Abstract. Water isotope-enabled coupled atmosphere–ocean climate models allow for exploration of the relative contributions to coral stable oxygen isotope (δ18Ocoral) variability arising from sea surface temperature (SST) and the isotopic composition of seawater (δ18Osw). The unforced behaviour of the isotope-enabled HadCM3 coupled general circulation model suggests that the extent to which inter-annual δ18Osw variability contributes to that in model δ18Ocoral is strongly spatially dependent, ranging from being negligible in the eastern equatorial Pacific to accounting for 50% of δ18Ocoral variance in parts of the western Pacific. In these latter cases, a significant component of the inter-annual δ18Osw variability is correlated to that in SST, meaning that local calibrations of the effective local δ18Ocoral–SST relationships are likely to be essential. Furthermore, the relationship between δ18Osw and SST can be non-linear, such that the model interpretation of central and western equatorial Pacific δ18Ocoral in the context of a linear dependence on SST alone leads to overestimation (by up to 20%) of the SST anomalies associated with large El Niño events. Intra-model evaluation of a salinity-based pseudo-coral approach shows that such an approach captures the first-order features of the model δ18Osw behaviour. However, the utility of the pseudo-corals is limited by the extent of spatial variability seen within the modelled slopes of the temporal salinity–δ18Osw relationship.

Water isotope-enabled coupled atmosphere/ocean climate models allow for exploration of the relative contributions to coral stable oxygen isotope (δ¹⁸O_coral) variability arising from Sea Surface Temperature (SST) and the isotopic composition of seawater (δ¹⁸O_sw). The unforced behaviour of the isotope-enabled HadCM3 Coupled General Circulation Model affirms that the extent to which inter-annual δ¹⁸O_sw variability contributes to that in model δ¹⁸O_coral is strongly spatially dependent, ranging from being negligible in the eastern equatorial Pacific to accounting for 50% of δ¹⁸O_coral variance in parts of the western Pacific. In these latter cases, a significant component of the inter-annual δ¹⁸O_sw variability is correlated to that in SST, meaning that local calibrations of the effective local δ¹⁸O_coral–SST relationships are likely to be essential. Furthermore, the relationship between δ¹⁸O_sw and SST in the central and western equatorial Pacific is non-linear, such that the interpretation of model δ¹⁸O_coral in the context of a linear dependence on SST alone may lead to overestimation (by up to 20%) of the SST anomalies associated with large El-Niño events. Intra-model evaluation of a salinity-based pseudo-coral approach shows that such an approach captures the first-order features of the model δ¹⁸O_sw behaviour. However, the utility of the pseudo-corals is limited by the extent of spatial variability seen within the modelled slopes of the temporal salinity–δ¹⁸O_sw relationship

International audienceThe position of the southern boundary of the Pacific warm pool is shown to have been stable since the early Pleistocene, based upon a planktic foraminiferal Mg/Ca-derived reconstruction of subtropical sea surface temperature in the Coral Sea. This contrasts with previous reconstructions showing warm pool contraction from the north and east and means that the early Pleistocene warm pool was more hemispherically asymmetric than its present configuration. The latter was not established until ˜1Ma, supporting a strengthening of the northern Hadley Cell, which was not replicated in its southern counterpart, prior to the Mid-Pleistocene Transition

[1] Cenozoic proxy records of both the isotopic composition of dissolved inorganic carbon in the oceans (d 13 C DIC ) and deep ocean carbonate preservation show significant periodicities in the range 400-500 kyr. Sensitivity analysis of the ocean carbon cycle to potential variability on this timescale in patterns of global oceanic primary productivity and/or continental weathering fluxes is performed using a 7 box ocean-atmosphere model. The data constraints imposed by Plio-Pleistocene proxy records of d 13 C DIC , carbonate preservation, and atmospheric pCO 2 variability allow different scenarios to be evaluated. Forcing with the global oceanic primary production ratio of inorganic and organic carbon leads to the relative phases of response most consistent with the proxy data. However, only when changes also occur in total marine primary production can the observed relative amplitudes of d 13 C DIC and pCO 2 variability also be reproduced. This scenario is consistent with oscillations between a highly productive, coccolithophore-rich global ocean and a less productive, diatom-rich one. Such oscillations have been proposed to originate from the orbital eccentricity cycle leading to changes in seasonality and thus silica utilization in the Southern Ocean. However, the period of response in model d 13 C DIC is always close to that of the forcing function used, and thus a significant discrepancy remains between the orbital eccentricity cycle period of 413 kyr and the observed ∼500 kyr d 13 C DIC periodicity seen in the Pleistocene.

We present 12 seasonally resolved δ 18 O profiles of giant clams (Tridacna sp.) from the Huon Peninsula, Papua New Guinea, spanning discrete periods of time (9-38 years) over the past 60 ka. The interannual anomaly time series of these shells are used to reconstruct interannual variability which can predominantly be attributed in the modern climate to the El Niño-Southern Oscillation (ENSO) in this region. We have found a significant reduction in interannual δ 18 O variance during the early Holocene, whereas during Marine Isotope Stage 3 there were at least some periods with variance not significantly different to the twentieth century. We show that ENSO variability seen during the late twentieth century is rare but not unprecedented within glacial climates.