AN ABSTRACT OF THE THESIS OF
RICHARD WILLIAM COUCH for the DOCTOR OF PHILOSOPHY
(Degree)
(Name)
in
GEOPHYSICS
(Major)
presented on
J I9
DateY
Title: GRAVITY AND STRUCTURES OF THE CRUST AND SUBCRUST IN THE NORTHEAST PACIFIC OCEAN WEST OF
WASHINGTON AN
Abstract approved:
RITISH COLUMBIA
Redacted for Privacy
Gravity measurements made with LaCoste-Romberg surface
ship gravity meter S-9, along 20, 000 ku-i of track line, off Washing-.
ton an& British Columbia indicate the region is approximately in
isostatic equilibrium. Free-air anomalies over the region average
near zero except along the base of the continental slope. A -80 to
-100 mgalfree-air anomaly occurs west of the Queen Charlotte
Islands over the partially sediment-filled Queen Charlotte trough.
A free-air anomaly greater than -l50mgal occurs over the Scott
Is lands fracture zone off the north end of Vancouver Island. The
fracture zone, 170 to 240 km long, exhibits minor bathymetric
relief and is interpreted as a graben-like structure filled with
approximately 6 km of sediments. Five crustal and subcrustal
cross sections consistent with seismic refraction and gravity data
indicate that the Mohorovicic (Moho) discontinuity is at depths of
approximately 25 to 28 km in the Insular Belt of British Columbia
and 19 km in western Washington. In the areas of the Cascadia
Abyssal Plain and continental rise off British Columbia, the com-.
bination of near-zero gravity anomalies and thin crust indicates that
mantle densities are less than under the adjacent continent and Tufts
and Alaskan abyssal plains. Still lower densities occur in the upper
mantle immediately beneath the Juan de Fuca and Explorer ridges
where Moho depths are less than 7 km.
Gravity and Structures of the Crust and Subcrust in the Northeast
Pacific Ocean West of Washington and British Columbia
by
Richard William Couch
A TNESIS
submitted to
Oregon State University
in partial fulfillment of
the requirements for the
degree of
Doctor of Philosophy
June 1969
APPROVED:
Redacted for Privacy
Professor of Oceanogphy
in charge of major
Redacted for Privacy
Chairm\an of the Departr4ient of Oceanography
Redacted for Privacy
Dean of Graduate School
Date thesis is presented
Typed by Gwendolyn Hansen for
Richard William Couch
AC KNOW LEDGEMENTS
This research was performed under the direction of Dr. Peter
Dehlinger, Professor of Geophysics at Oregon State University. I
am grateful to Dr. Dehlinger for providing the opportunity to do this
research, for his many suggestions in the analysis of the data and for
editing the manuscript.
I am grateful to Messrs. E. Robey Banks and Michael Gemperle
for their assistance in making gravity measurements at sea and in
applying computer operations to gravity problems. Their willingness
and ability to help were a constant encouragement.
Appreciation is also extended to Dr. John N. Gallagher for
assistance in making gravity measurements at sea and for many
helpful discussions about the research.
Professor Gunnar Bodvarsson and Professor Tjeerd H.
van Andel reviewed the thesis and made many helpful comments.
This thesis reflects the patience and encouragement of Carol,
my wife.
During my tenure as a graduate student, I was the recipient of
a most appreciated National Aeronautics and Space Administration
fellowship.
This research was supported by the Office of Naval Research
under contracts Nonr 1286(09) and (10). For this assistance I am
very grateful.
TABLE OF CONTENTS
Page
INTRODUCTION
1
HISTORICAL DEVELOPMENT
3
THEORETICAL CONSIDERATIONS
6
PREVIOUS WORK
9
Physiography
Geology
Geophysics
Seismic Refraction and Reflection Measurements
Earthquake Observations
Heat Flow
Magnetics
Gravity
NEW GRAVITY DATA
Surface-Ship Gravity Measurements
Reduction of Gravity Data
Distribution of Gravity Measurements
The Free-Air Gravity Anomaly Maps
INTERPRETATION OF THE FREE-AIR GRAVITY
ANOMALY MAPS
Gravity Anomaly Trends
Isostatic Equilibrium in the Northeast Pacific Ocean
Fjords of the British Columbia Coast
Gravity Anomaly Trends in Dixon Entrance, Hecate
Strait and Queen Charlotte Sound
The Continental Shelf Off Washington and
Vancouver Island
The Randsenken Effect Along the Continental Margin
of Washington and British Columbia
The Queen Charlotte Trough
A Postulated East-West Fracture Zone Near
51° N Latitude
9
12
14
14
19
25
26
31
36
36
41
46
48
51
51
52
54
55
58
60
63
65
Page
The Scott Islands Fracture Zone
67
Seamounts
69
The Explorer Trough
The Cascadia Abyssal Plain
72
CRUSTAL AND SUBCRUSTAL CROSS SECTIONS ACROSS THE
CONTINENTAL MARGINS OF WASHINGTON AND BRITISH
COLUMBIA
82
84
Construction of the Crustal and Subcrustal Cross
Sections
Sources of Error in the Cross Sections
The Central Washington Cross Section
The South Vancouver Island Cross Section
The Scott Islands Cross Section
The Moresby Island Cross Section
The Dixon Entrance Cross Section
The Continental Margin Structures off Washington
and British Columbia
84
88
90
96
102
109
115
119
CONCLUSIONS
122
REGIONAL TECTONICS
124
BIBLIOGRAPHY
133
APPENDICES
148
Appendix 1. The Reduction of Gravity Measurements
Appendix 2. Supplementary Data
Appendix 3. Two Dimensional Gravity Computations
148
161
171
LIST OF FIGURES
Figure
1. Physiographic diagram of the northeast Pacific Ocean
off Washington and British Columbia.
2.
3.
Page
11
Location map of seismic refraction lines and heat flow
stations off the coasts of Washington and British
Columbia.
15
Earthquake epicenters in the Pacific Northwest.
21
4. Total magnetic anomaly map of the region west of
Washington and British Columbia.
28
Map of gravity meter track lines off the coasts of
Washington and British Columbia.
33
Free-air gravity anomaly map of the region off the
west coast of Washington and British Columbia from
the Columbia River to Cape Scott.
42
Free-air gravity anomaly map of the region off the
west coast of British Columbia from Cape Cook to
Dixon Entrance.
43
Bathymetric chart of the Explorer trough and Scott
Islands fracture zone.
74
Bathymetric, gravity and magnetic profile across
Explorer trough along line X-X'.
76
Bathymetric, gravity and magnetic profile across
Explorer trough along line Y-Y'.
77
Bathymetric, gravity and magnetic profile across
Explorer trough along line Z-Z'.
78
12.
Location map of crustal and subcrustal cross sections.
85
13.
Graph of the Nafe-Drake relation between compressional wave velocity and rock density.
87
5.
6.
7.
8.
9.
10.
11.
Figure
14. Central Washington crustal and subcrustal cross
section BB'.
15.
South Vancouver Island crustal and subcrustal cross
section CC'.
Page
91
97
16.
Scott Islands crustal and subcrustal cross section DD'.
101
17.
Free-air gravity anomaly map of the Scott Islands
fracture zone.
106
Total magnetic intensity map of the Scott Islands
fracture zone.
10.7
Bathymetric, gravity and magnetic profile across the
Scott Islandsfracture zone along line SS'.
108
Moresby Island crustal and subcrustal cross section
EE'.
110
Total magnetic intensity map off Moresby Island.
113
18.
19.
20.
21.
22. Dixon Entrance crustal and subcrustal cross sectithI
FF'.
116
23. Diagram illustrating equation parameters for com24.
puting gravity over a cone structure.
165
Polygon illustrating equation parameters for twodimensional gravity computations.
171
LIST OF TABLES
Page
Table
Information on the gravity anomaly size and gravity
station density in different physiographic provinces
47
2.
Seamount load and compensation parameters
.71
3.
Seamount locations, dimensions and gravity
170
1.
GRAVITY AND STRUCTURES OF THE CRUST AND SUBCRUST
IN THE NORTHEAST PACIFIC OCEAN WEST OF
WASHINGTON AND BRITISH COLUMBIA
INTRODUCTION
The northeast Pacific Ocean is an area of great interest to
geologists and geophysicists because of its diverse structures, particularly as these structures have global geotectonic implications.
Descriptions and studi.es of various parts of this region have been
made over a period of years.
Bathymetric recordings showed a smooth floor, termed by
Hurley (1960) the Cascadia Abyssal Plain, immediately west of the
generally narrow continental shelf and steep continental slope off
Washington and British Columbia. Between Tufts Abyssal Plain, an
area of smooth sea floor approximately 300 miles west of Washington
and British Columbia, and the Cascadia Abyssal Plain lies a region
of abyssal hills and mountains. This region, first termed the Ridge
and Trough Province by Menard (1955b),is now considered a seg-
ment of the world-encircling mid-ocean ridge and rise system
(Menard, 1960, 1964). In this same region Raff and Mason (1961)
made magnetic measurements which showed the now classical long
magnetic anomaly lineatilons. Vine and Mathews (1963) suggested that
the magnetic anomalies, which are now found paralleling the mid-
ocean ridge system in all oceans, represented a magnetic recording
of sea floor spreading, a still controversial hypothesis first
postulated by Hess (1962) and Dietz (1961). Only recently have suf-
ficient data become available to permit a study of the structures and
dynamics of this region.
The surface of the solid earth west of Washington and British
Columbia is covered by approximately 3. 5 km of water; consequently,
direct observations of the physical and structural form of the earth
in this region are limited. In this environment geophysical measurements provide valuable indirect observations which contribute greatly
to the understanding of the earth beneath the sea.
This study is an effort to provide a description of the earth's
gravity field in the oceanic region contiguous with Washington and
British Columbia and to relate these measurements to possible
structures of the earth' s crust and subcrust. The initial part of this
study is concerned with the construction of free-air gravity maps of
the region which help to delineate and delimit the major structures of
the region. The final portion of this study combines the gravity
information with other geophysical data, such as seismic refraction
measurements, to form hypothetical crust and subcrustal cross sections of the major geologic structures, notably the Juan de Fuca
Ridge- -a portion of the mid-ocean rise system- -and the continental
margin region where the relatively thin oceanic crust adjoins the
much thicker continental crust.
3
HISTORICAL DEVELOPMENT
In the early part of the twentieth century measurements of the
earth's gravity field were made with a torsion balance in the search
for oil and heavy minerals. The torsion balance, which actually
measures the horizontal gravity gradient, was supplemented in the
early 1930's by the pendulum, which was usually used to measure
relative values of the acceleration of gravity.. The gravinieter which
measures small differences in gravity has superseded both the
torsion balance and pendulum for obtaining gravity measurements,
particularly over limited areas.
The early use of gravity measurements in studies of the gross
subsurface structures of the earth was concerned with the isostatic
balance of the earth's outer layers. Hayford and Bowie (1912)
developed a method of determining isostatic gravity anomalies which
showed that, regionally, the continental U. S. is in isostatic
equilibrium. Measurements of gravity at sea began in 1923 with a
three-pendulum apparatus devised by Vening Meinesz (1929) which
operated in a submerged submarine. Measurements by Vening
Meinesz (1934) combined with extensive land measurements, indicated
that at least regionally the earth's outer layers are in isostatic
equilibrium. In addition, Vening Meinesz (1934) measured some of
the largest negative anomalies known; these occur along the arcuate
4
oceanic trenches.
These large negative anomalies, indicating exten-
sive mass deficiencies, were used to support the tectogene theory of
the down-warping of the earth's crust due to forces deep within the
earth.
Additional measurements of gravity at sea were made with
pendulums by the Lamont Geological Observatory of Columbia Univer-.
sity (Worzel,
1965)
over oceanic ridges, along trenches and across
the transition zone between continental and oceanic crusts. These
data were combined with geological and other geophysical data, par-
ticularly seismic refraction results, which established the depths of
the various layers of hypothetical crustal sections in the areas of
study. Talwani etal.
(1959b)
adapted computer techniques to the
calculation of crustal sections. Densities can be obtained from
crustal and subcrustal velocities determined by seismic refraction
studies by the use of empirical relations such as those proposed by
Nafe and Drake (1961) based on data from a large number of core
samples. Illustrative of this work is the cross section of the Island
of Puerto Rico and the adjacent trench (Talwani etal. ,
1959a),
which
shows large variations in the depth to the Mohorovicic (Moho) discontinuity.
These techniques were applied by Dehlinger etal.
(l967a)
to
the Mendocino Escarpment west of northern California and indicated
that in this region the material beneath the crust has, in addition to
5
variations in depth to the Moho, horizontal variations in density.
Dehlinger eta]. (1967b) combined their gravity data with seismic
measurements reported by Shor eta]. (1968) to analyze structures
across Juan de Fuca Ridge and the transition zone between the con-
tinental crust of Oregon and adjacent oceanic crust. This work
indicates a; relatively low mantle density beneath Juan de Fuca Ridge
and a relatively thin continental crust under western Oregon. The
techniques developed in this and other similar research are applied
in this study to the interesting and controversial area west of
Washington and British Columbia.
THEORETICAL CONSIDERATIONS
A force of mutual attraction exists between all bodies.
According to Newton's law, the force of attraction between two bodies
is directly proportional to their masses and inversely proportional
to the square of the distance between them. In equation form
m1m2
where F is the force of mutual attraction directed
r2
along a line joining the two point masses m1 and m2 which are
F=
separated by the distance r..
The proportionality constant
is
y
termed the universal gravitational constant and equals 6. 673 X
io8
(cg unit).
A force defined at every point in space describes a force field.
When a force field is continuous in an open region, and the integral
of the force along a path joining two points in the field is independent
of the path between the points, there exists a potential function
such that
F
U
vU (e.g. Apostol, 1957). Because work is independent
of the path,such a force field is termed conservative and indicates that
the force
is equal to the gradient of the potential
U.
The
earth's potential is usually assumed positive in geophysical studies.
Consequently, the difference in potential U1
between two points
is defined as the negative of the difference in potential energy W1
between the two points.
7
The Newtonian gravitational field is a conservative force field
and, therefore, all field parameters may be defined in terms of
U
y
the gravitational potential of the point mass
r
at the location of measurement of U. Thus, if mass is dis-
where
m1
U
tributed with density
throughout a volume V,
p (x, y, z)
gravitational potential at an exterior point P,
'x,y,z,
by the equation
U
x0 ,y0,z 0
=
S
'V
p(x,y,z)
r(x,y,z)
the
is given
dv.
v
The fundamental problem in the determination of the density
distribution in the earth's crust and subcrust is the inverse problem
of potential theory, i. e. determining the source from its potential
field. The inversion of the potential field equation above may be
accomplished by application of Gauss' theorem and vector calculus
(e.g. Wylie, 1951). Outside the body the potential satisfies Laplace's
Equation
V2U
equation VU
0.
Within the body the potential satisfies Poisson's
-4ir p (x, y, z). In geophysical research actual field
data are measurements of forceper unit mass. Since F
F = VZU
U,
and Poisson's equation may be written:
p (x, y, z) =
4ii
F(x, y, z)
This relation indicates that a unique solution to the inverse problem
or an exact description of the mass distribution requires knowledge
of the gravitation.l potential or its gradient throughout all space.
Because in actual field surveys the acceleration of gravity is known
only on a plane above the unknown distribution, a unique determina-
tion of mass-structural configurations is not possible from gravity
data alone. Nevertheless gravity data provide a powerful geophysical
constraint on the many possible crustal and subcrustal structure
models which might be postulated. The accepted range of rock
densities, the geological concepts of horizontal layering with the
increase of density downw3rd in the earth's crust and mantle and the
addition of other geophysical constraints, such as provided by
seismic and magnetic measurements, reduce further the number of
possible structural models and allow reasonable hypotheses to be
formed concerning subsurface structures and mass distributions
within the earth.
PREVIOUS WORK
All applicable geological and geophysical data must be con-
sidered to determine the structure of the region west of Washington
and British Columbia realistically. A considerable amount
of
geological and geophysical survey work has been done in this region,
particularly during the last decade. Detailed bathymetric mapping
has outlined the major geologic features, and bottom sampling has
indicated some recent and contemporary processes involved in their
modification. Geophysical measurements, most notably magnetic
and heat flow, show the region to have an unusual geologic history
while earthquake studies indicate that tectonic processes are still
continuing.
The following sections briefly outline previous geological and
geophysical studies of the earth's crust and mantle in the oceanic
region west of Washington and British Columbia.
Phys iograpy
Murray (1941) published summaries of the early bathymetric
measurements obtained by the U. S. Coast and Geodetic Survey in the
northeast Pacific Ocean. Menard and Dietz (1951) described later
soundings of the U. S. Coast and Geodetic Survey. Papers by Menard
(1955a and 1955b) and Gibson (1960) discuss the submarinetopography
10
of the Gulf of Alaska, showing the larger seamounts and structural
features of the region. Physiographic studies by Hurley (1960) and
Menard (1964) delineated and named the major physiographic
provinces of the northeast Pacific Ocean.
McManus (1964, 1967a,
1967b) has reviewed and extended these studies. Figure 1 is a
physiographic map (after Hurley, 1960) showing the larger features
of the area of study and their relationship to the North American
continent.
Dominating this portion of the northeast Pacific Ocean is a
section of the world-encircling Oceanic Rise system a chain of high
relief features extending from the continental margin off northern
California to the continental margin off northern Vancouver Island.
McManus (1964) named the segment of the oceanic rise west of
Washington and Vancouver Island, Cobb Rise. The Juan de Fuca
Ridge, determined primarily from magnetic measurements (Vine and
Wilson, 1965), lays along the eastern sector of Cobb Rise. Between
Cobb Rise and the continental slope off Washington and Vancouver
Island is the depositional topography of Cascadia Abyssal Plain.
The Alaskan Abyssal Plain adjoins the Cobb Rise on the northwest
A seamount chain beginning west of the Queen Charlotte Islands
extends northwesterly across the Alaskan Abyssal Plain to the
Aleutian Trench southeast of Kodiak Island, Alaska. West of Cobb
Rise is the Tufts Abyssal Plain, separated on the north from the
II
WELKER
GUYOT
55
11
I
I
/
'ç'"
"
IU'
125°
I
I
I
PHYSIOGRAPHIC DIAGRAM
SEAMOUNT'BAKER FAN_\
(
ALASKAN
2/
I
I
OFTHE
NORTHEAST PACIFIC OCEAN
R. J. HURLEY 960
DICKENS
PSEAMOUNT
o
550
>' TOPOGRAPHY AFTER
( 4'c_
GRAHAM
Li
4
#10
'l
IfflhJjfljjJ
MOUNTAINS
HILLY AREAS
FRACTURE ZONE
MORESBY
BOWIE
A B VS SAL
ISLA
S BANK
' \,
OSHAWA
IO,.
,
SEAMOU T
WHITNEY RIDGE
PLAIN
QUEEN
CNARLOTTE
50
CMT0NTRS
\CHARLOTTE SOUND
)
SCHOPPE RIDGE
SCOTT
FRAC. ZONE
-
..
.
RIDGE
0°
,
-.
M TS.
FAN
TUFTS ABYSSAL PLAIN
JASCADIA
JJi11ABYSSAL
4R
I
1400
I
I
I
I
135°
I
I
COL R
JLIL
130°
L...............L...................._.._.......L...............L.I
Figure 1. Physiographic diagram of the northeast Pacific Ocean off
Washington and British Columbia (after Hurley, 1960).
ioce
.50
12
Alaskan Abyssal Plain by the Whitney, Schoppe and Peters ridges.
The Alaskan and Tufts abyssal plains have water depths near 4 km
while the Cascadia Abyssal Plain has depths of about 3 km. The high
relief of Cobb Rise extends in its most prominent feature;-Cobb
Seamount- -to within 40 meters of the surface. More typically, however, Cobb Rise has water depths near 2. 8 km and displays elongate
hills and furrows of less than 200 meters of relief which parallel the
rise crest.
The continental slope, described by Shepard (1938), Hurley
(1960), McManus (1964), and McManus and McGary (1967)features
continental borderland topography west of Washington and glaciated
topography north of the Strait of Juan de Fuca. West of the Queen
Charlotte Islands the steepness of the continental slope suggests a
fault 5 carp. Glaciation has produced substantial relief in the Dixon
Entrance, Hecate Strait and Queen Charlotte Sound portion of the con-
tinental shelf and adjacent fjords of British ColumbIa. The continental shelf west of Vancouver Island and Washington is relatively nar-
row, glaciated and incised with numerous submarine canyons.
Geology
Very few studies of dredged rock samples are reported for the
region west of Washington and British Columbia. The studies are
generally restricted to rocks obtained from seamounts of the deep
13
ocean.
Nayudu (1962) and Engel and Engel (1963) describe olivine-rich
to olivine-poor basalts from CobbSeamount. Palagonite tuff breccia
was also obtained off Cobb seamount and from Bowie Seamount off
the Queen Charlotte Islands (Nayudu, 1962). Nayudu, Chastam and
Woodruff (1968) report studies of prophyritic, vesicular and fine grained basalts and palagonite tuff breccia dredged from Eickelberg
Ridge.
Budinger (1967à)pointed out that the basalts of high pyroxene
content from Cobb searnount, Mendocino Ridge and the experimental
Mohole site (off southern California) differ chemically not only from
the nearby continental basalts of the Pacific Northwest (Waters, 1955)
but also from other basalts in the Pacific Basin (Engel and Engel,
1963).
Budinger and Enbysk (1967b) have reported a crushed whole
rock potassium-argon age of 27 ± 6 m. y. for a basalt boulder dredged
from 380 meters depth on the northeast slope of Cobb éeamount.
Dymond eal (1968) report a whole rock sample potassium-argon
age of 1.6 m.y. for a basalt boulder collected from the top of Cobb
eamount at a depth of 34 meters. The 27 m. y. age is difficult to
reconcile with the 3.5 m.y. age assigned to Cobb seamount by
magnetic dating (Vine, 1966),whereas the 1.6 my. age is in reasonable agreement.
14
Menard (1955a) and Hurley (1960) described sediment dispersal
in the region. Cobb Rise forms a barrier to western transport of
terrigenous sediments which, in the northern Cascadia Abyssal Plain,
are derived principally from Vancouver Island and possibly from the
Queen Charlotte Sound (Duncan, 1968). The general distribution of
the sediments in this region has been discussed by Hurley
(1960). Nelson (1968) described sediments onthe Astoria
Fan. Royse (1967) and Nelson et al, (1968) described the occurrence of Mazama ash, a key sediment time horizon and indicator of
sediment transport, on the continental slope of Washington and Oregon
and in the Cascadia Abyssal Plain.
Geophysics
Seismic Refraction and Reflection Measurements
Shor (1962), Milne (1964) and Shoretal. (1968) reported the
results of seismic refraction measurements in the oceanic region
west of Washington and British Columbia. Tatel and Tuve (1955)
and Richards and Walker (1959) presented the results of seismic
refraction measurements made in the adjacent continental region.
The locations of these measurements are shown in Figure 2. Milne
and White (1960) and White and Savage (1965) reported additional
seismic measurements about Vancouver Island.
8°
56
136°
I°
134°
30°
28°
126°
24°
122°
20°
118°
116°
I
I
114°
F
,f'
SEISMIC
56°
REFRACTION PROFILES
HEAT
410
1120
FLOW VALUES
Mk23
I 54°
4° I
20
60
1520
UkI8
2°I
I 50
500l
I'
I 48
262
8° I
S
G7
7
I
080
46
46°I
58
SF4
515'
I 56j 73
2
6
/
42°
40°
138°
136°
34°
132°
63
130°
128°
26°
24°
22°
20°
________ 44°
118
116°
114°
112°
Figure 2. Location of seismic refraction lines (Shor, 1962; Mime, 1964;
Shor etal. 1968) and heat flow stations (Foster, 1962; Von
Herzen, 1964; and Mesecar, 1967) off the coasts of Washington
and British Columbia.
L71
L1
Reversed seismic refraction lines G 13 through G 18 of Shor
etal. (1968) across Juan de Fuca Ridge show that the ridge is a surface expression of an upbowed crust and mantle. The crest and flanks
of the ridge exhibit low mantle velocities. Mantle arrivals over the
ridge crest are weak or absent suggesting that the Moho is unformed
or discontinuous near the crest.
The sediments in Cascadia Abyssal Plain are thickest near the
base of the continental slope. Mantle depths beneath Cascadia
Abyssal Plain, however, are relatively shallow 10 to 11 km where
water depths are 2-i/ kilometers. Refraction work on the continental shelf suggests a structural basement high near the outer edge of
the slope and large variations in layer and crustal thickness along
the Oregon-Washington shelf as well as across the shelf.
Milne (1964) shot a reversed refraction profile near the southem
edge of the Alaskan abyssal plain which indicated an abnormally
thin crust of 11.0 km. The reversed refraction profiles reported by
Shor (1962) west of the Dixon Entrance show that although the sedi-
mentary section thickens, toward the continent, the Moho is at depths
of 10 to 11 km. Shor (1962) also noted that the Moho becomes
shallower near the foot of the continental slope, similar to the outer
ridge of many. trenches, and suggested that a buried trench may lie
between the ridge and the slope. He also indicated that large. varia-
tions in thickness of the sedimentary layer in Dixon Entrance implies
17
large-scale faulting. Mime (1964) made similar seismic observations in Dixon Entrance.
White and Savage (1965) report an unreversed profile based on
the April, 1958,1. 5 kiloton Ripple Rock explosion in Discovery Pas-
sage, British Columbia. This profile was interpreted to indicate an
average P wave velocity of 7.8 km/sec and a corresponding average
crustal thickness of 31 kilometers beneath the Coast Range, the
Interior Plateau and the Rocky Mountains of British Columbia. No
intermediate layer refracted arrivals were noted. Extensive refraction studies (Mime and White, 1960; White and Savage, 1965) in the
vicinity of Vancouver Island indicate that considerable variation
occurs in both velocity and thickness of upper crustal structures.
Richards and Walker (1959) reported the results of a reversed
seismic profile shot in the Alberta Plains of western Canada. This
work indicates a depth of 43 km to the mantle, an average Pn velocity
of &. Z km/sec, a depth of Z9 km to the intermediate layer, and a
P* (intermediate layer) velocity of 7. 2 km/sec.
From their refraction studies in Washington, Tatel and Tuve
(1955) conclude that the crust is relatively thin, 19 km, beneath
Puget Sound and that the Pn velocity is 8. 0 km/sec. East of the
sound they conclude that the crust is 30 km thick. The 30 km depth
agrees well with the depth given by White and Savage (1965) for the
Coast Range about 250 km to the north.
Hess (1964) attributed variations in mantle velocities in the
Pacific Ocean to velocity anisotropism. He suggests (Hess, 1965)
there may be a 5% to 10% azimuthal velocity variation due to aniso-
tropic propagation, such that low velocity paths should parallel the
magnetic anomalies and high velocity paths lie perpendicular to the
magnetic anomalies (and consequently the active mid-ocean ridge-
rise systems). In the region of study the magnetic lineations
(Figure 4) trend generally north-south while nearly all the seismic
refraction profiles (Figure 2) trend east-northeast. Since the
seismic refraction profiles intersect the magnetic lineations at
nearly a constant angle in this region it is unlikely that velocity
variations along the same line are due to anisotropic propagation;
rather, they are more likely to be a function of temperature,
petrology or interface irregularities.
Hamilton (1967) described results of seismic reflection studies
in the Gulf of Alaska, which showed sediment thicknesses of 680
meters in the southeast part of the Alaska Plain and 470 meters in
northern Tufts Plain. Turbidite thickness decreased with distance
from the continental sources. Reflection profiles (Hamilton and
Menard, 1968) across the northern end of Juan de Fuca Ridge to the
continental terrace off Washington indicate 100 to 200 meters of
turbidites near the eastern edge of Juan de Fuca Ridge, 600 meters
near the center of Gas cadia Plain and 1600 meters near the base of
19
the continental slope. Hamilton and Menard (1968) concluded that
probable uplift of the ridge occurred after turbidite deposition began.
Hamilton (1967) also noted that the turbidite layers are undistorted,
contrary to what might be expected from crustal spreading, and concluded that the whole area has been quiescent since-middle Tertiary.
Bennett and Grim (1968) and Grim etal. (1968) have described
seismic reflection observations on the continental shelf off
Washington, which show a broad spectrum of tectonic activity.
Broad folds are observed on La Perouse bank; flat-lying beds exhibiting high angle normal faults occur west of the Olympic Peninsula;
and gentle folds and piercement folds with exposed diapiric cores
are noted south of Quinault Canyon.
Earthquake Observations
The oceanic region west of the coast of North America, between the Columbia River and Dixon Entrance, is part of the great
circum-Pacific seismic belt. The early compilation of Gutenberg
and Richter (1941) indicated that the circum-Pacific seismic belt
branches in the vicinity of southern California. The westerly branch
of the belt, consisting of predominantly shallow earthquakes, passes
out to sea in the vicinity of Cape Mendocino, California and re-
enters the continental region in the vicinity of the Queen Charlotte
Islands and southeast Alaska.
Recent compilations by McManus
20
(1965), Mime (1967), Bolt etal. (1968), and Tobin and Sykes (1968)
indicate that the earthquake epicenters of this region show preferred
trends which appear to relate to known physiographic features.
Figure 3 shows the distribution of earthquake epicenters for this
region which occurred between 1899 and 1960.
In southern California the earthquake activity of the continental
margin region of North America is associated with the San Andreas
fracture system. Northward, the earthquake activity extends along
the coast of northern California, across the Mendocino Escarpment
and then northwesterly along the Blanco fracture zone. After an
apparent offset, earthquake activity begins anew west of Vancouver
Island. It then extends northwesterly along the seaward side of the
Queen Charlotte Islands into southeast Alaska where it appears to
join the Fairweather-Chugach-St. Elias and Denali fault systems
(Benioff, 1954, 1962; St. Amand, 1957; Tobin and Sykes, 1968).
Compilations of earthquake activity in Washington by Rasmussen
(1967) and in Oregon by Berg and Baker (1963) indicate the relative
quiet of southern Washington and Oregon. This area of anomalous
quiet between belts of moderate earthquake activity apparently extends
westward over Cascadia Plain. The major portion of the earthquake
energy released in the northwest North America portion of the
circum-Pacific seismic belt occurs along the shelf-slope region of
southeast Alaska and British Columbia. Lesser amounts of energy
21
YUkON
ALASkA
XX
x
6.
X
MAGNITUDE
U.S
M.T
S
SSITISN
'L.
X
COUINSIA
55.
X
N
N
U
5
X
X
X
QUEEN
CNAIILOT1E
ISLANDS
X
.X
5o
X
XX
X_..... -
X
X
...
I$.AI
X
X
X
X
GET
$01110
.
45.
X
x
N
X
X
N
.N
.
N
N
X
XX
S
N
N
N
..
N
X
N
x
N
X
S
XX.
X,X
N
X
X
CAPE
BLANCO
S
XX
X
X
.
S
N
N
o.
N
X
135.
I
125
I
XXX
X
X S CAPE
Xk5 5J
130
S
X
X
N
120
I
Figure 3. Earthquake epicenters in the Pacific Northwest
(after Mime, 1967).
X
22
are released in California and in Puget Sound, Washington (Mime,
1967).
The Juan de Fuca Bidge area is comparatively aseismic except
at the extreme northern end. Epicenters of numerous small earthquakes at the north end of the ridge appear to be continuous with the
band of earthquake activity associated with the Explorer Ridge-Scott
Island fracture zone-Queen Charlotte Trough complex. It is difficult
to assign the earthquake activity of this region to any one physlographic feature. One possible exception is the concentration of earthquakes located by Tobin and Sykes (1968) near 500 N lat., 129. 6° W
long. Precision Depth Recorder (PDR) records indicate that, at this
position, the Explorer Ridge exhibits a'rift valley-like structure.
Recent faulting is suggested by the absence of sediment on tilted
benches evident along the trench walls. The PDR records also show
that approximately 40 to 50 km southeast of the north tip of the trench,
ponded sediment is severely tilted.
The
absence or distortion of
sediment in this area suggests recent or contemporary tectonism
probably related to the observed seismicity.
The linaments exhibited by the earthquake epicenters of Figure
3, particularly along the Queen Charlotte Islands, suggest transcurrent faulting. However, in discussing the predominance of trans-
current faulting (determined from fault-plane solutions) in circumPacific earthquakes, Hodgson (1957) notes that the zone from
23
southeast Alaska to California is unique in that all types of faulting
occur.
Six fault-plane solutions have been reported in the region of the
continental margin between Cape Mendocino and the Columbia River
(Tobin and Sykes, 1968; Hodgson and Milne, 1951; Hodgson and
Storey, 1954; A1germissen et a1., 19 65), Epicenters of the
April 13, 1949,and April 29, 1965,earthquakes are located in the
southern Puget Sound area. The preferred fault plane solutions for
these earthquakes indicate thrust faulting for the April 13, l949,.
shock south of Tacoma, Washington (Hodgson and Storey, 1954) and
normal faulting for the April 29, 1965, shock near Seattle, Washington
(Algermissen etal., 1965).
The epicenter of the earthquake of June 23, l946,is in the Strait
of Georgia (between Vancouver Island and the mainland). The fault
plane solutions of this earthquake indicate normal faulting. The pre-
ferred solution contains a right lateral strike-slip component with the
fault plane oriented N 23° W and dipping 600 east (Hodgson and Mime,
1951).
The earthquakes of August 22, 1949, and March 31, 1964,
occurred in the Alaskan Abyssal Plain immediately west of Queen
Charlotte Sound. The fault-plane solution of the August earthquake
indicates normal faulting with the more probable solutions oriented
N 56° E and dipping 67° east (Hodgson and Mime, 1951). A right
24
lateral transcurrent component of motion is associated with the preferred solution. Tobin and Sykes (1968) have indicated a transcurrent
fault solution for the March earthquake. The fault-plane solution for
right lateral motion indicates a strike of N 15° W.
The earthquake of August 22, 1949, was located in Dixon
Entrance between southeast Alaska and the Queen Charlotte Islands.
The corresponding fault- plane solutions indicate transcurrent faulting.
The preferred solution has a strike of N 29° W, a dip of 77° east and
exhibits right lateral motion. This solution is consistent with
observed faulting visible on adjacent land masses (Sutherland-Brown,
1966) and with the motion of the Queen Charlotte fault postulated by
Benioff (1954), St. Amand (1957), and Wilson (l965b).
Stauder (1967) has noted that the limitations imposed by seismic
wave propagation and the geographic distribution of seismic stations
favor transcurrent fault-plane solutions, particularly in cases of
shallow focus earthquakes. Tobin and Sykes (1968) state that in their
earthquake mechanism solutions strikes and dips may be uncertain
by as much as 300.
The limited number of fault-plane solutions with their expected
uncertainty are probably insufficient to describe the tectonics of this
region. They do suggest, however, that the continental margin be-
tween Cape Mendocino and Dixon Entrance is a region of block faulting
between two regions, southeast Alaska and California, which exhibit
25
predominantly right lateral trans current faulting.
Heat Flow
Foster (1962), Von Herzen (1964), and Mesecar (1968) have
reported heat flow values for the region west of Oregon, Washington,
and British Columbia. A composite map of these data is shown in
Figure 2. No geophysical interpretation of the heat flow data or
correlation of these data with other geophysical measurements has
been reported. The average of 25 heat flow values between 44° N
lat. and 55° N lat. is 2, 45 cal/cm2/sec, approximately double the
presently accepted world average.
Von Herzen and Uyeda (1963) noted that the highest heat flow
values observed over the East Pacific Rise occur in two narrow zones
approximately parallel and symmetrically oriented with respect to the
rise crest.
They suggest these observations can be accounted for by
the existence of an upward-moving mantle convection current beneath
the crest of the rise and lateral spreading of the material near the
surface.
The values of the limited number of heat flow stations within a
region extending 300 km to either side of Juan de Fuca Ridge show
an asymmetry. The average of nine values west of Juan de Fuca
Ridge is 1.89 cal/cm2/sec. The average of 16 values east of the
ridge is 2. 77 cal/cm2/sec. Duncan (1968) has indicated average
26
rates of deposition for Cascadia Plain during Pleistocene time of 0. 12
to 0. 18 cm/yr, and rates of 0. 02 to 0. 003 cm/yr during the postglacial period. The high deposition rates for Pleistocene time
occurred in the area of high heat flow, suggesting the equilibrium heat
flow rates may be 10 to 40% greater than those observed. A higher
heat-flow rate, east of Juan de Fuca Ridge, is also suggested by the
anomalously high value of 7.27 cal/cm2/sec near the north end of
the ridge, in an area which has had a lower sediment deposition rate.
The differences in the average heat-flow rates east and west of
Juan de Fuca Ridge appear significant and suggest major differences
in either temperature or conductivity of the crust or subcrust in these
areas.
The anomalously low heat flow value (0.80 cal/cm2/sec) west
of Juan de Fuca Ridge may be explained as due to a locally disturbed
environment (Von Herzen, 1964). Heat flow values in the proximity
of the Queen Charlotte Trough appear normal; however, west of
Dixon Entrance, they appear unexplainably high.
Magnetic s
Raff and Mason (1961) and Vacquier etal. (1961) have presented
results of an extensive magnetic survey in the Northeast Pacific which
extends from 40 N lat, to 520 N lat. and from the continental slope
0
of Oregon, Washington, and British Columbia westward approximately
27
400 km. Emiliaetal. (l968a) presented magnetic maps which extend
the magnetic survey between Cape Mendocino and the Strait of Juan de
Fuca eastward over the continental slope and shelf.
Magnetic measurements were made over the Queen Charlotte
Trough and Scott Island fracture in conjunction with Oregon State
University gravity surveys. Total field magnetic intensity maps based
on these data are shown in Figure 18 and Figure 2 1. Magnetic
measurements were also obtained along two tracklines west of
Graham Island in 1966 during the YAQUINA long cruise I (YALOC I).
These two lines, combined with two lines west of Graham Island
reported by Pitman and Hayes (1968), indicate the general magnetic
pattern of the area west of Graham Islard. The data from the above
surveys, compiled and shaded in Figure 4, show the magnetic anomaly
pattern of the region of the Northeast Pacific discussed in this thesis.
The magnetic anomalies, delineated by these surveys, are long
linear features having a predominantly north-south trend. Two of the
magnetic lineations observed over Cascadia Plain extend landward to
the shelf break, apparently crossing the continental rise and slope
unaltered. Calculations of source depths (Raff and Mason, 1961;
Emilia et al.,, 1 968a) indicate large variations in depth to anomaly
sources. These depths range from the water-crust interface to near
the crust-mantle boundary with a tendency for the source depths to
increase landward. Emiliaetal. (1968b) explained the nearshore
28
1400
135°
130°
125°
550
120°
55
50°
500
450
45°
140°
135°
130°
125°
Figure 4. Total magnetic anomaly map of the region west of Washington
and British Columbia (a compilation from Raff and Mason,
1961; Vacquier etal. 1961; Emilia etal. , 1967a; Pitrnan and
Hayes, 1968, see text).
120°
29
magnetic anomalies as due to probable seaward extensions of coastal
volcanics,
Talwanietal. (1965) noted that a magnetic anomaly profile
obtained across Juan de Fuca Ridge was similar to magnetic profiles
obtained over the East Pacific Rise in the South Pacific. They concluded that Cobb Rise is part of the mid-ocean ridge system with "the
important difference that the eastern flank of the rise is under the
continent." Vine (1966) interpreted the magnetic anomaly pattern as
reflecting sea-floor spreading, contemporaneous with reversals in the
earth's paleomagnetic field. In this hypothesis the spreading axis west
of Washington and Vancouver Island is represented by Juan de Fuca
Ridge. The sea floor is thought to spread at equal rates (approxi-
mately 2.9 cm/yr) from the N 300 E trending axis of the ridge toward
the continent and oceanward. Large- offsets of the spreading axes are
interpreted as due to transform faulting (Wilson, 1965a, 1965b). This
hypothesis has been extended by Heirzleretal. (1968), Pitman and
Hayes (1968), and Pitmanetal. (1968).
These investigators indicated
the ages of each of the magnetic anomalies as based on comparative
spreading rates of other mid-ocean rises and rifts.
Morgan (1968a, 1968b) postulated that the separating of large
crustal blocks generates the observed magnetic lineations. He suggests that a Northeast Pacific crustal block is moving northwesterly,
away from the North American continent crustal block toward the
III]
Aleutian and Kurile-Kamchatka trenches. The fracture of separation
is indicated by Gorda and Juan de Fuca ridges and the Blanco fracture
zone. Numerous oblique faults (or lineations) offset the magnetic
patterns (Raff and Mason, 1961). Morgan (1968b) interpreted these
lineations as indicating small blocks moving independently of each
other, Pavoni (1966) suggested that these lineations are due to
wrench faulting, mostly left lateral, which dissected an older crustal
unit. An oblique fault is noted offsetting the anomaly which defines the
spreading axis of Juan de Fuca Ridge. This suggests that at least
some of the oblique faulting has occurred in recent time.
Although the interpretation of the magnetic anomalies of Juan de
Fuca Ridge suggests a presently active spreading ridge (Vine and
Wilson, 1965; Pitman and Hayes, 1968) other geophysical evidence
suggests it is probably quiescent (Hamilton, 1967; Tobin and Sykes,
1968).
Bostrom (1967) has suggested that the spreading axis between
the Blanco and Queen Charlotte transform faults is presently being
reactivated beneath western Oregon and Washington.
The gradients and intensities of the magnetic anomalies imply
a crustal source. Model studies by Mason (1958) suggest an anomaly
source located in the crustal layer beneath the sediments. Vine and
Wilson (1965) suggest the origin is predominantly due to remanent
magnetization existing in the oceanic basalt layer wherein juxtaposed
normal and reversely polarized rock generate the observed steep
31
magnetic gradients. Theoretical analysis by Bott (1967) supports the
interpretation of a shallow source coupled with normal and reverse
magnetization. Ozima et al, (1968) have noted a more than 3 fold
increase in saturation magnetization of submarine basalts when heated;
they suggest that the magnetic lineations may reflect the effect of
localized heating of the basaltic crust.
Gravity
Harrison etal, (1957) and Worzel (1965) have published pendu-
lum measurements west of Washington and British Columbia. No
analysis of these data has been reported. It is interesting to note that,
although the region west of Washington and British Columbia exhibits
large positive and negative gravity anomalies, none of the 43 measure-
ments of Harrison etal, (1957) or Worzel (1965) are located so as to
indicate the existence of these anomalies.
Woollard and Strange (l96Z) have published regional free-air
and Bouguer anomaly maps of the Pacific Basin, Estimates of the
depth to mantle, based on these gravity data, were presented.
The USC&GS (Erickson, 1967) has made
surface ship measurements along a gravity meter check line west of
Cape Flattery, Washington. The Dominion Observatory of Canada
(Tanner, 1966), in a continuing gravity program, has made bottom
gravity meter measurements in the region of Dixon Entrance, Hecate
Strait and Queen Charlotte Sound. No analysis of these data are
reported and the gravity values are unpublished. Figure 5 shows the
station positions of the pendulum measurements of Harrison et al.
(1957) and Worzel (1965), the position of the USC&GS surface ship
measurements (Erickson, 1967), and the area of bottom meter
measurements of the Dominion Observatory.
Dehlinger etal. (1966a) have reported gravity measurements
along the Inside Passage of Alaska and British Columbia.
These
measurements show a rapid decrease in free-air anomalies toward
the continental interior in traverses along the fjords. Banks (1969)
reduced 104 of these rneasurementsto isostatic and Bouguer
anomalies. His results indicate a small residual tendency of the
gravity anomalies to decrease toward the Coast Range mountains.
He also notes a change in the probable depth of compensation north
of Juneau, Alaska and south of the San Juan Islands, Washington.
Figure 5 shows the locations of the sea gravity measurements
described above.
Heiskanen (1938) and the Dominion Observatory of Canada
(Stacy, 1967) reported land gravity values in the nearshore regions
of British Columbia including the Queen Charlotte Islands. Although
Heiskanen (1938) presented isostatic anomalies, no other analysis
was reported.
Walcott (1967) presented a Bouguer anomaly map of southwestern
33
140
56
J38
136
134
132
130
128
26
24
22
156
54
.4.
-52
50
0
48
4611
40
__-_i.ih
Figure 5. Map of surface ship gravity meter track lines off the coasts of
Washington and British Columbia. Map also locates underwater
gravity meter measurements of the Dominion Observatory of
Canada, pendulum stations of Harrison (1957) and Worzel (1965)
and computed seamount values (App, 2).
34
British Columbia which includes results of previous work in the area
by Miller (1929), Garland and Tanner (1957), White and Savage (1965),
and Dehlingeretal. (l966b). In his analysis of the gravity anomalies
he concluded that the coastal anomalies suggest a structural analogy
between Vancouver Island and the Queen Charlotte Islands. He also
presented a gravity profile from Vancouver Island across the Strait
of Juan de Fuca to the Olympic Peninsula. This profile shows a steep
gravity gradient over the Strait of Juan de Fuca which Walcott (1967)
suggests indicates large differences in crustal thicknesses on either
side of the strait.
Woollard and Rose (1963) and Sauers (1966, unpublished) have
presented Bouguer anomaly maps of Washington without interpretation.
Stuart (1961) constructed a Bouguer anomaly map of the Olympic
peninsula and Puget Sound of Washington and presented a two-
dimensional crustal model that extends from the Olympic Mountains
to the Puget Sound. His model shows a high density block of volcanic
rocks extending from the Olympic Mountains where they are exposed
to the area between Seattle and Tacoma, where they lie 16 to 20 km
deep.
Danes etál. (1965) from a detailed gravity survey of the southern
Puget Sound area, concludes that the entire Puget Sound area is a
major regional gravity low. He presents a north-south cross section
to illustrate his analysis, which shows large variations in the depth
35
to Moho.
Dehlingeretal. (1968) have combined gravity data west of
Oregon, land gravity data in Oregon and Washington (Thiruvathukal,
1968; Sauers, 1966 unpublished) and layer depths and velocities
obtained from seismic refraction work (Shor etal.
,
1968; Hill and
Pakiser, 1966) to construct two-dimensional crust- subcrustal cross
sections. These cross sections extend from Tufts Abyssal Plain
eastward across Juan de Fuca Ridge and Oregon to central Idaho.
One cross section passes through southern Washington and the other
through north central Oregon. These postulated cross sections
indicate the existence of a low-density mantle beneath Juan de Fuca
Ridge, a relatively thin crust (23 km) in western Oregon and a crust
which thickens to more than 40 km beneath the Cascade Range,
eastern Oregon and western Idaho.
NEW GRAVITY DATA
Surface-Ship Gravity Measurements
The important sea gravity measurements described in the
previous section, although of use in geodetic studies, are insufficient
in detail to delineate the major anomalies and do not provide adequate
control for structural analysis. The Department of Oceanography at
Oregon State University made surface ship gravity measurements
from 1964 through 1967 to provide a sufficient density of gravity
stations to map the area off Washington and British Columbia. These
measurements, combined with adjacent land measurements, provide
control for an analysis of the major structures. The initial traverses
made in 1964 and 1965 were made perperidicularto the coasts of
Washington and British Columbia in anticipation of the structural
form of the continental margin. These early measurements, how-
ever, revealed new structures, structural trends and areas exhibiting
unexpectedly large negative anomalies. To delineate and study these
gravity features in detail, additional measurements were made during
1966 and 1967 along traverses oriented generally perpendicular to the
observed anomalies. Figure 5 shows the surface-ship gravity meter
track lines.
Gravity measurements were made with LaCoste-Romberg
37
gimbal-suspended,surface-ship gravity meter S-9. Gravity meters
operating on moving platforms are subjected to horizontal and verti-
cal accelerations in addition to the acceleration of gravity. These
accelerations may be as much as several magnitudes greater than
the anomaly sought, To achieve an accuracy of a few parts per mu-
lion, three corrections are made to the gravity meter measurement:
1) a horizontal or Browne correction for lateral accelerations
acting on the meter; Z) a vertical correction for nonlinear
responses of the gravity meter under high vertical accelerations and
3) an Ebtvbs correction for the change in the earths centrifugal force
due to movement of the ship along the earth's surface. Each of the
three corrections is subject to uncertainties. These uncertainties
plus the uncertainty in ship position, due to navigation inaccuracies,
limit the accuracy of these gravity measurements to about ± 5 mgal
(Dehlinger etal., 1966à). Uncertainties in the best measurements
are estimated to be ± 3 mgal.
The gimbal suspension of meter S-9 permits the gravity meter
sensing element to swing freely in response to horizontal ship
accelerations.
The horizontal accelerations cause the meter to
swing or be deflected from the vertical. The sensing element of the
gravity meter thus measures the vector sum of gravity plus the
negative of the horizontal acceleration, Gravity is obtained from the
solution of the relati.on g0
2
- g
where
g = an average
value of gravity for the earth,
g
measured gravity and 8 = angle
of deflection of the sensing element from the vertical. This equation
is known as the Browne correction or horizontal correction (Browne,
1937) and is applicable for long-period accelerations. LaCoste and
Harrison (1961) have presented a comprehensive analysis of surfaceship gravity meter operations involving simultaneous periodic hori-
zontal and vertical accelerations. Their analysis, which included
fourth order terms, showed the Browne correction equation is correct for the LaCoste and Romberg gimbal suspended meter, provided
0 remains small and that gravity values are obtained over a length
of time that is long compared with ship acceleration periods.
The measurement of the horizontal acceleration or Browne
correction is accomplished by measuring the angle 0 or the deflection of the meter from the vertical. This measurement requires a
stabilized reference which, in meter S-9, is provided by a longperiod inertia bar (LaCoste, 1967). The free period of the inertia
bar in meter S-9 is two minutes. Dehlinger (1964) described the
reliability of gimbal suspended gravity meters at sea and the effect
of damping of the inertia bar with particular reference to meter S-9.
He noted that measurements can be accurate to within ± 3 mgal at
Browne corrections up to 300 mgal and that errors became
systematically more positive with increasing Browne correction. A
statistical study by Dehlinger etal. (l966a) indicated the root mean
39
square (rms) uncertainty in the S-9 Browne correction was less than
2 mgal at Browne corrections between 0 and 400 mgal. No adjustment
or modification of the S-9 horizontal accelerometers was made subsequent to the above study and consequently it is expected that Browne
correction errors in this study are approximately the same as those
previously reported. No corrections or adjustments of the gravity
measurements were made to compensate or correct errors due to
inaccurate horizontal acceleration determinations.
The vertical component of measured gravity (g) on a ship at
sea is actually a measurement of gravity plus vertical acceleration.
Vertical and horizontal accelerations observed on ships are 10,000
to more than 100, 000 times greater than the desired gravity meter
accuracy. These accelerations are of relatively short duration, however, and therefore easily removable by filtering. Filtering of the
vertical acceleration in the LaCoste and Romberg gravity meter is
accomplished electronically. Gravity meter and filter responses
must be extremely linear to avoid errors in gravity indications
(Harrison and LaCoste, 1968). The LaCoste and Romberg gravity
meter S-9 was designed to operate with vertical accelerations of
± 50 gals, on the assumption that vertical and horizontal accelerations
experienced at sea were approximately of the same size. It was later
determined that vertical accelerations very often exceeded the design
limit and consequently exceeded the assumptions of linear response.
40
The causes and results of nonlinear operation have been discussed
by LaCoste (1967). Factory linearity adjustments have been made to
the LaCoste and Romberg meter S-9; however, the linearity has been
noted to change with time, making it necessary to both retest the meter
periodically and to apply different correcting equations to the
measured gravity. The parameters, method and empirical equations
used to correct LaCoste and Romberg meter S-9 measurements were
described by Dehlinger etal. (1966a). The particular equations used
to correct the data used in this study are listed in Appendix 1.
The west to east component of ship velocity increases the
centrifugal acceleration experienced by the meter and thus decreases
the measured gravitational acceleration. This is termed the EtviSs
effect (described in Heiskanen and Vening Meinesz, 1958, p. 158)
and can be computed from the relation
= Zcosv mgal where
is the angular velocity of the earth (0.00007Z92/sec),
latitude of the point of observation and v
4
is the
is the ships west to east
velocity component. The correction so determined is the Etvs corr ection.
The estimated uncertainty associated with gravity meter
measurement, recording, reading, drift, and base station tie is ± 1
mgal. The estimated uncertainty in the vertical and Browne correction is ± 3 mgal when sea conditions produce Browne corrections
41
less than 400 mgal. The estimated uncertainty In Eötvös correction
is ± 1 n-igal (± 0.2 knot and ± 1.00 course heading). The total
estimated rms uncertainty, exclusive of position error, is
(32
+ 12 +
12)1/2
± 3.3 mgal.
Loran A and radar were the principal aids to navigation used
for the gravity surveys west of Washington and British Columbia. In
areas where radar ranging is most accurate, such as the fjords of
British Columbia, an estimated uncertainty in gravity measurement
due to position error, may be as low as ± 0.5 mgal (Dehlinger etal.,
1966a).
In the fringe areas of radar navigation where a combination
radar fixing and dead reckoning are used, gravity uncertainties due to
navigation errors may reach ± 3. 5 mgal. The rms uncertainty in
gravity measurement due to the positional accuracy attainable with
Loran A varies from ± 3. 0 mgal over the continental shelf to
6 mgal
over the Tufts and Alaskan abyssal plains.
Consequently, the total rms uncertainty in the sea gravity
measurements west of Washington and British Columbia ranges from
less than± 3.4mgaltonear± 7.0 mgal. For this reason, three
levels of uncertainty are assigned the contours of the free-air
anomaly maps shown in Figures 6 and 7.
Reduction of Gravity Data
The LaCoste and Romberg sea gravity meter produces a reading
___
134'
135°
51°
133°
r ____;
-:
42
132°
\
°
\
\
:
;/J) r
i[
125°
126°
_:- T -
.'o
L
\.o/
131°
7_((
130°
Figure 6. Free-air gravity anomaly map of the region off the west
coast of Washington and British Columbia from the
Columbia River to Cape Scott.
129°
'°
1/)
_
_
128°
127°
126°
:
I_
-t
125°
123°
124°
0
(/
132°
--- -
k
/
(/
..
127°
/\
'
/)
Ltiip4
-
28°
129°
7jrr
_
,1 )
130°
131°
'
I
124°
---
-
0l
_
8
23°
l2/
=-/,
-;a-----------
55.
39°
l38
i-
137°
36°
-:-
35°
- -
]c
'\
/_-
\
\
\J
--
,
\
,\
43
I3I
0
I3O
29°
-
J28
I27
126
55
)
\
t! I1
,)
\
132°
:
I
I
-
I33
134°
FREE-RGRAVITYANOMALYMAP
-I T1'
1
WEST COAST OF BRITISH COLUMBIA
!II
b
OCTOBER, 967
z
_ _ ____ _
I
/7
2
52
:J\
_)
H'
\
\\
/(
j \\J
\
NN
5
r\\\
50
39°
38°
37°
36°
35°
1'igure 7. Free-air gravity anomaly map of the region off the
west coast of British Columbia from Cape Cook to
Dixon Entrance.
I___________
34°
33°
132°
Th
131°
30°
29°
28°
27°
50
126°
44
which when corrected by the Etvs, Browne and vertical corrections
and multiplied by a meter (proportionality) constant is directly
related to the earth's gravitational acceleration at the point of
measurement. The LaCoste and Romberg meter measures difference
in gravity rather than absolute gravity; consequently, to determine
the earth's gravitational acceleration at a point the meter reading is
compared or tied to a base station. Gravity measurements made
aboard the USCGC YOCONA in 1964 and 1965 were tied to the
USC&GS station US 80 (Duerkson, 1949) in Astoria, Oregon and trans-
ferred to the ship using a Worden land gravity meter. Gravity
measurements made aboard the R/V YAQUINA in 1965, 1966 and
1967 were tied to the Oregon State University base station at the
Marine Science Center (Berg and Thiruvathukal, 1965) in Newport,
Oregon. These two base stations are tied to the U. S. Standard
Station in Washington, D. C.
The output of the LaCoste and Romberg gravity meter S-9 is
recorded by six analogue recorders. The recorders log meter reading, frequency and amplitude of vertical motion, displacement of the
meter from operating center, Browne correction and two orthogonal
components of horizontal acceleration.
The meter reading, vertical
correction, and Browne correction are determined from these data.
The ship's position, velocity, and heading, and an Etvs correction,
are determined from data provided by the ship's navigation log. The
45
meter constant, base station gravity value and base station meter
reading are logged constants determined prior to a survey.
Observed gravity, as a function of space and time is determined
from these data by using the relation:
where
g
= observed gravity,
V
k = meter constant,
m = on station meter reading,
meter reading,
rection, and
mbk + mk + E + V
g
c
mb
base
Ec = Etvbs cor-
= vertical correction. The data obtained aboard
the USCGC YOCONA and R/V YAQUINA were digitized by hand and
reduced using a computer program written in Fortran IV for the
Control Data Corporation (CDC) 3300 digital computer.
For mapping and analysis, observed gravity values were
reduced to free-air anomalies. Free-air anomalies are defined by
the relation:
F.A. = g
+
where F. A. is the free-air anomaly,
f
C
is the free-air correction and
g
t
is observed gravity,
g
is theoretical or normal
gravity. The free-air correction is a correction for elevation above
a sea level datum and is zero at sea level. Theoretical or normal
gravity was determined from the 1930 International Gravity Formula:
g = 978. 000 (1 + 0.0052884 sin
where
4
- 0. 0000059 sin
is the latitude of the station (e. g.
,
22) gals.
Heiskanen and Vening
Meinesz, 1958). The theoretical or normal gravity was programmed
46
for use on the CDC 3300 computer and incorporated in the general sea
gravity reduction computer program. The computer print out indicated
the latitude, longitude and free-air anomaly of each gravity station.
These data were then plotted and contoured by hand.
Distribution of Gravity Measurements
Figure 5 shows the location of the track lines along which surface- ship measurements of gravity were obtained. The total of the
track lengths is approximately 11,000 nautical miles (20, 000 km) and
the total number of surface-ship stations is approximately 15, 400.
These figures indicate a linear station density of one station approximately every 1.4 nautical miles (2. 5 km).
The addition of the pendulum stations reported by Harrison et al.
(1957) and Worzel (1965), the underwater gravimeter measurements
of the Dominion Observatory of Canada (Stacy, 1967), the surface-
ship measurements of the USC & GS and coastline land measure-
ments (Woollard and Rose, 1963; Sauers, 1966; Walcott, 1967; Stacy,
1967) increases the number of gravity stations which control the contours of Figures 6 and 7 to approximately 16, 100. The area mapped
is about 635, 000 km2.
The areal station density is then about one
station per 40 km2.
A station density of one per 40 km2 indicates a relatively thin
47
coverage when compared to a typical land gravity survey. These
figures represent an average but in the actual mapping, the track
lines and consequently the data points have an increased areal density
in the regions of steep gravity gradients. Further the high density of
stations along the track lines show the anomaly wavelength to be
expected in the region of measurement.
If a gravity anomaly can be assumed equidimensional (circular)
then the half wavelength of the anomaly observed along the track line
which crosses the anomaly approximates the diameter of the anomaly.
In this manner the number of gravity stations which describe an
anomaly can be estimated. Table 1 shows the average half wave-
length for gravity anomalies of different physiographic provinces.
Table 1. Information on the gravity anomaly size and gravity station
density in different physiographic provinces
Gravity anomaly Approximate Station
half wavelength anomaly area density per
Cascadia Abyssal Plain
Washington Continental
Shelf
Cobb Rise
Alaskan Abyssal Plain
Hecate Strait-Queen
Charlotte Sound
(km)
(km2)
anomaly
7
40
1.0
18
250
22
380
45
1600
6.3
9.5
40.0
12
115
2.8
The Cascadia Plain exhibits the shortest wavelength anomalies
of any province (and the smallest amplitudes) even though water depths
in this region are near 1500 fathoms (3000 meters). Anomaly wave-
lengths, although of comparative size, are shorter in the Dixon
Entrance-Hecate Strait-Queen Charlotte Sound area than on the con-
tinental shelf off Washington. The Alaskan Plain has the longest
anomaly wavelengths--twice as long as observed over the complex
topography of the Cobb Rise. Table 1 also indicates that the contours
of the anomalies of the Gas cadia Plain and the Dixon Entrance-Hecate
Strait-Queen Charlotte Sound areas are inadequately controlled by
available gravity measurements. However, because of the uneven
distribution of stations, the Cascadia Plain and the Dixon EntranceHecate Strait-Queen Charlotte Sound anomaly contours are better con-
trolled than indicated, while the anomaly contours of the Alaskan
Plain have considerably less control than indicated by Figure5
The
actual gravity control for each anomaly can be ascertained by cornparing Figures 6 and 7 with Figure 5.
The Free-Air Gravity Anomaly Maps
The geographic area covered by the free-air gravity anomaly
maps of Figures 6 and 7 extends from the Columbia River of Oregon
and Washington north to the Dixon Entrance and Alaska and from the
coastline of Washington and British Columbia west to a line
49
approximately 450 kilometers west of and parallel to the edge of the
continental shelf.
The free-air gravity anomalies are derived primarily from
surface ship gravity measurements obtained along the traverse lines
shown in Figure 5. The methods used in making the measurements,
the reduction of the measurements and methods of determining the
uncertainties in the processed data are outlined in the previous section and discussed in Appendices. 1, 2 and 3. Supplementary data and
methods of obtaining the final contours are discussed in Appendix 4.
The free-air gravity maps have three levels of contouring; solid
lines, long broken lines and short dashed lines indicating differences
in root mean square uncertainties in measurement. The solid lines
have an estimated rms uncertainty of ± 5 mgal. Solid contours occur
in areas in which weather conditions, navigation and meter operation
have resulted in better than average surface-ship sea gravity measure-
ments. Within the areas of solid contours are regions in which the
rms uncertainty in measurement is considerably better than ± S mgal.
For example, Dehlinger etal. (1966b) have estimated an rms uncertainty of ± Z mgal in measurements along the fjords of the Inside Passage of British Columbia and Alaska. Solid contours also imply that
the half wavelength of the anomaly described by the contours is equal
to or greater than the spacing of the available data. A possible
exception to this implication is the flat field of Cascadia Plain. This
50
region is discussed further under the subheading; Cascadia Abyssal
Plain.
The long broken lines indicate an estimated rms uncertainty of
± 7 mgal. The accuracy of the data in areas described by broken lines
has been limited by such factors as marginal meter operation, limited
navigation control, and adverse sea conditions. Variations of the
smoothed data along the traverse lines suggest that the half wave-
length of the observed anomalies is, on the average, at least of the
same order as the spacing of the available data, particularly in the
areas west of the continental slope.
The dashed lines are inferred and are based on isolated gravity
values, observed gravity gradients, or known bathymetric trends
and gradients. Dashed contours are, therefore, estimations of an
unknown gravity field.
Undoubtedly the gravity field of this region is more complex
than that indicated by the contours of Figures 6 and 7. The limitations
of surface ship gravity measurements and navigation facilities necessitates data smoothing which partially removes the high frequency
anomalies and limits the useful size of the survey grid. However,
the data are sufficiently accurate and numerous to delineate and per-
mit analyses of major structural features.
51
INTERPRETATION OF THE FREE-AIR GRAVITY
ANOMALY MAPS
The large density contrast between sea water and the underlying
sediments and rock and the varying proximity of the heavier material
cause the contours of the free-air anomalies to "follow" the bathy-
metric features. That is, the overlying water is essentially "trans
parentt' to gravity measurements. The density contrast between ocean
sediment and basement rock also leads to a partial transparency of the
sediment; however, the smaller density contrast between sediment
and basement rock necessitates larger volume and/or density changes
for a given gravity variation. Therefore, in areas in which the con-
tours of the free-air gravity anomalies do not follow the bathymetric
contours large variations in sediment thickness and/or severe
changes in the structure or density of the subsurface rock can be
expected.
Gravity Anomaly Trends
The free-air gravity anomaly contours in the area between the
Columbia River and Cape Cook trend approximately N 500 W, and
change in the area between Cape Cook and Dixon Entrance to approxi-
mately N 40° W. The trend of the anomalies parallels in general the
continental shelf, the coastline and the great structures of the
52
Western Cordillera. This trend is typified by the strong negative
anomalies of the Queen Charlotte Trough and Scott Islands fracture
zone and by the adjoined positive highs over Eickelberg Ridge.
In the physiographic provinces that McManus (1967b)termed the
Continental Rise and northern Cobb Rise, the anomaly contours are
extensive and irregular. However, in some instances they are somewhat symmetrical and strongly positive, outlining seamounts which
typically exhibit the shape of an elliptical cone with the long axis of
the basal ellipse along the regional trend.
The positive free-air anomaly over Bowie Seamount, the nega-
tive anomaly associated with the Explorer Trough, and a small positive anomaly northeast of Cobb Seamount appear as the only large
features with a dominant northeast-southwest trend.
Numerous anomalies cover Cascadia Plain. These positive and
negative anomalies are of small amplitude and area and show no
apparent dominant trend.
Isostatic Equilibrium in the Northeast Pacific Ocean
Isostatic equilibrium means that at some depth- -termed the
depth of compensation--the overlying crust and subcrustal materials
are in hydrostatic equilibrium. More generally the lighter rigid
crustal blocks of the earth are "floatingt on a more dense plastic
substratum. The observed topographic expressions of crustal blocks
53
such as mountains on a small scale or continents on a large scale are
compensated at depth by either lighter 'roots" or lower density
materials that extend beneath the Moho. An isostatic gravity anomaly,
based on measured gravity and assumptions of the general earth struc-
ture, indicates the state of compensation of the crustal elements.
Negative anomalies over oceanic regions indicate under compensation
and positive anomalies over compensation. Since the free-air reduc-
tion is essentially a type of isostatic reduction, corresponding to the
case when topographic masses are considered compensated at zero
depth (Heiskanen and Vening Meinesz 1958, p. 209) the average
regional free-air anomalies over the ocean directly reflect the isostatic status of the region, Therefore, oceanic areas having a negative free-air anomaly are considered to be under compensated or to
have a mass deficiency above the isopiestic level, Areas having a
positive anomaly have a mass excess and are considered over cornpensated.
The gravity anomalies of the abyssal plains, seamount province
and continental shelves, shown in Figures 6 and 7, undulate about the
zero level with amplitudes of approximately ± 20 rngal indicating that
regionally the area is approximately in isostatic equilibrium. Excep-
tions to this are relatively small areas which show large negative or
positive anomalies. Such areas are: the Queen Charlotte trough,
Scott Island fracture zone, numerous searnounts, the Explorer trough
54
and fjords along the coast of British Columbia. Anomalies associated
with these features do not necessarily imply deviations from isostatic
equilibrium but rather special mechanisms of anomaly generation or
isostatic compensation. The isostatic equilibrium of these special
cases will be discussed separately under the subheading pertaining to
each feature.
The region approximately enclosed by 50° and 510 N lat. and
133° and 137° W long, has an average free-air anomaly of about -20
mgal. Whether this negative region reflects a deviation from iso-
static equilibrium either as unusual crustal or subcrustal mass distributions, or insufficient or incorrect data is not known. This negative tendency is also discussed in association with a postulated fracture zone near 50. 50 N lat.
Fjords of the British Columbia Coast
Dehlinger etal, (1966b) described gravity measurements along
the fjords of the Inside Passage of Alaska and Canada. The measurernents indicated free-air gravity anomalies increasing negatively
toward the continent. The negative increase was tentatively attributed
to increasing elevations in terrain from the coastline toward the
cordillera. Additional measurements suggest that the free-air gravity
contours parallel the sides of the fjords, i.e. follow the topography
rather than cross them as suggested by the published map. This is
55
illustrated by the contours in Portland Inlet and Douglas Channel.
Although the free-air anomalies rapidly become more negative toward
the continent, isostatic reductions by Heiskanen (1938) and Banks
(1969) indicate that isostatic anomalies show only small negative
values along the fjords of British Columbia. This indicates that the
previously glacially loaded fjords--now drowned valleys--are now
near isostatic equilibrium and any residual isostatic anomaly is
probably local to the particular fjord. The small negative isostatic
anomalies near the heads of the fjords may be associated with the
very recent removal of the glacier ice near its most recent ice thickness es.
The trend of the negative free-air anomalies originating in
fjords can be traced across Hecate Strait and Queen Charlotte Sound
to the edge of the continental shelf and suggests either an extension
of the glacial channels or a structural continuity.
Gravity Anomaly Trends in Dixon Entrance, Hecate Strait
and Queen Charlotte Sound
Figure 7 shows a negative free-air gravity anomaly beginning in
Clarence Strait (550 N lat.
,
13].°50' W long.) which trends southward,
turning southeast off the northeast tip of Graham Island. From this
poinl it parallels the coast to the vicinity of 530 N lat.
130020
I
W long.,
where it turns again nearly straight south passing over the continental
56
shelf at 51° 30' N lat.
,
130° 30' W long. This feature is reflected in
the bathymetry of the area as a trough or bathymetric depression
between Clarence Strait and the northeast tip of Graham Island. Else-
where the limited bathymetry of the area suggests that it is sediment
filled with only small local depressions to hint at its location until it
nears the edge of the continental shelf.
This feature, originating in a fjord and passing diagonally across
the continental shelf, suggests that, rather than a synclinal depression as observed along the continental shelves west of Washington and
Oregon (described below) it reflects a glacial scour channel or glacial
valley now filled with sediment composed largely of glacial till. Den-
sities ranging from 1.5 to 2.5 gm/cm3 might be expected for
Pleistocene to Recent glacial till under the shallow water of the continental shelf. If a density of 2.0 gm/cm3 is assumed for this glacial
till and a density of 2. 7 gm/cm3 for the basement rock, the negative
gravity anomalies indicate a fill thickness ranging from 300 to 1000
meters. Higher densities would give a correspondingly thicker fill
of sediment and lower densities a thinner fill. Maximum depths of
fill might reach 3000 m.
A gravity low can also be observed beginning in the Portland
Inlet turning southeastward along the Grenville and Principle channels
and joining with the negative gravity anomaly of the Douglas Channel.
The Douglas Channel gravity low then passes south-southwestward
57
across Hecate Strait to the continental shelf terminating at the same
point as the Clarence Strait low. An additional negative gravity alignment can be noted beginning in the Fisher-Burke channels, crossing
Fitzhugh and Queen Charlotte sounds and joining the previously des-
cribed anomalies. The trend of these joined gravity anomalies
passes across the edge of the continental shelf at approximately
510301 N lat. and 130°30' W long. These channels are also partly
reflected in the bathymetry and apparently partially concealed with
sediment fill of similar thicknesses.
If these negative gravity lows are in fact caused by filled relic
glacial valleys, they indicate that valley glaciers in this region
passed southeast along Dixon Entrance and Hecate Strait and west-
ward across Queen Charlotte Sound, coalescing at a common point
on the continental shelf southeast of Moresby Island near the head
of the Moresby sea channel. A low ridge suggested by the positive
gravity anomaly between the northwest tip of Graham Island and Dali
Island, Alaska may have partially blocked access to the sea through
the Dixon Entrance during the glacial epoch.
These negative gravity anomalies and possibly filled glacial
valleys may also be partially structurally controlled. Tectonic structures and known faults of this region parallel these gravity lows over
a considerable portion of their length. Further, catalogued earth-
quake epicenters of this area locate along or near the axis of these
gravity lows.
The small magnitude of the gravity anomalies sug-
gests the earthquake activity is more likely to be due to continuing
tectonism rather than to isostatic adjustment.
The Continental Shelf Off Washington and Vancouver Island
The area of the continental shelf west of Vancouver Island
exhibits negative free-air anomalies of approximately -50 mgal
southwest of Nootka Island, nearly -70 mgal west of Cape Flattery,
and approximately -60 mgal west of Grays Harbor and Willapa Bay.
These anomalies are large, generally elongate, and trend parallel to
the shoreline. Smaller negative and positive anomalies occur between
the three larger anomalies. North of Cape Cook the anomalies of the
continental shelf are obscured by the extensive Scott Island fracture
zone.
South of the Columbia River the character of the gravity
anomalies on the shelf changes. A free-air gravity map, prepared
by Dehlinger etal. (1968) for the area west of Oregon shows a considerable number of small irregular positive and negative anomalies
along the continental shelf. The magnetic anomaly maps of the con-
tinental shelf west of Oregon also show many small 'irregular
anomalies with considerable magnetic relief (Ernilia, etal. , 1968a).
By comparison, the gravity and magnetic anomalies of the continental
shelf west of Washington and Vancouver Island are of strikingly
different character. They show fewer, more regular anomalies of
greater areal extent but of comparable magnitude.
The negative gravity anomalies on the shelf off Washington trend
toward the Olympic Mountain negative Bouguer anomaly and may be
genetically related. The Olympic Mountain anomaly, however, trends
northwest-southeast while the associated shelf anomalies turn, near
the coastline, to a north-south alignment. The negative anomaly west
of Cape Flattery is at least partly due to the bathymetric relief of
Astoria Canyon. The negative gravity anomaly along the continental
shelf off Vancouver Island appears confined to the shelf region and
may represent a sediment-filled synclinal depression.
A small gravity high superposed on the regional low west of the
Olympic peninsula extends from the- coast of Washington near 43.5°
N latitude west then northwestward to the closed gravity high off
Vancouver Island near Barkley Sound. Figure 4 indicates a positive
magnetic anomaly of similar areal extent which occurs in the same
location. The small gravity high joins a similar gravity high on land
which appears coincidental with the outcropping of Eocene volcanics
about the perimeter of the Olympic Mountains. The gravity and magnetic anomalies suggest that the Eocene volcanics observable on the
Olympic peninsula may extend westward beneath the continental shelf
as far as the continental slope off northern Washington and southern
Vancouver Island.
The most striking gravity anomaly along the Washington con-
tinental shelf is the - 60 mgal linear minimum west of Grays Harbor
and Willapa Bay. This anomaly appears to terminate at Astoria
Canyon or suffer a right lateral offset of about 15 km across the can-
yon. Two possible explanations of this are: (1) the sediment-filled
synclinal depression represented by the free-air gravity anomaly is
actually offset by a transcurrent fault along the Astoria CanyonColumbia River lineation, or (2) the gravity low south of Astoria Canyon has had an apparent westward displacement due to the upwarping
of Tertiary volcanics which, according to Snavely and Wagner (1964),
form a northward plunging anticlinorium in northwest Oregon. In
this latter explanation the seaward displacement of the anomaly is due
to tilting of the continental shelf and the residual sharp east-west
anomaly structure over Astoria Canyon is due primarily to the bathymetry of the canyon.
Carison (1968) suggested that step-like offsets of the Astoria
Canyon indicate that it may have been cut along a zone of structural
weakness. Minor faults and anticlines having strikes parallel with
the lower Columbia River are observed in northwestern Oregon.
The Randsenken Effect Along the Continental Margin of
Washington and British Columbia
Isostasy implies that thick blocks of continental crust sink
deeper into the more dense subcrustal material than do thinner blocks
61
of oceanic crust. Therefore, when continental and oceanic crustal
blocks are juxtaposed, the edge of the thick continental block pro-
jecting deeper into the substratum than the oceanic block produces a
step in the crust-subcrust boundary. This Randsenken (sunken-edge)
of the continent produces a negative free-air anomaly parallel to and
near the base of the continental slope and a positive anomaly parallel
to and generally over the landward side of the edge of the continental
shelf. The amplitude of the positive anomaly is somewhat reduced by
the reverse Randsenken effect of the deep water layer oceanward of
the continental slope. The Randsenken along a coastline such as the
Pacific coast of North America generally produces a negative free-
air gravity anomaly of -60 to -100 mgal. The larger negative
anomalies lie along transition zones having the steeper continental
slopes. Typical negative free-air gravity anomalies along the west
coast of North America due to the Randsenken effect are: Southeast
Alaska -80 mgal, Queen Charlotte Island -80 to -100 mgal,
Washington -20 mgal, Oregon -60 to -100 mgal, northern California
-60 mgal to -70 mgal and Baja California -85 mgal. These data are
taken from the free-air anomalies outlined in this thesis and values
near the base of the continental slope in the northeast Pacific Ocean
given by Worzel (1965), Vening Meinesz (Heiskanen, 1938),
Dehlinger et al. (1 967a), and Dehlinger et al. (1 967b).
Figure 7 shows the negative free-air anomaly due to the
62
Randsenken of the continent from the Dixon Entrance to Cape Cook.
The continental slope in this region exhibits a very steep gradient
and consequently, generates a large Randsenken effect gravity
anomaly. The negative free-air anomaly is exaggerated west of the
Scott Is lands due to the superposition of a fracture zone mass deficit
on top of the expected Randsenken effect.
Strikingly, the negative gravity anomaly of the continental
Randsenken is exceptionally small west of southern Vancouver Island
and Washington. Bathymetric maps of this region, prepared by
McManus (1964), show a continental slope having a shallow two step
gradient; however, the bathymetric relief is insufficient to account for
the near absence of a negative free-air anomaly. The gravity anomaly
along the continental slope of Washington and southern Vancouver
Island suggests that either the transition from oceanic to continental
crust is more gradual than expected, or that the crust of the continental block is thinner than expected.
The Randsenken effect or the gravity anomaly observed by
Worzel (1965) along the east coast of North America is approximately
half as large as observed along the west coast of North America.
This difference implies that either the transition from continental to
oceanic crust occurs more abruptly along the west than the east coast
or that the density contrast between continental and oceanic rocks,
including rocks of the upper mantle, is greater along the west coast
than the east coast.
63
The Queen Charlotte Trough
Menard and Dietz (1951) outlined, in their early study of the
submarine geology of the Gulf of Alaska, an irregular elongated
trough along the west side of the Queen Charlotte Islands. St. Amand
(1957) states that the continental shelf is narrow in this region, with
the 100 fathom contour only about 1 mile offshore.
He also postulates
on the basis of the narrow shelf and straightness of the western shore
that a fault defines the oceanward shoreline of the Queen Charlotte
Islands. Hurley (1960), in his study of the geomorphology of the
northeast Pacific, includes fathograms of the continental shelf and
slope and the abyssal plain west southwest from Dixon Entrance and
vest-northwest of Graham Island. These fathograms indicate a steep
continental slope terminating in a sharp scarp at the abyssal plain.
Hurley does not show a trough west of the Queen Charlotte Islands on
his physiographic map. Gibson (1960) presented a bathymetric
traverse west of Moresby Island which shows a complex bench on the
continental slope and a trough or depression at the base of the slope.
Gibson's map "The Submarine Topography in the Gulf of Alaska" out-
lines a "Queen Charlotte Trough" that is approximately 300 km long,
with an axis approximately 55 km west of the shoreline of the Queen
Charlotte Islands, The U. S. Hydrographic Office map (1952) of this
area shows only a small depression west of Moresby Island, The
64
configuration or existence of 'The Queen Charlotte Trought still
remains somewhat questionable at this writing. For this reason it
is difficult to compare the bathymetry and free-air gravity anomalies
in this region.
The gravity map indicates a gravity low approximately 150 km
in length west and northwest of Graham Island, a more complex
gravity low west of Moresby Island, and a small basin shaped low
southwest of the south end of the Queen Charlotte Islands. These
last two lows are separated by a gravity high which suggests two seamount structures connected by a ridge. Gravity gradients along the
steep continental slope west of Moresby Island reach 20 mgal/km.
A large part of the negative gravity anomalies west of the extremely
narrow continental shelf, in the area about the Queen Charlotte
Islands, can be attributed to the Randenken of the continent,
indicating that this area is essentially in isostatic equilibrium and
suggesting that a very deep sediment filled trench is unlikely. How-
ever, the basinal structure suggested by the gravity contours southwest of the south end of the Queen Charlotte Islands has a shallower
depth and a less negative gravity anomaly than the gravity minimum
to the north suggesting a possible ponding of 1 to 3 kilometers of
glacially derived sediment in this basin. The glacial silt streaming
down the continental slope near 510301 N lat. and l30°30'W long, was
apparently blocked in its northerly flow by the hills and ridge outlined
65
by the gravity high separating the two gravity minima.
A Postulated East-West Fracture Zone Near 510 N Lat.
Menard (1955b) postulated the existence of an eastwest fracture
zone about 600 miles north of the Mendocino fracture zone. This was
based on the observed regularity in spacing of the then known E-W
trending fracture zones which segment the eastern Pacific Ocean.
Hurley (1960) discussed this fracture zone further in his study of the
geomorphology of the northeast Pacific Ocean, suggesting that Peters
Ridge is near the expected position and has the proper trend. He also
suggested this trend disappears beneath the Queen Charlotte fan to the
east. This east-west lineation is in alignment with the westward set
of the north tip of Vancouver Island. This set is visible also in the
continental shelf west of the Scott Islands and is emphasized by the
gravity anomaly contours of the same area. The gravity anomaly
contours between 50° and 51° N lat. , however, do not show a pattern
which would be expected over a sediment-buried ridge, escarpment
or trough, such as those associated with known fracture zones of the
eastern Pacific Ocean.
The gravity data between 50° N and 51° N lat. and west of 132°
W long, have greater uncertainty than the data between 132° W long.
and the coast of Vancouver Island and may be of insufficient accuracy
to discern a small fracture.
Dehlinger etal. (1967a), in a study of the Mendocino Escarpment, have shown that in addition to the variation of gravity due to the
bathymetric relief, there is also a part of the anomaly over the
fracture zone which is due to density-mass differences in the upper
mantle. These differences are reflected in the free-air anomaly map
as a predominantly negative region south of the escarpment and zero
or slightly positive anomalies north of the escarpment. In the
vicinity of the postulated fracture zone between 500 N and 51° N lat.
and 1320 to 137° W long, the anomalies are predominantly negative
while to the north and west of this region the anomalies are near zero
or positive. While the data are inconclusive they do suggest the pos-
sibility of variations in distribution of mantle material similar to that
hypothesized beneath the Mendocino fracture zone,
If the Scott Island fracture zone as outlined by the 100 mgal
contour is interpreted as at least partly due to tension, then the
expected motion along an east-west fault located just south of 51° N
lat. would be right lateral contrary to that which would be expected
from the observed hboffsetlt of the continental shelf west of the Scott
Islands.
The gravity data do not generally substantiate the existence of
the postulated fracture but are inconclusive particularly if the fracture zone has its eastern terminus west of 132° W longitude.
67
The Scott Islands Fracture Zone
Benioff (1962) postulated a dextral transcurrent fault westsouthwest of the coasts of Alaska and British Columbia, extending
from approximately 48° N lat., 127° 30' W long, along a great circle
to 60° N lat., 140° 30' W long. Wilson termed this the Queen
Charlotte Islands Fault, suggesting that t is a transform fault rather
than a transcurrent fault.
fault near 48° N lat.
,
He also relocated the southern end of this
128°20' W long. McManus (l967b), from a
study of the bathymetry, described the area between Cobb Rise and
the northern end of Vancouver Island as consisting of fracture zone
topography. The bathymetric map of the region (Figure 8) shows
long narrow ridges (50°lO' N lat.
,
129°10' W long.), trending north-
west, with a relatively smooth floor between- -suggestive of sediment-filled troughs.
Along a coast with a relatively narrow continental shelf and
steep continental slope approximately 80 to 100 mgal of the observed
negative free-air gravity anomaly over the ocean may be attributed
to the Randsenken of the continent. For this reason the -100 mgal
closed contour will be used to define the area of the Scott Islands
fracture zone, This fracture zone actually lies partly on the continental shelf and slope between the Scott Islands and Quatsino Sound;
and, therefore, the -100 mgal contour line is probably a conservative
estimate of the actual areal size of thj.s tectonic feature.
This structural feature of the earth's crust defined by a large
negative free-air gravity anomaly was first noted on a traverse by
the R/V YAQUINA in August, 1965. The traverse passed in a north-
easterly direction between the Scott Islands and the northern tip of
Vancouver Island nearly over the region of the greatest negative
anomaly. Because this feature was first noted off the Scott Islands
and is nearly centrally located with respect to the islands, the Scott
Islands fracture zone was adopted as a best descriptor.
The Queen Charlotte fault as described by Benioff (1962) and
Wilson (1965b) actually lies along the extreme southwest edge of this
fracture zone. Further, the areal gravity outline is not what would
generally be expected for a trans current fault, suggesting that,
although this fracture zone is probably associated with the Queen
Charlotte fault, it is also an entity in itself.
This feature is approximately 35 km in width, at its widest
point, and 175 km long. It has the general appearance of a complex
graben or a group of down-faulted blocks which, after normal faulting,
have been subjected to right lateral shear. If in fact the maximum
gravity lows, indicated by small closed hatchured contours, can be
assumed to have been aligned originally along a northeast trend, a
30 km right lateral offset can be postulated between the gravity low
associated with the north end of the Explorer trough and the central
anomalies of the Scott Islands fracture zone.
If this anomaly is interpreted as a graben, it is most likely
filled with glacial till ubiquitous to this region. Assuming a density
contrast of 0. 3 gms/cm3 between upper crust and loose till (2.5-2.2)
and between lower crust and more consolidated till (2. 7-2. 4) maxi-
mum fill depths of 4 to 6 km might be expected.
Seamounts
Cobb Rise and the adjacent area toward the northwest contain
numerous seamounts (Figure 1). These seamounts seldom occur as
solitary prominences but rather as contiguous groups with a general
northwest trend. The method of determining the free-air gravity
anomaly value above the peaks of the seamounts and the average sea-
mount densities is described in Appendix 4. The location and
dimensions of 44 seamounts occurring in the area of the free-air
gravity anomalies of Figures 6 and 7 are given in Table 6. Figures
6 and 7 show that a negative free-air gravity anomaly occurs near
the base around solitary seamounts and along adjoined seamounts.
The larger negative free-air anomalies are most prominent along
the southwest and northeast sides of seamounts even near solitary,
symmetrical seamounts such as Union searnount (49°32. 5' N lat.,
132 041. 6' W long.).
This indicates that the substructural trend is
also predominantly northwest-southeast.
70
The negative anomaly surrounding the central positive peak
anomaly suggests that the seamounts in this region are in or nearly
in isostatic equilibrium. The average ratio of the area of compensation, defined by a circle about the seamount axis approximately
enclosing the negative ttmoat" anomaly, and the area of the load,
defined by a circle approximately enclosing the positive peak anomaly,
is 7. That is the weight of the conical seamount, sustained by the
crust, is distributed over an area approximately 7 times the area of
the base of the seamount. Load and compensation areas for 10 sea-
mounts are given in Table 2.
A comparison of the area of load, defined by the positive freeair gravity anomaly and the area of load defined by the physical base
of the seamount shown on bathymetric charts, shows they are, within
the accuracy of measurement, very nearly identical in the region
south of 50° N lat. However, north of 50° N
late
the load area
defined by gravity is approximately 5 times the load area defined by
bathymetry. Since the structure of seamounts above the ocean floor
is probably similar, this difference in load areas suggests that dif-
ferences in structure or constituents of the crust or upper mantle
occur between these two areas. In this respect McManus (1965)
noted that the seamounts of the Alaskan Abyssal Plain appear to occur
on rises.
If it is assumed that the source of magma to form a seamount is
Table 2. Seamount load and compensation parameters
Physical Dimensions
No.
ra
Table 3
(km)
3
9.9
18.4
18.4
8.3
18.4
16.2
16.5
11.0
17.5
7.4
5
6
14
15
25
28
37
38
40
h
(km)
2.02
2.73
3.06
1.67
2.76
2.55
2.85
1.83
2.95
1.91
Gravity Dimensions
r
v
rp
(km3)
(km)
(km)
(r/r)
216
966
8.6
34.2
16.0
43
18.5
4.1
47.8
38.5
21.4
6.6
98
4.0
169
60
9.3
15.9
5.8
240
109
160
24
10.4
15.3
2.3
1082
119
977
700
815
233
937
1080
19.2
7.8
14.3
13.5
12.8
26.4
35.6
19.3
30.6
38.5
34.2
47.7
100.5
40.7
Average
ra
=
h
=
v
=
r
=
p
rn
=
=
E.
g
=
7.4
4.6
8.1
7.1
3.3
8.0
4.5
7
2
h
(m)
(mgal)
22.8
21
2.1
151
19.4
108
10
radius of seamount base (km)
height of seamount (km)
volume of seamount (km3)
radius of positive F. A. anomaly (km)
radius of negative F. A. anomaly (km)
(V/p r2) (2. 4/3. 3) x
(m)
yL p t h = . 095
h (mgal)
I-
72
a subcrustal layer beneath the area of compensation, the thickness of
this layer may be determined from the volume of the seamount
obtained from bathymetry and the area of compensation. Since the
source material of the effusive has a density near 3. 3 gm/cm3 and
the resulting seamount a density near 2. 4 gms/cm a proportional
adjustment of the source layer thickness is made. The range of
thickness of such a layer is 20 to 170 meters with an average of 108
meters. In transferring the material of such a layer to the surface
as a seamount, a negative free-air anomaly would be formed in the
area of compensation. This anomaly would range from 2 to 23 mgal
with an average of 10 mgal. This figure is in good agreement with
the average anomaly observed about seamounts. It is not possible
by this means to say whether the material in the layer beneath the
area of compensation has actually erupted on the surface or whether
it has been displaced horizontally out of the area of compensation by
a surface load derived from a deeper source. It does, however,
indicate that isostatic equilibrium is achieved and that the weight of
the searnount is not likely sustained by the elastic strength of the
crust.
The Explorer Trough
Gibson (1960) depicts the Explorer trough on his map of the
submarine topography in the Gulf of Alaska as a 130-mile long, 5-10-
73
mile wide depression extending from approximately 48°20' N lat.,
131° W long., to 50°20' N lat., 129°40' W long. This topographic
feature is shown as a nearly linear feature trending N 15° E on the
longer southern and central portion and bending to N 500 E at the
northern end. Gibson presents cross-sections of the Explorer trough
suggesting they have the form of a thrust fault with a raised escarpment on the eastern edge and favors a tectonic origin for this
topographic feature.
Figure 8 shows a portion of a bathymetric map by McManus
(1964) which portrays the topography about the Explorer trough. This
map shows the location of the Explorer seamount (49°05' N lat.
130°45' W long..) andtheSethinole seamount (49°45' N lat. , 129°50' W
long.) with respect to the Explorer trough and indicates that the
northern end of the trough (50°05' N lat. , 129° 45' W long. ) terminates
abruptly against a long narrow northwest-southeast trending ridge.
The free-air gravity contours outlining the Explorer trough
show a pattern similar to the bathymetric maps of Gibson (1960) and
McManus (1964); however, the anomaly contours appear more sinuous
and irregular. The southern end of the trench encircles the
Explorer seamount anomaly in the manner of a moat typical of large
seamounts. The central portion of the negative free-air gravity
anomaly is elongate and narrow with the axis of the -50 mgal anomaly
displaced to the northwest of the topographic axis. The magnitude
74
51°
132°
1310
1300
129°
I28° z:,i
500
500
49°
490
4R°
480
132°
1310
130°
129°
Figure 8. Bathymetric chart of the Explorer trough and Scott
Islands fracture zone (after McManus, 1964) showing
location of profiles S-S', X-XT, Y-Y', and Z-Z'.
128°
75
and shape of the gravity anomaly suggest that the central portion of
the trough is partly filled with sediment. The free-air gravity
anomalies over the northeast end of the Explorer trough show two
-50 mgal depressions. The northeast gravity anomaly low is
directly correlative with the bathymetry; however, the gravity
anomaly approximately 20 km to the south has no apparent topographic
expression implying a total filling by approximately 1.5 km of sedi-
ment of a pre-existing basinal or graben structure.
Two possible explanations of the plan view character of the
Explorer trough reflected by the sinuous free-air gravity anomalies
are: 1) the trough is a series of circumstantially interconnected
moats about the Explorer seamount, Seminole seamount and the end
of the Dellwood seamount range; or 2) the trough originally was a
more linear feature, formed by tectonic activity, which has been
reshaped in the south and central areas by volcanic flows from
adjacent seamounts and in the north end by transcurrent faulting
normal to the trough axis.
In the first case the trough postdates the
seamounts and in the second case the trough pre-dates the seamounts.
Figures 9, 10 and 11 show three selected gravity, magnetic and
bathymetric profiles across and approximately orthogonal to the
Explorer trough. Figure 8 shows the location of the three profiles:
Figure 9 along line X-X, Figure 10 along line Y-Y' and Figure 11
along Z-Z'.
7£)
ALONG LINE XX'
568
a
60
566
50
40
30
20
564
562
10
>
Ii-
0
-10
560
-20
-30
-40
-50
-60
558
556
-J
554
552
GRAVITY
1.5
2.0
2.5
I-
3.0
3.5
0
5
10
15
20
25
DISTANCE
30
35
40
45
50
(km)
Figure 9. Bathymetric, gravity and magnetic profile across
Explorer trough along X-X'.
77
ALONG LINE V-V
560
20
559
I0
558
557
-
556
-20
555
554
553
-50
552
GRAVITY
2.0
2.5
0
5
10
20
15
0/STANCE
25
30
(km)
Figure 10. Bathymetric, gravity and magnetic profile across
Explorer trough along Y-Y'.
;
78
ALONG LINE Z - Z'
560
\
/
20
558
E
0
1
556
/
/'Y
I0
//
0
1
1'N/-'\
\\j (
552
()
::
k
550
548
MAGNETICS
-30
OBSERVED GRAVITY
546
COMPUTED GRAVITY
-40
2.5
p 03
S
-
I
\
3.5
p2.I
/\
E
\/
/
\
I.
/
/
4.0
\/\\
S
-.I
p2.9
4.5
p2.9
S
/
50
I
0
0
20
30
I
40
I
50
I
60
DISTANCE
I
70
I
80
I
I
90
100
110
120
(k,n)
Figure 11. Bathymetric, gravity and magnetic profile across
Explorer trough along Z-Z'.
2
79
The bathymetric profile over the northeast end of the Explorer
trough along line X-X' indicates a 3. 5 km deep trough or
trench with steep scarp-like walls. Benches occur on the northwest
wall. Scarps of the northwest wall appear to have slopes of 600 or
greater while the southeast wall slope is near 30°. The angle of the
walls or scarps suggests gravity faulting along the northwest wall
and low angle normal faulting along the southeast wall possibly indica-
tive of tensional forces normal to the trough axis. Several benches
occur on the northwest wall which exhibit little or no sediment cover.
Projection of the wall slopes downward to intersection suggests that
sediment fill in the trough axis is 0. 1 km or less. No significant
amount of sediment cover or fill is detectable on the ridge structures
which flank the northern end of the Explorer trough. PDR records
from approximately 40 to 50 km southeast of the trench axis show
ponded sediment which is severely tilted.
The bathymetry along profile Y-Y' near the mid-length of the
Explorer trough shows a relatively small trough approximately 7 km
wide and 300 m deep. The sediment fill in this portion of the
Explorer trough, estimated from gravity measurements, is 0. 5 to
1 km.
The topography along line Z-Z', indicated on precision depth
records, shows a basinal feature approximately 60 km wide and 300
to 400 m deep in the center. Free-air gravity anomalies show a
sharp 20 mgal negative depression over the axis of the topographic
depression suggestive of a sediment-filled trough. To estimate the
thickness of the sediment fill (or the actual trough depth) two sub-
surface 2 dimensional structures were assumed and the gravity
anomaly calculated for each. A density of 2. 1 gm/cm3 was used for
the sediment and a density of 2. 9 gm/cm3 for the underlying base-
ment rock.
The structure, indicated by the open triangles connected by
dashed lines, produces the gravity anomaly shown by the open
triangles. Although the calculated gravity anomaly, shown by the
open traingles, does not coincide with the details of the observed
gravity anomaly (shown by the solid 1ine)it does exhibit the same
amplitude. The model structure shows an estimated minimum
anticipated sediment thickness of 0. 6 km in the filled trough.
The structure indicated by the solid circles connected by a
solid line produces the gravity anomaly shown by the solid circles.
The gravity anomaly produced by this structure coincides well with
the observed gravity anomaly and the model structure shows an
estimated maximum sediment thickness or trough depth of 1.7 km.
The effect of two-dimensional models and the high density
contrast (2.9 gm/cm3 - 2. 1 gm/cm3 = 0.8 gm/cm3) cause these
calculations to produce conservative estimates of sediment fill
thicknesses. The sediment thickness and consequently the actual
EiJ
relief of the Explorer trough south of the Explorer seamount is
therefore probably near 1.5 km.
Magnetic measurements were made concurrently with the
gravity measurements along profiles X-X', Y-Y' and Z-Z'. The
total magnetic intensity exhibits a relief of 1 600 gammas over the
northeastern end of the trough, 800 gammas in the central section
and a relief of approximately 500 gammas over the southwestern end
of the trough near the base of the Explorer seamount. The gradient
of the regional magnetic field is approximately perpendicular to the
magnetic profiles. Therefore, removal of the regional field to pro-
duce a magnetic anomaly causes little change in the magnetic profile
except to change the reference level. Figures 9, 10 and 11 show
the magnetic anomalies are strongly positive over the troughand the
axis of the magnetic anomaly is displaced towards the southeast with
respect to the trough axis. The magnetic anomalies are very nearly
centered on the southeast scarp of the trough. Each magnetic profile shows a dip or decrease of 100 to 200 gammas near the magnetic
axis and also a greater amplitude of the southeast edge of the positive
anomaly than the northwest edge.
The amplitudes of the gravity and magnetic anomalies increase
along the strike of the Explorer trough from southwest to northeast
whereas the sediment thickness in the trough decreases from
approximately 1.5 kilometers to 0.1 kilometers. Maps of the free-air
82
gravity anomalies (Figure 6), magnetic anomalies (Figure 4) and the
bathymetry (Figure 8) suggest that the changes in anomaly values and
sediment thickness may not be gradual but rather that the Explorer
trough may be segmented. The absence of significant amounts of
sediment in the northeast end of the trough suggests recent deforma-
tionand earthquake epicenters about the immediate trough area
indicate continuing tectonism whereas the trough south and east of
the Explorer searnount appears vestigial.
CascadiaAbyssal Plain
The contours of the free-air gravity anomalies in the area
designated by Hurley (1960) as the Gas cadia Abyssal Plain are small
in amplitude and areal size and exhibit relatively smooth closed outlines. The half wave length of the free-air gravity anomalies along
the survey ships track line indicate that a number of these anomalies
have smaller dimensions than the survey grid. For this reason
additional small anomalies can be anticipated in the regions about
48° N lat.
,
128° W long, and 46° N
120° W long.
A1so it is
quite possible that with additional measurements the anomalies
indicated in this region may subdivide into many smaller ones.
These small anomalies suggest the presence of sediment-covered
hills of low relief. The magnitude of the gravity anomalies, as suming a density contrast between sediments and abyssal hills of 0. 6 to
0.7 gms/cm3, indicates that these abyssal hills may exhibit an undulat-
ing relief of 300 to 400 meters which is completely covered by sediment. Examination of PDR records obtained across the Cascadia
Abyssal Plain shows the tops of a number of hills protruding through
the sediment (Kulm, 1967), particularly near the western flank of the
Gas cadia Abyssal Plain where the sediment thins.
A seismic reflection profile across the north end of the Gas cadia
Plain (Ewing etal.
,
1968) indicates that the negative free-air
anomaly near 48.8° N lat.
,
128.8° W long, is due to a graben.
The
graben trends about N 30° E, generally parallel to Explorer ridge
and trough, and is partially filled with sediment 0.5 to 1. 2 km thick.
Earthquake epicenters (Tobin and Sykes, 1965) in the vicinity of the
grab en indicate continuing tectonism.
The thickness of sediment reported (Shor etal. , 1968;
Hamilton, 1967; Duncan, 1968) in the Cascadia Abyssal Plain is sufficient to produce a free-air anomaly of 10
to
60 mgal when super-
imposed on an isostatically-balanced crust. An observed zero average
anomaly over the Cascadia Abyssal Plain strongly suggests the region
has subsided to accommodate the additional load.
This subsidence
would be proportional to the sediment thickness and of the order of
380 meters near the center of the Gas cadia Abyssal Plain.
CRUSTAL AND SUBCRUSTAL CROSS SECTIONS ACROSS THE
CONTINENTAL MARGINS OF WASHINGTON
AND BRITISH COLUMBIA
Construction of the Crustal and Subcrustal Cross Sections
Figure 12 shows the location of five 2-dimensional crustal and
subcrustal sections across the continental margins of Washington and
British Columbia which were constructed to study the structural
transition from ocean to continent. Each of the sections is approximately 800 to 1000 km in length. They begin in the Tufts and Alaskan
abyssal plains and extend eastward to the continental interior east of
the coast ranges. A 50 km vertical dimension was arbitrarily
selected for the cross sections. On the basis of seismic refraction
investigations (e.g. James and Steinhart, 1966), isostasy (e.g.
Heiskanen and Vening Meinesz, 1958), previous studies (e.g. Worzel,
1965) of continental margins, and the wave lengths of the observed
gravity anomalies, major changes in structure in the zone of transition were expected to occur within 50 km of the earth's surface.
Land elevations on the continents were obtained from topographic
maps of the US Geological Survey and the Dominion Observatory of
Canada. Free-air gravity anomalies at sea are based
on measure-
ments made with gravity meter S-9 along track lines which very
closely approximate the surface trace of the cross sections. Water
140
138
136
rfl
,..
,,.
,',n
I1f
54,
52'
56
¶I
__
_____
54
__ __ -
52
co
Ui
depths were determined from PDR measurements obtained
simultaneously with the gravity measurements. Depths to layers in
the cross sections were obtained from seismic refraction measurements.
The location of the seismic surveys are shown in Figure 2.
Seismic refraction results were extrapolated laterally along isobaths
when they did not coincide exactly with the section profiles. The
topographic, bathymetric and seismic refraction results were scaled
onto working drawings and then connected by straight lines so that the
scaled cross section consisted of joined polygons. The cross sections
were effectively extended 5000 km in each direction to prevent end
effects in the final computation.
The observed seismic velocities were converted to densities
using the relation shown in Figure 13. This empirical relation,
established by Nafe and Drake (1961) is based on: 1) measurements
on sediments and sedimentary rocks (V< 6 km/sec), 2) Birch's
measurements at 10 Kbar for various igneous rocks, and 3) the
Bullen A model of the mantle (Nafe, 1967), The Nafe-Drake relation
between compressional wave velocity and rock density is generally
considered more reliable in sedimentary and upper crustal rocks than
in materials of the mantle. The densities obtained from the seismic
wave velocities using the Nafe-Drake curves were assigned to the
appropriate polygons of cross section.
87
_ff4.J
rai
LSI
-II.i±IIIi
7
-- -H-
i;=
5
---H-i-----
3
2
EEEEEEEEEEEE
:1.]
2
3
Figure 13. Graph of the Nafe-Drake relation between cornpressional wave velocity and density (Nafe and Drake,
1961).
4
The vertical component of gravitational acceleration at points
along the earth's surface due to the sw-n of the polygons of a cross
section was computed using the line integral method as applied by
Talwani et al. (1959b) (see Appendix 3). The computed gravity was
compared with observed gravity, and iterative adjustments of the polygons were made until observed and computed gravity values coincided.
A common crustal and subcrustal section 50 km in depth, was
adopted for the ocean end of the five cross sections. The common
section is a computational reference which permits the five cross
sections to be compared in three dimensions. The section, which is
not shown on the cross sections, consists of five layers: 1) a 3. 7 km
water layer of density 1.03 gm/cm3, 2) a 0.5 km sedimentary layer
of density 2.15 gm/cm3, 3) aO.8 km upper crustal layer of density
2.74 gm/cm3, 4) a 5.0 km lower crustal layer of density 3.00
gm/cm3, and 5) a 40.0 km mantle layer of density 3.30 gm/cm3
A free-air gravity anomaly of zero is assumed over the section.
Sources of Error in the Cross Sections
Bouguer gravity anomalies were used for structural control
over land.
The Bouguer correction assumes a density of 2. 67
gm/cm3 which may differ considerably from the true density in some
places. Walcott (1967), in constructing a Bouguer anomaly map of
Vancouver Island, estimates the anomaly error may be as large as
6 mgal. Sea gravity measurements have an rms uncertainty of
approximately SmgaJ.. Changes in gravity of approximately S nigal in
the Bouguer and free-air anomalies would cause less than 0. 5 km
vertical change in a Moho depth of the cross section,
Shor (1962) indicated the accuracy of seismic refraction
velocities is difficult to assess and attributes an "error of determination' of 0. 05 km/sec to good stations. An error of 0. 05 km/sec
in the velocity determination will result in an error of 0. 2 km in
layer depth of a deep-sea profile and about 0. 5 km in a shelf profile
(Shor, 1962). Errors in refraction studies may also occur due to
uncertainties in ship position particularly in areas of large structural
gradients. Shor etal, (1968) estimates an uncertainty of ± 5 km in
position in refraction studies off Oregon.
The velocity-density relation of Nafe and Drake (1961) is an
average fit to many rock types. The composition of mantle rocks is
unknown; and rock types postulated for the mantle,which have been
measured in the laboratory under conditions believed to exist in the
upper mantle, indicate that for a velocity of 8 km/sec densities may
range from 3. 0 gm/cm3 to 3. 7 gm/cm3 (e. g. W o olla r d,
1 9 6 2).
The empirical relation between crustal thickness
indicated by seismic measurements and changes in surface elevation
suggest mantle densities are between approximately 3. 28 gm/cm3
and 3.46 gm/cm3 (Woollard, 1962).
Clearly the thicknesses
and densities of upper mantle materials can be varied to conform
with the observed gravity anomalies. However, seismic refraction
control and a small range of reasonable densities in the crust restrict
acceptable layering for the models.
A uniform mantle is assumed below 50 km. If, however, dif-
ferences in the upper mantle beneath oceans and continents extend
greater than 50 km as has been suggested as possible by MacDonald
(1963), lateral density contrasts in the upper mantle would be
reduced.
The Central Washington Cross Section
The central Washington crustal and subcrustal section BB! is
shown in Figure 14. The western end of the central Washington sec-
tion is located in Tufts Abyssal Plain, near 45. 3° N lat.,
134.2° W long. The cross section strikes N 74° E cutting
obliquely across Juan de Fuca ridge and the Cascadia Abyssal Plain.
The cross section is normal to the trend of the continental slope and
shelf and crosses the coast north of Gray's Harbor, Washington.
It traverses Puget Sound near Seattle and has its eastern terminus in
the foothills of the Cascade Range east of Everett, Washington near
48.0° N lat., 120. 7° W long.
The short wave length gravity anomalies over Juan de Fuca
Ridge are due primarily to the effects of bottom topography. The
100
-J
q
(.0
50
-50
I-. -100
-150
0
5
(0
15
330
/
'\ \\\
/
i
327
327
502
295
20
25
20
25
ci.
I0
Lu
Lu
30
30
S
CENTRAL WASHINGTON
CRUST A
SUBCRUSTAL CROSS SECTION BB
FREE-AIR AND BOUGUER GRAVITY ANOMALY PROFILE
35
i/
\\\
40
COWIJTED SRAVITY0000
SEISMIC REFRACT1OI LINES (AFTER SHOR H at, 19681
DENSITES OF LAYERS IN GMiCM3
45
35
OBSERVED GRAVITY -
/
40
332
OREGON STATE UNIVERSITY
050100
HORIZONTAL SCALE
KILOMETERS
45
MARCH 1968
50
50
Figure 14.
Central Washington crustal and subcrustal cross section BB'.
'.0
92
average free-air anomalies across the ridge and across the Cascadia
Abyssal Plain, shown in Figure 14, are near zero, indicating that
these two provinces are essentially in isostatic equilibrium. Adjacent
to the ridge in Tufts Abyssal Plain, a small number of gravity
measurements indicate a negative free-air anomaly. Beginning at
approximately 127° W long, near the base of the continental slope
gravity anomalies become very irregular and show a gradual
decrease toward Puget Sound.
Gravity data on land are Bouguer anomalies (Woollard and Rose,
l963 uncorrected for topography; Sauers, 1966, unpublished). Gravity
anomalies in the Puget Sound region are not consistent with values
expected for the observed elevations. Danes étal. (1965) noted that
Puget Sound region exhibits a regional gravity low in addition to
individual large negative anomalies. Isostatic reductions by
Heiskanen (1938) and Banks (1969) indicate a -100 mgal negative
isostatic anomaly in the vicinity of Seattle.
It may be concluded
from these observations that the Puget Sound region is not, at least
locally, in isostatic equilibrium. The free-air and Bouguer
anomalies suggest that the region of isostatic disequilibrium may
extend westward as far as the continental slope.
Refraction lines (Shor etal.
,
1968) across the ridge show the
mantle to be as shallow as 7 km under the crest and to dip away from
the ridge to depths of about 10 km under the adjacent Cascadia and
93
Tufts plains. Velocities in the mantle are low on the ridge (7. 7 km/
sec or less) and slightly low (7.9 to 8.0 km/sec) under the adjacent
plains.
Crustal layering and mantle velocities in western Washington
are not defined by seismic refraction data; however, they are in
general agreement with the results of Tatel and Tuve (1955) who
indicate a depth to the mantle of 19 km in Puget Sound and 30 km
east of PugetSound. Crustal velocities beneath the Puget Sound
region show large variations (Neuman, 1959). Mantle velocities are
near 8.0 km/sec (Tatel and Tuve, 1955).
Seismic refraction measurements indicated by heavy lines on
the figure define the crustal layers of cross section BB' beneath Juan
de Fuca Ridge and Gas cadia Abyssal Plain. The seismic ties along
the continental shelf and slope were extrapolated a short distance
along isobaths from measurements off the Oregon-Washington coast.
The layers indicated by dashed lines are less reliable than those
indicated by solid lines. The lines of parallel dashes outline the
sides of polygons used in the computations and are used to suggest
either a gradual density transition or an indefinite structural
dimension.
The upper curve in Figure 14 shows the gravity anomalies. The
solid line represents observed free-air anomalies at sea and Bouguer
anomalies on land, and the circles represent calculated anomalies.
94
The sections assume a constant density beneath an arbitrary
depth of 50 km. For this condition a mantle density of 3. 32 gm/cm3
beneath Washington and 3. 30 gm/cm3 beneath Tufts Abyssal Plain
i s consistent with the observed gravity anomalies. A lower mantle
density of approximately 3. 27 gm/cm3 appears to exist under Juan de
Fuca Ridge and Cascadia Plain down to a depth of 50 km; a lower
density is also possible in which case the mantle layer would be less
thick.
The gravity data require the presence of a still lower density
material in the upper mantle beneath Juan de Fuca Ridge. The density
and configuration of the low density mantle material beneath the
ridge can be varied provided the gravity anomalies are satisfied.
Dehlinger etal. (1968) also postulated a low density upper mantle
beneath Juan de Fuca Ridge and similarly beneath the Gorda Ridge
(Dehlinger etal., l967a).
At the continental margin refraction results indicate the
Moho discontinuity dips steeply beneath the continent. Few
data on crustal layers are available for Washington. Based on a
mantle density of 3. 32 gm/cm the section indicates a depth to the
Moho of 19 km in western Washington, 36 km beneath Puget Sound
and approximately 34 km beneath the western flank of the Cascade
Range. The 19 km depth west of Puget Sound and 34 km east of
Puget Sound agree well with the seismic refraction results of Tatel
95
and Tuve (1955). Danes etal. (1965) attribute the large negative
gravity anomaly in the vicinity of Seattle to a 10 km deep sedimentary
basin. The cross section agrees in general with their interpretation
and suggests displacements of the crust which form the sedimentary
basin may extend to mantle depths.
A large distortion in the sedimentary structures near 125° W
long, is required by the gravity profile. The inferred structure may
be interpreted as severe folding of the sedimentary layers or normal
faulting of the layers with the west side down relative to the east side.
Gravity data cannot distinguish between these two interpretations.
Von Hueneetal. (1969) concluded, on the basis of seismic
profiler records, that very little if any horizontal crustal compression has occurred along continental margins where there are no
oceanic trenches. Very small graben-like features, filled with
sediment of density 1.90 gm/cm3, are observable on the cross section near 126° W long. These structures have also been observed
on seismic reflection records obtained by the University of Washington, and appear to be filled with unconsolidated sediments
(McManus, l967c).
These structures suggest crustal
tension.
A seismic reflection profile, reported by Ewing etal. (1968),
across the continental slope and Cascadia Plain approximately 200 km
north of the Washington cross section, shows a sediment-filled trough
at the base of the continental slope. The trough appears to have been
formed by faulting of the basement rock near the base of the slope
with the oceanic side down relative to the continental side. Inter-
preted as a normal fault, this structure also suggests tenson along
the continental slope off Washington.
The occurrence of tension along the continental slope would sug-
gest that the structure inferred from the gravity measurement is a
large normal fault offsetting the upper sedimentary layers and
extending into the lower crustal layer. The trough along the scarp
formed by the fault is apparently sediment filled.
The South Vancouver Island Cross Section
The south Vancouver Island cross section CC' is shown in
Figure 15. The western end of the south Vancouver section is located
in Tufts Abyssal Plain, near 44.7° N lat., 132. 7°W long. The
cross section strikes N 53° E cutting obliquely across Juan de Fuca
idge and the Cascadia Abyssal Plain. The section is normal to the
trend of the shelf and crosses the coast of Vancouver Island near the
south side of Barkley Sound. It traverses the Coast Mountains and
has its eastern terminus on the east flank of the Rocky Mountains,
near 52.3° N lat., 117.l°.W long.
The average free-air gravity anomalies over Juan de Fuca
Ridge and Cascadia Abyssal Plain are near zero, indicating that these
0
3
-Sc
(A
0C
IC
3-
I'-
'SC
0
0
0(
:
at,
to
:
15
20
3C
508
25
25
SOUTH VANCOUVER ISLAND
CRUST *2(1 SJBCRUSTAL CROSS SECTION CC'
FREE-AIR *2(1 BOUSUER GRAVITY AP4ALY PROFLE
ORSERVED
40
908IZQVThL
0
50
sraE
68
35
66 DIETERS
CJTED AN088ALY 00000
SEISMIC REFRAC11ON UNES (AFTER 514CR NI. 960SENSTES
45
NALY
LAYERS GOACM'
OREGON STATE INVERSITY
APRL 968
50
45
50
Figure 15. South Vancouver Is land crustal and subcrustal cross
section CC'.
J0
two provinces are essentially in isostatic equilibrium. The free-air
gravity anomalies become negative over the continental slope and dip
sharply negative over the continental shelf. The negative anomaly
over the shelf may be interpreted as reflecting a sedimentfilled
synclinal depression.
Bouguer anomalies on land are based on gravity studies
reported by Miller and Hughson (1936), Garland and Tanner (1957),
and Walcott (1967). Isostatic gravity anomalies reported by Miller
and Hughson (1936), Garland and Tanner (1957), Walcott (1967), and
Banks (1969) indicate that regionally Vancouver Island, Georgia
Strait and the Coast Ranges of British Columbia are
in
isostatic
equilibrium.
Seismic refraction measurements reported by Shor etal. (1968)
provide data on crustal layers and velocities in the mantle across
Juan de Fuca Ridge and Cascadia Abyssal Plain. Refraction results
reported by Richards and Walker (1959), which indicate a Moho depth
of 43 km beneath the Alberta Plains were used for control along the
east end of the south Vancouver section. The crustal layers and
mantle velocities in western British Columbia are not defined by
seismic refraction data. However a Moho depth of 34 km under the
Interior Plateau between the Coast and Rocky mountains of British
Columbia is near the 31 km depth determined by White and Savage
(1965) from an unreversed refraction profile.
Beneath the continent, layer thicknesses and depths depend on
the assumed densities; consequently, the structures are indicated by
dashed lines. Garland and Tanner (1957) attribute the smaller
Bouguer anomalies of interior British Columbia to anomalous
densities in the upper crust and the larger anomalies to warping of the
base of the crust. The crustal structure in Figure 15 shows a similar
interpretation.
White and Savage (1965) concluded that the crust of Vancouver
Island is 45 km thick. Ellis etal. (1968) have reanalyzed the seismic
explosion data obtained by the Dominion Observatory in the Vancouver
Island region from 1953 to 1963 and conclude that the Moho is deeper
than 56 km. These estimates of Moho depth are based on the absence
of identifiable Pn arrivals at distances greater than 400 km. The
depth to Moho beneath Vancouver Island as indicated on the South
Vancouver Island section is approximately 27 km.
A crustal thickness of 56 km would require an intermediate
crustal layer of density 3. 22 gm/cm3 approximately 42 km thick to
contain the same overburden mass as indicated for the Cascadia
Abyssal Plain. The high density and thickness of the intermediate
layer required by a deep Moho seem excessive for an acceptable
crustal section.
Walcott (1967) presented a gravity profile in a line of section
across Juan de Fuca Strait which extended north over the Vancouver
100
Island gravity high and south over the Olympic Mountains low, He
postulated that the anomaly is produced by the edge affect of two
juxtaposed crustal slabs of different density and thickness.
He
assumed that the thickness of the crust below Vancouver Is land is
51 km based on the results of White and Savage (1965) and showed that
a crustal block beneath the Olympic Mountains 33 km thick and 0. 2
gm/cm3 less dense than the Vancouver Island crustal block would
satisfy the gravity requirements. If, however, the crust is
assumed to be 19 km thick beneath western Washington in accord
with the Washington section,the same density contrasts (Walcott,
1967) lead to .a 28 km thickcrustbeneath Vancouver Island in
good agreement with the South Vancouver Island section.
The Washington and Vancouver Island sections suggest that
approximately half of the large gravity anomaly observed over the
Strait of Juan de Fuca is due to an 8 to 10 km vertical step in the
Moho and half due to an approximately 0. 2 gm/cm3 average density
contrast in the upper crustal rocks.
Crustal deformation at the base of the slope, apparent on the
Washington (Figure 14) and Scott Islands (Figure 16) sections, is not
present on the South Vancouver Island section. Figure 6 shows a
saddle in the gravity anomalies where the section crosses the juncture between abyssal plain and continental slope. The discontinuity
in the crustal deformation along the base of the slope at this location
101
may be due to an unexplained structural change or simply due to a
change in strike of the edge of the continental block.
The Scott Islands Cross Section
The Scott Is lands crustal and subcrustal section DD' is
illustrated in Figure
16.
The western end of the Scott Islands section
is located in Tufts Abyssal Plain, near 47. 30 N lat., 135.2° W
long. The section strikes N 55° E and intersects the continent
between the north end of Vancouver Island and the Scott Islands.
Its eastern terminus is in the Interior Plateau of British Columbia
near 52.40 N lat. , 12 5.
10
W 1 ong
The average free-air
anomalies over Cobb Rise and Explorer Ridge are near zero,
indicating these two regions are approximately in isostatic
equilibrium.
A large negative anomaly associated with the Scott Islands
fracture zone occurs over the continental slope. The free-air
anomaly is positive landward of the shelf break between the north
end of Vancouver Island and the Scott Is lands and near zero over
Queen Charlotte Strait. Bouguer anomalies along the Inside Passage
of British Columbia, reported by Banks
(1969),
decrease rapidly
toward the Coast Mountains.
Average free-air anomalies over Queen Charlotte Strait and
isostatic anomalies along connecting fjords (Banks,
1969)
are near
50
so
0
0
-4
-50
-50
- -100
-too
L
I50
5
5
132
130
N55E.
0
2.40
0
103
0
274
5
--------,'--__
2.00
-
C
2S0
5
I
\1
2.74
10
3
/
2O
I.
3.
IIIj
;JIJliI11IIrI
20
3.00
25
25
30
35
SCOTT ISLANDS
4
30
CRUST AND SUBCRUSTAL CROSS SECTION DD
FREE-AIR AND BOLJGUER GRAVITY ANALY PROFILE
OBSERVED GRAVITY
35
332
COtI'UTED GR/ffY 0000
40
SEISMC REFACTION LUES (AFTER MLPL 964)DNS1TIES
LAYERS Il GMICM3
40
NOR14TAL SCALE
0
45:
OREGON STATE INVERSITY
MAY
968
50
KILOTERS
100
45
50
50
Figure 16. Scott Islands crustal and subcrustal cross section DD'.
0
103
zero indicating that the area about the continental end of the Scott
Is lands section is in isostatic equilibrium.
Seismic refraction measurements reported by Milne (1964)
show a mantle velocity of 8. 4 km/sec and a relatively thin, 10 km
thick crust in the southern part of the Alaskan Abyssal Plain. These
measurements provide data on crustal layers and velocities on the
west end of the section. Beneath Cobb Rise, Explorer Ridge, and
the continent layer, thicknesses and depths depend on assumed den-
sities; consequently, the structures are indicated by dashed lines.
Gravity values of Miller and Hughson (1936) and Garland and Tanner
(1957) and the seismic refraction results of Richards and Walker
(1959) were extrapolated over the Alberta Plains to provide com-
putational control for the east end of the section. The section shown,
however, is terminated in the Interior Plateau of British Columbia.
If near normal crustal densities are assumed for Cobb Rise and
Explorer Ridge, light material must be postulated in the upper
mantle to contain the same over-burden mass as indicated in the
Alaskan Abyssal Plain. Figure 16 shows an upper mantle structure
similar to the structure shown in Figure 14 beneath Juan de Fuca
Ridge. A material of low density, here assumed to be 3. 18 gm/cm3,
overlies a less low density material of 3. 27 gm/cm3 and extends
from Eickelberg Ridge near 132.8° W long, to the continental
shelf.
104
The Moho off the north end of Vancouver Is land is shown as 27
km deep; if lighter average crustal densities are assumed the Moho
will be correspondingly shallower. The Bouguer anomaly over the
coast mountains is attributed to relatively light rocks in the upper
crust and a crust extending to depths of approximately 35 km.
No attempt was made to fit exactly the short wave length gravity
anomalies caused primarily by the very irregular topography of Cobb
Rise and Explorer Ridge. The negative anomaly near
131.10W long.
suggests a sediment-filled 2 km deep moat along the west side of
Explorer seamount, similar to the trough along the south side. The
sharp negative anomaly near 129.
70
W long, is primarily due to the
median trough (Explorer trough) of Explorer Ridge. The higher
elevation of Explorer Ridge is shown compensated by a thicker section
of the 3. 18 gm/cm3 mantle material.
The large negative anomaly of the Scott Islands fracture zone
near 129° W long, is interpreted on the section as a graben. A
crustal block of 2. 8 gm/cm is shown downthrown 6 km and the
resulting valley filled with two 3 km thick sediment sections of
average densities 2, 0 gm/cm3 and 2. 4 gm/cm3.
Ewing etal. (1968) reported a seismic reflection profile which
passes near the north end of the Scott Islands anomaly. The profile
shows normal faulting and narrow-filled trenches near the base of the
slope but no evidence of folding. The interpretation of the negative
105
anomaly of the Scott Islands fracture zone as crustal downfaulting is
supported by the seismic reflection observations.
The large negative anomaly associated with the Scott Islands
fracture zone is shown in Figure 17, The -100 mgal contour used as
an outline indicates that the structure is not a simple graben. The
irregular anomaly contours suggest that either it is composed of a
large number of small blocks or that a large graben has been subjected to shear and consequently deformed.
Figure 18 shows the total magnetic intensity observed over the
Scott Islands fracture zone. This zone exhibits very little magnetic
relief. The magnetic highs near 50.7° N lat. , 129.1° W long, and
50. 3° N lat., 128. 3° W long, coincide with changes in the contours
of the free-air anomalies and are probably due to near-surface basement rocks along the outer edge of the continental shelf. The large
magnetic highs near 50.0° N lat,
129. 7° W long, and 49.9° N lat.,
129. 6° W long, are the north end of a long magnetic lineation and in
this area coincide with the negative gravity anomalies associated with
the north end of the Explorer trough.
Figure 19 shows a composite profile along line S-S' across the
Scott Is lands fracture zone where the observed gravity anomaly is a
minimum. The location of profile S-S' is shown in Figure 8. The
gravity and magnetic anomalies are negative from the continental
shelf to the abyssal depths of the sea floor, The minimum (-160 mgal)
106
1300
- - - - - - -
128°
129°
r
510
510
I
500
500
1290
Figure 17. Free-air gravity anomaly map of the Scott
Islands fracture zone.
1280
107
130°
p - - - - - _ - - - - qo
SCOTT
J
ISLANDS
TOTAL MAGNETIC INTENSITY
-
1
CONTOUR INTERVAL
\\'\.
51°
510
010
:
'1'
-
50°
50°
129
Figure 18.
Total magnetic intensity map of the Scott Islands
fracture zone.
128
I
SCOTT ISLAND FRACTURE ZONE
a
a
GRAVITY, MAGNETIC,
E
BATHYMETRIC PROFILE
ALONG LINE S-S
-c
C-,
IId
z
C,
4
+250
a
0
a
E
e-.
aI&I
Id
0 -100
0
400
800
-250 -200
1200
1600
-500 -300
2000
2400
0
20
40
60
DISTANCE (kilometers)
Figure 19. Bathymetric, gravity and magnetic profile across
the Scott Islands fracture zone along line S-S'.
I.1.'
109
in the free-air anomaly 60 km from the southwest end of the profile
may indicate a fault at the base of the slope. The bathymetric highs
at 8 km and 26 km are named Paul Revere Ridge and Winona Ridge
respectively on a bathymetric map of the region compiled by McManus
(1964).
The Moresby Island Cross Section
The Moresby Island crustal and subcrustal section EF' is
shown in Figure 20. The western end of the Moresby Island section
is located in the Alaskan Abyssal Plain near 49.9° N lat. , 138.2° W
long. The section strikes N 42° E across the Alaskan Plain and inter-
sects the continent near the center of Moresby Is land. The section
crosses Hecate Strait and intercepts the Coast Mountains in the
proximity of Douglas Channel. It5 eastern terminus is in the Interior
Plateau of British Columbia near 55° N lat., 126° W long.
The average free-air anomalies over the Alaskan Abyssal Plain
are near zero indicating that the plain is in approximate isostatic
equilibrium. A large negative free-air anomaly occurs over the
continental slope off Moresby Island. Bouguer anomalies are
positive over Moresby Island (Stacy, 1967). Free-
air anomalies in
Hecate Strait average near zero, and Bouguer
anomalies (Banks, 1969) decrease rapidly along the Douglas channel
toward the Coast Mountains.
100
100
-.4
50
50
0
O
-50
-50
K
-(00
-100
'
H
-ISO
K
-I5O
5138
05
134
136
130
ZL
I28
267
I 03
200
210
297
'A -
.1 \I
L'.,
-
10
273
_-_
I
/
276
-j_
--------
IS
-
20
EL
-5
297
10
IS
- 20
25
30 -
MORES BY
ISLAND
CRUST AND SUBCRUSTAL CROSS SECTION EE
35 -
FREE-AIR AND BOUGUER GRAVITY ANOMALY PROFILE
OBSERVED GRAVITY
332
COMPUTED GRAVITY 000000
40 -
SEISMIC REFRACTION LINES (AFTER SHOR,I962) -
40
DENSTIES OF LAYERS IN GM/CM3
45 -
OREGON STATE UNIVERSITY
MAY
1968
HORIZONTAL SCALE
0
50
00
KILOMETERS
50
50
Figure 20. Moreeby Island crustal and subcrustal cross section EE'.
C
111
Average free-air anomalies over Hecate Strait and isostatic
anomalies along Douglas Channel (Banks, 1969) are near zero,
indicating that the area about the continental end of the Moresby
Island section is in approximate isostatic equilibrium.
Shor (1962) reported the results of seismic refraction measurements along a profile southwest from Dixon Entrance in the Alaskan
Abyssal Plain. The Moho is about 1.5 km shoaler between Hodgkins
Ridge and the Queen Charlotte trough than west of Hodgkins Ridge.
Mantle velocities are approximately 8. 3 km/sec west of Hodgkins
ridge and 8. 1 km/sec east of the ridge. A long compound refraction
profile from the northern tip of Graham Is land east to Celestial Reef
suggests that Moho depths are approximately 25 km near Dixon
Entrance.
The Alaskan Plain shows no marked changes in topography
between the seismic refraction profile of Shor (1962) and the similar
refraction measurements near the southern edge of the Alaskan Plain
reported by Milne (1964).
The seismic and bathymetric observa-
tions suggest the southeast sector of the plain has a uniform structure.
The refraction results of Shor (1962), therefore, were extrapolated
southeast along isobaths to provide a guide to crustal and subcrustal
layer depths in the Alaskan Plain and the depth of the Moho beneath
Hecate Strait. Structures of the upper crust beneath Moresby Island
and east and west of Moresby Is land are not controlled by seismic
112
measurements and, therefore, are shown as dashed lines.
The Moresby Island section indicates the Moho shoals to
approximately 9 km immediately west of the Queen Charlotte troug.h
and then descends to a depth near 25 km beneath Moresby Island and
Hecate Strait. The densities assumed for the crust at the east end
of the section result in a Moho depth of 40 km beneath the Coast
Mountains. The combined gravity and seismic observations require
a decrease in density of the upper mantle beneath the continental rise.
Short wave length gravity anomalies near 134° W long, suggest
faults or intrusions in the upper crust beneath the surface sediment
layer.
The wave lengths of the gravity anomalies are approximately
the same as the wave lengths of the magnetic anomalies in the region.
However, magnetic measurements were not made along the same
track line as gravity measurements; consequently, the results of a
correlation of gravity and extrapolated magnetic data were inconclusive,
Figure 21 shows total magnetic intensities southwest of Moresby
Island. The magnetic lineations observed over the Alaskan Plain
trend N 20 E off Moresby Island and appear to terminate at the base
of the N 43° W trending continental slope. The gravity and magnetic
anomalies show little correlation; the structure producing the welldefined gravity high at 52.1° N lat., 131.6° W long, causes only
a small perturbation in the magnetic field. The magnetic high over
113
Ie
I
5
N
530
MORESBY ISLAND
TOTAL MAGNETIC INTENSITY
CONTOUR INTERVAL
00 gammas
50 gammas
/ /
-- / Ii
I,
',
I
Ii
I
I
I
Ill
7
:
I
It
I
I
1
I
)
Ii
-
I'
e 4
'I
52
-
:
59
S
10
S
63
64
(a/lu'
1310
Figure 21.
Total magnetic intensity map of the area
off Moresby Island.
114
Moresby Island may be associated with the syntectonic plutons,
described by Sutherland-Brown (1966), which are concentrated along
the western flank of the Queen Charlotte Islands.
The interpretation of the crustal structure of Moresby Is land is
in good agreement with the structure proposed by Sutherland-Brown
(1966) based on geologic observations except that the large fault along
the west side of Moresby Island is shown dipping away from the island
rather than under the island. The actual dip of the predominantly
right lateral strike-slip fault is unknown.
Sutherland-Brown (1966) described a graben-like trough of
early Cretaceous age in the central area of the Queen Charlotte
Islands and also large northwesterly faults along the eastern flank
of the islands which, in addition to right lateral strike-slip movement, exhibit normal, east block down displacement. This suggests
east-west tensional stresses over the region contemporary with
transcurrent motion along the Denali-Queen Charlotte Islands fault
system off the west coast of the Queen Charlotte Islands. In a
region of tension normal or transcurrent faulting rather than thrust
faulting is expected; and, therefore, the large trans current faults
along the west side of the Queen Charlotte Islands may dip seaward
and have a normal fault component with the ocean down relative to
the islands.
Sutherland-Brown (1966) indicated that folding is less important
115
than faulting in the Insular Belt of British Columbia; consequently,
the short wave length gravity anomalies in Hecate Strait are interpreted as caused primarily by abrupt changes in water depths and
faulting and/or glacial gouging of the upper crustal layers.
The Dixon Entrance Cross Section
The Dixon Entrance crustal and subcrustal section FF' is
shown in Figure 22.
The western end of the Dixon Entrance section
is located in the Alaskan Abyssal Plain near 52.90 N lat.
long.
138.90 W
The section strikes N 69° E and intersects the continent in
Dixon Entrance off the north end of Graham Island. Its eastern
terminus is in the Interior Plateau of British Columbia near 55. 6° N
lat. , 1 27. 20 W long.
The average free-air anomalies over the Alaskan Plain, except
for the local anomaly over Hodgkins Ridge, are near zero indicating
the region is essentially in isostatic equilibrium. A large negative
anomaly occurs over the continental slope. The average free-air
anomaly is slightly positive landward of the edge of the continental
shelf and decreases to near zero in the eastern part of Dixon
Entrance. Bouguer anomalies along the Pearse Canal and the Portland
Canal, reported by Banks (1969), decrease rapidly toward the Coast
Mountains. Average free-air anomalies over the eastern part of Dixon
5C
50
C
0
-50
-oo
-150
j-In
5
138
136
-5
134
132
130
128
-
0
2.I0____
103
5
216
-
0
5
0
15
(5
20
25
3O
20
7,
25
XON
ENT
CE
30
RUST AND SUSCRUSTAL CROSS SECT
35
-
FREE-AIR AND BOUGUER GRAVITY ANOMALY PROFILE
068ED GRA(A
40
SEISMIC REFRACTION LINES
/
(AFTER SHOR. 9a
DENSITES OF LAYERS IN G$4C&13
MARCH
/
-
40
I
HORIZONTAL SCALE
0
50
00
OREGON STATE UNWERSITY
45
35
3.32
COMPUTED ONAV1TY 0000
1968
-
KILOMETERS
45
/
Figure 22. Dixon Entrance crustal and subcrustal cross section FF'.
50
-j
117
Entrance and isostatic anomalies along the Portland Canal (Banks,
1969) are near zero indicating that the area about the eastern end of
the cross section is approximately in isostatic equilibrium.
Shor (1962) reported the results of seismic refraction measurernents along a profile southwest from Dixon Entrance into the Alaskan
Abyssal Plain. The profile coincides closely with the Dixon Entrance
section and provides references for the crustal and subcrustal structures in the Alaskan Plain and in Dixon Entrance. Mime (1964)
reported a reversed profile shot in a north-west,south- east direction
off the north tip of the Queen Charlotte Islands. These results pro-
vide data on the structure of the upper crust near the edge of the continental shelf.
The Dixon Entrance section indicates the Moho shoals to
approximately 9 km beneath the continental rise immediately west of
the continental slope off Dixon Entrance and then descends to approxi-
mately 25 km beneath Dixon Entrance. The densities assumed for the
crust at the east end of the section result in a Moho depth near 40
km beneath the Coast Mountains. The combined gravity and seismic
observations require a decrease in the density of the upper mantle
beneath the continental rise.
The gravity anomaly near 1 36. 5° W long, is due to Hodglcins
Ridge and seamount. The two-dimensional analysis inaccurately
portrays the three-dimensional structure; however, a moat can be
118
noted along the ridge flanks, and regional isostatic compensation is
suggested.
Figure 4 shows magnetic anomalies determined from magnetic
measurements made along the Dixon Entrance gravity profile. The
results of a correlation of the magnetic anomalies and the short wave
length gravity anomalies between Hodgkins Ridge and the continental
slope were inconclusive.
Shor (1962) reports that an unreversed seismic profile at the
edge of the continental shelf west of Graham Island indicates a sedi-
mentary layer 3 km thick. The cross section indicates sediment 3
km to 5 km thick near the edge of the continental shelf. The large
negative anomaly over the continental slope is attributed to a thick
sedimentary wedge along the slope and associated faulting of the
crust with the ocean side down relative to the continental shelf. The
maximum sediment thickness actually occurs landward of the axis
of the Queen Charlotte trough. The short wave length gravity anomaly
near 134.2° W long, superimposed on the large negative anomaly
may mark the location of a fault near the base of the slope.
A high density crustal block beneath the outer shelf is probably
related to the substructure of the Queen Charlotte Islands. Both
seismic refraction (Shor, 1962) and gravity measurements suggest
extensive faulting and structural changes in the upper crust of the
continental shelf and Dixon Entrance. The negative anomaly near
119
131.8° W long, is associated with an extension of the topography of
Clarence Strait in southeast Alaska.
The Continental Margin Structures off Washington
and British Columbia
The five crustal and subcrusta]. cross sections are effectively
connected on the western end by a common section; consequently,
the structures outlined on the sections may be compared and, with
the aid of Figures 6 and 7, their lateral extent estimated.
The central Washington and south Vancouver sections indicate
a Moho depth of 7 to 9 km and a relatively light material in the upper
mantle beneath Juan de Fuca Ridge. The Scott Islands section also
suggests a Moho depth of 9 km and a similar low density material
in the upper mantle beneath the Explorer Ridge. In the Scott Islands
section the low density upper mantle is a consequence of the higher
topography of Cobb Rise and Explorer Ridge without an associated
increase in the free-air gravity anomaly, The region, bordered on
the northeast by the Scott Island fracture zone and on the southwest
by Eickelberg Ridge and Cobb Seamount, exhibits numerous sea-
mounts and an average depth less than the surrounding abyssal regions
whereas the free-air gravity anomaly averages near zero over both
the region of seamounts and the abyssal plains. This implies that
the entire region of approximately 135, 000 km2 may have a relatively
120
shallow Moho and light material in the upper mantle.
If, however, a shallow Moho and light upper mantle materials
are confined to the immediate vicinity of Explorer and Juan de Fuca
ridges Moho depths in the rest of the region will be approximately 12
to 13 km. In this event the crustal thickness, exclusive of the water,
will be 9 to 10 km compared to a crustal thickness of approximately
6 km beneath the surrounding abyssal plains.
All five cross sections show the upper mantle, in the region
extending from the coastline out to a line which closely approximates
the 3500 m (1800 fm ) isobath to be slightly less dense than the upper
mantle of Tufts and Alaskan abyssal plains farther west. The region
of low density mantle apparently extends from the Mendocino escarp-
ment (Dehlinger etal., 1967a; Dehlinger etal., 1968) to and possibly
including the continental rise off southeast Alaska.
A deep trough is located near the base of the continental slope
off British Columbia and extends from Dixon Entrance to the Strait
of Juan de Fuca. Off the Queen Charlotte Islands the trough is
associated with the Queen Charlotte trough, a bathymetric depression;
and west of Vancouver Island it is exemplified by the Scott Islands
fracture zone, with an estimated sediment thickness of 6 km or more.
West of Washington a smaller sediment-filled trough occurs
along the base of the slope. Gravity measurements suggest a filled
trough is also located on the continental slope parallel to the trough
121
near the base of the slope. A relatively high density ridge or anticlinal structure just landward of the edge of the continental shelf
apparently extends the length of the Insular Belt of British Columbia.
This structure is absent off Washington. The central Washington and
Oregon sections (Dehlinger et al., 1968) indicate that the !igranitic
layer of the continent appears to be absent in southwestern Washington
and northwestern Oregon.
The central Washington sections shows a Moho depth of 19. 5
km under western Washington. A vertical offset of 8 to 10 km occurs
in the Moho in the vicinity of the Strait of Juan de Fuca. The cross
sections of British Columbia show mantle depths under the Insular
Belt of approximately 25 to 28 km. The east end of the sections are
based on little structural control,but they suggest that Moho depths
are approximately 35 to 40 km beneath the coast mountains of Canada
and Cascade Range of northern Washington.
122
CONCLUSIONS
Near zero free-air anomalies west of Washington and British
Columbia indicate the entire region is approximately in isostatic
equilibrium and that isostatic adjustment is more rapid than tectonic
upheaval. The regional gravity low of Puget Sound and the Olympic
Peninsula, which suggests isostatic inequilibrium, extends westward to the edge of the continental shelf. Isostatic compensation of
individual searnounts in the region appears to occur over an area
approximately 7 times the basal area of the seamount.
Except for isolated seamounts, free-air anomalies, as elsewhere
in the Pacific Ocean (Dehlinger, 1969), are not correlative with
topography. Over Juan de Fuca Ridge free-air anomalies show no
apparent relation to the ridge nor to the magnetic anomalies which
define the ridge.
A free-air anomaly of approximately -100 mgalis associated
with the partially sediment-filled Queen Charlotte trough west of the
Queen Charlotte Islands. Off the north end of Vancouver Is land a
free-air anomaly greater than -150 mgaloccurs over the Scott Island
fracture zone. An anomaly less than -50 mgal occurs along the base
of the continental slope off Washington. The negative anomaly along
the base of the continental slope off British Columbia and Washington
is caused by the dip of the Mohorovicic discontinuity, lateral density
123
differences in the upper mantle and a sediment-filled trough.
In the region between the coastline and a line which approxi-
mately follows the 3500 m isobath, a mass deficiency must occur in
the upper mantle where densities are slightly less than in the Tufts
and Alaskan abyssal plains and in the adjacent continent. Still lower
densities appear to occur in the upper mantle beneath the Explorer
and Juan de Fuca ridges.
Explorer Ridge, proposed to be an active locus of sea-floor
spreading, exhibits moderate s eismicity, asymmetric wall structures
and magnetic anomalies, and a median trough with approximately two
kilometers of relief and little sediment fill.
Moho depths are near 25 to 28 km in the Insular Belt of British
Columbia and 19 km in western Washington. Moho depths appear to
be near 20 km in western Oregon (Dehlingeretal. , 1968;
Thiruvathukal, 1968) and 21 to 23 km in central California (Tatel and
Tuve, 1955; Thompson and Taiwani, 1964). The relatively thin crust
in the region between the continental shelf and coast mountains of
British Columbia, the Cascade Range in Washington and Oregon and
the Sierra Nevada in California, which includes the coast ranges of
California, Oregon and Washington, is apparently characteristic of
the transition from oceanic to continental structure in western North
America.
124
REGIONAL TECTONICS
Two tectonic processes active in the region west of Washington
and British Columbia are: 1) a geotectonic process causing the for-
mation of the mid- ocean, ridge-rise system, and 2) isostatic adjustment tending to reduce hydrostatic imbalances caused by the lateral
transport of material.
Morgan (1968b) and LePichon (1968) interpreted the placement
and configuration of the mid-ocean ridges and their related faults and
magnetic anomaly patterns as a consequence of the rotation, displacement and interaction of rigid crustal blocks moving about the earth's
surface on a deeper, more fluid layer. Each of the rigid blocks is
bounded by rises where new crustal material is formed as the blocks
separate, trenches or young fold mountains where crustal materials
are being destroyed, and great faults along which the blocks slide
past one another. The seismological studies of Isacks etal. (1968)
give impetus to this interpretation, particularly with respect to the
destruction of crustal material in the vicinity of the deep trenches.
West of Washington and British Columbia the Juan de Fuca and
Explorer ridges correspond to the boundary of separation between the
north Pacific Ocean block and the North American continent block.
The Queen Charlotte fault is a great fault along which the northeast
Pacific Ocean is moving northwestward relative to British Columbia
125
and southeast Alaska eventually underthrusting southeast Alaska and
the Bering Sea in the vicinity of the Aleutian Trench.
An analysis of fault plane solutions of earthquakes in Washington
and off British Columbia (Hodgson and Milne, 1951; Hodgson and
Storey, 1954; Algermissen and Harding, 1965; Tobin and Sykes, 1968;
and Gallagher, 1969) indicates the strike of the minimum horizontal
stress axis is approximately N 90° W in western Washington and
the Strait of Georgia and N 300 W to N 600 W off the north end of
Vancouver Island, in Hecate Strait and north of Graham Island. This
suggests the entire region, north of the Strait of Juan de Fuca, is
subject to tensional stress in a general northwest-southeast direction.
The graben-like structures, indicated by the gravity and seismic
reflection studies, which occur in the north end of the Cascadia Plain,
along the base of the continental slope and on the continental slope
and shelf support the interpretation of the fault-plane solutions as
indicating regional tension.
The separation of crustal blocks along Juan de Fuca Ridge
causes the formation of new crust along the ridge axis and consequently an extension of the crustal blocks. Deep-tow magnetometer
measurements (Larson and Spiess, 1969) and model analysis (Emi.lia,
1969) suggest that along active ridges new crust may form as narrow
intrusions such as dikes (Bodvars son and Walker, 1964) or diapirs
(Maxwell, 1968). The formation of new crust along Juan de Fuca
126
Ridge implies either a net transfer of crust out of the area or destruction of crust in an unfamiliar manner, i. e. without earthquakes,
compressive features along continental margins or prominent gravity
anomalies. The patterns of stress and related faulting of the region
are interpreted here to indicate that no underthrusting of the continent
is occurring along the continental margin of Washington and British
Columbia. This implies northwestward migration of Juan de Fuca
Ridge.
The asymmetrical patterns of topography, gravity and heat
flow with respect to the ridge axis are not inconsistent with the concept of a migrating ridge. The ridge substructure, shown by the
crustal and subcrustal cross sections extends into the upper mantle,
indicating the upper mantle is also part of the crustal block in motion.
Vine (1966, 1968) proposed a 3 cm/yr spreading rate perpendicular to the axis of Juan de Fuca Ridge. The trend of the topographic
structures and gravity anomalies of the region and the strike of the
Queen Charlotte fault suggest that separation along the ridge axis is
slightly oblique and at a rate near 3. 2 cm/yr. If the ridge is migrat-
ing, the crust on the east side is stationary, whereas the crust on the
west side is moving northwest with a velocity near 6. 4 cm/yr. The
distance between the two magnetic anomalies identified as occurring
on the ridge crest 10 million years B, P. (Vine, 1968) indicate an
average velocity of separation parallel to the Queen Charlotte fault
near 8 cm/yr. The additional crustal extension (or higher velocity
1 27
of separation) is apparently accommodated by strike-slip faulting
oblique to the ridge axis, clockwise rotation of the spreading axis,
and crustal extension by normal faulting.
The magnetic ages combined with the concept of a migrating
ridge indicate the ridge axis was located along the continental slope
off Washington and Oregon 10 million years ago. Vine (1966) noted
a marked change in the character of the magnetic anomalies near the
10 m. y. magnetic date. Ewing and Ewing (1967), in a study of the
sediment distribution on the mid-ocean ridges, noted a large change
in sediment thickness near anomalies with a 10 m. y. magnetic date.
They indicate a period of relative quiescence in sea-floor spreading
between late Mesozoic or early Cenozoic and 10 m. y. ago. LePichon
(1968) suggests each cycle of the intermittent spreading is marked by
a reorganization of the global pattern of motion. Menard and Atwater
(1968) interpret changes in strike of transform faults in the North
Pacific as indicators of changes in direction of block separation.
Stonely (1967) noted that Mesozoic and Cenozoic sediments depoisited
in a trough along a rapidly subsiding shelf in southeast Alaska were
deformed in late Cenozoic time by continuous movement of the Pacific
Ocean floor relatively toward the continental margin.
These observa-
tions suggest a change in geotectonic motion and a consequent change
in the regional stress pattern of the northeast Pacific in late Tertiary
time; Juan de Fuca Ridge, therefore, could have formed near the
base of the continental slope off Washington and Oregon approximately
10 m y. ago in response to a changing stress pattern.
128
The continental slope between northern California and the Strait
of Juan de Fuca presents a natural point of rifting between the nearly
en echelon San Andreas and Queen Charlotte faults.
The uplift of the
Coast Range and formation of the high Cascades (Baldwin, 1959) are
contemporary with the formation of the ridge 10 m. y. B. P. The
period of quiescence between the late Mesozoic and the occurrence
of rifting of the continental and oceanic blocks near the slope base is
contemporary with the filling of the Coast Range geosyncline. The
direction of relative motion between the continental and oceanic
blocks is oblique to the ridge axis; consequently, shear--predominantly NW trending right-lateral strike slip and NE trending
step faulting- -may have occurred along the continental margin prior
to and during the formation of the ridge and separation of the crustal
blocks. This hypothesis does not preclude pre-Tertiary thrusting
along the continental margin.
The apparent rotation of Juan de Fuca Ridge is due to the
tendency of active ridges to orient themselves normal to ridge terminating transform faults (Menard and Atwater, 1968). Ridge for-
mation, by block separation as megascale tension or cleavage fracturing and intruding or by crustal thinning and diapiric injection, is
expected to occur so as to involve a minimum of deformation energy.
Ridges tend toward an orientation normal to the trend of the regional
tensional stress vector and will rotate by growth so as to minimize
129
shear stress and shear fracture along the spreading axis. Explorer
Ridge, which formed approximately 4 m. y. B. P., is now normal to
the crustal velocity vector, the Queen Charlotte transform fault and
the continental margin of British Columbia. Prior to the formation
of Explorer Ridge the Scott Islands fracture zone acted as a sector
of the north-bounding transform fault of Juan de Fuca Ridge.
Analysis of magnetic anomalies (Vine and Wilson, 1965;
Morgan, 1968a, b) suggest Juan de Fuca Ridge is an active locus of
sea-floor spreading. Hamilton (1967), based on the observation of
undisturbed turbidities, suggests the Juan de Fuca portion of the East
Pacific Rise has been quiescent since middle Tertiary time. Portions of the ridge axis appear covered with sediment (McManus,
1965).
The free-air anomalies indicate isostatic adjustment has
occurred and no vertical tectonic imbalance is present. Earthquake
epicenters are notably absent along the central portion of the ridge
but occur along or near the ends of the ridge (Tobin and Sykes, 1968).
Duncanetal. (1969) note a small rift aligned with the axis of the
ridge on the south end, and Gallagher (1969) has obtained a fault plane
solution which suggests normal faulting parallel to the south end of the
axis of Juan de Fuca Ridge. These observations suggest that
spreading may not occur simultaneously all along the ridge and that
part of the ridge--the center portion- -may be temporarily quiescent.
The fault or fracture system connecting Juan de Fuca and Explorer
1 30
ridges and the Explorer Ridge and Queen Charlotte fault are, as yet,
undefined.
An analysis of fault-plane solutions of earthquakes off northern
California and southern Oregon (Byerly, 1938; Tobin and Sykes, 1968;
Bolt etal., 1968; and Gallagher, 1969) indicates:tbads of minimum
horizontal stress trends east-west, whereas the stress axis off
British Columbia trends northwest-southeast. The difference in
stress axes suggests that a north-south component of tension may
exist between the two regions. At present the north-south stress
component may be relieved by crustal extension in the form of normal fault components and possibly the formation of a small amount
of new crust along the Blanco fracture zone.
Vine (1966), Pavoni (1966) and Peter and Lattimore (1969) inter-
preted the smaller offsets in the linear magnetic anomalies noted by
Raff and Mason (1961) as strike-slip faulting. The offsets strike
nearly 30° to 60° to major structures of the region (e.g. Juan de
Fuca Ridge) and are consistent with regional tension in a NW-SE
direction. It is also possible that some of the offsets in the magnetic
anomalies may have formed in situ; the offsets may have propagtedin
the crust similar to the manner whereby crystal or lattice defects
are propagated or extended in crystals during growth. The magnetic
offsets or pseudo-faults may have started or been guided along old
fractures or crustal weaknesses. It is noted that the "Sovanco fault,
131
a major left-lateral strike-slip fault postulated by Pavoni (1966),
based on magnetic offsets and used in the arguments of Peter and
Lattimore (1969), passes through and is normal to a graben (Ewing
and Ewing, 1968) located near 49.00 N
128.8° W long. The
directions of stress necessary to form both features concurrently
are irreconcilable. Further the destruction of large quantities of
old crust required in their interpretation of the magnetic offsets is
incompatible with the structure of the region resulting. from the inter-.
pration of the combined gravity and seismic reflection measurements,
i. e. there is no evidence of sinking, down warping, overthrusting or
superposing of crustal blocks except possibly along the east end of the
Mendocino escarpment.
Sediment thickness increases eastward in the Cascadia Abyssal
Plain consistent with the concept of a migrating ridge or spreading sea
floor. In some areas the sediment surface appears to slope away
from Juan de Fuca Ridge. This may be in part attributed to uplift
of the ridge and in part to an isostatic adjustment or subsidence of
the oceanic crust proportional to sediment thickness. The importance
of isostatic adjustment is realized when it is noted that Cascadia
Plain contains nearly 68, 000 km3 of sediment, a sufficient amount
(via subcrustal mass displacement) to elevate the Coast Range and
continental shelf off Washington and Oregon 1000 meters. The near-
zero average free-air anomalies, thin crust and high heat flow of
132
Cascadia Plain suggest that isostatic adjustment is relatively rapid;
at least more rapid than tectonic upheaval.
Although the present erosion rate per unit area is 4 to 5 times
greater along the Coast Range than east of the Puget-Willamette
trough, most of the sediment in Gas cadia Plain comes from the
Columbia River drainage basin (Kulmetal., 1968). The analysis of
gravity anomalies observed in Oregon (Thiruvathukal etal., 1969)
indicates a slight mass excess in the Coast Range Mountains,
whereas the Columbia Plateau has a mass deficiency. This suggests
a possible phase lag in subcrustal displacement between Gas cadia
Plain and the continental interior or renewed ridge activity near the
continental margin.
The Rocky Mountain Trench (Garland and Tanner, 1957; Leech,
1965) closely parallels the large negative free-air gravity anomalies
of the Queen Charlotte trough, Scott Islands fracture zone and Puget
Sound depression. The Rocky Mountain Trench and the aligned
gravity minima flank the east and west sides, respectively, of the
Canadian Cordillera. If these two features are related, a late
Cretaceous or early Tertiary age is suggested for the structures
defined by gravity minima. The relatively thin crust of western
British Columbia may be a Mesozoic relic of oceanic underthrusting
of the continental margin.
133
BIBLIOGRAPHY
Algermissen, S. T., S. T. Harding, K. Steinbrugge and W. Cloud.
The Puget Sound, Washington, earthquake of April 29,
1965; preliminary seismological report and preliminary
engineering report. Washington, D. C., U. S. Coast and
1965.
Geodetic Survey.
51 p.
Apostol, T. M. 1957. Mathematical analysis. Reading, Mass.,
Addison-Wesley.
553 p.
Baldwin, E. M. 1959. Geology of Oregon. Eugene, distributed by
University of Oregon Bookstores. 136 p.
Banks, E. R. 1969. Isostatic and Bouguer gravity anomalies along
the Inside Passage of Alaska and British Columbia. Masterts
thesis. Corvallis, Oregon State University. 56 numb. leaves.
Benioff, H. 1954. Orogenesis and deep crustal structure: Additional evidence from seismology. Bulletin of the Geological
Society of America 65: 385-400.
Movements on major transcurrent faults.
In: Continental drift, ed. by S. K. Runcorn. New York,
Academic. p. 103-134.
1962.
Bennett, L. C., Jr. and M. Grim.
1968. Investigation of the continental shelf structure off the coast of Washington. (Abstract)
Transactions of the American Geophysical Union 49: 209.
Berg, J. W., Jr. and C. D. Baker.
1963.
Oregon earthquakes,
1841-1958. Bulletin of the Seismological Society of America
53: 95-108.
Berg, J. W., Jr. and J. V. Thiruvathukal.
Gravity base
station network, Oregon. Journal of Geophysical Research
1965.
70: 3325-3329.
Blackman, R. B. and J. W. Tukey. 1958. The measurement of
power spectra. New York, Dover. 190 p.
Bodvarsson, G. and G.P.L. Walker. 1964. Crustaldriftinlceland.
Geophysical Journal of the Royal Astronomical Society 8: 285300.
134
Bolt, B. A., C. Lomnitz and T. V. McEvilly. 1968. Seismological
evidence on the tectonics of central and northern California
and the Mendocino Escarpment. Bulletin of the Seismological
Society of America 58: 1725-1768.
Bostrom, R. C. 1967. Ocean-ridge system in northwest America.
Bulletin of the American Association of Petroleum Geologists
51: 1816-1832.
Bott, M. H. P. 1967. Solution of the linear inverse problem in
magnetic interpretation with application to oceanic magnetic
anomalies. Geophysical Journal 13: 313-323.
Browne, B. C. 1937. The measurement of gravity at sea. Monthly
Notices of the Royal Astronomical Society, Geophysics Supplement 4: 211-279.
Budinger, T. F. l967a. Cobb Seamount. Deep-Sea Research
14: 191-201.
Budinger, T. F. and B. J. Enbysk. 1967b. Later Tertiary data
from the East Pacific Rise. Journal of Geophysical Research
72: 2271-2274.
Bye rly, P. 1938. The earthquake of July 6, 1934, amplitude and
first motion. Bulletin of the Seismological Society of America
28: 1-14.
Carison, P. R.
1968. Marine geology of Astoria Submarine Canyon.
Ph. D. thesis. Corvallis, Oregon State University. 259 numb.
leaves.
Danes, Z. F., M. M. Bonno, E. Brau, W. D. Gilham, T. F.
Hoffman, D. Johansen, M. H. Jones, B. Malfait, J. Matsen
and G. 0. Teague. 1965. Geophysical investigations of the
southern Puget Sound area, Washington. Journal of Geophysical Research 70: 5573-5580.
Dehlinger, P. 1964. Reliability at sea of gimbal-suspended gravity
meters with 0. 7 critically damped accelerometers. Journal
of Geophysical Research 69: 5383-5394.
135
Dehlinger, P.., E. F. Chiburis and M. M. Collver. 1965. Local
travel time curves and their geologic implications for the
Pacific Northwest states. Bulletin of the Seismological
Society of America 55: 587-607.
Dehlinger, P., R. W. Couch and M. Gemperle. 1966a. Gravity
meter measurements corrected for vertical accelerations.
Journal of Geophysical Research 71: 6017-6023 (with correction in Journal of Geophysical Research 72: 4280).
Dehlinger, P., M. Gemperle, R. W. Couch and D. W. Hood.
1966b. Free-air gravity anomalies along the Inside Passage
of British Columbia and Alaska. Journal of Geophysical
Research 71: 6011-6015.
Dehlinger, P., R. W. Couch and M. Gemperle. 1967a. Gravity
and structure of the eastern part of the Mendocino escarpment. Journal of Geophysical Research 72: 1233-1247.
Dehlinger, P., R. W. Couch, W. A. Rinehart and M. Gemperle.
1967b. Free-air gravity anomaly map west of Oregon.
Salem, Oregon, Dept. of Geology and Mineral Industries.
1 sheet.
Dehlinger, P., R. W. Couch and M. Gemperle. 1968. Continental
and oceanic structure from the Oregon coast westward across
the Juan de Fuca Ridge. Canadian Journal of Earth Sciences
5: 1079-1090.
Dehlinger, P. 1969. Gravity and its relation to topography and
geology in the Pacific Ocean. (Submitted to the American
Geophysical Union for public ation)
Dietz, R. S. 1961. Continent and ocean basin evolution by
spreading of the sea floor. Nature 190: 854-857.
Duerkson, J. A. 1949. Pendulum gravity data in the United States.
Washington, D.C. 218 p. (U.S. Coast &Geodetic Survey.
Special Publication no. 244.)
Duncan, J. R., Jr.
1968. Late Pleistocene and postglacial sedimentation and stratigraphy of deep-sea environments off
Oregon. Ph. D. thesis. Corvallis, Oregon State University.
222 numb. leaves.
136
Duncan, J. R., W. J. Morgan, W. G. Melson, T. Simkin, H. Banks
and D. Gottfried. 1969. Juan de Fuca Ridge bathymetry:
independent evidence of sea-floor spreading. (Abstract)
Transactions of the American Geophysical Union 50: 185.
Dymond, J. R., N. D. Watkins and Y. R. Nayudu. 1968. Age of
the Cobb Seamount. Journal of Geophysical Research 73: 39 773979.
Ellis, R.
K. H. Tseng and W. G. Milne. 1968. A new model
for the crust in the vicinity of Vancouver Island. (Abstract)
Transactions of the American Geophysical Union 49: 711.
I'4.,
Emilia, D. A., J. W. Berg, Jr. and W. E. Bales.
1968a. Magnetic
anomalies off the northwest coast of the United States. Bulletin
of the Geological Society of America 79: 1053-1062.
l968b. Oceanic extension of coastal volcanics:
northwestern Oregon. The Ore Bin 30: 2 1-28.
Emilia, D. A. 1969. Numerical methods in the direct interpretation
of marine magnetic anomalies. Ph. D. thesis. Corvallis,
Oregon State University. (In preparation)
Engel, C. G. and A. E. J. Engel. 1963. Basalts dredged from the
northeastern Pacific Ocean. Science 140: 1321-1324.
Erickson, B. H. 1967. Geophysicist, USC&GS, Pacific Northwest
Laboratory. Personal communication. Seattle, Washington.
September 7.
Ewing, J. and M. Ewing. 1967. Sediment distribution on the midoce.n ridges with respect to spreading of the sea floor.
Science 156: 1590-1592.
Ewing, J., M. Ewing, T. Aitken and W. J. Ludwig. 1968. North
Pacific sediment layers measured by seismic profiling. In:
The crust and upper mantle of the Pacific area, ed. by L.
Knopoff, C. L. Drake and P. J. Hart. Washington, D. C.
p. 147-173. (American Geophysical Union. Geophysical Monograph no. 12.) (National Research Council. Publication no.
1687.)
Foster, T. D.
Heat-flow measurements in the northeast
Pacific and in the Bergin Sea. Journal of Geophysical Research
1962.
67: 2991-2993.
1 37
Gallagher, J. N. 1969. A method for determining the source
mechanism in small earthquakes with application to the Pacific
Northwest region. Ph. D. thesis. Corvallis, Oregon State
University. (In preparation)
Garland, G. 11 and J. G. Tanner.
Investigations of gravity
and isostacy in the southern Canadian Cordillera. Publications
of the Dominion Observatory (Ottawa) 19: 169-222.
1957.
Gibson, W. M. 1960. Submarine topography in the Gulf of Alaska.
Bulletin of the Geological Society of America 71: 1087-1108.
Grim, M. S., L. C. Bennett and R. A. Harman.
1968. Shallow
seismic profiling of the continental shelf off Gray's Harbor,
Washington. (Abstract) Transactions of the American Geophysical Union 49: 209.
Gutenberg, B. and C. F. Richter. 1941. Seismicity of the earth.
New York. 131 p. (Geological Society of America. Special
Paper no. 34.)
Hamilton, IE. L. 1967. Marine geology of abyssal plains in the Gulf
of Alaska. Journal of Geophysical Research 72: 4189-4213.
Hamilton, E. L. and H. W. Menard. 1968. Undistorted turbidites
on the Juan de Fuca Ridge. (Abstract) Transactions of the
American Geophysical Union 49: 208.
Harrison, J. C., G. L. Brown and F. N. Spiess.
1957. Gravity
measurements in the Northeastern Pacific Ocean.
of the American Geophysical Union 38: 835-840.
Transactions
Harrison, J. C. and W. C. Brisbin. 1959. Gravity anomalies off
the west coast of North America, I: Seamount Jasper. Bulletin
of the Geological Society of America 70: 929-934.
Harrison, 3. C. and C. E. Corbato.
1963. The Mt. Wilson calibration range, new geodetic gravity values in the western United
States and some unpublished submarine measurements.
Malibu, California, Hughes Research Laboratories. 9 numb.
leaves.
Harrison, J. C. and L. J. B. LaCoste.
1968. The performance of
LaCoste-Romberg shipboard gravity meters in 1963 and 1964.
Journal of Geophysical Research 73: 2163-2 174.
1 38
Hayford, J. F. and W. Bowie. 1912. The effect of topography and
isostatic compensation upon the intensity of gravity. Washington, D.C. 132 p. (U.S. Coast and Geodetic Survey. Special
Publication no. 10.)
Heirtzler, J. R., G. 0. Dickson, E. M. Herron, W. C. Pitman III
and X. LePichon. 1968. Marine magnetic anomalies, geomagnetic field reversals, and motions of the ocean floor and
continents. Journal of Geophysical Research 73: 2 119-2135.
Heiskanen, W. A. 1938. Catalogue of the isostatically reduced
gravity stations. Helsinki. 139 p. (International Association
of Geodesy. Isostatic Institute. Publication no. 5.
Heiskanen, W. A. and F. A. Vening Meinesz. 1958. The earth and
its gravity field. New York, McGraw-Hill. 470 p.
Hess, H. H. 1962. History of ocean basins. In: Petrologic
studies: a volume to honor A. F. Buddington. New York.
Geological Society of America, p. 599-62 0.
Seismic anisotropy of the upper most mantle
under oceans. Nature 203: 629-631.
1964.
1965. Mid-oceanic ridges and tectonics of the sea
floor. In: Submarine geology and geophysics, ed. by W. F.
Whittard and R. Bradshaw. Proceedings of the 17th symposium
of the Coiston Research Society, Bristol, England, 1965. London, Butterworths. p. 317-332.
Hill, D. P. and L. C. Pakiser.
Crustal structure between
the Nevada test site and Boise, Idaho, from seismic refraction
measurements in the earth beneath the continents. Washington,
D.C. p. 391-419. (American Geophysical Union. Geophysical
Monograph no. 10.) (National Research Council. Publication
no. 1467.)
1966.
Hodgson, J. H. and W. G. Mime. 1951. Direction and faulting in
certain earthquakes in the North Pacific. Bulletin of the
Seismological Society of America 41: 22 1-242.
Hodg son, J. H. and R. S. Storey. 1954. Direction of faulting in
some of the larger earthquakes of 1949. Bulletin of the Seismological Society of America 44: 57-84.
139
Hodgson, 3. H. 1957. Current status of fault-plane studies - a
summing up. In: The mechanics of faulting with special reference to the fault-plane work, ed. by J. H. Hodgson. Publications of the Dominion Observatory (Ottawa) 20: 413-418.
Hurley, R. J. 1960. The geomorphology of abyssal plains in the
northeast Pacific Ocean. La Jolla. 105 numb. leaves.
(Scripps Institution of Oceanography.
SIO Reference 60-7.)
Technical Report no.
Lsacks, B., 3. Oliver and L. R. Sykes. 1968. Seismology and the
new global tectonics. Journal of Geophysical Research 73:
5855-5900.
James, D. E. and J. S. Steinhart.
Structure beneath the
continents: A critical review of explosion studies, 1960-1965.
In: The earth beneath the continents, ed. by 3. 5. Steinhart
and T. 3. Smith. Washington, D.C. p. 293.333. (American
Geophysical Union. Geophysical Monograph no. 10.) (National
Research Council. Publication no. 1467.)
1966.
Kuim, L. D. 1967. Assistant Professor of Oceanography, Oregon
State University, Dept. of Oceanography. Personal communication. Corvallis, Oregon. February 4.
Kuim, L. D., K. F. Scheidegger, J. V. Byrne and J. J. Spigai.
1968. A preliminary investigation of the heavy mineral suites
of the coastal rivers and beaches of Oregon and northern
California. The Ore Bin 30: 165-179.
LaCoste, L. J. B. and J. C. Harrison. 1961. Some theoretical
considerations in the measurement of gravity at sea. Geophysical Journal of the Royal Astronomical Society 5: 89-103.
LaCoste, L. J. B. 1967. Measurement of gravity at sea and in the
air. Reviews of Geophysics 5: 477- 526.
Larson, R. L. and F. N. Spiess.
1969.
East Pacific Crest: a
near-bottom geophysical profile. Science 163: 68-71.
140
Leech, G. B. 1965. The Rocky Mountain Trench. In: The world
rift system. Report of a symposium, Ottawa, Sept. 1965.
Ottawa. p. 307-329. (Geological Survey of Canada. Paper
66-14.)
Le Pichon, X. and M. Taiwani. 1964. Gravity survey of a seamount
near 35°N, 46°W in the North Atlantic. Marine Geology 2: 262277.
Le Pichon, X. 1968. Sea-floor spreading and continental drift.
Journal of Geophysical Research 73: 3661.
MacDonald, G. J. F. 1963. Deep structure of continents. Reviews
of Geophysics 1: 587-665.
Mason, R. G. 1958. A magnetic survey off the west coast of the
United States between latitudes 32° and 36°N and longitudes
1210 and 128°W. Geophysical Journal of the Royal Astronomical Society 1: 320-329.
Maxwell, A. 1968. Continental drift and a dynamic earth.
Scientist 56: 35-51.
American
Gravity and structure of Cobb Seamount. Unpublished manuscript. Corvallis, Oregon State University, Dept.
of Oceanography.
McKay, A.
1966.
McManus, D. A. 1964. Major bathymetric features near the coast
of Oregon, Washington, and Vancouver Island. Northwest
Science 38: 65-82.
1965 Blanco Fracture Zone, northeast Pacific
Ocean. Marine Geology 3: 429-455.
1967a. Classification of submarine physiography
in the Gulf of Alaska. Northwest Science 41: 118-125.
l967b. Physiography of Cobb and Gorda Rises,
northeast Pacific Ocean. Bulletin of the Geological Society of
America 78: 527-546.
l967c. Associate Professor of Oceanography,
University of Washington, Dept. of Oceanography. Personal
communication. Seattle, Washington. March 23.
141
McManus, D. A. and N. McGary. 1967. Topography of an intermediate borderland, the continental terrace off Washington.
(Abstract) In: Program of the 1967 Annual Meetings of the
Geological Society of America. New Orleans. p. 145-146.
Menard, H. W. and R. S. Dietz. 1951. Submarine geology of the
Gulf of Alaska. Bulletin of the Geological Society of America
62: 1263-1285.
Menard, H. W. 1955a. Deep sea channels, topography and sedimentation. Bulletin of the American Association of Petroleum Geologists 39: 236-255.
Deformation of the northeastern Pacific
basin and west coast of North America. Bulletin of the Geological Society of America 66: 1149-1198.
1955b.
1960.
The East Pacific Rise. Science 132: 1737- 1746.
1964. Marine geology of the Pacific. New York,
McGraw-Hill. 271 p.
Menard, H. W. and T. Atwater. 1968. Changes in direction of sea
floor spreading. Nature 219: 463-467.
Mesecar, R. S. 1968. Oceanic vertical temperature measurements
across the water-sediment interface at selected stations west
of Oregon. Ph. D. thesis. Corvallis, Oregon State University.
99 numb. leaves.
Miller, A. H. 1929. Gravity in western Canada. Publications
of
tbe
Dominion Observatory (Ottawa) 8: 241-330.
Miller, A. H. and W. G. Hughson. 1936. Gravity and isostasy in
Canada. Publications of the Dominion Observatory (Ottawa)
11: 81-134.
Milne, A. R. 1964. Two seismic refraction measurements: North
Pacific Ocean Basin and Dixon Entrance. Bulletin of the Seisrnological Society of America 54: 41-50.
Milne, W. G. and W. R. H. White. 1960. A seismic survey in the
vicinity of Vancouver Island, British Columbia. Publications
of the Dominion Observatory (Ottawa) 24: 145-154.
142
Mime, W. G. 1967. Earthquake epicenters and strain release in
Canada. Canadian Journal of Earth Sciences 4: 797-814.
Morgan, W. J. 1968a. Pliocene reconstruction of the Juan de Fuca
Ridge. (Abstract) Transactions of the American Geophysical
Union 49: 327.
1968b.
Rises, trenches, great faults, and crustal
blocks. Journal of Geophysical Research 73: 1959-1982.
Murray, H. W. 1941. Submarine mountains in the Gulf of Alaska.
Bulletin of the Geological Society of America 52: 333-362.
1961. Physical properties of marine
sediments. Lamont Geological Observatory Technical Report
2 (CU-3-61, June). New York.
Nafe, J. E. and C. L. Drake.
1967. Professor of Geology, Columbia University,
Dept. of Geology. Personal communication. New York.
November 3.
Nafe, J. E.
Nayudu, Y. R. 1962. A new hypothesis for origin of guyots and
seamount terraces. In: The crust of the Pacific Basin, ed.
by G. A. MacDonald and H. Kuno. Washington, D. C.
p. 171-180. (American Geophysical Union. Geophysical
Monograph no. 6.) (National Research Council. Publication
no. 1035.)
Nayudu, Y. R., C. Chastain and J. L. Woodruff. 1968. Morphology
and petrology of two seamounts on Eickelberg Ridge. In:
(Abstract) Abstracts for 1967. Boulder, Colorado. p. 343.
(Geological Society of America. Special Paper 115.)
Nelson, C. H. 1968. Marine geology of Astoria Deep-sea Fan.
Ph. D. thesis. Corvallis, Oregon State University. 287 numb.
leaves.
Nelson, C. H., L. D. Kulm, P. R. Carlson and J. R. Duncan. 1968..
Mazama ash in the northeastern Pacific. Science 161: 47-49.
Neuman, F. 1959. Crustal structure in the Puget Sound area.
Travaux Scientifique de la Union Geodesique et Geophysique
Internationale, sec. Seismologie, ser. A, fasc. 20, p. 153-l67.
143
Ozima, Minoru, Mituko Ozitna and Ichiro Kaneoka. 1968. Potassium-.
argon ages and magnetic properties of some dredged submarine
basalts and their geophysical implications. Journal of Geophysical Research 73: 711-723.
Pavoni, N. 1966. Tectonic interpretation of the magnetic anomalies
southeast of Vancouver Island. Pure and Applied Geophysics
63: 172-178.
Peter, G. and R. Lattimore. 1969. Magnetic structure of the Juan
de Fuca-Gorda ridge area. Journal of Geophysical Research
74: 586-593.
Pitman, W. C., III and D. E. Hayes. 1968. Sea floor spreading in
the Gulf of Alaska. Journal of Geophysical Research 73: 65716580.
Pitman, W. C., III, E. M. Herron and 3. R. Heirtzler.
1968.
Magnetic anomalies in the Pacific and sea floor spreading.
Journal of Geophysical Research 73: 2069-2085.
Raff, A. D. and R. G. Mason. 1961. Magnetic survey off the west
coast of North America, 40°N latitude to 52°N latitude. Bulletin
of the Geological Society of America 72: 1267- 1270.
Rasmussen, N. 1967. Washington State earthquakes 1840 through
1956. Bulletin of the Seismological Society of America 57:
46 3-476.
Richards, T. C. and D. J. Walker. 1959. Measurement of the
thickness of the earth's crust in the Albertan plains of western
Canada. Geophysics 24: 262-284.
Rinehart, W., R. C. Bowen and E. F. Chiburis.
1964. Airport
gravity base station network in Oregon. The Ore Bin 26: 37-56.
Royse, C. F., Jr. 1967. Mazama ash from the continental slope off
Washington. Northwest Science 41: 103-109.
Sauers, J.
Bouguer gravity anomaly and topographic map of
Washington. Unpublished research. Seattle, University of
Washington. 1 sheet.
1966.
Shepard, F. P.
1938. Submarine canyons: distribution and longi-
tudinal profiles. Geographical Review 28: 439-451.
144
Shor, G. G., Jr.
1962. Seismic refraction studies off the Coast of
Alaska: 1956-1957. Bulletin of the Seismological Society of
America 52: 37-57.
Shor, G. G., Jr., P. Dehlinger, H. K. Kirk and W. S. French.
Seismic refraction studies off Oregon and northern California. Journal of Geophysical Research 73: 2175-2194.
1968.
Shurbet, G. L. and J. L. Worzel.
1955.
Gravity anomalies associated
with seamounts. Bulletin of the Geological Society of America
66: 777-782.
Snavely, P. D., Jr. and H. C. Wagner.
1964. Geologic sketch of
northwestern Oregon. Bulletin of the U. S. Geological Survey
1181-M: Ml-M17.
St. Amand, P. 1957. Geological and geophysical synthesis of the
tectonics of portions of British Columbia, the Yukon Territory
and Alaska. Bulletin of the Geological Society of America
68: 1343-1370.
Stacy, R. 1967. Geophysicist, Dominion Observatory of Canada,
Gravity Division. Personal communication. Ottawa, Canada.
April 14.
Stauder, W. 1967. The present status of fault-plane solutions from
P and S waves and the relation of these solutions to tectonic
stress fields. In: Proceedings of the Vesiac conference on the
current status and future prognosis for understanding the source
mechanism of shallow seismic events in the 3 to 5 magnitude
range. Vesiac report. Ann Arbor, University of Michigan,
Willow Run Laboratories. p. 3-24.
Stonely, R. 1967. The structural development of the Gulf of Alaska,
sedimentary province in southern Alaska. Journal of the Geological Society of London 123: 25-57.
Gravity study of crustal structure in western
Washington. In: Short papers in the geologic and hydrologic
sciences. Washington, D.C. p. 273-2 76. (U.S. Geological
Survey. Professional Paper 424-C, article 48.)
Stuart, D. J.
1961.
145
Sutherland-Brown, A. 1966. Tectonic history of the insular belt of
British Columbia. In: A symposium on the tecktonic history and
mineral deposits of the western Cordillera. Vancouver, 1964.
Montreal. p. 83-87. (Canadian Institute of Mining and MetaUurg
Spec ial Volume no. 8.)
Taiwani, M., G. H. Sutton and 3. L. Worzel. 1959a. A crustal
section across the Puerto Rico Trench. Journal of Geophysical
Research 64: 1545-1555.
Taiwani, M., J. L. Worzel and M. Landisrnan. 1959b. Rapid gravity
computations for two-dimensional bodies with application to the
Mendocino submarine fracture zone. Journal of Geophysical
Research 64: 49-59.
Taiwani, M., X. LePichon and J. R. Heirtzler. 1965. East Pacific
Rise: The magnetic pattern and fracture zones. Science 150:
1109-1115.
Tanner, J. G.
1966.
Gravity. Canadian Geophysical Bulletin 19:
4-17.
Tatel, H. E. and M. A. Tuve. 1955. Seismic exploration of a continental crust. In: Crust of the earth, ed. by A. Poldervaart.
New York. p. 35-50. (Geological Society of America. Special
Paper 62.)
Thiruvathukal, J. V. 1968. Regional gravity of Oregon. Ph. D.
thesis. Corvallis, Oregon State University. 92 numb. leaves.
Thiruvathukal, J. V., J. W. Berg, Jr. and D. F. Heinrichs.
1969.
Regional gravity of Oregon. (Submitted for publication to the
Bulletin of the Geological Society of America)
Thompson, G. A. and M. Talwani. 1964. Geology of the crust and
mantle, western United States. Science 146: 1539-1549.
Tobin, D. G. and L. R. Sykes. 1968. Seismicity and tectonics of
the northeastern Pacific Ocean. Journal of Geophysical
Research 73: 3821-3846.
U.S. Hydrographic Office. 1952. Chart no. 1410 N. Washington,
D.C.
146
Vacquier, V., A. D. Raff and R. E. Warren. 1961. Horizontal
displacements in the floor of the northeastern Pacific Ocean.
Bulletin of the Geological Society of America 72: 1251-1258.
Vening Meinesz, F. A. 1929. Theory and practice of pendulum
observations at sea. Delft, Netherlands Geodetic Commission.
Gravity anomaly maps of the Indonesian ArchiVol. 2.
Report of the gravity expedition in the Atlantic of 1932 and the
interpretation of the results. Deift, Netherlands Geodetic
1934.
pelago. In: Gravity expeditions at sea, 1923-1930.
Commission. 208 p.,
Vine, F. J. and D. H. Matthews.
1963.
Magnetic anomalies over
ocean ridges. Nature 199: 947-949.
Vine, F. J. and J. T. Wilson.
1965. Magnetic anomalies over a
young oceanic ridge off Vancouver Island. Science 150: 485-
489.
Vine, F. J.
Spreading of the ocean floor: new evidence.
Science 154: 1405-1415.
1966.
1968. Magnetic anomalies associated with mid-ocean
ridges. In: The history of the earth's crust, ed. by R. A.
Phinney. Princeton, Princeton University. p. 73-89.
Von Herzen, R. P. and S. Uyeda. 1963. Heat flow through the
eastern Pacific Ocean floor. Journal of Geophysical Research
68: 4219-4250.
Von Herzen, R. P. 1964. Ocean-floor heat-flow measurements west
of the United States and Baja California. Marine Geology 1:
225-239.
Von Huene, R., D. W. Scholl and G. G. Shore, Jr. 1969. Hypotheses
of thrusting at continental margins in light of recent marine
geophysical observations. (Abstract) Transactions of the
American Geophysical Union 49: 757.
Walcott, R. I. 1967. The Bouguer anomaly map of southwestern
British Columbia. Vancouver. 74 p. (University of British
Columbia, Institute of Earth Sciences. Scientific Report no. l5)
147
Waters, A. C. 1955. Volcanic rocks and the tectonic cycle. In:
Crust of the earth, ed. by A. Poldervaart. New York. p. 703722. (Geological Society of America, Special Paper no. 62.)
White, W. R. H. and J. C. Savage. 1965. A seismic refraction and
gravity study of the earth's crust in British Columbia. Bulletin
of the Seismological Society of America 55: 463-486.
Wilson, J. T. 1965a. A new class of faults and their bearing on continental drift. Nature 207: 343- 347.
Transform faults, oceanic ridges and magnetic
anomalies southwest of Vancouver Island. Science 150: 482-485.
196 Sb.
Woollard, G. P. 1951. A gravity reconnaissance of the island of chu.
Transactions of the American Geophysical Union 32: 558-568.
1962. The relation of gravity anomalies to surface
elevation, crustal structure and geology. Madison. (University
of Wisconsin. Geophysical and Polar Research Center. Research
Report no. 62-9.)
Woollard, G. P. and W. E. Strange. 1962. Gravity anomalies and
crust of the earth in the Pacific Basin. In: The crust of the
Pacific Basin, ed. by G. A. MacDonald and H. Kuno. Washington, D. C. p. 60-80. (American Geophysical Union. Geophysical Monograph no. 6.) (National Research Council. Publication
no. 1035.)
Woollard, G. P. and J. C. Rose.
1963. International gravity measurements. Tulsa. Society of Exploration Geophysicists. 518 p.
Worzel, J. L.
1965. Pendulum gravity measurements at sea, 19361959. New York, John Wiley. 422 p.
Wylie, C. R., Jr.
1951. Advanced engineering mathematics. New
York, McGraw-Hill. 640 p.
APPENDICES
148
APPENDIX 1
GRAVITY DATA REDUCTION TO FREE-AIR ANOMALY
The gravity measurements used to construct the free-air
anomaly maps of the region west of Washington and British Columbia
shown in Figures 6 and 7 were made with LaCoste-Romberg gimbal
suspended surface ship gravity meter S-9 from 1963 through 1967.
Positioning, a critical factor in obtaining gravity measurements
at sea, was determined principally by Loran A. Radar fixes were
obtained where possible and sunlines, star fixes and radio positions
were obtained on occasion as supplementary data. Dead reckoning
was used where all other methods were obviated. All positions were
replotted in the laboratory from the original data on a working track-
line chart. The times at which the ship turned were obtained from
the analogue gravity records, and changes in engine rotation rates,
stearage and sea condition were obtained from the ships log. All
data pertinent to the ships navigation were added to the track line
chart, and lines of best fit were sequentially constructed from one
point of change in velocity or heading to another. After a preliminary
reduction of the gravity measurements the positions of the track line
end points were re-evaluated and errors corrected or minor modifications made. Positional accuracy and consequently gravity
149
measurements improved with experience.
Measurements were obtained aboard the BERTHA ANN in July,
1963; the USC&GS SURVEYOR in February, 1964; the USCGC
YOCONA in August, 1964, and May, 1965; and the R/V YAQUINA
in June, 1965, July, 1966, April, 1967, and September, 1967. When
using LaCoste-Romberg meter S-9, errors in gravity measurement
occur due to a nonlinear response to vertical acceleration
(Dehlinger etal.,, 1966a LaCoste, 1967). Corrections to the vertical
errors are obtained using the relation
Ve =
a(R-R)A2
where the vertical acceleration is given by the relation
A
B
vT
k(B)
peak to peak amplitude of instantaneous beam motion
logged on analogue recorder,
T = period of beam motion in seconds,
R
average beam position, and
R, a, and k
laboratory determined constants.
The constants as determined by LaCoste-Romberg for the era of
operation were:
= 40, a
2.3 X 10, k = 23
21 Feb 61
R
13Nov64
R95,a2.0X104,k23
1 50
4 Dec 64
40, a
R
=
1.1 X
k
2. 3 X 10, k
R°
=
50.4, a
66
R
=
-160, a
=
3 Mar 66
R
54, a
1.8 X 10, k
17 Mar
24 Feb
65
=
=
23
=
22
17.8 X 10, k =
=
22
29
Because of the change in the parameters with time (e. g. Mar
65 to Feb 66) in an unknown manner empirical equations were deter-
mined by statistical analysis from data obtained over the Newport
marine gravity range. The equations used to determine the vertical
correction for the surveys are:
2.94X105B2F2(95-R);
BERTA ANN
July, 1963
USC&GS SURVEYOR
February, 1964 2. 94X105B2F2(95-R);
USCGC YOCONA
August, 1964
2. 94X105B2F2(95-R);
USCGC YOCONA
May, 1965
3.09Xl05B2F2(43.9R);
R/V YAQUINA
June, 1965)
R/V YAQUINA
July, 1966
NONE
R/V YAQUINA
April, 1967
5. 6X105B2F2 (1 36. 4-R)
R/V YAQUINA
September, 1967 NONE
LBB(3. 79X102F+0.
where the frequency of instanteous beam motion F
724)
1/T.
Modification of the amplitude B in the correctionequ.ation usedin
1965 compensates for the attenuation of the higher frequencies by an
RC filter in the phase comparator circuit of the gravity meter. The
frequency response of the meter was improved in 1966.
An addition of an automatic reader to the gravity meter in 1965
1 51
allowed a higher data sampling rate but extended the response.time
of the gravity meter. The time constants of meter 5-9 varied from
approximately 1 minute to nearly 5 minutes. Filter or overaging
time constants can cause phase shift and variable attenuation of
the higher frequency spectral components of the observed gravity
anomalies. A time offset of 4 minutes was used to compensate for
the meter time delay when operating with the automatic reader.
Hanning smoothing (e.g. Blackman and Tukey, 1958, p. 15) was used
on automatic reader data sampled at four minute intervals. A
mathematical analysis showed that neither the meter response and
transfer function nor the hanning smoothing significantly distorted
the observed gravity field.
A computer program to reduce the gravity measurements to
free-air anomalies was written by Mr. M. Gemperle in FORTRAN
IV for use on the CDC 3300. The program parameters are:
152
INPUTS
1.
Calibration curve - converts gravity meter counter units to
milligals.
Input is 50 data points - at even l00's of gravity meter counter
units
2.
F OR MAT (8 F 10.0)
Base Station Data, etc.
a) Base meter reading (in counter dial units)
b) Base gravity value
c) Year and month of cruise
d) Effective date of results
e) Four. letter code for ship
f) Four. letter code for base station
g) Code for punching output cards (blank no cards; 1 card)'
h) Meter drift
i) Code for interpolation subroutine to find location of points
on track lines with even lO's free air anomaly (blank - no
interpolation; 1-interpolation)
j) Time constant for averaging circuit
FORMAT (2F10.1, 4(lx,A4), 3X, 12, F5.l, IS, F5.1, F5.2)
3.
Navigation Cards
a) Time and lat and long for beginning of line
b) Time, lat, and long for end of line.
c) Profile number
FORMAT (2 (3F2.0, F3.0, F5.1, F4.0, FS.1, 2X), AS)
4.
Gravity Station Cards
a) Time of reading
b) meter reading (in meter units)
c) depth in fathoms
d) freq.
e) amp
f) average reading line
g) Browne correction
h) quality (poor or blank)
i) Test for last card in profile (1-last card; blank-not last card)
j) Test for first card in profile (1-firstcard; blank-notfirst card)
FORMAT (3F2.0, F9.1, SF5.0, 1X, A4, Ii, 9X, 15)
1 53
C) I TT PT IT S
1.
Base station information
a) Date effective
b) Zero gravity (gravity corresponding to zero meter units)
c) base meter reading
d) base gravity
e) drift
f) Filter time constant
FORMAT (18H1RESULTS EFFECTIVE, ZX, A4, 3X, 13HZERO
GRAVITY=, FlO. 1, Z1H BASE METER READING=, F7. 1, 15H
BASE GRAVITY, FlO. 1, 8H DRIFT, F5. 1, 5X, 4HRC,
F6. 2)
2.
Navigation Data
a) Profile number
b) Speed
c) Course
d) Time, lat, and long of first point
e) Time, lat, and long of end point
FORMAT (12HOPROFILE NO., A5, 2X, 6HSPEED, F5. 1, 2X,
7HCOURSE=, F5.O, 2X, 7HPOfl\TT 1, 2X, 312, F5.O, F6. 1,
F5.O, F6.1, 2X, 7HPOINT 2, ZX, 312, F5.O, F6.l,F5.O,
F6. 1)
3.
Gravity Station Data
a) Year and month
b) DTG - Day, hour, mm
c) Depth in meters
d)
Latitude and longitude of station
e) Gravity meter reading (meter units)
f) Eotvos Correction
g) Normal (theoretical) Gravity
h) Meter reading in milligals
i) Gravity corrected for Eotvos effect
j) Gravity corrected for Eotvos and vertical nonlinearity
k) Browne correction
1) Vertical correction
m) Free air anomaly [j) - g)J
n) Quality (poor or blank)
o) Averaged (filtered) free air anomaly
[either 1/3 (FAA1...j+FAAm+FAA1+1) or 1/4 (FAA.j+
2FAA+FAAm+i)]
Outputs 1, 2, and 3 are standard outputs to high speed line printer.
Optional outputs are as follows:
1 54
4.
Even l0's free air anomaly output on line printer gives each
even multiple of 10 mgal free air anomaly interpolated between
measurement stations.
a) Interpolated free air anomaly
b) Latitude and longitude
FORMAT (1H, 20X, Fl0.1, F5.0, F6.l, Fl0.0, F6.l)
5.
Punched card #1
Col
1-10
11-18
19-28
29-34
35-43
44-52
53-57
Year, month, day, hour, mm
Latitude
Longitude
Depth (meters)
Observed gravity
Normal gravity
Free air anomaly
(58-62 Simple Bouguer
never ) 63-68 Terrain corrected Bouguer
used
69-73 Isostatic anomaly
¼75-76 Reliability
77-80 Date effective
1
6.
Punched Card #2
Col 1-10 Year, month, day, hour, mm
11-16 Latitude
17-23 Longitude
24-2 7 Ship (4 letter code)
28-31
32-37
38-42
43-48
49-51
52-54
55-57
58-62
63-67
68-74
75-77
78-80
Base station (4 letter code)
Base gravity value (minus 900 gals)
Base meter reading
Zero gravity (minus 900 gals)
Drift
Course
Speed
Eotvos correction
Meter reading
Eotv.os corrected gravity
Browne correction
Vertical correction
The program listing is:
PPAN1 MATNX
155
"IN A,KK
nT1Fr'lsI'II Y(5fl) .A(.l) '.j''N1.fl ..'JAI
?-
Fc'1Ar(A4)
17fl FPAAF (2
.U,F3.Q,F5.1
r3
P4AI(1pH0pqFTIF
,)i() A'
7'C1'T j,2X,T?,F.,F.1
I
F. 1
41
1)
A7 (7X,3F?.fl,F7.1,c.(.! x,A4.T I
I
7 FAT (1H ,ix,A4.1X,T)
1
c' !'Y1L41
4 10.1 ,FS .0' FQ 1, F9
1
FQAT(1H ,lx,A4,1x,3J?
,'.().fc.1 ,
I. i
y , A4
4FIo.1,F5o,F.I,Fq.1,7.A4,F7.1,1
I
F'AAT(,F1r1.1 ,1X,A4,1X,A4,I,A4,1X,4,3*,1,',F., 7'
c0 F1AT(1RHIFSIJITS FFFCTTV'X.A4,,1H/EC
I ?1i
3ASFT 'FtFP kFaDINr,F7.1 ,16H sc
H
2
F1=,fS.1,,X,4I-IPC ,%7)
1'4 F'A1(241TrAI ANMCL'Y AVEAfZ
,V1,.1/2 iT"(p(
1 I,
fl
,FO)
1MS
"r
'O FO'lAT(F1fl.0)
Aflt1'7O) Y
Fr(1,,00) IPC
F V TX y
= cii
T F 3 1'1
M = j = F T 'I
S = 0.
?7
PA1,1O/IIF
Ir0 PFAD(l'g)
JF (MP)400,40l ,40()
TAVF=SU//
4r
lITF(3,1fl4)T\',7
STP
4rfl
FXP(..T/C)
IMp*00149.
fl=T .49
3 PFAr)(1,1?'))TA1 ,TA?,TA.11 ,,2,A?
ki C1,L3?,P
1
TF(TAI )403,100.403
43 A[ ATAL1*60.+Ai2
Al N(AI 1*60. .AL2
RI ATkL1*60,+H
LN(3i '60.PLC?
.TA2.TA3ThO.
TRTF1*,4. +Y,T3/(0.
V!AT(ALAT4LAT)*2.9qF_(6/2.
Afl
IF (ALAIRLAT) 1(1,19,10
19 IF (ALCHLN(1 11., 12,13
11 C15TF
r,
r
c()
r,129)
11 C=r.5*PTF
10 TFTA=ATAN(DFPA/(ALAT, T)(
IF (THF1P)2l ,3,23
IF (ALAi_PLAfl?4,?9O,?,
'4 CARS(THFTA).2.O*°IF
'I
f,
ril00
CARS(TkETA)+PTF
(3y30O
'3 TFiALAtRLAfl?7,fl,2
'7 (=T-WT
ri
'i r?,T-1 ,Ti',T,4I ' '.
TscT'
TO 3r
(.
156
AS(FiFT4).PTF
C)
r,Tr3OO
DT=iV1c(DFPAk)
?01
r,CT3O1
3o
.1
pFFn=uS1/lTk-TA)
1 1 41 = 1 A 1
I I 471
1141=141
TI
=1
TTP2T7
F (=C * A 0
wPTTF(3jl)l)PN,cPEFr),cflFr,TTA1 ,tT",1TA1.AL1 ,AL,,ALrl,AL)',
TT11 , LIR?, IT1.4L) .L?'q1C
I
r?
2
I
( , AMAr
1X1X1*?4. +1X2.1X3/60.
I TX'l =TX
T TX ? TX?
ITXITX1
TXE 1=XLT/60.
XL 1 = I XLI
XL2XLAT_XL1*,fl.
I XL? = XL? * 10.
X'1X CNG/60.
XL
= I X Cl
XLC2=XLrrs4t-XLC1*6O. +0.Oc
I XL 0? XL C? * tO.
TF5CALF.E:Q.0. 1000,1001
1001 FASTCSTN(C)*SPFED
P4T=XLA1*2.9OMR2F04
P.4I72.*PHT
FOTCR7.5CCS (P-4T ) *EASTC
1F(KK.EQ.l)SO,l
IF(TX-IyY.LT.O.11#,7171,lr.
TF(ARS(rV).LI.O.1)7(,,77
77 Fv=(ECTCR_ETVCS)*EXPTA_TX)*./r,C)
qC)
1
FCTCREoTCP_EV
76 TF(KTFST.EQ,1)7,84
7S FTVCSETCP
TX
TA
4 T4FCG=978O49.*(J.,,OO524*(cT,(r1))**?_c.9E_()c*(STN(1I4T,),**?)
J=rALE.O,O14q
PJ,49
CZECc,,Y(J),(SCALE*0.fl1_R)*tvtj+1)_V(.J))
C C P c, PC R , P + ECT CP
VEPT-EP(FRFQ,AMP,P)
IF ( ik ,E.r,l) 71, 70
70 VEPT=VT_CVERT_VEPT2)*FxP((TXxX_TX)*M)./QC)
71 VT2VcRT
TXXXTX
IF ARC Wt,GT .400, CR, ARS ( VFRT) . tT.1S.} 72 7
72 ItIJALzIDCCR
73 TRt6RgCfPC,R+VPT
FANOM:TPUGR-THFCG
S UM= SUM F A NCM
77+1.
flPgDP*1 ,R2AP
1000 CALL VELCCITV(OFP,XLAT)
Ay4(rjFT5.o*AMA(
IF fl(K.E0.fl700,701
7r0 KKK=KK
1=1
157
Jjj=o
7fl7 K(IIAL(IpIJ4L
YM(T)YOMC
JtJP'iK(1'1 )ITX1
JIJNI<CI.7)ITX2
JUNK (I 'U TTX3
ACT,1) rnFpTH
A CT ,?)XL1
A(I,3)XL2
A(T,4)XLCI
-
A(T,5)aYLC2
..-
------
TF(SCALF.E(3.O.) 1001.1004
1003 ACT ,6)A (I,7):A(I,g)=A(r,q)=A(T .10)=A(I.11 )A(1;-12)*o.
A(T,13)AC1,14)=0,
IC 1005
G
0n4 A(T,$)ccALE:
A(T,7)FCTCR
A CT ,R) IHECG
BGR
AC 1,9)
ACT ,!0) .CCRGR
ACT,11)TRUGP
-
A(1,13)2VERT
ACT, 14)FANCM
...
IOnS ACT,15)XLAT
--
-
------
.
A CT ,18),,TX
A(I,16)rXLCPIG
ACT, i9 .AMAGNET
-----
---- --
GCIC(800,60,$10)I
r,0 WRIT
-
-
3,7)yMCl,cJNKC1,K,K=l.3,A(1,J) ,J1,14) ,KOUAL(1) ,A(i,IQ)
GCIC0
-- ...... -- ............. __
7n1 IF(KKK.FQ.1)70?,704
7n2 KKKO
I
GCTC707
74 1z3
GT707
BiO TF(A(3.1R)-A(2,IR) .GI.O.fl7.Cp.A(?,1-A(1,18).Gr,fl.07)900,Q01
9n0 WRIT 3,7)vMC2,cJuNK2,,(,K1,3,A2,J.j1,j L,YC.4tI9.1
JJJ=0
(01C902
C WFIGHTEJ) AVFPAGE
9o1 A(?,17)u(A(1,j4).,*A(2,4).AC3,1.4))/4. _____
IF (15.E0,O)452,453
__________ ___________________
453 KKaJJJJJJ.1
CALLTNTFR
1
...?
850
,A(2.q)
___________
____
WRITC3,7)YM(3),CJUNK(3,K),KI,3),(A(3,J),J1,l4),K0UAL(3),A(.iQ)
851 oCR20L1.19
____ ______
DR20Mal,2
80 A(M,L)aA(M+1,L)
_____________
IF(K.TEST.EQ.1)8SO.851
DCR21NI .2
8I JUNK(N.T)JtJNKCW.1,!)
00R22KZ1 ,2
Kr3UAL(K) KQ(JAL (I(,1)
822 YM(K)sVp4(X.1)
A° IF(KTEST.EQJ)9,2
.
FtJPJCTTCI ER(FRFt,AMP,
C
t5
)
vETT(AL EopCR AS ADJUSTED 7/S/A7
CC.6OE-Q5
.....-.
PCC*AMP*AMP*FPEQ*FRF*(13.4.
RETURN
F Mn
-
--- -.-
.___ ..
159
SLJRCIJTTNE INTER
CMMCN A,KK
OTMENSIeNA(3,19)
1Or0 FPMAT(1H ,2OX,F10.1,F5.O,Ff,.,F1fl,o,FA.1)
FA!\IOM=Ac2,17)
XL A1*A C;, 15)
XLNG5A(2,j6)
---
--
---- ______
IF (l(PC.E0,1)300,301
3r'l IFA=FANM/10.
JFA*pA2,1O,
K=IABSyFA-JFA)
IF (ARS(FANOM-F4?) .GT.ABS (FAN3M.FA?) ) 30P,303
3o
(K.1
3n3 IF(K.EQ.0.3R.K.r,T.10)300,04
3n4 nC2001a1 ,K
tFCFA2.T.FANCM) 310,305
35 TFCFA2.LT.O,)3Q,3O7
3rF, A?JM=(JFA,j-1).1O
r,010315
____
3o7 APJHCJFA,)*1
G3T0315
310 TF(FA2.IT.0.)311,312
3i1 ANOMJFA-fle10
(CT315
32 AJOM(JFA-I+1)*1Q
--------------------
3iS OELTAsARS( (ANCM-FA2RFANM-FA,fl
CLAr.XLAT2+(XLAT-XLAT2)*DELTA
CL3N(IXLCN2+ (XLCNG-XLON2, *DELT4
ILAT!CL&T/60,
DLAT*ILAT
CMIN!CLAT-I)LAT!AQ,
I LNG.CLONG/6O
PjGRIJCNG
CLMI WsC,.CNG-DLCNG60,
-
2ñ0 WRTTEC3i1000)ANCMLAT,CMTN,flLCNr,.CLMtN
3o0 F42.FAN,M
XLAT2.XLAT
xXLN
RTUPN___________
-
----
P
StIRR3)1TNE vECITY(DEP,xIAT
r SrUNr) VFLOCyTY CORRFCTICNS FP DFP1'4S
r rCRPECIICNc FOR NORTH
TF(PEP.i T.IOO.)1.2
?
1
RFIIJRN
F Nfl
EAST PAC!f1._..
161
APPENDIX 2
SUPPLEMENTARY GRAVITY DATA
The bathymetry of region of the northeast Pacific Ocean west
of Washington and British Columbia exhibits large topographic con-
trasts between physiographic provinces similar to those observable
in the diverse geologic structures of the adjoining continental coastal
regions. The greatest problem in studying this region gravimetrically
was obtaining an areal density of measurements sufficient to allow
detection and analysis of the major structures of the region. Surface
ship gravity measurements along the track lines indicated in Figure 5
provided approximately 15, 400 data points. However, these points
are concentrated along track lines leaving large unsurveyed areas
which could contain major structural features. Additional valuable
data were obtained from the literature and from personal files.
Harrison etal. (l957) reported 10 pendulum measurements
obtained in a submarine west of the Columbia River and seven
measurements southwest of Vancouver Island.
He estimated an rms
uncertainty of ± 3.8 mgal for these measurements. Worzel (1965)
has reported the values of 28 pendulum measurements made in submarines in the area west of Washington and British Columbia. He
estimated a rms uncertainty of ± 3. 6 mgal for these measurements.
162
These predominantly deep ocean pendulum measurements provided
data between track lines and also an independent check of the surface
ship gravity meter measurements. The location of the pendulum
stations of Harrisonetal. (1957) and Worzel (1965) are shown on
Figure 5.
Approximately 280 underwater gravimeter measurements were
made available by the gravity Division of the Dominion Observatory
of Canada (Stacy, 1967 personal communication). No estimation of
the accuracy of these measurements has been reported; however
these underwater measurements are expected to be three to five
times more accurate than the available surface-ship data. The under-
water meter measurements are spaced at 10-15 km intervals in portions of Dixon Entrance, Hecate Strait and Queen Charlotte Sound.
The areas in which these measurements were made are indicated in
Figure 5 by hatchure lirie.
The U. S. Coast and Geodetic Survey check the performance of
their surface-ship gravity meter at sea by retracing a previously
measured track line called a signature line." The USC&GS
(Ericson, 1967 personal communication) made available approximately
90 gravity measurements along a signature line off Cape Flattery,
Washington.
Considerable assistance in extending gravity contours shoreward
was obtained by using available land gravity measurements made along
163
shore lines. Land gravity data in the Queen Charlotte Islands and
along some of the fjords of British Columbia was made available by
the Gravity Division of the Dominion Observatory of Canada (Stacy,
1967 personal communication). A Bouguer gravity map of
southern British Columbia by Walcott (1967) was of considerable help
in defining the gravity anomalies along the continental shelf off
Vancouver Is land. The sea values west of Washington were joined
to the contours of an unpublished Bouguer anomaly map by Sauers
(1966).
The rms uncertainties of land gravity measurements are
near ± 0. 3 mgal; approximately an order of magnitude better than sea
gravity measurements.
In contouring the reduced measurements to form Figures 6 and
7 two considerations were applied: 1) the gravity field is a potential
field and is therefore everywhere smooth and continuous, 2) the high
density contrast between the water and the material of the ocean bottom cause gravity gradients to follow bathymetric gradients. The
first consideration indicates that when measurements are made at a
distance from the anomaly producing source (i. e. in deep water)
gravity gradients are less. This suggests that both limited smoothing
and sympathetic contouring of scattered data generally result in the
best graphic approximation of the gravity field. The second con-
sideration suggests that in continuing a gravity anomaly contour from
a point defined by measurements to a second similar point, the most
164
likely path follows the isobath or isobathic trend between the two
points. This is not always possible in detail but generally the gravity
anomaly trends may be considered coincident with bathymetric trends.
In some areas, however, bathymetric contours are defined by insufficient data and both the gravity and bathymetry outline only the
largest of features.
In certain areas, although the bathymetry is reasonably well
known, the gravity field is unknown and cannot reasonably be
extrapolated from available data. Most notable are areas containing
seamounts which have closed contours. Ships tracks usually do not
pass over searnount peaks, and therefore a number of closed contours
remain which are not tied to any gravity measurement.
To help define gravity contours about seamounts the gravity
anomaly value over the seamount summit was calculated for 44 seamounts in the region mapped in Figures 6 and 7. Seamounts for
which calculations were made were selected on the basis of size,
symmetry and distance from areas in which the free-air gravity
anomaly about the base of the seamount would be greatly different
from zero.
The equation, from which the expected anomaly may be deter-
mined, is based on the assumption that seamounts may be approxi-
mated by a cone structure. The dimensions of the cone are indicated
in Figure Z3 and the derivation of the equation is as follows:
165
x
Figure 23. Diagram illustrating equation parameters for
computing gravity over a cone structure.
For the solid cone in Figure 23 the vertical acceleration of gravity
at point o is given by the equation
Zir R D+h
zr
dzdrdO
S (r + z2 3/2
=
(1)
ooD
where
D
apex depth
h
cone height
a
radius of cone base
=
gravitational constant
= cone density
p
and
R=
(z-D)
Integrating with respect to
e
yields
R D+h
g
=
z
2 rr
zr
SS (r2 +z 2 3/2
oD
dz dr
(2)
166
Integrating with respect to r
yields
D+h
hS
=
a
z
h
[ {[i]
a
z -2Dz+D 21/2
}
2
a1
_Idz
hj
(3)
letting
c
h2
[(i;) +1]
Equation (3) reduces to
D+h
a
z
(4)
and simplifying yields
[{DZ(c1)}2 {D2(c1)+2Dh(ci)+ch2}l/2]
= 2v
+ln
D
c
dz
(cz2-2Dz+D 21/2-S
$D
Integrating with respect to
D+h
zdz
1/2 2
1/2
D(c-l)+c
[D (c-i)]
1/2 2
D(c_1)i-ch+c
21/2 +a
(5)
[D (c_.I)-i-2Dh(c_1)+ch ]
Equation (5) was written in Fortran IV with minor adaptations
for running on a CDC 3300 computer.
The gravity anomaly expected at point
o
of Figure 23, i. e.
above the seamount summit, depends on the average density of the
material within the conic structure. Nayudu (1962) and Engel and
Engel (1963) have reported the density of some volcanic rocks
obtained by dredging in this region. The densities of these surface
167
rocks, however, cannot necessarily be expected to represent the
average density of seamounts of this region.
Shurbet and Worzel (1955) used a density of 2.84 gm/cm3 in
approximating the free-air gravity anomaly observed over two seamounts in the Gulf of Maine. Dehlinger etal. (1967a)also noted that
a density of 2.84 gm/cm3 could best remove the effect of topography
in a two-dimensional analysis in the vicinity of Cobb seamount.
Harrison and Brisbin (1959), in conducting a geophysical investigation of Jasper seamount found the observed gravity anomaly could
best be approximated using a density of 2. 3 gm/cm3. LePichon and
Talwani (1964) in investigating a seamount in the north Atlantic Ocean
discuss the possible densities to be assumed in forming a Bouguer
anomaly which exhibits a minimum relief over the seamount. They
suggest a range between 2. 3 and 2.9 gm/cm3 and use 2. 6 gm/cm3
in their analysis of possible seamount structures.
Further analysis (Woollard, 1951, McKay, 1966)
indicates that to actually fit the observed gravity anomaly with a
model a structure and substructure must be assumed for the seamount which consists of masses having large variations in density.
In particular seamount slope densities are typically light; 2. 0 to
2. 3 gm/cm3 while core densities must be heavy; 3. 0 to 3. 3 gm/cm3.
No attempt was made to generate a seamount model which would
give the expected anomaly over the seamount.
A cone model having
a uniform average density was assumed to enable estimating the
gravity anomaly to be expected over the peak of the seamount. In
this respect, the flank and basal anomalies of seamounts, shown in
Figures 6 and 7, are at least partly defined by actual measurement
and only the maximum anomaly is dependent on calculated values.
To arrive at the best average density for seamounts in the
region west of Washington and British Columbia, calculated values
were compared with measured values for Cobb and Union seamounts.
Cobb and Union seamounts were the only two seamounts where
measurements were obtained over or nearly over the seamount peak.
The density of best average fit was 2. 43 gm/cm3 (or a density con-
trast between rock and water of 1.4 gm/cm3). Peak anomaly values
of 44 seamounts were calculated using this density. The dimensions
and resulting anomalies of these seamounts are given in Table 3
and their location is shown by the x's of Figure 5. These calculated
values were used as an aid in constructing the free-air gravity
anomaly contours of Figures 6 and 7.
Linear interpolation between gravity measurements was used
in areas where no other control was available.
pkA1 SFAMNT
j-9
mo FOPMAT(14)
iril FOPMAT(4F10.2)
2n0 rPMAT(cqH GRAVITY CORRECTION
TON///6I4
DEPTH (F'MS)
(M(4L)j)
P SEAMOUNT USTN
HEII4T (FMS) PASE (T)
2o1 FOPMAT(1H
PRI'T 2rO
REA( jOr,N
PTF3.14159
CA.67
COJST2.*PIE*CA
I
I IT4I
READ lOi,AD,Apl,Aw,RHO
WXAW*j .53 184
HAH*,OrjR2RR
DsAr*,Orj 8288
AW/2.
CaH*H/(*A)+l.
F*f)* (Ce.)
FjPST(j,/C)*( (D*F)*.5-E)
x4uM=F+(c*D*F)** .5
XDEN5Fr*H+E(C**.5)
XSECAL'G (XNIJM/XDEN)
SECONDD/(C**CP) )X5EC
Ga C
(FIRST.SECCMF+A)
PRINT 21,AD,AH,AW,PHC,G
(yN) i,2,2
2 CNTINUF
FN
3200 rCRTRAN F)!AGNOSTIC RESULTS - FOR
SEANT
r
CCNE AQ!T
RHO
FT)O,
170
Table 3. Seamount locations, dimensions and gravity
Apex
Base
No.
1
2
Name
Thompson
Bear
3
4
5
6
Cobb
Location
W. Long.
N. Lat.
46
46
46
46
46
46
7
47
8
47
47
47
47
47
47
48
48
48
48
48
48
9
10
11
12
13
14
15
16
17
18
19
Warwick
Springfield
20
21
22
23
Eickelberg
Heck
24
25
26
27
28
29
30
Union
31
32
33
34
35
36
37
38
39
40
Graham
Bowie
41
42
43
Hodgkins
Denson
Dickens
44
Brown
48
48
48
48
02.5
02.5
22.0
31.5
41.8
45.6
13.0
21.0
33.9
54.0
54.4
57.0
58.9
04.1
04. 1
09.0
16.3
20.0
22.0
24.5
30.8
31.5
38.4
45.0
55.9
07.9
19.0
32.5
46.0
48.8
49.0
54.0
19.5
24.5
35.8
39.5
44.6
45.5
48
48
49
49
49
49
49
49
49
50
50
50
50
50
50
52 22. 1
53 18.0
53 30. 5
54 07.5
54 32. 8
54 53.5
Diameter
(nt. nil.)
Height
(lath)
Depth
686
1220
1120
800
1505
1687
764
280
380
600
195
740
880
1090
124
150
38
27
670
46
855
30
28
60
38.0
26.5
37.5
58.0
20.4
49.0
29.5
44.0
19.8
39.0
31.2
58.0
56.0
11.8
48.6
25.2
133 10.4
5.4
11.8
10.8
12.0
20.0
20.0
10.0
10.0
9.5
9.0
20.0
4.0
18.0
560
850
595
687
1000
570
920
1520
640
1460
129 14. 5
129 54. 5
129 23.3
7. 5
675
9. 1
626
835
1420
885
720
128
130
130
130
131
130
131
131
130
129
130
131
129
130
132
130
133
09.0
129 37.0
130 08.0
131 48.0
133 52.4
132 17.5
133 50.0
132 41.6
131 46.0
131 57.0
133 27.0
132 06.5
131 27.5
132 13.0
130 42.5
131
131
130
06.5
16.5
9.0
7.0
12.0
8.0
11.0
18.0
11.0
8.0
8.6
17.6
15.0
18.0
18.0
10.0
7.0
14.0
8.0
6.5
5.6
10.4
8.0
6.3
52.8
134 06.0
135 42.0
9.2
12.0
19.0
135 59. 5
137 45.0
136 58. 4
138 36.0
15. 3
8.0
18. 3
7.0
610
1405
700
880
1570
861
708
1050
750
700
853
910
656
745
1077
1010
1626
1318
1051
1340
788
(lath)
13
863
660
880
580
280
880
390
625
874
565
432
565
780
990
495
937
919
158
789
942
690
745
859
747
490
844
655
293
490
24
432
599
260
812
Gravity
(mgl)
24
85
72
52
26
50
121
15
108
34
31
52
104
55
34
28
100
44
59
127
45
27
67
37
27
29
57
30
32
69
66
143
93
50
108
32
171
APPENDIX 3
TWO-DIMENSTIONAL GRAVITY COMPUTATIONS
The perimeter of a two-dimensional structure can be approxi-
mated by an n-sided polygon. Taiwani, et al. (1959b)
derived expressions for the vertical component of gravitational
acceleration due to an n-sided polygon at any point. These authors
obtained for the vertical acceleration at a point P due to the attrac-.
tion of the polygon in Figure 2.4.
P
-.<
x
Figure 24. Polygon illustrating equatio panameters for.
two-dirensicai gravity computatthn
172
z
V = 2Gp
where
G
is the gravitational constant,
p
is the density, and
cos 0, (tan0.-tan.)
1
z.
1
1
e cosO j+l
1
1
(tan 0
1
1
-tan)
i+l
1
]
z.
-1
O,tan
X.
1
1
1
4.tan -1
z.
-z.
1+1
1
-x.
1+1
0
i+l
= tan
a.X
1
The expression for
Z.
1
1
X1
r
i+l
iJ
J
i+l
i+lI z -z
L
I
may be simplified as the following cases:
CaseA: If X.=O
1
z = -a1 sin4 cos
[0+_
}]
173
Case B:
0
lix.
1+1
Z. = a. Sin 4
Cos
tan4. log e
4
{COs 8. (tan 0. - tan4 )}
1
1
CaseC: IfZ.Z.i+1
1
Z. = Z.(O.i+1 - 0.)
1
1
Case D: If X.
1
X.
i+l
1
X. log
Z.
1
1
Cos 0
e
CosO.1+1
Case E: If 0.
1
Z.
0
1
Case F: If X. = Z. = 0
1
z.
1
=0
1
CaseG: IfX.i+1
z.=0
1
Z.
i+1
0
+
[.
1
1
]
174
The computer program for two-dimensional cross sections
which uses the method of Taiwanietal. (1959b) computes the
attractions along a profile of a two-dimensional polygonal
body.
The program, written in FORTRAN IV was used on a
CDC 3300 computer. The program was initially written by Mr.
W. A. Rinehart and extended and modified by Dr. E. F. Chiburis
and Mr. M. Gemperle. The program parameters are:
175
Input Cards
1) General Control
a) Cross section identification
b) First location where gravity is to be computed
c) Kilometers between points of gravity computation
d) Total number of points where gravity is to be computed
e) Number of blocks in model
f) Code to indicate if model check is to be performed
(blank - no check; any characters - check)
FORMAT (A8, F6.O, F4.O, 215, A3)
2)
First card of each block
a) Density of block
b) Number of corners in block
c) Block number
FORMAT (F5.O, 13, A8)
3) Corner of block (polygon)
a) X-coordinate of corner
b) Z-coordinate of corner
c)
block identification
4) Last card of deck - If model check is specified, a card is
required at end of the deck giving limits of model.
Output (High speed line printer)
1)
Profile identification
FORMAT (1H1, A8l1)
2) Computed gravity
a) X-location
b) Computed gravity
c) Computed gravity of point minus computed gravity at first
point using density of 1.0 for all blocks. This serves as
a check to make sure no gaps (equivalent to zero density)
or overlaps (equivalent to double density) exist in model,
and helps to locate errors when they occur. All values
should be zero except for edge effects.
FORMAT (1H, Fl0.l, F15. 2, ZOX, F17.4)
176
3) Model check - residual at three station points is computed.
This residual should be zero if the data has been read
correctly.
FORMAt (28H (model-check), Station 1 , F8. 2/
F8. 2 /
28H (model-check), Station 2
28H (model-check), Station 3 , F8. 2!!')
,
Listing of Block Densities and block numbers
FORMAl' (1H, A8, F1O.2)
4)
5) Block identification error - output occurs if block identification is not consistent in input data.
FORMAT (41H COORDINATE IDENTIFICATION ERROR IN
CASE, A5)
The program listing is:
177
pp'r,pAM TAL.wANT
rTMFNSIpJ PX(?),x(200),7(2),TCT(2OO),CMK1(2flfl),K2(2rr
DT'lFNStN CHV.3(?O0) ,A10C200) ,I'7(PflO) ,TXT(200)
FCPP.lAT(6R,F6.O,F4.O,2I,A3)
2 FCPMAT(FS,fl,13,AR)
900 F°MAT(1H ,AR,FI0.2)
1
3fl1 FDMAT(2FR.n,Ac,
35 F1AT(41H COCQrTN4TE TDF'TTTeATIN ERPOR IN Cr ,A5)
3 FPiAT(
BFR.r,)
11 FCPJ44T(4)
200 FMAT (IH ,FIO.1,F15.2.2r)X,F17.4)
21 FcMAr(1H1,AR//)
2n5 FQ4AT (2811
qMO0EL-CHFCK),STATTN la,FR.2/ 28H
,STATT
2,FR., / 2I4
P1=3. 1'597
100
p4j) l,SAM,PX(l),STEPqM,M,MCWI(
IF (M.E.0) øi ,7?
7o1 ST°
7')2 PRTNT2O1 ,SAJ4
TTT (1) r.
TT C1)0
flrl12 M
PX (1) =Px (I-i) +STEP
P7 (J)0,
ITT I T)C).
o IT(I)=.0
JJj=0
1061'4=),M
52
IF (JJJ.0.l )50,703
703 pEAfl2,4',J,S4LT
PPT1T 9r)0,SALT,PHQ
300 T*l,j
PFAp3o1,xn ,Zn
r
3oo
oc 32 y.2,j
TFCAII)(1) .1E.At0(T-1) )700,302
7o PRTPT305,AT0(1
302 CNTTNUF
50 KJ+1
X(K)X()
7(K) zZ (
SUM.0
f)D101I1,M
DOSJU1 'K
S X(J).X(J)-PX(1)
OCl?2Jl ,K
IF(X (J) )6,7,8
6 rM.ATANZJ/XCJ).P
IF(PI-TH-.00000I)47,4-7,9
47 114-PT
7 TH.5P1
-601C
$ r14.ATAN(Z(J)/X(J))
Fcn4-.00O001)45,45,q
(MflFLCHECK)
3,F&,2 //
178
45 T'40
9 TF(J.NE.l)l0,704
704 T141=TH
GC T
17
10 Ti42=TH
LJ-1
A7(J)7(L)
RX(J)x(L)
JJ=
IF(X(L).F'.O.)70S,706
JJi
7(Th IF(*(J),E.0.)707.708
o7 JJ=2
70R TF(A,EQ.0.)709,710
JJ=3
710 IF(P.E(.0.)711.712
ii JJ4
712 TF(T41.JE.TH2)19,713
s=0.
7i
r,r102
19 IF(X(L),NE,0.)??,714
714 IF(7(L).NE.0.)2P,715
715 S0.
r,rc 102
?
TF(y(J).NE.O,)pc,716
716 TF(7CJ).NE.O.)2c,717
7i S.
102
25 GCT(2b,27,28,2q,3cfl,JJ
26 S.(X(J)*X(J)*Z(L)/(A*A+X(J)*(J)))*(TH2,5*PIe(A/X(J) )0(AL(Z(L))
1-.5*ALO( (XLI)
X LI) .Z (J)* (J)
0C10102
27 S(X(L)*X(L)*Z(J)/(A*A+X(L)0X()))*(TH1.5*PI.(A/X(L))*(AL(Z(J))
1..S*ALOX(L)*XL),Z(L)*nL)
S-S
G31C102
2
S7(L)*(TH2_TH1)
GTQ102
29 VRSQRT C (Z (J)
Z CM +X (J) .X Cj) ) / CZ(L) Z CL) .X (L 'X L)
SX(t_.)*ALOG((XCL)/X(J))*V)
-
GOT1O2
O TX(J)(7(J)*R/A)
u=R*4/ (qeB,A*4)
V=71.T2
SzT*tJ*(V.W)
102 SUMSUM+S
TH1T2
122 CCNTTN'k
DQ1O3J1,K
13 X(J)X(j)+PX(j)
TF(MCI4K.EQ.0)101,718
718 IF(T.NE,l)60,710
79 CHK1(IN)=SIJM
-
)
179
TF(.JjJ.F().1)10J,Afl
0 IF IT .JtM/2) ,1 ,7?O
?O CK2(I'l)S1JM
fl
T
(i.NE.1)(,2,77
7?1 CIrSIiM
F? I
S'le13.34
111(1) 111 (I) 1)UM
StiMSUM*4)1*1 3
TCT (I) TCT (
e
) +M
101 StJM.()
106 rJTTNtJr
IF (,JJJ.FT.0)53,72?
7?
!C 1000 T=?,M
1000 ITT (Kr) T11 (KT )-TIT (1)
TI! (1) O.
PP1T
00,(PX(T),TT(I),TTT(T),T1,M)
IF (MCHK.FQ,O) lnn,723
723 JJJ1
TC-4K10.
TCl4P2O.
TC'4K30.
I)51 Izi ,N
TCHK1=TCHKI+CHK1 (I)
TCNK2TCHK2+CNK7 IT)
51 TC3T(HK3+CHKI(I)
307 PF4n3 (xU ,Z(fl ,I1,4)
N I
J=4
53 PsTCH(ICHK1(1)
R2=THK-CHK2 (1 1
P3TCHK3-CHK3 Ill
pTNT20c,w1 ,P2,p3
(T'o0