Taphonomic trade-offs in tropical marine death assemblages: Differential time averaging, shell loss, and probable bias in siliciclastic vs. carbonate facies

Radiocarbon-calibrated amino-acid racemization ages of individually dated bivalve mollusk shells from Caribbean reef, nonreefal carbonate, and siliciclastic sediments in Panama indicate that siliciclastic sands and muds contain significantly older shells (median 375 yr, range up to ∼5400 yr) than nearby carbonate seafloors (median 72 yr, range up to ∼2900 yr; maximum shell ages differ significantly at p < 0.02 using extreme-value statistics). The implied difference in shell loss rates is contrary to physicochemical expectations but is consistent with observed differences in shell condition (greater bioerosion and dissolution in carbonates). Higher rates of shell loss in carbonate sediments should lead to greater compositional bias in surviving skeletal material, resulting in taphonomic trade-offs: less time averaging but probably higher taxonomic bias in pure carbonate sediments, and lower bias but greater time averaging in siliciclastic sediments from humid-weathered accretionary arc terrains, which are a widespread setting of tropical sedimentation.