
• Evaluate groundwater uses based on chemical parameters
• Identify the basic inorganic constituents in groundwater
• Identify the driving factors that control the concentration of
inorganic constituents, including:
pH and Eh
Temperature
Total dissolved solids
Dissolved gases
• Identify how the driving factors that control the
concentration of inorganic constituents influence:
Dissolution and precipitation
Redox potential (Eh)
Adsorption
Hydrolysis
Carbonate equilibrium
• List the chemical characteristics important to the
concentration of organic constituents in groundwater
-------
• Define the terms weight fraction of organic carbon in the soil
(foc), organic-carbon partition coefficient (Koc), distribution
coefficient (KD), retardation factor (RD), and octanol-water
partition coefficient (Kow)
• Define dense nonaqueous phase liquids (DNAPLs) and light
nonaqueous phase liquids (LNAPLs).
-------
NOTES
HYDROGEOCHEMISTRY
PRIMARY DRINKING
WATER STANDARDS
Inorganics
Microbiological
Pesticides/herbicides
Volatile organic compounds
Radioactivity
SECONDARY DRINKING
WATER REGULATIONS
Chloride
Color
Copper
Corrosivity
Fluoride
Foaming agents
Iron
Manganese
Odor.
PH
Sulfate
Total dissolved solids
Zinc
9/93
Hydrogeochemistry
-------
NOTES
QUALITY DETERMINES USABILITY
Taste
Odor
Poisons
Fluoride
Nitrate
Iron
Hardness
Sediment
Dissolved solids
isaUw
DOMESTIC
pH
Acidity
Alkalinity
Silica
Hardness
Sediment
Dissolved solids
INDUSTRIAL
Boron
Alkalinity
Sodium-calcium ratio
Dissolved solids
^ fiyfa j^jj
IRRIGATION
INORGANIC
GEOCHEMISTRY
SURFACE WATER
CHEMICAL COMPOSITION
• Rain water
• Seawater
• River water
Hydrogeochemistry
9/93
-------
NOTES
SURFACE WATER
COMPOSITION
Chemical Rain Water River Water
Ca
++
0.015
0.075
0.027
All concentrations in mg/L
0.020
38.000
10.000
SURFACE WATER
COMPOSITION
Chemical Rain Water River Water
Na+
cr
F-
All concentrations in mg/L
0.220
0.072
20.000
2.900
24.000
0.300
SURFACE WATER
COMPOSITION
Chemical Rain Water River Water
S04" 1.100 51.000
HCO~ —- 113.000
NO; -- 2.400
o
All concentrations in mg/L
9/93
Hydrogeochemistry
-------
NOTES
DOMESTIC WATER QUALITY
Example
Bolton Well Field
Great Miami River Aquifer
DOMESTIC WATER SUPPLY
Bolton Plant-Great Miami River
Chemical Raw Water
Finished
Water
Fe++
Ca++
Mg++
0.136 0.034
90.000 31.000
21.000 20.000
All concentrations in mg/L
DOMESTIC WATER SUPPLY
Bolton Plant-Great Miami River
Chemical Raw Water
Finished
Water
Na+ 33.800 32.000
K+
Cl" 54.000 59.000
F" 0.250 1.020
All concentrations in mg/L
Hydro geochemistry
9/93
-------
DOMESTIC WATER SUPPLY
Bolton Plant-Great Miami River
Chemical Raw Water
Finished
Water
so;
HCO-
NO;
54.000
—
2.200
51.000
—
2.310
All concentrations in mg/L
OTHER WATER
QUALITY PARAMETERS
Chemical _ . ... . _. ... , „ , Great Miami
Parameter RalnWater R'ver Water Seawater R|ver
Hardness —- 138
TDS 1.609 232
pH 4.9 7.4
R. / F.
6581.55 C. 318/164
34500 463 / 323
8.0-8.4 7.4 / 9.3
All concentrations in mgIL
C. Calculated; R. Raw Water; F. Finished Water •
GROUNDWATER QUALITY
• Chemicals/compounds present
• Chemical concentration
• Subsurface distribution
NOTES
9/93
Hydrogeochemistry
-------
NOTES
NATURAL ORGANIC CONSTITUENTS
Constituent
mg/L
Bicarbonate (HC03) 150-200
Carbonate (C03) 150-200
Calcium (Ca) 25-30
Magnesium (Mg) 25-30
Chloride (Cl) 250
Fluoride (F) 0.7-1.2
Iron (Fe) >0.3
Manganese (Mn) >0.05
Sodium (Na) 20-170
Sulfate(S04) 300-1000
CHARACTERISTICS
• Hardness
• pH (or hydrogen ion activity)
• Specific electrical conductance
• Total dissolved solids (IDS)
HARDNESS
Expressed as calcium carbonate in
milligrams per liter or grains per
gallon of water
One grain is equivalent to 17 mg/L
Hydrogeochem is try
9/93
-------
NOTES
HARDNESS
Type
mg/L
Soft 0-60
Moderately hard 61-120
Hard 121-180
Very hard > 180
DRIVING FACTORS
DRIVING FACTORS
• pH and Eh
• Temperature
• Total dissolved solids
• Dissolved gases
• Aquifer and soil mineral composition
9/93
Hydrogeochemistry
-------
NOTES
Chemical Processes
Affected by These
Driving Factors
CHEMICAL PROCESSES
• Dissolution and precipitation
• Carbonate equilibrium
• Hydrolysis
• Adsorption
• Redox potential
DISSOLUTION/PRECIPITATION
NaCI
Dependent on:
Solubility
PH
Temperature
Hydrogeochemistry
9/93
-------
DISSOLUTION PRECIPITATION
ca+++ co;
Mg + CO"
Na++ Cf
++
Ca+
CaCO3
MgCO3
NaCI
CaSCX
CARBONATE EQUILIBRIA
H2C03
H +HCO,
o
++ C0r
CaCO, ^Ca + C0~
HYDROLYSIS
R-X + H20
R-X + OH"
R-OH + H + X
-> R-OH +X
NOTES
9/93
Hydrogeochemistry
-------
NOTES
ADSORPTION
Partitioning of elements
Cation exchange capacity
(CAT)ION EXCHANGE
CAPACITY
• Retards movement of chemical constituents in
groundwater
• Amount of exchangeable ions in
milliequivalents per 100 grams soil at pH = 7
ADSORPTION/DESORPTION
Physical
Electrical
Q Q Q Q
^^^^^^^^^^^}
Clay
Hydrogeochemistry
10
9/93
-------
NOTES
OXIDATION/REDUCTION
REACTIONS
Oxidation = reaction resulting in
a loss of electrons
Reduction = reaction resulting in
a gain of electrons
Eh = "redox" potential
Low or negative Eh = reduction
High or positive Eh = oxidation
SOIL ZONE REACTIONS
(02>0.1 ppm)
78° mv
*"
Fe +
HS ~-> SO"
sulfide oxidation
iron oxidation
nitrification N
manganese oxidation Mn -^ Mn<+
iron sulfide oxidation Fe2S^Fe 3++ SO «
aerobic respiration CHjO +02 >COa + H>0
SATURATED ZONE REACTIONS
denitrification N03-^ N2(gas)
manganese reduction Mn4*-^ Mn2*
iron III reduction Fe3i-> Fe2+
sulfate reduction SO^' -^ HS"+ H2S
methane fermentation CH20-^ CH4(gas)
LU
nnn
Recharge zone
D.O. > 1.0
"+" Eh
"-"Eh D.O. < 1.0
•4~ To discharge zone
9/93
11
Hydrogeochemistry
-------
NOTES
Eh"
(volts)
THE LIMITS OF WATER
STABILITY
O2+2H + 2e~ = 2H20
H2+ O -2e~= H2O
LU
Trans/f
/0na/
Hydrogeochemistry
12
9/PJ
-------
Mine
waters
4.0
0.8V
WATER IN CONTACT WITH
ATMOSPHERE
Normal Aerated
Rain Streams ocean saline
water residues
pH
-Eh
10.0
0.3V
LU
Tran
s'tfona/
TRANSITIONAL WATERS
Bog waters Groundwater
3.0
0.1 V
pH
Eh
9.0
-0.2V
NOTES
9/93
13
Hydrogeochemistry
-------
NOTES
.
HI
pH
WATER ISOLATED FROM
ATMOSPHERE
Saturated Euxenic Organic-rich
.. marine saline
soils
environment waters
5.0
-0.1 V
pH
Eh
11.0
-0.5V
ORGANIC CHEMISTRY
Hydrogeochemistry
14
9/93
-------
NOTES
CHEMICAL-SPECIFIC
CHARACTERISTICS
Chemical phase (solid, liquid, gas)
• Solubility
• Vapor pressure
• Specific gravity
* Koc
• Kow
VOLATILIZATION
ftfr
Dependent on: Vapor press
1-
sure
Henry's Law constant
CHEMICAL DEGRADATION
Hydrolysis
UV Photolysis
nrvr t nnr
9/93
15
Hydrogeochemistry
-------
NOTES
BIODEGRADATION
Tetrachloroethylene •
-^. Trichloroethylene
•^ Vinyl chloride
>H2OandC02
• Microbe
Eh
Redox potential
SOIL ZONE REACTIONS 78° ™
(02>0.1 ppm)
sulfide oxidation HS "-^ SO2"
iron oxidation Fe2+-> Fe 3+
nitrification NH4-^N03
manganese oxidation Mn -^ Mn*+ yj
iron sulfide oxidation Fe2S^Fe 3++ SO 4
aerobic respiration CHjO +O2 -^COj + h^O
k
SATURATED ZONE REACTIONS °
denitrification NO3-^ N2(ga8)
manganese reduction Mn4+-^ Mn2*
iron III reduction Fe3i-^ Fe2+
sulfate reduction S024-^HS"+H2S
methane fermentation CH20-^ CH4(a,0) „.„
•£.(}(} mv
Hydrogeochemistry
16
9/93
-------
REDUCTION OF MANGANESE (Mn)
AND IRON (Fe)
3 Mn02 -I- 18 H+ + 6 OH~= 3 Mn+++ 12 H20
8 H++ 2 Fe O = 4 Fe+++ 4 KO + 0,
23 t c
REDUCTION OF SULFATE
HS + 4H20 = S0^+ 9H + 8e
ADSORPTION
Partitioning of elements
Cation exchange capacity
NOTES
9/93
17
Hydrogeochemistry
-------
NOTES
PARTITIONING OF ELEMENTS
<^~ I just love anions ~^)
Well, so do I HI
ADSORPTION OR
DISTRIBUTION COEFFICIENT
K
ORGANIC CARBON FRACTION
(foe)
The fraction of the aquifer solid
material that is organic carbon
Hydrogeochemistry
18
9/93
-------
ORGANIC CARBON
PARTITION COEFFICIENT (Koc)
The distribution coefficient for the
organic solute between water and
natural solid organic matter
K
d
oc ^ 'oc
logKd = logKoc+logf(
oc
K
d
T
om
log Kd = log Kom + log fom
NOTES
9/93
19
Hydrogeochemistry
-------
NOTES
Koc = 1.72 x K
om
OCTANOL-WATER
PARTITION COEFFICIENT (Kow)
The adsorption of the nonpolar organic
molecules to the solid organic material in the
formation
OCTANOL
+
WATER
OCTANOL
SOLUTE
SOLUTE
SOLUTE
RATIO OF THE AMOUNT OF SOLUTE THAT PARTITIONS
OUT OF THE AQUEOUS PHASE ONTO THE SOLID ORGANIC MATTER
for Toluene
CH_
log Koc = 0.72 log Kow + 0.49
Hydrogeochemistry
20
9/93
-------
= Kocfoc
log K d = 0.72 log Kow + log foc + 0.49
Kd = distribution coefficient
KOC = organic carbon partition coefficient
KQW = octanol-water partition coefficient
foc = organic carbon fraction
Kd for Toluene
log Kd = log Koc + log f
'oc
log Kd = 0.72 log Kow+ log foc + 0.49
Koc for Benzene
log KOC = - °-54 '°gs + °-44
NOTES
9/93
21
Hydrogeochemistry
-------
NOTES
Kd for Benzene
log Kd = log Koc + log foc
log Kd = - 0.54 log S + log f oc + 0.44
RETARDATION FACTOR
Rd = 1 +
(Kd)(pb)
n
pb
n =
= Retardation factor (unitless)
= Distribution coefficient (ml/g)
Kd = KQC foe
= Bulk density (g/cc)
Porosity (decimal fraction)
RETARDATION
R = 1 +Pb x Kd
n
R = Retardation factor
ft, = Bulk density
Kd = Distribution coefficient = K Oc fa
n = Porosity
Contaminant Velocity:
'••ft
vx = Contaminant velocity
v = Groundwater flow velocity
Rx = Retardation factor for contaminant x
Hydrogeochemistry
22
9/93
-------
NOTES
DENSITY STRATIFICATION
Dependent on:
Specific gravity ^«
Solubility fft
S.G. < 1
SOLUBLE
S.G. > 1
PLUME STRATIFICATION
Unconfined aquifer |g| Groundwater flow |;£f.: •.:
'//, Bedrock
Y//////.
9/93
23
Hydrogeochemistry
-------
Migration of Chlorophenolic Compounds
at the Chemical Waste Disposal Site
at Alkali Lake, Oregon —
1. Site Description and Ground-Water Flow
by James f. Pankow2, Richard L. Johnson*, James E. Houckb,
Susan M. Brillante1, and W. Jerry Bryanb
ABSTRACT
The hydrogeology of the chemical wiste disposal site
in the closed basin at Alkali Lake, Oregon has been
examined. Interest in the sice is due to the burial
(November 1976) of 25,000 drums of herbicide manu-
facturing residues in unlincd trenches on the playa of the
basin. Included in the wastes were large amounts of chloro-
phenols and polymeric chlorophenoxyphenols. The flow of
the alkaline (pH =» 10) ground water in the site area is
driven by: (1) springs which create a mound cast of the site;
ar.d (2) the sump effect of "West Alkali Lake," a topo-
graphic low to the west of the site. Porosity, bulk mass
densities, and grain-size distributions were determined. At
one piezometer, the depth to ground water ranged between
0.9 m and 2.2 m. With the bottoms of the trenches in
which the chemicals were buried between 0.60 and 0.75 m
below the level of the ground surface, the bottom portions
of the trenches may, at least occasionally, be in direct
contact with the ground water.
INTRODUCTION
The by-products of a wide variety of chemical
processes are often disposed of together—nonuni-
formly, and noninstantaneously in one chemical
disposal site. Such disposal usually results in
complicated ground-water contaminant plumes
which are difficult to model. The disposal of
chemical waste in the Alkali Lake Basin (Lake
County, Oregon, Figure 1) does not follow this
typical model. Wastes there "were received over a
fairly narrow time frame. In addition, nearly all of
the wastes were from one chemical manufacturing
operation (the production of chlorophenoxy
herbicides). As a result, most of the individual
compounds making up the waste were chemically
similar (chlorophenolic), differing primarily by
"Oregon Graduate Center, 19600 N.W. Walker Road,
Bcavcrton. Oregon 97006.
bNEA, Inc.. 10950 S.W. 5th St.. Ste. 380, Bcavcrton,
Oregon 97005.
Received December 1983, revised July 1984,
accepted July 1984.
Discussion open until March 1, 1985.
Vol. 22, No. 5-GROUN'D WATER-Scptcmber-October 1984
molecular weight. Thus, while the value of the
partitioning coefficient of the contaminants
between the soil matrix and the ground water (Kj)
may vary between the compounds, the physico-
chemical processes which control their values in
the soil and ground-water media will be similar.
Early interest in the Alkali Lake area was
oriented towards the mining of soda (NajCO3).
Claims were first filed by an Oregon firm in the
late 1800s. These claims chinged ownership several
times. They were purchased by Chem Waste, Inc.
(Portland, OR) in 1967 for the purpose of
establishing a 4 ha waste chemical storage site. The
location for the site was selected such that it was
just inside the mining claim boundary. Areas
further inside the playa would have been subject to
greater amounts of playa water as well as greater
hauling distances. The site was licensed by the
Oregon Department of Agriculture (ODA) in 1968
for pesticide waste storage. Little or no prior study
was carried out to determine the suitability of the
site to receive chlorophenolic wastes. Storage of
wastes at the site began in 1969. The feasibility of
using shallow land application to degrade the
wastes was investigated in 1970 on several plots of
land near the site (Goulding, 1973). The largest of
the plots was 4 ha in area (Figure 1). The possibil-
ity of using the waste material as a rangeland-
improving herbicide was also investigated (Figure
1).
By late 1971, a total of twenty-five thousand
206 1 (55 gallon) drums of manufacturing wastes
from the production of 2,4-D (2,4-dichloro-
phenoxyacetic acid), and MCPA (4-methyl-2-
chlorophenoxyacetic acid) had been stockpiled on
pallets at the site. The wastes represented primarily
the distillation residues ("still-bottoms") which
resulted during the separation of desired chloro-
phenols from a phenol chlorination process
mixture. Included in the still-bottoms were various
chlorophenols and a large variety of polymeric
593
Reprinted by permission of the Ground Water Publishing Company. Copyright 1984.
All rights reserved.
-------
Fig. 1. Topographic map of Alkali Like pl*y« ind
surrounding arei. Major contours ire it 200 foot intervals
with supplemental contours at 100 foot intervili. (Prepired
on the basis of maps obtained from the Defense Mapping
Agency Topographic Center, Washington, O.C.)
chlorophcnoxyphcnols (CPP) (Pankow etal.,
1981). In addition to the still-bottoms, the waste
materials also included the herbicides 2,4-D, and
MCPA. All available evidence indicates that no
2,4,5-T (2,4,5-trichlorophenoxyacctic acid) or
2,4,5-T wastes were present in the disposed
materials. The structures of these various
compounds are presented elsewhere (Johnson
etal.. 1984a).
In 1971, the Oregon Department of Environ-
mental Quality (ODEQ) and ODA stopped addi-
tional wastes from being hauled to the site.
Between 1972 and 1974, ODEQ pursued cleanup
of the site by its owners through State courts, but
ultimately lost. The State of Oregon condemned
the site and took possession in October 1976. In
July 1976, remedial action funds were requested
and subsequently received from the Oregon Legis-
lature. Given the general corrosiveness of the waste
(the chlorophenols, CPP, and phenoxy herbicides
are all acids) (Johnson et al., I984a) by this point
in time, many of the drums had begun to leak. In
October 1976, the U.S. Environmental Protection
Agency (EPA) Alkali Lake Task Force cook
samples from five different barrels and tested them
for solubility and pH when mixed both with the
local ground water and with deionized water. Their
report, dated December 1976. indicates that on the
average, 70% of the material in the drums was
soluble in the highly alkaline (pH & 10), local
ground water (EPA, 1976). The wastc/ground-
water volume or mass ratio used in the solubility
tests was not indicated. When 2 g of each of the
five samples were mixed with 100 ml of deionized
water, pH values of 4.5, 7.0, 5.0, 5.0, and 8.0 were
obtained. The acidic pH values are in the range
expected based on the pKj [-log (acidity con-
stant)) values for chlorophenols. The neutral to
alkaline pH values were probably due to the
presence in some of the barrels of basic residues
from the alkaline coupling of 2,4-dichlorophenol
with chloroacctic acid to give 2,4-D. A contract
was let by ODEQ to crush and bury the drums in
12 shallow (0.60 to 0.75 m deep), unlined trenches
130 m long and 20 m apart (EPA, 1976; ODEQ,
1977a). This operation, carried out in November
1976, converted the storage site into a disposal
site. The major portion of the wastes were there-
fore injected into the ground-water system in a
narrow time period. Although the water table in
the area of the site is very shallow (typically only
1 to 3 m deep), it was hoped that the location of
the site inside of the closed Alkali Lake basin
would limit the movement of the contaminants in
the alkaline ground water.
HYDROLOGY, GEOCHEMISTRY,
AND GEOLOGY
The site is located on the northwestern edge
of the Basin and Range Physiographic Province
(Fenneman, 1931). This area is characterized by a
large grabcn occurring between two dramatic,
north-south-trending fault scarps, Abcrt Rim on
the east (840 m high) and Winter Ridge on the
west (360 m high). As is typical for the Province as
a whole, the graben contains a variety of closed
basins. The Lake Abert and Summer Lake basins
(2,200 and 1,000 km', respectively) have been the
most studied (Donath, 1958; Phillips and Van
Denburgh, 1971; Van Denburgh, 1975). In Pleisto-
cene times, "Lake Chewaucan" occupied a large
portion of the Lake Abert and Summer Lake
basins. A second Pleistocene lake once occupied
the Alkali Lake basin. In terms of that former lake.
the Alkali Lake basin has a drainage area of
750 km7 (Mundorff, 1947). Abert Rim bounds the
Alkali Lake basin on the east. The land to the west
slopes gently upward (400 m gain over 10 km), and
possesses numerous small eolian deflation areas
(diameters up to 2 km).
The site is located on the southwest edge of
594
-------
Fig. 2. Photograph and tuperimpoied topographic map of the lite vicinity. Contours are shown at 5 foot intervals. All
unlabeled contours north of the site are 4265 foot contours. Unlabeled contours east of the site are 4260 contours. (Prepared
with the (distance of U.S. EPA (1983) data.]
the 5 km diameter playa in the Alkali Lake basin
(Figures 1 and 2). The playa serves as a sump for
both surface- and ground-water discharge. Two
flowing wells were drilled in the playa area by
Stott (1952), well 5N1 (45 m deep), and well
18R1 (90 m deep) (Newton and Baggs, 1971). The
regular flow of the former well has created a small
marshy environment. The natural surface- and
ground-water discharges lead to the formation of a
shallow ephemeral lake. Standing water is occasion-
ally within 100 m of the site. As mentioned above,
the water table at the site is usually 1 to 3 m below
the ground surface. As is typical for closed basins,
net evapotranspiration exceeds net precipitation in
the playa area. Ground water flowing into the
playa is fresh [total dissolved solids (TDS) = 200 to
500 mg/1; specific conductance = 100 to 250
-------
Specific Conductance Measurements
Specific conductance measurements were
made using a Model 33 YSI (YSI, Inc.. Yellow
Springs, OH) portable conductivity meter. During
field monitoring, the calibration and linearity of
the meter were checked twice daily with standards
ranging from 600 to 50,000 pmhos. Single-point
calibration corrections were made several times
daily. At the beginning of the field monitoring
program, several standing columns of water were
bailed from the PVC wells prior to sampling. How-
ever, this practice was discontinued later in the
study since the high hydraulic conductivity in the
area of the site causes a flushing of the water over
the slotted interval. Measurements were made in
the PVC wells by lowering the probe down to one
meter below the water surface, and waiting one
minute for the signal to stabilize. In order to
obtain more spatial resolution than would be
provided by the PVC wells, additional points were
sampled using 1.8 to 2.5 m long, 0.64 cm O.D.,
0.46 cm I.D. type 316 stainless steel (SS) tubes
(unslotted, open-bottomed). After placing a 0.45
cm O.D. rod inside each of the tubes, they were
pushed 1.5 to 2.3 m into the ground by hand. The
inner rod was then retracted leaving a clear
sampling tube. A hand vacuum/pressure pump
(Nalge, Inc., Rochester, NY) was attached to the
steel tube via FEP Teflon tubing, and 100 ml were
withdrawn and placed in a 125 ml vial. The con-
ductivity was measured after allowing the
electrodes to equilibrate for one minute.
RESULTS
Bulk Mass Density, Porosity, Grain-Size, and
Mineralogical Determinations
Bulk mass density and porosity values found
for several samples obtained near Well 25 ranged
between 0.85 to 0.95 and 0.60 to 0.70, respective-
ly. The two soil samples subjected to grain-size
distribution analysis were similar (Figure 6). In
both cases, 90% of the material possessed diame-
ters greater than 0.006 mm. Very little material
was found in the operationally-defined "clay" size
range. Examinations with a petrographic micro-
scope revealed that both samples consisted largely
of fine glass (volcanic ash) only partly devitrified,
often entirely vitreous. The index of refraction of
n = 1.51 indicated a silica content near 68% and a
rhyolitic or rhyodacitic composition. An occasion-
al fragment of pumice was observed. In addition,
the samples contained fine particles of calcircand
small amounts of plagioclase feldspar, quartz,
basaltic glass, ortho and clinopyroxenc, and diatom
g | | | fSSSS S ! i 9-
" " G««IN till (,,«!>
Fig. 6. Percent of toil material below a given grain size vi.
grain size for two sample! taken near Well 34.
tests. While the sample obtained at 2.4 m was
found to contain a small quantity of clinoptilolite
(a zeolite), almost no clay minerals were found.
This observation is consistent with the nature of
the size distributions. Atterburg rests applied to
the material passing the number 40 screen gave
liquid limits, plastic limits, and Atterburg Number;
of 70%, 48%, and 22 and 73%, 58%, and 15 for the
samples obtained at 1.2 and 2.4 m, respectively.
These results arc also consistent with the detection
of little clay mineral material, though: (1) visual
inspection of materials obtained in the augering of
some sample wells has suggested the presence of
some clays in portions of the aquifer; and (2) Jones
and Weir (1983) have found authigenic clay
materials in nearby Lake Abert.
Ground-Water Flow System
The locations of the PVC wells are shown in
Figure 5. The water table maps (Figure 7) were
drawn using data from that well series. Since some
of the wells were installed at separate times, each
of the maps shows the location of the sampling
points which provided the database for that map.
Figure 8 presents water level data as a function of
time. All data have been plotted relative to a
datum level of 1,000 cm as the height of the casing
on Well 2. The general direction of ground-water
flow is westward. As mentioned earlier, this direc-
tion results from the relative locations of the
springs, West Alkali Lake, and the ground-water
conduit provided by the partially-filled
topographic low.
For the annual cycles presented in Figures 7
and 8, the highest and lowest water table levels in
the area of the site occur in March to April and in
September to November, respectively. The large
evapotranspiration losses which occur in the basin
598
-------
22 Nov. 1981
II Sipl. 1982
26 Moy 1982
Fig. 7. Water tible maps in centimeters obtained during November 1981, and February, May, and September 1982. The
points at which data wen taken for the preparation of the maps are ihown in each of the figurei. All data are relative to the
datum: top of Well 2 casing - 1,000 cm.
in the summer months [maximal losses at nearby
Lake Abert occur in July to August (Van Denburgh,
1975)) arc no doubt largely responsible for this
cycle. Since the average annual precipitation in
tiie basin is only 17.5 cm, other than promoting
evaporation, the decrease in precipitation during
the summer months probably plays a minimal role.
If summertime losses at West Alkali Lake were
substantially greater than at other areas near the
site, the hydraulic gradient across the site would be
expected to maximize during the summer. This
docs not occur. Rather, as Figure 9 shows, it tends
900
UJ
cr
Lul
800
7OO
WELL 8
77 78 79
82 83 84
80 81
YEAR
Fig. 8. Water level at Welli 2 and 8 ai a function of time
over the period 1977 to 1984.
0.002
r-
•z.
UJ
O
0.001
rr
o
77 78 79
80 81
YEAR
82 83 64
Fig. 9. Hydraulic gradient between Well! 2 and 8 at a
function of time over the period 1977 to 1984.
599
-------
Fig. 10. Specific conductance isoplcths (April 19831,
/jmhos/cm x 10'.
to maximize during February to June and has its
lowest values during September to November. This
may be because of large summertime evapotrans-
piration losses from the playa area east of the site.
While the ground-water mound caused by the
springs to the east of the site remains in place
throughout the year, such losses could cause
sufficient decay of the mound to allow a decrease
in the east to west gradient. The seasonal variation
in the water table would then be due to the inter-
play between the evaporranspiration losses in the
various areas. Variable spring output could also
explain the observed temporal behavior of the
'gradient and the water table, though such cycling is
not common in springs discharging from volcanic
rock aquifers (Todd, 1980). The causes of the
seasonal variation in the hydraulic gradient
continue to be investigated by the authors.
The average value of the annual fluctuation in
the water table height at Well 2 was 0.6 m (Figure
8). The actual average fluctuation must be greater
because discrete and not continuous data were
obtained. This fluctuation, however, docs not
provide an estimate of the cvapotranspiration
losses from the playa area since a substantial
amount of the fluctuation is no doubt due to the
growth and subsidence of the ground-water mound
near the site. The depth to ground water at Well 2
ranged between 0.9 m (April 1982) and 2.2 m
(October 1977). With the bottoms of the trenches
in which the chemicals were buried between 0.60
and 0.75 m below the level of the ground surface
(ODEQ, 1977a), the bottom portions of the
trenches may, at least occasionally, be in direct
contact with the ground water. Since historicaJ
data for nearby Lake Abert indicate a water level
fluctuation of more than 5 m in the period since
the early 1930s (Phillips and Van Denburgh, 1971),
600
there is a possibility that the wastes will be hydrau-
lically lifted out of the trenches and onto the playa
some time in the future. Given the freshness of the
springs to the east of the site and the high salinity
of the near-surface playa ground water, the specific
conductance contours obtained near the site
(Figure 10) confirm that the springs are important
in determining the rate and direction of ground-
water movement. Since the seasonal water table
maps do not show major changes in the direction
of ground-water flow, it is likely that the conduc-
tivity contours are similar throughout the year.
The fact that the low conductivity water flows
directly along the major axis of the site has proven
convenient in plume modeling work to be
described elsewhere.
SUMMARY AND CONCLUSIONS
The fact that the playa surface was initially
selected for chemical storage was due to the
purchasabiliry of the mining patent whose bound-
aries closely followed the playa boundaries. The
subsequent disposal of the waste in trenches at the
same site is unfortunate because: (1) the water
table is close to the ground surface; and (2) there
is a strong local ground-water flow. The direction
of ground-water movement is westward throughout
the year. The conductivity contours confirm the
directional nature of the flow as well as the .
importance of the springs east of the site as a
source of a local ground-water mound. With •
hydraulic gradients across the site, hydraulic
conductivities, and porosities of the order of
2.0 X 10- to 1.2 X 10'J,.0.01 to 0.10 cm/s, and
0.65, respectively, local ground-water velocities of
0.3 to 16.0 cm/day would be inferred if the total
porosity was available for flow. Since Johnson
etai (1984b) have shown that the porosity
available for flow at this site is only 0.01 to 0.05,
the actual ground-water velocities are much higher,
i.e. in the range of 3.9 to 1,040. cm/day. The
manner in which this ground-water flow has
influenced the shape of the compound-dependent
contaminant plume downgradicnt of the site is the
subject of the second paper in this series (Johnson
et a/., 1984a).
ACKNOWLEDGMENTS
We express our appreciation to John A.
Cherry for many helpful discussions. We also
appreciate the permission to work at the Alkali
Lake Chemical Disposal Site granted to us by the
Oregon Department of Environmental Quality. •
This work was financed in part with Federal funds
-------
from the United States Environmental Protection
Agency (U.S. EPA) under Grant Number 808272.
The contents do not necessarily reflect the views
and policies of the U.S. EPA nor docs mention of
trade names or commercial products constitute
indorsement or recommendation for use.
REFERENCES
Donath, F. A. 1958. Basin-Range Structure of South
Central Oregon. Ph.D. Thesis, Stanford University.
Environmental Protection Agency. U.S., Region X, Seattle,
Washington. 1976. Report of the Alkali Lake Talk
Force. 14 pp.
Environmental Protection Agency, U.S.. Environmental
Monitoring Systems Laboratory, Las Vegas, Nevada.
1983. Topographic Map of Alkali Lake, Oregon.
Project No. AMD 83060 JO 44.10.
Fenneman, N. M. 1931. Physiography of the Western
United Stttci. McGriw-Hill, New York, N.Y. 534 pp.
Coulding. R. L. 1973. The Alkali Lake Project: Soil
Biodegradation of Pesticide Manufacturing Work,
Lake County, Oregon. Report. Environmental Health
Sciences Center, Oregon State University, Corvallis,
Oregon 97331.
J«hn»n, R. L. and J. F. Pankow. 1984. Unpublished work.
Oregon Graduate Center, Beaverton, Oregon 97006.
Johnson, R. 1_, S. M. Brillantc, J. F. Pankow, J. E. Houck,
and L. M. Isabcllc. 1984a. Migration of chloro-
phenolic compounds at the chemical waste
disposal site at Alkali Lake, Oregon. 2. Contaminant
distributions. Jn press, Ground Water.
Johnson, R. L., R. T. DeCeiar, J. F. Pankow, and J. A.
Cherry. 1984b. Push-pull tests in the characterimion
of ground water flow in fractured media. In
preparation.
Jones, B. F. and A. H. Weir. 1983. Clay minerals of Lake
Abcn, an alkaline, saline lake. Clays and Gay
Minerals, v. 31. pp. 161-172.
Mundorff, N. L. 1947. The Geology of Alkali Lake Basin,
Oregon. Master's Thesis, Oregon State University.
National Oceanic and Atmospheric Administration. 1984.
Clirtutological Data, Oregon, May 1961-Presem.
Environmental Data Service, Ashville, Maryland.
Newton, V. C. Jr., and D. Baggs. 1971. Geologic Evalua-
tion of the Alkali Lake Disposal Site. State of Oregon
Department of Geology and Mineral Industries, Open
File Report. July 1.
Oregon Department of Environmental Quality. 1977a.
Alkali Lake Disposal Project Monitoring Report
No. I.June 14, 1977.8pp.
Oregon Department of Environmental Quality. 1977b.
Alkali Lake Disposal Project Monitoring Report
No. 2. November 17, 1977. 5 pp.
Oregon Department of Environmental Quality. 1978. Alkali
Lake Disposal Project Monitoring Report No. 3.
December 22, 1978. 8 pp.
Oregon Department of Environmental Quality. 1979. Alkali
Lake Disposal Project Monitoring Report No. 4.
October 1. 1979.6pp.
Oregon Department of Environmental Quality. 1981. Alkali
Lake Disposal Project Monitoring Report No. 5.
January 5, 1981. 7 pp.
Oregon Department of Environmental Quality. 1982. Alkali
Lake Disposal Project Monitoring Report No. 6.
February 12. 1982.9 pp.
Pankow, J. F.. L. M. Isabelle. and D. F. Barofsky. 1981.
The identification of chlorophenoxyphenols in soil
and water samples by solvent extraction and field
desorption mass spectrometry. Anal. Chim. Acta.
v. 124, pp. 357-364.
Phillips, K. N. and A. S. Van Denburgh. 1971. Hydrology
and Geochemistry of Aben, Summer, and Goose
Lakes, and Other Cloud-Bum Lakes in South-Central
Oregon. U.S. Geological Survey Professional Paper
502-B. U.S. Govt. Printing Office, Washington, D.C.
91 pp.
Stott, W. J. 1952. Investigation of Saline Deposits in
Southern Oregon. Bonneville Power Administration
Study Contract No. 1BP-7748 to University of
Portland. 60 pp.
Todd, D. K. 1980. GroundwaterHydrology. Second
Edition. Wiley and Sons, N.Y. 535 pp.
Van Denburgh, A. S. 1975. Solute Balance at Abert and
Summer Lakes, South-Central Oregon. U.S.
Geological Survey Professional Paper 502-C. U.S.
Govt. Printing Office, Washington, D.C. 22 pp.
Walker, G. W. and C. A. Rcpenning. 1965. Geological Map
of the Adel Quadrangle. U.S. Geological Survey Map
1-446. U.S. Govt. Printing Office, Washington, D.C.
James F. Pankow is an Associate Professor in the
Department of Chemical, Biological, and Environmental
Sciences at the Oregon Graduate Center in Beaverton,
Oregon. In 1973 be received bis B^. degree in Chemistry
from the State University of New York at Bingbamton. He
received a Ph.D. in 1979 in Environmental Chemistry from
the Department of Environmental Engineering Science at
the California Institute of Technology. His research
interests include the transport, fate, and analysis of organic
chemicals in the environment.
Richard L. Johnson is a Graduate Student Research
Assistant in the Department of Chemical, Biological, and
Environmental Sciences at the Oregon Graduate Center in
Beavenon, Oregon. In 1973 be received bis B-S. degree in
Chemistry from the University of Washington. His research
interests include the transport, fate, and modeling of
contaminants in ground water.
James E. Houck is a Senior Scientist at NEA, Inc., in
Portland, Oregon. In 1971 be received bis BS. in Chemistry
from the University of Arizona. He received bis Ph.D. in
Chemical Oceanography from the University of Hawaii in
1978. His research interests include the analysis, monitor-
ing, end modeling of contaminants in the air and water
environments.
Susan M. Brillante is a graduate student research
assistant in the Department of Chemical. Biological, and
Environmental Sciences at the Oregon Graduate Center in
Beaverton, Oregon. She received her B./. in Chemistry
from Loretto Heights College in Denver. Colorado, in 1966.
Her research interests include analytical organic chemistry
and the fate and modeling of contaminants in air fnd vtaier
systems.
W. Jerry Bryan is a field technician for NEA, Inc.,
and the Oregon Graduate Center. He received bis B.S. in
Geology from Southern Oregon Slate College in 1982.
601
-------
Migration of Chlorophenolic Compounds
at the Chemical Waste Disposal Site
at Alkali Lake, Oregon — 2. Contaminant
Distributions, Transport, and Retardation
by Richard L. Johnson', Susan M. Brillante1, Lome M. Ijabelle4,
James E. Houckb, and James F. Pankow*
ABSTRACT
The behaviors of five chlorophcnols and three chloro-
phenoxyphenols (CPPi) hive been investigated it the
chemical wucc dispotil sice at Alkali Lake, Oregon. All of
the compound! demonstrated similar trends in areal distri-
bution hydraulically downgradieni from the site. The
transport distances for the di- and trichlorophenols were
influenced greatly by their ionization in the high pH (~10)
ground water. In batch equilibrium experiments, these
compounds were found to have Kp values of ~0.0 for the
soil and ground water taken from the site. While also large-
ly ionized at pH °" 10, a tetrachlorophcnol, pcntachlorophc-
nol, and the CPPi demonstrated substantial sorption in the
batch equilibrium experiments as well as retardation
relative to the di- and trichlorophenols at the lite. The
retardations observed relative to 2.6-dichlorophcnol were
las than predicted bated on the batch equilibrium results.
Possible reasons include cosolvcnt effects due to the plume
itself, nonuniform contaminant distributions at the time of
the original burial, the fractures which arc present in the
aquifer, and a decreasing ground-water velocity with
distance westward of the site. Evidence is presented to
support the last reason. These results show, for the first
time, well-behaved concentration contours embodying
compound-dependent retardation in the transport of
sorbing and nonsorbing organic compounds from an
existing waste disposal site.
'Department of Chemical, Biological, and Environ-
mental Sciences, Oregon Graduate Center, 19600 N.W. Von
Neumann Dr., Beavenon, Oregon 97006.
bNEA, Inc., 10950 S.W. 5th St.. Ste. 380, Beavcrton,
Oregon 97005.
Received December 1983, revised November 1984.
accepted January 1985.
Discus'sion open until March 1, 1986.
652
INTRODUCTION
The widespread contamination of ground
water by organic compounds requires that the
processes which control their transport in ground-
water systems be understood. Sorption has been
recognized as playing a fundamental role in retard-
ing the movement of organic compounds (Freeze
and Cherry, 1979; Roberts et al. 1982). When
studying the retardation of compounds in contami-
nated ground-water systems, one must also
consider the possibility of confounding effects due
to: (1) the presence of large concentrations of
contamination-related organic compounds which
may lead to decreased retardation through cosolva-
tion; (2) spatially nonuniform and compound-
dependent contamination; (3) time-dependent
source strength (s); (4) the presence of fractures in
the geological medium comprising the zone of con-
tamination; and/or (5) changes in the ground-water
velocity or irregularities in the hydraulic conduc-
tivity (Kh) with distance downgradient of the
source. All of these processes and conditions must
be considered in the study of migration of contam-
inants at the Alkali Lake, Oregon chemical waste
disposal site (Pankow et al., 1984).
The equilibrium partition coefficient
[Kp, g sorbed solute/g soil (dry-weight basis)] of a
compound is commonly determined by means of
batch equilibrium experiments. Both the role of
concentration in sorption (O'Connor et al., 1980;
Karickhoff, 1981, 1983, 1984;Schwarzenbach and
Westall, l981;Chiou«o/., 1983) and the kinetics
of the sorption process (Karickhoff, 1983) may be
studied by this technique. The value of Kp depends
Vol. Z3, No. 5-CROUND WATER-Scptembcr-October 1985
Reprinted by permission of the Ground Water Publishing Company. Copyright 1985.
All rights reserved.
-------
upon the nature of the soil, the solute, and to a
lesser extent the nature of the aqueous phase. The
soil organic carbon (SOC) content of the soil is
generally the most important factor influencing
sorption. It is usually denoted foc where
' g of SOC/g of soil
(1)
When the SOC content is on the order of
0.1% or greater (foc > 0.001), it has been possible,
especially for nonionizing compounds, to predict
the degree of sorption of a given solute for a
variety of soils based on the foc of the soils. Under
such conditions, Kp may be estimated as
Kp = Kocfoc (2)
where Koc (g sorbed solute/g organic carbon) is the
organic carbon distribution coefficient (Karickhoff,
1981). If a compound can ionize by protonation or
deprotonation, the literature Koc value for that
compound is usually reported for the neutral form
of the compound. For a compound that can become
negatively charged (e.g., a phenol), the negative
form will generally not sorb to the neutral to
negatively-charged soil organic matter as well as
will the neutral form. In such a case, pH-dependent
ionization can lead to a much reduced Kp value.
In practice, most modeling of the transport of
sorbing organics has employed the Freundlich
isotherm S = KpCn where S = (g sorbed solute)/
(g soil); C = (g dissolved solute)/(cmj pore water);
and n is a constant usually less than 1.0. Linearity
of the isotherm is often assumed (n = 1).
For flow through homogeneous porous media
with linear sorption, the degree of retardation
relative to the ground-water velocity is given by
R =
(3)
where R = retardation factor (dimensionless);
Pb = soil bulk density (g/cm3); and 6 is the soil
porosity. For a nonsorbed compound, Kp = 0, and
the migration velocity equals the actual physical
ground-water velocity. For a sorbed compound,
the migration velocity will be less (by a factor of
R) than the physical ground-water velocity. R will
be given correctly by equation (3) only when the
assumptions of equilibrium and linear sorption are
satisfied. When a relative retardation factor Rr is
computed as the migration velocity of a sorbed
compound relative to that for a nonsorbed com-
pound, under these conditions, Rr will equal R as
given by equation (3).
When fractures are present in the medium of
interest, solutes will diffuse into zones of the
aquifer matrix where the water is immobile and
there is no advective flow. This process has been
much discussed recently as a mechanism for
retardation. These zones can be either porous
media between fractures (Freeze and Cherry, 1979;
Grisak and Pickcns, 1980, 1981; Sudicky and
Frind, 1982), or simply small volumes of aquifer of
low hydraulic conductivity (van Genuchtcn, 1974).
In such systems, even compounds with Kp values
of 0 will move more slowly than the physical
velocity of the ground water in the mobile regions
since even nonsorbed compounds will diffuse into
the immobile regions. A principal difference
between the effect of "matrix" diffusion and that
of sorption is that the former is less compound-
specific. That is, while the matrix diffusion coeffi-
cients for different species in a given matrix
material can vary over a single order of magnitude,
Kp values can vary over many orders of magnitude.
The mechanism of retardation in a mobile/
immobile system rests in the fact that all com-
pounds spend a fraction of their time in the
immobile zone. If the mobile water moves suffi-
ciently slowly that sorption and physical partition-
ing equilibrium is reached rapidly between the
mobile and immobile water, then the retardation
relative to the physical velocity of the mobile
water will be given by (Johnson, 1984)
(l+Kppb/fl)(H-0imB/flmfb)
(4)
where 6 is the overall porosity of the system,
6mf is the porosity of the region in which the
mobile water is flowing, 8\m is the porosity of the
immobile region, B is the average immobile region
half width, and b is the average mobile region half-
width. As noted above, even when Kp is zero, R is
greater than 1.0. In fact, although Kp for water
itself is 0.0, the velocity of tagged (e.g., tritiated)
water relative to the mobile-water velocity also
would be retarded relative to the physical velocity
of the mobile water. This is the definition of
retardation that is usually implicit in discussions of
mobile/immobile-water systems (Feenstra et al.,
1984). Unlike the porous medium case then, in
mobile/immobile systems, the actual time-averaged
velocity of a specific water molecule is less than
the physical velocity of water in the mobile
regions.
Since the migration velocity of a nonsorbed
compound in a mobile/immobile system is an
obvious reference for considering the behavior of
other compounds, it is often of interest to com-
pute the retardation of a sorbed compound relative
to that of 2 nonsorbed compound. When equilib-
rium is established quickly between the mobile and
653
-------
immobile water, that Rr value will be given by
equation (3): the factor involving the specific
mobile and immobile parameters in equation (4)
cancels out. In this case, the system acts like one
in which ground water is moving at the velocity it
would have if all of the porosity was open and
available for flow. This is the equivalent porous
medium (EPM) case.
If mass transport limitations prevent the rapid
establishment of an equilibrium partitioning (both
physical and sorptive) between the mobile and the
immobile regions, then: (1) the velocity of both
nonsorbed and sorbed compounds will be closer to
the velocity of the mobile water; and (2) since the
velocity of a sorbed compound will be closer to
that for a nonsorbed compound, the Rr as com-
puted by the ratio of the former to the latter will
be less than R as given by equation (3). (This mass
transport kinetics effect is phenomenologically
distinct from slow sorption/dcsorption chemical
kinetics.)
The Rr concept is useful in the study of field
data. When carrying out controlled laboratory
tracer experiments in simulated soil media, well-
defined physical boundaries and tracer source
functions are employed, and retardation factors are
conceptually well-defined. At a contaminated field
site, however, such well-defined features are
usually absent. As a result, it may well become
necessary to measure: (1) the transport distances
of the breakthrough fronts of various contaminants
relative to some hopefully uniformly meaningful
physical boundary; and (2) the retardation of
various contaminants, relative to the front of one
of the contaminants which hopefully may be
assumed to be nonsorbing and conservative. The
mcaningfulncss of an Rr factor produced by this
mechanism will suffer from inaccuracies if:
(1) the various contaminants of interest moved
relative to one another before their arrival at the
selected transport reference boundary; (2) the
various contaminants were deposited in the source
area at different times and/or in different, non-
uniform distributions; (3) the subjective assign-
ment of the location of the various breakthrough
fronts cannot be done in a meaningful manner;
(4) a spatial variability of the ground-water
velocity is (has been) present; and/or (5) relatively
large amounts of dissolved organic compounds
which affect sorption are (have been) present in
the ground water.
Depending upon their exact nature, the first
four conditions could lead to either a decrease or
increase in the Rr values. As an example of the
fourth condition, if the ground-water velocity
decreases away from the source, the relatively fast-
moving nonsorbing compounds will experience the
decreasing velocity first while the more retarded
compounds will remain longer in the region of
higher velocity: the Rr values will be reduced. The
fifth condition will generally lead to a reduction in
the measured Rr values due to either: (a) an
increased water solubility of the sorbing com-
pounds; and/or (b) an overloading of the sorbing
SOC. For mobile/immobile-water systems which
are complicated by any of the above five condi-
tions, if Rr values less than those predicted by
equation (3) arc observed, it may be difficult to
ascertain the extent to which disequilibrium
between the two types of regions is responsible.
To date there have been very few field-scale
studies of the transport and retardation of organic
compounds in undisturbed ground-water systems.
The water-table aquifer system at the chemical
disposal site at Alkali Lake, Oregon (Pankow et at.,
1984) provides a good opportunity to study some
of the factors controlling these phenomena.
Chlorophenolic organic compounds with Kp
values ranging from —0.0 to greater than 50 arc
present in the waste materials. Some compounds
have migrated over 600 m in 8 years. Field
examination of the aquifer materials has indicated
the presence of a large number of fractures (and/or
bedding plane openings) spaced from mm to cm
apart (Johnson, 1984). We believe that this system
comprises a well-defined mobile/immobile-water
system.
In this paper, the Rr values of the sorbing
compounds calculated relative to a nonsorbing
compound have been compared to Rr values calcu-
lated using equation (3) together with Kp results
from batch equilibrium experiments. The differ-
ences between the retardation factors obtained by
these two methods are discussed with regard to
various physical and chemical processes. The role
of ground-water movement through regions of
higher mobility at this site is examined.
CHEMICAL WASTES AT ALKALI LAKE
During the period 1960 to 1970, considerable
quantities of the herbicide 2,4-D (2,4-dichloro-
phenoxyacctic acid) were manufactured for use as
a defoliant in Vietnam (Young et al., 1978). Large
amounts of 2,4-D by-product wastes were disposed
of at Alkali Lake (Pankow et al., 1984). The 2,4-D
was produced by the ethcrification of 2,4-dichloro-
phenol (2,4-DCP) and chloroacetic acid. Much of
the 2,4-DCP used in this process was manufactured
654
-------
3JC - Q
Fig. 1. Products and wait* by-producu in the lynthesit of
2,4-D (where n and p - 1 to 5, m - 1 to 4, end x > 1).
by the direct Cl, chlorination of phenol (Figure 1).
This step led to the production of a majority of the
unwanted by-products. These included 2,6-di-
chlorophenol (2,6-DCP); 2,4,6-trichlorophcnol
(2,4,6-TCP); 2,3,4,6-tetrachlorophenol (TeCP);
pentachlorophenol (PCP) (Figure 2); and a host of
polymerization products known as the chloro-
phenoxyphenols (CPPs) (Figure 1). The CPPs were
formed by the linear addition of chlorophenol
units. This process led to CPP dimcrs, trimers,
tetramers, pcntamers, etc. Since the coupling
process can occur at any one of the chlorine
positions, and since a variety of different chloro-
2,4 -OCR
(2. * -dichlorophtnol)
2.6-DCP
(2.6-dieMoropHtiiol)
2.4,6-TCP
<2.«.6-lMCl»lorophii>ol)
TeCP
12.3.1.6-ltuocho
Fig. 2. Structurei of chlorophenoh.
phenols were present in the phenol chlorination
process mixture, a wide variety of CPP compounds
were formed.
After the Cl, chlorination step, the desired
2,4-dichlorophcnol was separated from the by-
products by distillation. The "still-bottom"
mixture has been found to include oligomers with
up to at least five rings, and a total of from two
to seven chlorine atoms (Pankow et al., 1981). A
large number of structural isomers (compounds
with the same overall formula, but different
specific structures) will occur in the CPP
compound class.
The various waste compounds display a wide
range of water solubilities. Many of the chloro-
phenoh arc quite acidic, in pan because of the
presence of chlorine atoms on the phenol ring. The
ground water flowing beneath the disposal site at
Alkali Lake is very alkaline (pH =• 10). As a result,
most of the chlorophenols exist primarily as anions
and arc therefore quite soluble. The CPPs, while
also rather acidic, contain a hydrophobic phenoxy
group substituent. They are therefore less soluble,
and show greater sorption to the soil. This paper
will consider the distributions of eight different
chlorophenolics (five chlorophenols and three
CPPs) at the Alkali Lake site.
EXPERIMENTAL
Sampling
Sample bottles (40 or 125 ml capacity) were
washed with warm water and laboratory detergent,
rinsed with dcionized water, dried, and then rinsed
with- methylene chloride. Samples were collected
on April 30 and May 1, 1984 from 1.6 cm O.D.
(0.9 cm l.D.) PVC tubes. Since the PVC tubes were
slotted to three m below the water table, "samples
obtained with them reflected an approximate con-
centration average over that interval. (Contamina-
tion is limited primarily to the top three m of the
ground water.) Samples were collected using a
hand-operated vacuum pump (Nalge,-Inc.,
Rochester, NY). Because the chlorophenolics are
primarily in an ionized state at pH 10, their vapor
pressures were very low, and it is unlikely that any
losses occurred during the reduced pressure
sampling. Upon return to the laboratory.and .prior
to analysis, the samples were stored at 4°C.
2,4-Dichlorophenol, 2,6-Dichlorophenol, and
2,4,6-Trichlorophenol Determinations
Sample work-up was carried out using a
special apparatus (Figure 3) by means of the
following steps: (1) 10 ml of sample placed in
655
-------
/
•Stp-pik'g
C.lirW.. '
n
i
\
\
\
^
p — -x.
r
Fig. 3. Apparatus for sample work-up and pauage through
"Scp-p*k" C-18 cartridge.
apparatus; (2) stirrcr activated, pH electrode
inserted, and sample acidified to pH 2-3 with
6 N HCl to protonatc all organic acid analytes;
(3) electrode rinsed with 2 ml of organic-free
water-, then'removed; (4) vessel capped; (5) Waters
Assoc. (Milford, MA) C-18 Sep-Pak cartridge
placed on vessel, and vessel pressurized to 10 psi
with nirrogcn gas to obtain 5 ml/min flow rate
until the sample was exhausted; (6) Sep-Pak
removed, then aspirated for 30 s to remove residual
water; (7) organic compounds elutcd with 2 ml of
methylene chloride, the first drop (residual water)
discarded; (8) volume reduced to 1.0 ml using a
micro Kuderna-Danish/Snyder column apparatui
and a 95°C water bath; (9) 5 p\ of an external
standard (ES) solution in methylene chloride
added containing 10 pg/pl of meta-chloropheno!,
and 10 pi of an ES solution in methylene chloride
containing 100 ng/pl each of chryscne and fluor-
anthcne; (10) 50 mg of anhydrous sodium sulfate
added for desiccation; (11) concentrated extract
transferred to a precleaned 3.5 ml amber glass
minivial (Pierce Chemical, Rockford, IL);
(12) sample stored at 4°C; and (13) extract
analyzed by gas chromatography (GC). Standard
chemicals were obtained from Chem Service (West
Chester, PA).
The use of the term "external standard" (ES)
here is as described by others (Research Triangle
Institute, 1983). An ES is a standard compound
which is added to the sample extract just prior to
the analytical determination step: it is "external"
to the work-up. The ES allows compensations to
be made for (1) changes in instrument response
between calibrations, and (2) variations in volume
of sample injected into the analytical instrument.
An "internal standard" (IS) is added to the sample
prior to the work-up to monitor the efficiency of
the extraction process: it is "internal" to the
work-up.
The analyses took place without derivitization
using fused silica capillary gas chromatography
(GC) with helium carrier gas on a Hewlett-Packard
(Palo Alto, CA) 5880A capillary GC equipped with
a flame ionization detector (FID). [Samples were
also run on a Finnigan (Sunnyvale, CA) 4000
GC/mass spectrometer/data system (GC/MS/DS)
to ensure that no misidcntifications or coelution
problems would go unnoticed.) The column used
was a 30 m, 0.25 mm I.D., 0.25 pm film thickness
DB-5 (SE-54) fused silica capillary column (J&W
Scientific, Rancho Cordova, CA). The carrier gas
linear velocity used was 20 cm/s (at 175°C). With
the injector at 275°C and the FID at 310'C,
1.0 pi volumes were injected splidess at 45°C. The
GC temperature program used was: hold at 45*C
for two min; program to 175°C at 10°C/min; then
hold at 175"Cfor 5 min.
A three-point quadratic calibration curve
(based on peak areas) was developed for the
GC/FID analyses for each compound using the
HP 5880A system software. Samples were run
following calibration. For quality assurance (QA),
the calibration was randomly verified by running
an intermediate concentration standard. If the
response varied by more than ±10% from the
known concentration, the calibration was repeated.
In addition, for every set of ten samples, a
duplicate, a spiked (recovery) sample, and a blank
were also analyzed. The latter was prepared by
processing 10 ml of deionized water through the
concentration procedure.
2,3,4,6-Tetrachlorophenol (TeCP), Pentachloro-
phenol PCP), and Chlorophenoxyphenol (CPP)
Determinations
The sample extracts described above for the
di- and trichlorophenol determinations were also
analyzed for TeCP, PCP, and for three CPPs
denoted CL2D2, CL3D3, and CL4D2. These ire all
656
-------
chlorophcnoxyphenol dimcrs, with two, three, and
four chlorines, respectively. Specific structures for
these compounds could not be obtained because
reference compounds arc not currently available.
The compounds fluoranthcnc and chryscne,
added during the sample work-up, served as the ES
compounds. Analyses were carried out by injection
of 1.0 t>\ of the extract using an on-column injector
and the same type of column used in the chloro-
phcnol analyses. The column was mounted in an
HP5790 GC substituted for the Finnigan 9610 GC
on the Finnigan GC/MS/DS (Pankow and Isabelle,
1984). The transfer line, source, and MS manifold
temperatures were maintained at 225, 225, and
100°C, respectively. The helium carrier gas linear
velocity used was 30 cm/s (at ambient tempera-
ture). The on-column injections were carried out
with the oven at 80°C. The GC temperature
program used was: immediate temperature
program at lO'C/min to 320°C, then hold at
320*C for 4 min. The analytes were detected using
multiple ion detection (MID, or "selected ion
monitoring"). The ions monitored for each of the
compounds were (in their order of GC elution):
'TeCP (131, 232. 234); PCP (200, 266, 268);
CL2D2 (184. 252. 254); fluoranthene, ES
compound (101, 202); CL3D3 (225, 254, 288.
290); CL4D2 (146, 322. 324); chrysenc. ES
compound (114, 228); and 5-chloro-2-{2,4-di-
chlorophenoxy)-phenol (Irgasan DP300) (146,
288, 290). For QA. every tenth sample was
anaJyzcd both in duplicate and spiked with a-
recovery standard. TcCP and Irgasan DP300 were
used as recovery standards. The latter was obtained
from Ciba-Geigy, Basil, Swiucrland.
Bulk Mass Density and Porosity Determinations
Core samples of soil material taken down-
gradient from the site in the saturated zone con-
tained intact angular blocks which appeared
relatively undisturbed by sampling. Measuring
1 -2 cm across, they were representative of the
aquifer material. To determine density, the blocks
were weighed and their volume determined by
immersion in water. They were rewcighed to verify
that no appreciable amount of water had been
absorbed, then dried at 50°C to a constant weight.
The bulk mass density was calculated as the dry
mass to volume ratio. The porosity was calculated
as the weight loss to volume ratio.
Soil Organic Carbon (SOC) Determinations
The soil was obtained and composited from
the d-epth range of 1-3 m from a location 10m east
of WeU 2. The well locations arc given by Pankow
el al. (1984). Composite samples of l.Og were
treated with 5 ml of a 5% stannous chloride,
3 N HC1 solution, then heated for 4 hours at 50°C
under vacuum to volatilize inorganic carbon
(carbonate and bicarbonate) as CO,. Small aliquots
(10 mg) of the dried soil were weighed, then
combusted in an apparatus developed by Johnson
et al. (1985) based on previous work by Johnson
(1981). The combustion was carried out at 600°C
in a 10% 0,-90% He gas mixture. The CO, formed
was catalytically converted to methane, then
measured with an FID. Calibration of the FID
response was carried out using injections of
methane. The stannous chloride was used in the
carbonate volatilization step to prevent premature
oxidation of the SOC (Allison and Moodie, 1965).
Batch Equilibrium Experiments
Batch equilibrium experiments were carried
out using contaminated water and uncontaminatcd
soil from the site. The water used was collected at
Well 38. The soil was the same composite for
which the SOC values were obtained. Three sets of
experiments were carried out, each in triplicate. In
the first, 20 ml of Well 38 water (spiked with the
control compound naphthalene at 12 mg/1) and 2 g
of dry soil were mixed in a 35 ml glass via). In the
second experiment, 2 ml of Well 38 water was
diluted with 18 ml of Well 2 water, spiked with
naphthalene at 12 mg/1, and mixed with 2 g of soil
in a 35 ml vial. The third experiment was identical
to the second, except that each sample was spiked
with an additional 500 mg/1 of 2,4-DCP. The latter
was done to investigate if the presence of high
concentrations of the chlorophenol would decrease
sorption of the other compounds. The samples
were equilibrated by end-over -end rotation at
20 i 3°C for 24 hours (30 inversions/min). Each
sample was then centrifuged for 15 min. An
amount of 12-15 ml of the supernatant was with-
drawn into a 20 ml syringe, then forced through a
glass fiber prcfilter followed by a silver membrane
filter (Sclas Corp., Huntingdon Valley, PA). Ten
ml was then processed as described above for
chlorophenol determinations.
The ground water at the site ranges between
7-12*C. This is ~10°C lower than the temperature
at which the sorption measurements were made.
Because sorption generally increases with
decreasing temperature (Karickhoff, 1984), it is
expected that the sorption of the compounds at
the site will be somewhat greater than predicted
by the sorption measurements.
657
-------
Fig. 4. Distribution of 2,4
-------
Fiy. 6. Distribution of 2.4,6-uichloroph«nol (2,4,6-TCP). Contours are given in units of mg/l.
RESULTS
2,4-DCP, 2,6-DCP. and 2.4,6-TCP Distributions
The distributions of 2,4-DCP, 2,6-DCP, and
2,4,6-TCP in the ground water at Alkali Lake are
given in Figures 4-6. The sampling points for which
data were obtained and used in the drawing of the
contours arc shown on the figures. These three
chlorophcnols display almost identical patterns.
This implies that these chlorophcnols (1) have
similar source functions, (2) have been subject to
the same hydrology, and (3) have experienced
similar degrees of sorption or lack thereof. Based
on the pH of the ground water (=-10), the acidity
of these compounds, and their low molecular
weights, one could predict that sorption would be
similarly unimportant for all three. Because the
sampling network used was fairly extensive and the
analytical precision quite good (relative standard
deviation 5%). it is possible to discern a narrowing
of the contaminant plume ~ 150 m downgradicnt
of the western edge of the site. This narrowing may
be caused by irregularities in the KH of the aquifer
surrounding the zone of contamination. This will
be discussed further below.
The true north-south line which passes
through the northwest -corner of the site was
selected as the uniform transport reference line
(TRL) from which to measure transport distances.
This TRL, marked on the figures, was selected
because: (1) the maximum concentrations of most
of the chlorophcnolics (even the retarded ones
discussed below) arc.near this line; and (2) Rr
values calculated using transport distances
measured from this line will be upper bounds on
Rr values relative to other, if any, more "meaning-
ful" reference lines to the east (i.e., upgradicnt).
For each compound, the specific point
selected from which to measure transport distances
was the point on the TRL at which the concentra-
tion of the compound was at a maximum. In
Figures 4-6, boxes are then used to mark the loca-
tions of the positions (as located on the line of
shallowest rate of descent, i.e., approximate plume
center line for each compound) at which the con-
centration drops to 2% and 25% of the reference
poini value. These arc the cndpoints of the 2% and
25% transport distances which arc listed in Table 1.
The concentrations of the three chlorophcnols
659
-------
Fig. 7. Diitribution of 2,3.4.6-titnchlorophenol (TtCP). Contours tr« givtn in unitt of /jg/l.
Fig. 8. Diitribulion of penlichlorophenol (PCP). Contouri tic given in uniti ol pg/l.
660
-------
Table 1. Diilance Traveled (X) and
Relative Retardation (Rr) Valuei
Com-
pound X(2%) X(2}%) R,<2*)' R,(25K) K,(pred)c
2.4-DCP
2.6-DCP
2.4,6-TCP
TeCP
PCP
CL2D2
CL3D3
CL4D2
'R,(2*)»
bR,(25*)-
£R,(pred)-
400. 250.
420. 270.
400. 270.
330. 210.
260. 40.
230. 70.
280. 150.
.0
.0
.0
.2
.6
.8
.5
210. 40. 2.0
retardation factor relative
2,6-DCP - X2 d.DCpU*)'
retardation factor relative
2.6-DCP • X26.DCP(25*
R(pred)/R2 6.DCP(pred).
1.1
1.0
1.0
1.3
6.8
3.9
1.8
6.8
to
to
/X(25%)
1.0
1.0
1.0
3.5
13.5
34.
20.
40.
drop to 2% of their reference values at distances of
400 to 420 m. The 50% transport distances are not
given in Table 1 since their comparatively low
values are subject to much error, particularly for
the more retarded compounds to be discussed
below.
TeCP and PCP Distributions
Due to the larger relative standard deviations
for the low (»jg/l) level determinations of TeCP and
PCP (14% and 19%, respectively), it was not
possible to draw the concentration contours of
these two compounds (Figures 7 and 8) with the
same degree of detail present in the 2,4-DCP,
2,6-DCP, and 2,4,6-TCP contours. The minimum
detectable concentrations were nevertheless an
order of magnitude below the lowest values
reported. The general patterns of contamination
were similar to those for the other three chloro-
phcnols, though the concentrations of TeCP and
PCP were ~ 100 times lower than those of the
former. Using the same criteria described above,
the concentrations of TeCP and PCP dropped to
2% of their reference values at distances of
approximately 330 and 260 m, respectively (Table
1). These distances are shorter than those for the
other three chlorophenols.
CPP Distributions
The concentration distributions of the CPP
compounds CL2D2, CL3D3, and CL4D2 are
presented in Figures 9-11. The minimum detect-
able concentrations were an order of magnitude
fig. 9. Distribution of CL2O2. Contours ire given in unhi of pg/l.
66}
-------
Fig. 10. Distribution of CL3D3. Contours «f« givtn in units of pg/l.
* •• \
*$&&\ •
.^J^S^V.:.^-;-1..-
y-ji: »^?r»-w;-, 1W.".-..--: •
Fig. 11. Diilribution of C14D2. Conlouri >re given in uniti o( ps/l.
662
-------
below the lowest values reported. The concentra-
tions of these compounds dropped to 2% of their
reference values it distances of approximately 230,
280, and 210 m, respectively (Table 1). These
distances are less than those for most of the
chlorophenols.
Porosity, Density, and Soil Kp and SOC Values
Values for the overall soil porosity (6)of
from 0.60 to 0.70 were obtained. The overall bulk
mass density (pt>) values were in the range of 0.90
to 0.95.
For the concentration ranges studied, the
results from the three batch equilibrium experi-
ments indicated no significant dependencies of any
of the Kp values on the concentration of sorbing
analyte, or on the presence of 500 mg/1 of added
2,4-DCP. The Kp values and the associated
standard deviations presented in Table 2 were
therefore calculated from the average of the results
of the total of nine experiments. The Kp values for
the chlorophcnolics studied here ranged from 0.0
to 28. (Table 2). The control compound naphtha-
lene gave a mean Us value for Kp of 16. i 2.
The measured SOC values for the soil on a dry
weight basis gave a mean ± Is value of 2.4 ± 0.3%.
The number of replicates was 12. These determina-
tions were complicated because of the high levels
of carbonate present in the samples. The good
precision suggests that the carbonate-removal step
included in the SOC analysis procedure was
effective, though the possibility of slightly high
SOC values remains. The inclusion of the control
compound naphthalene in the Kp experiments
permitted an independent measure of the SOC
T*ble 2. Measured Kp Values for Alkali Lake Soil/Water
System (pH=»10, T - 20°C), pK, Valuei. and
Literature Koc Values for pH < pK.
Compound Kf pK, Kc(
Naphthalene
2,4-DCP
2.6-DCP
2.4,6-TCP
TeCP
PCP
CL2D2
CL3D3
CL4D2
16. 12.
0.0 1 0.5
0.0 1 0.5
0.0 1 0.5
1.8 1 1.0
9.51 1.8
24. 1 5.
14. 1 3.
28. 15.
7.8
6.8
6.2
5.4
4.7
NA
NA
NA
870.'
545.b
NA
1.070.b
6.640.b
32.900.b
NA
NA
NA
1 Karickhoff, 1981.
bSchellenbcrg el at., 1984.
NA * not available.
content of the soil. Literature values for the Koc
for naphthalene are available. Karickhoff (1981)
cites a value of 870, and Mabey et al. (1982) cite 2
value of 940. By equation (2), these K0c values
give SOC values of 1.8% and 1.7%, respectively.
(The error in the Kp value/or naphthalene has not
been propagated into these SOC values since the
available Koc values arc in any event averages for. a
number of different soils, and data on their corre-
sponding statistical variability are not readily
available.) The ratio of the average of these two
values to the measured SOC value is 0.73, i.e., near
1.0.
DISCUSSION
All of the chlorophenols and the CPPs demon-
strated similar trends in area) distribution
hydraulically downgradient from the site (Figures
4-11). The direction of transport is consistent with
the previously reported slope of the water table
(Pankow et al.. 1984). The transport distances for
the di- and trichlorophenols were no doubt
influenced greatly by their ionization in the high
pH ground water. Indeed, as seen in Table 2, these
compounds were found to have Kp values of —0.0
for the soil and ground water taken from the site.
These results are in agreement with those of Miller
and Faust (1973) and Schelknbcrg et al. (1984).
Both groups report that while the protonated
forms of these phenols sorb (Koc > 0 for
pH < pK,, see Table 2), their sorption to SOC
decreases to zero when the pH exceeds their pK»
values by at least one unit. The pH —10 ground
water at the site is 2.2 to 3.8 units above the pK,
values for the di- and trichlorophenols {Table 2). It
may therefore be concluded that the similarities of
the di- and trichlorophcnol distributions were
indeed due in large part to the similarly nonsorbing
natures of these three compounds at pH =» 10.
While also largely ionized at pH =• 10, TeCP
and PCP demonstrated substantial sorption (Table
2). These results are also in agreement with results
obtained by Schcllenberg et al. (1984). Similarly,
despite the fact that the CPPs will surely also be
present largely as anions at pH =» 10. they demon-
strated substantial sorption. Therefore, the fact
that TeCP, PCP, and the CPPs demonstrated
retardation relative to the di- and trichlorophenols
may be understood in terms of their nonzero Kp
values.
Of the three nonsorbing di- and trichloro-
phenols, it may be possible to treat 2,4-DCP,
2,6-DCP, and 2,4,6-TCP as approximately con-
servative. To evaluate the effect of sorption on the
663
-------
transport of TeCP, PCP, and the CPP, therefore. Rr
values referenced to 2.6-DCP were: (1) calculated
using the 2% and 25% transport distances (Rr(2%)
and Rr(25%)) of the compound and those of
2,6-DCP; and (2) predicted using equation (3)
and the experimental Kp values (Rr(pred)). The
results are presented in Table 1.
It should be pointed out that the most
meaningful measured Rr values for the presumed
constant source input function would have been
the ones based on the 50% concentration fronts of
the sorbed and nonsorbed compounds. Rr values
based on the 2% or 25% fronts will be deflated
relative to the 50% front due to differing relative
effects of dispersion on the sorbed and nonsorbed
compounds. Such differences will increase with
increasing sorption and/or decreasing defined-
frontal concentration. For TeCP, this effect
could have caused a decrease in the Rr(2%) and
Rr(25%) values of only -9 and 15%, respectively
(Johnson, 1984). For CL4D2, the most sorbed
compound, this effect could have caused a decrease
in the Rr(2%) and Rr(25%) values of -30 and 45%,
respectively (Johnson, 1984).
If the input function was not a constant, but
rather decreased in time due to depletion of the
finite mass of the source contaminants, that would
lead to an inflation of the Rr(2%) and Rr(25%)
values. This is the case because the nonsorbed
compounds will tend to suffer relatively greater
depletion at the source. This in turn requires that
greater transport distances be inferred in order to
reach 2% or 25% of the depleted source concentra-
tion values. Such inflation would tend to counteract
deflation caused by the use of transport distances
for contamination fronts defined at concentration
values less than 50% of the source reference value.
Therefore, it may be concluded that only a
small amount of deflation of the Rr values is
possible. For the sorbing compounds (Kp > 0),
however, the observed R,(2%) and Rr(25%) values
arc in fact smaller than the Rr(pred) values by
substantial amounts (Table 1). Indeed, had the
sorption measurements been carried out at 10°C
rather than 20*C, it may be expected that the
Rr(pred) values in Table 1 would have been even
larger.
The possible reasons remaining for the differ-
ences between the measured and predicted R,
values include: (1) cosolvent effects leading to i
decreased retention of TeCP, PCP, and the CPPs;
(2) nonuniform contaminant distributions at the
time of the original buriil, e.g., a time zero center
of mass for2,6-DCP further to the east than the
time zero centers of masses of TeCP, PCP, and the
CPPs; (3) the fractures in the aquifer; and/or
(4) a decreasing ground-water velocity or irregulari-
ties in Kh with distance westward of the site.
Cosolvent effects could be due to either
actual solvents in the contaminant plume and/or
the high levels of di- and trichlorophcnols. Large
quantities of solvents are not believed to have been
present in the waste, nor have any been detected in
high concentrations. Since the batch equilibrium
experiments demonstrated no changes of the Kp
values with overall phenol concentration nor with
the addition of 500 mg/1 of 2,4-DCP, cosolvent
effects due to the high concentrations of chloro-
phenols in the plume itself also do not appear
important.
Figures 4-11 provide evidence that the various
compounds were not distributed perfectly uni-
formly in the site. It does, however, appear that
the trenches of waste near the selected transport
reference line were the most contaminated in the
compounds of interest to this study since the
concentrations of the nonsorbed and sorbed
compounds alike seem to be highest in that area.
Most are more or less uniform in.concentration
along.the TRL. (PCP is an exception with its locally
very high concentrations in the northwest corner
of the site (Figure 8). It is believed that this
different character of the PCP distribution in the
site artificially inflates its Rr(2%) and Rr(25%)
values. Nevertheless, they are still substantially less
than the Rr(pred) for this compound.) Moreover,
when the data for all of the compounds are
replotted after integrating along transects perpen-
dicular to the direction of flow (a process which
should remove dependence on the transverse
irregularities in the source functions), the Rr(2%)
and Rr(25%) values are still lower than the corre-
sponding Rr(pred) values (Johnson, 1984).
The fact that the ground water moves in
fractures in the soil is also not believed to be
responsible for the reduced R,(2%) and Rr(25%)
values. It has been determined that diffusive
equilibrium between the matrix and the fractures
is approached within approximately two hours
(Johnson, 1984). This short time period implies
that matrix diffusion limitations are probably not
responsible for the reduction of the Rr(2%) and
Rr(25%) values, and that the Alkali Lake system
behaves as an EPM.
Water-level measurements indicate that with
increasing distance westward of the site, the
hydraulic gradient first decreases, then increases,
then decreases again as West Alkali Lake is
664
-------
approached (Johnson, 1 984). The decrease and
subsequent increase in the gradient is most likely
due to a hydraulic conductivity-defined constric-
tion located approximately 150 m west of the
western edge of the site. This constriction is very
likely the cause of the narrowing observed in the
di- and trichlorophenol distributions. It appears
from the contaminant distributions that all of the
compounds have reached the constriction. The
leading edges of the contaminant distributions of
the nonsorbed compounds have largely stagnated
on the downgradient side of the constriction. The
leading edges of the contaminant distributions of
the sorbed compounds, however, are still moving.
The effect will be to reduce the measured Rr(2%)
and Rr(25%) values. This is believed to be the
primary reason why these values are lower than
the Rr(pred) values.
CONCLUSIONS
The arcal distributions of eight chlorinated
phenolics hydraulically downgradient from a
chemical disposal site have been presented. These
results show, for the first time, well-behaved
concentration contours embodying compound-
dependent retardation in the transport of sorbing
and nonsorbing organic compounds from an
existing waste disposal site. The trends in relative
retardations of the compounds arc consistent with
the Kp values determined in batch equilibrium
experiments carried out using samples of the native
soil and ground water (pH =» 10).
While the trend in observed retardations of
the chlorophenolics is correct, the magnitudes of
the relative retardations are less than those
predicted using Kp values determined from the
batch experiments. This is probably the result of
irregularities in the-Kh values downgradient of the
site. Cosolvent effects due to the plume itself,
nonuniform contaminant distributions, and the
fractures in the aquifer arc believed to have played
only a minor role in this regard.
ACKNOWLEDGMENTS
We express our appreciation to John A.
Cherry for many helpful discussions. We also
appreciate the permission to work at the Alkali
Lake Chemical Disposal Site granted to us by the
Oregon Department of Environmental Quality.
This work was financed in part with Federal funds
from the United States Environmental Protection
Agency (U.S. EPA) under Grant Number 808272.
The contents do not necessarily reflect the views
and policies of the U.S. EPA nor does mention of
trade names or commercial products constitute
endorsement or recommendation for use. The
sponsorship of ground-water contamination
research at the Oregon Graduate Center by the
Northwest Environmental Research Center
(NWERC) is also gratefully acknowledged.
REFERENCES
Allison. L. E. jnd C. D. Moodic. 1965. In Methods of Soil
Analysis, C. A. Black (Ed.). American Society of
Agronomy, Inc., Madison, Wl. pp. 1380-1396.
Chiou, C. T., R. E. Porter, and D. W. Schmedding. 1983.
Partition equilibria of nonionic organic compounds
between soil organic matter and water. Environ. Sci.
Technol. v. 17, pp. 227-231.
Fccnstri. S.. J. A. Cherry, E. A. Sudicky, and 2. Haq. 1984.
Matrix diffusion effects on contaminant migration
from an injection well in fractured sandstone. Ground
Water, v. 22, pp. 307-316.
Freeze, R. A. and J. A. Cherry. 1979. Groundwater.
Prentice-Hall, Englewood Cliffs, NJ. 604 pp.
Grisak, G. E. and J. F. Pickcns. 1980. Solute transport
through fractured media: 1. The effect of matrix
diffusion. Water Resources Research, v. 16. pp.
719-730.
Gris»k. G. E. and J. F. Pickens. 1981. Ah analytical
solution for solute transport through fractured media
with matrix diffusion. J. Hydrology, v. 52, pp. 47-57.
Johnson, R. L. 1981. Design and evaluation of a thermal-
optical method for the analysis of carbonaceous
aerosols. M.S. Thesis, Oregon Graduate Center,
Beaverton, OR 97006.
Johnson, R. L. 1984. The groundwatcr transport of chloro-
phenolics in a highly fractured soil at Alkali Lake,
OR. Ph.D. Thesis, Oregon Graduate Center.
Beaverton, OR 97006.
Johnson, R. L.. M. E. Anderson, and j. F. Pankow. 1985.
The determination of soil organic carbon by oxygen
combustion with CO; conversion to methane and
subsequent flame ioniiation detection. Unpublished
work, Oregon Graduate Center, Beavenon, OR 97006.
Karickhoff, S. W. 1981. Semi-empirical estimation of
sorption of hydrophobic pollutants on natural sedi-
ments and soils. Chemosphere. v. JO, pp. 833-846.
Karickhoff, S. W. 1983. Sorption kinetics of hydrophobic
pollutants in natural sediments. Extended Abstracts,
Div. of Environ. Science, 186th National American
Chemical Society Meeting, Washington. D.C.
Karickhoff, S. W. 1984. Organic pollutant sorption in
aquatic systems. J. Hydraulic Engineering, v. 110,
pp. 707-735. .
Mabcy, W. R.. J. H. Smith, R. T. Podoll. H. L. Johnson,
T. Mill. T.-W. Chou. J. Gates. 1. W. Partridge.
H. Jiber, and D. Vandenbcrg. 1982. Aquatic fate
process data for organic priority pollutants. EPA
Report No. 440/4-81-014.
Miller. R. M., and S. D. Faust. 1973. Sorption from solu-
tion by organo-clay: III. The effect of pH on sorption
of various phenols. Env. Letters, v. 4, pp. 211-223.
O'Connor, C. A., P. J. Wierenga, H. H.Cheng, and K.C.
Doxtadcr. 1980. Movement of 2.4.5-T through large
soil columns. Soil Science, v. 1 30. pp. 157-162.
665
-------
Pankow, J. F.. L. M. lubcllc.ind D. F. Barofsky. 1981.
The identification of chlorophenoxyphenoU in soil
and water samples by solvent extraction and field
desorption mass speetrometry. Anal. Chim. Aeta.
v. 124, pp. 357-364.
Pankow.J. F.. R. L. Johnson, J. E. Houck. S. M. Brillanic,
and W. J. Bryan. 1984. Migration of chlorophcnolic
compounds at the chemical waste disposal site at
Alkali Lake, Oregon. 1. Site description and ground-
water flow. Ground Water, v. 22, pp. 593-601.
Pankow, J. F. and L. M. Isabellc. 1984. Interface for the
direct coupling of a second gas chromatograph to i
gas chromatograph/mass spectrometer for use with a
fused silica capillary column. Anal. Chcm. v. 56,
pp. 2997-2999.
Research Triangle Institute. 1983. Master Analytical
Scheme for the Analysis of Organic Compounds in
Water, v. III.
Roberts, P. V., M. Reinhard, and A. J. Vaiocchi. 1982.
Movement of organic contaminants in groundwater:
Implications for water supply. J. American Water
Works Assoc. v. 14, pp. 408-413.
Schcllenbcrg, K., C. Leucnbergcr. and R. P. Schwarzenbach.
1984. Sorption of chlorinated phenols by natural
sediments and aquifer materials. Environ. Sci. and
Technol. v. 18. pp. 652-657.
Schwtnenbach, R. P., and J. Westall. 1981. Transport of
nonpolar organic compounds from surface water to
groundwater. Laboratory sorption studies. Environ.
Sci. Technol. v. 15. pp. 1360-1367.
Sudicky, E. and E. 0. Frind. 1982. Contaminant transport
in fractured porous media: Analytical solutions for a
system of parallel fractures. Water Res. Research.
v. 18. pp. 1634-1642.
van Gcnuchten, M. T. 1974. Mus transfer studies in sorbing
porous media. Ph.D. Thesis, New Mexico State Uni-
versity. Las Cruccs, NM.
Young. A. L., J. A. Calcagni. C. E. Thalken. and J. W.
Trcmblay. 1978. The toxicology, environmental fate,
and human risk of herbicide orange and its associated
dioxin. USAF OEHLTechnical Report OEHL
TR-78-92.
Richard L. Johnson is a joint Postdoctoral Research
Associate in the Department of Earth Sciences at the Uni-
versity of Waterloo in Waterloo, Ontario, and a Research
Scientist in the Department of Chemical, Biological, and
Environmental Sciences at toe Oregon Graduate Center in
Beaverton, Oregon. In 197J be received bis B.S. degree in
Chemistry from the University of Washington. In I9S4. be
received his Ph.D. in Environmental Chemistry from the
Department of Chemical, Biological, and Environmental
Sciences at the Oregon Graduate Center. His research
interests include the transport, fate, and modeling of con-
taminants in ground water in both porous and fractured
systems.
Susan M. Brillante is a graduate student Research
Assistant in the Department of Chemical. Biological, and
Environmental Sciences at the Oregon Graduate Center.
She received her B.A. in Chemistry from Loretto Heights
College in Denver. Colorado in 1966. Her research interests
include organic analytical chemistry and the fate of con-
taminants in ground-water systems.
Lone M. Isabelle is a Research Associate in the
Department of Chemical, Biological, and Environmental
Sciences at the Oregon Graduate Center. He received bis
B.A. in Chemistry from San Francisco State College in
1971 and bis M.S. in Organic Chemistry from California
State University at San Francisco. His research interests
include the application ofGC/MS/DS instrumentation in
the determination and study of organic compounds in the
environment.
James E. Houck is a Senior Scientist at NEA, Inc. in
Portland, Oregon. In 1971 be received bis B.S. in Chemistry
from the University ofAriiona. He received bit Ph.D. in
Chemical Oceanography from the University of Hawaii in
' 1978. His research interests include the analysis, monitor-
ing, and modeling of contaminants in the air and water
environments.
James F. Pankow is an Associate Professor in the,.
Department of Chemical. Biological, and Environmental
Sciences, and Director of the Groundwater Research
Laboratory at the Oregon Graduate Center in Beaverton,
Oregon. In 1973 be received bis B.A. in Chemistry from the
State University of New York at Binghamton. He received
a Ph.D. in 1979 in Environmental Chemistry from the
Department of Environmental Engineering Science at the
California Institute of Technology in Pasadena. California.
His research group has been involved in the development
and application of sensitive analytical methods incorporat-
ing capillary column GC/MS/DS techniques for the
determination of trace organic contaminants in ground-
water systems. His group employs these methods in studies
which examine and model many aspects of the processes
which control the transport and fate of organic chemicals in
the environment.
666
-------
TREATMENT
TECHNOLOGY.
Using the Properties of Organic Compounds to Help
Design a Treatment System
by Evan Nycr. Gary Boeticher, and Bridget Morello
I have decided 10 provide the physical/chemical and
ireatabiliiy properties of 50 compounds in my column
for this issue. The physical/chemical parameters of Ihe
compounds can be used to help evaluate data generated
during remedial investigations. The treatability parame-
ters can be used as a basis for the preliminary design
of a treatment system that will remove organic com-
pounds from ground water.
The biggesi obstacle in designing a treatment system
is where to begin. Typically, the two main slaning points
I have seen applied in designing a treatment system are
laboratory trcalability studies and "by-the-book"
design. Neither of these methods are accurate or effi-
cient. I n laboratory trealabiiity studies, the designer gen-
erally submits a ground waler sample to the laboratory
for purposes of simulating full-scale treatment units.
Laboratory treatabiliiy studies, however, cannot be used
as a direct simulation of most organic treatment pro-
cesses. (This issue will be discussed in deiail in my next
column). Textbooks should never be used as "cook-
books" for Ihe design of a treatment system. The cook-
book recipe simply uses every treatment method availa-
ble for removing organic compounds and sizes unil
operations based on values supplied in the textbook.
The final design uses all the treatment units in series.
Textbooks, including my own. should be used for general
knowledge and reference purposes only, not {or design
data.
The treatment system designs I have worked on have
always been preceded by complete evaluation of the
properties of the compounds. While I would not proceed
directly to full-scale installation based strictly upon anal-
ysis of compound properties, they can provide several
insights for final design. Most important, the properties
of compounds can indicate critical points of a design
and areas requiring further data. These areas can then
be further evaluated in laboratory and field pilot tests.
The main physical/chemical propenies that should
be evaluated prior to design arc solubility, specific grav-
ity, and octanol/water coefficient. These properties
mainly help us understand data generated during
remedial investigations. However, they will have some
inpui in the treatment system design as will be discussed.
Solubility
Solubility is one of the most important propenies
affecting the fate and transport of organic compounds.
. The solubility of a compound is described as the maxi-
mum dissolved quantity of compound in pure water at
a specific temperature. Solubilities of most common
organic compounds range from 1 to 100,000 ppm at
ambient temperature. However, several compounds
exhibit higher solubilities, and some are infinitely solu-
ble. Highly soluble compounds are easily transported
by the hydrologic cycle, and tend to have low adsorption
coefficients (or soils and low bioconcentration factors
in aquatic life. Highly soluble compounds also tend to
be more readily biodegradable.
Solubility usually decreases as temperature increases
due to an increase in water vapor pressure at the liquid/
gas interface. Escaping molecules then force larger num-
bers of gas molecules out of solution.
When reviewing the results from a ground waler
study, the concentrations of organic compounds should
• be related to the solubilities of those compounds. For
example, high concentrations of a non-soluble com-
pound may indicate the presence of a pure compound
NAPL. Therefore, the treatment system should be
designed -with the capability to treat pure compounds.
Table) presents the solubility values for 50 organic com-
pounds.
Specific Gravity
Specific gravity is a dimensionless parameter derived
from density. The specific gravity of a compound is
defined as the ratio of the weight of a compound of a
given volume and at a specified temperature to the
weight of the same volume of water at a given tempera-
ture. The specific gravity of a water at 4 C is usually
used as a basis because Ihe density of water at 4 C is
1.000 g/mL.
In environmental analysis, the primary reason (or
knowing the specific gravity of a compound is to deter-
mine whether liquids will float or sink in water. Pure
compounds that are lighter than water will form a layer
on top of the water. Organic compounds that are heavier
than waier will move through Ihe aquifer until they are
F.ll 1001 CU'MR HI
Reprinted by permission of the Ground Water Publishing Company. Copyright1991
All rights reserved.
-------
TABLE 1
Solubility for Specific Organic Compounds
1
•)
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
Compound .
Acenaphthene
Acetone
Aroclor 1254
Benzene
Benzo(a)pyrene
Benzo(g.h.i)perylene
Benioic acid
Bromodichloromeihanc
Bromoform
Carbon leirachloride
Chiorobenzene
Chloroeihane
Chloroform
2-Chlorophenol
p'Dichlorobenzenc (1,4)
1.1-Dichloroethane
1.2-Dichtoroethane
1.1-Dichloroethylcne
cis-l,2-Dichloroeihylene
trans-1.2-Dichloroeihylcne
2.4-Dichlorophenoxyaceiic acid
Dimethyl phthalale
2.6-Dinitrotoluene
1.4-Dioxane
Eihyl benzene
bis(2-Eihylhexyl)phthalate
Heplachlor
Heiachlorobenzene
Hexachloroelhanc
2-Hexanone
Itophorone
Methylene chloride
Methyl elhyl keione
Methyl naphthalene
Methyl lert-butyl ether
Naphthalene
Nitrobenzene
Penlachlorophenol
Phenol
1. 1.2.2 -Tetrachloroelhane
Tetrachloroethylene
Tetrahydrofuran
Toluene
1.2,4-Trichlorobenzene
1.1.1-Trichloroelhanc
1.1.2-Trichloroethanc
Trichloroethylene
2.4.6-Trichlorophenol
Vinyl chloride
o-Xylenc
Solubility
Imf/L)
3.42
1x10° •
1.2x10-'
1.75x10"
1.2xlO-J
7x10-
2.7X101
4.4x1 01
3.0UI03
7^7x10=
4.06x10*
5.74x1^
8.2x1 03
2.9x10*
7.9x10'
5.5x10*
8.52X103
2.25x1 01
. 3JxlO"
6JXI01
6.2x10*
4J2X101
1J2X101
4JU105
UZxlO2
2^5x10-'
1.8x10-'
6xl(TJ
5x10'
1.4x10*
1-2x10*
2x10*
2.68X103
2^4x10'
4.8
3.2x10'
1.9XIO*
1.4x10'
9JxlO*
2.9x1 03
UxlO2
3x10-'
5J5X102
3x10'
UxlO1
4.5X103
. l.lxlO3
BxlO1
2.67XI03
1.75XI02
Reference
2
2
1 (A)
2
2
2
2
1 (B)
1 (A)
1 (A)
2
.1 (A)
2
2
(A)
(A)
(A)
(A)
(A)
2
2
2
2
1 (A)
2
2
1 (A)
2
1
2
1 (B)
1 (A)
2
3
2
2
1 (B)
1 (A")
2
1 (A)
4
1 (A)
2
1 (A)
1 (A)
1 (A)
2
1 (A)
1 (C)
TABLE 2
Specific Gravity for Specific Organic
Compounds
* Solubility of ljDOO.000 m|/L assigned because of reponed •infmiie solubil-
ity" in [he literature.
\.Supfrfund Public Health Evaluation Manual. Office of Emergency and Re-
medial Response Office of Solid Waste and Emergency Response. U.S.
Environment*I Protection. Atencv. 1986.
A. Environment*) Cnieru and Aueument Office (ECAO). EPA. Health
E/lecu Atteawnenu for Specific Chemicals. 1982.
B. Mabey. W.R., J.H. Smith. R.T. Rodoll, H.L- Johnson. T. Mill. T.W.
Chou. J. Cam, I.W. Patndte. H. Jaber. and D. Vanderberf. "Aquatic
Fate Process Data for Organic Pnority Pollutants," EPA Contract Noi.
68-01-3867 ind 68-OV298I by SRI International, for Moouonng ind
Data Suppon Division. Office of Water Regulations and Standard}.
Washington. D.C. 1982.
C Dawson. et al.. Physical/Chemical Propenic* of Hazardous Waste Con-
stituents, by Southeast Environmental Research Laboratory for U.S.
EPA. 1980.
2. U.S.EPA "Basics of Pump-and-Trcat Ground-Water Remediation Tech-
nology" EPA/600/8-901003. Robert S. Kerr Environmental Research Labo-
ratory, March 1990.
3. Manufacturer's data: Texas Pctrochemtcati Corp.. Gasoline Grade Methyl
ten-butyl ether Shipping Specification and Technical Data. 198ft.
J CKC Handbook of Chemutiy and Phwcs. 71st Edition. CRC Press. Ohio.
1990.
Compound
1 Acenaphlhcne
2 Acetone
3 Aroclor 125J
4 Benzene
5 Benzo(a)pyrene
6 Benzo(g.hu)perylene
7 Benzoic acid
8 Bromodichloromethane
9 Bromoform
10 Carbon tetrachloridc
11 Chlorobenzene '
12 Chloroelhane
13 Chloroform
14 2-Chlorophenol
15 p-bichlorobenzene (1.4)
16 1.1-Dichloroethane
17 1.2-Dichlorocthane
18 1.1-Dichloroeihylene
19 cis-1.2'Dichloroe!hylene
20 trans-1.2-Dichloroethylene
21 2.4'Dichlorophenoxyacetic acid
22 Dimethyl phthalale
23 2.6-Dinitroioluene
24 1,4-Dioxanc
25 Ethvlbenzene
26 bis(2-Eihy!hexyl)phthalaie
27 Heplachlor
28 Hexachlorobenzene
29 Hexachloroelhane
30 2-Hcxanone
31 Isophorone
32 Methylene chloride
33 Methyl elhyl keione
34 Methyl naphthalene
35 Methyl len-butyl ether
36 Naphthalene
37 Nitrobenzene
38 Penlachlorophenol
39 Phenol
40 1.1^2-Tetrachloroelhane
41 Tetrachloroethylene
42 Tetrahydrofuran
43 Toluene
44 1.2.4-Trichlorobenzene
45 1.1.1-Trichloroelhanc
46 1.1.2-Trichloroethanc
47 Trichloroethylene
48 2.4.6-Trichlorophenol
49 Vinyl chloride
50 o-Xylene
Specific
Gravity Reference
1.069 (95W) 1
.791 1
U (25!) 3
.879 1
1.35 (25' j 4
NA
1J16 (2S-/4') 1
2.006 (15V4-) |
2.903 (I5-) 1
1.594 1
1.106 1
.903 1
1.49 (20-C liquid) J
1.241 (18.2V1S') 1
1.458 (21') 1
1.176 1
1.253 1
1.250 (15') 1
1 .27 (25'C liquid) 2
1.27 (25'C liquid) 2
1.255 6
1.189 (25*/25*)
1.283 (111')
1.034
^67
.9843
1.57
2.044
2.09
.815 (18*W)
.921 (25')
1.366
.805
1.025 (14'W)
.731
1.145
1.203
1.978 (22')
1.071 (25V4-)
1.600
1.631 (15'/4-)
^88 (21V4')
.866
1.446 (26')
1.346 (15V4-)
1.441 (25JV4')
1.466 (20-/20-)
1.490 (75V4-)
.908 (25V25')
.880
* Speafic gravity of compound at 20"C referred to water at 4*C
(20"/4*) unless otherwise specified.
NAcNol Available
1. Lanti's Handbook of Qttmistry. llth edition, by John A. Dean.
McGraw-Hill Book Co.. New 1973.
2. Hazardous Chemicals Data Book, 2nd edition, by G. Weiss.
Noyes Dan Corp.. New York. 1986.
3. U.S. Public Health Service Agency for Toxic Substances and Dis-
ease Registry. 'Draft Toxicological Profile for Selected PCBs."
November 1987.
4. U.S. Public Health Service Agency for Toxic Substances and Dis-
ease Registry. "Draft ToxicologicarProfile for Benzo(a)pyrene.~
October")987.
5. Verschueren. Karel. Handbook of Envtronmental Deia on
Orfomf Chemicals, 2nd edition, Van Nosirand Reinhold Co..
New York. 1983.
6. Merck Index. 9lh edition. Merck and Co. Inc.. New Jersev. 1976.
82
Fall 1991 GWMR
-------
fully adsorbed by soil panicles or until (hey encounter
an impenetrable layer. Table 2 presents the specific gra-
vities of 50 organic compounds.
Octanol/Water Partition Coefficient
The ocianol/water partition coefficient (K0..) is
defined as the ratio of a compound's concentration in
the octanol phase to its concentration in the aqueous
phase of a two-phase system. Measured values lor
organic compounds range from 10~3 to )07. Lew K.ov.
values (< 10) are considered hydrophilic and tend to
have higher water solubility. High K0» values (> 10*)
are very hydrophobic.
K.ow values for organic compounds are used to evalu-
ate fate in the environment. The parameter can be
related to solubility in water and bioconcentration
effects, but it is mainly used to relate to soil/sediment
adsorption. When combined with the organic content
of the soil, lv,w values can be used to predict the amount
of material adsorbed in the soil and the retardation
factor for movement through the aquifer.
When pure compounds are lost to the environment,
it is important to know where they are likely to be found.
Soluble compounds will migrate with the surface water
which will infiltrate the aquifer and migrate with the
ground water. Non-soluble compounds will be adsorbed
on the soil. However, if the mass of organic compounds
exceeds the adsorptive capacity of the soil, the com-
pounds will continue to migrate until they reach the
aquifer. Compounds with low specific gravity will be
retained at the surface of the aquifer, and compounds
with high specific gravity will continue to move vertically
through the aquifer. Table 3 presents KOW values for 50
organic compounds.
The physical/chemical properties presented here will
help the reader understand where compounds of con-
cern might be in the ground and/or aquifer. These pro-
perties are also necessary for use in designing treatment
systems such as oil/water separators and liquid/liquid
extractors.
The main treatabihty parameters thai should be used
to help design a treatment system are strippability (Hen-
ry's law constant), adsorbability, and biodegradability.
Henry's Law Constant
Generally for non-ideal solutions, Henry's law stales
that the equilibrium partial pressure of a compound in
the air above the air/water interface is proportional to
the concentration of that compound in the water. Hen-
ry's law can be expressed as follows:
PA = HAXA
where:
PA = Partial pressure of a compound in liquid at equi-
librium with gas (aim)
HA = Henry's laws constant (aim)
XA = Mole fraction of a compound in gas (mole/mole)
Therefore. Henry's law constant expresses the amount of
chemical partitioning between air and water at equilibrium.
TABLE 3
Octanol Water Coefficients (K,,w)
for Specific Organic Compounds
]
2
i
4
5
6
1
S
0
10
11
12
13
14
15
16
n
IE
19
20
21
2:
23
2<
25
26
21
2S
29
30
31
32
33
34
35
36
37
38
39
40
41
4;
43
44
45
46
47
46
49
50
Compound
Acenaphlhene
Acetone
Aroclor 1254
Benzene
benzo(a)pyrene
bcnzo(g,h.i )pe rylene
Benzoic acid
Bromodichlorome Inane
Bromoform
C«rbon tetrachloride
Chlorobenzene
Chloroeihane
Chloroform
2-Chlorophenol
p-Dichlorobenzene (1,4)
l.l-Dich)oroelh>nc
1.2-Dichloroethanc
1.1-Dichloroelhylene
cii-1.2-Dichioroethylene
lrans-1.2-Dichloroelhylene
2.4-Dichlorophenoiyaceiic acid
Dimethyl phlhalate
2.6-Diniiroioluene
1.4'Dioxane
Ethylbcnzcne
bit(2-Ethylheiyl)phthalate
Hepuchlor
Heiachlorobenzene
Hexachloroe thane
2-Hexanone
Isophorone
Meihylene chloride
Methyl ethyl keione
Mcihyl naphthalene
Methyl len-butyl ether
Naphthalene
Nitrobenzene
Peniachlorophenol
Phenol
1 .1 ,2.2-Teirachloroeihane
Telrachloroethylcnc
Teirahydrofuran
Toluene
1.2.4-Trichlorobcnzenc
1 .1.1 -Trichloroelhanc
1.1.2-Trichlorocthanc
Trichloroethylenc
2.4.6-Trichlorophenol
Vinyl chloride
o-Xylene
K..
1.0x10'
6x10-'
1.07x10*
1.3xlO:
l.lSxlO*
3.24x10*
14X10'
7.6x10'
2.5x10'
4.4x10'
6.9x10-'
33x10'
9.3x10'
1.5x10'
3.9x10'
6.2x10'
3.0x10'
6.9x10'
5.0
3.0
6.5x10'
1.3x10'
LOxlO1
1.02
1.4x10'
9-SxlO3
Z-SUKT1
1.7X105
3.98xltT
23x10'
5.0x10'
1.9x10'
1.E
UilO*
NA
2.8x10'
7.1x10'
l.OxlO3
2.9x10'
2.5x10'
3.9x10'
6.6
UxlO-'
2.0x10'
3.2x1 0:
2.9x10'
2.4x10*
7.4x10'
2.4x10'
B.9xlO-'
Reference
2
1 (D)
1 (A)
•>
•t
-1
I
1
1 (B)
1 (A)
1 (A)
2
HA)
2
2
1 (A)
1 (A)
1 (A)
1 (A)
1 (A)
2
4.
2
2
i (A)
2
2
1 (A)
2
3
2
1 (B)
1 (A)
2
2
2
1 (B)
1 (AJ
2
i 1>386? and 6&43-2981 by SRI International, (or Monitoring and
Data Support Division. Office of Water Regulation* and Siandardi.
Washington. D.C, 19S1
C Dawton ei al.. Phyucal/Oiemicat Propenies of Hazardous Waste Con-
ftituenu. by Southeast Environmental Research Laboratory lor U.S.
EPA. ,980
D. Handbook ef Exwotunttual Data for Organic Oumtcets. Van
Kostnnd Retnhoid Co. New York. 2nd Edition. 1983.
"L \J£. EPA "Basio of Pump*and-Treai Cround-Witer Remediation Tech-
txJofy." EPA/60&-6-9Qr{XU. Roben S. Kerr Eitvironmentat Research Labo-
ratory. March 1990.
3. Lvman. Warren J.. el at. *Roe«rch and Devclopmenl of Methods for Esti-
mating Phytiocochcmical Propenies of Organic Compound! of Environ-
mental Concern," June 1981.
4. EPA Dnfi Document "Hazardous Waste Treatment. Storage and Disposal
Faciliiiei fTSDF) Air Ermtstoni Model.' April 1989.
Fall 1991 GWMR
-------
TABLE 4
Henry's Law Constants
for Specific Organic Compounds
Compound
Henrys La»
Constant* aim mj
wiier/mj air Reference
1 Accnaphthcne
2 Acetone
3 Aroclor 1254
4 Benzene
5 Benzo(»)pyrene
6 Benzo(g,hj)perylene
7 Benzole acid
8 Bromodichloromelhane
9 Bromoform
10 Carbon teirachloride
11 Chlorobenzene
12 Chloroeihane
1} Chloroform
14 2-Chlorophenol
15 p-Dichlorobenzene (1.4)
16 1.1-Dichloroethane
17 U-Dichloroetnane
18 1.1-Dichloroeihylene
19 cis-1.2-Dichloroethylene
20 uans-1.2-Dichlorocthylene
21 2.4-Dichlorophenoxyacetic acid
22 Dimethyl phthalaie
23 2.6-Dinitrotoluene
24 1.4-Dioxane
25 Ethylbenzcne
26 bis(2-Ethylhexyl)phlhalate
27 Hepiachlor
28 Hexachlorobenzene
29 Hcxachloroeihane
30 2-Hcxanone
31 Isophorone
32 Methylene chloride
33 Methyl ethyl ketone
34 Methyl naphthalene
35 Methyl ten-butyl ether
36 Naphthalene
37 Nitrobenzene
38 Pcmachlorophenol
39 Phenol
40 1.1^2-Tetrachloroethane
41 Tetrachloroethylenc
42 Teirihydrolunn
43 Toluene
44 1.2.4-Trichlorobenzene
45 1,1,1-Trichloroethane
46 1.1.2-Trichloroethane
47 Trichlorocthylene
48 2.4.6-Trichlorophenol
49 Vinyl chloride
50 o-Xylene
1 - at water temperature of 68"F
5.1
0
150
230
.1
0
0
127
35
1282
145
34
171
0.93
104
240
51
1841
160
429
10
0
.2
.6
359
0
46
37.8
138
1.6
.3
89
1.16
3.2
196
20
j 2
o!l5
0.017
21
1035
2 •
217
128
390
41
544
.2
355.000
266
5
1
5
1
5
5
5
. 1
3
1
2
5
1
2
5
5
5
5
1
5
5
2
5
5
5
1
2
2
1
4
5
•)
±
2
5
1
5
1
5
1
2
1
5
3
1
1. per Hydro Croup Inc.. 1990
2. Solubility and vapor phase pressure data from Handbook of Envi-
ronmental Data on Organic Chemicals. 2nd Edition, by Karel Vers-
chueren. 1983. Van Nosirand Remhold Co.
3. Michael C Kavanaugh and R. Rhodes Truisel. "Design of Aeration
Towers to Strip Volatile Contaminants from Drinking Water"
Journal AWWA. December 1980. p. 685.
J. Coskum Yuneri. David F. Ryan. John j. Callow. Miral D. Gurol.
"The Effect of Chemical Composition of Water on Henry's Law
Constant." Journal WPCF. Volume 59. Number 11. p. 954. Novem-
ber 1987.
5. U.S. EPA. "Basics of Pump-and-Treat Ground-Water Remediation
Technology." EPA/600-ft-90/003. Robert S. Kerr Environmental
Research Laboratory. March 1990.
Aeration is a technology often employed in water
treatment applications to strip the concentration of vola-
tile organic compounds (VOCs) from water. The con-
trolling factor in removal of VOCs from water is the
rate of transfer from the liquid phase (water) to the gas
phase (air) until equilibrium is established. The transfer
rate of VOCs from water via aeration depends upon
the temperature of both the water and the air. as well
as the physical and chemical properties of the VOCs.
Water temperature changes of as little as 10 C can result
in threefold increases in Henry's law constants. In a gas-
liquid system, the equilibrium vapor concentration of a
VOC can be computed from the compound specific
Henry's law constant and total system pressure.
Generally, the greater the Henry's law constant (i.e.,
greater than 160 aim), the more volatile a compound.
and the more easily it can be removed from solution.
Henry's laws constants can be computer modeled to
develop a preliminary design and cost estimate for an
air stripper. Table 4 presents Henry's law constants for
50 organic compounds.
Carbon Adsorption Capacity
Activated carbon has variable effectiveness adsorb-
ing organic compounds. Low molecular weight, polar
compounds are not well absorbed. High molecular
weight, non-polar compounds such as pesticides, poly-
chlorinated biphenyls, phthalaies, and aromatics are
readily adsorbed.
Activated carbon adsorption isotherm data can be
used to evaluate the carbon adsorptive capacity for
organic compounds. These data may be used to com-
plete an initial estimate of the organic mass that carbon
will adsorb. Since the main cost of carbon adsorption
is carbon, this mass data can be used as a preliminary
basis for cost estimation. Table 5 presents carbon
adsorption capacity values for 50 organic compounds.
Biodegradability
Organic compounds are transformed by biochemical
reactions in the environment and in engineered unit
operations. Biodegradation of organic compounds occur
aerobically and/or anaerobically depending on the
molecular structure of the chemical and the environ-
mental conditions. Engineered bioremediation is neces-
sary to enhance natural processes that are usually less
than optimal in the environment.
The first and most important parameter to evaluate
before implementing bioremediation is determining
whether the compound is degradable. the most effective
biodegradation mechanism (aerobic vs. anaerobic), and
the biodegradation rate. From an ecological point of
view, chemicals that are completely degradable. but
slow, can be persistent in the environment.
Biodegradation potential has been reviewed and can
be categorized as degradable. persistent, and recalcitrant.
Readily degradable refers to compounds that have passed
biodegradability tests in a variety of aerobic environments.
Degradable also refers to compounds that are normally
degraded in tests but not necessarily in the environment.
Fill 1991 GWMR
-------
TABLE 5
Adsorption Capacity for Specific
Organic Compounds
Adwrption
Capaciti
(mg rompound/p
carbon I
Compound at 500 ppb Reference
1 Acenaphlhene 155 4
2 Acetone 4> 1
3 Aroclor 1254 NA
4 Benzene 80 1
5 Benzo(a)pyrene 24.8 4
6 Benzo(g.h.i)perylene 6.? 4
7 Benzoic acid 40 (at pH = 3) 4
8 Bromodichloromethane 5 4
9 Bromoform 13.6 4
10 Carbon leirachlohde 6.2 2
11 Chiorobenzene 45 3
12 Chloroethane 0.3 4
13 Chloroform 1.6 I
14 2-Chlorophenol 38 3
15 p-Dichlorobenzene (1.4) 87.3 4
16 1.1-Dichloroethane 1.2 4
17 1.2-Dichioroelhane 2 2
18 1.1-Dichloroethylenc 34 4
19 cii-U-Dichloioethylenc 9 5
20 trans-U-Dichloroethylene 2.2 4
21 2.4-Dichlorophenoxyacetic acid NA
22 Dimethyl phlhalate 91.2 4
23 2.6-Diniiroioluene 116 4
24 1.4-Dioxane 0.5-1.0 5
25 Elhylbenzene 18 1
26 bis(2-Ethylhexyl)phlha!aic 3995 4
27 Heptachlor 631.5 4
28 Hexachlorobenzene 42 3
29 Hexachloroethane 74.2 4
30 2-Hexanone <13 5
31 Isophorone 244 t
32 Methylenc chloride 0.8 3
33 Methyl ethyl kctonc 94 1
34 Methyl naphthalene 150 5
35 Methyl ten-butyl ether 6.5 5
36 Naphthalene 5.6 3
37 Nitrobenzene 50.5 4
3S Peniachlorophenol 100 3
39 Phenol 161 1
40 1.1.2.2-TctrachloroethanE 6.2 4
41 Tetrachlorocthylene 34.5 2
42 Tetrahydrofuran <0.5 5
43 Toluene ' 50 I
44 1,2.4-Trichlorobeniene 126.6 4
45 1.1.1-Trichloroethanc 2 2
46 1.1.2-Trichloroelhane 3.7 4
47 Trichloroethylene 1S.2 2
48 2.4.6-Triehlorophenol 179(atpH«3) 4
49 Vinyl chloride TRACE 3
SO o-Xylenc , 75 4
NA e Not Available
1. Verschuren. Karel. Handbook of Environmental Data on Organic
Chemicals. New York: Van Nostrand Rcinhold. 1983.
2. Uhler. R.E. el al. Treatment Alternative for Croundwater Con-
tamination. James M. Montgomery. Consulting Engineers.
3. Sienzel, Mark. Letter of Correspondence to Evan Nycr. August
21. 1984.
4. U.S. EPA "Carbon Adsorption Isotherms for Toxic Organic."
EPA-600/8-80-023. Municipal Environmental Research Labora-
tory April 1980.
5. Roy. Al. Calgon Carbon. 1991.
TABLE 6
Disappearance or Biodegradation Potential
for Specific Organic Compounds
1
1
3
4
5
6
1
8
9
10
1)
12
13
14
15
16
11
IE
19
JO
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
4£
49
50
holes:
Compound
Acenaphlhene
Aceionc
Aroclor 1254
Benzene
Bcnzo(a)pyrene
Benzo(g.h.i)perylene
Benzoic acid
Bromodichloromelhane
Bromolorm
Carbon tetrachloride
Chlorobenzene
Chloroethanc
Chloroform
2-Chlorophenol
p-Dichlorobenzene (l,4j
1,1-Dichioroeihane
1 ,2-Dichloroelhanc
l.l-Dichloroethylene
cis-U-Dichloroethylene
irans-1.2-Dichloroethylenc
2,4.Diehlorophenoxyacetic acid
Dimethyl phihaiaie
2.6-Dinitroioluene
1.4-Dioiane
Elhylbenzenc
bis(2.Elhylheiyl)phihalate
Hepiachlor
Hexachlorobenzene
Hcxachloroelhane
2-Hexanone
Isophoronc
Melhylene chloride
Methyl ethyl ketone
Methyl naphthalene
Methyl ten-butyl ether
Naphthalene
Nitrobenzene
Pentachlorophenol
Phenol
1 .1 .2.2-Tetrachloroethanc
Tetrachloroeihylenc
Tetrahydrofuran
Toiuent
1 ,2.4-Trichlorobenzene
1 .1 .1 -Trichloroethane
1 ,1.2-Trichloroethane
Trichloroethylene
2.4.6-Trichlorophenol
Vinyl chloride
c-Xylenc
D - Detrtdabk
P • Peruneni
B...
defradabilht Ref
D
D
P.D
D
P.D
P.D
D
P.D
P.D
P.D
D
D
P.D
D
P.D
P.D
P.D
P.D
P.D
P.D
D
D
D.P
P.R
D
P.D
P.R
P.R
D
D
D
D
D
D
NA
D
D
P.D
D
P.D
P.D
D
D
P.D
P.D
P.D
P.D
D
P.D
D
R « Recatouani
NA • Not Available
erenre
;
1 *
1
2.J
2.i
?
1
1
6
1
1
1
1
1
)
2
5
2.5
8
1
2.5
2
1
2
5
5
1
1
2
1
)
2.5
1
7
1
2
Rclcrencu:
1. Guiiom comptkd in E.K. Nyer, Crouiufweirr Jrtaimtnt Technology. 2nd
E4~ In Produnion
2. Dragu/.. J.. TV Soil CAemittry of Hazardous Motet tab. The H&urdoiu
Material! Coetrol Research Iminuit. 19S&. pp. 367-3~
3. bcttard. DI_ "Bacterul TranilomnaticMa of Polychlorinaied Biphenyb."
In: titoifchnoiofv and Btoacfrwietion D. Ktnely. A. Chakrabany. Ci.
Omcnn (£01.) Advances in Applied Bioiecnnok»ty Seriei. VbL 4. Poniolio
Pub. Co.. The WcxxUandt. Tuai. 199C.
4, "Qvancieruaiioo and Laboratory Soil 7rcaubi)ity Studies lor Creosote
and PenucUorephenoI Sludge* and Conummatcd Soil." EPA: Wuhm|-
ton. D.C. 198&. EPAMOCV7-88^i.
5. Finer. P. J. Cbudoba, BioeefrmdabtitiY of Organic Substances in itW
AoMoti'c £ji*wtMme/u. CRC Preu, 1990
6. Vojtcl. T.M. PI. McCany. Tratulormiiioru ol H*k>feiiaied Aliphatic
Compound,- Env. So. Technol.. 21. 732-736. 1987.
7. Votskj)-. V.T- CP. Gr*d). "Toiieaty of Setecied RCRA Compoundi to
Activated Sludge Mkroorpantsmi." Journal WPCF. Vol. 60. No. 10. 1850,
IVfti,
8. Klecka. C.M, SJ. Consoii. "Removal of 1.4-Dioxanc fiom \Vaiiewaici.~
Journal of Huarooui Material!. 13. 16I-16S. 1986.
Fall 1991 GWMR
85
-------
Persistent refers to chemicals thai remain in the envi-
ronment for long periods of lime. These compounds are
not necessarily "non-degradable." but degradation
requires long periods of acclimation or modification of the
environment to induce degradation. Recalcitrant refers to
compounds that are non-degradable.
From the literature, each compound must be evalu-
ated to determine the estimated time to complete the
transformation of the chemical under optimal condi-
tions. If the time period is acceptable, treatability and
pilot plants can then be initiated. Table 6 presents bio-
degradation potential for 50 organic compounds.
We can combine these treatability properties with
our experience in full-scale design and generate a theo-
retical preliminary design. This design can be used to
generate a preliminary cost estimate. Based upon this
data, we can eliminate the technologies that obviously
will not work. This data will also show us which com-
pounds are controlling the designs. We can then go back
and confirm their concentrations in the field, and test
the actual treatment in laboratory and pilot plant tests.
I hope you find these tables to be a convenient source
of important information. I encourage you, however,
not to use the data as a final basis for full-scale design.
Evan K. Nyer is an expert in the research and appli-
cation of technology to ground water cleanups. As vice
president with Ceraghty & Miller Inc., he is responsible
for engineering services including hazardous and solid
waste management, environmental and natural
resource management, remediation activities and
designing treatment systems for contaminated sites
throughout the United Stales and in foreign countries.
He has designed more than 100 ground water treatment
systems.
Nyer travels throughout the country leaching treat-
ment techniques at seminars and universities. He has
written numerous papers on ground water decontami-
nation and other water and waste water cleanup tech-
niques. He is responsible for bringing to the field many
innovative techniques for biological treatment of water,
soils, and in situ treatment and the application of exist-
ing technologies to ground water contamination. He is
a member of the Water Pollution Control Federation,
The National Water Well Association, The American
Institute of Chemical Engineers, and The American
Society of Civil Engineers.
Bridget Morello received a B.S. Che from the Uni-
versity of South Florida in 1987 and is currently work-
ing for Geraghry it Miller^ Process Group in Tampa,
Florida. She is mainly involved in treatabiliry evalua-
tion and design of ground water treatment systems.
Gary Boettcher is a project scientists with Geraghry
& Miller Inc. in Tampa, Florida. He received his B.S.
degree in microbiology from the University of South
Florida and is currently pursuing a Master of Public
Health (MPH) degree. He is involved in investigation,
treatability, and design of biological remediation
systems.
86
Fall 1991 GWMR
-------
Section 8
-------
GROUNDWATER FLOW RATES
AND MODELING
PERFORMANCE OBJECTIVES
At the end of this lesson, participants will be able to:
• Describe the physical, chemical and biological processes that
affect groundwater flow and contaminant transport
• List three factors that groundwater models can predict
• List three problems associated with groundwater models used
in groundwater assessment
• Identify misuses of groundwater models.
-------
NOTES
GROUNDWATER FLOW
RATES AND MODELING
PHYSICAL PROCESSES
• Advection
• Hydrodynamic dispersion
• Molecular diffusion
• Density stratification
• Immiscible phase flow
• Fractured media flow
CHEMICAL PROCESSES
• Oxidation-reduction reactions
• Radionuclide decay
• Ion exchange
• Complexation
• Cosolvation
• Immiscible phase partitioning
• Sorption
9/93
Groundwater Flow Rates and Modeling
-------
NOTES
BIOLOGICAL PROCESSES
• Microbial population dynamics
0
•
•
Substrate utilization
Biotransformation
Adaptation
• Cometabolism
1
2
Conceni
1
k
0
I^B Distance from source •
Advection
+
DARCY'S LAW
Q = KIA
• Q = discharge
• K = hydraulic conductivity
• = hydraulic gradient
• A = area
Groundwater Flow Rates and Modeling
9/93
-------
long path
short path
fast
t
o
0)
o
o
O
I
Distance from source
Pore
size
Path
length
Friction
in pore
Advection
plus
dispersion
Distance
NOTES
9/93
Groundwater Flow Rates and Modeling
-------
NOTES
c-*[
c =
o
V =
L =
t =
erfc
CONCENTRATION
AT DISTANCE "L"
= longitudinal dispersion coefficient
solute concentration at source
average linear velocity
distance
time
= complementary error function
t
c
o
13
^
c
0)
o
o
O
1
0
\~"
Advection
plus
retardation
^
•• Distance from source MHlfe
RETARDATION
R = 1 +A x Kd
n
R = retardation factor
A =
ry =
bulk density
distribution coefficient = (KoC )(foc)
n = porosity
Contaminant Velocity:
vx = contaminant velocity
v = ground water flow velocity
Rx = retardation factor for contaminant x
Groundwater Flow Rates and Modeling
9/93
-------
NOTES
Hypothetical contaminant plume
with a small transverse dispersivity
Waste
o>
l
2
CD ,
Hypothetical contaminant plume
with a large transverse dispersivity
Waste
i
2
o
.
0.2 \
\\
0.3
Continuous source
Groundwater flow
t
0 I1 12 13
One-time source
9/93
Groundwater Flow Rates and Modeling
-------
NOTES
DNAPLSOyRCE
^^
Residual DNAPL->
'///////
/ Lower permeability strata ./I
////////
-------
NOTES
:::::]]]]]]]\]\ Diffusion :
jjl::::::: into rock
*;;*;:;;*;;*;;*
Fractured porous rock
PRIMARY MODEL TYPES
Defined by Objective
Screening models
In-depth environmental fate models
MODELS CAN PREDICT
• Spatial variation
• Temporal variation
• Parameter variation
9/93
Groundwater Flow Rates and Modeling
-------
NOTES
MODEL DIMENSIONS
ONE-DIMENSIONAL
TWO-DIMENSIONAL
THREE-DIMENSIONAL
2-D
Prediction
Reality
MODELING PROBLEMS
• Lack of appropriate modeling
protocols and standards
• Insufficient technical support
• Inadequate education and training
• Widely used, but selection and use
inconsistent
Groundwater Flow Rates and Modeling
9/93
-------
NOTES
MOST COMMON EPA MODELS
Name
MODFLOW
HELP
RANDOM WALK
USGS-2D
USGS-MOC
Relative Use
29
24
21
20
19
KEYS TO SUCCESSFUL
USE OF MODELS
Proper imput data and parameter
estimates
Effective communication
Understanding the limitations of the
model
G.I.G.O.
Garbage in = Garbage out
The first axiom of computer usage
9/93
Groundwater Flow Rates and Modeling
-------
NOTES
Selection Criteria for Mathematical
Models Used in Exposure
Assessments:
Ground-Water Models
U.S. EPA. 1988. EPA/600/8-88/075.
MODELING PROCESS
• Problem characterization
• Site characterization
• Model selection criteria
• Code installation
• Model application
I Ground-Water Flow \
| .Water, Table or Confined Aquifer? \
I Porous Media or Fracture Flow? I
•
I 1. 2. or 3 Dimensional? \
| Single Phase or Multi-Phase? |
Homogeneous or Heterogeneous?
Hydraulic Conductivity. Recharge,
Porosity, Specific Storage
I Single Layer or Multi-Layer? \
(Constant or Variable I
Thickness Layers? \
•
| Steady-State or Transient? \
Select the Appropriate Analytical or
Numerical Ground-Water Flow Code
or
Continue with the Decision Tree
and Select a Combined
Ground-Water Flow and
Contaminant Transport Model
Groundwater Flow Rates and Modeling
10
9/93
-------
NOTES
WATER TABLE Oft
CONFINED AQUIFER
• Water table
• Confined (majority)
• Combination of both
POROUS MEDIA/
FRACTURE FLOW
• Porous media (majority)
• Fractured media
• Combination of both
1,2, OR 3 DIMENSIONAL
• 3 dimensional (preferred)
• Availability of data
9/93
11
Croundwater Flow Rates and Modeling
-------
NOTES
SINGLE PHASE OR
MULTI-PHASE
Few models available for multi-phase
flow
HOMOGENEOUS OR
HETEROGENEOUS
Homogeneous if:
- Conductivity values are within
an order of magnitude
- Recharge, porosity, and storage
values vary less than 25%
SINGLE LAYER OR
MULTI-LAYER
Single layer aquifer if:
- Hydraulic conductivities are
within an order of magnitude
- Hydraulic gradients and porosity
are within 25%
- Flow direction is same
Groundwater Flow Rates and Modeling
12
9/93
-------
NOTES
CONSTANT OR VARIABLE
THICKNESS LAYERS
• Constant or uniform if thickness
changes less than 10%
STEADY-STATE OR TRANSIENT
• Steady-state if water table fluctuations
are less than 10%
• Transient is difficult to implement
Select the appropriate analytical or
numerical groundwater-flow code
or
• Continue with the decision tree and
select a combined groundwater-flow
and contaminant transport model
- Compatible but separate models
- Combined flow and transport
model
9/93
13
Groundwater Flow Rates and Modeling
-------
NOTES
c
I Contaminant Transport I
^wf^
Point, Line, or Areal Source?
TH~
D
| Initial Value or Constant Source? \
St
Nurr
^B
1, 2, or 3 Dimensional?
TH~
1 Dispersion? \
^^^^^wf^
Adsorption?
• Temporal Variability
• Spatial Variability
•
Degradation?
• 1st Order/and Order
• Radioactive Decay
•
Density Effects?
' Thermal and/or Concentration
m
tlect the Appropriate Analytical
erlcal Contaminant Transport C
or
ode
POINT, LINE, OR
AREAL SOURCE
• Point source: pipe outflow or well
• Line source: trench
• Areal source: waste lagoon or
agricultural field
• Volume source: volume in
groundwater
INITIAL VALUE OR
CONSTANT SOURCE
• Instantaneous pulse
• Continuous release
- Constant
- Variable
Groundwater Flow Rates and Modeling
14
9/93
-------
NOTES
1,2, OR 3 DIMENSIONAL
3 dimensional unless lower dimension
is justified
1 dimensional generally predicts
higher concentrations
DISPERSION
Represents spreading of solute
caused by mechanical mixing
Difficult to measure in the field
Requires field calibration
ADSORPTION
Temporal/Spatial Variability
Process whereby dissolved chemicals
become attached to solids
Current practice: lump chemical and
biological processes into retardation
9/93
15
Groundwater Flow Rates and Modeling
-------
NOTES
DEGRADATION
1st/2nd Order - Radioactive Decay
Degradation results from:
- Biological transformations
- Hydrolysis
- Other chemical reactions
DENSITY EFFECTS
Thermal and/or Concentration
Naturally occurring situations normally
not affected by density
Landfill leachates often affected by
density
Select the appropriate analytical or
numerical contaminant transport code
- Transport model compatible with
groundwater-flow model
- Combined groundwater-flow/
contaminant transport model
Groundwater Flow Rates and Modeling
16
9/93
-------
NOTES
SOURCES OF MODELS
AND MODEL INFORMATION
Superfund Exposure Assessment Manual
Chapter 3 - Contaminant Fate Analysis
(35 Models)
U.S. EPA. 1988. 540/1-88/001.
SOURCES OF MODELS
AND MODEL INFORMATION
National Ground Water Association
6375 Riverside Drive
Dublin, Ohio 43017
614761-1711
1-800-551-7379
SOURCES OF MODELS
AND MODEL INFORMATION
International Groundwater Modeling Center (IGWMC)
Institute for Ground-Water Research and Education
Colorado School of Mines
Golden, Colorado 80401-1887
(303) 273-3103
9/93
17
Groundwater Flow Rates and Modeling
-------
Section 9
-------
PROBLEM 1
Flow Net Construction
-------
PROBLEM 1: FLOW NET CONSTRUCTION
GENERAL DISCUSSION
Groundwater-level data can be used to determine direction of groundwater flow by constructing
groundwater contour maps and flow nets. A minimum of three observation points are needed to
calculate a flow direction. The procedure is first to relate the groundwater field levels to a
common datum - map datum is usually best - and then accurately plot their position on a scale
plan, as in Figure 1. Next draw a pencil line between each of the observation points, and divide
each line into a number of short, equal lengths in proportion to the difference in elevation at each
end of the line.
The next step is to join points of equal height on each of the lines to form contour lines (lines of
equal head). Select a contour interval which is appropriate to the overall variation in water levels
in the study area. The direction of groundwater flow is at right angles to the contour lines from
points of higher head to points of lower head.
This simple procedure can be applied to a much larger number of water-level values to construct
a groundwater-level contour map such as the one in the example. First, locate the position of
each observation point on a base map of suitable scale, and write the water level against each
well's position. Study these water-level values to decide which contour lines would cross the
center of the map. Select one or two key contours to draw in first.
Once the contour map is complete, flow lines can be drawn by first dividing a selected contour
line into equal lengths. Flow lines are drawn at right angles from this contour, at each point
marked on it. The flow line are extended until the next contour line is intercepted, and are then
continued at right angles to this new contour line. Always select a contour which will enable you
to draw the flow lines in a downgradient direction.
PROBLEM 1: THE THREE-POINT PROBLEM
Groundwater-flow direction will be determined from water-level measurements made on three
wells at a site as depicted in Figure 1:
A. Given:
Well Number Head (meters)
1 26.26
2 26.20
3 26.08
B. Procedure
1. Select water-level elevations (head) for three wells as depicted in
Figure 1.
9/93 1 Flow Net Construction
-------
N
S>
a
1
WELL 2
( head, 26.20 m )
25 50
100
METERS
Figure 1
WELL1
( head, 26.28 m )
WELLS
( head, 26.08 m )
-------
2. Select the well with water-level elevation between the other wells (Well 2)
3. Draw a line between Wells 1 and 3. Note that somewhere between these
wells is a point, labeled A in Figure 2, where the water-level elevation at
this point is equal to Well 2 (26.20 m).
4. To determine the distance x from Well 1 to point A the following
equation must be solved (see Figures 3, 4, and 5):
X
5. Distance Y is measured directly from the map (200 m) on Figure 3.
H,, H2, and H3 represent head or water-level elevations from their
respectively numbered wells.
6. After the x distance is calculated, groundwater-flow direction based on the
water-level elevations can be constructed 90° to the line representing
equipotential elevation of 26.20 m as depicted on Figure 6.
C. Problem:
Based on the previous instructions for the three point solution use the three head
or water-level elevations depicted on Figure 7 and determine the groundwater-
flow direction for this site.
9/93 3 Flow Net Construction
-------
I
I
'
a
N
WELL1
( head, 26.28 m )
WELL 2
( head, 26.20 m )
Point A
1
0 25 50
100
WELLS
( head, 26.08 m )
METERS
Figure 2
-------
N
1
I
I
£>
WELL 2
( head, 26.20 m )
0 25 50
100
WELL1
( head, 26.28 m )
Point A
WELLS
( head, 26.08 m )
METERS
Figure 3
-------
I
I
S1
( 26.28 - 26.20 ) ( 26.28 - 26.08 )
X 200
X = 80
Figure 4
I
-------
vo
I
N
X = 80 m
WELL 2
( head, 26.20 m )
0 25 50 100
WELL1
( head, 26.28 m )
WELLS
( head, 26.08 m )
METERS
Figure 5
-------
s
I
a
oo
N
WELL 2
( head, 26.20 m )
0 25 50
100
METERS
WELL1
( head, 26.28 m )
Groundwater-Flow
Direction
WELL 3
( head, 26.08 m )
Figure 6
-------
500
320,
1 MILE
Figure 7
9/93 9 Flow Net Construction
-------
B
D
A '
Map
View
FIGURES
DEVELOPMENT OF CONTOUR LINES
Consider an island in a lake and the patterns made on it when the water level recedes. The
shoreline represents the same elevation all around the island and is thus a contour line (Figure
8A). Suppose that the water level of the lake drops 10 ft and that the position of the former
shoreline is marked by a gravel beach (Figure 8B). Now there are two contour lines, the new
lake level and the old stranded beach, each depicting accurately the shape of the island at these
two elevations. If the water level should continue to drop in increments of 10 ft, with each
shoreline being marked by a beach, additional contour lines would be formed (Figures 8C and
8D). A map of the raised beaches is in essence a contour map (Figure 8E), which represents
graphically the configuration of the island.
Flow Net Construction
10
9/93
-------
PROBLEM 2: EXAMPLE OF FLOW NETS AND HYDRAULIC GRADIENTS
Purpose
The exercise will employ basic principles defined in the determination of groundwater-flow
directions. Groundwater gradients (slope of the top of the groundwater table) will be calculated
as shown in the three-point problem (problem 1) in this problem.
Key Terms
head—The energy contained in a water mass produced by elevation, pressure and/or
velocity. It is a measure of the hydraulic potential due to pressure of the water column
above the point of measurement and height of the measurement point above datum which
is generally mean sea level. Head is usually expressed in feet or meters.
contour line—A line which represents the points of equal values, i.e. elevation,
concentration, etc.
equipotential line—A line which represents the points of equal head of groundwater in an
aquifer.
flow lines—Lines indicating the flow direction followed by groundwater toward points of
discharge. Flow lines are always perpendicular to equipotential lines. They also indicate
direction of maximum potential gradient.
Procedure
1. Review Figures 8-11.
2. Select an appropriate contour interval that fits the water-levels available and the size of
the map on Figure 9. (Twenty-foot contour intervals should be appropriate for this
problem.)
3. Draw the equipotential lines on the map (Figure 10) interpolating between water-level
measurements in a similar manner as in problem 1.
4. Construct flow lines perpendicular to the equipotential lines drawn in step 3 (see
Figure 11.)
9/93 11 Flow Net Construction
-------
101.9
96.2
99.6
e
102.0 Q
100.8
Scale: 1" = 425'
88.9
94.8
94.8
,91.0
99.1
102.4
101.9
101.8
FIGURE 9
WELL LOCATIONS AND HEAD MEASUREMENTS
100
101.9
®
102.0
100.8
Scale: 1" = 425'
88.9
FIGURE 10
EQUIPOTENTIAL LINES WITH WELL HEAD MEASUREMENTS
Flow Net Construction
12
9/93
-------
100'
88.9
101.9
FIGURE 11
FLOW LINES ADDED TO EQUIPOTENTIAL LINES AND
CALCULATION OF HYDRAULIC GRADIENT
5. The hydraulic gradient is calculated by measuring the scale distance between equipotential
lines along a flow line that crosses the site, and dividing that value into the calculated
change in head across the same distance (H2 - Hj)
#1 - H2 Atf
For example (see Figure 11):
Head at A = 100' (H,)
Head at B = 90' (H2)
Measured distance between the points is 1200' (L)
Head at point A minus head at point B divided by the distance between the points equals
hydraulic gradient (slope from point A to point B).
100 feet - 90 feet = 10
1200 feet 1200
= 8.3 x 10'3 feet/foot
Select a distance on your contour map between two contour lines and compute the
gradient.
9/93
13
Flow Net Construction
-------
PROBLEM 3: FLOW NET CONSTRUCTION
After completing the contour map in Problem 2, plot a profile of the site's groundwater surface
at Y-Y' on Figure 12.
Procedure
Vertically project the contour lines that intersect line Y - Y' on map in Figure 12,
to the dashed line labeled Y-Y' below this map and above the graph for the
site's profile.
Be sure to label each mark on this line with it's respective elevation.
Plot each elevation point on the graph using the vertical scale given.
Connect the points on the graph. You have now constructed the site's
groundwater-surface profile. This profile also represents a cross section.
Flow Net Construction 14 9/93
-------
400
'/32°
Y
380
380
520
400 • 360 N
360*
g, 360
500
- - 300
360
1 MILE
420
Y
*320
.420 X
400.
360
400
420
'420
Figure 12
Y
r
500
400
LJJ
LLJ
300
9/93
15
Flow Net Construction
-------
PROBLEM 4: FLOW NET CONSTRUCTION
Site Name: Bakers Quarry
Location: Tippersville, Maine
SITE HISTORY/OPERATION
The quarry operation began in 1905 providing construction grade granite locally and was closed
in 1928 when the volumes of groundwater seeping into the pit made it economically unfeasible to
continue mining (Figure 13). The site was abandoned and the pit filled with water. The owners
of the quarry declared bankruptcy and ownership fell to the city of Tippersville in lieu of
delinquent tax payments.
The quarry was used as a swimming hole and occasional dump site for locals until 1958, when
several children drowned. The site was fenced and patrolled to prevent swimming. Uncontrolled
dumping by individuals and local industry increased dramatically with the swimming ban.
Dumping took place around the rim of the quarry and periodically the bulldozer from the town
landfill was used to push material into the pit. Gradually the pit was filled and several fires
forced the town to terminate dumping in 1971. The surface of the site was covered with local
material, primarily sand and gravel.
The site gained notoriety when an area-wide survey identified it as a potential industrial dump
site. A preliminary site investigation, started on 4/14/82, including sampling a spring located
approximately 25 feet from the limits of quarrying. Priority pollutant analysis of this sample
identified ppm levels of polychlorinated biphenyls and trichloroethylene. Results from this
preliminary investigation were used to justify a more extensive hydrogeologic study of the site.
ELEMENTS OF THE HYDROGEOLOGIC INVESTIGATION
The first step of this investigation was to do a literature review of geologic information. A
discussion with a local amateur geologist revealed a paper from a geologic investigation
performed during active quarrying. Information from this study and observations at an outcrop
on-site provided a geologic background for the investigation. The quarry material is a slightly
gneissoid biotite-muscovite granite. Several dikes were identified in the quarry wall.
The probable high permeability and infiltration rate of the less consolidated waste material
compared to that of the granite, could cause groundwater mounding in the pit area. Potential
mounding, and inadequate information about groundwater flow direction dictated a ringing of the
site with monitoring wells.
Twenty-two monitoring wells were planned and installed at the site from 10/1 to 11/14/82.
Eleven were installed in bedrock, the unconsolidated zone sealed with steel casing and grouted.
Eleven monitoring wells were installed in the unconsolidated heavily weathered bedrock or
unconsolidated zones. For the purpose of this problem set you will only be using data from the
eleven wells listed in Table 1. An explanation of this data is depicted in Figure 14.
9/93 17 Flow Net Construction
-------
400'
800' 1200'
Site Boundary
FIGURE 13
SITE MAP - BAKERS QUARRY, TIPPERSVILLE, MAINE
Flow Net Construction
18
9/93
-------
TABLE 1
MONITORING WELL DATA
WELL
NUMBER
MW 1
MW2
MW3
MW4
MW5
MW6
MW7
MW 8
MW9
MW 10
MW 11
(a)
TOP OF
CASING
ELEV.
(feet)*
87.29
89.94
88.04
82.50
82.50
72.50
80.58
86.03
114.01
108.67
105.07
(b)
GROUND
SURFACE
(GS)
ELEV.
(feet)*
84.79
87.99
85.44
79.80
80.05
69.50
78.28
83.53
111.21
106.67
103.37
(c)
GROUND-
WATER
ELEV.
(feet)*
80.49
84.69
75.29
72.40
73.40
67.50
74.78
76.93
92.36
93.97
94.97
(d)
WELL
DEPTH
(feet
below GS)
151.9
103.05
103.1
102.3
102.45
99.6
99.5
99.2
99.9
98.7
102.1
BOTTOM
OF WELL
ELEV.
(feet)*
-67.11
-15.06
-17.66
-22.50
-22.40
-30.10
-21.22
-15.67
11.31
7.97
1.27
(e)
BED-
ROCK
DEPTH
(feet
below GS)
7.5
7.5
2.0
14.0
8.5
9.0
8.0
8.5
10.5
10.8
2.5
* Datum: mean sea level
9/93
19
Flow Net Construction
-------
MW 1
ca;>
87.29
84.79
151.9
7.5
Bedrock
80.49
Datum Csea I eve I
Top of casing elevation Cfeet}
Ground surface elevation Cfeet}
Groundwater elevation Cfeet}
Wei I depth below ground surface
Bedrock depth
FIGURE 14
MONITORING WELL ELEVATIONS
F/ow Net Construction
20
P/P3
-------
DEVELOPMENT OF SITE PROFILES
Purpose
The development and comparison of topographic profiles across the site will help the student to
understand the variability of the surface terrain usually found on most of the larger sites. The
water-table profile will also be constructed. These profiles will be useful during the next
problem, "Flow Nets and Determination of Hydraulic Gradients."
Procedure
1. To construct cross-section lines, lay the edge of a piece of paper along the cross-section
line selected and draw a straight line. Mark the location of the monitoring wells along
the edge of the paper. (The placement of some wells may need to be projected, as not all
of the wells lie along a straight line.)
A - A' MW 9, MW 2, and MW 4 (in that order)
B-B' MW 1, MW8, andMW7
C - C' MW 11, MW 3, MW 7, and MW 5
NOTE: One should be aware that projection of wells to a cross-section line could cause
distortions which might affect interpretation of the distribution of subsurface geology or soil.
2. Using the graph paper provided, transfer these well locations to the bottom of the page
along the horizontal axis.
3. The vertical axis will represent elevation in feet. Mark off the elevations in ten foot
increments. Each division of the graph will represent an elevation increase of two feet.
4. Graph the ground surface elevation for each of the chosen monitoring wells. (This
information is found in the monitoring well data, Table 1.)
5. Graph the groundwater elevations for these same locations.
6. Repeat this procedure for the other cross-sections lines.
7. Compare the topographic profile to the water table-profile. Are they identical? After
looking at this data, are there any conclusions that can be drawn?
9/93 21 Flow Net Construction
-------
PROBLEM 2
Geologic Cross-Section Construction
-------
PROBLEM 2: GEOLOGIC CROSS-SECTION CONSTRUCTION
Materials necessary to construct cross section:
Graph paper. Use vertical scale of one graph paper division per 20 feet.
Colored geologic map from geologic report that covers the area surrounding the
location of the cross section. The cross section is labeled on the map as A - A'.
Description of major geologic units from a geologic report.
Six water well logs which include a driller's description of the sediments and rocks
encountered during the drilling of each well. Location of the wells are depicted on
the geologic map and the following topographic quadrangle map. These logs will be
used to construct the geologic cross section.
Topographic quadrangle map depicting the ground surface elevation along this cross
section. This map will be used to construct the surface profile at the cross section.
Ruler and colored pencils will be supplied upon request.
Procedure to follow in the construction of this cross section:
• The instructor will divide the students into groups after discussing the materials
included with this exercise.
• Each group will first select an individual to construct a common surface profile that
the other students will use to put their soil boring interpretation for each well on the
cross section. This person will use the same procedure employed earlier to construct
the profile in the flow net exercise.
• Each group will then determine who will evaluate the soil boring information from
each well log and define which major geologic units are present. This interpretation
will be based on the geologic information already provided. Hint: The geologic map
indicates the major unit at the top of each soil boring.
• When the major unit intervals are defined on each boring, this information will be
constructed on the graph paper provided using the vertical scale given above. It is
suggested that one use the same symbols as those used to describe the major geologic
units in the geologic report.
• After the surface profile is completed, the constructed boring log for each well will
be transposed onto this profile. This is done by placing the top of each constructed
well log at its appropriate surface location on the profile. This will eliminate the need
9/93 1 Cross-Section Construction
-------
of calculating the elevation for the top of each major geologic unit found in each
boring.
• Once this is accomplished each group will attempt to correlate these major geologic
units between the six soil borings.
Cross^Section Construction 2 9/93
-------
Depth of Soil
Boring
0 - 5.0 ft
5.0 - 10.0 ft
10.0 - 15.0 ft
1 C A 1 7 C fi.
Ij.U - 1 1 .J It
17.5 - 20.0 ft
20.0 - 22.5 ft
Lithologic
Description
Tan, silty clay
(SC)
Light brown,
sandy silt
(SS)
Brown fine-grain
_ rt%- j
sand
/"CC\
(F5>)
Dark brown,
coarse-grain
sand
(CS)
Reddish brown
gravel
(G)
Red Clay
(C)
©
x_x
•s
cl
&
0.0
-
5.0
-
100
-
15.0
20.0
-
Boring Log
Example
#1
• •
.
.
. .
t t
• • •
o o o
e o o o
o o o o
... *
00_0000000000
•
QoQoooQooo
o o o°°o o°°°
° O O O O O O O
OOOOOOOO
OOOOOOOO
OOOOOOOO
OOOOOOOO
Boring Log
Example
#2
SC
SS
FS
CS
G
c
9/93
Cross-Section Construction
-------
8
S?
CLAY PLUG
s
a
1'
HORIZONTAL
BEDDING
SCOUR AND FILL
PREVIOUS _
FLOOD PLANE
VARIOUS
X-BEDDING
STYLES
PLANAR
X-BEDDING
TROUGH X-BEDDING
MASSIVE
GRAVEL BED
CHANNEL LAG
-------
GRAIN SIZE DECREASES
FINING UPWARDS SEQUENCE
POOR
SORTING
GOOD
STREAM VELOCITY DECREASES
9/93
Cross-Section Construction
-------
DESCRIPTION OF MAP UNITS
Symbol Description
Quaternary Sediments
Qa Alluvium or stream deposits(Holocene)—Composed of silt-, sand-, and
gravel-size sediment. These deposits are found in floodplains, river terraces,
and valley bottoms. The vertical sequence of sediment grain size decreases
from bottom to top. Locally the unit includes lacustrine (lake) and paludal
(swamp or marsh) clays and silts, eolian (wind blown) sand deposits in
depressions.
Qfg Flood deposits (Pleistocene)—Poorly sorted, stratified mixture of boulders,
cobbles, gravel, and sand resulting from multiple episodes of catastrophic
outbursts from glacial-dammed lakes, such as glacial Lake Missoula. The
Little Spokane River valley was one of the main channelways for outburst
flood waters from this ancient lake. Each flood event is represented by a
sequence of sediment grain size decreasing from bottom to top.
Qls Alluvial fan deposits (Pleistocene)—Composed of unstratified and poorly
sorted (heterogeneous and anisotropic) clay-, silt-,sand-, and gravel-size
sediment. Some fan deposits contain large blocks of rock as much as 8 meters
in diameter.
Ql Loess (Pleistocene)—Composed of light- to medium-brown, unstratified eolian
particles of clay, silt, sand, and volcanic ash. The loess mantles the Columbia
River basalt and is most commonly found on the tops of low hills and
plateaus where erosion by water has been minimal.
Qglf Lacustrine and flood deposits (Pleistocene)—Composed of light-gray, friable,
sediment of clay, silt, and fine sand near the top of this sequence of deposits.
This sediment overlies flood deposits composed of stratified mixtures of
boulders, cobbles, gravel, and sand. Each flood deposit is represented by a
sequence of sediment grain size decreasing from bottom to top.
Columbia River Basalt Group—Tertiary (Miocene)
Mvwp & MvgN2 Wanapum basalt flows and upper flows of the Grande Ronde basalt.
Me, Latah Formation—Gray to tan to yellow-orange siltstone, claystone, and
minor sandstone of lacustrine and fluvial depositional environments.
Intrusive Igneous Rock —Cretaceous
Kiat. The symbol represents all Cretaceous and Tertiary-age intrusive igneous rocks
including the Mount Spokane granite.
Cross-Section Construction 6 9/93
-------
Figure 1 Regional Geologic Map
Mead Landfill Site
...).ooqp_
fEET
15000 20000 25000
-------
WATER WELL REPORT
BTATE Or WASHINGTON
I) OWNER:
Addr
2) LOCATION OF WELL: county _..2.£Q..Kj3
and dULanc* from McUon or vubdlvUlon corner
(3) PROPOSED USE; Dom«*Uc Jg' Industrial Q Municipal D
Irritation Q Ta«l Well Q Other D
(4) TYPE OF WORK: ft^Vr'e n^"n°t( wt" 1
New well "^ Method: Dux Q Bored Q
D**p«n*d D Cable |J Driven D
Reconditioned Q ftlr RoUry^J Jetted Q
(5) DIMENSIONS: Diameter of well f"j. Q icchca.
Drllltil /PO n Depth of completed «-f II . »O , ,,ft.
(6) CONSTRUCTION DETAILS:
Casing installed: *T>0 - DUO,. fjom ._.^.._.. ft. to „.£.!?-_ ft.
Perforations: Y«* {3. No'rf
Typ>* of ptrforMor u**d , ,
,, , p-rfQrr»H«"* from ft la ft
Screens: yM Q NO Q ^-~ / / \ f( <
tin ^ <^L^ Q /^»
Typ«.T.y..=£. - »*P J5-... „_ ft.
UAtcria] u*ed in **»J C^^ ^*^t Q^f& ^*/' . ..
Did *Jiy ruiLi conUin unut»ble water) Yes D No ££
Typ* of water? , D*pth o' « 5"
/8^ WATER LEVEL*S* l-*nd-«uri»ce «Jrvatlon ft? '/' ^ •
Sutic Uvtl L/ ..£__. ft. b«low lop of wtU D»le
ArtxiJan water la controlled by
(Cap, vaJve. etc.)
/Q\ IXTFI 7 TFCTQ- Drawdown U amount water level la
\U) rrr.l-1-i irou. lowered below italic level
Wai a pump Uil made? Yej Q No ft U ym. by whomt
Yield: 2O fal./min. with ft. drawdown after oO (Hvv'^OT.
... -
_
Aecovtry data atf
(10) WELL LOG:
Formation: Dejcribe by color, character, nit of material and xructurc. one
Ihou Ihickneai ot aqui/en and the kind and nature o/ the matrnal m rarh
atralum penetrated, with at Uajl one entry for inch chanpc o( formation
MATTJUAL
^a ^.A • Coovg/r <^v« l*i
^I'lj-sJ-tm*. vA^erliieAJitA i^j/^-l^^i, h* &
bJ-fc i^-tv^24*,- -SairtpLe _ > | ^7 crl rCfcC. ^-e
.
,r _
TROM
•v
, 5'
10
"2-O
4o
TO
5~'
/6'
2or
4 o'
Jerz '
Work runed ^ ^^ /^ . IB ~Ll. Complrled CTfef.S../..S.-... !»..(»./..
WELL DRILLER'S STATEMENT:
Thli wc.ll w»i drilled under my jurisdiction and this report is
true to the but of my knowledge and belief.
NAME _ _
(Panon. firm, er corporaUon) (Trpc or pnni)
AddrtM -,
(Will Driller)
Mcr.n*e No .....,,..., Dale i '®
(USt ADDITIONAL SHZTTS ET NTCISSARY)
-------
WATER WELL REPORT
STATE Or WASHINGTON
1) OWNER:
Addr
2) LOCATION OF WELL: county
Bearing and dliLance from Mellon or aubdlvuton corner
JAiii— - .H(iLit .tikl-i s.cJ7.... T.2.7....N.. R.44 J=j*
(3) PROPOSED USE: Domestic $_ Industrial D Municipal D
IrrifaUon Q Te«t WeU D Other Q
t4) TVPF OF WORK' Owi.er'i number of well 2.
(i) i xrt or YYun.iv. (lf more Uu) onel ^._
N«w well ]S£ Method: Dux D Bored D
Deepened Q , Cable D Driven D
Reconditioned D A* *~ Rotary & Jetted Q
(5) DIMENSIONS:
Drilled ?..3?. -..n.
Diameter of well P- ^ — Inchea.
Depth of completed well j2.2& ft.
(6) CONSTRUCTION DETAILS:
Casing installed: ^.0-
Ttu-eaded Q
Welded D
" DlAm. from ft. to
" Dl*m. from ft. to
*?— ft.
ft.
ft.
Perforations: y« Q N,
Tytx of perforator uud_
SIZE of perforation! —
In. by
pcrforatloni from •_>.—. ft. to
perforation* from — ft. to
. pvriorattani (ram ft. to
_ In.
ft.
ft.
ft.
Screens:
Htnaj
Typ«..X v
Diam. .A'.
Dlim
NO D
/.. ,i
UJe,\\
del
Slot Hie ...iCL_ trom .%.?..0... ft. to -2..S5. ft.
Slot lUe from ft to ft.
/-, i i j
Gravel packed:
NO 0 su« of
Crkvel pliced iiom ---------- ........ ---- ft. to ..........
/O—ZQ
________ ;__ .....
..... — ft.
Surface seal: Y«J^ NO Q TO what depth) ,.- ft.
Material Uied In «tl C*iKM»A«a.r.....&jKiC«i..i:
Did any ktrata contain unusable water) Yea Q No Q
Type of water? Depth of atrata
Method of ***lini rtr»u off
PUMP: Man
Type: —
lfe!_^.tf
HJ>_-/-V_..
Uand-iurface elevation
above mean a«a level..
SUlic level -[&{.X2 ft. below top of well Dale
ArttJLlaa preuure lb*. per e
Hecovtrr i1"' thickncu o/ acjui/tri and the kiixxf and nature o/ the matrnal in tarh
ttratum perijtrated. with at leajt onr entry /or each chonor o/ /ormation.
MATE7UAL I TROM \ TO
'Jit. "S g c
C-^^'^J
2
O
4-
^
n
/o
7 n
#0
Work mrted
. Computed ..... >(S<3.a.?..0.... 1B.6-3
WELL DRILLER'S STATEMENT:
ThU well w»* drilled under my jurisdiction and this report U
true to the best of my knowledge ind belief.
NAME..
(Pir»OD. firm, or corporaUon)
(Type or pnnt)
Wu a chemical anaJjrala made) Yee Q No Q
(USr ADDITIONAL KHTTTS IT NICtSSAnY)
(Weil Driller)
Liceni* No_ Date
IB..
ICY OMM-X)
-------
WATER WELL REPORT
STATE OF WASHINGTON
1) OWNER: N.m.,._:
Addrcta,
2) LOCATION OF WELL:
Hint and distance from a*ctlon or subdivision corner
>— »i<*U ..^v, s.e_../Z T.2.7.N.. R..43 J^.
(3) PROPOSED USE: Domestic ^gf ladunrl»l D Municipal D
Irrigation Q T««t Well Q Other D
(4) TYPE OF WORK: ^rnVre uinbone°l' *'" ^
New well & Uclhod: Due Q Bored Q
D»ep«ne |Y f~ <:r \ ^ —.
nirnm 4.^"cifit fir, 2*O tnm 2^0 *, 10 3/0 f,
Gravel packed: Ye»#^ No 0 Siir of travel :'.*!
Did usy ruit* conLun uauuble water? Ye* Q No Q
Typ« of wiier? D«pih of *1r»tJ
s~ i r-
/7\ PUMP- w -.,,,.,,,, .•. vj. ~. Ll^~ r IA (A c\ "r~ O JS
(*8^ WATER LEVELS' I-and-»urface elevation / Cj 3Q
Static level -' *£. ......... .ft. b«low top of well Date _
Arvrilan water U controlled by
(Cap, valve, etc.)
/O^ XWT T 'T*PQT^- Drawdown U amount water level lj
\V) rrr.J^Jj AC-31 J. lowered b«low rtatlc level
Waj a pump UH made) Y« Q No [S( If ye«. by whom?
Yield: oi f7 I al./irvln. with ft. drawdown afur J? , O hn.
.
... .
Recover? dii» end the kiiuf and naiurr of the matrnal in 'tact.
ttratum p*rvelralcd. unh at leojt on« entrv /or »orh chonpc of formation
MATTIUAL
^>VfiA->-dL ^ Cp civ \e 'S&v rf
#£****« ASaloose.
i
k>r>Li\ tjf
/ »J-J -i >/^
Jya^Lfi i av/n*\\T-c uJI
/ CLt*d <^A * fJlo^J,
*l»~~ . *
^_
£^*(-rte^a*l^cL^^hr*ij^&y^ j-p'
^tre^j-^. ^ L&
\/es<<**-l0v tk? So.
<7/l/f2w ^ ^
TROM
b
l£>
_2O
3o
40
S'O
bo
•70
tyo
Jon
1 16
/ j £^>
td-(\
I$T>
/&&
J-bo
2^O
TO
'o
2o
So
4o
So
to
r?O
fo
f^C~O
I'O
1^0
/
A5b
/C»o
^fof~>
2*?£>
5.5 o
-^ "P IP^^ Complftfd JH>^-30 , )o rX;
WELL DRILLER'S STATEMENT:
Thli well wu drilled under my Jurisdiction and thLi report U
true to the but of my knowledge and belief.
NAME _ ....
(Pej
Addrr"
Ucrnn No
»on. ftrm, or corporation) (Type or pnni)
(Well Driller)
Dale 18
(USr ADDITIONAL SHITTS D* NTCrSSARY)
ICY o&o-i-rc
-------
WATER WELL REPORT
STATE Or WASHINGTON
1) OWNER:
Addr
2) LOCATION OF WFT.I.; county 'Sfi&K
Be^nnf and dliLanct from McUon or subdivision corner
(3) PROPOSED USE; Dom«tic hjf IndurtrlaJ D Municipal Q
f*~
Irritation D T««t Well Q Other D
(4) TYPE OF WORK: '$"££
N«w well ££.
• D*«p«n*d D
Reconditioned Q
"''
4 _
Method: Dux D Bored D
Cable D Driven D
Al'/RoUry^ J«tted D
(5) DIMENSIONS:
Drilled _..ft.
of well S-'-:Zi Inchea.
Depth of completed w»U n.
.!?...
.7*.
(6) CONSTRUCTION DETAILS:
Casing installed: A.O - num. from .-.±S. n. to r.l/.^- ft.
Threaded D —" Dlain. from ft. U> n.
WtJded D " DUm. from ft. to ft.
Perforations: yM Q^ No 5£
Type of perforator ua*d .—- .....
SIZE of perforation! .-
In. by
perforation* from i_>— ft. to
— p«r
HJ>
_,
-------
WATER WELL REPORT
STATE Or WASHINGTON
1) OWNER: N.m._
Addr
2) LOCATION OF WELL: c«umy ......
*nd dUUtnct from xcllon or »ubdlvU.on corner
5EJ:£j£lL- S.£.Vt .lLtt s.,_..^.... T. 2.7 N .. „ .43
(3) PROPOSED USE: Domestic ^laduttrUl D Municipal D
Irrigation D Te*t WeU D Other D
(4) TYPE OF WORK: ^"rnVr. u^boncV *"" -^~
New well ^H Method: Dur D Bored D
D«*p«ne-tll itf& ft,
(6) CONSTRUCTION DETAILS:
Casing miLQlled: ^•*~f - r^nm fn»rn *^ #t ta r'— '**: «
Threaded Q „, " DUm. Irom ft. lo . ft.
Perforations: YM Q. NO {^
Typ« of perforator u**d™™ ....
SIZE of fttrtoraUonj .— in. by . — In.
perforations from .-- ft. to ._...* . .. ft
p^ff ijfptinm from ft to f*
Screens: YU^ NO n ,-p i (A) (\ <
Minvficl'irrr'i N"*>» , >J ^ ^ ^So*^ \A^'£, (. | -Jcy^^*i^
«- T^iy/ <,(_»-///!• fio i. ji».
Typ«.-.~Jt..tt_5LJkitJi;iim .E,.>ri. . Model No
Gravel packed: Ye«^ NO 0 si« of travel:
Criivel placed from ft. to _ — ft.
„ , ,
Mpttria) vucd In teaJ CX^yL^r^A
Did JLoy rtriU conUan unuuible water? Yei Q No Q
Typ« of water? . D«pth of straU....
(7) PUMP- MinjUifiu-r1' "-"ii ^-^U Hlt HP /£?
(8) WATER LEVELS: ^v"."^^' JiT^'el". . . _/Je&CL_ft.
Sutic level 1^2^..Ll. ft Kelnw lop of well Hate
AnciJ&n water ii controlled by -
(Cap. vaJve. etc.)
fQ\ XXTFT T TTP^TQ- Drawdown U amount water level Ij
V»; vrr.i-.Li at-aio. lowered b*low rtauc level
Waj « pump Uri roade? Yej Q No tt If ye«. by whom?
Yluld: 1 S"O lal./mln. with ft. drawdown afur { .0 hri.
.. .
.. .
Recovery data (Urae tAkea u uro when pump turned off) (water UveJ
measured from well top to water level)
Time Woter L-tvt\ Time Water I-rv«l Time Water L*v«l
Date of l«n
(10) WELL LOG:
Tormatlon: Describe by color, character, rvri of malerval and structure one
•hou thickneiJ o/ aqui/rri and the kind and nature o/ thr matrnal in tact.
Uratum p*Tvetrotcd. uiith at leajt one entrv /or each chonpc o/ formation
MATXJUAL
•*>SJU- Ost+ft fO+iA
t "^&4*-Q £*V\.C\ d**t*\
^%64<& QAA A ."5*^v '-^-'/ sifn** *5^^v^r^-Y ^
f\e^.^ a^J r^.^- i^JLa-^- Irso^fs^
$ c^oltfS q~y~*~~
^,^t^^c a^ a kr-~-z.
•^AXU^^ a <; a la+-<*+-JL, y.
*(yva*~/ 'cyv si»e •£i*7c-/t..»v 5
A /*y /
PM J&v*. s
TROM
'o
fo
20
3>&
4(j
S~o
(f O
?£>
8O
t%0
loo
no
/2o
>3f>
/4o
f-so
l(aO
77o
l&o
2.^0
TO
(0
2<3
3 <*
4-0
<-0
Cxp
?o
So
9-D
/<9T3
IfO
/20
I3o
I4o
ISo
• c^>^^
f f £?
XSC3^
_2^D
_p ^
-------
WATER WELL REPORT
STATE Or WASHINGTON
DOWNER:
2) LOCATION OF WELL; c«
Bt-.nni and diiimce from atcUon or mbdlvUlon corner
mlr ............ ..
L- -tj£_i<. JfcULy. *•*-
.- T.. ...*..
(3) PROPOSED USE: Domestic ^ Industrial D Municipal Q
Irritation D T«*t Well D Other D
(4) TYPE OF WORK: ^fwmc0rrtt l^""n°' wt" *O
Ntw well £&. Method: Due Q Bored D
D«p«ne-' t t S •? *
•f* "P I/ / *Cx /» ft <^> •> . j 1 XI
Typ< — IT-i ,fctr— S?fi(l.._Jat-> .. Model No_
ni»r-i Slnt ilrf .. from ,,, , fl, In ,, , , . fL
Gravel packed: Yeilal^ NO Q sue of »r«vei: ..
Did any nrata contain unusable water} Ytj Q • No Q
Typ« oY water? „ ™_ D«plh of ytr«L»
/^ J P ^
(7) PUMP- Hf«i"»"*"«'-'- Wf^i fcVwMq-Vr>5
w*--^ . So< w I^VA tfv s i «9 1* up .5^
^8^ \VATER LEVELS' X-*nd-§urf»ce eJrvatlon .X'lOOP)
^ ' //9>f/x * abovt mean Ma level — -L^:: — Li_^t.
SUtiC .rvtl '.'5- - '"'/ «- t**lnw inp nf w»1) n»t*
(Cap. vaJve. etc.)
/Q^ TXTT T TPCTC;- Drawdown U amount water level lj
\V) rr£0-.lj ICOIO. lowered b«low n«Uc level
Wai a pump l«t roade) Yti Q No 0 If yam. by whom)
Yield: 2L .D »al./mln. wllh ft. drawdown afur /, O hri.
... -
.. .
Hecoviry o-*ti -rC«r^ ^ ' / ^^1 ^
l
-fv^trVw'^c Q Ll'nst^ OL^^-C.1
t
TROM
O
*?&
"B/o
TO
0
2%o
J/&
3 So
i
Work rtantd KT^-^- '7 ,10 -52. Complrttd..3L'ti.^..
-------
PROBLEM 3
Aquifer Tests
-------
NOTES
AQUIFER TESTS
GROUNDWATER AND
CONTAMINANT MOVEMENT
• Position and thickness of aquifers and
aquitards
• Transmissivity and storage coefficient
• Hydraulic characteristics of aquitard
• Position and nature of boundaries
• Location and amounts of groundwater
withdrawals
• Locations, kinds, and amounts of pollutants
AQUIFER RESPONSE DEPENDS ON:
• Rate of expansion of cone of depression
- Transmissivity of aquifer
- Storage coefficient of aquifer
• Distance to boundaries
- Recharge
- Impermeable
9/93
Aquifer Tests
-------
NOTES
Limits of cone
of depressioi
Land surface.
/v
Cone of |
depression ' 1
Flow lines
/'
Aquiclude
Unconfined Aquifer
Limits of c
of depres*
one Land surface -~^_
i'lon\s' Potentiometr
Drawdown — ">
:::::: Aquiclude : "
- -:-:-:|
ic
Q
t
A
,.-
surface ^^ ^s.
•-v N
x \
. ^\Cone of
depression
'.'..'. .'.'.'i
Aquiclude < •.•.•,.'.;.,:
Confined Aquifer
AQUIFER TEST METHODS
Step drawdown/well recovery tests
Slug tests
Distance-drawdown tests
Time-drawdown tests
Aquifer Tests
9/93
-------
NOTES
STEP DRAWDOWN
Well Recovery Tests
Well is pumped at several successively higher
rates and drawdown is recorded
Purpose
- Estimate transmissivity
- Select optimum pump rate for aquifer tests
- Identify hydraulically connected wells
Advantages
- Short time required
- One well required
SLUG TESTS
Water level is abruptly raised or lowered
Used in low yield aquifers (<0.01 cm/s)
SLUG TESTS
Advantages
• Can use small-diameter well
• No pumping - no discharge
• Inexpensive - less equipment required
• Estimates made in situ
• Interpretation/reporting time shortened
9/93
Aquifer Tests
-------
NOTES
SLUG TESTS
Disadvantages
Very small volume of aquifer tested
Only apply to low conductivities (0.0000001 to
0.01 cm/s)
Transmissivity and conductivity only estimates
Not applicable to large-diameter wells
Large errors if well not properly developed
Do not give storativity
DISTANCE-DRAWDOWN TESTS
Advantages
• Can also use time-drawdown
• Results more accurate than single well test
• Represent more of aquifer
• Can locate boundary effects
DISTANCE-DRAWDOWN TESTS
Disadvantages
• Requires multiple piezometers or
monitoring wells (at least 3 wells)
• More expensive than single well test
• Must handle discharge water
• Requires conductivities above 0.01 cm/s
Aquifer Tests
9/93
-------
NOTES
TIME-DRAWDOWN TESTS
Advantages
Only one well required
Tests larger aquifer volume than slug test
Less expensive than multiple-well test
TIME-DRAWDOWN TESTS
Disadvantages
• Pump turbulence may interfere with
water-level measurements
• tests smaller aquifer volume than
multiple-well test
• Must handle discharge water
• Requires conductivities above 0.01 cm/s
THEIS METHOD
First formula for unsteady-state flow
- Time factor
- Storativity
Derived from analogy between
groundwater flow and heat flow
Laborious method
- Log-log paper
- Curve matching
More accurate than Jacob method
9/93
Aquifer Tests
-------
NOTES
THEIS'S ASSUMPTIONS
•
•
•
•
•
•
•
Aquifer is confined
Aquifer has infinite areal extent
Aquifer is homogeneous and isotropic
Piezometric surface is horizontal
Carefully controlled constant pump
rate
Well penetrates aquifer entirely
Flow to well is in unsteady state
-
i
I
^
Potentiometric ^
^^^'surface Q
.__. _ . . ^. .
Xt"
Drawdown N T
H CorTfjning layer jj$il WSiL
(Confined aquiferj ~- r
Confining layer
1
1
, .L
Cone of
depression
*. * ^
—
•
ynj
m
THEIS EQUATION
T = transmissivity
_ QW(u) Q _ djscharge (pumping
4TS
W(u) = well function
s = d
_ 4Ttu s „
S - r2 bo
t = tir
rawdown
torage coefficient
ne
rate)
r = radial distance
Aquifer Tests
9/93
-------
NOTES
WELL FUNCTION - W(u)
W(u) = -0.577216-l
and u =
4Tt
S = storage coefficient
t = time
r = distance
T = transmissivity
W(u) is an infinite exponential series and cannot
be solved directly
JACOB METHOD
• Somewhat more convenient than Theis's
method
- Semilogarithmic paper
- Straight line plot
- Eliminates need to solve well function
W(u)
- No curve matching
• Applicable to:
- Zone of steady-shape
- Entire zone if steady-state
JACOB'S FORMULA
T = transmissivity (ft2/day)
T = Q = pump rate (ft3/min)
As = change in drawdown (ft/log cycle)
_ 2.3 Q _ 2.3 gal_ 1.440min ft
~ X X X
_
4TAs ~ 4T rnn day 7.48 gal X ft"
_ 35Q
~ As
so that Q is now expressed in units of in gallons per minute
9/93
Aquifer Tests
-------
NOTES
JACOB DERIVATION
35Q K_l
T~ As b
T = transfnissivity is square feet per day
Q = pump rate in gallons per minute •
As = change in drawdown in feet
over one log cycle
K = hydraulic conductivity in feet per day
b = aquifer thickness in feet
Land surface
\
Cone of depression — — "^J
(unsteady shape)
/ • . -3;
3
River
\ TT /
^ \iV
Confining layer i"i|^&/' : " '|
'4r— ' — ^
^-^ .->
3
r River
\ T7 /
J^" \ /
''.".' . . :' . .:..U(' / i ]
4— 4- , ,%*».-•!
4—4— • ; V» " <
(2)
NONEQUILIBRIUM
Aquifer Tests
9/93
-------
NOTES
(3)
EQUILIBRIUM
9/93
Aquifer Tests
-------
TABLE 1
PUMPING TEST DATA
Q = 109 GPM
Pumping Time
(minutes)
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
18
20
22
24
26
28
32
35
40
45
50
55
60
90
120
b = 20 feet
Drawdown measured from top of casing
(feet)
6.1
6.5
7.5
8.0
8.6
9.5
10.5
11.2
12.0
13.0
14.0
15.5
17.0
18.0
19.3
20.5
23.5
25.2
26.7
28.2
29.5
30.5
32.0
34.5
36.6
38.5
40.5
42.0
43.5
50.1
54.8
9/93
11
Aquifer Tests
-------
PROBLEM 4
Groundwater Investigation
-------
PROBLEM 4: GROUNDWATER INVESTIGATION
LEAVINGS RESIDENCE
On October 12, 1982, the Bettendorf, Iowa, Fire Department was called to the Leavings residence
with complaints of gasoline vapors in the basement of the home.
October 16, 1982, the Leavings were required to evacuate their home on an indefinite basis until the
residence could be made safe for habitation.
PERTINENT KNOWN FACTS
The site of contamination is a residential neighborhood in Bettendorf, Iowa.
It backs on commercially zoned property, which has only been partially developed to date. The
residential area is about 10 years old, and contains homes in the $40,000 to $70,000 range. There
was apparently some cutting and filling activity at the time it was developed.
Within 1/4 mile to the northwest and southwest, eleven reported underground storage tanks are in
use, or have only recently been abandoned.
• 1000 feet northwest to two tanks owned and operated by the Iowa Department of
Transportation (IDOT)
• 700 feet southwest to three in-place tanks owned initially by Continental Oil and now
by U-Haul. One reportedly leaked. (Bettendorf Fire Department (BFD))
• 1200 feet southwest to three tanks owned and operated by an Amoco service station,
no reported leaks (BFD)
• 1200 feet southwest to three tanks owned and operated by a Mobil Oil service station,
no reported leaks (BFD)
Adjoining neighbors have complained about several trees dying at the back of their lots (8 and 10).
No previous problem of gasoline vapors reported at these locations.
The general geologic setting is Wisconsin age loess soils mantling Kansan and Nebraskan age glacial
till. Valleys may expose the till surface on the side slope. Valley soils typically consist of the
colluvial and alluvial silts.
Your previous experience in this area includes a geotechnical investigation of the hotel complex
located west of Utica Ridge Road and northwest of the Amoco service station. Loess soils ranged
from twenty-two feet thick on the higher elevations of the property (western half) to 10 feet thick
on the side slope. Some silt fill was noted (five to seven feet) at the east end of the hotel property.
Loess soils were underlain by a gray lean clay glacial till which apparently had groundwater perched
9/93 1 Groundwater Investigation
-------
on it. Groundwater was typically within ten to fifteen feet of ground surface. This investigation was
performed eight years ago and nothing in the boring logs noted hydrocarbon vapors observed. It
should be noted that these observations were not routinely reported at that time.
Other projects in the area include a maintenance yard pavement design and construction phase testing
project at the IDOT facility located northwest of the Leavings residence. Loess soils were also
encountered in the shallow pavement sub-grade project completed three years ago. It was noted in
the firm's records that the facility manager had a minor gasoline spill a year before and that it had
been cleaned up when the tank was removed and replaced with a new steel tank. The second tank
apparently was not replaced at that time.
In the Leavings residence, vapors are very strong and the power has been shut off. Basement
windows have been left open to reduce the explosion potential.
OBJECTIVE
Your consulting geoenvironmental engineering firm has been retained by the attorney representing
the Leavings to:
1. Determine the source of the hydrocarbon contamination. This is not an emergency
response action.
2. Be prepared to defend the data you obtain and the conclusions you draw by means
of litigation.
3. Consider possible site remediation plans and recommendations that will make the
home habitable again.
BUDGET
The allowable budget to develop the field exploration is $25,000.
INFORMATION AVAILABLE
Interviews of neighbors. IDOT. station managers, and U-Haul
Lot 9: The trees are in pretty good condition. The house was vacant. Mrs. Leavings let
you in and asked you to put any cigarettes out before entering, just in case. She wanted to
have a house, not a hole, to come back to. You observe six inches of free product that looks
and smells like gasoline in the open sump pit in the basement. The power was cut so the
water level in the sump was allowed to rise. The fluid level in the sump was about three feet
below the basement floor level.
Neighbors: They lost several trees in back yards during the past spring. They contacted the
commercial developer behind their homes and complained that the fill that was placed there
Groundwater Investigation 2 9/93
-------
several years ago has finally killed their trees. They got no satisfaction from the developer.
Both said that when you find out where the gas came from let them know so they can sue
someone too. They noted that this past September and October were unusually wet (lots of
rainfall).
IDOT: The manager remembers your people testing his parking pad. Says the one
underground storage tank (UST) was replaced in 1979 while the second tank was installed
when the facility was first built in 1967. Both the original tanks were bare metal tanks. The
old one has always had gasoline while the newer one was the diesel tank. There are no
inventory records or leak testing records. He has never had any water in his tanks. He will
check with his supervisor to have the USTs precision leak tested.
U-Haul: The manager says the station used to be a Continental Oil station with three USTs.
One 6000 gal UST unleaded was kept in service for their fleet. It was found to be leaking
a month ago. They had originally been installed by Continental in 1970 when the station was
built. He has no idea how much was lost.
Mobil: The manager was pleasant until he found out what you wanted. You did learn that
he built the station in 1970 and installed three USTs at that time. The manager would not
answer any further questions.
Amoco: The manager wasn't in, but you talked to an assistant and got his phone number.
When you called later the manager said he was aware of the leaking tank at U-Haul and was
anxious to prove the product was not from his station. He said they installed three USTs for
unleaded, premium and regular in 1972. A diesel UST was installed in 1978. The tanks are
tested every two years using the Kent Moore (now Petrotite) test method. The tanks have
always tested tight. No inventory control system is being used at present. If you want to
put monitoring wells on his property, just let him know and he'd be happy to help out.
Review of Bettendorf City Hall records
An existing topographic map and scaled land use map are available.
Ownership records indicate the land was previously owned by Mr. and Mrs. Ralph
Luckless. Zoning at that time was agricultural only. The clerk said she had known
them prior to the farm sale in 1964. That section was used mostly for grazing cattle.
It was too steep for crops. She said she remembered a couple wooded valleys in that
field. A stream used to run along where Golden Valley Drive is now and that kids
used to swim in it (get muddy in it). The other valley was between Golden Valley
Drive and where all that fill is now near U-Haul and Amoco. You may want to talk
to the current property owner about that.
The current owner of the undeveloped property is Mr. M. Forester (developer) with
an Iowa City, IA address.
There is no record of storm or sanitary sewer lines along Utica Ridge Road south of
Golden Valley Drive. Storm and sanitary sewer lines run along Spruce Hills Drive.
9/93 3 Groundwater Investigation
-------
Iowa Geological Survey
There are no records of any wells in the section.
Adjoining section wells indicate top of bedrock at about 650 feet mean sea level (MSL). The
uppermost useable aquifer is the Mississippian for elevation 350 feet to 570 feet MSL. The
materials overlying the Mississippian are Pennsylvanian shales and limestone.
Soil Conservation Survey maps
The 1974 edition indicates "Made Land" over nearly all of the area not designated as
commercial zone. "Made Land" normally indicated areas of cut or fill.
Interview with developer - Mr. M. Forester
He bought the property in question in the 1960's. He developed the residential area first and
some of the commercial development followed. About forty acres remain undeveloped to
date. He is looking to locate a shopping center there if the economy ever turns back around.
He remembers getting a lot of cheap dirt and fill when the interstate cut went through about
one half mile west in the late 1960's. He filled in a couple of good sized valleys at that time.
He has a topographic map of the area after it was filled.
He would be more than happy to help out in any way possible. If you need to put any wells
on the property just let him know ahead of time. There are no buried utilities on the
property except behind the residential property.
Groundwater Investigation 4 9/93
-------
ASSIGNMENT: PHASE 1 FIELD INVESTIGATION
TABULATION OF FEES FOR PHASE 1 FIELD INVESTIGATION GROUP
WORK SHEET #1 | # UNITS
Recommendation for making residence
habitable
Field Exploration - mobilization
seismic refraction survey
earth resistivity survey
terrain conductivity
soil gas survey
soil boring with photo ionization detector -
25 feet deep max - grouted shut
COST
$500 LS
(lump sum)
$500 LS
not available
not available
not available
$1500/ac
$500 ea
TOTAL
$
$
$
6
$
$
Monitoring wells
2" PVC
1 5 ft screen - 25 ft deep
2" stainless steel
1 5 ft screen - 25 ft deep
well security - locking protector pipe
$1200 ea
$1700ea
$300 ea
$
$
$
Aquifer testing:
lab permeabilities
slug test w/interpretation
pump test w/interpretation
24 - 36 hr test
not available
not available
not available
6
Q
6
Chemical analysis (under C-O-C procedures)
priority pollutants
total hydrocarbons by IR
BTEX by GC
Field investigation engineering analysis and
report
not available
not available
15%
$2000 min
TOTAL COST:
6
v
6
$
9/93
Groundwater Investigation
-------
SCHEDULE OF FEES AND ESTIMATE OF COSTS*
Recommendations for making residence habitable
Field Exploration:
mobilization
seismic refraction survey
earth resistivity survey
terrain conductivity
soil gas survey
soil boring with PID
25 ft deep - grouted
monitoring well - 2" PVC
15 ft screen - 25 ft deep
monitoring well - 2" stainless steel
15 ft screen - 25 ft deep
well security - locking protector pipe
aquifer testing
lab permeabilities
slug test with interpretation
pump test with interpretation
(24 - 36 hr test)
chemical analysis (C-O-C procedures)
priority pollutants
total hydrocarbons by IR
BTEX by GC
field investigation engineering analysis
and report
$500
$500
$2,500
$2,500
$1,000
$1,500
$500
$1,200
$1,700
$300
$200
$300
$5,000
$1,000
$100
$2,000
(minimum)
LS
LS
/acre
/acre
/acre
/acre
ea
ea
ea
ea
ea
ea
ea
ea
ea
LS
Remediation Study:
remedial option evaluation
report preparation
agency coordination
$8,000
(minimum)
$2,000
$2,000
LS
LS
LS
* fees and cost estimates are for classroom purposes only.
Groundwater Investigation
9/93
-------
Remediation implementation and operation
excavate and dispose $150 cubic yard
air stripper
install winterized system $45,000 ea
operation for one year $5,000 ea
carbon adsorption unit
water phase installation $15,000 ea
operation for one year $10,000 year
recovery system (20 - 25 ft trench)
trench installation $75 linear ft
operation for one year $2,000 ea
well point
installation $2,000 ea
operation for one year $2,000 year
construction observation and testing 20%*
analytical costs (total hydrocarbons by
IR, BTEX by GC)
recovery system $2,400 month
monitoring well system (quarterly) $150 well
This is based on the remediation system construction and installation costs, not operational costs.
9/93 7 Groundwater Investigation
-------
Scaled Land Use
Iowa" DOT
Maintenance
Facility
9/93
Groundwater Investigation
-------
600 N
Predevelopment
Topographic Map
1000 S
Groundwater Investigation
10
9/93
-------
600 N
Existing Topographic Map
9/93
11
Groundwater Investigation
-------
§
CM
O
o
O
o
o
o
8
o
§
IP
IT
§
to
S
o
s
o
o
•f
o
o
....
'
tin
en
—
DD
+
—
1 4
' en
c
t—
+
<-
D
•n
^
\
+ ,l
mm*
. . . . ^
<
•
3?
H
GROUP
h — ~~
*
ip" "
| Soil gas survey
hit " " miss "— "
^^f^
/
Monitor
fill
loess
I till
alluviu
nonde
i free p
| gw ele
<^>
A4— •
cm
ing well
m
tection
roduct
nation
i 9
m
i
. t
— , . — i ,
. ..
Soil borings
fill
i IOGSS
till
alluvium
[ hit "+"
| miss
ti ti
-
1
500 N
400 N
300 N
200 N
100 N
0
7
100 S
200 S
300 S
400 S
11
500 S
600 S
700 S
14
800 S
900 S
1000 !
Groundwater Investigation
12
9/93
-------
J K L
GROUP
4
7
11
en
C=]
CZI
P/P3
13
Groundwater Investigation
-------
GROUP
MONITORING WELLS
FILL
LOESS
ALLUVIUM
TILL
NON DETECTED
DISSOLVED PRODUCT
FREE PRODUCT
WATER ELEVATION
1
2
3
4
5
6
7
8
9
10
SOIL BORINGS
FILL
LOESS
ALLUVIUM
TILL
HIT( + )
MISS (-)
A
B
C
D
E
F
G
H
I
J
9/93
15
Groundwater Investigation
-------
PROBLEM 5
Nomograph
-------
The information in Problem 5: Nomograph is reproduced or adapted from the following article:
Kent, D.C., W.A. Pettyjohn, and F.E. Witz. 1982. Methods for the Prediction of
Leachate Plume Migration, pp. 246-261. In: Proceedings of the Second National
Symposium on Aquifer Restoration and Ground Water Monitoring. May 26-28,
1982, The Fawcett Center, Columbus, Ohio. D.M. Nielsen (ed).
This information is reproduced by permission of the National Ground Water Association. Copyright
1982. All rights reserved.
-------
Proceedings
of the
Second National Symposium
on Aquifer Restoration
and
Ground Water Monitoring
May 26-28,1982
The Fawcett Center, Columbus, Ohio
Edited by
David M. Nielsen, Director of Research
National Water Well Association
Worthington, Ohio
Sponsors
National Center for Ground Water Research
U.S. Environmental Protection Agency
National Water Well Association
Published by
National Water Well Association
500 W. Wilson Bridge Road
Worthington, Ohio 43085
Produced by
Water Well Journal Publishing Company
500 W. Wilson Bridge Road
Worthington, Ohio 43085
NATIONAL WATER WELL ASSN.
LIBRARY
-------
Table 1
Definition of Terms
Primary Variables: Units
C := Concentration of leachate at a specified time and distance (M/L3)
X = Distance from source where concentration of leachate is computed. Distance is (L)
measured in direction of ground-water flow (perpendicular to gradient)
y = Transverse distance measured from the centerline of ground-water flow (Assumed (L)
to be zero in the nomograph)
t = Sample time from beginning of leachate source flow (T)
Aquifer Parameters:
m = Effective aquifer thickness or zone of mixing (L)
n = Effective porosity of aquifer or zone of mixing (Dimensionless)
V = Velocity of ground-water flow within voids:
estimated directly or from: _ KI
where:
K = Coefficient of permeability or hydraulic conductivity of aquifer or zone of
mixing: .. _ T
m
I = Gradient of ground-water flow (Dimensionless)
Transport Parameters:
Dx = Longitudinal dispersion coefficient (mixing rate) with respect to distance in x direc- (LV~)
tion and time: estimated directly or from:
Dx = ax V+D*
where:
or, = Longitudinal dispersivity (L)
D* = Molecular diffusion coefficient, which is assumed to be negligible for velocities (L2/T)
typical of permeable aquifers. D* may be the dominant process in aquitards
where oxV would be negligible.
Dy = Transverse dispersion coefficient (mixing rate) with respect to distance in the y (LJ/T)
direction and time: estimated directly or from:
Dy = oy V + D*
where:
xory = Transverse dispersivity (L)
or estimated as:
D = Dx'divided by a ratio, which commonly ranges between 5 and 10 for medium
to coarse sand aquifers .
247
-------
R
y
Table 1
Definition of Terms
(Continued)
Retardation factor estimated directly or from:
Vd
where:
(Dimensionless)
(M/L3)
(Dimensionless).
(LVM)
pb = Bulk density of aquifer medium
n, = Total porosity
Kd = Distribution factor for sorption on aquifer medium (from sorption isotherm
column studies)
V = Velocity of ground water C-/T)
Vd = Observed velocity of leachate for a given concentration and chemical species (L/T)
Coefficient for radioactive or biological decay. For no decay, the value of y is one. • (Dimensionless)
(Assumed to be one in the nomograph.) Calculated from:
y \ • / *» * — • i
where:
X = Decay constant = Og' L
tV2 = Halflife: time when half of the original mass 'remains
Source Rate of Leachate:
QC0 = Mass flow rate estimated directly or obtained from the product of:
Q = Volume flow rate estimated directly or from:
Q = Aq
where:
A = area of source
q = recharge rate
C0 = Initial concentration
Intermediate Variables (Used for Nomograph only):
X0 = A characteristic dispersion length or scale factor given by:
(1/T)
(T)
(M/T)
(LVT)
V?v
TD = A characteristic dispersion time or scale factor given by:
(L/T)
(M/L3)
(L)
(T)
r
QD = A characteristic dilution-dispersion flow
_ Q0 = n m VDxDy
-------
PROBLEM 5: NOMOGRAPH
A SIMPLE GROUNDWATER MODEL TO EVALUATE
CONTAMINANT PLUME MIGRATION
INTRODUCTION
Groundwater models are used to evaluate the fate of contaminant migration in groundwater.
Typically this migration is depicted as a plume. The size and shape of the plume is dependent on
many interactive factors such as the hydraulic parameters of the aquifer, the compositional
complexities and concentrations of the contaminants, the length of time contaminants were injected
into the groundwater, and the heterogeneity and compositon of the aquifer's geologic framework.
More complicated models are capable of assimilating on a grid or other data distribution system,
which covers the potential area of concern, the interaction of these factors in order to predict the
plume's geometry. Obviously, the more data available to input into the model, the more time
required to compute this information and to evaluate the extent of migration. Of course this also
requires more money for lengthy computations. Fortunately there are less complex models available
that are not as costly and time consuming. Unfortunately these models are less sensitive to the
variance of factors that control plume migration. They are considered more of an approximation or
screening device to quickly evaluate the extent of contamination.
The nomograph is one such model. It has the capability of quickly estimating the potential distance
and time a contaminant plume migrates downgradient from the source. Other benefits include quickly
evaluating the placement of monitoring wells to further characterize the plume and possibly
controlling migration offsite, and providing an inexpensive predictive method. However, it can only
evaluate the concentration of one chemical component within this plume. When there are many
contaminants found at a site it is suggested that the most mobile or conservative contaminant should
be considered first. This should provide a worst case senario in evaluating the maximum length the
plume has migrated from the source.
Computer models are based on an attempt to define the interaction of physical properties of the
aquifer and the contaminant in terms of mathematical formulas. The nomograph is not different in
this respect. It utilizes a variance to the Wilson-Miller equation (Wilson and Miller, 1978) as shown
below:
QCa
C = V ! exp
x,x
where:
9/93 1 Nomograph
-------
The Wilson-Miller equation was formulated to predict a two dimensional plume in a uniform
groundwater-flow environment. The equation shown above provides scale factors based on physical
parameters which are known or can be calculated. The scale factors in this equation are used as
ratios with the primary parameters of time (t), distance (x), and the mass flow rate from the source
or the product of volume flow rate (Q) times the initial concentration of the contaminant (C0). These
ratios are expressed as t/TD, x/XD, and QC0/QD. These scale factors are defined below as:
T
~~
x .5-
**• n —
D V
where:
Dx = longitudinal dispersion coefficient or dispersion in the downgradient direction
(i.e., x direction),
Dy = transverse dispersion coefficient or dispersion in the crossgradient direction
(i.e., y direction),
V = seepage velocity of the groundwater,
Rd = retardation of the contaminant,
n = effective porosity of the aquifer, and
m = total aquifer thickness.
It is important to note that the y distance of contaminant migration is ignored in the nomograph
model.
Values of Dx and Dy are site-specific parameters and are dependent on the homogeneity and
isotropism of the aquifer, advection rate of the contaminants, hydrodynamic dispersion, and the
reaction potential of the contaminants to the aquifer's matrix or geology. Therefore these values are
difficult to determine. Typically values can be obtained through laboratory experiments but are
usually underestimates due to the size of the aquifer sample. There are also field methods which can
be used to determine these coefficients, but are beyond the scope of this discussion. If the reader is
interested in further discussion on this matter he/she should review the modeling section of this
manual, Freeze and Cherry (1979), Fetter (1988), and Driscoll (1986).
Nomograph 2 9/93
-------
Seepage velocity (V) is calculated if hydraulic conductivity (K), hydraulic gradient (I), and effective
porosity are known for the aquifer using the formula as shown below:
n
Retardation coefficient (Rd) can also be calculated using the following formula:
R =
d
where:
Kd = distribution coefficient of the contaminant between the groundwater and
aquifer,
PB = bulk density of the aquifer matrix, and
6 = the total porosity within in the aquifer which is typically a higher value than
the effective porosity used earlier.
Distribution coefficient is also further explained in the modeling references given earlier in this
discussion.
APPLICATIONS
As stated earlier, the nomograph is designed to provide a simple technique to estimate one of the
following problems:
la. The concentration (C) is determined from a given distance (x) and for a specified time
(t),
Ib. The maximum concentration (C) that might occur over a long period of time
usually defined as steady state conditions,
2a. The distance (x) where a specified concentration of contaminant will exist given some
time interval (t),
2b. The maximum distance (x) a contaminant might migrate under steady state
conditions, or
3. The time (t) when a known or specified concentration (C) of a contaminant will
migrate to a selected location downgradient of the source.
9/93 3 Nomograph
-------
EXAMPLE
A disposal facility in South Farmingdale, Nassau County, New York began receiving cadmium- and
hexavalent chromium-enriched electroplating wastes in 1941. The waste was dumped into three pits
on location and immediately began infiltrating into an unconfined, shallow glacial aquifer. By early
1960 the migrating plume had reached 4,200 feet (ft) downgradient (x), 1,000 ft crossgradient (y),
and 70 ft into the aquifer. The thickness of the aquifer (m) in this area varied according to soil
borings from 100 to 140 ft with an average of(110 ft. Groundwater velocity was estimated between
0.5 and 1.5 feet per day (ft/dy). Specific yield (Sy) of the aquifer was estimated at 35. According
to site records it was reported that 200,000 to 300,000 gallons per day (gpd) of waste fluids were
discharged into the three pits. Infiltration rate (q) was estimated at a rate of 7,600 inches per year
(in/yr). Chromium concentration in the waste averaged 31 milligrams per liter (mg/1). The combined
area of these pits was measured at approximately 15,470 square feet (ft2). Retardation of the
contaminants was not a factor at this site due to the lack of free clay and organic matter within the
surficial aquifer. Based on experimental data obtained from similar sites in the area, longitudinal or
downgradient dispersivity (ax) and transverse or crossgradient dispersivity (ay) were assumed to be
70 ft and 14 ft respectively. A summary of this information is provided in Table 1 below:
TABLE 1
m = thickness of aquifer 110.0 ft
Sy = specific yield 35.0
V = groundwater velocity 1.5 ft/dy
ax = longitudinal dispersivity 70.0 ft
oty = transverse dispersivity 14.0 ft
Rd = retardation coefficient 1.0
C0 = initial concentration of contaminant 31.0 mg/1
A = area of discharge of contaminant 15,470.0 ft2
q = infiltration rate of contaminant 7,600.0 in/yr
Before using the nomograph some basic assumptions and calculations are necessary. First, effective
porosity (n) of the aquifer is not known for this site. However Sy was reported as 35. Without further
information one will estimate n at 35% based on the Sy value. Dispersion coefficients can be
calculated from dispersivity values already given in Table 1 using the following formulas of Dx =
axV and Dy = ayV where V is velocity of the groundwater found in Table l.The results of these
calculations are 105 ft2/dy and 21 ft2/dy, respectively.
For a nonpoint source for contaminants, such as at this site, the volume flow rate is estimated using
the formula Q = Aq. Values of A and q are given above in Table 1 . Converting units of q from
in/yr to ft/dy requires the following calculation:
yr )l2in) \365dy dy
The value for Q can now be calculated as 26,763 ft3/dy using the above formula Q = Aq. The value
of Q can also be computed from the discharge rate of 200,000 gpd of waste in the pits by using the
Nomograph 4 9/93
-------
conversion formula of 1 ft3/?. 48 gallons to change the units of gpd to fifVdy. The result of this
conversion yields a value for Q of 26,738 ft3/dy which is similar to the previous value of Q.
Now the mass flow rate, expressed as QC0, is (26,763 ft3/dy)(31 mg/1). Changing the units of mg/1
to pounds per cubic feet (Ib/ft3) requires the following conversion:
31 mg = (31mg\ 1kg (2.2lb\ lQ?l lm = 0.00195 Ib
I ~( I )(l06mg)(lk8)(lm*)(35ft*)~ ft3
The mass flow rate (QC0) of (26,763 ft3/dy)(31 mg/1) is equivalent to 52 Ib/dy.
From the above computations and assumptions we now have sufficient information to calculate the
scale factors for the previously decribed applications for the nomograph as shown below:
D - 10ft
D V 1.5ftIdy
_ __ . m
V2 (l.5ft/dy)2
QD = nmD = (0.35) (110ft) J(lQ5ft2/dy) (2\ft2ldy) = l,SQSft3/dy
Now consider the three applications of the nomograph. In order to use the nomograph, the ratios
used in the nomograph must be calculated as shown in Figure 1.
9/93 5 Nomograph
-------
NOMOGRAPH FOR
PLUME CENTER-LINE
CONCENTRATION
STEADY STATE
F (t —«)
500
IPQO
2.0OO
5.OOO
KDDOO
20.0OO
50DOO
A 100
,000 IO.OOO IOO.OOO
HO4
X
XD
-------
Application la and 1b
la. Find the concentration (C) of chromium at a known distance (x) downgradient and
time (t). Assume for this example x = 4,200 ft and t = 2,300 dy. Follow Figure 1
for the solution to this problem.
— = ' * = 60 (This value is located at point A)
— = ' ^ = 49.3 (This value is located at point E)
TD 46.61dy
(26,163ft3 1 dy)($\mgll)\ Afoet ,,,-,• , • , .;. -,m
= - — - - J ' •r/v - ^— - = 458.9 mg// (This value is located at point D)
QD
To find the concentration (C) draw a line vertical from point A to the intersection with the
t/TD value of 49.3 located as point B on Figure 1. Then draw a line horizontally from point
B to point C. Now from point C draw a straight line through the QC0/QD value of 458.9
mg/1 (point D) and intersecting the vertical bar graph representing the concentration (C) of
the contaminant, in this case chromium, under these conditions (point E). The solution to the
problem is 2.6 mg/1 of hexavalent chromium found 4,200 ft downgradient after 2,300 days
of migration from the disposal pits.
Ib. Find the maximum concentration of hexavalent chromium for the same distance
downgradient as given in application la but for a longer time period (greater than 10
years). In this solution, as depicted in Figure 1 also, one must project the x/XD value
vertically to the steady state line which represents a time (t) approximately greater
than 10 years. This is depicted as point F on Figure 1. The procedure is now similar
to application la ultimately projecting a line through point D and intersecting the
vertical concentration bar graph at point H. Under these conditions, 20 mg/1 of
hexavalent chromium is predicted.
Application 2a and 2b
2a. Determine the distance (x) downgradient where a selected concentration (C) of a
contaminant, such as hexavalent chromium, will occur at a given time (t). This
application is important if one is interested in evaluating chromium's breakthrough
concentration above its MCL (maximum contaminant level) of 0.05 mg/1 according
9/93 7 Nomograph
-------
to drinking water standards. Using the same value for QC0/QD of 458.9 mg/1 as
represented as point D on Figure 2 , one now locates the concentration of 0.05 mg/1
on the vertical concentration bar graph. This is depicted as point A on Figure 2.
Starting at point A project a straight line through point D to point B on the
nomograph of Figure 2. From point B construct a line horizontally across the
nomograph. Now one must determine the amount of time (t) the migration of the
contaminant has occurred in order to determine the time ratio of t/TD. For example,
one might be interested in how far downgradient this concentration of 0.05 mg/1 will
occur after 4,667 days. The time ratio computes to 100 using the time scalor (TD) of
46.67 days and is depicted as point C on Figure 2. Projecting a line vertically down
from point C to the x/XD value of 150 (point E), the distance (x) downgradient this
concentration occurs after 4,667 days is computed as 10,500 ft or approximately 2
miles. For a shorter time (t) period , such as 466.7 days the ratio t/TD equals 10
(point F), x/XD is determined as 23. Therefore the concentration of 0.05 mg/1 of
hexavalent chromium is estimated at 1610 ft downgradient from the source after
466.7 days have passed.
2b. Find the distance downgradient the concentration of 0.05 mg/1 will occur under
steady state conditions. Unfortunately the steady state line on the nomograph is not
intersected as depicted on Figure 2. For this scenario, one can only assume that the
distance downgradient is greater than 7,000,000 ft according to the computation of
100,000 (x/XD) times 70 ft (XD). Let us hope your agency has remediated this site
before steady state is ever achieved!
Nomograph 8 9/93
-------
\a\jjfc 2 '
ioi* 2 A * 2 4
NOMOGRAPH FOR
PLUME CENTER-LINE
CONCENTRATION
STEADY STATE
:; (mg/l)
; (mg/l)
500
IDOO
5,000
KDOOO
20,000
'50000
I°°E «So 1,000 10,000 100,000
X
XD
-------
Application 3
3. Find the time (t) required for a known concentration (C) to migrate to a specified
distance (x) downgradient of the source. Let us consider how long it will take for 1
mg/1 of hexavalent chromium to reach a private water supply well located 5280 ft (x)
downgradient of the contaminant source. First one must locate the concentration (C)
of 1 mg/1 on the vertical concentration bar graph. This is depicted as point A on
Figure 3. Using the same value of 458.7 mg/1 for the ratio of QC0/QD (point D),
construct a straight line from point A through point D to the nomograph (point B).
Then project a horizontal line from point B through the graph. Next project a vertical
line from point C representing the value of 75.45 as computed for x/XD = 5,280 ft/
70 ft until it crosses the previously constructed horizontal line (point E) as shown on
Figure 3. At this intersection the value of t/TD is equal to 60 or t is 2,800 days
according to the calculation TD times 60 where TD equals 46.67 days. It is important
to note that should point E occur above the steady state line for the value x/XD of
75.45 on the nomograph, the concentration (C) of 1 mg/1 will not be found at this
distance of 5,280 ft downgradient from the source.
Nomograph 10 9/93
-------
NOMOGRAPH FOR -
PLUME CENTER-LINE :
CONCENTRATION -
STEADY STATE
(Ib/fn;[irs5
500
IPOO
2POO
5,000
KDjDOO
20,000
50,000
1,000 10,000 100,000
HO4
-------
PROBLEM
The Alkali Lake chemical storage site in eastern Oregon became a chemical waste disposal site in
November, 1983 when 25,000 55-gallon capacity drums were crushed and buried in 12 shallow
(0.60 to 0.75 meters deep) unlined trenches, 130 meters (m) long and 20 m apart. The drums
contained chlorophenolic "still bottoms" or distillation residues from the manufacturing of herbicides
2,4-D (2,4-Dichlorophenoxyacetic acid) and MCPA (4-Methyl-2-chlorophenoxyacetic acid) for the
Vietnam War. The average concentration of 2,4-D in the drums was determined to be 200 mg/1.
Burial of these drums injected this chemical waste directly into the groundwater system beneath this
site at the rate of 8,313 gallons per day. A site investigation was conducted eight years later after
many complaints concerning the strange odor and taste in the water from private wells. The results
of the site investigation obtained the following aquifer parameters or properties:
m = aquifer thickness 30 m (98.43 ft)
n = effective porosity 5 % (0.05)
0 = total porosity 65 % (0.65)
PB= bulk density 0.95 g/cm3
K = hydraulic conductivity 0.1 cm/sec (283 ft/dy)
I = hydraulic gradient 0.0002
V = seepage velocity 1.132 ft/dy
Dx= longitudinal dispersion coefficient 60 ft2/dy
DY= transverse dispersion coefficient 12 ft2/dy
QC0= mass injection rate 8,313 gpd x 200 mg/1
Kd= distribution coefficient 0.5 cm3/g
Rd= retardation coefficient 1.73
Unfortunately this investigation was concluded and a site report submitted to the State Department
of Environmental Protection before it was discovered that the well search was not complete.
Somehow, a municipal water-supply well located 1,700 ft downgradient was overlooked on the initial
well search for potentially impacted water-supply wells within a one-mile radius of the site. Since
the water from this well was not sampled during the investigation, how can one estimate the
concentration (C) of 2,4-D in this well given the following ratios for the nomograph:
t/TD = 36, x/XD = 32, and QC0/QD = 1,696 mg/1.
Nomograph 12 9/93
-------
PROBLEM
Calculations and Unit Conversions
AQUIFER THICKNESS (m)
30mf3.281.ftl,
m =
30m| 3.2
1.0 1 1.
Om J
HYDRAULIC CONDUCTIVITY (K)
0.1 cmT 60.0sec| 60.0minl 24.0hrJ 1.0in T 1.0.ft
sec }l.0min| l.Qhr J[ l.Ody }2.54cm| 12.0m
dy
SEEPAGE VELOCITY (V)
dy[ 0.05 j
V = 1.132^-
dy
RETARDATION COEFFICIENT (Rd)
Rd = 1.0
Rd = 1.0
0.5cm3[0.95g| 1.0
g [ cm3 10.65
9/93
13
Nomograph
-------
TIME SCALE FACTOR
Rd = 1.0 + 0.73 = 1.73
T
=
T =
l
1.73|60.0yi;2 dy2
1-01 dy
= 103.8 dy __
D 1.28 y
Q =
DISCHARGE RATE (Q)
1.0ft3
dy
IAS gal
dy
FLOW SCALE FACTOR
QD = nm JDXDY
QD = (0.05)(98.0>*)
Qo =
dy2 dy
Nomograph
14
9/93
-------
LENGTH SCALE FACTOR
Dx
Xn = -?
P/P3
15
Nomograph
-------
NOMOGRAPH FOR
PLUME CENTER-LINE
CONCENTRATION
STEADY STATE
500
ipoo
2POO
5,000
'
20.00O
5CLOOOI
IO
-6_
IO'3
IO'2
QC0
H QD
! (mg/l)
HO I
102
QD I '
(lb/ft3)
!'°1
I02J
•
I03j
10
5J
-iJHO3
!LI04
-107
-108
IHO
9
,000 10,000 100,000
X
XD
HO
HO-'
-i
,-2
c
(mg/l)
-10
HO2
Ho3
HO4
-------
Section 10
-------
APPENDIX A
Sampling Protocols
-------
GENERALIZED GROUNDWATER SAMPLING PROTOCOL
Step
Goal
Recommendations
Hydrologic measurements Establish nonpumping water level
Well purging
Sample collection
Filtration/preservation
Field determinations
Field blanks/standards
Sample storage,
transportation, and chain
of custody (COC)
Remove or isolate stagnant H20,
which would otherwise bias
representative sample
Collect samples at land surface or
in well bore with minimal
disturbance of sample chemistry
Filtration permits determination of
soluble constituents and is a form
of preservation; it should be done
in the field as soon as possible
after sample collection
Field analyses of samples will
effectively avoid bias in
determining
parameters/constituents that do
not store well (e.g., gases,
alkalinity, and pH)
These blanks and standards will
permit the correction of analytical
results for changes that may
occur after sample collection.
Preserve, store, and transport
with other samples.
Refrigerate and protect samples to
minimize their chemical alteration
prior to analysis. Document
movement of samples from
collector to laboratory.
Measure the water level to
±0.3 cm (±0.01 ft)
Pump water until well purging
parameters (e.g., pH, T, Q'\
Eh) stabilize to ±10% over at
least two successive well
volumes pumped
Pumping rates should be
limited to ~ 100 mL/min for
volatile organics and gas-
sensitive parameters
For trace metals, inorganic
anions/cations, and alkalinity.
Do not filter TOC, TOX, or
other volatile organic
compound samples; filter other
organic compound samples
only when required
Samples for determining gases,
alkalinity, and pH should be
analyzed in the field if at all
possible
At least one blank and one
standard for each sensitive
parameter should be made up
in the field on each day of
sampling. Spiked samples are
also recommended for good
QA/QC.
Observe maximum sample
holding or storage periods
recommended by EPA.
Documentation of actual
holding periods should be
carefully performed. Establish
COC forms, which must
accompany all samples during
shipment.
Adapted from: U.S. EPA. 1985. Practical Guide for Ground-Water Sampling. EPA/600/2-85/104.
Robert S. Kerr Environmental Research Laboratory, Ada, OK.
9/93
Sampling Protocols
-------
APPENDIX B
References
-------
REFERENCES
Allen, H.E., E.M. Perdue, and D.S. Brown (eds). 1993. Metals in Groundwater. Lewis
Publishers, Inc., Chelsea, MI.
Aller, L., T.W. Bennett, G. Hackett, R.J. Petty, J.H. Lehr, H. Sedoris, D.M. Nielsen, and J.E.
Denne. 1989. Handbook of Suggested Practices for the Design and Installation of Ground-Water
Monitoring Wells. EPA 600/4-89/034. National Ground Water Association Publishers, Dublin,
OH.
AIPG. 1985. Ground Water Issues and Answers. American Institute of Professional Geologists,
Arvada, CO, 1985.
Bachmat, Y., J. Bredehoeft, B. Andrews, D. Holtz, and S. Sebastian. 1980. Groundwater
Management: The Use of Numerical Models, Water Resources Monograph 5. American
Geophysical Union, Washington, DC.
Back, W., and R.A. Freeze. 1983. Chemical Hydrology. Benchmark papers in Geology/v.73.
Hutchinson Ross Publishing Co., Stroudsburg, PA.
Barvis, J.H., J.G. McPherson, and J.R.J. Studlick. 1990. Sandstone Petroleum Reservoirs.
Springer Verlag Publishing.
Bear, J. 1972. Dynamics of Fluids in Porous Media. American Elsevier, NY.
Bear, J. 1979. Hydraulics of Groundwater. McGraw-Hill, New York, NY.
Bear, J., D. Zaslavsky, and S. Irmay. 1968. Physical Principles of Water Percolation and Seepage.
UNESCO.
Bennett, G.D. 1989. Introduction to Ground-Water Hydraulics: A Programmed Text for Self-
Introduction. Techniques of Water-Resources Investigations of the United States Geological Survey.
United States Government Printing Office, Washington, DC.
Benson, R.C., et al. 1984. Geophysical Techniques for Sensing Buried Wastes and Waste
Migration. EPA 600/7-84/064.
Bitton, G., andC.P. Gerba. 1984. Groundwater Pollution Microbiology. John Wiley & Sons, New
York, NY.
Bouwer, H. 1978. Groundwater Hydrology. McGraw-Hill Book Co., New York, NY.
Carter, L.W., and R.C. Knox. Ground Water Pollution Control. Lewis Publishers, Inc., Chelsea,
MI.
Cedergren, H.R. 1977. Seepage, Drainage and Flow Nets. Second edition. John Wiley & Sons,
New York, NY.
9/93 1 References
-------
Cole, J.A. (ed). 1974. Groundwater Pollution in Europe. Water Information Center Inc., Port
Washington, NY.
Collins, A.G., and A.I. Johnson (eds). 1988. Ground-Water Contamination: Field Methods.
American Society for Testing and Materials.
Davis, S.N., and R.J.M. DeWiest. 1966. Hydrogeology. John Wiley & Sons, New York, NY.
Dawson, K.J., and J.D. Istok. 1991. Aquifer Testing. Lewis Publishers, Inc., Chelsea, MI.
DeWiest, R.J.M. 1965. Geohydrology. John Wiley & Sons, New York, NY.
Dobrin, M.B. 1960. Introduction to Geophysical Prospecting. McGraw-Hill, New York, NY
Domenico, P.A. 1972. Concepts and Models in Groundwater Hydrology. McGraw-Hill, New
York, NY.
Domenico, P.A. 1990. Physical and Chemical Hydrogeology. John Wiley & Sons, New York,
NY.
Dragun, J. 1988. Soil Chemistry of Hazardous Materials. Hazardous Materials Control
Research Institute, Silver Spring, MD.
Drever, J.I. 1988. Geochemistry of Natural Waters. Second edition. Prentice-Hall, Inc.,
Englewood Cliffs, NJ.
Driscoll, F.G. 1986. Groundwater and Wells. Second edition. Johnson Division, St. Paul, MN.
Everett, L.G., L.G. Wilson, and E.W. Hoylman. 1984. Vadose Zone Monitoring for Hazardous
Waste Sites. Noyes Data Corporation.
Fetter, C.W., Jr. 1980. Applied Hydrogeology. Charles E. Merrill Publishing Co., Columbus,
OH.
Freeze, R.A., and W. Back. 1983. Physical Hydrogeology. Benchmark Papers in Geology/v. 72.
Hutchinson Ross Publishing Co., Stroudsburg, PA.
Freeze, R.A., and J. Cherry. 1979. Groundwater. Prentice-Hall, Englewood Cliffs, NJ.
Fried, J.J. 1975. Groundwater Pollution. Elsevier Scientific Publishing Co., Amsterdam.
Garrels, R.M., and C.L. Christ. 1987. Solutions, Minerals, and Equilibria. Harper and
Corporation Publishers.
Gibson, U.P., and R.D. Singer. 1971. Water Well Manual. Number 4101. Premier Press,
Berkeley, CA.
Harr, M.E. 1962. Groundwater and Seepage. McGraw-Hill, New York, NY.
References 2 9/93
-------
Heath, R.C. 1987. Basic Ground-Water Hydrology. USGS Water Supply Paper 2220. U.S.
Geological Survey.
Heath, R.C., and F.W. Trainer. 1992. Ground Water Hydrology. National Ground Water
Association, Dublin, OH.
Hem, J.D. 1989. Study and Interpretation of the Chemical Characteristics of Natural Water.
United States Geological Survey Water Supply Paper 2254. U.S. Government Printing Office,
Washington, DC. .
Hillel, D. 1971. Soil and Water: Physical Principles and Processes. Academic Press, New York,
NY.
Hoehn, R.P. 1976-77. Union List of Sanborn Fire Insurance Maps Held by Institutions in the U.S.
and Canada. Western Association of Map Libraries. Santa Cruz, CA.
Johnson, A.I., C.B. Pettersson, and J.L. Fulton (eds). 1992. Geographic Information Systems
(GIS) and Mapping - Practices and Standards. American Society for Testing and Materials.
Kranskopf, K.B. 1967. Introduction to Geochemistry. McGraw Hill, Inc., New York, NY.
Kruseman, G.P., and N.A. de Ridder. 1990. Analysis and Evaluation of Pumping Test Data. ILRI
Publication 47. International Institute for Land Reclamation and Improvement, Wageningen, The
Netherlands.
Larkin, R.G., and J.M. Sharp, Jr. 1992. On The Relationship Between River-Basin
Geomorphology Aquifer Hydraulics and Ground-Water Flow Direction in Alluvial Aquifers.
Geological Society of America Bulletin, v. 104, pp. 1608-1620.
LeBlanc, R.J. 1972. Geometry of Sandstone Reservoir Bodies, pp. 133-190. In: American
Association of Petroleum Geologists Memoir 18. Underground Waste Management and
Environmental Implications. T.D. Cook (ed). 412 pp.
LeRoy, L.W. 1951. Substance Geologic Methods. Colorado School of Mines.
Lohman, S.W. 1979. Ground-Water Hydraulics. Geological Survey Professional Paper 708. U.S.
Government Printing Office, Washington, DC.
Mackay, D., W.-Y. Shiu, and K.-C. Ma. 1992. Illustrated Handbook of Physical-Chemical
Properties and Environmental Fate for Organic Chemicals. Volumes I, II, and III. Lewis
Publishers, Inc., Chelsea, MI.
Mandel, S., and Z.L. Shiftan. 1981. Groundwater Resources: Investigation and Development.
Academic Press.
Matthess, G. 1982. The Properties of Groundwater. John Wiley & Sons, New York, NY.
9/93 3 References
-------
Mazor, E. 1991. Applied Chemical and Isotropic Groundwater Hydrology. Halsted Press (a
division of John Wiley and Sons Inc.), New York, NY.
McDonald, M.G., and A.W. Harbaugh. 1988. A Modular Three-Dimensional Finite-Difference
Ground-Water Flow Model. Techniques of Water-Resources Investigations of the United States
Geological Survey. United States Government Printing Office, Washington, DC.
McWhorter, D., and O.K. Sunada. 1977. Ground-Water Hydrology and Hydraulics. Water
Resources Publishing, Ft. Collins, CO.
Montgomery, J.H., and L.M. Welkom. 1990. Groundwater Chemicals Desk Reference. Lewis
Publishers, Inc., Chelsea, MI.
Morrison, R. 1983. Groundwater Monitoring Technology. Timco Mfg. Company, Prairie du Sac,
WI.
Morrison, R.T., and R.N. Boyd. 1959. Organic Chemistry. Allyn and Bacon, Inc.
NGWA. 1991. Summaries of State Ground Water Quality Monitoring Well Regulations by EPA
Regions. National Ground Water Association, Dublin, OH.
NWWA. No date. Selection and Installation of Well Screens and Ground Packs: An Anthology.
National Water Well Association, Dublin, OH.
Niaki, S., and J.A. Broscious. 1987. Underground Tank Leak Detection Methods. Noyes Data
Corporation Publishers.
Nielsen, D.M. (ed). 1991. Practical Book of Ground-Water Monitoring. Lewis Publishers, Inc.,
Chelsea, MI.
Nielsen, D.M., and A.I. Johnson (eds). 1990. Ground Water and Vadose Zone Monitoring.
American Society for Testing and Materials.
Nielsen, D.M., R.D. Jackson, J.W. Gary, and D.D. Evans. 1972. Soil Water. American Society
of Agronomy, Madison, WI.
Nielsen, D.M., and M.N. Sara (eds). 1992. Current Practices in Ground Water and Vadose Zone
Investigations. American Society for Testing and Materials.
Palmer, C.M., J.L. Peterson, and J. Behnke. 1992. Principles of Contaminant Hydrogeology.
Lewis Publishing, Inc., Chelsea, MI.
Pettyjohn, W.A. (ed). 1973. Water Quality in a Stressed Environment. Burgess Publishing,
Minneapolis, MN.
Pettyjohn, W.A. 1987. Protection of Public Water Supplies from Ground-Water Contamination.
Noyes Data Corporation, Park Ridge, NJ.
References 4 9/93
-------
Polubarinova-Kochina, P.Y. 1962. Theory of Groundwater Movement. Princeton University Press,
Princeton, NJ.
Powers, P.J. 1981. Construction Dewatering: A Guide to Theory and Practice. John Wiley &
Sons, New York, NY.
Princeton University Water Resources Program. 1984. Groundwater Contamination from
Hazardous Wastes. Prentice-Hall, Inc., Englewood Cliffs, NJ.
Remson, I., G.M. Hornberger, and F.J. Molz. 1971. Numerical Methods in Subsurface
Hydrology. Wiley-Interscience, New York, NY.
Sanborn Map Company. 1905. Description and Utilization of the Sanborn Map. Pelham, NY.
Sanborn Map Company. 1905. Surveyor's Manual for the Exclusive Use and Guidance of
Employees of the Sanborn Map Company. Pelham, NY.
Summers, W.K., and Z. Spiegel. 1971. Ground Water Pollution, A Bibliography. Ann Arbor
Science Publishing, Ann Arbor, MI.
Sun, R.J. 1978-84. Regional Aquifer-System Analysis Program of the U.S. Geological Survey
Summary of Projects. U.S. Geological Survey Circular 1002.
Telford, W.M., L.P. Geldart, R.E. Sheriff, and D.A. Keys. 1976. Applied Geophysics.
Cambridge University Press, Cambridge, England.
Todd, O.K. 1980. Ground Water Hydrology. Second edition. John Wiley & Sons, New York,
NY.
Todd, O.K., and D.E.G. McNulty. 1976. Polluted Groundwater. Water Information Center, Inc.,
Port Washington, NY.
Travis, C.C., and E.L. Etnier (eds). 1984. Groundwater Pollution, Environmental & Legal
Problems. American Association for the Advancement of Science, AAAS Selected Symposium 95.
U.S. EPA. 1984. Geophysical Techniques for Sensing Buried Wastes and Waste Migration.
EPA/600/7-84/064. U.S. Environmental Protection Agency.
U.S. EPA. 1985. Practical Guide for Ground-Water Sampling. EPA/600/2-85/104. U.S.
Environmental Protection Agency.
U.S. EPA. 1985. Protection of Public Water Supplies from Ground-Water Contamination: Seminar
Publication. EPA/625/4-85/016. U.S. Environmental Protection Agency.
U.S. EPA. 1986. RCRA Ground-Water Monitoring Technical Enforcement Guidance Document.
OSWER-9950. U.S. Environmental Protection Agency.
9/93 5 References
-------
U.S. EPA. 1986. Superfund State Lead Remedial Project Management Handbook. EPA/540/G-
87/002. U.S. Environmental Protection Agency.
U.S. EPA. 1987. Data Quality Objectives for Remedial Response Activities Example Scenario:
RI/FS Activities at a Site With Contaminated Soil and Ground Water. EPA/540/G-87/004. U.S.
Environmental Protection Agency.
U.S. EPA. 1987. Superfund Federal Lead Remedial Project Management Handbook.
EPA/540/G-87/001. U.S. Environmental Protection Agency.
U.S. EPA. 1988. Guidance of Remedial Actions for Contaminated Ground Water at Superfund
Sites. EPA/540/6-88/003. U.S. Environmental Protection Agency.
U.S. EPA. 1988. Selection Criteria for Mathematical Models Used in Exposure Assessments:
Ground-Water Models. EPA/600/8-88/075. U.S. Environmental Protection Agency.
U.S. EPA. 1988. Superfund Exposure Assessment Manual. EPA/540/1-88/001. U.S.
Environmental Protection Agency.
U.S. EPA. 1988. Technology Screening Guide for Treatment of CERCLA Soils and Sludges.
EPA/540/2-88/004. U.S. Environmental Protection Agency.
U.S. EPA. 1989. Ground-Water Monitoring in Karst Terranes: Recommended Protocols &
Implicit Assumptions. EPA/600/X-89/050. U.S. Environmental Protection Agency.
U.S. EPA. 1990. Basics of Pump-and-Treat Ground-Water Remediation Technology.
EPA/600/8-90/003. U.S. Environmental Protection Agency.
U.S. EPA. 1990. Catalog of Superfund Program Publications. EPA/540/8-90/015. U.S.
Environmental Protection Agency.
U.S. EPA. 1990. Handbook Ground Water Volume I: Ground Water and Contamination.
EPA/625/6-90/016a. U.S. Environmental Protection Agency.
U.S. EPA. 1990. Quality Assurance Project Plan. U.S. Environmental Protection Agency,
Emergency Response Branch, Region VIII.
U.S. EPA. 1990. Subsurface Contamination Reference Guide. EPA/540/2-90/001. U.S.
Environmental Protection Agency.
U.S. EPA. 1991. Compendium of ERT Ground Water Sampling Procedures. EPA/540/P-91/007.
U.S. Environmental Protection Agency.
U.S. EPA. 1991. Compendium of ERT Soil Sampling and Surface Geophysics Procedures.
EPA/540/P-91 /006. U.S. Environmental Protection Agency.
References 6 9/93
-------
U.S. EPA. 1991. Ground-Water Monitoring (Chapter IT of SW-846). Final Draft. U.S.
Environmental Protection Agency, Office of Solid Waste.
U.S. EPA. 1991. Handbook Ground Water Volume II: Methodology. EPA/625/6-90/016b. U.S.
Environmental Protection Agency.
Van Der Leeden, F., F.L. Troise, and D.K. Todd. 1990. The Water Encyclopedia. Second
edition. Lewis Publishers, Inc., Chelsea, MI.
Practical Applications of Ground Water Models. National Conference August 19-20,1985. National
Water Well Association, Dublin, OH.
Verruijt, A. 1970. Theory of Groundwater Flow. Gordon & Breach Sciences Publishing, Inc.,
New York, NY.
Walton, W.C. 1962. Selected Analytical Methods for Well and Aquifer Evaluation. Bulletin 49,
Illinois State Water Survey.
Walton, W.C. 1970. Groundwater Resource Evaluation. McGraw-Hill, New York, NY.
Walton, W.C. 1984. Practical Aspects of Ground Water Modeling. National Water Well
Association, Dublin, OH.
Walton, W.C. 1989. Analytical Groundwater Modeling. Lewis Publishers, Inc., Chelsea, MI.
Walton, W.C. 1989. Numerical Groundwater Modeling: Flow and Contaminant Migration. Lewis
Publishers, Inc., Chelsea, MI.
Wang, H.F., and M.P. Anderson. 1982. Introduction to Groundwater Modeling. W.H. Freeman
Co., San Francisco, CA.
Ward, C.H., W. Giger, and P.L. McCarty (eds). 1985. Groundwater Quality. John Wiley &
Sons, Somerset, NJ.
Wilson, J.L., and P.J. Miller. 1978. Two-Dimensional Plume in Uniform Ground-Water Flow.
Journal of Hydraulics Div. A. Soc. of Civil Eng. Paper No 13665. HY4, pp. 503-514.
9/93 7 References
-------
APPENDIX C
Sources of Information
-------
SOURCES OF INFORMATION
SOURCES OF U.S. ENVIRONMENTAL PROTECTION AGENCY DOCUMENTS
Center for Environmental Research Information (CERI) (no charge for documents)
Center for Environmental Research Information (CERI)
ORD Publications
26 West Martin Luther King Drive
Cincinnati, OH 45268
513 569-7562
FTS 8-684-7562
Public Information Center (PIC) (no charge for public domain documents)
Public Information Center (PIC)
U.S. Environmental Protection Agency
PM-211B
401 M Street, S.W.
Washington, DC 20460
202 382-2080
FTS 8-382-2080
Superfund Docket and Information Center (SDIC)
U.S. Environmental Protection Agency
Superfund Docket and Information Center (SDIC)
OS-245
401 M Street, S.W.
Washington, DC 20460
202 260-6940
FTS 8-382-6940
National Technical Information Services (NTIS) (cost varies)
National Technical Information Services (NTIS)
U.S. Department of Commerce
5285 Port Royal Road
Springfield, VA 22161
703 487-4650
l-800-553-NTIS(6847)
Superintendent of Documents
Government Printing Office
202 783-3238
9/93 1 Sources of Information
-------
SOURCES OF MODELS AND MODEL INFORMATION
Superfund Exposure Assessment Manual
EPA/540/1-88/001, April 1988
Chapter 3 "Contaminant Fate Analysis" - 35 models
National Ground Water Association
National Ground Water Association
6375 Riverside Dr.
Dublin, OH 43017
614 761-1711
International Groundwater Modeling Center (IGWMC)
Paul K. M. van der Heijde, Director IGWMC
Institute for Ground-Water Research and Education
Colorado School of Mines
Golden, CO 80401-1887
303 273-3103
303 273-3278 (fax)
Groundwater Flow Model
Soil and Water Conservation Society (SWCS) Student Chapter
Iowa State University
3510 Agronomy Hall
Ames, IA 50011
515 294-7850
Cost: $384.00 (including shipping)
UST Video: Groundwater Cleanup
Industrial Training Systems Corp.
20 West Stow Road
Marlton, NJ 08053
609 983-7300
Cost: $595.00
Sources of Information 2 9/93
-------
GEOPHYSICS ADVISOR EXPERT SYSTEM VERSION 2.0
Gary R. Olhoeft, Jeff Lucius, Cathy Sanders
U.S. Geological Survey
Box 25046 DFC - Mail Stop 964
Denver, CO 80225
303 236-1413/1200
U.S. Geological Survey preliminary computer program for Geophysics Advisor Expert
System. Distributed on 3.5" disk and written in True BASIC 2.01 to run under Microsoft
MS-DOS 2.0 or later on IBM-PC or true compatible computers with 640k or greater memory
available to the program. No source code is available.
This expert system program was created for the U.S. Environmental Protection Agency,
Environmental Monitoring Systems Laboratory, Las Vegas, Nevada. The expert system is
designed to assist and educate non-geophysicists in the use of geophysics at hazardous waste
sites. It is not meant to replace the expert advice of competent geophysicists.
COMPREHENSIVE LISTING OF AERIAL PHOTOGRAPHY
U.S. Department of Agriculture, ASCS
Aerial Photography Field Office
2222 West 2300 South
P.O. Box 30010
Salt Lake City, UT 84130-0010
801 524-5856
9/93 3 Sources of Information
-------
0 0
0 0
(N/ «-
>•
err
CD
i
i
i
I
• . • •'
{-
lijlfe
.^:i;i>;i;i,ii
670
"?.
•:
:3V'i f
$
^,
w
O
O
00
*
O O O O O O
O O O O O O
r^ «c in ^ fo rsi
O
2 c
Contaminant Distribution - Dec. 1982
.; J; I';*!
• •*• >v»!»'(
: i|,^|3i
Sf*v . |>»
5S in
££::*$&«&?:&
£>mv
StfS
J?P
m
MR
i 1
It ^
fti
tlSuili
t!
f
®l.
:•:*!
1
li
U1*^
ftjil
^8gM^V:ii!
•^^^|;i«>"
>•
~--
670
cp
i
i
...
S
JT-;
jj
1
it1.
f>^<
...
660
.......
i|i:
jllf
;
;
;
>^ ••
: ':• f///i
^. ..-«,-...
650
u. ........
— „„.«..
=?wk
m
m$ i\ >-$&ffiffllff
HiP: :
ini !
im»i
• : :**
• •'
.... —
i
9
— -
>
^ •••
f~—
•
''660
1
640
:
^•:^i
;•;'
li— .....
L
w-iV^ITt
--
-i
>...*.....
i-
i
,^-
i
1V^
V"
r\>^
S
^ ' "^
-—
640
650
iH Amoco Free Produi
HI Amoco Dissolved
gg£3 UHaul Free Produc
^ Uhaul Dissolved
^3 Mixed Dissolved
:t
i i
• i
GRID FORM 1
5 600
cnn N
DUU n
400
300
200
100
100
200
300
400
500 S
. 600
. 700
BOO
900
- 100 35
-------
500 N
- 400
- 300
200
100
- 100
200
300
. 400
500 S
600
700
800
900
000 S
-------
o
o
fsl
o
o
*
o
o
o
o
o
OS
o
o
CO
o
o
r»
o
o
*e
*
o
o
in
o
o
V
O O
o o
« f>J
o
o
Recovery Operations -
y 600
500
- 400
. 300
. 200
.. 100
May 1983
Recovery
(45 ft)
-i- i-
•f !- >- !-
•630
+ 4 100
+—-----..• >.-.V
-f -!-
Recovery
1- -1- + -J- -=—>- -}-
Amoco Free Product
Amoco Dissolved
UHaul Free Product
Uhaul Dissolved
Mixed Dissolved
GRID FORM 1
800
900
1000
-------
600
500 N
- 400
. 300
. 200
... 100
July 1987
Recovery Operations
Amoco Free Product
Amoco Dissolved
UHeul Free Product
UHeul Dissolved
Mixed Dissolved
Disposed Soil Area
500 S
900
1000 S
-------
Table 2
Project No. 782563
SlUQ-ln Test Results
honltonng Well HW-6
Test No. 1
|tUpS« TTTWILWpSW IV..
R»v
Min
0.12
024
0.36
0.72
123
15
1.75
2.01
Seconds
72
14.4
21.6
T*WT
Otpth
Ft
2.09
2.51
2.76
432| 323
73.0
90.0
105.0
120.6
2.73J 163.0
325) 195.0
425
6.14
255.0
36S.4
7B3| 469.8
3.61
3.95
439
4.72
3.07
5.32
6.06
n
19.91
19.49
1924
18.77,
1839
16XJ5
17.61
1728
16.93
16.68
15.94
724J 14.76
8.12J 1338
VT-TU Art-no;
1JDO
0.96
0.93
0.88
0.84
0^0
0.76
0.72
0£S
0^6
058
0.45
036
951 1 570.6I 8.76J 1324J 029
1124| 674.4|
9.35| 12.65|
023|
r= 1 in.
L = 8.7 ft
R = 2.25 in.
K r 3.3E-6 ft/sec or l.OE-4 cm/sec
2.8E-1 ft/day or 1.0E2ft/yr
To= 466.7 sec.
Note: K is calculated based on Hvorslev Method (1951)
OD
-02'
-0^
-0.8
-1JO-
-12'
-1.4-
-\£-
100
200
300
Tvn»1n
400
5X
700
-------
Investigation Results - Phase
70
.40
Contamination
Based on
Phase 1 Study
Groundwster
Contours
Dec 1982
-------
500 N
100
TCEC - Phase 1 & 2
Diagram
200
300
400
500 S
600
700
800
900
1000 5
-------
FENCE DIAGRAM
LEVINGS GEOENVIRONMENTAL
ASSESSMENT
<^?&s
iist^i^sai^i^uiiaiJiji^iiiiiiiii^w:-,-;
tJtS**ff*M**^1)4*'l& -^--'"'"M-!^^^"""" • '.'•
•.Y-.V-.V«l51ipii»i
vy.v^uJiafe
ya:!!s:;va:!;i!'!H^^^^^
ss5^;!?fS^fefflii!2^H^K^g»fi^nE!nfa^^
2f^'5iK«=SSsSffiaa^SSSZa!.^=SiilSS5^^
SSaifiii^aSi
sm^s&tmm^f^s?m-^
^.^2a^^ns^y.:r.::|a:t«
^^gf?t^'^^"''-'!3^^"y5f*^"^
«:s2i>S'ii!JS!£SffiSSJ!E2JS£F"'
^gj^^ew^y^^^iivy^.-
-------
North-South Profile
Station 700 W
UHAUL
TANK\
!ttm«&*;KKS&
^8Silpg^||^ip||pi
^^i^liiiiissjiiiii^
CLAYEY SILT -FILL
SILT, TRACE CLAY -LOESS
CLAYEY SILT -ALLUVIUM
STRlcCELSAND -GLACIAL TILL
-i780
—-- 770
- 750
740
730
'20
710
U.S. GOVERNMENT PRINTING OFFICE: 1993- 300-571, / 83098
------- |