Scale dependence of environmental and physiological correlates of δ 18 O and δ 13 C in the magnesium calcite skeletons of bamboo corals (Gorgonacea; Isididae)

We examine in detail δ 18 O and δ 13 C in the calcite internodes of bamboo corals as potential proxies of physiological and environmental variability, through (a) a “core top” calibration that includes specimens from a wide range of habitats and environmental conditions and (b) a comparison of high resolution serial point analyses along radial growth axes of a sub-set of specimens with each other, with instrumental temperature and salinity records, with growth rates and with a nominal skeletal proxy for temperature (Mg/Ca) in the same specimens. At the whole-of-specimen level, δ 18 O and the intercept of the strong within-specimen regression of δ 18 O against δ 13 C correlates highly with ambient temperatures at slopes that are identical to those reported for other marine biogenic carbonates (−0.22 per °C). δ 13 C varies predominantly with apparent specimen-mean growth rate. It also correlates with the slope of the within-specimen covariance between δ 18 O and δ 13 C, which in turn is distributed bi-modally among specimens and linked to differences in apparent growth rates. Within-specimens, variability in δ 13 C, and to a lesser extent δ 18 O, correlates between specimens collected in the same region and differs between regions, implying an environmental effect, but the factors involved for either isotope ratio are unclear. Correlations between δ 18 O and temperature (and Mg/Ca) range from positive to negative among specimens and appear to vary over time even within specimens. The mismatch between the consistent temperature-dependence of δ 18 O at the whole-of-specimen level and the mixed relationship within-specimens can be reconciled by assuming an unknown temperature-dependent factor affecting δ 18 O during the growth of Isidid calcite. The contrast between the results of the “core top” temperature calibration for δ 18 O, which are consistent with studies of other carbonates, and the apparently more complex suite of factors affecting both δ 13 C and δ 18 O within specimens strongly suggests that calibrations based on a “core top” approach be applied cautiously to environmental reconstructions based on variations of a nominal proxy within specimens.