Palaeomagnetic study of Oligocene (24–30 Ma) lava flows from the Kerguelen Archipelago (southern Indian Ocean): directional analysis and magnetostratigraphy

Abstract

We report palaeomagnetic determinations in order to estimate the palaeosecular variation of the geomagnetic field for the upper Oligocene period from a southern site. We measured the anisotropy of magnetic susceptibility (AMS) of basalts from five sections to check the reliability of applying tectonic correction and concluded that no stratigraphic corrections were needed. A positive reversal test implies that secondary components were sufficiently removed from the characteristic remanent magnetization. Also, the sampling was sufficiently random to accurately reflect upper Oligocene palaeosecular variation. Calculations of the between-flow angular standard deviation ( S B ) from the geographic pole show values between 21.2 and 21.8°. This palaeosecular variation estimate agrees with Model G for palaeosecular variation [J. Geophys. Res. 96 (B3) (1991) 3923], which is axisymmetric and assumes an equatorial symmetry of the geomagnetic field. The virtual geomagnetic poles (VGPs) from the Mont des Ruches section seem to exhibit two polarity changes: reverse–normal (R–N) first, and in more detail a second normal–reverse (N–R) one. Finally, assuming that no polarity chrons are missed by any of the sections, we compare the magnetostratigraphy we defined to the global geomagnetic polarity scale of [Geophys. J. Int. 129 (1) (1997) 176] using previously obtained radiometric dates to correlate the sections [J. Petrol. 39 (4) (1998) 711; Earth Planet. Sci. Lett. 174 (2000) 313; J. Petrol. 43 (7) (2002) 1341]. The precise age constraints furnished by these correlations allow an estimation of eruption rates from 0.30 to at least 1.31 km/my, with wide variations from section to section and an increase of eruption rate for younger sections. However, due to the non-uniqueness of the correlations, the age limits of the sections and consequently the extrusion rate estimations can vary. Moreover, assuming the Mont des Ruches section that has probably recorded all the magnetic polarity intervals, the Mont de la Rabouillère and Mont de la Tourmente intermediate VGP directions may indicate unrecorded polarity chrons. The eruption rates calculated can thus be overestimated.