Magnetostratigraphy and Environmental Magnetism of the RS15-LC100, RS15-GC101, RS15-GC102, and RS15-GC107 cores from the Ross Sea

Abstract

The Antarctic ice sheets are sensitive to changes in ocean temperature and are predicted to retreat in the coming centuries. Geological reconstructions of ancient ice margin retreat can offer insights into the potential future rates of change. However, a lack of well-dated sedimentary records near the ice margin in the Ross Sea has restricted the reconstruction of ice sheet retreat since the Last Glacial Maximum and the subsequent oceanographic changes.

Paleomagnetic studies are capable of providing the precise age control required for paleoclimate reconstruction from marine sedimentary successions and also enable correlation between geographically distant records, enabling comparisons to be made on both regional and global scales. The study of magnetic grains within sediment (environmental magnetism) can provide additional insights into changing oceanographic conditions and can be used to reconstruct changing terrestrial erosion rates and sediment delivery patterns.

This thesis discusses the paleomagnetic study of four sedimentary cores: RS15- LC100; RS15-GC101; RS15-GC102; and RS15-GC107. The cores were recovered from the Drygalski and Glomar Challenger basins on the Ross Sea continental shelf, and further offshore on the abyssal plain by the Korean Polar Research Institutes’ vessel the R/V Araon, in 2015. They are located in regions with greatly varying depositional and oceanographic environments, and range in age from Holocene to mid-late Pleistocene. Paleomagnetic and environmental magnetic studies are sparse in this region, with a lack of well constrained age models.

U-channel subsamples were measured at the Otago Paleomagnetic Research Facility (OPRF) in a 2G Enterprises superconducting magnetometer. Each sample underwent alternating field (AF) demagnetisation, and the data analysed on orthogonal component vector plots. Puffin Plot software was used to determine the Characteristic Remanent Magnetisation (ChRM) of the sediment and to produce a magnetostratigraphy. In addition, magnetic mineral studies (hysteresis, IRM, FORC and TDMS) were carried out in order to create a record of the environmental magnetism.

Paleomagnetic age models based on secular variation or relative paleointensity were not able to be constructed for RS15-LC100, RS15-GC101, and RS15-GC102, due to the lack of a coherent ChRM. The ChRM records for RS15-LC100 in particular were entirely unable to have a magnetostratigraphy produced due to significant variations in magnetisation direction down the entire core. These cores are likely be Holocene in age and will rely on radiocarbon dating to produce an age model. Rock magnetic data indicates that magnetite is the dominant remanence carrying mineral, with potential indicators of the diagenetic magnetic mineral greigite present in some cores which may contribute to the poor data quality.

A paleomagnetic age model was successfully constructed for RS15-GC107. Downcore variation in ChRM inclination aligned with the calculated Geocentric Axial Dipole inclination and the magnetostratigraphy that was correlated with the geomagnetic polarity timescale. Because biostratigraphic constraints were unavailable two age models were produced. Both models contain the Brunhes-Matuyama reversal boundary, with the preferred model, which results in the least varying sedimentation rate placing the boundary (C1n - C1r.1r, 0.774 Ma) at approximately 530 cm down core. This model also places the upper boundary of the Jaramillo chron (C1r.1r - C1r.1n, 0.990 Ma) at approximately 725 cm, and the lower boundary of the Jaramillo chron (C1r.1n - C1r.2r, 1.071 Ma) at approximately 775 cm. Calculated sedimentation rates varied between 0.68 cm/kyr and 0.72 cm/kyr. Small quantities of IRD are present throughout the core, as evidenced by discrete intervals with poor quality of demagnetisation data. TDMS and FORC analyses did not indicate the presence of greigite or other minerals that may carry a CRM.