Fig. 7

Salinity, shell-derived δ ¹⁸O_sw, and Ba/Ca ratios of core-top G. ruber tests. a Shell-derived δ ¹⁸O_sw and calcification salinity plot of core-top samples in this study and the literature. Shell-derived δ ¹⁸O_sw values are calculated using the Mg/Ca-derived temperature and core-top δ ¹⁸O_c of G. ruber. The data from KY, A7, 403, and 404 (gray squares) are from Kubota et al. (2010), Sun et al. (2005), Lin et al. (2006), and Chen et al. (2010), respectively. Error bars show the propagation error of ±0.28 ‰ (1σ). Seawater data (green circles) and their regression line (green broken line, Fig. 4) are also shown with the core-top data. Core-top data fall close to the local δ ¹⁸O_sw–salinity regression line. b Ba/Ca _{G. ruber} ratios and calcification salinity plot of all core-top samples in this study. The replicated analyses of Ba/Ca _{G. ruber} ratios at each core site show high variability, resulting in the range from 0.66 to 2.82 μmol mol⁻¹. The calculated shell Ba/Ca ratio in equilibrium with the Kuroshio water is 0.5–0.72 μmol mol⁻¹. Lower values at HR2 and HR6 are within this range. Higher shell Ba/Ca ratios than this Kuroshio water range might be attributable to barite contamination or calcification in low saline, Ba-enriched CDW