Survival of the 1%: Consequences of a Two-Phase Dynamic of Aragonitic Shell Loss and Stabilization for the Temporal Resolution of Proxy Data

Abstract:

The strongly time-averaged nature of molluscan and other biogenic carbonate skeletons in seabeds is an under-appreciated uncertainty in paleoenvironmental inference using geochemical proxies. An extensive shell-dating program on the southern California shelf using AMS-calibrated amino-acid racemization corroborates the strongly right-skewed “L-shape” of shell-age frequency distributions (AFDs) found elsewhere. In California, the median age of aragonitic bivalve shells (2-7 mm, no size effect detected) is generally <100 y (5 of 8 assemblages have medians <50 y), and maximum measured ages are ~2550 to ~11,900 y. Modeling reveals that shells undergo an initial high disintegration rate λ₁ (~decadal half-lives) but shift abruptly, within the first ~500 y postmortem, to a 100-fold lower disintegration rate λ₂ (~ millennial half-lives) at sequestration rate τ (burial and/or diagenetic stabilization). This drop permits accrual of a long tail of very old shells even when sequestration is very slow, allowing only a minority (<1%) of shells to survive the first phase (Tomasovych et al. 2014 Geology). We suspect that permanent diagenetic stabilization may be necessary to ensure that shells do not revert to λ₁ after temporary sequestration in favorable pockets within the mixed layer. Preliminary SEM shows significant coarsening of inner shell-layer crystallites, consistent with Ostwald ripening. New cores from 50-m water depth document such L-shaped AFDs in each 2-cm increment within the upper ~25 cm of the seabed, the local penetration depth of callianassid shrimp, indicating both upward and downward advection of shells. AFDs from deeper core increments, cut off from the input of new shells, have flat (uniform) shapes with millennial-scale median ages; shells at these depths represent the 1% that survive λ₁ and are then almost impervious to further destruction. The importance of sequestration (stabilization) to the prolonged persistence of shells and the likely remodeling of shell microstructure, all within the mixed layer, and the coarser temporal resolution of permanently buried shell assemblages, underscore the need to acquire a suite of replicate measurements per increment and to assume that some porewater equilibration of original shell has likely occurred.