Berylium-10 production rate and relative paleointensity at low latitude across the last geomagnetic reversal

High resolution records of 10Be production and relative paleointensity have been obtained across the last reversal (Matuyama-Brunhes) from low latitude sediments of the equatorial Indian Ocean (Maldives area). Both magnetic and geochemical analyses were performed from the same single samples to avoid any artifical depth offset. The authigenic 10Be concentrations were normalized with respect to 9Be in order to correct for potential environmental effects, while the relative paleointensity was derived from the remanent magnetization intensity after accounting for changes in magnetic concentration within the sediment. Similarities between the magnetic concentration parameters and the 9Be variations open interesting perpectives to extract the influence of persistent disturbing noise in signals of relative dipole intensity. Both paleointensity records are characterized by large deviations with similar amplitude but relatively different shapes during the reversal. In contrast to most previous studies, the perfect synchronism of the 10Be/9Be peak with the NRM/ARM dip during the last reversal indicates that bioturbation and post-depositional processes did not affect the magnetization in these sediments. We obtain an age of 778ka for the onset of the transition derived from the reversal angle, while the intensity minimum occurs at 770ka +/- 2ka. These estimates remain younger than the radiometric age of 783ka inherent to the compilation of the volcanic records. We will discuss whether this difference is significant. Outside the reversal, the situation appears more complex since the 10Be production rate does not show a great deal of coherence with relative paleointensity. At this very low latitude site, the berylium peak was likely favored by the presence of multiple geomagnetic poles during the transitional period which, added to the weakness of the field, amplified the penetration of cosmic rays. In contrast, during polarity intervals with strong dipolar intensity, 10Be production is expected to be extremely limited, unless large inter-latitudinal mixing occurred in the atmosphere. The present results suggest that interlatitudinal atmospheric mixing was not very efficient, and therefore that the 10Be signal may not provide a pertinent record of geomagnetic intensity during periods of stable polarity when recorded at low latitudes.