Paleomagnetism of the Jurassic Transantarctic Mountains revisited — Evidence for large dispersion of apparent polar wander within less than 3 Myr

The Transantarctic mountains predominantly consist of Jurassic continental flood basalts (Kirkpatrick) and sills (Ferrar) emplaced during the earliest phase of the break-up of Pangea. Published ages, based on a variety of geochronological methods all agree rather well and cluster around 180 Ma suggesting emplacement during a rather short time interval of not more than 3 Myr. Paleomagnetic studies, mostly carried out between the 60s and 90s of the last century, however, yield two well defined but significantly different groups of paleomagnetic pole positions plotting either in intermediate latitudes (A) or at latitudes exceeding 60°S (B). Pole positions belonging to group A are generally interpreted to be of primary origin whereas the significance of group B poles remains unclear. Here we report new data from the Kirkpatrick basalts of Northern Victoria Land, Antarctica, where 157 oriented paleomagnetic samples were taken, covering about 800 m of stratigraphy and 23 volcanic flows in Gair Mesa (73.4666°S; 162.8666°E). After removal of a steep magnetic overprint with rather low coercivities straight linear segments of exclusively normal polarity trending toward the origin of projection are identified in 151 samples from 22 sites. Maximum unblocking temperatures do not exceed 580 °C and maximum coercivities not 60 mT. The resulting mean virtual geomagnetic pole (VGP) for the Gair Mesa plots at 66.4°S; 227.7°E (α₉₅ of 6.2°, k = 25.7) comparable to group B poles. Reflecting light microscopy reveals that the magnetic inventory of all samples analyzed is dominated by magnetite showing shrinking cracks and broad ilmenite lamellae, the latter being diagnostic for high temperature oxidation. Analysis of the distribution of the 22 pole positions obtained here shows that secular variation has successfully been averaged (S_b = 15.9°) suggesting that the rocks studied here have recorded the time averaged geomagnetic field during early mid Jurassic times. Based on these new results we postulate that both clusters A and B of pole positions reflect primary magnetizations and that the Kirkpatrick basalts of Gair Mesa have been emplaced in a rather short normal polarity time interval of 3 Myr. If our reasoning is correct, the apparent polar wander path for the latest early Jurassic might be more complex than previously thought. Whether this complexity is evidence for massif True Polar Wander remains speculative.