Abstract
The trace element composition of the calcium carbonate shells of foraminifera preserved in marine sediments is a key tool for reconstructing past ocean environments. The size of the shells is known to influence some elements but as studies usually focus on restricted size ranges, the effect of size on the trace element composition of shells from different species of foraminifera is largely unknown. To provide better constraints on trace element incorporation, well-preserved, size separated and mono-specific samples of planktonic foraminifera from a surface sediment sample have been analysed. Clear differences in patterns of trace element incorporation are observed between spinose and non-spinose species that cannot simply result from differences in depth habitats. Spinose species have photosymbionts that moderate incorporation of pH-sensitive elements such as B and U, and possibly Mg, with larger individuals apparently hosting more photosymbionts when calcifying their shells. Furthermore, spinose species can be clearly distinguished from non-spinose species as they have higher B/Ca due to their shallower depth habitat and photosymbiont activity. Size effects on the incorporation of Li, Mg, Mn, Sr, Ba, and Na, vary between species. For example, Li/Ca decreases with size of Globigerinoides ruber and Mn/Ca increases with size of Globorotalia tumida . Na/Ca ratios are highest in smaller shells of Trilobatus sacculifer (both with and without a sac-like final chamber), G. ruber (pink), and Neogloboquadrina dutertrei , and decrease with increasing shell sizes below 425 µm. Overall, deeper dwelling non-spinose species have higher Ba/Ca by a factor of 10 compared to the spinose species, much more than the 20% increase in seawater Ba concentrations with depth in the upper water column. Assuming a relationship between precipitation rate and calcite Sr/Ca ratios, the similarity of Sr/Ca values between spinose and non-spinose species indicates calcification rates are comparable. As the bulk of calcification in planktonic foraminifera takes place over a month, this suggests larger or thicker shells do not calcify significantly faster than smaller or thinner shells. Instead, we suggest larger (thicker) shells calcified more often and probably grew during times of optimal environmental conditions or food availability. Strongly significant (p < 0.01) positive correlations between shell Li/Mg and Mn/Ca ratios with apparent calcification depth and seawater neutral density may be useful for studying the water depth influence on trace element incorporation.