Abstract

X-ray fluorescence (XRF) core scanners have advantages in nondestructive measurements of the elemental compositions of sediment cores with high stratigraphic resolution. To estimate elemental flux using sediment compositions measured by the XRF core scanner, total mass flux must be precisely calculated using the dry bulk density (DBD) and linear sedimentation rate records. In this study, we investigated a method for quantitatively estimating the DBD of sediment core samples with high resolution using the XRF spectra obtained by an XRF core scanner. We chose the Quaternary hemipelagic sediments collected from three different water depths in the Japan Sea during the Integrated Ocean Drilling Program Exp. 346. They primarily constitute clay and silty clay with varying amounts of biogenic silica and biogenic carbonate. Conventionally, the DBDs of deep-sea sediments were determined using onboard moisture and density (MAD) measurements. However, because MAD measurements take time, the stratigraphic resolution of the data is not high enough to reflect the composition changes of the sediments detected by the XRF core scanner analysis. In this study, a multiple regression analysis was conducted between MAD-based DBD as an objective variable and various combinations of XRF core scanner outputs, such as scattering X-ray intensities and peak intensities of the elements that reflect changes in grain composition, as explanatory variables. The results of the multiple regression analysis indicated that the introduction of ln(Cl/Ti) in addition to X-ray scattering improves the accuracy of DBD estimation, whereas the introduction of ln(Si/Ti), ln(Ca/Ti), ln(Fe/Ti), and ln(Br/Ti) does not significantly improve the accuracy. The inclusion of Cl for DBD estimation is important for marine sediments which contain Cl in the pore water. The estimation of the DBD of sediment cores solely using the results of the XRF core scanner enabled the DBD estimations of the stratigraphic positions exactly the same as where element composition was determined with high-spatial resolution by XRF core scanner analysis. This will enable us to estimate content of each element per unit volume of the sediment with high-spatial resolution and precision, thereby enhancing our ability to reconstruct past geochemical cycles with higher time resolution and accuracy.