Understanding Geomagnetic Dipole Behaviour and Climatic Fluctuations over the Past 380,000 Years Using Cosmogenic 10Be and Paleointensity Records

Paleomagnetic field reconstructions provide us with the access to the long-term evolution of geodynamo mechanisms. The longest continuous paleomagnetic records originate from measurements of thermoremanent magnetization of ocean crust, although the sources of magnetization and the process governing changes in these records are still debated. Reconstructions based on magnetic signal in marine sediments are another way to obtain continuous paleomagnetic field variations spanning several million years in case of relative paleointensity (RPI) and cosmogenic ¹⁰Be records. These two independent proxies appear to align particularly well during periods of low dipole strength, such as geomagnetic excursions and reversals. This agreement is especially compelling because previous joint analyses of environmental proxies, RPI, and cosmogenic ¹⁰Be in ice and marine sedimentary cores indicate that both records can be influenced by non-geomagnetic modulations. Stacking a large number of RPI records from different depositional environments has been proposed to eliminate the site-related effects. However, this approach does not allow for a fair estimation of the relative climate versus geomagentic contributions to the overall paleomagnetic record. Moreover, when global climatic controls are embedded in individual records, stacking alone fails to remove these artefacts. 

In the present study, we use Principal Component Analysis (PCA) to evaluate the extent to which geomagnetic, climate and environmental modulations influence both RPI and ¹⁰Be signals. The effectiveness of this method in separating geomagnetic and climatic contributions was recently demonstrated on a ¹⁰Be signal from the North Atlantic site MD95-2016. Recognising that PCA performance strongly depends on the number of available records, we have extended this analysis to a globally distributed 30 RPI and 4 ¹⁰Be records spanning the past 380 ka. This global dataset enables us to compare the efficiency of each recording mechanisms during 8 geomagnetic excursions and 11 marine isotopic stages. We then compare independently derived geomagentic components from both ¹⁰Be and RPI records against previously elaborated RPI and Be - based VADM curves. Meanwhile, other contributions are interpreted in terms of global and regional climatic variations.