Shell layer variation in trace element concentration for the freshwater bivalve Elliptio complanata

Abstract

While the elemental chemistry of marine mollusk shells contains valuable environmental information, extending these interpretations to freshwater bivalve shells may be problematic, given the wide range of aqueous chemistries that exist in freshwater environments. To better understand the significance of these records, 20 bivalves were collected from four freshwater streams to determine the geochemical relationships that exist between the bivalve shell and their sources streams. The concentrations of manganese (Mn), copper (Cu), strontium (Sr), barium (Ba), and calcium (Ca) were analyzed by laser ablation inductively coupled plasma mass spectrometry across the inner (INL) and outer nacreous layer (ONL) of each shell for comparison to stream data collected over the period of shell growth. Within an individual shell, the content of Mn, Sr, and Ba were significantly higher in the INL than the ONL, while Cu concentrations were similar. Strontium and Ba concentrations co-varied between the two layers in 1:1 relationships, while Mn displayed a preferential enrichment in the INL that increased from 2:1 to 5:1 as the Mn content of the shell decreased. Each elemental profile can be correlated between shell layers, except for the trace element Cu, which appeared to be more closely related to the organic content of the shell. These results suggest that the major element geochemistry of the shell layers differentially reflect the aqueous chemistry of the water in which a bivalve lived, but that these records are most likely overprinted by physiological processes specific to the mantle tissue from which a particular shell layer is secreted. Distribution coefficients (shell _{M /Ca}:water _{M /Ca}, where M = trace element of interest) were calculated using the median molar elemental concentration for each shell (by layer), and the appropriate water concentration. Bivalves from a contaminated site were excluded from this analysis because their shells were anomalous in size and color. For the remaining 15 shells, distribution coefficients (INL and ONL) were: Mn (0.50 and 0.21), Sr (0.26 and 0.17), and Ba (0.05: INL only). Inner nacreous layer values were comparable to the upper end of published estimates for freshwater bivalves and fish otoliths, while the ONL values were comparable to the lower range of values. Inclusion of shells from the contaminated site resulted in the calculation of distribution coefficients that fell outside the range of published estimates. These results suggest that exposure to pollutants may have a bearing on the biological factors that control the elemental concentration of bivalve shells in freshwater environments. While researchers generally avoid sampling shell material from the INL because of the potential for shell dissolution, the similarity in elemental profile trends observed here suggests that both the INL and ONL record accurate geochemical information, but in distinct forms. An obvious advantage of INL analysis is the relative increase in elemental concentration, but this is afforded at the expense of spatial resolution. Considered collectively, these results suggest that valuable information can be gleaned from the elemental concentration of freshwater bivalve shells when care is taken in the choice of material from which these records are extracted.