B.D. Johnson

Spreading history of the eastern Indian Ocean and Greater India's northward flight from Antarctica and Australia

Recent information from magnetic surveys and from deep-sea drilling allows Sclater and Fishers Late Cretaceous and Cenozoic reconstructions of the eastern Indian Ocean to be extended back almost to the beginning of the Cretaceous Period. After a short phase of spreading off northwestern Australia in Middle and Late Jurassic time, Greater India and Antarctica-Australia dispersed near the beginning of Cretaceous time (130 m.y. B.P.) with the opening between them of a landlocked sea; between 110 to 120 and 105 m.y. B.P., Greater India cleared Antarctica-Australia, and the sea floor generated between them became continuous with the rest of the Indian Ocean. In Santonian time (80 m.y. B.P.), a new pattern of rapid spreading (as much as 17.5 cm/yr) began and caused an oceanic part of the Indian plate to be transferred to the Antarctic-Australian plate. Rapid spreading continued nearly to the end of Paleocene time (53 m.y. B.P.). With the inception at this time of spreading between Antarctica and Australia, three plates (Indian, Antarctic, and Australian) spread at a slow rate until the end of early Oligocene time (32 m.y. B.P.). At 32 m.y. B.P., the separate Indian and Australian plates became united, as they are today, while Antarctica remained a separate plate. Together with paleomagnetic and other determinations that show that Southeast Asia lay at or north of the Equator, the trail of sea floor generated during Greater Indias northward flight implies that Southeast Asia rotated westward across this trail to its present position no earlier than middle to late Miocene time (10 m.y. B.P.).

Greater India's place in Gondwanaland and in Asia

Abstract Evidence from the eastern Indian Ocean implies a Greater India in Gondwanaland. Translated without distortion to Asia, the northeast part of Greater India is found to correspond in shape with the Tibetan Plateau. With geophysical evidence, this suggests that the northeast part of Greater India underlies, presumably by underthrusting, the Tibetan Plateau.

A revised fit of East and West Gondwanaland

Abstract Recently available evidence provides the basis for a revised fit between East and West Gondwanaland before break-up in the Late Jurassic: deep-sea drilling shows that the entire Falkland Plateau is probably underlain by continental crust, marine geophysical studies off southeast Africa indicate large areas of thinned continental or transitional crust; palaeomagnetic studies show that the western side of Madagascar lay alongside equatorial East Africa; and from the pattern of sea-floor spreading between Madagascar and India we deduce that the southern half of the western margin of India cannot have lain, as customarily shown, alongside the eastern margin of Madagascar, but must have lain farther south. This information about Madagascar provides the crucial link between East (Antarctica, Australia, India) and West (South America, Africa) Gondwanaland. The rest of East and West Gondwanaland is brought into contact so that the Falkland Plateau opposes the margin of Antarctica between 10° and 15°E and the southern part of South America fits without deformation into the Weddell Sea re-entrant of Antarctica. In terms of the continuity of geological features and the cluster of pre-break-up palaeomagnetic poles, the revised fit is at least as favourable as that of Smith and Hallam (1970). In its close match of the continental outlines and its harmony with the pattern of subsequent sea-floor spreading, the revised fit is superior to previous reconstructions.

An equivalent source model of the satellite-altitude magnetic anomaly field over Australia

The long-wavelength magnetic anomaly field at 400–700 km elevation over Australia measured by the Pogo series satellites is modeled by an equivalent source technique. Magnetic moments of dipoles in a regular latitude-longitude grid at the Earths surface are found by a least squares procedure such that the resultant anomaly field makes a best fit to that observed. The fit improves steadily as dipole spacing is decreased, but with diminished returns, so that below a spacing of about 2.7° improvement is insignificant. The fit at this spacing is about 0.5 nT, as compared with an observed anomaly field (at 450 km) of around ±8nT. The distribution of magnetic moments is converted to a model of apparent magnetization contrast in a layer of arbitrary constant thickness. The distributions reflect regional variations in gross magnetization within, and thickness of, the magnetic crust. For source spacing greater than 2.5°, the magnetization values contour easily and show good correlations with tectonic provinces. For source spacing finer than about 2.5°, spurious oscillatory effects begin to enter the solutions, and the magnetization distributions are not physically meaningful. The apparent magnetization contrast models can be transformed to variable-thickness models if independent estimates of local magnetic crustal thickness are available.

Early spreading history of the Indian Ocean between India and Australia

Abstract A revised model of seafloor spreading between India and Australia from the inception of spreading 125 m.y. to the change to a new system at 90 m.y. stems from the wider recognition of the M-series of magnetic anomalies off the southwestern margin of Australia, from a revised pole of opening between Australia and Antarctica, and by the extension in the central Wharton Basin of the Late Cretaceous set of magnetic anomalies back to 34. The phase of spreading represented by the later anomalies has been extended back to 90 m.y. in order to give a resolved pole that describes the rotation of India from Australia consistent with the M-series anomalies, DSDP site ages, and fracture zone trends. An abandoned spreading ridge in the Cuvier Abyssal Plain indicates a ridge jump within the first ten million years of spreading. Elsewhere, two kinds of ridge jump (one to the continental margin of Australia or India, the other by propagation of the spreading ridge into adjacent compartments thereby causing them to fuse), are postulated to account for other observations.

Seafloor constraints on the reconstruction of Gondwanaland

Abstract M-series magnetic anomalies in the Mozambique Basin, off Dronning Maud Laud, and off the western margin of Australia, taken with other oceanic evidence, are used to test a new fit of Gondwanaland in which Madagascar adjoins the northern part of the western margin of India, the Falkland Plateau opposes part of Antarctica, and the southern part of South America fits without deformation into the Weddell Sea re-entrant of Antarctica. The first stage of spreading, from 150 m.y. to 125 m.y., is between East and West Gondwanaland, during which Madagascar, as part of East Gondwanaland, rotated half-way to its present position. The second stage, from 125 m.y. to 105 m.y., began with the separation of South America from Africa, and Antarctica-Australia from India-Madagascar, and ended with Madagascar reaching its final position with respect to Africa. In the third stage, from 105 m.y. to just before 90 m.y., spreading continued except between India-Madagascar and Africa which were part of a single plate. A major reorganisation of spreading systems at the end of the third stage saw India separate from Madagascar. In our reconstructions, a slight mismatch between the modelled transform faults and the observed fracture zone on the eastern side of the Dronning Maud set suggests that only small adjustments are required to the new reconstruction of Gondwanaland.

Prominent magnetic anomaly along the continent-ocean boundary between the northwestern margin of Australia (Exmouth and Scott Plateaus) and the Argo Abyssal Plain

Abstract A prominent positive magnetic anomaly, with amplitude as great as 1400 nT, lies along the lower slope between the northern Exmouth Plateau, characterized seismically by faulted layered reflectors, and the Argo Abyssal Plain, characterized by hyperbolic reflectors. Because the slope of the northern Exmouth Plateau is floored by continental crust, as shown by Triassic and Early Jurassic volcanics and sediments deposited before the Middle Jurassic inception of spreading in the Argo Abyssal Plain, the anomaly lies along the continent-ocean boundary (COB). To the northeast of the northern Exmouth Plateau, the anomaly, here with an amplitude no greater than 700 nT, is traced along the lower slope of the Rowley Terrace and Scott Plateau, and across the expanded continental rise north of 13°S, and indicates that the Rowley Terrace and the bulk of the Scott Plateau are continental, and that only the lower part of the rise, north of 13°S, is oceanic. In this area, an oceanic ridge rises above the adjacent breakup unconformity in the Scott Plateau to impound a deep sedimentary basin, like that impounded by the Outer Voring Plateau off Norway. The COB anomaly is modelled as a shallow two-dimensional magnetic body with rectangular cross-section and 40–80 km wide that extends up to 20 km seaward of the COB, and is interpreted as a complex of rift-related dykes in the continental crust and adjacent oceanic crust.

Contraints on the cenozoic position of Sundaland

Abstract The Cenozoic ocean-floor path of the continental fragment, Greater India, is overlapped by the present western part of Malaysia and Sumatra which are now part of a coherent continental block, Sundaland. This part of Southeast Asia must consequently have lain further east during the Cenozoic. The past positions of Greater India, combined with published paleomagnetic data indicating that Sundaland has lain near the Equator since the Permian and rotated anticlockwise since the mid-Cretaceous, are used to reconstruct constraints on the relative motions of Sundaland and the Indian—Australian plate in 10 m.y. intervals. We show that the northern part of Sundaland has rotated a minimum of 550 km westward with respect to India in the last 50 m.y. (since Early Eocene) with most of the rotation occurring in the latter half of the Cenozoic. Accepting geological evidence for an even larger Cenozoic sinistral shear between Sundaland and Australia, we construct a model consistent with ocean-floor and paleomagnetic constraints in which Australia and Sundaland made their closest approach between 10 and 20 m.y. ago (Miocene). The S-shape of the Banda Arcs may have formed since mid-Miocene from an initially linear, E-W trending pair of arcs by the interaction of the large sinistral shear between Sundaland and Australia and the collision of the leading edge of Australia with these paired arcs commencing approximately 15 m.y. ago.

An equivalent layer magnetization model for Australia based on Magsat data

Abstract An equivalent layer magnetization model for Australia and adjacent oceanic areas is presented. The model is obtained by linear inversion of Magsat anomaly data measured in the altitude range 325–550 km. The anomaly data set has been isolated from the raw data set by use of models of the core field and very long wavelength external fields, and is internally consistent. Certain major structural features of the Australian continent are geographically associated with magnetization anomalies. A first-order difference is seen between the Tasman Zone and the Precambrian cratonic areas: magnetization anomalies are much more subdued in the former, possibly reflecting a shallowing of the Curie isotherm within the crust. A profile of the vertical integral of magnetization is presented for a crustal section extending from the Gawler Block to the southeast coast. It is shown that the magnetization variations are probably due partly, but not wholly, to depth to Curie isotherm variations; gross magnetization variations among at least three distinct crustal units must be involved.

Palaeomagnetic comparison of a new fit of East and West Gondwanaland with the Smith and Hallam fit

Abstract Palaeomagnetic data for five time intervals in the Phanerozoic are compared for a new fit of East (Antarctica, Australia, India) and West (Africa, Madagascar, South America) Gondwanaland and the hitherto durable fit of Smith and Hallam. The dispersion of the Palaeomagnetic poles is marginally less on the Smith and Hallam fit for four time intervals and marginally greater for the remaining one (Permo-Triassic). The Permo-Triassic poles of East Gondwanaland are evenly distributed between India and Australia and the decreased dispersion of the poles for this period on the new fit of East and West Gondwanaland is paralleled by the decreased dispersion of the poles for both India and Australia. Within the limitations of the analysis imposed by the data, the palaeomagnetic comparison shows that there is little to choose between the two fits.