Scale-dependent effects of growth rate on temperature reconstruction in a marine biogenic carbonate

• Among specimens, the slope of mean Mg/Ca ratios is canonical with temperature • Within specimens, the slopes are steeper and vary negatively with growth rate • Intra-colony effects appear to reflect growth rate-dependent variable microstructure • Different slopes among and within colonies explain discrepancies in reconstructions • Similar microstructural effects on proxies may be widespread in marine calcifiers Analyses of a temperature proxy—Mg/Ca ratios—in the calcitic skeletons of bamboo corals (Octocorallia: Keratoisididae and Mopseidae) show that the effects of ambient temperature and growth rate on Mg/Ca (and oxygen isotope ratios) differ depending on the scale of the analysis. At the whole-of-specimen scale, Mg/Ca ratios are canonically positively correlated with temperature and influenced by growth rate-dependent entrapment and diffusion. However, within specimens, Mg/Ca is much more steeply positive with temperature and correlates negatively with growth rate. Within specimens, the slope of the (Mg/Ca)/temperature relationship itself varies with growth rate, which is hypothesized to result from micro-scale shifts in the relative abundance of slow and fast growing crystal types that, in turn, differ in their affinity for Mg++. This variable slope critically affects proxy-based temperature reconstruction, can underlie frequent mis-matches in marine calcifiers between environmental records and proxy-based reconstructions, and suggests cautious application within-specimens of proxies developed and calibrated using geographic “core top” comparisons.