U31B-01 08:05h
Jan Hospers' development of the geocentric axial dipole hypothesis and defense of field reversals
When Jan Hospers arrived at Cambridge in 1949 he already had planned to use paleomagnetism as a stratigraphic tool to investigate Icelandic lava flows during summer 1950. He did not, however, plan to work on the problem of reversals or use paleomagnetism to test continental drift. At the time he did not know about reversals, and he thought continental drift was a relic of past geological thought. Although Ben Browne remained Hospers' official supervisor, Keith Runcorn, who arrived at Cambridge in January 1950, became his mentor. He encouraged Hospers and provided key contacts. Hospers returned with 22 samples from his first collecting trip. He observed that approximately half his samples were reversely magnetized, that they appeared to be clustered around the axis of rotation, and that he needed a statistical procedure for reducing the observed scatter. Runcorn asked R. A. Fisher to help with the statistics. Fisher developed the statistical analysis, and Hospers (1951) argued for field reversals and the geocentric axial dipole hypothesis (GAD). Runcorn encouraged Hospers to obtain samples from Northern Ireland and to return to Iceland. Hospers' new work led him to make two firm and important contributions to paleomagnetism. Hospers (1953-1954) offered the first extensive empirical support for serial reversals and GAD. His support for field reversals included a thorough discussion of L. N\'{e}el's self-reversal hypotheses. The empirical support he marshaled in favor of GAD, which extended back through the Miocene, provided a firm base on which K. Creer, E. Irving, Runcorn, and other paleomagnetists could present arguments favoring mobilism. He also constructed a rudimentary polarity time scale.
U31B-02 INVITED 08:20h
The First Paleomagnetic Polar Wander Path
At the end of 1952, having completed my work on the Cambridge astatic magnetometer, I was motivated to embark on a "Preliminary Paleomagnetic Survey of Rocks from the British Isles" by exciting results obtained by two fellow research students:- Jan Hospers' proposal of the axial dipole hypothesis and Ted Irving's discovery of strongly oblique Pre-Cambrian paleomagnetic directions, substantiated by the oblique Triassic directions obtained by John Clegg's group. Geologists advised me to collect from palaeontologically well dated rock formations. But these turned out to be very weakly magnetized and thereafter I concentrated on purple and dark red coloured rock formations. By the end of July 1954 I had compiled a table of nine Period-mean paleomagnetic directions spanning the last 600 Myr. I passed a copy to Keith Runcorn to include in a talk (co-authors Creer and Irving) scheduled for the August 1954 Rome Assembly of IAGA. Meanwhile, background reading took me to Gutenberg's "Internal Constitution of the Earth (1951)" where I came across (Fig 12) paths of the north pole proposed by Kreichgauer (1902) and by Koppen and Wegener (1924). This prompted me to calculate a paleomagnetic polar wander path. I presented this at the Annual Meeting of the British Association for the Advancement of Science held at Oxford on September 8th. An artist's representation of it was published in Time Magazine of September 24th 1954 where the accompanying text records that I stressed that similar work on other continents would be necessary to distinguish whether the continents had drifted independently or whether the sole mechanism had been polar wander. On my return to Cambridge, Maurice Hill informally suggested that I should indicate precision, so for each pole I calculated semi-major and minor axes of the ellipse of confidence corresponding to the radius of confidence of each mean direction. These were shown in my Ph.D. thesis, where also I calculated a paleomagnetic pole for John Graham's North American Silurian Rose Hill Formation and argued tentatively that it's location, somewhat to the east of the British Silurian pole, is qualitatively consistent withWegnerian drift.
U31B-03 08:35h
Paleogeography, Plates and the Role of Latitude
In the 1920s, Wegener, Koppen and Argand created the first mobilistic paleogeography, which for many reasons did not prosper. Mobilism was revived in the 50s and 60s through measurements of long-term motion of crustal blocks relative to earth's spin axis (apparent polar wander APW and latitude change, 1954) and relative to each other (plate tectonics 1967/8). In the past decade, seismic and geodetic arrays and campaigns have observed compatible short-term motions. Creer's discovery of apparent polar wander (APW) which we celebrate today, initiated the development of a new mobilistic paleogeography based on physical measurement of latitude change and rotation of crustal blocks; attempts to make unambiguous reconstructions encountered difficulties not the least of which was the adamantine fixism of earth science institutions. However, the advent of plate tectonics a decade later (1967/8) provided the natural spherical framework for integrating not only measurements of tectonic motion on which it was initially based, but also APW and measurements of latitude change; this was done regionally in 1971 and globally in 1973. At this point the new paleogeography can be said to have matured; over the last 30 years it has become the basis for reconstructing the evolution of earth's surface features vital for understanding evolution of life, long-term climate change, measuring large-scale deformation in mobile belts, etc. I shall illustrate the key steps in this progression from Creer's first path to the present day.
U31B-04 INVITED 08:50h
A Journey with Runcorn
S. K Runcorn was a central figure in the early history of Paleomagnetism and he contributed a great deal to the way the subject developed. I joined him as a research student in the autumn of 1955.This was after the first polar wandering curves were published by Creer,Irving and Runcorn in 1954.the important development of this idea in the next three years was carried out by Runcorn's group;which in the winter of 1955-56 moved from Cambridge to Newcastle -on- tyne. The new data from North America lead to a reevaluation of the meaning of polar wandering when intercontinental comparisons were make. It became clear as the data accumulated from North America mostly by Runcorn and from Australia by Irving that in order to explain the data that movement between the continents must have taken place. I will describe what led Runcorn to change his mind in the spring of 1956.and what it was like to work with Keith in these interesting years.
U31B-05 09:05h
APWPs: Critical Building Steps and Potential for Future Geodynamical Studies.
Among other items, the Apparent Polar Wander Paths (APWP) of individual tectonic blocks contain information on geodynamical events from both shallower lithospheric sources (e.g. continental breakups and collisions) and deeper mantle sources affected by convection. Recent advances in the Earth Sciences, such as provided by global tomography or climate modeling, have emphasized the essential need for accurate reconstructions of the Earth's surface (blocks, plates and their boundaries and topographies), in order for instance to compare the initial positions of downgoing slabs or emerging hot spots with deeper mantle features, or to understand why and how time-varying climates and biomes may be related. A half century after its initial pionneering contributions to the formulation of continental drift and plate tectonics, paleomagnetism remains an invaluable tool which will allow us to solve a very large range of problems concerning both internal and external geodynamics (solid, fluid and bio-spheres). The accurate determination of the APWPs of crustal and lithospheric blocks remains one of the main goals that must be pursued by the paleomagnetic community. Based on two examples, one extending from Permian to Present, the other in the Late Proterozoic to Early Paleozoic, I will discuss various aspects of the construction of APWPs and reconstruction of past plate positions. Critical steps involve: 1) how are original paleomagnetic data selected? 2) how are the kinematic models used to transfer data from one plate to the other determined ? 3) how can poles coming from deformed (e.g. rotated) regions be used ? and 4) how good is the geocentric axial dipole (GAD) assumption, which is fundamental for reliable plate reconstructions ? Particular emphasis will be given to this last topic, since the GAD hypothesis has recently been challenged, with suggestions that significant long-term octupolar contributions might have existed from the Precambrian throughout to the early Tertiary. These might account for the low inclinations observed in central Asia in the Cretaceous and early Tertiary, or for the misfits of Pangea (Pangea A or B?). GAD hypothesis appears to be essentially correct and APWPs still have a bright future.
U31B-06 09:20h
The Case for Pangea B: Paleomagnetic Contributions from Adria
The pre-drift Wegenerian model of Pangea is almost universally accepted, but debate exists on its pre-Jurassic configuration since Ted Irving introduced Pangea B. We review Permian and recently acquired Jurassic-Cretaceous paleomagnetic data from para-autochthonous regions of Adria such as the Southern Alps, which we show to be broadly consistent with "African" APWPs. Paleomagnetic data from para-autochthonous Adria can therefore be used to bolster the Gondwana APWP in the poorly known Late Permian-Triassic time interval. Adria paleopoles are integrated with the Gondwana and Laurasia APWPs and used to generate a tectonic model for the evolution of Pangea. The Early Permian paleopole of Adria from radiometrically dated igneous rocks, in conjunction with the coeval Gondwana and Laurasia paleopoles again from igneous rocks, support Pangea B. The use of paleomagnetic data strictly from igneous rocks excludes artifacts from sedimentary inclination error as a contributing explanation for Pangea B. The ultimate option to reject Pangea B is to introduce a significant zonal octupole component in the Late Paleozoic time-averaged geomagnetic field. Our dataset consisting entirely of paleomagnetic directions with low inclinations from sampling sites confined to one hemisphere show that the effects of a zonal octupole field contribution cannot explain away the paleomagnetic evidence for Pangea B. We therefore regard the paleomagnetic evidence for an Early Permian Pangea B as robust. Because the Late Permian/Early Triassic and the Middle/early Late Triassic paleopoles from Adria and Laurussia support Pangea A, the phase of transcurrent motion between Laurasia and Gondwana that caused the Pangea B to A transition occurred essentially in the Permian. It took place after the cooling of the Variscan mega-suture and lasted ~20 m.y., with an average relative plate velocity of approximately 15 cm/yr. Finally, we review geological and paleomagnetic evidence in support of an intra-Pangea dextral megashear system. In particular, we present paleomagnetic data from Corsica and Sardinia that, during the Permian, were presumably caught into the transcurrent plate boundaries between Gondwana and Laurasia and dissected away in variably rotated crustal blocks.
U31B-07 INVITED 09:35h
Origin and History of Magnolias
The classic disjunct distribution of living magnolias between southeast Asia and the Americas has been a puzzle for more than a century and a half. We propose a scheme for the origin and history of magnolias to explain this distribution by integrating paleogeographic, paleoclimatic, paleobotanical, and phytogeographic data. Our scheme is based on paleomagnetically determined latitudes. We assume that the moist warm temperate climate favoured by most extant magnolias has always been their preference. Molecular analyses (largely chloroplast DNA) reveal that several North American species are distinct and basal forms suggesting that magnolias likely originated in North America, a conclusion supported by the fossil record. We identify four evolutionary stages: (1) Ancestral magnolias originate in the Late Cretaceous of North America in high mid-latitudes ($45\deg$-$60\deg$N) at low altitudes under greenhouse climates. (2) During the exceptionally warm climate (super-greenhouse) of the Eocene, magnolias spread eastwards via the Iceland-Faroes isthmuses, to Europe and then across west and central Asia possibly to east Asia, still at low altitudes and high mid-latitudes. (3) With global cooling from the mid-Cenozoic, magnolias shift their core range to lower mid-latitudes ($30\deg$-$45\deg$N), become extinct in Europe and southern Siberia, breaking the once continuous distribution into two. (4) In the late Cenozoic, as ice-house conditions develop, magnolias diversify rapidly and expand southward into moist warm temperate uplands in newly uplifted mountain ranges of South and Central America, southeast Asia and the High Archipelago between southeast Asia and Australia. The late Cenozoic evolution of magnolias is characterized by the impoverishment of northerly species and diversification of southerly species. Thus, the centre of origin is not the center of diversity today. Disjunction at the generic level likely occurred as part of the mid-Tertiary southward displacement of the northern hemisphere "boreotropical" flora, assisted by the development of north-south water barriers to floral interchange, especially the Turgai Straight across eastern Siberia in the Late Eocene.
U31B-08 09:50h
A Short Review of True Polar Wander
Polar wander on Earth has been suggested since the 19th century. With the discovery that continental drift accounted for much of the apparent polar wander of continents, one could ask whether there was a remaining ("true") fraction in polar wander not accounted for by plate tectonics, which would be a characteristic of "Earth as a whole". TPW results from conservation of angular momentum in a rotating, deformable body. Seen from the surface of the Earth, TPW appears as the wave-like propagation of the Earth's bulge, whose rate of motion is controlled by mantle viscosity. The bulge adjusts in about 104 years, the characteristic time for glacio-isostatic rebound. Paleomagnetic poles (used to derive APWPs) and oceanic data (used to derive plate kinematic models) can be blended to produce a "synthetic " APWP for all plates. Motion of hotspots with respect to plates can then be integrated to derive an estimate of TPW: this displays in succession a standstill at 160-130 Ma, a quasi-circular track from 130 to 70 Ma (rate 30 km/m.y.), a standstill at 50-10 Ma and faster motion up to the present (rate 100 km/m.y.). Suggested episodes of superfast TPW seem to be artefacts. Field geometry is unlikely to severely alter TPW estimates. A legitimate concern is that the analysis is not truly global (it fails to encompass the Pacific plate). And there are ongoing debates on the fixity of hotspots with respect to each other. We find little evidence for significant inter-hotspot motion (larger than 5 km/m.y.) either within the Pacific or Indo-Atlantic hemispheres. Other authors do find some motion (e.g. between Hawaii and Louisville). We suggest that (primary) hotspots form two slowly deforming subsets in the two geodynamically distinct hemispheres. The two subsets would have been in slow motion for the last 45 m.y., but in faster motion prior to that. Other authors (Gordon) conclude that there is no significant motion in the past 125 m.y. It has been suggested that a quasi-discontinuous 90° rotation is possible in which the rotation axis will align with a new axis of maximum non-hydrostatic moment of inertia (inertial interchange event IITPW). Oscillatory polar wander is thought by some to be documented in the Precambrian, which is attributed to TPW about a long-lived inertial axis inherited from the super-continent of Rodinia (Evans). Modeling of TPW has made significant progress. Seismic tomography is used to infer 3D maps of density heterogeneities that drive flow in the viscous mantle. The maximum speed of polar wander driven by mantle convection is about 100km/m.y. A significant viscosity increase in the lower mantle is required to bring TPW rates closer to observed values. Recent modeling (Steinberger) involves limited, predictable inter-hotpot motion. One can still derive a TPW curve in the "mean mantle" reference frame that takes hotspot motions into account. This captures most of the features of the observed TPW. Possible links between 1) TPW episodes, or major changes between TPW episodes, or IITPW events, 2) reorganizations in the geometry of subduction zones (e.g. avalanches), or plume and superplume generation, and 3) major biotic changes (e.g. mass extinctions) will likely keep analyses of polar wander very much alive in the coming decade, despite the somewhat bothering feeling that true polar wander still remains an elusive geophysical phenomenon.