Dataset: Cross Seamount subfossil coral radiocarbon data and age modeled dates

Living and subfossil deep-sea proteinaceous corals (Kulamanamana haumaeae) were collected from 447 and 415 m respectively from Cross Seamount (19°N, 158°W), southwest of the Big Island of Hawaii, in 2004 using the DSRV Pisces V. See related dataset (https://www.bco-dmo.org/dataset/884969) for results from living corals.

The polyps of live-collected corals were removed and all skeletons were rinsed first with seawater, followed by freshwater, before being air dried on deck. In the lab, ~0.7 cm thick cross-sectional disks were sectioned from near the base of each skeleton, then polished and mounted to glass slides.

A computerized Merchanteck micromill was used to drill a radial transect from the outside to the inner layers of the coral skeleton cross section at 0.1 mm resolution, yielding 2-3 mg of powdered skeletal material per sample interval. All 168 samples from the live-collected coral were analyzed (155 samples in duplicate); while of the 455 subfossil samples drilled, 193 were analyzed, with a greater sample density analyzed from the outermost 10 mm. Bulk powdered coral material (~0.4 mg) was enclosed in tin capsules for bulk δ13C and δ15N analyses via continuous-flow isotope-ratio mass spectrometry (IRMS) using a Carlo-Erba elemental analyzer connected to an Optima Isotope Ratio Mass Spectrometer (IRMS). C/N ratios were simultaneously determined. The analytical error for these measurements, as determined from analysis of standards of known composition (pugel and acetanilide), was ±0.2‰ for both δ13C and δ15N values and ±0.1 for C/N ratios. To test for the impact of secondary authigenic carbonate in the sample matrix, a subset of the same samples from the subfossil coral (1–2 mg each, n = 7) were acidified using 1 N HCl for 20 h under refrigerated conditions, followed by filtration onto a 0.22 μm glass fiber filter. The filter was dried overnight at 45°C before material was transferred into a tin capsule and analyzed following the same procedure as described above.

Radiocarbon dating was performed on nine sample aliquots (~1 mg total material) each spread ~5 mm apart along the skeleton cross section transect. Samples were first fumed with concentrated HCl and subsequently dried before being transferred with methanol into 6 mm short pre-baked (400C, 60 minutes) quartz tubes and dried. The 6 mm tubes were then placed inside larger 9 mm pre-baked (400C, 60mins) quartz tubes with an appropriate amount of pre-baked (400C, 60 mins) CuO, prior to pumping off excess atmosphere, sealing, and combusting to produce CO2. The CO2 after cryogenic purification was converted to graphite and analyzed at the Lawrence Livermore National Laboratory. Results were corrected for δ13C and background 14C using similarly handled 14C-free coal. Radiocarbon results were transformed into calibrated years before present (yr BP). These ages and subsequent age models were generated with Calib 8.2 (Stuiver et al. 2021) using a local reservoir (ΔR) correction of -177 ± 16, (Druffel et al., 2001; Guilderson et al., 2021) and the Marine20 database (Heaton et al. 2020) and linearly interpreted to provide a continuous age-model.

Issue Report:

For subfossil coral (Coral_Sample_ID: PV588), samples were less than 100 micrograms carbon which is why the analytical errors are large.