Evaluating color reflectance spectral methods for estimating biogenic contents of South China Sea sediments

The biogenic contents of marine sediment cores are useful proxies for revealing past oceanic and climatic variability. Color reflectance spectral methods provide a rapid and non-destructive avenue for acquiring high-resolution biogenic content data from the split surfaces of marine sediment cores. This study presents new results on the applications of color reflectance methods for determining the biogenic contents of sediment cores retrieved from the South China Sea (SCS), a dynamic setting that receives biogenic components such as carbonate, total organic carbon (TOC), and opal from the sea, and terrestrial sediments transported by the Pearl River and Kao-Ping River from the north, and the Sunda Shelf from the south. Our study found that the color reflectance methods that are based on a regionally limited calibration provide more optimal estimates of the biogenic content of marine cores from the SCS than do basin-scale calibration methods. Though the regional calibrations appear to be successful, our analyses indicate the complexity of applying color reflectance methods in marine sediment core studies. We also performed a controlled experiment that was designed to quantitatively test how the carbonate content increased the influence of color reflectance changes, using outcropped, Pleistocene-aged sediment samples taken from the “Moon World (MW)” in southwestern Taiwan as control data. The results of this experiment suggest that the regression equations for estimating CaCO3 content based on core data are not solely driven by input variables responding directly to CaCO3 content changes, and that the composition of terrestrial sediment inputs into the SCS are complicated both spatially and temporally. Our studies also suggest that a better understanding of the compositional changes of terrestrial sediment inputs into the SCS will be required to minimize the uncertainties of biogenic content estimations by color reflectance methods.

► We present new color spectra equations for estimating sediment biogenic contents.
► We design a control experiment to test the sensitivity of the color equations.
► Our results indicate the complexity in using color spectra methods.