INTRODUCTIONThe Southwest Indian Ridge (SWIR) is the
type example of an ultraslow spreading center with a current full spreading rate of 14 km/m.y. But how long has this ridge been ultraslow? Previously presented spreading histories for the SWIR indicate faster spreading rates prior to anomaly 18 time (A18, ca. 40 Ma; Cande and Kent, 1995) than at present, with a decrease from slow to ultraslow sometime between A13 (ca. 33 Ma) and A6 (ca. 20 Ma) (Bergh and Norton, 1976; Fisher and Sclater, 1983; Molnar et al., 1988; Patriat and Segoufi n, 1988). This lack of precision could be attributed to the fact that magnetic anomalies at ultraslow ridges are relatively diffi cult to identify, but it is more likely due to lack of data combined with the deceivingly linear trends of the SWIR fracture zones, which were thought to indicate a single, simple spreading phase during the last 40 m.y.
At fi rst glance, the smooth curvilinear frac-ture zone trends of the SWIR for ages <40 Ma appear consistent with stable plate motion, but newly identifi ed magnetic anomalies tell a differ-ent story. In this work we present evidence that a dramatic spreading rate decrease occurred along the SWIR ca. 24 Ma and, more generally, that a major change in spreading rate can happen with-out apparent change in spreading direction. We also discuss this event within the context of global plate motion at the time, the possibility of “event propagation” along plate tectonic boundaries, and the correlation of the spreading rate change at the SWIR with global plate motion events.
DATA ANALYSISThis study uses previously collected magnetic
anomaly profi les (Cannat et al., 2006; Hosford et al., 2003; Sauter et al., 2001; Sclater et al.,
1997) and unpublished transit ship tracks. At slow to ultraslow spreading ridges, complete and easily identifi able magnetic anomaly sequences are rare. They are most often observed along narrow swaths of seafl oor that form the central part of spreading segments that, due to segment propagation, are not always oriented parallel to fracture zones. It is therefore unusual that
survey transects are so well positioned as to record distinct identifi able magnetic anomaly sequences. As a result, profi les exhibiting long uninterrupted series of magnetic anomalies at the SWIR are few, especially on its remote southern fl ank.
Anomaly shapes at slow spreading ridges are particularly sensitive to the frequency of magnetic fi eld inversions. Since A18 time, magnetic fi eld inversions have been relatively frequent, which tends to make the forms of A13, A8, and A6 more diffi cult to identify than those of the A21–A18 sequence. We therefore began anomaly identifi cation by calculating two models: one with a rate of ~29 km/m.y. to produce anomaly forms that correspond well with the easily recognizable A21–A18 magnetic anomaly sequence, the other with the present ultraslow spreading rate of ~15 km/m.y. for the period A5–A0 (e.g., Lemaux et al., 2002;
From slow to ultraslow: A previously undetected event at the Southwest Indian Ridge at ca. 24 MaPhilippe Patriat Laboratoire de Géosciences Marines, CNRS-UMR 7154, Institut de Physique du Globe de Paris,
4 place Jussieu 75252 Paris cedex 05, FranceHeather Sloan Environmental, Geographic, and Geological Sciences, Lehman College, City University of New York,
250 Bedford Park Blvd., Bronx, New York 10469, USADaniel Sauter Institut de Physique du Globe de Strasbourg, UMR7516 CNRS-ULP, Ecole et Observatoire des Sciences
de la Terre, 5 rue Descartes 67084 Strasbourg cedex, France
ABSTRACTChanges in plate motion are thought to be recorded in the trend of fracture zones, even
though fracture zones provide no information about the spreading rate. Using newly compiled published and unpublished magnetic data from the Southwest Indian Ridge, we calculated fi nite rotation poles for A13, A8, and A6, from which we determined a 50% decrease in spreading rate from slow to ultraslow at ca. 24 Ma not accompanied by a signifi cant change in spreading direc-tion. This spreading rate decrease is concurrent with changes in plate motions at only two of the four adjoining plate boundaries. Finally, we discuss the possible relationships of this event with other absolute or relative plate motion events that occurred at ca. 24 Ma at the global scale.
Keywords: mid-ocean ridges, global tectonics, kinematics, magnetic anomalies.
3
4
5
6Dep
th (
km)
–200–100
0100200
Mag
netic
ano
mal
y (n
T)
–200–100
0100200
2
3
4
5
–200–100
0100200
100 km
A5A6A8 A5 A6 A8
A6 A8 A13 A18 A21A6A8A13A18A21
29 km/m.y.
15 km/m.y.15 km/m.y.
Dep
th (
km)
Mag
netic
an
omal
y (n
T)
Profile md47
N S
100 km
md47
SWIRAfricanPlate
Antarctic Plate
Africa
SEIR
CIR
Figure 3
Figure DR1
RTJ
Figure 1. Magnetic anomaly profi le md47 compared with variable-spreading-rate syn-thetic anomaly profi les. Profi le md47 magnetic anomaly forms A21 to A8 correspond to synthetic profi le calculated with a spreading rate of 29 km/m.y. (bottom). The 15 km/m.y. synthetic profi le (top) matches profi le md47 A6 to A0. Shading indicates the period during which the spreading rate decrease occurred. Map shows location of profi le md47, Figure 3, and Figure DR1 (see footnote 1). CIR—Central Indian Ridge; RTJ—Rodriguez triple junction; SWIR—Southwest Indian Ridge; SEIR—Southeast Indian Ridge.
208 GEOLOGY, March 2008
Schlich and Patriat, 1971). Using these two sets of anomaly identifi cations as anchors, we adjusted the intervening spreading rate to obtain synthetic anomaly forms that closely resemble observed forms for the remaining time periods (A18–A5). The rate adjustments were made by comparing the synthetic model to the 1100-km-long profi le md47, the only long profi le oriented perpendicular to the spreading axis that does not cross discordant zones (Fig. 1). For ages ca. 33–26 Ma (A13–A8), the form of the md47 anomalies corresponds to the synthetic profi le with a spreading rate of 29 km/m.y. For ages <20 Ma (A6 and younger), the 15 km/m.y. syn-thetic profi le is a much better match. We were then able to validate the variable spreading rate model at shorter, ridge-perpendicular profi les (Fig. 2). The synthetic anomaly forms match the observed anomalies well, particularly the A13–A7 sequence, the most diffi cult to identify along the SWIR. Having established the validity of our variable spreading rate model with obser-vations along well-oriented profi les, we used it to identify A6, A8, and A13 along the remaining profi les compiled between the Prince Edward and Melville fracture zones, making many new or improved identifi cations (Fig. 3).
SPREADING RATE CALCULATION: FROM SLOW TO ULTRASLOW
To obtain the spreading rate and direction for the periods A13–A8 and A6–A5, we determined the fi nite rotation poles for A13, A8, and A6 (see the GSA Data Repository1) and used a pre-viously calculated pole for A5 (Lemaux et al., 2002). Finite rotation poles were calculated by superposing sets of conjugate anomaly identifi -cations at as great a distance as possible along the ridge to produce well-constrained poles that accurately describe plate motion without the use of fracture zone trends (Patriat and Segoufi n, 1988). We chose one set of conjugate anomaly identifi cations close to the Melville fracture zone and another at a distance of more than 2000 km, near the Prince Edward fracture zone. The accu-racy of the resulting fi nite poles is determined by calculating 95% confi dence ellipses for all identifi cations using an accepted method (Royer and Chang, 1991) (Fig. 3; see GSA Data Repos-itory for parameters of confi dence ellipses). The error ellipses indicate a high degree of accuracy. The poles produce satisfactory superposition of magnetic anomalies: Identifi cations made within a single spreading segment on one plate,
once rotated, fall in the same segment on the conjugate plate, indicating that the plate motion determination is self-consistent (Fig. 3).
Spreading rate and direction were calculated using the stage poles (A13–A8 and A6–A5) deduced from the fi nite rotation poles. For the section of the SWIR near longitude 51°E, the spreading direction for the period between A13 and A8 (N16°E) varies only 13° from that cal-culated for A6 to A5 (N3°E). In sharp contrast to this small change in the spreading direc-tion, the spreading rate for this same section decreases ~50%, from 29.5 km/m.y. for the period A13–A8 to 14.2 km/m.y. between A6 and A5, the present ultraslow spreading rate. Assuming that the change in rate was instan-taneous at some time after A8 and before A6, we calculate that it occurred at 24.2 Ma, A6C time (see GSA Data Repository). The close cor-respondence between the data and the synthetic anomaly forms, which were calculated with an instantaneous spreading rate change (Fig. 2), lends support to this assumption.
EVIDENCE FROM TRIPLE JUNCTION EVOLUTION
At either end of the SWIR is a triple junc-tion of divergent plate boundaries, the Rodri-guez triple junction to the east where the SWIR meets the Central Indian Ridge (CIR) and the Southeast Indian Ridge (SEIR) and the Bouvet triple junction to the west where it meets the Southern Mid-Atlantic Ridge (SMAR) and the South American–Antarctic Ridge (SAAR). During the period between A13 and A5, the spreading rate and direction of the fastest
ridges at each of the triple junctions (SEIR and SMAR) remained roughly constant (Patriat and Segoufi n, 1988; Shaw and Cande, 1990). Given these conditions, the velocity triangles for both triple junctions indicate that large changes in spreading direction on the CIR and SAAR should have resulted from the 50% decrease in spreading rate of the SWIR, a pre-diction that appears to be supported by kine-matic reconstruction and the satellite-derived gravity map of Smith and Sandwell (1997) (Fig. 4). Trends of the Egeria fracture zone and other major fracture zones at this time at the CIR shift from WSW-ENE to SW-NE. Spread-ing direction deduced from CIR fi nite poles changed from N87°E to N56°E (Patriat and Segoufi n, 1988). The SAAR spreading direc-tion changed from NW-SE to E-W just before A6 time (20 Ma) when the ridge broke into very short segments offset by long transform faults (Barker and Lawver, 1988).
Note that the Bouvet triple junction veloc-ity triangle has been constructed using rates and directions deduced from the rotation poles determined here and so does not take into account postulated relative motion on the Nubia-Somalia boundary (Lemaux et al., 2002; Royer et al., 2006). Our new A13, A8, A6, and A5 identifi cations west of the Andrew Bain fracture zone fi t unexpectedly well with their conjugate identifi cations when rotated about the newly calculated poles. Thus, we fi nd no evidence of signifi cant differential motion between the east and west of the Andrew Bain fracture zone for ages ≥11 Ma. This discrepancy with previous fi ndings may be explained by differing iden-
D profile th99aC profile ata9510_TR
A profile mdop04_TA
Distance from the axis (km)
Mag
netic
ano
mal
y (n
T)
-200-100
0100200
-200-100
0100200
B profile md66
N S29 km/m.y. 29 km/m.y.14.5 km/m.y. 14.5 km/m.y.
A 18 A 13 A 8 A 6 A 5 A 18A 13A 8A 6A 5
Dep
th (
km)
-200-100
0100200
100 200 300 400 5006
0100200300400500 0
3
4
5
Figure 2. Magnetic anomaly identifi cations along four profi les (see Fig. 3 for locations) using variable rate model (bottom): 29 km/m.y. for A21 to A6C (24 Ma) and 14.5 km/m.y. for A6 to A1 (present). Shading and model as in Figure 1.
1GSA Data Repository item 2008054, supplemen-tary methods detailing the synthetic model calcula-tion, time of spreading rate change calculation, fi nite poles and confi dence ellipse parameters, and new magnetic anomaly identifi cations west of Andrew Bain fracture zone, is available online at www.geosociety.org/pubs/ft2008.htm, or on request from [email protected] or Documents Secretary, GSA, P.O. Box 9140, Boulder, CO 80301, USA.
GEOLOGY, March 2008 209
tifi cations of A5 along profi les on the African plate: We interpret as A5B the anomalies previ-ously identifi ed as A5 (see Fig. DR1).
DOES THE SWIR SPREADING RATE CHANGE CORRESPOND TO A GLOBAL PLATE TECTONIC EVENT?
Our fi ndings indicate that a change in plate motion occurred along the SWIR ca. 24 Ma: a 15 km/m.y. decrease of the already slow spread-ing rate, but no signifi cant change of spreading direction. It is worth noting that another similar plate motion event (PME), change in spreading rate, but not direction, has been observed in the Indian Ocean at A31 time (ca. 68 Ma) when the spreading rate became ultrafast along the SEIR (Patriat, 1987). While marked changes in plate boundary geometry and plate motion are observed at the CIR and SAAR, we fi nd that little or no change occurred on the SEIR or the SMAR at the time of the SWIR event, indicating a PME may occur along a particular path affecting some plate boundaries but not others. Whether the SWIR PME was part of a global PME occurring at ca. 24 Ma can only be assessed by analyzing evidence of transmission along adjoining plate boundaries.
Several important events occurred on the global plate boundary system between 30 and 20 Ma. Rifting began separating the African and Arabian plates at ca. 30 Ma, followed by the opening of the Gulf of Aden along the Sheba Ridge, which began spreading at <20 Ma (d’Acremont et al., 2006). Silver et al. (1998) proposed an abrupt slowing of absolute motion of the African plate when it collided with Eurasia at ca. 30 Ma. A major PME took place in the Pacifi c at ca. 24 Ma: The Farallon plate split to form the Nazca and Cocos plates (Handschumacher, 1976), accompanied by a marked change in spreading direction appar-ent in both magnetic anomalies and fracture zone trends. In the northern Pacifi c the spread-ing rate decreased dramatically (Cande and Kent, 1992), while in the southwestern Pacifi c the spreading direction changed along the Macquarie Ridge (Cande and Stock, 2004). Are these events related to the SWIR PME? The answer lies in the tracing of the path of plate motion change along the adjoining net-work of plate boundaries and evaluating it in light of global absolute plate motion.
The path of the SWIR PME can be traced along the CIR and SAAR. Transmission via
the CIR reaches successively the Carlsberg and Sheba Ridges. A dramatic spreading rate increase occurred at the Carlsberg Ridge, the boundary between India and Africa (Mercuriev et al., 1996), in conjunction with the opening of the Gulf of Aden along the Sheba Ridge. Trans-mission of the SWIR PME appears to dissipate along this nascent African-Arabian boundary. Transmission via the SAAR leads to the Scotia plate boundaries, composed, at A6C time, of a complex and evolving system of divergent and convergent boundaries (Engles et al., 2005; Livermore and Woollett, 1993). As yet we have no direct evidence that the PME was transmit-ted beyond the boundaries of the Scotia plate to the boundaries of the Pacifi c plate. Ultimately, transmission of a PME may be absorbed when the event path reaches diffuse plate boundaries that typically surround small plates (Gordon, 1998), which have been referred to as “buffer plates” by Anderson (2002).
36°E
36°E
40°E
40°E
44°E
44°E
48°E
48°E
52°E
52°E
56°E
56°E
60°E
60°E
64°E
64°E
48°S 48°S
44°S 44°S
40°S 40°S
36°S 36°S
32°S 32°S
28°S 28°S
G
G
Mel
ville
FZ
Nov
ara
FZ
Atla
ntis
II F
Z
Gau
ss F
Z
azel
le F
Z
allie
ni F
Z
Indo
med
FZ
RTJ trace
RTJ trace
Dis
cove
ry F
Z
Prin
ce E
dwar
d F
Z
Mar
ion
FZ
Eric
Sim
pson
FZ
Magnetic
Anomalies
A6
A8
A13
md47
c
a
d
b
60°W 50°W 40°W
0°
10°N
20°N
Finiterotationpoles
A5A6A8A13
South America
Figure 3. Magnetic anomaly identifi cation A6, A8, and A13 plotted on the Southwest Indian Ridge tectonic map. Background bathymetry derived from satellite sea-surface altimeter (Smith and Sandwell, 1997). Anomaly identifi cations appear as solid symbols, and conju-gate rotated anomaly identifi cations appear as open symbols. Dashed lines indicate location of profi le md47 and Figure 2 profi les. a—profi le mdop04_TA; b—profi le md66; c—profi le ata9510_TR; d—profi le th99a; FZ—fracture zone; RTJ—Rodriguez triple junction. Inset shows location of fi nite rotation poles and 95% confi dence ellipse for Africa-Antarctica rela-tive plate motions. Poles and ellipses for A6, A8, and A13 were calculated for this work; the A5 pole and ellipse are from Lemaux et al. (2002).
CIR A8-A13
CIR A5-A6
SW
IR
SWIR
SMAR
A5-A6SAAR A8-A13
SEIR
60°E
60°E
62°E
62°E
64°E
64°E
66°E
66°E
68°E
68˚E
24°S 24°S
22°S 22°S
20°S 20°S
A B
D
18°W
18°W
15°W
15°W
12°W
12°W
9°W
9°W
6°W
6°W
3°W
3°W
0°
0°
60°S 60°S
57°S 57°S
C
A6A8
SAAR
SWIR
SEIR
CIR
SM
AR
Scotia Pl.
African Pl.
Antarctic Plate
S. Am.Plate
Ind.Aus.Pl.
c
b
CIR
Figure 4. Predicted and observed changes at the Rodriguez and Bouvet triple junctions in response to rate change at the SWIR. A: Location map. The Rodriguez triple junc-tion is to the east where the SWIR meets the CIR and the SEIR. The Bouvet triple junction is to the west where the SWIR meets the SMAR and the SAAR. B: Triple junction velocity tri-angles predict spreading direction change at the CIR and SAAR, constant plate motion for the SMAR and SEIR assumed. The period A13–A8 is shown in red and A6–A5 in black. C and D: Evidence for plate motion change between A8 and A6 in satellite-derived gravity maps of CIR (C) and SAAR (D). CIR—Central Indian Ridge; SEIR—Southeast Indian Ridge; SWIR—Southwest Indian Ridge; SMAR—Southern Mid-Atlantic Ridge; SAAR—South American–Antarctic Ridge.
210 GEOLOGY, March 2008
It is diffi cult to imagine that events in the Pacifi c occurred independently of those in the Indian Ocean. The hypothesis of a relationship between these events assumes a common origin for the postulated global event and the possi-bility of its transmission taking several mil-lion years. The origin could be the collision of Africa with Eurasia. The sudden halt of African absolute motion in combination with continuing unchanged spreading at the SMAR led to a rapid increase in the westward absolute motion of the South American plate and deformation along its western leading edge (Silver et al., 1998), result-ing in relative motion change at this boundary. The potential for tracing a continuous PME path from the SWIR PME to Pacifi c reorganization is intriguing and remains to be confi rmed by global kinematic reconstruction at ca. 24 Ma.
CONCLUSIONSOur fi ndings indicate a decrease in the SWIR
spreading rate from slow to ultraslow at ca. 24 Ma with no signifi cant change in spreading direction. A path of correlative changes in plate motions can be traced along adjoining plate boundaries in both directions, although only two of the four immediately adjacent boundaries show any sig-nifi cant adjustment. This absence of evidence of plate motion change in the SWIR fracture zone trends and in spreading at the SMAR and SEIR may have disguised the importance of this event. Are there common characteristics for plate boundaries that remain stable while their triple junction neighbors respond to PMEs by reorga-nization and plate motion change?
It seems likely that the SWIR PME is part of a global event that may have been initiated by the collision of Africa with Eurasia beginning at ca. 30 Ma. The apparent cause-and-effect link between events occurring at 30 Ma and 24 Ma raises another question: What is the duration of the transition from an assumed global event initiated by collision to the fi nal new stable plate motion phase?
Study and comparison of global PMEs may represent a new and challenging direction lead-ing to a better understanding of the driving forces of plate tectonics.
ACKNOWLEDGMENTSThis study was made possible by data collected
during transits of R/V Marion-Dufresne. We are grate-ful to Institut Paul Emile Victor and the chief scientists for their cooperation, especially B. Ollivier for collect-ing the data. Most of the profi les west of 45°E come from the data set of H. Bergh. M. Tivey provided data from surveys east of the Atlantis fracture zone. We also thank Steve Cande for insightful discussion and the reviewers, Anne Briais and Sharon Mosher, for their thoughtful critique and helpful comments. Insti-tut de Physique du Globe contribution 2300.
REFERENCES CITEDAnderson, D.L., 2002, How many plates?: Geology,
v. 30, p. 411–413, doi: 10.1130/0091-7613(2002)030<0411:HMP>2.0.CO;2.
Barker, P.F., and Lawver, L.A., 1988, South American–Antarctic plate motion over the past 50 Ma, and the evolution of the South American–Antarctic Ridge: Geophysical Journal International, v. 94, p. 377–386, doi: 10.1111/j.1365-246X.1988.tb02261.x.
Bergh, H.W., and Norton, I.O., 1976, Prince Edward fracture zone and the evolution of the Mozam-bique Basin: Journal of Geophysical Research, v. 81, p. 5221–5239.
Cande, S.C., and Kent, D.V., 1992, A new geomag-netic timescale for the Late Cretaceous and Cenozoic: Journal of Geophysical Research, v. 97, p. 13,917–13,951.
Cande, S.C., and Kent, D.V., 1995, Revised calibra-tion of the geomagnetic polarity timescale for the Late Cretaceous and Cenozoic: Journal of Geophysical Research, v. 100, p. 6093–6095, doi: 10.1029/94JB03098.
Cande, S.C., and Stock, J.M., 2004, Cenozoic recon-structions of the Australian–New Zealand–South Pacifi c sector of Antarctica, in Exon, N., et al., eds., The Cenozoic Southern Ocean: Tectonics, sedimentation and climate change between Australia and Antarctica: American Geophysical Union Geophysical Monograph 151, p. 5–18.
Cannat, M., Sauter, D., Mendel, V., Ruellan, E., Okino, K., Escartin, J., Combier, V., and Baala, M., 2006, Modes of seafl oor genera-tion at a melt-poor ultraslow-spreading ridge: Geology, v. 34, p. 605–608, doi: 10.1130/G22486.1.
d’Acremont, E., Leroy, S., Maia, M., Patriat, P., Beslier, M.-O., Bellahsen, N., Fournier, M., and Gente, P., 2006, Structure and evolution of the eastern Gulf of Aden: Insights from mag-netic and gravity data (Encens-Sheba MD117 cruise): Geophysical Journal International, v. 165, p. 786–803, doi: 10.1111/j.1365-246X.2006.02950.x.
Engles, G., Livermore, R., Fairhead, J.D., and Morris, P., 2005, Tectonic evolution of the west Scotia Sea: Journal of Geo-physical Research, v. 110, B02401, doi: 10.1029/2004JB003154.
Fisher, R.L., and Sclater, J.G., 1983, Tectonic evolu-tion of the Southwest Indian Ocean since the Mid Cretaceous: Plate motions and stability of the pole of Antarctica/Africa for the last 80 Myr: Geophysical Journal of the Royal Astronomical Society, v. 73, p. 553–576.
Gordon, R.G., 1998, The plate tectonic approxima-tion: Plate nonrigidity, diffuse plate bound-aries, and global plate reconstructions: Annual Review of Earth and Planetary Sciences, v. 26, p. 615–642, doi: 10.1146/annurev.earth.26.1.615.
Handschumacher, D.W., 1976, Post-Eocene plate tectonics of the eastern Pacifi c: American Geo-physical Union Geophysical Monograph 19, p. 177–202.
Hosford, A., Tivey, M., Matsumoto, T., Dick, H., Schouten, H., and Kinoshita, H., 2003, Crustal magnetization and accretion at the Southwest Indian Ridge near the Atlantis II fracture zone, 0–25 Ma: Journal of Geophysi-cal Research, v. 108, p. EMP 9-1–EMP 9-23, 10.1029/2001JB000604.
Lemaux, J., Gordon, R.G., and Royer, J.Y., 2002, Location of the Nubia-Somalia boundary along the Southwest Indian Ridge: Geology, v. 30, p. 339–342, doi: 10.1130/0091-7613(2002)030<0339:LOTNSB>2.0.CO;2.
Livermore, R.A., and Woollett, R.W., 1993, Seafl oor spreading in the Weddell Sea and southwest Atlantic since the Late Cretaceous: Earth and
Planetary Science Letters, v. 117, p. 475–495, doi: 10.1016/0012-821X(93)90098-T.
Mercuriev, S., Patriat, P., and Sochevanova, N., 1996, Evolution de la dorsale de Carlsberg: Evidence pour une phase d’expansion très lente entre 40 et 25 Ma (A18 à A7): Oceanologica Acta, v. 19, p. 1–13.
Molnar, P., Pardo-Casas, F., and Stock, J., 1988, The Cenozoic and Late Cretaceous evolu-tion of the Indian Ocean: Uncertainties in the reconstructed positions of the Indian, African and Antarctic plates: Basin Research, v. 1, p. 23–40.
Patriat, P., 1987, Reconstitution de l’évolution du système de dorsales de l’océan Indien par la méthode de la cinématique des plaques: Paris, Terres Australes et Antarctiques Françaises, 308 p.
Patriat, P., and Segoufi n, J., 1988, Reconstruction of the Central Indian Ocean: Tectonophysics, v. 155, p. 211–234, doi: 10.1016/0040-1951(88)90267-3.
Royer, J.Y., and Chang, T., 1991, Evidence for relative plate motions between the Indian and Australian plates during the last 20 m.y. from plate tectonic reconstructions: Implications for the deformation of the Indo-Australian plate: Journal of Geophysical Research, v. 96, p. 11,779–11,802.
Royer, J.Y., Gordon, R.G., and Horner-Johnson, B.C., 2006, Motion of Nubia relative to Antarctica since 11 Ma: Implications for Nubia-Somalia, Pacifi c–North America, and India-Eurasia motion: Geology, v. 34, p. 501–504, doi: 10.1130/G22463.1.
Sauter, D., Patriat, P., Rommevaux-Jestin, C., Cannat, M., Briais, A., Bergh, H., Boulanger, D., Deplus, C., Grindlay, N., Isezaki, N., Mendel, V., Mével, C., Thibaud, R., Tisseau, C., Vanney, J.-R., Whitechurch, H., and Yama-moto, M., 2001, The Southwest Indian Ridge between 49°15′E and 57°E: Focused accretion and magma redistribution: Earth and Planetary Science Letters, v. 192, p. 303–317.
Schlich, R., and Patriat, P., 1971, Mise en évidence d’anomalies magnétiques axiales sur la branche ouest de la dorsale médio-indienne: Comptes Rendus de l’Académie des Sciences Série IIa: Sciences de la Terre et des Planètes, v. 272, p. 700–703.
Sclater, J.G., Munschy, M., Fisher, R.L., Weatherall, P.A., Cande, S.C., Patriat, P., Bergh, H., and Schlich, R., 1997, Geophysical synthesis of the Indian/Southern Oceans, part 1: The Southwest Indian Ocean, Scripps Institution of Oceanog-raphy Reference Series 97-06: San Diego, La Jolla, Scripps Institution of Oceanography, University of California, p. 45.
Shaw, P.R., and Cande, S.C., 1990, High-resolution inversion for South Atlantic plate kinematics using joint altimeter and magnetic data: Journal of Geophysical Research, v. 95, p. 2625–2644.
Silver, P.G., Russo, R.M., and Lithgow-Bertelloni, C., 1998, Coupling of South American and African plate motion and plate deformation: Science, v. 279, p. 60–63.
Smith, W.H., and Sandwell, D.T., 1997, Global seafl oor topography from satellite altim-etry and ship depth soundings: Science, v. 277, p. 1956–1962, doi: 10.1126/science.277.5334.1956.
Manuscript received 11 July 2007Revised manuscript received 5 November 2007Manuscript accepted 6 November 2007
