Embed Size (px)
Citation preview
AN ABSTRACT OF THE THESIS OF
Annette V. deCharon for the degree of Master of Science
in Oceanography presented on October 14. 1988.
Title: Structure and Tectonics of Cascadia Segment. Central Blanco Transform Fault Zone
Abstract approved:
Robert W. Embley
Seismic-reflection profiles and SeaBeam bathymetry are used to examine the structural
development of the Cascadia Segment, a region of back-tilted blocks flanking a central
depression within the Blanco transform fault zone. The depressed position of the Cascadia
Segment has caused this area to act as a sediment trap for bottom transported material
moving through Cascadia Channel (Duncan, 1968; Griggs & Kulm, 1970). The existence
of thick turbidite sequences atop tectonic blocks this an ideal place for seismic-reflection
imaging.
Escarpment slopes and fault dip angles in Cascadia Segment range between 110 and
31°, which are, on average, shallower than values reported from slow-to-intermediate-rate
spreading centers. The perched turbidite sequences are assumed to have originated within a
central rifting basin. The lack of unconformable sequences within the back-tilted turbidites
indicates that the tectonic blocks of Cascadia Segment have been uplifted in a continuous
manner. Thus, a quantitative analysis of tilt of turbidite sequences versus distance from the
present central rifting basin, Cascadia Depression, is used to understand the structural
development of the Cascadia Segment. The high degree of tilt seen in turbidite sequences
proximal to the present central rifting basin is consistent with a model in which turbidite
flows are deposited within a basin that is being uplifted and spreading away from a narrow
extensional zone. Individual tectonic blocks rotate at least 6° as they move through the
terraces of Cascaclia Segment. Extension of the eastern terrace has been accommodated by
rotation of three large, discrete blocks whereas the western terrace has extended through
block rotation and a large number of normal faults with smaller throws. Irregular
morphology of the flanking blocks of this spreading center may be attributed to short-term
(on the order of 106 years) asymmetric spreading. Turbidite tilt data and fault orientations
demonstrate conclusively that the muting of topography from the rift valley to the rift
mountains greater than 30 km off the central spreading axis is accomplished by: (1)
movement along outward-dipping normal faults in the rift mountains, and (2) back-tilting
of tectonic blocks within the terraces. The possibility of reverse motion along pre-existing
Redacted for Privacy
inward-dipping faults in the rift mountains is not refuted by tilt data from Cascadia
Segment.
Sediment thickness on the floor of Cascadia Depression varies between less than 50 m
and greater than 850 m. Basement doming, evident in the center of the depression, may be
caused by the intrusion of sills into the sediments. There is a great disparity in basement
slope between the eastern half (average slope = 11.5°) and western half (average slope =
4.2°) of Cascadia Depression. Asymmetry of basement slope seems to be established in the
central rifling basin and is propagated outward onto the terraces until isostatic equilibrium is
reached. The local slope of Cascadia Depression's floor is the result of interaction between
two components: rate of movement along faults within the basin and the rate at which
turbidite-derived sediments can bury the faults. At present, movement along low-angle
fault surfaces within Cascadia Depression is occurring at a higher rate than sediments are
being delivered to the basin. Duncan (1968) has estimated that the rate of turbidite
sediment accumulation in this area decreased rapidly about 12,000 years ago. Fault angles
and offset values were used to calculate extension rates averaged over this time period. A
progressive decrease in opening rate from 5.8 cm/yr to 4.1 cm/yr is evident from the
northern to central parts of Cascadia Depression, respectively. This discrepancy in
estimated opening rates is probably the result of: (1) "true" short-term asymmetric opening
from north-to-south; (2) the masking of surface offset along faults towards the southern
part of Cascadia Depression by ponding of turbidites; or (3) a combination of both. The
calculated rates of opening support a model in which Cascadia Segment is a "trapped"
spreading segment from an episode of rift propagation (Embley et al., in press). The
calculated extension rate of Cascadia Depression coupled with the morphologic similarity
between Cascadia Segment and slow-to-intermediate mid-ocean ridges support the idea that
this is an area of active seafloor spreading.
The tectonic evolution of Cascadia Segment is compared to existing models of rift
valley evolution and crustal generation processes at mid-ocean ridges, especially in the
context of steady state versus non-steady state theories of rift valley development (i.e.,
Davis & Lister, 1977; Kappel & Ryan, 1986). The high subsidence rate of the rift valley
floor (1.8 cm/yr averaged over 6,660 years) coupled the calculated horizontal extension rate
(about 5 cm/yr averaged over the past 12,000 years) may indicate that the current magma
supply has waned, thus allowing extension to become the dominant mechanism at Cascadia
Depression. Therefore, recent tectonics may support the idea that the balance between
extension and volcanism at Cascadia Segment may be episodic in nature (after Kappel &
Ryan, 1986).
Structure and Tectonics
of Cascadia Segment, Central Blanco Transform Fault Zone
by
AnnetteV. deCharon
A THESIS
submitted to
Oregon State University
in partial fulfillment of
the requirements for the
degree of
Master of Science
Completed October 14, 1988
Commencement June, 1989
APPROVED:
Adjunct Professor of Oceanography in charge of major
Dean of College of Oceanography
Dean of GraduateS9i
Date thesis is presented October 14. 1988
Typed by researcher for Annette V. deCharon
Redacted for Privacy
Redacted for Privacy
Redacted for Privacy
ACKNOWLEDGEMENTS
During the course of my graduate work, I have come to know and respect many
people, both professionally and personally. I would like to express my gratitude to all
those who have encouraged me and facilitated the completion of this thesis. I have a great
amount of admiration for my major advisor, Dr. Bob Embley, who has continually
supported me as a student and a friend. My defense committee, Drs. LaVerne Kuim, Shaul
Levi, and Wayne Seim, made a great contribution to content of the thesis and I certainly
appreciate their efforts. Much of the seismic-reflection data used in this study was
generously provided by Dr. Bill Normark of the U.S. Geological Survey. His quick (yet
thorough) edit of the first draft of the thesis hastened its completion and I am grateful for
his help. Discussions with Drs. Clive Lister and P.J. Fox helped with the interpretation of
the tectonic and structural aspects of the work. I wish to thank Drs. Fred Spiess, John
Delaney, and Clive Lister for access to data collected during the 1987 cruise aboard the R/V
Melville. Lingyun Chiao spent many hours generating a seismic-reflection profile that was
used in this study. Computer-digitization of seismic-reflection records was possible thanks
to the help of David Borg-Breen of NOAAIPMEL, Andy Lau and Marijke Van Heeswijk.
Figure 4 was drafted and sent to me (at a moment's notice) by Gini Curl of NOAA/PMEL.
Three of the people who have had a great influence on my "budding" career in Marine
Geology are Drs. Jim McClain, Dallas Abbott, and Jim Franklin. Their good humor,
excellent advice, and enthusiasm will remain with me for a long time.
My position as a graduate research assistant through the NOAA "Vents" program
allowed me to work closely with people from Corvallis and Newport. My first year-and-a-
half of graduate school was a pleasant experience thanks to support of the Oceanography
faculty, students, and staff. Friends such as Dave M., Collin, Tom, Cindy, Yip, Bruce
A., Bruce F., Bruce S., Anne P., Karen, Dana, Beth, and the "Bull Pen" crowd made that
time more enjoyable. The Secher Ave. gang will always be remembered for its
contribution to my sanity: Jana, Collin, Tom, Heidi, Chris, Lucy, Jack, Rambo, etc. The
tolerance and support of Newport folks warrants special recognition. Many thanks to Bob,
Steve, Chris, Susan, Bruce, Michele, Maria, Andy, Marijke, Tom, Saskia, Karen D.,
Anne D., Jessica, Robin, and Pete/Bud.
I want to let my family know that I appreciate all the support and sacrifices that they
have made for me. The encouragement given to me by my parents, Daye, and Elwood
helped me to survive many rainy (and the few sunny) days in ol' Ory-gun.
This research was supported by the National Oceanic and Atmospheric Administration
grant NA-87-ABH-00018.
TABLE OF CONTENTS
Page
INTRODUCTION ................................................................... 1
The Blanco Fracture Zone and its basins ................................. 2
Cascadia Segment........................................................... 2
PREVIOUSWORK................................................................. 7
Regional studies of the Blanco transform fault zone .................... 7
Area-specific studies within the Blanco transform fault zone 10
DATA ANALYSIS AND METHODS ............................................. 15
Seismic-reflection data ...................................................... 15
SeaBeam data processing................................................... 17
STRUCTURE AND SEDIMENTATION OF THE CENTRAL BLANCO
TRANSFORM FAULT ZONE BETWEEN 129°20W AND 128°00'W:
CASCADIASEGMENT............................................................ 18
Statistical study of Cascadia Segment turbidite sequences ............. 18
Fault angles and orientations in Cascadia Segment: comparison with
models of slow-to-intermediate spreading mid-ocean ridges........... 20
Tilt measurements of turbidite reflectors about central axis............ 24
Summary of the tectonic evolution of Cascadia Segment ............... 30
SMALL-SCALE STRUCTURE WITHIN A SEDIMENTED EXTENSIONAL
BASIN: CASCADIA DEPRESSION ............................................. 33
Isopach map of Cascadia Depression sediment.......................... 33
Cross-sectional profiles across Cascadia Depression ................... 36
Classification of acoustic facies using 3.5-kHz reflection records 38
Fault orientations within Cascadia Depression........................... 41
Estimated spreading rate of Cascadia Depression over the past
12,000 years ................................................................. 47
Summary of the tectonic evolution of Cascadia Depression............ 52
CONCLUSIONS AND PROPOSED FUTHER STUDY ....................... 54
Models of crustal accretion at spreading centers: steady state versus
non-steady state rift valley formation..................................... 55
Is there hydrothermal activity within Cascadia Depression' ........... 59
How crustal accretion may be occurring at Cascadia Depression 60
Futurestudies................................................................. 65
BIBLIOGRAPHY .................................................................... 67
LIST OF FIGURES
Figure Page
1. Location map of Blanco transform fault zone within the northeast
PacificOcean ........................................................ 3
2. SeaBeam map of Cascadia Segment ...................................... 4
3. Seismic-reflection profile from Cascadia Segment ...................... 5
4. Embley et al.'s (in press) model of possible sequence of events over
the past 5 Ma outlining the development of the Blanco
transform fault zone ................................................. 8
5. Cores 6609-19 and 6609-20 (from Duncan, 1968) ..................... 11
6. NOAA watergun seismic record of Cascadia Depression .............. 14
7. Plot of ships' tracks within Cascadia Segment showing seismic-
reflection data navigation lines .................................... 16
8. Depth (100 m intervals) of reflectors versus tilt......................... 21
9. Fault dips and morphotectonic terminology as applied to Cascadia
Segment ............................................................. 22
10. Morphotectonic terminology and plot of tilt values in Cascadia
Segment assuming turbidites originally horizontal ............. 25
11. Macdonald & Atwater's (1978) models of rift valley-to-rift mountain
transition ............................................................. 29
12. Single-channel seismic-reflection profile showing relative motion
alongfaults .......................................................... 31
13. Isopach of minimum sediment thickness within Cascadia Depression 34
14. Plot of ships' tracks within Cascadia Depression showing seismic-
reflection data navigation lines.................................... 35
15. Isometric view of seven topographic cross-sections within Cascadia
Depression .......................................................... 37
16. Examples of four 3.5 kHz echo-character types used in this study
from profiles within Cascadia Depression.(based on Damuth,
1975 and 1980; Damuth et al., 1983)............................ 39
17. Distribution of four 3.5 kHz echo-character types within Cascadia
Depression (based on Damuth, 1975 and 1980; Damuth et al.,
1983) ................................................................. 40
18. Fault orientations based on SeaMARC II side-scan sonar data, 3.5-kHz
reflection data and bathymetry within Cascadia Depression... 42
19. Seismic-reflection records showing approximate fault dips from the
northwest corner of Cascadia Depression ....................... 44
20. SeaMARC II side-scan sonar image of northern Cascadia Depression. 45
21. Unmigrated single-channel airgun seismic records showing some of
the fault dips used in calculation of extension within Cascadia
Depression .......................................................... 50
22. Map view of faults within Cascadia Depression and total spreading
rates indicated ....................................................... 51
23. Mechanisms for the formation of axial-valley topography at slow-
spreading ridges (from Deffeyes, 1970) ......................... 56
24. Models of steady state and non-steady state rift valley formation and
the expected distribution of sulfides within one stage of rift
valley development ................................................. 58
25. Seismic-reflection profiles from three sedimented spreading centers:
Guaymas Basin, Escanaba Trough, and Middle Valley 61
26. Seismic-reflection profile showing drag-folds at margin of volcanic
complex at center of Cascadia Depression ....................... 64
Table
LIST OF TABLES
Page
Statistical analysis of tilt measurements from eastern blocks 1 & 2... 19
Vertical and horizontal components of displacement rate from surface
offsets ................................................................ 48
STRUCTURE AND TECTONICS OF CASCADIA SEGMENT,
CENTRAL BLANCO TRANSFORM FAULT ZONE
INTRODUCTION
Geophysical data, such as SeaBeam bathymetry and SeaMARC side-scan sonar, has
been used to understand the complicated structure and morphology of the transform zones
of the world's oceans. One common feature within transform zones is the extensional, or
pull-apart basin, resulting from divergence along strike-slip faults (Fox & Gab, 1984).
On the other hand, Cascadia Segment, a morphologically and structurally complex area
within the Blanco transform fault zone (Plate 1), has evolved as the result of a long history
of active seafloor spreading. This adds another dimension of complexity to what we
know about the structure of oceanic transform fault zones. The existence of very short
spreading segment such as Cascadia Segment within a long transform fault zone may force
us to re-think the idea that stationary deep-seated processes must control where seafloor
spreading occurs.
The tectonic evolution of Cascadia Segment is an important factor in deciphering the
geologic history of the southern Juan de Fuca and northern Gorda Ridges. This area ofthe northeast Pacific Ocean has been affected by about 12 million years of ridge
propagation and transform reorientation. The existence of a short seafloor spreading
segment within the Blanco transform fault zone may help explain the complicated magnetic
anomaly patterns found at this area of the ocean floor.
The depressed position of the Cascadia Segment has caused this area to act as a
sediment trap for bottom transported material moving through Cascadia Channel (Duncan,
1968; Griggs & KuIm, 1970). The existence of thick turbidite sequences atop tectonic
blocks within Cascadia Segment makes this an ideal place for seismic-reflection imaging.
Seismic-reflection data from this area reveals the structure and tectonics associated with
seafloor spreading: a difficult task on sediment-free ridges.
Cascadia Depression, the largest rhomboidal basin within the Blanco transform fault
zone and centrally located Within Cascadia Segment, was proposed to be a site of active
seafloor spreading (Embley et aL, 1987). Cascadia Depression is analogous to
sedimented spreading centers found on mid-ocean ridge crests. Sedimented spreading
centers are considered possible analogues of some ancient sediment-hosted ore deposits.
The first section of this paper addresses the overall structure of Cascadia Segment in
light of models of rifting processes. The second section features an analysis of the
2
detailed bathymetry, structure and tectonics of the current site of active spreading,
Cascadia Depression.
The Blanco Fracture Zone and its basins
The Blanco Fracture Zone is a 350-km-long transform zone that connects the southern
Juan de Fuca and northern Gorda Ridges and includes a fossil transform segment to the
southeast (Figure 1). The active part of the fracture zone, the Blanco transform fault zone,
is a series of right-lateral, right-stepping strike-slip faults separated by extensional basins.
SeaBeam bathymetry reveals the morphological complexity of each of the basins (Plate 1).
From west to east, these basins are identified as the Blanco Trough (or West Blanco
Depression), East Blanco Depression, Surveyor Depression, Cascadia Depression and
Gorda Depression. Two of these basins, Surveyor Depression and Gorda Depression,
have rhomboidal shapes. Rhomboidal basins are also found on continents and result from
divergence along strike-slip faults (Aydin & Nur, 1982). The other basins within the
Blanco transform fault zone are morphologically more complicated. The Blanco Trough is
an elongate "hole" more than 2400 m deep that constitutes the westernmost end of the
active transform zone. East Blanco Depression has an overall rectangular shape and is
structurally more complex than simple rhomboidal pull-apart basins (Palmer et al., 1987;
Embley et al., in press).
Cascadia Segment
The structure and tectonics of Cascadia Segment, located in the central part of the
Blanco transform fault zone, is the focus of this study (Figure 2). The deepest part of the
basin (greater than 3000 m) has a rhomboidal shape. This deep area, hitherto referred to
as Cascadia Depression, measures 17 km long and 15 km wide. The structure of the
Cascadia Segment is complicated by back-tilted blocks that flank the long (N-S) axis of the
central basin and extend 50 km beyond it toward the east and west (Figure 3). In cross-
section, the Cascadia Segment with its deep central axis and back-tilted flanking blocks,
looks very much like a slow spreading mid-ocean ridge axis. The back-tilted blocks are in
some cases covered with thick turbidite sequences whose dominant source is the North
American Continent (Griggs & Kuim, 1973). Most of these sediments were transported
by turoidity currents via Cascadia Channel (Plate 1) and other smaller channel systems(Figure 2: Embley et al., in press; Griggs & Kuim, 1970).
The structure of portions of Cascadia Segment was initially summarized by Griggs &
Kuim (1973) and Ibach (1981). In this study, a more detailed examination of the
structural development of the Cascadia Segment is conducted using seismic-reflection,
3
132° 1300
i5 \%
cr;i '.. c
: -
1280 1260 124°
Canada
Washington
/
Oregon
-42°J)
1MENDOCINO F. -. California-_
40°132°
130° 128° 126° 124°
Figure 1: Location map of Blanco transform fault zone within the northeast Pacific Ocean.
I?970°00 -I29OOO -129'OOOO -I26OOO -12e10OO. -1263000 -IB2OOO -1281000 128'OO 00-IIi. 00, 00
-iI*4i-jc------jJi
Figure 2: SeaBeam map of Cascadia Segment. Contour interval is 100 m. SeaBeam data
collected by NOAA/NOS within the Blanco transform fault zone area in 1981,
1983 and 1984. Line AA refers to cross-section profiles seen in Figures 9 and
10. Dredge tracks are designated by dotted lines. Hydrocast locations are shown
as wavy patterns. Locations of cores 6609-19 and 6609-20 are shown as stars at
longitudes -12840' and -12829', respectively.
IMLII.
+700 gammas
-700 gammas
Figure 3: Seismic-reflection profile from Cascadia Segment. Vertical exaggeration isapproximately 15: 1. An electromagnetic sound source was used with frequencycentered at 200 Hz. A dual-channel hydrophone was used that measures both a
pressure signal and a velocity signal (vertical geophone). These two signalswere used together in order to distinguish between bottom returns and watersurface returns (Lister et al., 1987). The course of Cascadja Channel dissectsthe profile just east of the ea.sternmost tectonic block (directly above the lettersV.E.). Note the buried back-tilted block on the westernmost flank of Cascadia
Segment. Magnetic anomaly profile is shown below.
Ui
SeaMARC IT side-scan sonar, and SeaBeam bathymetric data. On the regional scale, these
data demonstrate that the morphology of Cascadia Segment has developed through seafloor
spreading. The acoustic penetrability of the turbidites on the flanking ridges of the
depression provides a means of calcuLting the amount of tilt these blocks have undergone
during the rifting process. Relative motion along faults is determined in the Cascadia
Segment, a formidable task at mid-ocean ridges. East-west morphologic asymmetry is
evident at Cascadia Segment and may be a result of asymmetric seafloor spreading. The
topographic development of Cascadia Segment is compared to models of rift valley
evolution and crustal generation processes at mid-ocean ridges. This, in turn, allows
examination of current models of topographic evolution at spreading centers (i.e.,
Macdonald & Atwater, 1978; Davis & Lister, 1977; Deffeyes, 1970).
Cascadia Depression is the prese.n locus of spreading within Cascadia Segment.
High-resolution 3.5-kHz reflection records and SeaMARC H side-scan sonar data indicate
that turbidite sediments, covering the Cascadia Depression floor, are cut by faults.
Acoustic basement doming at the center of the basin is revealed in a map of sediment
thickness distribution within Cascaclia Depression. A quantitative estimate of spreading
rate is made for Cascadia Depression using a method described by Davis & Lister (1977).
The relationship between turbidite accumulation and tectonic movement within Cascadia
Depression as it rifts apart is explored and compared to other sedimented spreading centers(Guaymas Basin, Gulf of California; Escanaba Trough , southern Gorda Ridge; andMiddle Valley, northern Endeavour Ridge).
PREVIOUS WORK
Regional Studies of the Blanco transform fault zone
Using seafloor magnetic anomaly patterns and the time scale of Heirtzler et al. (1968),Atwater (1970) proposed that the Blanco transform formed approximately 10 to 15 Ma.The magnetic anomaly patterns of the southern Juai de Fuca Ridge are complicated by riftpropagation along the ridge system Hey et al., 1980; Wilson et aI.,1984). The Blancotransform fault zone has extended progressively westward over the past 12 million years
(m.y.) by annexation of propagating offsets (Hey et al., 1980). Wilson et al. (1984)proposed that the western end of the Blanco transform fault was established as the
southward-directed rift propagation on the Juan de Fuca Ridge ceased about 5 Ma. Shortly
after this, in response to a major shift in the Juan de Fuca / Pacific pole of rotation,
northward-directed rift propagation eliminated a short ridge segment located north of the
first pseudofault (Wilson et aL, 1984). A more detailed model of Embley et al. (in press),
postulates that the Cascadia Segment is a short remnant of the ridge segment eliminated inWilson et al.'s (1984) model (Figures 4A, 4B). Embley et al. (1987, in press) speculatethat Cascadia Segment has been actively spreading at a full-rate of 6 cm/yr over the past 5million years. This is consistent with GLORIA side-scan sonar data that shows a fault
trace, coincident with the southern boundary fault of Cascadia Segment, continuous to atleast I 30°30'W (Figure 4E ; EEZ Scan 84 Scientific Staff, 1986). A single-channel
seismic-reflection record reveals a 14.0 m high fault scarp coincident with the GLORIA
side-scan fault trace at 130 lOW longitude. Yet another episode of southward-directed riftpropagation along the Juan de Fuca is proposed to have occurred at about 1.5 Ma (Figures4C, 4D; Embley et al., in press). At this time, the fossil transform located at the westernend of the Blanco transform fault zone was reactivated. Normal spreading at the southernJuan de Fuca Ridge, after cessation of propagation, has progressively shifted thepseudofaults to the east producing crust with typical ridge-parallel structures and magneticanomaly patterns between the ridge axis and pseudofault trace (Figure 4E).
The Blanco transform fault zone is seismically active (Tobin & Sykes, 1968; Bolt etal., 1968; Chandra, 1974; Couch et al., 1978; Rodgers, 1979), but because of poorazimuthal distribution of seismograph stations and poorly constrained crustal velocitymodels, the epicenters determined from P-phase data have routinely shown offsets of 20-30 km northward from the topographic expression of the fracture zone (Northrup, 1970).The use of T-phase data greatly improves this fit (Dunnebier & Johnson, 1967; Northrup,1970). A local seismicity study using sea-surface sonobuoy arrays recorded earthquakes
Figure 4: Embley et al.'s (in press) model of possible sequence of events over the past 5Ma outlining the development of the Blanco transform fault zone. Diagramsbased on unpublished magnetic modelling by D. Wilson, SeaBeam bathymetrvand GLORIA side-scan sonar data (EEZ Scan 84 Scientific Staff, 1986). A majorshift in the Juan de Fuca / Pacific pole of rotation occurs between 4A and 4B,causing a reorganization of the transform. The Cascadia Segment began at thistime as the result of an incomplete capture of the segment by the propagator. A
propagator impinged on the inactive extension of the transform fault between
diagrams 4C and 4D, reactivating it (about 1.5 Ma). Initiation of smallextensional basin at the eastern end of the transform (presently the GordaDepression) may have occurred as the result of the "death' ofa northward-propagating rift before it reached the Blanco Fracture Zone (1-2 Ma ?). Themodel suggests that the Surveyor and Blanco Depressions are the youngest of theBlanco extensional basins and have formed by a minor change in the relative
motion between the Juan de Fuca and Pacific plates.
NN
NN NI
NN
N
IJi,f
N
N/
\NN
NN
N)
-
N
NN
NNN
NNN
N
NN
NN NNN
/N
/NKN N
N
N
'
I'
N
NN
/
I4
N
-
N
N
--
N
Ii:
/
N
/NN'
/N
---
N
-41
N'
/\\
JI
I
I
I
II/I
I
I
I
6
-4\
s1,
$ffrj
/
10
along the northwesternmost Blanco Ridge and within the Cascadia Segment (Plate 1;
Johnson & Jones, 1978). During the deployment, a rnicroearthquake swarm of fourteen
events occurred within the Cascadia Depression. These events were interpreted as
extensional movements consistent with the seafloor spreading.
The gravity field over the Blanco transform fault zone is characterized by a negative
free-air gravity anomaly (Dehlinger et al., 1970) with the largest anomaly (-30 mgal)
located over the Blanco Trough. Melson (1969) notes a strong positive (greater than 1,400
gammas) magnetic anomaly over the Blanco Trough. This magnetic anomaly probably
reflects the presence of highly magnetized (Fe-Ti) basalts commonly found at propagating
rift tips (Vogt & Bycrly, 1976).
Area-specific studies within the Blanco transform fault zone
The steep northern scarp of Blanco Trough has been dredged and imaged in detail
using deep-towed sonr (Delaney et al., 1987). Rock types include hyalloclastites, pillow
basalts, diabase, greenstone, tectonic and hydrothermal breccia. Recent hydrothermal
activity is noted by SeIk (1978) and M. Leinen (pers. comm.) in the Blanco Trough.
Embley et al. (in press; Figure 4E) and Palmer et al. (1987) have proposed that the locus of
transform motion near the Southern Juan de Fuca ridge-Blanco transform intersection has
recently jumped to the north, opening the Blanco Trough. They interpret the East Blanco
and Surveyor depressions as zones of extension that served to "transfer strike-slip motion
to the main body of the Blanco Fracture Zone" as the Blanco Trough opened up (Figure 4;
Plate 1). The model of Embley et al. (in press; Figure 4E) depicts the Blanco Trough, East
Blanco and Surveyor Depressions as young (less than 1 Ma) extensional basins that formed
as a result of a minor change in the relative motion between the Juan de Fuca and Pacific
plates.
The magnetic anomaly over Cascadia Segment is of low amplitude and flattened in
nature (Figure 3; Lister et al., 1987). Three mechanisms proposed to account for very low
amplitude seafloor magnetic anomalies are hydrothermal leaching of iron-titanium oxides
(Levi & Riddihough, 1986), low specific remanence intensity of sills / dikes as compared
to extrusive basalts (Irving, 1970; Vogt et al., 1970), and crustal temperatures beneath
thick sediment that exceed the Curie points of unoxidized titanomagnetites (Johnson &
Atwater, 1977; Johnson & Flail. 1978; Marshall, 1978; Prévot et al., 1979).
Core data documents significant vertical tectonism within Cascadia Segment (Duncan,
1968). Piston core 6609-19, from Cascadia Depression, contains turbidite-derived
sediments with Mt. Mazama volcanic ash (6,600 y. in age) at the base (Figures 2 and 5).
Duncan (1968) concluded that the floor of the basin has subsided 120 m because Mt.
11
Figure 5: Cores 6609-19 and 6609-20 (from Duncan, 1968).
(A): Core 6609-19 is over 8 m in length and contains Mt. Mazama volcanic ash(6,600 y. in age) and turbidite-derived sediments (Duncan, 1968).
(B): Core 6609-20, was taken at a depth of 2600 m, from a fault-block ridge located
east of the Cascadia Depression. It shows a progressive change from sandturbidite deposits at the base, gradually reducing in number and thicknessupsection, to totally pelagic, foraminiferal sediment near the top. Studies ofturbidity current flows in this area (Duncan, 1968; Griggs & Kuim 1973; Kuim& Fowler, 1974) indicate that sand was not deposited higher than 250 m abovethe Cascadia plain or the channel. The estimated uplift of this fault-block ridge is450 m at a rate of 2 cm/yr (Duncan, 1968).
.1
A6609-19C04t SICOINV tPt
-- 4 -.
1I
F
Figure 5
COnC S(D'C'I1 T"S
a
B (North)
3.3
3-5
4.0
0 2000..L I
C14 Date6609-20
0 -- 14,850!200 '(OP
: 1
--- 16,450!750 '(OP
--o- 21,950!950 'I'BP
B' (South)
CISC4O/4
-
:__ ...' ---
..
4': .- ,.0
,.,
yE. 35
13
Mazama ash is present in turbidite sequences both up and down channel from Cascadia
Depression. Core 6609-20, from a fault-block east of Cascadia Depression, shows a
progressive change from sand turbidite deposits at its base, gradually reducing in number
and thickness upsection, to totally pelagic, foraminiferal sediment near the top (Figures 2
and 5). Duncan (1968) estimated the uplift of this fault-block ridge to be 450 m at a rate of2 cm/yr.
Hydrocasts taken in Cascadia Depression show maxima with respect to methane
between depths of 2600 m-3500 m (Embley et al., 1987; Figure 2). Methane anomalies
may be hyarothermal exhalations or stem from the heating of organic matter within the
turbidite sequences by volcanic sills intruded beneath the basin. Seismic reflection records
reveal what may be volcanic sills near the center of the depression (Figure 6). Dredge
samples obtained from CascaLlia Depression include mudstone, dolostone and siltstone
(Figure 2).
Local mechanisms, such as mass-wasting / backcutting on the south wall of Cascadia
Channel triggered by earthquake activity along the Blanco Ridge and turbidity currentsgenerated from slumping on the north wall of the Blanco Ridge, are proposed to account
for the distinct charmelized terrain located north of the Blanco Ridge (Plate 1; Embley,
1985). The downdropping of Cascadia Depression's floor may be a conthbuting factor tothe mass wasting along this section of Cascadia Channel.
According to the model of Embley et al (in press; Figure 4D), the easternmostextensional basin in the Blanco transform fault zone, the Gorda Depression, initiated as the
result of the cessation of a northward-propagating rift before it reached the Blanco Fracture
Zone (1-2 Ma ?). Hydrothermal activity is noted in the Gorda Depression by Hart et
al.(1986). Dredge hauls consist of hydrothermally altered basalts, greenschist gabbros and
greenschist breccias each representing "distinct stages of hydrothermal mineralization in the
evolution of seawater into mature hydrothermal solution and then cooling of these solutions
as they exit from oceanic crust" (Hart et al., 1986).
14
.1.. ... -.. Bk
II BIocki 1
.: ': '.
: ..;!:--..: ;j...
-5km
Figure 6: NOAA watergun seismic record of Cascadia Depression. Vertical exaggeration =9.7. Subcrustal reflectors, located near the center of the depression, may bevolcanic sills (Embley et al., 1987). The designations "Block 1" and "Block 2"refer to the statistical study described in Chapter One.
15
DATA ANALYSIS AND METHODS
Seismic-reflection data
Seismic-reflection data from Cascadia Segment used in this study comes from thefollowing sources:
(1) Single channel airgun and 3.5-kHz seismic-reflection records taken aboard the RJVS .P. Lee in 1985 by William Normark of the US Geological Survey;
(2) Watergun single-channel and 3.5-kHz seismic-reflection data collected by RobertEmbley in 1984 on the NOAA Ship Surveyor;
(3) One 3.5-kHz reflection profile taken within Cascadia Depression on the NOAAShip Discoverer in August, 1987; and
(4) A seismic-reflection profile taken in July, 1987 aboard the RJV Melville (Figure 3).Over 140 reflectors were digitized from seismic-reflection records in order to quantify
sediment thickness and the tilt of seismic layers. An interval velocity of 1.8 km/sec wasused (Kuim & von Huene et aL, 1973; Ibach, 1981) and the velocity-depth gradient withinthese sediments was assumed to be zero. This assumption is within the 95% confidencelevel for deep-sea sediments shallower than about 1080 m (Carison et al., 1986). Thecalculated tilt of very deep acoustic layers will be too low because of the onset of thevelocity-depth gradient at about 1080 m depth. The overall variation in maximum sedimentthickness atop the fault blocks at Cascadia Segment is from less than 300 m to greater than1100 m. The largest calculated error estimate within the Cascadia Segment tilt data is only2.8°.
Individual turbidite reflectors on the back-tilted blocks were selected by eye at aninterval of 1-2 reflectors every 100 m of depth. The digitized data was stored on a VAX750 / microVAX 2 operating system as two-way travel time (depth) versus absolute time.The reflection records were merged with navigation tracks that were likewise keyed toabsolute time (Figure 7). The digitized subsurface acoustic reflectors were placed into athree-dimensional grid: X (latitude), Y (longitude) and Z (depth as two-way travel time).Cross-sectional profiles (vertical exaggeration = 10:1) generated by a DISSPLA graphicssoftware program were assembled into a three-dimensional fence diagram model (afterDavis & Lister, 1977). The positioning of cross-sections relative to one another was aidedby correlation with SeaBeam bathymetry. Using this method, records from the varioussources at differing scales were reduced to one data base.
-t29' 20 00
L°L?\
-129'OO'OO' -12e.sOOo' -126'IOOO' -128'30OO' -1282000 -12810OO' -12800,00.
U' A
Figure 7: Plot of ships' tracks within Cascadia Segment showing seismic-reflection data
navigation lines. NOAA Ship Surveyor tracks are solid lines. RJV S.P. Leetracks are dashed lines. RJV Melville track is a dotted line.
S
-y z-
17
SeaBeam bathymetric data vvas a valuable resource in resolving the structure of Blanco
transform fault zone. The SeaBeam data base used in this study consists of smoothed ship
navigation merged with multi-beam echo sonar data. Using a VAX 7501 microVAX 2operating system and a RADIAN CPS-1 ("Contour Plotting System") software program,
SeaBeam data files were converted into evenly spaced gridded data sets at various scales.
The area of the seafloor ensonified by SeaBeam multi-beam sonar corresponds to a
rectangular area subtending angles of 600 by 2 213°. The standard swath width ensonified
by SeaBeam sonar is equivalent to 75% of water depth. Within the Blanco transform fault
zone, water depth is highly variable: ranging from less than 1600 m to greater than 4800 m
(Plate 1). Thus, evenly spaced trackline coverage during the initial SeaBeam survey of the
Blanco transform fault zone in 1983 left intermittent gaps between adjacent swaths. A field
operation designed to fill in some of the more severe gaps in SeaBeam sonar data was
carried out on the NOAA Ship Discoverer in 1987. Subsequently, in order to finalize the
bathymetry, it was necessary to edit existing SeaBeam sonar data and append the data
collected in 1987 to the earlier survey. This was done using a computer program that
allows blocks of sonar / navigation data to be edited based on absolute time. The final
result was a synthesis of SeaBeam bathymetric data from various smveys in the Blancotransform fault zone over a five-year period. These data were used to generate a multitudeof graphical plots such as bathymetric maps, cross-sectional profiles and isometric view
plots. These various plots proved to be a integral factor in the interpretation of structureand morphology of Cascadia Segment.
18
STRUCTURE AND SEDIMENTATION OF THE CENTRAL BLANCO
TRANSFORM FAULT ZONE BETWEEN 129°20W AND 128O(YW:
CASCADIA SEGMENT
By anaigy with sedimented spreading centers, suchas the Escanaba Trough (Atwater
& Mudie, 1973), the occurrence of back-tilted blocks covered with turbidite sequences that"fan out" away from a central depression suggests that the Cascadia Segment was formed
by seafloor spreading. It was shown by Atwater & Mudie (1973) that turbidite layers
perched on back-tilted blocks flanking the Gorda Rise (located north of the Escanaba
Trough) were deposited as a more continuous, nearly horizontal body at the time that the
underlying crust occupied the valley floor, during and shortly after its formation. Using
magnetic intensity and seismic-reflection data they showed that the turbidite-derived
sediments occupying the central rifting basin were uplifted and tilted along with the
underlying crust as it moved away from the spreading center. Therefore, based on the
assumption that perched turbidite sequences at Cascadia Segment originated within a central
rifting basin, seismic-reflection profiles and SeaBeam bathymetry were used to examine the
structural development of the Cascadia Segment. The lack of unconformable sequences
within the back-tilted turbidites (Figure 3) indicates that the tectonic blocks of Cascadia
Segment have been uplifted in a continuous manner. Duncan (1968) has estimated the rateof uplift on one block located east of Cascadia Depression to be 2 cm/yr (Figure 5).
A statistical analysis of tilt versus depth was conducted on the first two back-tilted
blocks located east of Cascadia Depression (Figure 6). These blocks were chosen for the
following reasons: (1) the high density of data lines provides good areal coverage of thefirst two eastern blocks (Figure 7); (2) one trackline crosses the trend of other data over
these two blocks (Figure 7); (3) seismic-reflection profiles show better resolution of deep
acoustic layers on less tilted blocks (in this case, those blocks located closerto Cascaina
Depression). A sampling interval of 100 m was used because of the spacing of reflectors
in the original seismic-reflection records (Table 1).
Shallow turbiclite layers are easier to image acoustically than deep turbidite layers. The
distribution of samples is skewed towards the shallow layers and hence the sample
distribution curve is not "normal" (see colunm "N" of Table 1). The statistics presented in
Table 1 are still valid and enlightening, however, because trends between acoustic tilt and
depth are evident within both block "1" and block "2" and between the two blocks. In
19
TABLE 1
STATISTICAL ANALYSIS OF TILT MEASUREMENTS FROM EASTERN BLOCKS1&2
(assuming an interval velocity of 1.8 km/sec)
BLOCK 1 BLOCK 2
Depth of N X S Range Depth of N X S Rangereflector (number (mean) (standard of tilts reflector (number (mean) (standard of tilts(m) of samples) deviation) () (m) of samples) deviation) ()
0-100 9 1.59 0.37 1.08-2.26 0-100 3 3.62 0.59 3.07-4.24
100-200 7 2.55 0.85 1.87-3.96 100-200 6 5.02 0.97 3.87-6.19
200-300 5 3.64 0.68 2.83-4.38 200-300 4 4.34 0.69 3.76-5.33
300-400 5 4.54 084 4.05-6.03 300-400 5 7.01 1.20 5.75-8.85
400-500 3 5.15 0.43 4.78-5.62400-600 4 7.68 1.28 6.57-9.52500-600 3 5.70 0.83 4.80-6.44
> 600 3 6.75 0.26 6.47-6.99 > 600 1 11.02
All Block 1 Data All Block 2 Datanumber = 35 number = 23mean = 3.60 mean = 5.8752=3.25 S2=4.21
SS 110.56 SS = 92.55
All DataF-TEST = 1.294T-TEST = 2.337
20
general, the mean tilt values increase with increasing depth intervals (Figure 8). Ranges of
tilt values, based on data gathered from different locations on each block (Figure 7) using
various types of seismic-reflection equipment, are shown in Table 1. Despite the variation
in data sources, the scatter of tilt values within the 100 m sampling intervals is low
Two statistical tests are used to compare the data in eastern blocks "1" and "2".
Application of the "F-test" shows at a geater than 95% confidence level that the variances
of the two data sets are equal (F=l.291; Fcrit=2.1). Results of the two-tailed "T-test"
indicate that there is a significant difference (greater than 95% confidence level) between the
mean values of blocks "1" and "2" (T=2.337; Tcrit=2.Ol). Together, these findings
support the observation that the tilt of individual blocks increases with distance away from
Cascadia Depression, the central spreading axis. Also, the equivalent variance of tilt values
in "Block 1" and "Block 2" is consistent with the idea that both were formed by the same
controlling process.
Fault angles and orientations in Cascadia Segment: comparison with models of slow-
to-intermediate spreading mid-ocean ridges
A classification scheme, derived from terminology developed by Macdonald (1986 and
1982) to describe the FAMOUS and TAG sites on the Mid-Atlantic Ridge, is used here to
compare the morphology of Cascadia Segment with mid-ocean ridge spreading centers.
The classification of geomorphologic features found commonly at seafloor spreading
centers is based upon: (1) fault throws; (2) dip of faults (to or away from the spreading
axis); (3) direction of relative motion on the faults (normal or reverse). The descriptive
terminology as it is applied to the Cascadia Segment is shown in Figure 9.
Note that the thick sediment cover on top of the bounding blocks tends to mask the
basement topography of Cascadia Segment. Although the surface expression of tectonism
at the Cascadia Segment is not perfecdy analogous to less sedimented ridges, the same
general processes probably control both types of spreading areas. Therefore, therelationship between faulting, back-tilting and crustal extension in Cascadia Segment lend
insight into how ridge-crest topography is created at other sites.
The "inner floor" is the relatively flat and deep central zone where crustal accretion and
crustal extension is maximum. The depth of the inner floor of Cascadia Segment varies
between 300 m and 560 m below the level of the flanldng inner walls. Corrected for
sediment thickness within Cascadia Depression, the basement of the inner floor ranges
between 500 m and as much as 1060 m below the level of bounding fault scarps.
According to the scheme of Macdonald (1982), this rift valley depth puts Cascadia Segment
into the slow-to-intermediate-spreading category. The proposed spreading half-rate of
2
C/)
C) 4C)
C)C) 6
8
10
1-iL
21
DEPTH (m)
C C C C C CC C C C C C-Lr CI
I I I I CC C C C C CC C-
C C CLt
0 BLOCK 1
+ =BLOCK2
Figure 8: Depth (100 m intervals) of reflectors versus tilt.
A
3A1
31 25 24 17' °
50 40 30' 20' 10' 6 10' 20' 30 40 50km
W. rift Western Inner Eastern E. riftmountains o terrace floor terrace mountains.E Lii
Lu
V.E.=5
Figure 9: Fault dips and morphotectonic terminology as applied to Cascadia Segment.
Vertical exaggeration is 5: 1. This classification scheme it based on morphology
used to describe the FAMOUS and TAG sites on the Mid-Atlantic Ridge
(Macdonald. 1986 and 1982). The terms used here include: inner floor, inner
walls, terraces, outer walls and rift mountains. The inner floor is the relativelyflat and deep central zone where crustal accretion / crustal extension is maximum.
Tlie'inner walls are the faults that bound the inner floor. The 'terraces
correspond to the topography created through block rotation and dip-slip motion
as extension occurs away from the central axis. The "outer walls" are defined as
the transition boundary between the relatively rugged terraces and the more
topographically subdued rift mountains.
k)
23
approximately 3 cm/yr clearly qualifies Cascadia Segment as an intermediate rate spreading
center (Embley et aL, in press; this paper). However, Macdonald (1986) has shown that
for spreading rates over 1.7 cnl/yr, a rift valley greater than 300 m deep can occur within
20 km of a triple junction or transform fault. Therefore, the "anomalously" deep rift valley
at Cascadia Segment does not contradict its calculated intermediate spreading rate.
The"inner walls" are the block faults that bound the inner floor. The slopes of the
inner walls at Cascadia Segment are 26° to the east and 20° to the west: gentler than those
reported from the slow-to-intermediate spreading Mid-Atlantic Ridge (Macdonald &
Atwater, 197S) and the slow-spreading Cayman Trough (CAYTROUGH, 1979). These
mid-ocean ridge escarpments typically have slopes of 50° to vertical.
There is a six degree disparity between east and west inner wall slope values. In
addition, the western inner wall is wider than the eastern inner wall (western inner wall=
4.7 km; eastern inner wall= 2.1 km). This may be the result of the rotation of tectonic
elements and a resulting decrease in slope of the western inner wall toward the northwest
corner of Cascadia Depression. The curved trend of Cascadia Depression's western wall
proximal to the northern bounding transform fault is apparent in SeaBeam bathymetry
(Figure 2). Fox & Gab (1982) attribute rotation of tectonic elements at ridge-transform
intersections to the transmission of shear stresses associated with the transform fault into
the spreading center domain.
The term" terraces" corresponds to the rugged topography created through block
rotation and dip-slip motion as extension occurs off the central axis. The angles of inward-
dipping faults in the Cascadia Segment terraces range from 1 i° to 31°: generally shallower
than those reported from slow-to-intermediate-rate spreading centers. Faults can be traced
at depth through the turbidite sediments. Therefore, the values of fault angles shown in
Figure 9 are true fault dips and not merely calculated from exposed scarp measurements.
The tops of the outer flanking escarpments (greater than 20 km from the central axis) are
less steep than the faults that formed them. Seismic-reflection records show that the
sediments on top of these blocks have been subjected to slumping suggesting that mass
wasting processes have tended to decrease the local escarpment slopes within the Cascadia
Segment terraces.
East-west topographic asymmetly is evident in the terraces. The extension in the
Cascadia Segment has resulted in a 5.7 km difference between the length of east (23.6 km)
and west (29.3 km) terraces. The expression of extension differs on either side of
Cascadia Depression. The western terrace is relatively smooth in contrast to the jagged
morphology of the eastern terrace. Extension of the eastern terrace has been accommodated
by rotation of three large, discrete blocks. The western terrace has extended through block
24
rotation and a large number of normal faults with smaller throws.
Morphological asymmetry between opposing flanks of the Mid-Atlantic Ridge has
been attributed to short-term (on the order of 106 years) asymmetric seafloor spreading
(Macdonald,1986). Asymmetric spreading for crust 0-5 m.y. in the Atlantic has been
reported at 45N (Loncarevic & Parker, 1971), 26°N (McGregor & Rona, 1975) and 6S
(van Andel & Heath, 1970). Assuming a opening half-rate of 3 cm/yr at Cascadia
Segment, five million years of asymmetric spreading could create topographic asymmetrycrust up to 150 km off-axis.
Morphologic asymmetry and its relationship to seafloor spreading rate varies as
reported from the Mid-Atlantic Ridge. Eberhart (1987) concludes that faster spreading
towards one flank at TAG sites 1 through 3 has resulted in a wider terrace in that direction.
Conversely, at TAG, narrower terraces coincide with slower spreading. The same
relationship between spreading rate and terrace width is evident at latitude 36°50'N on the
Mid-Atlantic Ridge ("Narrowgate" site; Macdonald, 1977). Using similar reasoning, faster
short-term spreading may have produced the wider and gentler slope of the Cascadia
Segment's western terrace whereas the relatively narrow, rugged eastern flank may be the
result of slower spreading. On the other hand, the disparity between the widths ofeastern
and western terraces at latitude 36°30'N on the Mid-Atlantic Ridge ("FAMOUS") does notcorrelate with asymmetric spreading rate (Ramberg & van Andel, 1977).
The "outer walls" are defmed as the transition boundary between the relatively rugged
terraces and the more topographically subdued rift mountains. At the Cascadia Segment,
the outer walls mark the transition between inward-dipping normal faults of the terraces and
the flattened topography of the rift mountains. One obvious mechanism of muting the
topography within the rift mountains in the Cascadia Segment is the occurrence of normalfaults that dip away from the central axis of extension (Figure 9). The outward-dipping
low-angle faults locally correlate with pre-existing block boundaries (i.e. the former "tops"of rifted blocks).
Tilt measurements of turbidite reflectors about central axis
The topographic development of the Cascadia Segment has been, in general, an
asymmetric process. Figure 10 shows the variation in amount of turbidite tilt within
individual fault blocks. A model of topographic evolution east and west of Cascadia
Segment can be developed through examination of turbidite and underlying basement tilt
versus distance from the central axis.
Tilt values in the western inner floor and inner wall (less than 10.0 km off-axis) vary
from -4.2° to 6.1° at Cascadia Segment:. A similar range of tilting (from -3.7° to 6.8°) was
25
Figure 10: Morphotectonic terminology and plot of tilt values in Cascadia Segment
assuming turbidites originally horizontal. Shown above the cross-sectionalprofile is a plot of turbidite layer tilt versus distance from the axis of spreadingbeyond about 3 km of the central spreading axis. Each data point marks theposition closest to the central axis (X 0 km) where that tilted layer is observed
on seismic-reflection records. Positive and negative values on the vertical axisrepresent those layers that tilt away from and toward the central axis,
respectively. "Hidden" tilts are those turbidite tilt values that are not seen awayfrom the central axis The envelopes of "hidden" tilts are shaded.
H. It ....S. '.. I
is.
:.,. i1'' '
:
SI
I
S A 4.1.1
..
I
I
:
'
10 20 30 4050km50 40 30 20 10
Inner Eastern E. riftmountaiflS
W. rift Westernfloor terrace omountaifls terrace
Ui
L
27
measured within eastern inner floor and inner wall, less than 7.0 km away from Cascadia
Depression's central axis. The high degree of tilt seen in inner wall and inner floor
turbidites is consistent with a model in which turbidite flows are deposited within a basinbeing uplifted and spreading away from a narrow extensional zone. The turbidite layers are
deposited during and shortly after the accretion of the underlying crust and are
subsequently uplifted and tilted as the crust moves away from the rift valley floor.
There are two trends of tilt within the eastern and western terraces: (1) the envelope of
positive tilt values; and (2) the negative tilts that are found at the bases of fault blocks
(Figure 10). The positive tilts are bound by distinct upper and lower trends. The
maximum tilt values correspond to the deepest acoustic layers and therefore best define the
tilt of the underlying basement. The data indicates that, in general, the basement of the
terraces increases its tilt away from the central axis in both the east and west directions.
The highest tilt achieved within the western half of the inner wall and inner floor (6.1°) is
slightly lower than that seen in the eastern inner wall and inner floor sequence (6.8°). In
the western terrace the basement rotates approximately 6° within 10 km of the central axis.
The basement rotates from 6.8° to 12.8° Within the entire eastern terrace, or 6° over 23 km.
Fault blocks on slow-spreading ridges are typically tilted 5° to 15° away from the
spreading axis (Atwater & Mudie, 1973; Macdonald & Luyendyk,1977). Macdonald
(1982) reports that some major blocks such as those bounding the outer walls of the rift
valley may be back-tilted in excess of 30°. Tilts of 3° to 5° and occasionally up to 10°
appear to be typical of intermediate-to-fast-spreading crust (Klitgord & Mudie, 1974).
The increase in minimum positive tilt values out to at least 50 km from the axiswithinboth the east and west terraces supports the contention that the blocks undergo additional
rotation away from the axis during the rifling process. According to a model proposed byDavis & Lister (1977) the blocks will 'freeze in "after a state of isostatic equilibrium is
reached, wherein the crust and lithosphere are strong enough to support any remainingnon-isostatic load. Off-axis extension through fault-block rotation and normal faulting are
two modifications to the "frozen in" topography within the terraces of Cascadia Segment.
"Hidden" tilts are those turbidite tilt values that are seen away from the central axis
(Figure 10, shaded region). The east-west difference in the envelope of "hidden" tilt valuesdemonstrates the asymmetry of block rotation in the terraces. The east-west asymmetry ofthe "hidden" tilt zones is another expression of the asymmetrical extension at CascadiaDepression.
Negative tilt values in the terrace areas are turbidite sediments that have been tilted
toward the central axis owing to their juxtaposition against active bounding faults (dragfolds). The sense of folds confirms that they abut active normal faults.
4.
Data from Cascadia Segment is examined and compared with three models proposed
by Macdonald & Atwater (1978) to account for the transition from the rugged rift valleyand terrace regime into the relatively smooth rift mountains on the Mid-Atlantic Ridge. Themodels, shown in Figure 11, include: (a) back-tilting of rift valley and terrace walls toapproximately horizontal within the terraces; (b) reverse faulting along pre-existing inward-dipping faults in the rift mountains; and (C) movement along outward-dipping normal faultsin the rift mountains. According to Macdonald & Atwater (1978), deep-tow studies of theMid-Atlantic Ridge indicate that the primaiy mechanism for transformation of the rift valleytopography is normal faulting along fault planes that dip away from the valley axis (Figure
1 1C). They find the other mechanisms, i.e. regional tilt (Figure 1 1A) and reverse faulting(Figure 1 1B), to be less important in the rift mountains of the Mid-Atlantic Ridge.
Figure 1 1A depicts a model in which the rift valley-to-rift mountain transitionat theFAMOUS site (37°N) on the Mid-Atlantic Ridge is accomplished by a modest rotation (5°-
9°) of the entire rift valley half-section as it passes into the rift mountains (Figure 1 1A;
Macdonald & Atwater, 1978; Macdonald & Luydendyk, 1977). This quantitative estimateof regional back-tilting is consistent with data from Cascadia Segment that demonstrates arotation of at least 6° occurs in the terraces. However, at Cascaclia Segment, the regional
slope from the rift valley center to the top of the outer walls (1 '-2°), as determined by
SeaBeam bathymetry, is significantly lower than that of the Mid-Atlantic Ridge at 37°N (5°-
9°). The disparity in regional slopes between Cascadia Segment and the Mid-Atlantic
Ridge, may suggest that the rotation of tectonic blocks at Cascadia Segment is not
necessarily the dominant mechanism of rift valley-to-rift mountain transition at this site.
Figure 1 lB depicts rift valley relief as being subdued by reverse motion along pre-
existing inward facing faults (Macdonald & Atwater, 1978). In this model, as new normalfaults are created near the rift valley , the relict normal faults are collapsed by reverse
motion at the outer wall and beyond into the rift mountains (Deffeyes,1970; Osmaston,1971; Harrison, 1974). Evidence from the outer walls and outermost terraces of CascadiaSegment may support the reverse faulting model. The possibility of reverse faulting and
reverse drag folding along the inward-dipping bounding fault of the eastern outer wall isconsistent with the very high positive tilt (highest = 12.8°) of the acoustic layers adjacent tothis fault (Figure 10). Tilted turbidite sequences are not discernable within the eastern
outer-wall block. However, the maximum tilt within the western outer wall, a value that
correlates with the tilt of the underlying basement, is 9.8°. This value fails well below the
maximum tilt trend established in the western terrace. Reverse faulting along the inward-dipping bounding fault of the western outer wall block could account for this lowmaximum tilt.
30
Macdonald & Atwater's (1978) third model, depicted in Figure 1 1C, proposes anoverprinting of the nft valley topography by normal faults that dip away from the valley
axis (Macdonald & Luydendyk, 1977; Needham & Francheteau, 1974). According to this
model, these may be new fault planes or reactivated faults that previously had small
offsets. These faults obtain much of their throw beyond the outer wall and thus
accomplish the transformation of rift valley into rift mountains. The outer walls / rift
mountains of Cascadia Segment are cut by outward-dipping faults (Figure 9). The senseof motion on these faults is determined in two ways: (1) drag-folding along outward-
dipping normal faults has resulted in the negative and shallow positive values of turbidite
tilt in the eastern rift mountains (figure 10); and (2) drag-folded turbidite-derived layers,
deposited within a former course of Cascadia Channel, record motion along shallow
outward-dipping normal faults in the eastern rift mountains (X=41 km, 44 km. Figure 10;
and Figure 12). The seismic-reflection profile shown in Figure 12 reveals what may be a
deeply buried block on the westernmost flank of Cascadia Segment. This block may havebeen dropped down along an outward-dipping fault.
Turbidite tilt data and fault orientations demonstrate conclusively that the muting of
topography from the rift valley to the rift mountains of Cascadia Segment is accomplished
by: (1) movement along outward-dipping normal faults in the rift mountains, and (2) back-
tilting of tectonic blocks within the terraces. The possibility ofreverse motion along pre-
existing inward-dipping faults in the rift mountains is not refuted by tilt data from Cascadia
Segment. These conclusions are in partial agreement with the observations from the Mid-
Atlantic Ridge which indicate that the primary mechanism for transformation of the rift
valley topography is normal faulting along fault planes that dip away from the valley axis
(Macdonald & Atwater, 1978). However, analysis of turbidite sequences atop tectonic
blocks at Cascadia Segment indicates that regional tilt is another significant factor in the
transition from rift valley-to-rift mountain topography.
The structure of Cascadia Segment is the result of long-term, active seafloor
spreading. The tectonic evolution of Cascadia Segment is revealed by examination of
tilted turbidite sequences lying atop the back-tilted blocks. The variation in tilt values with
distance away from the Cascadia Segment's central spreading axis demonstrates the
following:
(1) At Cascadia Segment, fault escarpment slopes are generally shallower than those
reported from the Mid-Atlantic Ridge (Macdonald & Atwater,1978) and Cayman
Trough (CAYTROUGH, 1979);
32
(2) The high degree of tilt achieved within the inner floors / inner walls at Cascaclia
Segment is consistent with a model in which turbidite flows are deposited in an
actively rifting basin;
(3) Off-axis extension through further fault-block rotation (at least 6°) and normal
faulting are two modifications to the topography within the terraces of Cascadia
Segment;
(4) The asymmetric expression of extension between the eastern and western flanks
of Cascadia Segment may be attributable to short-term (on the order of 106
years) asymmetric seafloor spreading. Extension of the eastern terrace has been
accommodated by rotation of three large, discrete blocks whereas the western
terrace has extended through block rotation and a large number of normal faults
with smaller throws;
(5) The muting of topography from the rift valley to the rift mountains of Cascadia
Segment is accomplished by two mechanisms including back-tilting of
tectonic blocks within the terraces and movement along outward-dipping normal
faults in the rift mountains. The possibility of reverse motion along pre-existing
inward-dipping faults in the rift mountains is not refuted by tilt data from
Cascadia Segment.
SMALL-SCALE STRUCTURE WITHIN A SEDIMENTED EXTENSIONAL
BASIN: CASCADIA DEPRESSION
The gross morphology of Cascadia Segment has been examined and attributed to long-
term seafloor spreading from a central axis. The present spreading axis within Cascadia
Segment is Cascadia Depression, a small rhomboidal basin. The short-term (on the orderof iø - lO years) vertical and horizontal tectonics within Cascadia Depression wereinvestigated using single-channel and 3.5-kHz reflection, SeaMARC II side-scan sonar,and piston core data. Also, because Cascadia Channel has been a source of turbidite
sediment to Cascadia Depression, it is possible to calculate extension of the basin over thepast 12,000 years using a method described by Davis & Lister (1977).
In order to better define the subsurface processes occurring at Cascadia Depression, a
sediment isopach map has been constructed. Figure 13 shows the contoured minimum
depth to acoustic basement using values from single-channel seismic-reflection recordstaken within Cascadia Depression. The data consists of ten tracklines that are orientedsemi-parallel to the short axis of the basin, two tracklines that cross the northwest corner ofthe basin, and a singular south-to-north trackline across the central zone of extension within
Cascadia Depression (Figure 14). The available data base is sufficient to contour the basin
area with reasonable certainty. The greatest detail, however, is achieved in the central partof the basin where trackline coverage is densest.
A relatively thin cover of sediment (less than 250 m) is found throughout the centraland northeast sections of Cascadia Depression. The thin sediment cover corresponds in ageneral way to a wedge-shaped topographic area delineated on Figure 2 by the 3500 mdepth contour. The 0" isopach is centered on the "V"-shaped topographic high at43°46.5'N / 128°41 .5W (Figure 13). This central area of high acoustic basement can beconsidered the present focus of extension in Cascadia Depression.
The slope of acoustic basement away from the central domed area within Cascadia
Depression is calculated taking into account bathymetric depth and depth to basement.There is a great disparity in basement slope between the eastern and western halves of thebasin. The gentler slope is toward the west (average = 4.2°) whereas the eastern slope ofacoustic basement is relatively steep (average = 11.5°). This asymmetry of basement slopebetween the west and east flanks of Cascadia Depression is consistent with data presentedin the previous section. Asymmetry of basement slope seems to be established in the
-43° 50'
3; (/Q
-43° 40'
DLfl
rj
Figure 13: Isopach of minimum sediment thickness within Cascadia Depression.
Contour Interval = 100 m; Shading Interval = 200 m.
35
__)(ID cx:(N
Figure 14: Plot of ships tracks within Cascadia Depression showing seismic-reflection
data navigation lines. NOAA Ship Surveyor tracks are solid lines. RJV S.P.
Lee tracks are dashed lines. RJV Melville track is a dotted line. NOAA Ship
Discoverer track is a dot-dashed line.
36
central rifting basin and is propagated outward onto the terraces until isostatic equilibrium isreached.
Another notable feature of the isopach map is the great depth to acoustic basementat
the northwest corner of Cascadia Depression (Figure 13). This corner of Cascadia
Depression may be analogous to the deep "nodal" basins found at ridge-transform
intersections (Fox & Gallo, 1984). The depth to acoustic basement at the southeastern
corner is unavailable because of the lack of data there. More seismic-reflection profiles are
needed from the southeast corner to test the idea that the corners of Cascadia Depression arestructurally analogous to ridge-transform intersections.
Cross-sectional profiles across Cascadia Depression
In order to gain some perspective on the overall structure of Cascadia Depression, aseries of NW-SE stacked profiles were generated from the SeaBeam data. These are
shown in Figure 15 as viewed from the southeast corner of the basin. The most noticeabletopography on the eastern half of Cascadia Depression is the dome-like feature whose
southernmost extent is evident in the third cross-sectional profile from the bottom (labelled
"doming" on Figure 15). The linear trend of the doming is parallel to the long axis of the
basin. The feature seems to "spread out" towards the north in a general "V" shape. Thisreflects the shape of the subcrustal "doming" seen on the isopach map (Figure 15).
The western half of Cascadia Depression is shallower and shows much more fine-
scale topographic variation than the eastern half (Figure 15). The topography here consists
of turbidite-derived sediments that have been faulted upward toward the western inner wailin a series of step-like faults. The faults have throws of 5 m to 168 m. The variation in
turbidite tilt within the largest (westernmost) sediment bench is -O.T to 3.5°.
The local slope of Cascadia Depression's floor is the result of interaction between two
components: rate of movement along faults within the basin and the rate at which turbidite-
derived sediments can bury the faults. At present, relative uplift of the topographic highs
(or relative sinking of topographic lows) in Cascadia Depression is occurring at a higher
rate than sediments are being delivered to the basin. Core 6609-19, taken within the deeper
part of the basin (Figure 2), contains only three turbidite sequences within the 86C cm ofsection younger than 6,600y (Duncan, 1968). This suggests that the recent number of
turbidite flows into Cascadia Depression has been small. The southeastern quadrant ofCascadia Depression has apparently acted as a local catchment basin for turbidites travelling
down Cascadia Channel (Figure 15). These relatively thick turbidites (3 sequences in 860cm of section) are probably of limited areal extent. Furthermore, Duncan (1968) found Mt.
Mazama ash in turbidite sequences both up and down channel from Cascadia Depression
37
V.E.10
South
Figure 15: Isometric view of seven topographic cross-sections within Cascadia Depression.Spacing between the seven cross-sections is 2 km and they are alignedperpendicular to the long axis of the basin. Vertical exaggeration = 10.
38
and concluded that the floor of the basin has subsided 120 m. Therefore, the floor of
Cascad.ia Depression has dropped at an average rate of about 1.8 cm/yrover 6,600 years.
Classification of acoustic facies using 3.5-kHz reflection records
In order to examine the recent tectonics of Cascadlia Depression, 3.5-kHz echo-
character types have been categorized. Four 3.5-kHz echo-character types are found within
Cascadia Depression. These include: (1) sharp bottom echo with parallel sub-bottom
reflectors; (2) irregular overlapping hyperbolic reflectors; (3) sharp bottom echo with
parallel sub-bottoms that exhibit sediment thickening against faults (growth faulting); and
(4) sharp flat bottom echo with undulating sub-bottom reflectors (after Damuth, 1975 and
1980; Damuth et al., 1983; See Figure 16). The distribution of these four echo-character
types within Cascadia Depression is shown in Figure 17. The data base used in this
categorization are the same as those used in preparation of the isopach map (Figure 14).
Based on the structural setting of Cascadlia Depression, the present base level of a deep-sea
channel, it is assumed that the echo types reflect various degrees of tectonic deformation of
turbidites.
Layered turbidite-derived sediment gives sharp bottom echoes with parallel sub-bottom
reflectors (Figure 16, line AX). This echo-character type is prevalent in the shallower
western half of Cascadia Depression where step-like faults cut turbidite sediment.
Likewise, turbidite-derived sediment is found in the northwest corner of the basin. Thesediment in this corner is cut by steep faults that are oblique to the regional direction of
spreading. Growth faulting is not evident in these turbidite sequences implying that
turbidite sedimentation and active faulting are not penecontemporaneous in this half of the
basin.
Irregular overlapping hyperbolic reflectors are associated with morphology controlled
by acoustic basement that either crops out or is buried by thin conformable sediments
(Figure 16, line BB'). The hyperbolic reflections in the middle of Cascadia Depression are
centered on the topographically high area whicl:, in turn, coincides with acoustic basement
doming. A submersible investigation of the domed area in Cascadia Depression did not
locate extensive outcrops; however, small, sediment-covered step-faults with subtle
outcrops were present (R. Embley, pers. comrn).
The 3.5-kHz reflection records from eastern Cascadia Depression reveal sharp bottom
echoes with parallel sub-bottoms that exhibit sediment thickening against faults ("growth
faulting"; Figure 16, line CC'). Growth faults are characteristically found wherever
unconsolidated sediment is deposited in basins that are actively growing in breadth and
depth. Sediment thickness is greater on the down-dropped side of the active fault. The
Uwest
6west.
39
B'east
1
-I,.:.
1 1 )!iIl
'-v
:' :';
Diieast
Figure 16: Examples of four 3.5 kHz echo-character types used in this study from profileswithin Cascadia Depression (based on Damuth, 1975 and 1980; Damuth etal..1983). Location of example records is shown in Figure 17.
Line AA' corresponds to layered turbidite-derived sediment chat give sharp bottom echoeswith parallel sub-bottom reflectors
Line BB' corresponds to irregular overlapping hyperbolic reflectors associated withmorphology controlled by acoustic basement that either crops out or is buried bythin conformable sediments
Line CC' corresponds to sharp bottom echoes with parallel sub-bottoms that exhibitsediment thickening against faults ('growth faulting").
Line DD' corresponds to sharp flat bottom echoes with undulating sub-bottom reflectors.This type of surface reflection may be associated with bedforms from a turbulentflow regime, tectonic deformation of sediment, or sediment that has undergonedifferential compaction, or some combination of these.
43050
/ J( %2*%4\(\
)
"IIR / I III' 'i.
I //([-'
lii:
I 'j7/f \.i i
I t - -. . I S - I_ . I I.
I \ 'c :--/1W I'jI ( \ \ / \_II-I (111
: :--
-_ -';'. ; /
.:.:...:...:.::.::II
.".-
/
\.'- '4.
Figure 17: Distribution of four 3.5 kHz echo-character types within Cascadia Depression
(based on Dantuth, 1975 and 1980; Darnuth et aL, 1983). Examples of echo-
character types are shown in Figure 16. Lines AA', BB, CC', and DD'
designate where examples of echo-characters shown in Figure 16 are located.
41
sediment thickness is consistently greater on the western side of the individual faults. Thisindicates that the blocks on the east of these growth faults are being uplifted relative to thewestern side. This agrees with the expected sense of motion on the eastern "half' of abasin that is being extended and uplifted away from the central axis. The limited areal
extent of surficial growth faults indicates that active faulting with concomitant
sedimentation has been restricted to the eastern half of Cascadia Depression in post-glacialtimes.
The southeastern corner of Cascadia Depression shows sharp flat bottom echoes withundulating sub-bottom reflectors on 3.5-kHz records (Figure 16, line DD'). This type ofsurface reflection may be associated with bedlorms from a turbulent flow regime, tectonic
deformation of sediment, sediment that has undergone differential compaction, or perhapssome combination of these factors (W. Normark, pers. comm.; C. Lister, pers. comm.).The exact origin of this echo-character type at Cascadia Depression is unclear.
Fault trends, determined using 3.5-kHz and single-channel seismic-reflection records,and SeaMARC II side-scan sonar data, are shown in Figure 18. The orientation of faultslocated on the eastern side of Cascadia Depression is about O30. Recall that these are theonly faults within Cascadia Depression that exhibit surficial growth faulting. This indicatesthat the eastern side of the neovolcanic zone is the only region within the basin where
turbidite sedimentation is occurring simultaneously with active fault movement. The long-term axis of spreading is slightly east of north, as reflected in the orientation of fault scarpsin Cascadia Segment (Figure 2). Therefore, the faults on the eastern half of Cascadia
Depression are about 30 degrees oblique to the axis of long-term spreading. The trend ofthe eastern faults probably reflects the short-term direction and mode of seafloor spreadingwithin Cascadia Depression. Faults of this orientation could stem from a relatively higherextension rate toward northern Cascadia Depression. A higher rate of opening toward thenorth is supported by several features within the basin: (1) the "spreading out" character ofthe bathymetry seen in cross-sectional profiles; (2) the shape of the subcrustal doming(Figure 13); and (3) the triangular wedge of hyperbolic reflectors seen in Figure 17.
The dominant trend of faults in the region west of the extensional axis is slightly westof north (353°-357°). Unlike the faults located in the eastern half of Cascadia Depression,the western step-like faults are located above the level of turbidite deposition withinCascadia Depression. The orientation of the western faults may reflect a relatively older
extensional regime, subparallel to the regional strike of fault block escarpments.The strike of the western faults"bends" gradually towards the northwest corner of
41A
Figure 18: Fault orientations based on SeaMARC II side-scan sonar data, 3.5-kHz
reflection data and bathymetry within Cascadia Depression. The 3.5-kHz
tracklines are those shown in Figure 14. The location of faults observed on the
3.5-kHz are shown by solid line segments (teeth point downward). Dotted
lines show inferred fault trends based on 3.5-kHz records. Inferred fault trends
between teeth are based on: (1) fault throw values; (2) measured fault dips from
single-channel seismic records; and (3) interpretation of detailed bathymetry.
The dot-dashed fault trends are taken from SeaMARC H side-scan sonar data
and correlated with exact fault locations from seismic records.
.4 L .4 C
28
43
Cascadia Depression, The observed trend of faults in the northwest corner ranges between120°-145° (NW-SE), about 35 - 60 degrees oblique to the regional axis of seafloor
spreading. Seismic-reflection records indicate that the faults cut through the sediment at
relatively steep angles (greater than 60°: Figure 19). The amount of throw on these faults issmall and the associated horizontal extension is minor.
The northern boundary of Cascadia Depression can be divided into distinct zonesbased on fault orientations, SeaMARC II side-scan sonar, and SeaBeam bathymetry data.
The northern boundary is divided into two areas: (1) between 128°44'W and 128°48'W
longitude ("western north wa1l); and (2) between 128°39'W and 128°44'W ("eastern north
wall"). Faults do not "bend gradually" into the trend of the transform fault at the westernnorth wall of Cascadia Depression. Rather, the faults truncate the western north wall athigh angles (Figure 18). The western north wall, as seen in SeaMARC II side-scan sonar
(Figure 20) and SeaBeam bathymetric data (Figure 18), is highly reflective, steep, and
scalloped in nature. Degradation of the eastern north wall of Cascadia Depression is
inferred from hyperbolic echoes seen on 3.5-kHz reflection records. The degeneration of
the slope is evident in SeaMARC side-scan sonar data as numerous en echelon fault scarps
located at the base of the eastern north wall (Figures 18 and 20). The trend of these faults
(1110 to 121°) is subparallel to both the orientation of the boundary itself (108°) and the
regional transform fault azimuth (118°). The degradation of the eastern north wall has
resulted in a slope much gentler than that of its western counterpart.
At ridge-transform intersections one flank of the ridge is accreted against the aseismic
limb of a fracture zone without experiencing strike-slip related tectonism (Fox & Gallo,
1984). In the case of Cascadia Depression, the eastern "half' of the north wall shouldexperience no strike-slip motion. Therefore, the gentle slope and en echelon faulting alongthe eastern north wall may be the result of mass wasting along a passive boundary. Thewestern north wall of Cascadia Depression, on the other hand, should accommodate
strike-slip motion. The steep, scalloped character of the western north wall supports theidea that it is tectonically active.
The active portion of the transform at the East Pacific Rise / Tamayo Transformintersection is a steep, sediment slope with scalloped scarp surfaces. The scarps are
probably produced by gravitational erosion that is triggered by strike-slip faulting of
underlying oceanic basement (Tamayo Tectonic Team, 1984). At Cascadia Depression,
the spatial correlation between truncating faults and the scalloping of the western north wall
may indicate that slope failure is triggered by movement along the truncating faults.
However, the high reflectivity of the western north wall on SeaMARC II side-scan sonardata suggests that this surface is sediment-free. The available seismic-reflection data cannot
ji ___i_,__ --
o0 0 ....
----------.;
I I
I.
I
-S
. - .. . ... .., - -
I -1 -
:-. .: ... . .: :.:'.: .:. -..:I T --SI_ i I
-_ A
J
VE- I
Figure 19: Seismic-reflection records showing approximate fault dips from the northwest
corner of Cascadia Depression. Vertical exaggeration is approximately 8: 1.
45
Figure 20: SeaMARC II side-scan sonar image of northern Cascadia Depression.(A) : Original SeaMARC II side-scan sonar image.(B): SeaMARC II side-scan sonar record with faults delineated. The northern
boundary wall of Cascadia Depression is divided into two areas: (1) between128°44'W and 128°48'W longitude ("western north wall"); and (2) between12839'W and l2844'W ("eastern north wall"). The western north wall ishighly reflective, steep, and scalloped in nature. The degeneration of theeastern north wall is evident as numerous en echelon fault scarps.
Figure 20
47
support or refute the idea that the northern wall of Cascadia Depression is free of sediment.
Further investigation is needed to resolve the fine-scale structure and tectonics of CascadiaDepression's northern wall.
SeaMARC II side-scan sonar data shows that the faults located about 5 km west ofCascadia Depression have a trend of 118°. This is equivalent to the azimuth of ridges andtroughs that define the Blanco transform fault zone as a whole. Therefore, at a distance of
5 km west of Cascadia Depression the bearing of faults parallels the azimuth of the
principal transform displacement zone or "PTDZ" (after Fox & Gab, 1984). The
calculated age of crust at this distance from the spreading center is about 165,000 y (half-spreading rate = 3 cm/yr).
The high density of seismic-reflection data across Cascadia Depression is the basis for
a quantitative estimate of spreading rate using a method described by Davis & Lister
(1977). Using fault throw values from 3.5-kHz reflection records, fault dip angles from
seismic-reflection data, estimates of turbidite sedimentation, and geometric relationships,
the amount of horizontal extension at the surface can be estimated.
Turbidite activity is a strong function of the glacial-interglacial-postglacial state of the
terrigenous source in this type of channel-fed basin (Davis & Lister, 1977). Duncan
(1968) has estimated that the rate of turbidite sediment accumulation in this area decreased
rapidly about 12,000 years ago. Prior to l2,000y, high rates of turbidite sedimentation
would have maintained a horizontal surface within Cascadia Depression, masking tectonic
deformation at the surface. Recall that piston core data indicates that the base level of
Cascadia Depression's floor was 120 m higher 6,600 years ago (i.e. the deepest part of thebasin was at approximately 3400 m depth). Assuming Duncan's (1968) estimated averagedrop rate of the floor (1.8 cm/yr) can be extrapolated back 12,000 years, then the"horizontal surface" which defined Cascadia Depression's floor at the end of the last glacialperiod would have been at a depth of about 3300 m. Figure 18 shows that, with the
exception of the major bounding faults (i.e., those faults with greater than 400 m ofthrow), all faults within the central part of Cascadia Depression (south of 43°48'N latitude)
are found at depths greater than 3300 m. Therefore, it was assumed for all calculations that
the surface displacement along the faults within the central part of Cascadia Depression
(with throws less than 400 m) is younger than 12,000 years.
Fault throw values are measured from 3.5-kHz records ("Surface offset" column of
Table 2). Assuming the faults have been actively displacing the once-horizontal surface ofthe basin for 12,000 years, an average vertical displacement rate is calculated over this
TABLE 2
VERTICAL AND HORIZONTAL COMPONENTS OF DISPLACEMENT RATE FROM SURFACE OFFSETS
Trackline Surface Vertical Fault Horizontal (north=1, offset offset dip displacement middie=2, (m) rate (°) rate
south=3) (crn/yr) (cni/yr)
1 17 0.14 19 0.41 1 45 0.38 13 1.62 1 15 0.13 20 0.34 1 25 0.21 20 0.45 1 10 0.08 25 0.18 1 5 0.04 21 0.11 1 7 0.06 20 0.16 1 5 0.04 30 0.07 1 10 0.08 23 0.20 1 155 1.29 30 2.24
----------------------------------------------------------- TflTAT. 7R
2 163 1.36 40 1.62 2 7 0.06 20 0.16 2 7 0.06 19 0.17 2 8 0.07 25 0.14 2 32 0.27 21 0.69 2 35 0.29 30 0.51 2 20 0.17 21 0.43 2 15 0.13 20 0.34 2 30 0.25 12 1.18 2 5 0.04 13 0.18
------------------------------------------------------------ TOTAL: 5.42
3 140 1.17 40 1.39 3 14 0.12 20 0.32 3 30 0.25 25 0.54 3 12 0.10 29 0.18 3 40 0.33 23 0.79 3 15 0.13 28 0.24 3 10 0.08 20 0.23 3 5 0.04 13 0.18 3 5 0.04 12 0.20
------------------------------------------------------------ TOTAL: 4.07
period. Thus, the "vertical offset rate" values (Table 2) obtained are simply the surfacefault throws divided by 12,000 years.
Figure 21 shows the western half of one of the three seismic-reflection profiles used inthese calculations. Fault dip (Table 2) was determined directly from the profiles using aninterval velocity value of 1.8 km/sec (Kuhn & von Huene et al., 1973) and a zero velocity-depth gradient because sediment thickness within Cascadia Depression is less than 800 m(Figure 13; Carison et al., 1986). Determination of fault dip using umnigrated seismic-reflection profiles was made more difficult because of diffractions which disrupt therecords (Figure 21). However, it was found that the faults cut through the entire section ofsediment, are low in angle (1Y-40°), and do not curve with depth. The measured anglesmatch well with the low regional fault dips (1 1°-31°; Figure 9). Using the simple geometricrelationship between the estimated vertical offset rate along each fault and the dip of thefault, the horizontal extension rate (averaged over the past 12,000 years) of each fault iscalculated (last column of Table 2).
The extension rates on the individual faults within each of the three tracklines areadded together and the total extension rate for each line is shown in Figure 22. The ratesobtained, 4.1 cm/yr, 5.4 cm/yr. and 5.8 cm/yr, match extremelywell with the expected fullspreadmg rate of approximately 6 cm/yr (Embley et al., in press). The extension withinCascadia Depression has apparently been accommodated along shallow faults. Clearly, thehorizontal extension rate is highly dependent upon the dip of the faults (i.e., in general, anyincrease in fault dip will serve to decrease the calculated horizontal rate of extension).However, at Cascadia Depression, it is found that fault dip values up to 40 degrees higherthan those measured (Table 2), will produce horizontal extension rates well within an orderof magnitude of the estimated spreading rate of 6 cm/yr.
A progressive decrease in opening rate from 5.8 cm/yr to 4.1 cm/yr is evident fromnorth to south, respectively (Figure 22). This discrepancy in estimated opening rates isprobably the result of: (1) "true" short-term asymmetric opening of Cascadia Depressionfrom north-to-south; (2) the masking of surface offset along faults towards the southernpart of Cascadia Depression by ponding of turbidites; or (3) a combination of (1) and (2).
Short-term asymmetric spreading may be documented at Cascadia Depression.
Averaged over 12,000 years, Cascadia Depression could have opened up at a higher ratetoward the north of the basin. Asymmetrical spreading may account for many features inthe basin including: the "V"-shaped trends in topography (Figure 15) and acousticbasement doming (Figure 13), and also the obliquity of some faults within the basin(Figure 18) to the long-term axis of spreading.
The method used to estimate horizontal extension rates within Cascadia Depression is
50
.
ti't l
St
i'1
SS
L'''5 1;1' 1
i
.1 a
U p4:. il.i.'. '': .tt!U
c.'J ;:.'
o
0 St
0 fS,'_. 5;' -. 'I
..,.,.... S
csJS
o :.Lt)
S
o
't1
TU : : :£
00
i:
I I..
t'4l .
'v....S .
I'..
00
:S ... ; 41
LU
'( Yq.:.'ti 0 50tçS
' t1 ".s v ,.. .
EI '
tk
?
:,
fri S tj
' &
5' I
11 4S s. 5U
; lSc \!-s I
:' :'.;\ 0
, '1I 1 "" .ji.d' "1
is5.. _iilt.> ittiv ::.
S
S . ,&.1\&, 1\tsI 1
1' ;i; q., ;. :.'' -. ..
PS 'v'.:. ,; . S
, -;5-f1 S
If
1
IL
SI /
Li.iSCSI 5islIt1
::
5Pr 1,'&; '. . -
I)
20.
0050
0
.0
0500
L0.
50
I . '.
S S5)
S (S r-1
5)
0,...
>.5
5-5.50,
51
a)(N (N
Figure 22: Map view of faults within Cascadia Depression and total spreading ratesindicated. Fault dips (top) were determined directly from seismic-reflection
records. Horizontal extension rates (bottom) were estimated usingfault throwvalues from 3.5-kHz reflection records, fault dip angles from seismic-reflectiondata, estimates of turbidite sedimentaiion,and geomethc relationships (Table 2).
52
contingent on the assumption that turbidite accumulation throughout the basin has been
insignificant relative to the fault motion over the past 12,000 years. However, this
assumption is probably not valid in the southeastern quadrant of Cascarlia Depression
where turbidite sediment accumulation has occurred within the past 6,600 years (Duncan,
1968). Therefore, the calculated honzontal extension rates at Cascadia Depression should
be anomalously low in this area. The values shown in Figure 22 support the idea that
sediment ponding has reduced the estimates of horizontal extension within the southeastern
quadrant of Cascadia Depression.
There is a disparity in the cumulative estimated horizontal extension rates between the
eastern and western sides of Cascadia Depression (Figure 22). The western "group" of
faults apparently accommodate a greater amount of extension on each of the three seismic-
reflection tracklines than their eastern counterparts. This east-west difference in extension
is, in part, a consequence of turbidite "masking" of fault offsets in the southeastern
quadrant of Cascadia Depression. Recent motion along the eastern boundary fault of
Cascadia Depression is suggested by the presence of growth structures at the base of the
fault (Figure 16C). Unfortunately, the amount of extension attributable to motion along
Cascadia Depression's eastern boundary fault cannot be calculated using the method
described above. The additional component of extension on the eastern boundary faultshould compensate for some (if not all) of the asymmetry in calculated horizontal extension
rates between the eastern and western sides of Cascadia Depression. Furthermore, any
supplemental extension will serve to increase the overall calculated horizontal extension
rates within Cascadia Depression.
Summary of the tectonic evolution of Cascadia Depression
The relationship between horizontal extension, vertical tectonics, and sedimentation of
Cascadia Depression is consistent with the idea that it is a center of seafloor spreading. The
temporal and spatial evolution of Cascadlia Depression as it rifts apart is revealed by
examination of acoustic basement morphology, determination of fault or itations andquantification of relative motion along faults. The following conclusions are made:
(1) Doming of the acoustic basement is occurring in the center of Cascadia Depression.
Subcrustal mechanisms that could account for the shallow acoustic basement in the
central basin are: (a) presence of sills; (b) uplift by volcanic doming along a linear
rift; or (c) diapiric intrusions such as serpentine;
(2) Asymmetry of basement slope seems to be established in the central rifling basin
and is propagated outward onto the terraces until isostatic equilibrium is reached;
53
(3) Extension within Cascadia Depression has apparently been accommodated along
shallow faults. Quantitative estimates of extension within Cascadia Depression
reveal spreading rates of 4.1 cm/yr, 5.4 cm/yr, and 5.8 cm/yr. The calculated ratesof extension support Embley et al.'s (in press) model in which Cascadia Segmentis a remnant 'trapped" spreading segment from an episode of rift propagation;
(4) Within Cascadia Depression, a progressive decrease in calculated opening ratefrom 5.8 cm/yr to 4.1 cm/yr is evident from north to south, respectively. This maybe the result of true short-term asymmetric opening of Cascadia Depression, or themasking of surface offset along faults towards the southern part of CascadiaDepression by ponding of turbidites;
(5) Short-term (on the order of iO years) asymmetric spreading may be occurring
within Cascadia Depression. Asymmetric spreading could explain the "spreading
out" character of bathymetry toward the northern depression, the irregular shape ofthe acoustic basement doming. the thangular wedge of hyperbolic reflectors in the
central depression seen in 3.5-kHz records, and the oblique orientation of growthfaults on the eastern side of the depression;
(6) The decrease in calculated opening rate towards the south may be the result ofsediment ponding in the southeastern quadrant of Cascadia Depression. Turbiditeaccumulation in this part of the basin has been insignificant over the past12,000 years (Duncan, 1968), contrary to one assumption of the method used to
calculate horizontal extension. Therefore, the calculated horizontal extension ratesshould be anomalously low in this area;
(7) Recent motion along the eastern boundary fault of Cascadia Depression is
suggested by the presence of growth structures at the base of the fault (Figure
16C). This additional component of extension should compensate for some (if notall) of the asymmetry in calculated horizontal extension rates between the easternand western sides of Cascadia Depression. Furthermore, any supplementalextension should increase the overall calculated horizontal extension rates withinCascadia Depression.
54
CONCLUSIONS AD PROPOSED FURTHER STUDY
The detailed interpretation of the morphologic and structural information derived from
SeaBeam bathymetric, single-channel reflection seismic and SeaMARC II side-scan sonar
data outlined in the previous chapters has provided important clues to the tectonic evolution
of this portion of the Blanco transform fault zone. A tectonic evolution of Cascadia
Segment is outlined in a model in which there has been a juxtaposition of several processes
occurring over the past several million years. In this model, a basin forms by seafloor
spreading, is filled with sediment and subsequently the crust is rafted away from the central
axis, through the terraces and out into the rift mountains. The predominance of
conformable turbidite sequences on the uplifted tectonic blocks at Cascadia Segment
indicates that uplift within and away from the central basin has occurred in a continuous
manner.
A progressive decrease in calculated opening rate over the past 12,000 years, from 5.8
cm/yr to 4.1 cm/yr, is evident from northern to central Cascadia Depression, respectively.
This may be the result of true short-term asymmetric opening of Cascadia Depression, or
the masking of surface offset along faults towards the southern part of Cascadia Depression
by ponding of turbidites. The high degree of tilt achieved within the inner floors / inner
walls of Cascadia Segment support the idea that turbidite flows are deposited into a basin
that is spreading away from a central axis. Asymmetry of basement slope, established in
the rifting basin, is propagated outward onto the terraces until isostatic equilibrium is
reached. Off-axis extension through further fault-block rotation (at least 6°) and normal
faulting are two mechanisms by which topographic relief is created within the terraces of
Cascadia Segment. Asymmetric seafloor spreading on the order of 106 years may account
for the asymmetric expression of extension between the eastern and western flanks of
Cascadia Segment. Extension of the eastern terrace has been accommodated by rotation of
three large, discrete blocks whereas the western terrace has extended through block rotation
and a large number of normal faults with smaller throws. The muting of topography from
the rift valley to the rift mountains of Cascadia Segment is accomplished by back-tilting of
tectonic blocks within the terraces and movement along outward-dipping normal faults in
the rift mountains. The possibiit of reverse motion along pre-existing inward-dipping
faults in the rift mountains is not refuted by tilt data from Cascadia Segment.
55
Models of crustal accretion at spreading centers: steady state versus non-steady state
rift valley formation
In light of the conclusions made about the tectonic development of Cascadia Segment,
further speculation about deep-seated processes occurring there can be addressed. The
proposed evolution of Cascadia Segment is compared to existing models of rift valley
evolution and crustal generation processes at mid-ocean ridges. This allows examination of
current models of topographic evolution at spreading centers, especially in the context of
steady state versus non-steady state (i.e. episodic) theories of rift valley development
(Kappel & Ryan, 1986; Macdonald & Atwater, 1978; Davis & Lister, 1977; Deffeyes,
1970, for example).
Atwater & Mudie (1968, 1973) reported that evidence from turbidite deposits and
benthic foraminifera from cores taken at the Gorda Rise indicate that faulted blocks were
uplifted along normal faults to form the rift valley rather than the valley floor being faulted
down. This model requires the inner floor to "ride out" passively into the terraces along a
series of normal faults in a steady-state manner. Several steady state models have been
proposed to explain rift valley generation on slow-spreading ridges including: (1) hydraulic
head loss and viscous drag in a narrow conduit of upward flowing asthenospheric material
(Sleep, 1969; Lachenbruch, 1973); (2' extension and necking of the lithosphere near the
axis (Tapponnier & Franceteau, 1978); (3) the disproportion in widths of the axial
extension zone and axial volcanic zone (Deffeyes, 1970; Palmasson, 1973; Anderson &
Noltimier, 1973); and (4) horizontal extensional stresses in a strong, brittle lithosphere that
thickens with distance from the ridge axis, the width of the axial valley being controlled by
plate thickness near the ridge axis (Morgan et al., 1987).
The morphologic resemblance between Cascadia Segment and the northern Endeavour
Ridge ("Middle Valley"), i. e. individual back-tilted blocks covered with turbidite
sequences that "fan out" away from a central depression, suggests that these two regions
are the result of similar deep-seated processes. In order to explain the evolution of
individual back-tilted turbidite-covered blocks at Middle Valley, Davis & Lister (1977)
modified Deffeyes (1970) steady-state model of ridge-crest topographic generation where
the material supply occurs over a broaaer region than extension (Figure 23). Cascadia
Depression's central zone of extension may be as wide as 6 km, if defined as where the
shallow faults that cut through the sediment extend through to the underlying acoustic
basement (i.e., about twice the width of the faulted basement shown in Figure 2').
Although Cascadia Depression's apparent zone of extension is wider than that proposed for
Middle Valley (Figure 23C, curve 1), the similarity of topographic profiles between the two
1 SJRF.8CE 11rEXTENSIONAL
2ATE
2
1
Figure 23: Mechanisms for the formation of axial-valley topography at slow-spreading
ridges (from Deffeyes, 1970). In each case, curves 1, 2, 3, and 4 represent
surface extensional velocity, material supply rate, integrated supply rate, and the
resultant topography, as functions of distance from the ridge crest axis.
(A) : The general case with the width of the supply zone greater than the surface
extension.
(B) : A model in which the crustal extension zone is narrow as suggested by deep-
tow magnetic data (Atwater & Mudie, 1973).
(C) : This model is similar to 23B, but with a rigid crust superimposed. The
formation, uplift, and outward rotation of normal-fault blocks from the axial
valley are a consequence of the formation of a rigid crust at the origin. Once the
blocks have reached a level of isostatic equilibrium, or once the crust and
lithosphere are strong enough to support any remaining isostatic load, the
blocks will "freeze in" as permanent features of the seafloor topography (from
Davis & Lister, 1977).
57
areas suggests that the general relationship between extension and crustal accretion at
Middle Valley also holds for Cascadia Depression. A discontinuous topographic profile
with roughly equidimensional blocks is caused by the addition of rigid crust to a model in
which the crustal extension zone is relatively narrow (Figure 23C). Rotation of blocks
away from the axis of crustal accretion occurs because of the change of material supply rate
with distance from the origin. Negative tilts, found within the inner floors / inner walls of
both Middle Valley and Cascadia Segment, may be the result of a non-steady state factor
such as the formation of a new axial valley where material supply is deficient (Figure 10;
Davis & Lister, 1977). A non-steady state nature of Cascadia Segment's rift valley is also
supported by the fact that the floor of Cascadia Depression has subsided 120 m in 6,600
years (Duncan, 1968).
Kappel & Ryan (1986) have proposed a non-steady state model of rift valley
development based on SeaMARC I side-scan sonar data from the Juan de Fuca Ridge
system (Figure 24). Kappel & Ryan (1986) use the term "elongate summit depression"
(ESD) rather than "rift valley" because the latter implies an exclusively tectonic origin for
the axial feature. Models representing both steady state and non-steady state origins for
ESD's are shown in Figure 24. Figure 24A depicts the ESD as a steady state feature that,
through time and in a conveyor belt fashion, transports crust (slice X, for example) upward
from the depression floor to the ridge flanks via faulting. Distinctly different is the episodic
model of Kappel & Ryan (1986) shown in Figure 24B. In this non-steady state model,
slice X, located at the summit of the crestal ridge at times Ti and T2, is collapsed
downward to become part of the ESD (at time T3). The ESD widens by further extension
during time T4; over the same relative time interval, slice Y moves away from the central
axis and down the crestal ridge. During phase Ti and T2 the ridge axis is a neovolcanic
province that may have a width many times broader than the axial extrusion zone (Figure
24B). Later, in phases T3 and T4, the axis becomes a dominantly tectonic province.
Kappel & Ryan's (1986) non-steady state model is in agreement with an early working
hypothesis of Normark (1976) that "volcanic extrusive activity at the rise crest and (or)
periods of extension and faulting are episodic". Clearly, volcanic extrusive activity at
Cascadia Segment is not as voluminous as that of spreading ridges hundreds of kilometers
in length. However, the model of Kappel & Ryan (1986) lends support to the idea that the
balance between extension and volcanism at Cascadia Segment may be episodic in nature.
Morphological evidence, such as the extreme depth of the inner floor and the predominance
of gentle fault dips in Cascadia Segment, indicates that extension is the more dominant
factor. Furthermore, the high subsidence rate of the inner floor (1.8 cnilyr averaged over
6,660 years) and the high calculated iorizontal extension rate (possibly greater than 5.8
A BTi Xr-
12
T 3
T4 t
C
T 1x
T2 AjT3T4xx
¶ IsoclYon
5. Refefence Piece of Seof1oc
Rift valley axisElongate summit depression)
Sulfides
2-; --'-z
f400m
5Km
Figure 24: Models of steady state and non-steady state rift valley formation and the
expected distribution of suffides within one stage of rift valley development.
(A) : The steady state model of rift valley topographic development (after Kappel&
Ryan, 1986). In this model, the crestal ridge cross-section does not change
significantly from Ti to T4, and slices X and Y follow the seafloor profile as
they move away from the ridge-axis. Slice X is faulted up from the elongate
summit depression (ESD) floor to become part of the flanks from T2 to T3, and
then 'unfaulted' during T4.(B) : Kappel & Ryan's (1986) non-steady state model of rift valley evolution. A
crestal ridge grows from Ti-to T2, the ESD collapses into the crestai ridge
during T3, and there is incipient growth of a new crestal ridge during T4.
There is not necessarily an equal amount of nine during each of these stages.
Note that the whole flank of the crestal ridge moves away from the ridge axis
intact as the ESD widens.
(C) : This schematic diagram integrates Stages T2 and T3 of the non-steady state
tectonic model (Figure 24B) with a model that predicts the disnibution of
sulfide deposits (Kappel & Franidin, in press). During this stage, the ESD is
wide and flcrured, and the 400'C isotherm is located within a thickened gabbro
cap. The ESD bounding faults provide the pathway for the hot fluids to reach
the seafloor.
59
cm/yr averaged over the past 12,000 years taking into account motion along the eastern
boundary fault) may indicate that the magma supply has waned in the recent past allowing
extension to become the current dominant mechanism at Cascadia Depression. If this is the
case, the present configuration of Cascadia Segment fits best between stages T2 and T3 of
the non-steady state model (Figure 24B). This scenario suggests that extension at Cascadia
Segment is going on regardless of the upwelling magma supply. The presence of
proximal, cold truncating transform boundaries may perturb the production of basaltic melt
in the rising asthenospheric wedge beneath the rift valley at Cascadia Segment (P. Fox,
pers. comm.; Fox & Gallo, 1982 and 1984).
Is there hydrothermal activity within Cascadia Depression?
Most seafloor hydrothermal deposits have been identified using deep-towed cameras
and submersibles while surveying mid-ocean ridge crests (i.e., RISE Project Group, 1980;
Ballard et al., 1981, 1982; Normark et al., 1983). Kappel & Ryan's (1986; Figure 24B)
non-steady state model of rift valley development provides the basis for explaining the
distribution of hydrothermal circulation and massive sulfide deposits based on regional
tectonic setting of northeast Pacific spreading centers (Kappel & Franidin, in press).
Confined hydrothermal circulation, restricted by impermeable sediment, results in high
upper crustal temperatures and intense sediment diagenesis, causing localized hydrothermal
deposits that are much larger than those found in unsedimented spreading centers (Lonsdale
& Becker, 1985; Holmes & Morton, 1986; Karlin & Lyle, 1986; Davis et al., 1987). The
conditions necessary to host hydrothermal activity at seafloor spreading centers include
(Kappel & Franidin, in press): (1) the availability of high temperature fluids (400°C;
Bischoff et al., 1983; Seyfried & Janecky, 1985); (2) focussing of fluids into conduits; (3)
transportation of fluids to the seafloor with a minimum of conductive cooling or mixing
with cold seawater; and (4) in order to preserve sulfide deposits, some form of covering,
such as sediment must be present (Kappel & Franidin, in press).
Recall that the present configuration of Cascadia Segment falls best between stages T2
and T3 of Kappel & Ryan's (1986) non-steady state model (Figure 24B). At this stage of
the non-steady state model (Figure 24C), volcanism has waned and the rift valley widens
by tectonic stretching. The model preseated in Figure 24C depicts the 400°C isotherm to
mirror the shape of a subcrustal magma chamber. In the case of Cascadia Segment and in
lieu of a distinct magma chamber, the insulating quality of turbidite sediment contained
within the rift valley basin (wherever greater than 300 m in thickness) is sufficient to
contain the heat necessary to attain the required temperature for formation of metalliferous
hydrothermal solution (after Kappel & Franidin, in press). The major bounding faults of
the rift valley should provide a pathway to the surface for hot and buoyant fluids.
An indirect indicator of hydrothermal activity at Cascadia Segment is the low
amplitude, flattened nature of the magnetic anomaly signature (Figure 3). The absence of
significant magnetic anomalies at sedimented spreading centers may be a positive indicator
of a closed system of hydrothermal activity. Levi & Riddihough (1986) propose that the
absence of magnetic anomalies at sedimented spreading centers is a result of hydrothermal
leaching of iron-titanium oxides, a consequence of a long fluid residence time in the basalt.
Longer residence time of the hydrothermal fluid should result in more complete alteration of
basalt than is expected at relatively more "open" (i.e., non-sedimented) hydrothermal
systems (Levi & Riddihough, 1986). Two other mechanisms have been put forward to
explain the flattened magnetic anomalies at sedimented spreading centers. Irving (1970)
and Vogt et al. (1970) propose that slower-cooling magma dikes / sills have larger grain
sizes than their extrusive counterparts, and thus correspondingly lower specific remanence
intensities and lower stabilities. The second model addresses the thermal conditions at the
base of sediment column where temperatures in the basalts may exceed the Curie points of
unoxidized titanomagnetites (Johnson & Atwater, 1977; Johnson & Hall, 1978; Marshall,
1978; Prévot et al., 1979). This situation would decrease the spontaneous magnetization of
the underlying rocks and reduce observed magnetic anomalies.
How crustal accretion may be occurring within Cascadia Depression
Sedimented spreading centers have become the focus of considerable interest as
possible modern analogs of some ancient sediment-hosted ore deposits. The mechanism(s)
by which crustal accretion is occurring within Cascadia Depression can be addressed by
comparing its small-scale seismic-reflection characteristics to other sedimented spreading
centers such as the Guaymas Basin within the Gulf of California, Middle Valley located at
the northernmost end of Endeavour Ridge, Northeast Pacific and Escanaba Trough at the
southern end of Gorda Ridge, Northeast Pacific. Sediment-hosted hydrothermal activity
exists at each of these locales. Volcanism within the Guaymas Basin and Middle Valley
occurs as intrusive sills (Gieskes et al., 1981). Escanaba Trough, on the other hand, has
numerous volcanic domes and edifices that protrude through the sediment cover along the
axial rift in a linear trend (Morton et al., 1987a, 1987b).
The Guaymas Basin, located within the Gulf of California, has total spreading rate of
6cm/yr. Intrusion is by emplacement of thin sills and thicker bysmalith-like bodies, which
episodically raise hills of folded and faulted semiconsolidated turbidites (Figure 25A,
cartoon; Lonsdale & Lawyer, 1980; Lonsdale et al., 1980). Einsele et al. (1980) report that
sill intrusions into highly porous sediments in the Guaymas Basin cause large-scale
Figure 25: Seismic-reflection profiles from three sedimented spreading centers are shown.
(A) : Cartoon (on left) suggesting mechanism of intrusion, faulting and folding, and
morphogenesis in axial rift valley of Guaymas Basin (from Lonsdale &
Lawyer, 1980). Note that intrusion of thin sills has little effect on relief,
because they made space for themselves by expelling pore water from intruded
sediments (Einsele et al., 1980). The seismic-reflection profile (on right)
crosses near a hydrothermal site at the Guaymas Basin. Note confused
diffracted echoes from the sill complex that underlies horizontal basin-floor
strata (from Lonsdale et al., 1980).
(B): Seismic-reflection profile near seafloor dome at the Escanaba Trough (from
Morton et al., 1987b). This record shows upward bowing of the basement
reflector and side-echoes associated with the volcanic edifice beneath the
seafloor dome. Marginal faulting of the sediment fill is also shown. Pictured
below is an interpretive cross-section showing schematic fluid circulation paths
through the volcanic edifice and overlying sedimentary rocks (from Morton et
al., 1987b).
(C) : Seismic-reflection profile from Middle Valley, northern Endeavour Ridge
(from Davis et al., 1987). Small topographic mound (enclosed by box) is a
site of active hydrothermal venting (J. M. Franklin, pers. comm.). The
sediment thickness here is about 300 m and the mound itself is located about 2
km from the eastern bounding normal fault scarp. Both the airgun seismic-
reflection profile and the 3.5-kHz profile (inset) show the mounds to be, in
general, acoustically transparent, with occasional interbedded strong reflectors
(Davis et al., 1987).
62
NW SEVOLCANIC
IPINNACLEBASIN FLOOR
HYOROTHRMALS
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - -B
MASSIVESULFIDE /
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - _.C
SW
Figure 25
63
expulsion of heated pore fluids, thus creating space for the intruding magma which can, in
turn, lead to initiation of a hydrothermal systems. The type of uplift depicted schematically
in Figure 25A (left) should induce discernable drag-folding of turbidite sequences at the
margins of the growing intrusions. Single-channel seismic-reflection data from the near the
hydrothermal site show a confused pattern of diffracted echoes from the sill complex
(Figure 25A, right).
The structure of the slow-spreading (about 2.3cm/yr half-rate) Escanaba Trough has
been described in detail by Morton et al. (1987a, 1987b; Figure 25B). Sulfide deposition
at two sites in the Escanaba Trough is associated with volcanic edifices that have been
faulted upward through the axial valley sediment. In order to account for the small amount
of turbidite sequence deformation associated with the uplift of these volcanic edifices,
Morton et al (1987b) concluded that "these edifices either predate much of the sedimentary
fill or have grown more-or-less contemporaneously with sedimentation". At the margins of
small domes located on the tops of the large edifices, Morton et al (1987b) note an abrupt
change in 3.5-kHz echo-characteristics from gentle pelagic drape to a highly deformed and
uplifted sequences. This is similar to the transition seen at the central domed area of
Cascadia Depression from growth faulted turbidite sediment to hyperbolic echoes (Figures
16 and 17).
The characteristic structure of Middle Valley, northern Endeavour Ridge, is shown by
a seismic-reflection profile in Figure 25C (from Davis et al., 1987). This profile crosses a
topographic mound which is a site of active hydrothermal venting (J.M. Franklin, pers.
comm.). The sediment thickness here is about 300 m and the mound itself is located about
2 km from the eastern bounding normal fault scarp. Both the airgun seismic-reflection
profile and the 3.5-kHz profile (Figure 25C) show the mounds to be, in general,
acoustically transparent, with occasional interbedded strong reflectors.
The character of the acoustic reflectors at the center of Cascadia Depression is revealed
in Figure 26. The subsurface "doming" at Cascadia Depression is defined by highly
reflective, chaotic hyperbolic returns on seismic-reflection records. This is in contrast to
the acoustically transparent nature of seismic-reflection records taken at Middle Valley
(Figure 25C). It is unclear, however, whether this "doming" is caused by shallow volcanic
sills (i.e., similar to the Guaymas Basin), the uplift a volcanic hill (i.e., analogous to the
Escanaba Trough), or perhaps, serpentine diapirism.
Examination of seismic-reflection records at Cascadia Depression reveals small-scale
drag folding in the turbidite sequences around the subsurface margins of the subcrustal
high-reflectivity zones. Subsurface drag-folding at Cascadia Depression does not show
increased deformation with depth thus indicating that diapiric intrusion is jj the cause of
.,H;;;II:;I::;,,.u,::,,,..
: ::.,:;: .
HI
III.
i.I I - I II
I I I.
3 II 11 II
2 I('I
I
Figure 26: Seismic-reflection profile showing drag-folds at margin of volcanic (?) complex
at center of Cascadia Depression. Vertical exaggeration is 3:1.
65
subsurface doming.
The extent of the drag-folding in turbidite reflector sequences around the subsurface
margins of the volcanic complexes at Cascadia Depression indicates that subsurface
processes generally postdate the sedimentary fill (Figure 26). This is in contrast to the
observations of Morton et al. (1987b) at the Escanaba Trough where a general lack of
extensive drag-folding at the margins of volcanic sequences indicates that the volcanic
edifices predate much of the sedimentary fill (Figure 25B, interpretive cross-section).
The type and extent of drag-folding seen in seismic-reflection records (Figure 26)
indicaies that the mechanism of crustal accretion at Cascadia Depression is probably by
growth of volcanic sills, perhaps in a bysmalith-like fashion (i.e., similar to that depicted in
schematically in Figure 25A, cartoon). Therefore, the mechanism(s) of crustal accretion at
Cascadia Depression may be directly analogous to those processes occurring at the
Guaymas Basin (Figure 25A, cartoon; Lonsdale & Lawyer, 1980). Lonsdale & Lawyer
(1980) report that "detached" faults at the Guaymas Basin intersect volcanic sifis and serve
as conduits for hydrothermal circulation (Figure 25A, cartoon). By analogy, the fault
surfaces at Cascadia Depression that intersect the subsurface "doming" (i.e., faults shown
in Figure 21) may serve to focus upwelling hydrothermal fluids from a volcanic sill
complex (Lonsdale & Lawyer, 1980; Kappel & Franklin, in press; Figure 24C).
Future Studies
This study has addressed the detailed structure and tectonics of Cascadia Segment,
using a variety of methods and many types of geophysical data. It has been demonstrated
that active seafloor spreading is the probable mechanism by which the morphology and
structure of Cascadia Segment has evolved. In order to test this hypothesis directly, further
exploration of the current site of active spreading, Cascadia Depression, is required. By
analogy to other sedimented spreading centers, specific target areas for investigation within
Cascadia Depression have been proposed.
Investigative surveys of Cascadia Depression can be carried out using a number of
research methods. A first-order field operation should include a detailed investigation of
the structures within Cascadia Depression using deep-towed side-scan sonar (i.e..
SeaMARC I). These type of data should indicate the true character and orientation of the
faults seen in seismic-reflection data, detect whether any extrusive basalts exist on the floor
of the basin, and reveal the presence of high-density, hydrothermally-altered sediment
(perhaps in the shape of of mounds) within this extensional basin. Detailed sonar records,
coupled with the available SeaBeam bathymetry, wifi serve to delineate the structures that
warrant further investigation.
Of key interest is the detection of sediment-hosted hydrothermal activity and its
associated sulfides withi Cascadia Depression. Water-sampling and salinity-temperature-
density measurements should reveal the presence of hydrothermally-derived constituents
within the water column such as radon, manganese, iron, and associated high
concentrations of methane. Heat-flow measurements within the floor of Cascadia
Depression should define the areas where hydrothermal activity may be occurring.
Seafloor photographic surveys could outline the extent of hydrothermal activity through
mapping of features such as talc deposits, metalliferous sediments, and high concentrations
of hydrothermally-associated organic matter such as bacterial mats. Material obtained in
piston cores and dredge hauls should show similar mineralogy to that found at known
sediment-hosted hydrothermal areas. Mineralogical assemblages from sediment-hosted
hydrothermal areas include pyrrhotite, sphalerite, pyrite, marcasite, barite, chalcopyrite,
talc, and chlorite (i.e., Einsele et aL, 1980; Lonsdale et al., 1980; Lonsdale & Lawyer,
1980; Davis et al., 1987; Morton et al., 1987a, 1987b).
On the regional scale, a first-order problem is to confirm that a progressive age
increase of turbidite-derived sediment in fact exists away from the central depression. This
can be done using a systematic program of dating core material from the individual back-
tilted tectonic blocks that flank Cascadia Depression. The availability of reliable dating
methods (i.e., carbon-14) and detailed SeaBeam bathymetric data makes this a worthwhile
endeavor.
The above discussion gives a general idea of the types of data needed to further our
knowledge of this area. Cascadia Segment's position as a possible center of crustal
accretion that has subject to long-term turbidite sedimentation provides an excellent
opportunity to understand the structure of spreading centers in general. Furthermore, the
existence of very short spreading segment within a long transform fault zone gives new
insight as to where and how crustal accretion can occur. Deciphering the tectonics and
structure of this small, enigmatic spreading segment will help explain the complicated
geologic history of this part of the ocean basin and, perhaps more importantly, may tell us
more about the structural and tectonic processes associated with seafloor spreading.
67
BIBLIOGRAPHY
Anderson, R.N., and H.C. Noltimier, 1973, A model for the horst and graben structure of
midocean ridge crests based on spreading velocities and basalt delivery to the ocean
crust, Geophys. J. Roy. Astron. Soc., v. 34, p. 137-149.
Atwater, T., 1970, Implications of plate tectonics for the evolution of western North
America, Geol. Soc. Amer. Bull., v. 81, p. 3513-3536.
Atwater, T. and J.D. Mudie, 1968, Block faulting on the Gorda Rise, Science, v. 159, p.
729-731.
Atwater, T. and J.D. Mudie, 1973, Detailed near-bottom geophysical study of the Gorda
Rise, J. Geophys. Res., v.78, p. 8665-8686.
Aydin, A. and A. Nur, 1982, Evolution of pull-apart basins and their scale independence,
Tectonics, v. 1, p. 9 1-105.
Ballard, R.D, J. Francheteau, T. Juteau, C. Rangan, W. Normark, 1981, East Pacific Rise
at 21°N: the volcanic, tectonic, and hydrothermal processes of the central axis,
Earth Planet. Sci. Lett., v. 55, p. 1-10.
Ballard, R.D., Tj. van Andel, and R.T. Holcomb, 1982, The Galapagos Rift at 86°W:
variations in volcanism, structure, and hydrothermal activity along a 30-km
segment of the rift valley, J. Geohys. Res., v. 87, p. 1149-1161.
Bischoff, J.L., R.J. Rosenbauer, P.J. Aruscavage, P.A. Baedecker, and J.G. Crock,
1983, Seafloor massive sulfide deposits from 21 °N, East Pacific Rise, Juan de
Fuca Ridge, and Galapagos Rift: bulk chemical composition and economic
implications, Econ. Geol., v. 78, p. 1711-1720.
Bolt, B.A., C. Lomnitz and T.V. McEvilly, 1968, Seismological evidence on the tectonics
of central and northern California and the Mendocino escarpment, Seismol. Soc.
Amer. Bull., v. 58, p. 1735-1767.
Carlson, R.L., A.F. Gangi and K.R. Snow, 1986, Empirical reflection travel time versus
depth and velocity versus depth functions for deep-sea sediment column, L
Geophys. Res., v. 91, p. 8249-8266.
CAYTROUGH, 1979, Geological and geophysical investigation the mid-Cayman Rise
spreading center: Initial results and observations, j Deep drilling results in the
Atlantic Ocean: Ocean crust, M. Talwani, C.G. Harrison and D.E. Hayes (eds.),
American Geophysical Union, Maurice Ewing Series, v. 2, p. 66-93.
Chandra, U., 1974, Seismicity, earthquake mechanisms and tectonics along the western
coast of North America from 42°N to 61°N, Seismol. Soc. Amer. Bull., v. 64, p.
1529- 1549.
Couch, R.W., C. Keinle and G. Connard, 1978, The continental margin of the Pacific
Northwest: A sumniaiy of the geological and geophysical data, School of
Oceanography, Oregon State University GTF 780816.
Damuth, J.E., 1975, Echo character of the Western Equatorial Atlantic floor and its
relationship to the dispersal and distribution of terrigenous sediments, Mar. Geol.,
v. 18, p. 17-45.
Damuth, J.E., 1980, Quaternary sedimentation processes in the South China Basin as
revealed by echo-character mapping and piston-core studies, flj Hayes, D.E. (ed.),
The tectonic and geologic evolution of Southeast Asian seas and islands, American
Geophysical Union, Geophysical Monograph 23, p. 105-125.
Damuth, J.E., R.D. Jacobi and D.E. Hayes, 1983, Sedimentation processes in the
Northwest Pacific Basin revealed by echo-character mapping studies, Geol. Soc.
Amer. Bull., v. 94, p. 381-395.
Davis, E.E. and C.R.B. Lister, 1977, Tectonic structures on the Juan de Fuca Ridge,
Geol. Soc. Amer. Bull., v. 88. p. 346-363.
Davis, E.E., W.D. Goodfellow, B.D. Bornhold, J. Adshead, B. Blaise, H. Villinger, and
G.M. LeCheminant, 1987, Massive sulfides in a sedimented rift valley, northern
Juan de Fuca Ridge, Earth Planet. Sci. Lett., v. 82, p. 49-61.
Deffeyes, K.S., 1970, The axial valley: A steady-state feature of the terrain, IN Johnson,
H. and B.L. Smith (eds.), Megatectonics of continents and oceans, Camden, N.J.,
Rutgers Univ. Press, p. 194-222.
Dehlinger, P., R.W. Couch, D.A. McManus and M. Gemperle, 1970, Northeast Pacific
structure, Maxwell, A. (ed.), The Sea, v. 4, John Wiley and Sons, New York,
p. 133-189.Delaney, J.R., F.N. Spiess, W.E. Colony, J.L. Karsten, D.E. Kelley and P. Nehlig,
1987, A complete deep-tow swath map of southern-facing "wall" of the Blanco
Trough, Juan de Fuca Ridge region, Eos Trans. A.G.U., v. 68, p. 1509.
Dunnebier, F.K. and R.H. Johnson, 1967, T-phase sources and earthquake epicenters in
the Pacific basin, Hawaii Inst. Geophys. Rep., RTG-67-24, Univ. Hawaii at
Manoa, Honolulu.
Duncan, J.R., 1968, Late Pleistocene and post glacial sedimentation and stratigraphy of
deep-sea environments off Oregon, Ph.D. dissertation, Oregon State Univ., 222 p.
Eberhart, G.L., 1987, The geology and tectonic setting of three hydrothermally active sites
along the Mid-Atlantic ridge, M.S. thesis, Univ. of Miami, 181 p.EEZ Scan 84 Scientific Staff, 1986, Atlas of the Exclusive Economic Zone, Western
Coterminous United States, U.S. Geological Survey Misc. Investigations.
Einsele, G., J.M. Gieskes, J. Curray, D.M. Moore, E. Aguayo, M.-P. Aubry, D. Fornari,
J. Guerrero, M. Kastner, K. Kelts, M. Lyle, Y. Matoba, A. Molina-Cruz, J.
Niemitz, J. Rueda, A. Saunders, H. Schrader, B. Simoneit and V. Vacquier, 1980,
Intrusion of basaltic sills into highly porous sediments, and resulting hydrothermal
activity, Nature, v. 283, p. 441-445.
Embley, R.W., 1985, A locally formed deep ocean canyon system along the Blanco
Transform, Northeast Pacific, Geo. Mar. Lett., v. 5, p. 99-104.
Embley, R.W., L.D. Kulm, G. Massoth, D. Abbott and M. Holmes, 1987, Morphology,
structure and resource potential of the Blanco transform fault zone, Scholl,
D.W., A. Grantz and J.G. Vedder (eds.), Geology and Resource Potential of the
Continental Margin of Western North America and Adjacent Ocean Basins--
Beaufort Sea to Baja California, Circum-Pacific Council for Energy and Mineral
Resources, v. 6, p. 549-561.
Embley, R.W., D. Wilson, A. Malahoff and R. Ganse, in press, Tectonics of the Blanco
transform fault - N.E. Pacific, 40 pp.
Fox, P.J. and D.G. Gallo, 1982, A mantle weld at ridge-transform intersections:
Implications for the development of oblique structures, Eos Trans. A.G.U., v. 63,
p.446.
Fox, P.J. and D.G. Gab, 1984, A tectonic model for ridge-transform-ridge plate
boundaries: Implications for the structure of oceanic lithosphere, Tectonophvsics,
v. 104, p. 205-242.
Gieskes, J, M. Kastner, G. Einsele, K. Kelts and J. Niemitz, 1981, Hydrothermal activity
in the Guaymas basin, Gulf of California: a synthesis, Curray, J.R., Moore,
D.G. et aL, mit. Repts. Deep Sea Drilling Project, v. 64, Washington (U.S. Govt.
Printing Office), p. 1159-1167.
Griggs, G.B. and L.D. Kulm, 1970, Sedimentation in Cascadia deep-sea channel, Geol.
Soc. Amer. Bull., v. 8, p. 1361-1384.
Griggs, G.B. and L.D. Kulm, 1973, Origin and development of Cascadia deep-sea
channel, J. Geophys. Res., v. 78, p. 6325-6339.
Harrison, C.G.A., 1974, Tectonics of mid-ocean ridges, Tectonophysics, v. 22, p. 301-310.
70
Hart, R., D. Pyle and J. Robbins, 1986, Hydrothermal sulfides, breccias and greenstones
from the Gorda Depression, Dept. of Geol. and Mi Res. 0en File Rept., v. 86-16.
Heirtzler, J.R., 0.0. Dickson, E.M. Herron, W.C. Pittman III and X. Le Pichon, 1968,
Marine magnetic anomalies, geomagnetic field reversals and motions of the ocean
floor, J. Geophys. Res., v. 73, p. 2119-2136.
Hey, R.N., F.K. Dunnebier and W.J. Morgan, 1980, Propagating rifts on mid-ocean
ridges, J. Geophys. Res., v. 85, p. 3647-3658.
Holmes, M.L., and J.L. Morton, 1986, Massive sulfide deposits from Escanaba Trough,
Gorda Ridge (abs.), Geol. Soc. Amer., Abs. with Prog., v. 18, p. 118.Ibach, D.H., 1981, The structure and tectonics of the Blanco Fracture Zone, M. S. thesis,
Oregon State Univ., 6Op.
Irving, E., 1970, The Mid-Atlantic Ridge at 45°N: XIV, Oxidation and magnetic properties
of basalt; review and discussion, Can. Jour. Earth Sci., v. 7, p. 1528-1538.Johnson, H.P., and T. Atwater, 1977, Magnetic study of basalts from the Mid-Atlantic
ridge, lat. 37°N, Geol. Soc. Amer. Bull., v. 88, p. 637-647.
Johnson, H.P., and J.M. Hall, 1978, A detailed rock magnetic and opaque mineralogical
study of the basalts of the Nazca Plate, Geophys. J. Roy. Astron. Soc., v. 52, p.45-64.
Johnson, S.H. and P.R. Jones, 1978, Microearthquakes located on the Blanco Fracture
Zone with sonobuoy arrays, J. Geonhys. Res., v. 83, p. 255-261.
Kappel, E.S. and J.M. Franklin, in press, The relationship between geological
development of ridge crests and sulfide deposits in the northeast Pacific Ocean.
Kappel, E.S., and W.B.F. Ryan, 1986, Volcanic episodicity and a non-steady state rift
valley along northeast Pacific spreading centers: evidence from SeaMARC 1J,
Geophys. Res., v. 91, p. 13,925- 13,940.
Karlin, R. and M. Lyle, 1986, Sediment studies on the Gorda Ridge, State of Oregon,
Dept. of Geol. and Mi md. Open-file Rept., v. 0-86-19, 76 pp.Klitgord, K.D. and J.D. Mudie, 1974, The Galapagos spreading center: A near-bottom
geophysical survey, Geophys. J. Roy. Astron. Soc., v. 38, p. 563-586.
Kulm, L.D. and G.A. Fowler, 1974, Oregon continental margin structure and stratigraphy:
A test of the imbricate thrust model, j Burk, C.A. and C.L. Drake (eds.), The
Geology of Continental Margins, Springer-Verlag, New York, p. 26 1-284.
Kulm, L.D., R. von Huene et aL, 1973, Initial Reports of the Deep-Sea Drilling Project, v.
18, Washington (U.S. Govt. Printing Office), 294 pp.
71
Lachenbruch, A.,1973, A simple mechanical model for oceanic spreading centers, J.
Geophys. Res., v. 78, P. 3395-3417.
Levi, S. and R. Riddihough, 1986, Why are marine magnetic anomalies suppressed over
sedimented spreading centers?, Geology, v. 14, p. 651-654.Lister, C.R.B., A. V. deCharon and L. Chiao, 1987, Profile delineates tilted sedimentary
blocks on both sides of Cascadia spreading center in the Blanco Transform, EOS
Trans. A.G.U., v. 68, n. 44, p. 1483-1484.
Loncarevic, B.D. and R.L. Parker, 1971, The Mid-Atlantic ridge near 45°N,17, Magnetic
anomalies and seafloor spreading, Canadian Jour. of Earth Sci., v. 8, p. 883-898.Lonsdale, P.F., and K. Becker, 1985, Hydrothermal plumes, hot springs, and convective
heat flow in the southern trough of the Guaymas Basin, Earth Planet. Sci. Lett., v.
3, p. 21 1-225.
Lonsdale, P.F., and L.A. Lawyer, 1980, Immature plate boundary zones studied with a
submersible in the Gulf of California, Geol. Soc. Amer. Bull., v. 91, p. 555-569.Lonsdale, P.F., J.L. Bischoff, V.M. Burns, M. Kastner, and R.E. Sweeney, 1980, A
high-temperature hydrothermal deposit on the seabed at a Gulf of California
spreading center, Earth Planet. Sci. Lett., v. 49, p. 8-20.
Macdonald, K.C., 1977, Near-bottom magnetic anomalies, asymmetric spreading, oblique
spreading and tectonics of the Mid-Atlantic Ridge near 37°N, Geol. Soc. Amer.
Bull., v. 88, p. 541-555.
Macdonald, K.C., 1982, Mid-ocean ridges: fme scale tectonic, volcanic, and hydrothermal
processes within the plate boundary zone, Ann. Rev. Earth Planet. Sci. Lett., v.
10, p. 155-190.
Macdonald, K.C., 1986, The crest of the Mid-Atlantic Ridge: models for crustal generation
processes and tectonics, fi Vogt, P.R. and B.E. Tucholke (eds.), The Geology of
North America, Volume M, The Western North Atlantic Region: Geological Society
of America, p. 51-68.
Macdonald, K.C. and T.M. Atwater, 1978, Evolution of mid-ocean ridges, Earth Planet.
Sci. Lett., v. 39, p. 3 19-327.
Macdonald, K.C. and B.P. Luydendyk, 1977, Deep-tow studies of the structure of the
Mid-Atlantic Ridge crest near 37°N (FAMOUS), Geol. Soc. Amer. Bull., v. 88, p.621-636.
Marshall, M., 1978, The magnetic properties of some DSDP basalts from the North Pacific
and inferences for Pacific plate tectonics, J. Geophys. Res., v. 83,p. 289-308.McGregor, B.A. and P.A. Rona, 1975, Crest of the Mid-Atlantic ridge at 26°N, j
Geophys. Res., v. 80, p. 3307-3314.
72
Melson, W.G., 1969, Preliminary results of a geophysical study of portions of the Juan de
Fuca Ridges and Bianco fracture zone: Environmental Science Services
Administration (ESSA) Tech. Memo., Coast and Geodetic Survey Tech. Memo. 6,
U.S. Dept. Comm.
Morgan, J., E. Parmentier and J. Lin, 1987, Mechanisms for the origin of mid-ocean ridge
axial topography: Implications for the thermal and mechanical structure of accreting
plate boundaries, J. Geophys. Res., v. 92, p. 12,823-12,836.
Morton, J.L., W.R. Normark, S.L. Ross, R.A. Koski, M.L. Holmes, W.C. Shanks, III,
R.A. Zierenberg, M. Lyle, and L.M. Benninger, 1987a, Preliminary report, cruises
L1-86-NC and L2-86-NC, Escanaba Trough, Gorda Ridge, U.S. Geol. Surv.
Open-File Rept., v 87-375-A, l3pp.
Morton, J.L., M.L. Holmes and R.A. Koski, 1987b, Volcanism and massive sulfide
formation at a sedimented spreading center, Escanaba Trough, Gorda Ridge, N.E.
Pacific Ocean, Geophys. Res. Lett., v. 14, p. 769-772.
Needham, H.D. and J. Francheteau, 1974, Some characteristics of the rift valley in the
Atlantic Ocean near 36°48' north, Earth Planet. Sci. Lett., v. 22, p. 29-43.
Normark, W.R., 1976, Delineation of the main extrusion zone of the East Pacific Rise at
latitude 21°N, Geology, v. 4, p. 681-685.
Normark, W.R., J.L. Morton, R.A. Koski, D.A. Clague, J.R. Delaney, 1983, Active
hydrothermal venting and sulfide deposits on the southern Juan de Fuca Ridge,
Geology, v. 11, p. 158-163.
Northrup, J., 1970, Accuracy of earthquake epicenters on the Gorda Ridge, Seismol. Soc.
Amer. Bull., v. 60, p. 265-267.
Osmaston, M.F., 1971, Genesis of ocean ridge median valleys and continental rift valleys,
Tectonophysics, v. 11, p. 387-405.
Palmasson, G., 1973, Kinematics and heat flow in a rift zone, with application to Iceland,
Geophys. J. Roy. Astron. Soc., v. 33, p.451-481.
Palmer, J., J.R. Delaney and F.N. Spiess, 1987, Recent deep-tow observations at the
Blanco Fracture Zone-Juan de Fuca Ridge intersection, Eös Trans. A.G.U., v. 68,
p. 1492.Prévot, J., A. Lecaille, and R. Hekinian, 1979, Magnetism of the Mid-Atlantic Ridge crest
near lat. 37°N from FAMOUS and DSDP results: a review, jj Talwani, M. et
al.(eds.), Deep Drilling Results in the Atlantic Ocean: Ocean Crust, v. Amer.
Geophys. Union Maurice Ewing Series, v. 2, p. 210-229.
73
Ramberg, I.B., and T.H. van Andel, 1977, Morphology and tectonic evolution of the rift
valley at latitude 36°30'N, Mid-Atlantic Ridge, Geol. Soc. Amer. Bull., v. 88, p.577-586.
RISE Project Group, 1980, East Pacific Rise: hot springs and geophysical experiments,
Science, v. 207, p. 142 1-1433.
Rodgers, G.C., 1979, Fault plane solutions, Juan de Fuca plate map, JFP-6, Pacific
Geoscience Center, Dept. Mines, Energy and Resources, Sidney British Columbia,
Canada.
Selk, B.W., 1978, The manganese-enriched sediments of the Blanco Trough: evidence for
hydrothermal activity in a fracture zone, M.S. thesis, Oregon State Univ., 137 p.Seyfried, W.E., Jr., and D.R. Janecky, 1985, Heavy metal and sulfur transport during
subcritical and supercritical hydrothermal alteration of basalt: influence of fluid
pressure and basalt composition on crystallinity, Geochim. Cosmo. Acta., v. 49, p.
2545-2560.
Sleep, N., 1969, Sensitivity of heat flow and gravity to the mechanism of seafloor
spreading, J. Geophys. Res., v. 74, p. 542-549.
Tamayo Tectonic Team, 1984, Tectonics at the intersection of the East Pacific Rise with the
Tamayo Transform Fault, Mar. Geophys. Res., v. 6, p. 159-185.Tapponnier, P., and J. Franceteau, 1978, Necking of the lithosphere and the mechanics of
slowly accreting plate boundaries, J. Geophys. Res., v. 83, p. 3955-3970.Tobin, D.G. and L.R. Sykes, 1968, Seismicity and tectonics of the northeast Pacific
Ocean, J. Geophys. Res., v. 73, p. 3821-3845.
van Andel, Tj.H. and C.R. Heath, 1970, Tectonics of the Mid-Atlantic ridge 6-8° south
latitude, Mar. Geophys. Res., v. 1, p. 5-36.
Vogt, P.R., C.N. Anderson, D.R. Bracey, and E.D. Schneider, 1970, North Atlantic
magnetic smooth zones, J. Geophys. Res., v. 75, p. 3955-3968.
Vogt, P.R. and G.R. Byerly, 1976, Magnetic anomalies and basalt composition in the Juan
de Fuca-Gorda Ridge area, Earth Planet. Sci. Lett., v. 33, p. 185-207.
Wilson, D.S., R.N. Hey and C. Nishimura, 1984, Propagation as a mechanism of
reorientation of the Juan de Fuca ridge, J. Geophys. Res., v. 89, p. 92 15-9225.
