Cenozoic Island Dolostones
A best-case test of geochemical proxies in ancient dolostones. Even where diagenesis looks as close to “textbook” low-temperature, seawater dolomitization as the rock record gets, carbonate clumped isotopes (Δ47) and cerium anomalies (Ce/Ce*) remain intact, while δ238U isotopic ratios have been overprinted.
Overview
Modern examples of low-temperature dolomite formation are rare, which makes Cenozoic “island dolostones” — like those on Grand Cayman and Cayman Brac (British West Indies) — one of the best natural laboratories available for testing how carbonate geochemical proxies hold up during dolomitization. These platforms are geologically young, have never been deeply buried, and the conditions responsible for dolomitization can be reasonably inferred.
A well-documented dolomitization front on Grand Cayman allows for these proxies to be directly tested. Using carbonate clumped isotopes (Δ47), dolostones along the periphery of the island formed at a temperature of 19.1 ± 1.4°C from a fluid with δ18OFLUID = 0.2 ± 0.5‰ VSMOW, indistinguishable from normal seawater. Petrographically, the dolomite crystals are pristine. By any standard, this is about as close to textbook, low-temperature seawater dolomitization as the geological record offers.
Then we measured uranium isotopes (δ238U) in that same dolomitization front. Despite pristine petrography and an unambiguous seawater-temperature and δ18OFLUID signature, δ238U in the dolostones is offset from both modern seawater and the coeval limestone: A clear diagenetic overprint. Cerium anomalies (Ce/Ce*), measured on the same samples, show no such offset at all. Two proxies, same rocks, same fluid history: One survives, one doesn't. Nothing in the texture or the clumped-isotope temperature would have warned us which.
Reconstructing ancient seawater chemistry from dolomitized rocks therefore requires a diagenetic correction factor for δ238U, calibrated to the specific depositional and diagenetic setting, rather than one fixed value applied everywhere. A proxy that fails under otherwise ideal, texturally pristine conditions is a reminder that every geochemical archive needs an independent check like this one before its signal is taken at face value.
Key Figures