Vertical tectonic movements of the crust in transform fracture zones of the Central Atlantic

Abstract

The most significant vertical movements of the oceanic crust in the Central Atlantic are characteristic of transverse ridges
confined to transform fracture zones. These movements are also recorded in some local depressions of the Mid-Atlantic Ridge
(MAR) and in older structures of deep-sea basins. The amplitude of such movements substantially exceeds that related to the
cooling of lithospheric plates. Vertical movements can be driven by various factors: the thermal effect of a heated young
MAR segment upon a cold plate, thermal stress, thermal energy released by friction in the course of displacement of fault
walls relative to each other, serpentinization of the upper mantle rocks in the transform fault zone, and lateral compression
and extension. The alternation of compression and extension that arises because of the nonparallel boundaries of the transform
fracture zone and the unstable configuration of the rift/fracture zone junction was the main factor responsible for the formation
of the transverse ridge in the Romanche Fracture Zone. The most probable cause of the vertical rise of the southern transverse
ridge in the Vema Fracture Zone is the change in the spreading direction. In general, the fracture zones with active segments
more than 100 km long are characterized by extension and compression oriented perpendicularly to the main displacement and
related to slight changes in the spreading configuration. It is impossible to single out ambiguously the causes of vertical
movements in particular structural features. In most cases, the vertical movements are controlled by several factors, while
the main role belongs to the lateral compressive and tensile stresses that appear owing to changes in the movement of lithospheric
blocks in the course of MAR spreading.

Full text

25
ISSN 0016-8521, Geotectonics, 2006, Vol. 40, No. 1, pp. 25–36. © Pleiades Publishing, Inc., 2006.
Original Russian Text © A.A. Peive, 2006, published in Geotektonika, 2006, No. 1, pp. 31–43.
INTRODUCTION
The basement topography in the Central Atlantic is
formed by tectonic movements of different directions
and amplitudes and owing to volcanic activity. Vertical
tectonic movements are of primary importance in the
formation of transform fracture zones (FZs) and posi-
tive structural features of both the Mid-Atlantic Ridge
and deep basins. The amplitude of vertical movements
with formation of large positive features in some FZs
amounts to many hundred meters during short-term
periods, a value which is inconsistent with the standard
spreading model. The study of these movements is
important for geology of the oceanic floor and provides
insight into the origin of Atlantic Ocean, 60% of whose
bottom is occupied by FZs. At present, Russian and for-
eign expeditions have studied the largest FZs of the
Atlantic Ocean and gained extensive geological and
geophysical data, which allow estimation of the contri-
bution of vertical tectonics to the formation of oceanic
structures and understanding of their mechanisms of
formation [6, 15, 17, 21, and others]. Scientists from
the Geological Institute of the Russian Academy of Sci-
ences have performed many expeditions in the Central
Atlantic, and important material on the Romanche,
Vema, Fifteen Twenty, and other FZs has been gained.
On the basis of these studies, large positive structural
units of FZs have been described in detail [1, 4, 6–8, to
name a few]. Therefore, the factual data concerning
these structures will be touched on only briefly and the
main attention will be focused on the most probable
models and mechanisms of their formation.
The Central Atlantic is characterized by develop-
ment of extended transform FZs such as the Romanche,
Vema, Kane, Fifteen Twenty, Oceanographer, etc. Their
principal feature is a transform valley and bounding
transverse ridges of different lengths and heights. At
junctions with segments of rift valleys, the transverse
ridges pass into the so-called inner and outer corner
highs. The transverse ridges rise above the coeval bottom
for many hundred meters. They are represented by either
continuous elevated lithospheric blocks, as in the Vema
FZ, or isolated tectonic fragments of crustal and upper
mantle rocks. The transverse ridges occur both in active
segments of FZs and beyond them, up to deep basins.
They are among the most prominent morphological fea-
tures that have resulted from vertical movements.
The transverse ridges include separate second-order
uplifts, the heights of which significantly exceed the
common level of the ridge [11, 30]. The largest among
these uplifts are highs in the eastern part of the northern
transverse ridge in the Romanche FZ and the southern
Vema FZ, as well as the uplift of the St. Peter and
St. Paul Rocks located on the northern transverse ridge
of the São Paulo FZ 200 km east of its western intersec-
Vertical Tectonic Movements of the Crust in Transform Fracture
Zones of the Central Atlantic
A. A. Peive
Geological Institute, Russian Academy of Sciences, Pyzhevskii per. 7, Moscow, 119017 Russia
Received May 23, 2005
Abstract
—The most significant vertical movements of the oceanic crust in the Central Atlantic are character-
istic of transverse ridges confined to transform fracture zones. These movements are also recorded in some local
depressions of the Mid-Atlantic Ridge (MAR) and in older structures of deep-sea basins. The amplitude of such
movements substantially exceeds that related to the cooling of lithospheric plates. Vertical movements can be
driven by various factors: the thermal effect of a heated young MAR segment upon a cold plate, thermal stress,
thermal energy released by friction in the course of displacement of fault walls relative to each other, serpenti-
nization of the upper mantle rocks in the transform fault zone, and lateral compression and extension. The alter-
nation of compression and extension that arises because of the nonparallel boundaries of the transform fracture
zone and the unstable configuration of the rift/fracture zone junction was the main factor responsible for the
formation of the transverse ridge in the Romanche Fracture Zone. The most probable cause of the vertical rise
of the southern transverse ridge in the Vema Fracture Zone is the change in the spreading direction. In general,
the fracture zones with active segments more than 100 km long are characterized by extension and compression
oriented perpendicularly to the main displacement and related to slight changes in the spreading configuration.
It is impossible to single out ambiguously the causes of vertical movements in particular structural features. In
most cases, the vertical movements are controlled by several factors, while the main role belongs to the lateral
compressive and tensile stresses that appear owing to changes in the movement of lithospheric blocks in the
course of MAR spreading.
DOI:
10.1134/S0016852106010031

26
GEOTECTONICS
Vol. 40
No. 1
2006
PEIVE
tion with the rift valley [12]. These islands are frag-
ments of the subcontinental mantle that have been
uplifted to the surface and are composed of peridotite
and of mylonite that developed from peridotite. It is
probable that this fragment of the upper mantle was
retained after the Equatorial Atlantic near the MAR was
opened as a result of multiple changes in the spreading
direction and displacements of transform FZ axes [12].
STRUCTURE OF TRANSVERSE RIDGES
IN THE ROMANCHE, VEMA,
AND KANE FRACTURE ZONES
The Romanche Fracture Zone.
The most elevated
part of the northern transverse ridge in the Romanche
FZ at a depth less than 2 km is located approximately
opposite to the southern intersection of the FZ with the
rift valley (Fig. 1) and consists of several uplifts called
highs A, B, C, and D from west to east [15]. These highs
were crossed by CMP profiles and dredged. As a result,
the representative data allowing interpretation of their
geologic structure have been obtained [4, 15].
High A is the most uplifted segment of the
Romanche FZ and reaches only 930 m in depth. Dredg-
ing near the summit yielded shallow-water reef lime-
stone that accumulated at a depth of approximately 50 m
and was subjected to subaerial diagenesis. Coral
remains date this limestone at about 5 Ma [15]. The
CMP record demonstrates that high A consists of an
upper layer 250–300 m thick that is separated from the
seismically opaque sequence by a distinct reflector.
According to the dredging data, this sequence is proba-
Romanche FZ (Rom 2) Rift valley of the Mid-Atlantic Ridge
West
East
Chain FZ
Rift valley of the Mid-Atlantic Ridge
Paleo-Romanche FZ (Rom 1)
Change in Romanche valley
Uplifted area
of transverse ridge
150 km
strike at 19
°
–20
°
W
Fig. 1.
Topography of the Romanche Fracture Zone deduced from satellite altimetry data [27].

GEOTECTONICS
Vol. 40
No. 1
2006
VERTICAL TECTONIC MOVEMENTS OF THE CRUST IN TRANSFORM FRACTURE ZONES 27
bly a fragment of the oceanic crust composed of
basalts, gabbroic, and ultramafic rocks. Highs B and C
are characterized by a similar structure. Thus, all these
highs may be regarded as subsided atolls formed on the
eroded oceanic crust. The subsidence rate of these
structural features was approximately 0.2 mm/yr, i.e.,
an order of magnitude higher in comparison with that
calculated from thermal cooling and subsidence of the
oceanic lithosphere [28].
The second type of fault-line highs is represented by
the isometric high D that is slightly elongated along the
fault. This high is isolated from the fault scarp by a rel-
atively gentle slope dipping southward. The multichan-
nel seismic profiling across the high in the latitudinal
direction revealed a thick (up to 4 km) complexly
deformed sequence with folds and thrusts [16]. The
sequence is composed of deformed volcanosedimen-
tary rocks. In the western part of the high, as is evident
from the dredged material [15], the sequence consists
of a lower unit composed of Lower Cretaceous pelagic
limestone and an upper unit formed by Paleocene–
Eocene arkose sandstone. The rocks are much older
than can be expected from the spreading rate calculated
for the axial zone of MAR. Furthermore, the occur-
rence of the Lower Cretaceous rocks near the eastern
intersection with the rift valley is inconsistent with the
paleogeographic reconstructions, according to which
the South and Central Atlantic segments merged in the
Albian [23, 26]. Such localization of the Lower Creta-
ceous sequence cannot be explained in terms of the
simple model of oceanic crust spreading. A deep basin
probably existed in the Equatorial Atlantic between the
Gulf of Guinea and South America prior to the main
stage of the South Atlantic opening that occurred
140 Ma ago.
The real height of the northern transverse ridge in
the Romanche FZ exceeds the level calculated from the
cooling of the lithosphere by 2 km in the area between
15°30
′
and
18°
W and by 3–4 km in highs A, B, C, and D.
The
Vema Fracture Zone
is one of the largest FZs
in the near-equatorial segment of the Atlantic. The
southern wall of the Vema Valley is marked by a trans-
verse ridge 310 km long, 4.5 km high (above the valley
bottom), and approximately 30 km wide at the base
(Fig. 2). This ridge is a raised undisturbed fragment of
the oceanic lithosphere [2, 14, 20]. The transverse ridge
100 km
West
East
Vema FZ
Rift valley of the Mid-Atlantic Ridge
Transverse ridge
Rift valley of the
Mid-Atlantic
Ridge
Lemm Fault
Fig. 2.
Topography of the Vema FZ deduced from satellite altimetry data [27].

28
GEOTECTONICS
Vol. 40
No. 1
2006
PEIVE
consists of segments separated by faults. The upper
parts of two of these segments are composed of shal-
low-water limestone. The summit of the transverse
ridge is located 2–4 km above the level, which is
expected in the case of normal thermal contraction.
This anomalous area originates at a distance of approx-
imately 100 km from the eastern intersection with the
rift valley and is traceable for 300 km. The highest part
of the ridge at a depth less than 600 m is located 70 km
west of the western intersection.
The limestone dredged at the latitude of
44°22
′
N
makes up a thick (about 500 m) carbonate platform
50 km long that was formed near the sea level in the
Middle Pliocene (3–4 Ma ago) [17]. Thus, the summit
of the transverse ridge subsided to its present-day posi-
tion (600 m) at a rate of 0.2–0.3 mm/yr. The age of this
block is approximately 35 Ma if an average spreading
rate of 1.2 mm/yr and a present-day distance from the
MAR axis of 360 km are assumed. The subsidence rate
of the cooling lithospheric plate is 0.01 mm/yr [28], a
value which is an order of magnitude lower than that
observed. The limestone located down the slope at the
same longitude was formed close to the ocean surface
between 5 and 14 Ma. Now, it is located within a depth
interval of 1.0–1.4 km that corresponds to a subsidence
rate of 0.1–0.3 mm/yr and the age of limestone dredged
from the uppermost slope of the ridge [16].
The
Kane Fracture Zone
. The transverse ridge
extends along the northern wall of the FZ valley for
200 km to the east of the eastern intersection with the
rift valley and is 15–40 km wide and 1.2 km high rela-
tive to the coeval crust (Fig. 3). According to the seis-
mic data, the Moho discontinuity is raised beneath the
Kane FZ and the transverse ridge. Located further south
is the
21°30
′
N FZ, which is not parallel to the Kane FZ in
its eastern segment. A divergence of
4°–6°
provided exten-
sion of the block between these FZs for 4–12 km [25].
The transverse ridge is marked by a positive Bou-
guer anomaly indicating that the basal part of the crust
beneath this ridge is characterized by an excess of mass.
The crust of the young plate has an anomalously low
density. The excessive mass can be explained by
East
West
50 km
Rift valley of the
Mid-Atlantic
Ridge
Kane FZ
Transverse ridge
Rift valley
of the Mid-Atlantic
Ridge
Fig. 3.
Topography of the Kane FZ deduced from satellite altimetry data [27].

GEOTECTONICS
Vol. 40
No. 1
2006
VERTICAL TECTONIC MOVEMENTS OF THE CRUST IN TRANSFORM FRACTURE ZONES 29
upwelling of the mantle material that compensated the
rise and flexure of the lithosphere. The low density of the
young plate could have been caused by crust fracturing
or a large volume of detrital material that formed in the
course of the tectonic rise of the transverse ridge [25].
VERTICAL MOVEMENTS BEYOND
TRANSVERSE RIDGES
Vertical tectonic movements beyond the rift valleys
and the active segments of transform FZ, deep-sea
basins included, occurred throughout the entire neotec-
tonic stage of the Atlantic’s evolution [5]. The forma-
tion of some isometric uplifts of the acoustic basement
and diapir-shaped, usually linear uplifts (swells) is related
to these movements [1]. These uplifts extend for many
tens of kilometers in the near-latitudinal direction and
pierce various sedimentary layers, deforming the latter.
Three groups of diapir-shaped swells have been
established in the area between 7
°
and
10°
N [8]. The
shallowest diapirs (250–2500 m) are situated to the
west of
42°
W; between
34°
and
40°30
′
W, they are sub-
stantially deeper (<400 m). The third group is located
east of
33°
W. Most of these diapirs are located at a
depth of 4.5–4.8 km. Their heights vary from 0.2 to
1.6 km [1]. The diapir-shaped structural features are
also noted northward in passive segments of the Mara-
thon and Mercury FZs. Deformations of sediments near
the diapirs are similar to deformations that are related
to salt and clay domes and expressed in tilting and fold-
ing of sedimentary layers [8]. At the same time, it is
hardly possible that the revealed diapirs have a similar
origin. Sporadic dredgings suggest that these diapirs
are composed of volcanic or deep holocrystalline rocks
that are occasionally intensely deformed and altered by
secondary processes. Two mechanisms responsible for
the diapir formation are most probable. The diapirs may
be related to the vertical tectonic movements caused by
(1) lateral extension or compression of fault-bounded
blocks of the oceanic crust or by (2) serpentinization of
ultramafic rocks that substantially increases the rock
volume and provides vertical squeezing of plastic mate-
rial along weakened linear zones, which, in general, are
parallel to transform faults, with the capture of the over-
lying second and third layers of the oceanic crust. Some
diapirs pass along the strike into the transverse ridges,
as is observed between the Doldrums and Vernadsky
FZs, or have similar relationships with sediments.
Thus, it may be concluded that at least some of the
transverse ridges and diapir-shaped swells are of the
same origin [8].
As can be judged from rare earthquakes, some struc-
tural features remain tectonically active, although most
of them are Pleistocene in age, as is evident from defor-
mations of the upper sedimentary layers [1].
Some isometric uplifts of the acoustic basement are
located in the Vema FZ [21]. In depressions filled with
sediments within the Sierra Leone Tectonic Zone
(
5°20
′
–5°40
′
N), horst-shaped uplifts 300–500 m high
have stepwise or tilted slopes and reveal intricate rela-
tionships with the sedimentary fill that indicate their
tectonic origin. The formation of the uplifts is still in
progress because they deform all sedimentary layers
[7]. These uplifts are structurally connected with larger
uplifts located within rift-related ridges. It seems that
intense differentiated vertical movements are closely
related to lateral latitudinal extension of a heteroge-
neous, intensely fractured MAR segment between 5
°
and
7°
N that formed in the geodynamic setting of dry
spreading [3].
MECHANISMS OF VERTICAL MOVEMENTS
DURING FORMATION
OF TRANSVERSE RIDGES
The fault valley is an integral part of transform FZs,
whereas the transverse ridges do not always accompany
this negative structural feature. The transform valley is
largely a result of tensile stress that attends the shearing
in active segments of transform FZs in the course of
spreading of lithospheric plates. This does not rule out
that particular segments of a transform FZ, including a
transform valley, could experience compression in
some periods of their development.
As concerns the cause of vertical movements in
some areas of the oceanic crust that lead to the forma-
tion of transform ridges, several models explaining this
phenomenon have been suggested, although none of
them are universal. These include models of the thermal
effect of a heated young MAR segment upon the cold
plate, thermal stress, thermal energy released by the fric-
tion of fault walls during their displacement relative to
each other, serpentinization–deserpentinization of the
upper mantle rocks in a transform FZ, compression and
extension during displacements along FZ, and others.
The thermal effect
of the heated young MAR seg-
ment on the cold plate. According to model calculations
[18, 22], the thermal anomaly that exists beneath the rift
valley heats the segment of the cold plate at the inter-
section between the rift and transform valleys and can
provide its uplifting (Fig. 4, model A). This assumption
is indirectly supported by the fact that the large uplifts
in the Romanche and Vema FZs are situated near such
intersections. In an FZ with a 30-Ma-old active seg-
ment, the temperature of the old lithosphere in the crust
area corresponding to 10 Ma at a distance of 20 km
from the fault axis should rise by
100°ë
to a depth of
70 km. The maximal rise in this case can be as large as
a few hundred meters [11]. Such an effect could pro-
vide uplifting by no more than 500 m at the eastern
intersection of the Romanche FZ [15]; this value is
insufficient for the Vema FZ as well. Moreover, this
model does not explain the vertical movements in FZs
beyond the region of contact with active rift segments
and the formation of transverse ridges only on one side
of the fault.

30
GEOTECTONICS
Vol. 40
No. 1
2006
PEIVE
Rise of the
plate edge
A
Cold
plate
Hot
plate
Transform fault
Heat transfer
B
Spreading axis
Compression Extension
Spreading axis
C
Lithosphere
Asthenosphere
D
12
Eroded
part
Transform valley
Transform fault
Transform fault
Eroded
part
Maximal
rise
Final
surface
Initial
surface
h
h
Fig. 4.
Models of vertical movements with formation of transverse ridges: (A) the thermal effect of the heated young segment of
the Mid-Atlantic Ridge [17, 21]; (B) thermal stress [18, 28]; (C) longitudinal melt flow [31]; (D) erosion of the lithospheric plate [9]:
(1) initial stage, (2) terminal stage.
The thermal stress
of the cooling of oceanic plates
in the course of their motion away from the spreading
centers results in reduction of their linear sizes (hori-
zontal contraction). The size reduction of a plate seg-
ment 150 km long can be as large as 300 m for a cooling
of
200°ë
[19, 29]. During the displacement along the
transform fault, marginal parts of the plate experience
alternating compression and extension (Fig. 4, model B).
The maximal stress is observed near axial parts of rift
valleys. The near-surface part of the lithosphere is sub-

GEOTECTONICS
Vol. 40
No. 1
2006
VERTICAL TECTONIC MOVEMENTS OF THE CRUST IN TRANSFORM FRACTURE ZONES 31
ject to compression, while the lithosphere proximal to
the Moho discontinuity is subject to extension [29]. As
a result, the arising stress bends the lithosphere. The
thermal stress caused by cooling and compression of
the lithosphere 20 Ma old results in its bending and a
rise of the transverse ridge by 150 m [24].
Friction-related heating.
The movement along
extended faults is accompanied by friction and release of
thermal energy that stimulates the temperature rise along
the fault plane by
200–400°ë
[18]. The thermal energy
resulting from the relative displacement of the northern
and southern walls of the Romanche FZ can provide a
rise of no more than 200 m in the transverse ridge. In this
case, the area subjected to heating and uplifting is 5 km
wide on both sides of the FZ’s axis [15].
Serpentinization–deserpentinization
of the upper
mantle rocks in transform FZs. As indicated from the
dredged samples, the transverse ridges are composed
largely of brecciated ultramafic rocks. Their serpentini-
zation gives rise to a significant increase in volume and
a decrease in density, phenomena which lead to a rise
of a transverse ridge. Complete serpentinization of a
5-km-thick peridotite block result in a rise of its upper
boundary by 2 km. The intensity of serpentinization
depends largely on the permeability of rock to seawater
and the process can reach the mantle/crust interface.
Fracture zones, being highly brecciated, are easily per-
meable to seawater. Deserpentinization accompanied
by subsidence of the transverse ridge may occur during
its drift near the heated intersection area. Serpentiniza-
tion of ultramafic rocks begins as the temperature drops
below 500
°
C, whereas deserpentinization begins as the
temperature rises above 500
°
C. In the Romanche FZ,
the isotherm at 500
°
C may have subsided below 10 km
[15]. Near the eastern intersection with the rift valley,
the thickness of the crust, which may be involved in
dehydration, is 6 km. Despite the significant contribu-
tion of serpentinization to the vertical component of
transverse-ridge movements and with the consideration
that this process usually affects no more than 50–80%
of ultramafic rocks, it is difficult to explain the forma-
tion of topography with an amplitude up to 4 km with-
out involving consideration of other mechanisms.
The longitudinal flow of a melt
beneath the MAR
axial zone. It may be assumed theoretically that a melt
moves beneath the MAR rift zone toward the intersec-
tion area and its motion is restricted by the cold plate,
phenomena that stimulates both the intense rise of the
plate margin and active volcanism with the vertical
growth of a transverse ridge [31] (Fig. 4, model C). The
transverse ridges actually host fresh basalts, but they are
composed largely of gabbroid and ultramafic rocks, a cir-
cumstance which prevents consideration of this mecha-
nism as a principal mode of transverse ridge formation.
Erosion of the lithosphere plate.
The erosion-related
unloading along a transform FZ results in the flexure
and rise of the plate margin along the transform bound-
ary. The rise does not exceed 35% of the height differ-
ence (
h
) between the levels of the lithosphere plate and
transform valley (Fig. 4, model D) [10]. The structure
of the transform fault provides for a certain step in
topography. Since the vertical surface does not remain
stable for a long time, the uplift will be subject to ero-
sion that results in the unloading and rising owing to
isostatic compensation. This mechanism is not univer-
sal but can reinforce other mechanisms. The difference
in height within most of the oceanic transform FZs
amounts to 60% (sometimes even more than 100%).
This model does not explain the origin of the initial
scarp across the strike of an FZ. Furthermore, the origin
of the transform valley itself is not quite clear. This phe-
nomenon is probably related to a strip of the anoma-
lously thin crust that formed at those ends of rift valleys
where dry spreading is prevalent. Extension perpendic-
ular to an FZ may result in formation of a valley, while
tectonic unloading may lead to the rise of FZ’s wall.
The erosion-related unloading explains only an addi-
tional rise: the shallower the fault valley, the greater the
contribution of tectonic unloading.
Mechanisms of vertical movements of transform
ridges caused by compression and extension along
unparallel converging and diverging boundaries of
transform faults or by changes in the direction of litho-
spheric plate motion with formation of variously ori-
ented transform zones are considered hereinafter for
the Romanche, Vema, and Kane FZs.
MOST PROBABLE CAUSES OF VERTICAL
MOVEMENTS IN THE ROMANCHE, VEMA,
AND KANE FRACTURE ZONES
Above, we considered several factors that can con-
trol the rise of transverse ridges in the Romanche and
Vema FZs. Most of them cannot provide the observed
amplitude of the crustal blocks’ rise.
The Romanche Fracture Zone.
The main cause of
vertical movements in the northern transverse ridge is
likely related to compressive and tensile stresses that
accompany the motion of the northern and southern
walls of the Romanche FZ relative each other. Such
stresses are evident from earthquakes within the FZ.
The lateral stresses may be induced by the following
factors: transform-fault bending, a slight change in the
direction of spreading, and changes in geometry of
plate boundaries adjusted to a new configuration in the
fault–ridge system.
The topographic maps of the Romanche FZ show
that its strike in the eastern part of the active segment
near
19°–20°
W changes from latitudinal to west–
northwestern (Fig. 1) and does not completely coincide
with the direction of motion of the lithospheric plates.
A difference of approximately
10°
results in compres-
sion and formation of a large high in this area during
movement along the fault. This phenomenon is also
indicated by mechanisms of recent earthquakes in this

32
GEOTECTONICS
Vol. 40
No. 1
2006
PEIVE
area. Similar processes could have operated in the past
as well.
The formation of highs A, B, C, and D in the north-
ern transverse ridge is related to the reorganization of
the FZ in the late Miocene (7–8 Ma). This stage corre-
sponded to the origination of the present-day dynamic
system of the Romanche FZ (Rom 2) with dying off and
closure of the paleo-Romanche (Rom 1) system [4, 15].
The factor most probably responsible for the rise of
blocks was compression of Rom 1 during the formation
of Rom 2 and movement along Rom 1 at an angle rela-
tive to the former’s azimuth (Fig. 5). In a rather short
Extension
New fault zone
Old fault zone
Old spreading
New spreading
axis
axis
Compression
New fault zone
Old fault zone
Old spreading
New spreading
axis
axis
A
B
Rift
valley
Old plate
Area under compression
Young plate
Rift valley
Old plate
Young plate
Fig. 5.
Vertical movements of transverse ridges determined by tectonic factors (change in the direction of plate motion, unparallel
boundaries of transform faults, etc.) in the Romanche FZ: (A) change in the stress field under a slight change in the direction of
spreading, (B) schematic interaction between old (northern) and young (southern) plates near the eastern intersection of the
Romanche FZ by oblique displacement along the fault.

GEOTECTONICS
Vol. 40
No. 1
2006
VERTICAL TECTONIC MOVEMENTS OF THE CRUST IN TRANSFORM FRACTURE ZONES 33
period (2–3 Ma), these blocks started to subside and
formed small deeps in the fault valley between 18
°
and
19
°
W. The change in stress fields was probably caused
by general geometric reorganization of the transform
fault–rift system in the eastern intersection area.
Block D, where the thick sedimentary sequence is
deformed into folds and complicated by low-angle
thrusts, exemplifies compressional deformation. Com-
pression in the junction area of the thick cold lithos-
phere of the northern wall of the fault and the young
thin lithosphere of its southern wall resulted in thrust-
ing of the young plate with the formation of a large
high. Model calculations confirm the theoretical proba-
bility of such an amplitude of the high [15].
The Vema Fracture Zone.
Among several potential
factors responsible for the formation of the southern
transverse ridge in the Vema FZ, the change in the
spreading direction accompanied by compressive and
tensile stresses is the most probable. It is assumed that
compression and formation of the southern transverse
ridge in this FZ was confined to the block between the
Vema transform FZ and the southerly located aban-
doned Lemm Fault [20]. Compression occurred before
10 Ma ago and was related to changes in geometry of
the fault–rift system [11] (Fig. 6). Since 10 Ma ago, the
position of the pole of the Central Atlantic opening
changed with reorganization of the spreading direction
and origination of the meridional tensile stress [30].
According to [20], the entire transverse ridge prob-
ably rose as a single block. This is indicated by similar
ages and depths of two carbonate platforms located on
the summit of the ridge at a distance of 100 km from
each other, as well as by principally similar internal
structure for a distance of over 100 km. The ridge is
sharply asymmetric in the longitudinal section. Its
northern slope is characterized by the practically undis-
turbed oceanic crust and upper mantle, while the south-
ern gentle slope is composed of basalts. The base of the
ridge is marked by a seismic boundary dipping south-
ward and nearly parallel to the oceanic floor. On the
basis of these data, the transverse ridge is interpreted as
an edge of the bent lithospheric plate separated from its
northerly part by the Vema FZ [20]. The period of
uplifting was probably less than 6.6 Ma.
The origin of Late Paleocene (55–58 Ma) limestone
in the middle part of the transverse ridge’s slope at a lat-
itude of
44°22
′
W is of particular interest. Under a
spreading rate of 1.1 cm/yr, only rocks of 35 Ma and
younger may occur at such a distance from the MAR
axis [17]. The occurrence of the Late Paleocene rocks
can be explained by complicated reorganization of the
Vema FZ with a change in the motion of this block from
0 Ma
Axis of the Mid-Atlantic
Ridge
Vema Fault
Carbonate
bank
Vema transverse ridge
Axis of the Mid-Atlantic
Ridge
Abandoned
nodal
depression
Active
nodal
depression
Nontransform
displacement
Orientation of structural elements
Lemm Fault scarp
Axis of the Mid-Atlantic
Ridge
2.2 Ma
Axis of the Mid-Atlantic
Ridge
Mid-Atlantic Ridge
Abandoned rift of the
6.6 Ma
0 50 km
Fig. 6.
Schematic evolution of the Vema and Lemm FZs [19].

34
GEOTECTONICS
Vol. 40
No. 1
2006
PEIVE
the eastern to the western direction after the northward
jump of the active segment of the fault [13].
The Kane Fracture Zone.
The topography of the
transverse ridge in the eastern part of the Kane FZ is
consistent with the model of the lithosphere flexure as
a result of tectonic erosion during normal faulting [25].
Approximately 8 Ma ago, the Kane FZ experienced
extension at its eastern intersection, an occurrence
which stimulated the formation of normal faults that
were displaced progressively eastward along the fault
for tens of kilometers (Fig. 7). The faults disturbed iso-
static equilibrium and brought about a rise of the lying
wall and subsidence of the hanging wall. The lithos-
phere flexure as a response to this faulting event led to
the formation of the transverse ridge in the Kane FZ
within the old plate and depressions within the young
plate. The amplitude of vertical displacements is pro-
portional to that of displacements along normal faults.
Extension in the oceanic crust between the Kane and
21°30
′
N FZs was in progress until 3 Ma ago, when the
change of the spreading direction terminated. Since
then, the plate’s edge has become stable [25]. This
model works well for insignificant (<5 km) extension
and normal faults that dip at angles of
<45°
; variations
in the crustal thickness virtually do not effect interpre-
tation. The amplitude of the ridge’s rise depends on the
magnitude of extension, the displacement along the
fault, and the thickness of the old plate. As such, this
8 Ma
Kane FZ
Old
direction of spreading
New
21°30′ N FZ
3 Ma
Kane FZ
21°30′ N FZ
Transverse ridge
Extensional depressions
Fig. 7. Model of the formation of transverse ridges in the Kane and 21°30′ N FZ [24].

GEOTECTONICS Vol. 40 No. 1 2006
VERTICAL TECTONIC MOVEMENTS OF THE CRUST IN TRANSFORM FRACTURE ZONES 35
model can be considered as a variety of the erosion
model.
The advantage of this model consists in the possibil-
ity of explaining the irregular distribution of transverse
ridges. Repeated changes in the direction of spreading
of lithospheric plates resulted in the discrete formation
of transverse ridges within a single fault zone.
CONCLUSIONS
The most significant vertical movements of the oce-
anic crust are characteristic of transverse ridges in
transform FZs, although they are also notable within
some MAR depressions and old deep-sea basins. The
amplitude of displacements is substantially higher in
comparison with that related to the cooling of lithos-
pheric plates.
Detailed studies have shown that alternation of the
compression and extension regimes that results from
unparallel boundaries of the transform faults and unsta-
ble configuration of the junction area between a rift and
a fault (changes in the spreading direction, jump and
progradation of rift valleys, migration of the transform
valley) were the main factors responsible for vertical
movements during the formation of the northern trans-
verse ridge in the Romanche FZ.
Among all factors providing for the formation of the
southern transverse ridge in the Vema FZ, the changes
in the spreading direction accompanied by reorganiza-
tion in geometry of the fault/rift junction area—progra-
dation and rollback of rifts, the jump of the MAR axis,
and the displacement of transform fault boundaries—
are the most probable.
Longitudinal extension in the eastern intersection
area of the Kane FZ stimulated normal faulting in the
fracture zone itself. The flexure of the lithosphere as a
response to the normal faulting gave rise to the formation
of transverse ridge in the Kane FZ within the old plate,
while depressions were formed within the young plate.
In general, FZs with active segments over 100 km
long are characterized by lateral extension and com-
pression perpendicular to the direction of the main dis-
placement that are related to the slight change in the
direction of spreading. In the course of extension, the
thick cold lithosphere opposite to an intersection area
precluded rift progradation and formation of a trans-
form FZ with a different strike. The duration, ampli-
tude, and position of a tension zone depend on (1) the
change in the angle of spreading direction, (2) the
length of a transform FZ, and (3) the spreading rate and
duration of the stage when prograding rifts acquire a
new position and a new transform FZ fits a new spread-
ing center [9].
It is impossible to name a single factor responsible
for vertical movements in particular structural features.
In most cases, vertical displacements are controlled by
combination of factors, among which the lateral tensile
stress associated with the formation of a fault valley
and the rise of one of its walls to form a transverse ridge
is the most important (Fig. 8). The local compressive
stresses that arise owing to geodymamic changes in the
motion of lithospheric plates in the course of their
spreading may play a substantial role in the formation
of large highs.
ACKNOWLEDGMENTS
This study was carried out under Program no. 14 of
the fundamental investigations of the Presidium of the
Russian Academy of Sciences, “Principle Problems of
Oceanology: Geology, Physics, Biology, and Ecology,”
and supported by the Russian Foundation for Basic
Research (project no. 03-05-64159) and the Ministry of
Industry and Science of the Russian Federation.
REFERENCES
1. A. O. Mazarovich, Geological Structure of the Central
Atlantic: Faults, Volcanic Edifices, and Seafloor Defor-
mations (Nauchnyi Mir, Moscow, 2000) [in Russian].
2. A. A. Peive, G. N. Savelieva, S. G. Skolotnev, and
V. A. Simonov, “Structure and Deformations of the
Crust–Mantle Boundary Zone in the Vema Fracture
Zone, Central Atlantic,” Geotektonika 35 (1), 16–35
(2001) [Geotectonics 35 (1), 12–29 (2001)].
3. A. A. Peive, G. N. Savelieva, S. G. Skolotnev, and
V. A. Simonov, “Tectonics and Origin of the Oceanic
Crust in the Region of “Dry” Spreading in the Central
Atlantic (7°10′–5° N),” Geotektonika 37 (2), 3–25 (2003)
[Geotectonics 37 (2), 75–94 (2003)].
4. A. S. Perfil’ev, A. A. Peive, Yu. M. Pushcharovsky, et al.,
“Romanche Fracture Zone: Structure, Origin, and Geo-
dynamics,” Geotektonika 29 (4), 3–14 (1994).
Transform
valley
Transverse
ridge
Fig. 8. Conceptual model of formation of a transform valley and a transverse ridge.

36
GEOTECTONICS Vol. 40 No. 1 2006
PEIVE
5. Yu. M. Pushcharovsky, A. O. Mazarovich, and S. G. Sko-
lotnev, “Neotectonics of the Oceanic Floor of the Central
Atlantic,” Geotektonika 39 (2), 1–14 (2005) [Geotecton-
ics 39 (2), 99–111 (2005)].
6. Yu. M. Pushcharovsky, A. A. Peive, Yu. N. Raznitsin, and
E. S. Bazilevskaya, Fracture Zones in the Central Atlan-
tic (GEOS, Moscow, 1995) [in Russian].
7. Yu. M. Pushcharovsky, S. G. Skolotnev, A. A. Peive,
N. S. Bortnikov, E. S. Bazilevskaya, and A. O. Mazarov-
ich, Geology and Metallogeny of the Mid-Atlantic
Ridge: 5°–7° N (GEOS, Moscow, 2004) [in Russian].
8. Structure of the Doldrums Fracture Zone, Central Atlan-
tic (Nauka, Moscow, 1991) [in Russian].
9. G. Baines, M. Cheadle, B. John, and N. Kusznir, “Origin
of Anomalous Uplift at Inside Corner Highs: the Impor-
tance of Transform Parallel Normal Faulting and Trans-
verse Ridge Formation,” EOS Trans. AGU 83 (47),
F1332 (2002).
10. C. Basile and P. Allemand, “Erosion and Flexural Uplift
along Transform Faults,” Geophys. J. Int. 151, 646–653
(2002).
11. E. Bonatti, “Vertical Tectonism in Oceanic Fracture
Zones,” Earth Planet. Sci. Lett. 37, 369–379 (1978).
12. E. Bonatti, “Subcontinental Mantle Exposed in the
Atlantic Ocean on St. Peter-Paul Islets,” Nature 345
(6278), 800–802 (1990).
13. E. Bonatti and K. Crane, “Oscillatory Spreading Expla-
nation of Anomalously Old Uplifted Crust Near Oceanic
Transforms,” Nature 300, 343–345 (1982).
14. E. Bonatti, M. Ligi, D. Brunelli, et al., “Mantle Thermal
Pulses below the Mid-Atlantic Ridge and Temporal Vari-
ations in the Formation of Oceanic Lithosphere,” Nature
423, 499–505 (2003).
15. E. Bonatti, M. Ligi, L. Gasperini, et al., “Transform
Migration and Vertical Tectonics at the Romanche Frac-
ture Zone, Equatorial Atlantic,” J. Geophys. Res. 99,
21779–21802 (1994).
16. E. Bonatti, Y. Raznitsin, G. Bortoluzzi, et al., “Geologi-
cal Studies of the Romanche Transform (Equatorial
Atlantic): A First Report,” Giorn. Geol. 53, 31–48
(1993).
17. E. Bonatti, R. Sartori, and A. Boersma, “Vertical Crustal
Movements at the Vema Fracture Zone in the Atlantic:
Evidence from Dredged Limestones,” Tectonophysics
91, 213–232 (1983).
18. Y. J. Chen, “Thermal Model of Oceanic Transform
Faults,” J. Geophys. Res. 93, 8839–8851 (1988).
19. B. J. Collette, “Thermal Contraction Joints in a Spread-
ing Seafloor as Origin of Fracture Zones,” Nature 251,
299–300 (1974).
20. K. Kastens, E. Bonatti, D. Caress, et al., “The Vema
Transverse Ridge (Central Atlantic),” Marine Geophys.
Res. 20, 533–556 (1998).
21. K. Kastens, K. Macdonald, S. P. Miller, and P. Fox,
“Deep Tow Studies of the Vema Fracture Zone, 2. Evi-
dence for Tectonism and Bottom Currents in the Sedi-
ments of the Transform Valley Floor,” J. Geophys. Res.
91, 3355–3367 (1986).
22. K. E. Louden and D. W. Forsyth, “Thermal Conduction
across Fracture Zones and Gravitational Edge Effect,”
J. Geophys. Res. 81, 4869–4874 (1976).
23. D. Nurberg and R. D. Muller, “The Tectonic Evolution
of the South Atlantic from Late Jurassic to Present,” Tec-
tonophysics 191, 27–55 (1991).
24. E. M. Parmentier and W. F. Haxby, “Thermal Stresses in
the Oceanic Lithosphere: Evidence from Geoid Anoma-
lies at Fracture Zones,” J. Geophys. Res. 91, 7193–7204
(1986).
25. R. A. Pockalny, P. Gente, and R. Buck, “Oceanic Trans-
verse Ridges: A Flexural Response to Fracture-Zone-
Normal Extension,” Geology 24, 71–74 (1996).
26. P. D. Rabinovitz and J. Lagabreque, “The Mesozoic
South Atlantic and the Evolution of Its Continental Mar-
gins,” J. Geophys. Res. 84, 5973–6002 (1979).
27. D. T. Sandwell and W. H. Smith, “Marine Gravity
Anomaly from Geosat and ERS-1 Satellite Altimetry,”
J. Geophys. Res. 102, 10039–10054 (1997).
28. J. G. Sclater, R. N. Anderson, and M. L. Bell, “Elevation
of Ridges and Evolution of the Central Eastern Pacific,”
J. Geophys. Res. 76, 7888–7915 (1971).
29. D. L. Turcotte, “Are Transform Faults Thermal Contrac-
tion Cracks?,” J. Geophys. Res. 79, 2573–2577 (1974).
30. T. Van Andel, R. Herzen, and J. Phillips, “The Vema
Fracture Zone and the Tectonics of Transverse Shear
Zones in Oceanic Crustal Plates,” Mar. Geophys. Res. 1,
261–283 (1971).
31. P. R. Vogt and G. L. Johnson, “Transform Faults and
Longitudinal Flow Below the Mid-Ocean Ridge,”
J. Geophys. Res. 80, 1399–1428 (1975).
Reviewers: N.V. Koronovskii and S.V. Ruzhentsev