The potential origins and palaeoenvironmental implications of high temporal resolution δ18O heterogeneity in coral skeletons

δ18O was determined at high spatial resolution (beam diameter ∼30 μm) by secondary ion mass spectrometry (SIMS) across 1–2 year sections of 2 modern Porites lobata coral skeletons from Hawaii. We observe large (>2‰) cyclical δ18O variations that typically cover skeletal distances equivalent to periods of ∼20–30 days. These variations do not reflect seawater temperature or composition and we conclude that skeletal δ18O is principally controlled by other processes. Calcification site pH in one coral record was estimated from previous SIMS measurements of skeletal δ11B. We model predicted skeletal δ18O as a function of calcification site pH, DIC residence time at the site and DIC source (reflecting the inputs of seawater and molecular CO2 to the site). We assume that oxygen isotopic equilibration proceeds at the rates observed in seawater and that only the aqueous carbonate ion is incorporated into the precipitating aragonite. We reproduce successfully the observed skeletal δ18O range by assuming that DIC is rapidly utilised at the calcification site (within 1 h) and that ∼80% of the skeletal carbonate is derived from seawater. If carbonic anhydrase catalyses the reversible hydration of CO2 at the calcification site, then oxygen isotopic equilibration times may be substantially reduced and a larger proportion of the skeletal carbonate could be derived from molecular CO2. Seasonal skeletal δ18O variations are most pronounced in the skeleton deposited from late autumn to winter (and coincide with the high density skeletal bands) and are dampened in skeleton deposited from spring to summer. We observed no annual pattern in sea surface temperature or photosynthetically active radiation variability which could potentially correlate with the coral δ18O. At present we are unable to resolve an environmental cue to drive seasonal patterns of short term skeletal δ18O heterogeneity.