Skeletal carbonate on the continental shelf: SEM evidence of early diagenetic alteration of the 1% of shells that persist for millennia in the mixed layer and thus enter the permanent record

An extensive program of dating aragonitic bivalve shells from the mixed-layer of the southern California shelf using AMS-calibrated amino-acid racemization has revealed strongly right-skewed shell-age frequency distributions: the median age of shells (2-7 mm) is quite young, generally <100 y, but the distribution includes a long tail of old shells ~2550 to ~11,900 y (Tomasovych et al. 2014 Geology). Modeling indicates that shells experience a high initial disintegration rate λ₁ (~decadal half-lives) but shift abruptly, within the first ~500 y, to a 100-fold lower disintegration rate λ₂ (~ millennial half-lives) at sequestration rate τ (burial and/or diagenetic stabilization). Although this drop permits the accrual of a long tail of very old shells, <1% of shells survive the first phase. In the fossil record, proxy geochemical and ecological data are extracted from such survivors, and so understanding the mechanisms of shell persistence in the mixed layer is critical to confident paleoenvironmental inference. We have hypothesized that permanent diagenetic stabilization within the mixed layer, which Pb-210 shows is 4-17 cm thick on this shelf, may be necessary to ensure that shells do not revert to λ₁ after temporary sequestration in pockets of favorable porewater. Our new SEM of shells of known, AMS-calibrated post-mortem age shows that macroscopic (10x) variation in shell color and luster is related to (1) loss of inter-crystallite organic matrix, probably by microbial maceration (chalky phase; within a few years post-mortem), (2) development of a surficial skim-coat of large, complexly interdigitating “jigsaw” crystallites, consistent with Ostwald ripening (shells re-acquire luster; patches develop within 10s y and coalesce over 100s to k y), and (3) development of a consistently fine-grained, apparently penetrative fabric, suggesting replacement of the original microstructure by a novel authigenic phase (gray stain; shells ≥~5 ka). Electron back-scattering diffraction is now being used to test for mineralogic changes as we better constrain the timing of phases 2 and 3: if the skim-coat suffices to rachet a shell into the 1% of survivors, the minute contribution to shell mass would have negligible blurring effect on proxy data even if this coat is calcitic.