ISOTOPIC RESPONSES TO INTERANNUAL CLIMATE VARIABILITY SIMULATED BY AN ATMOSPHERIC GENERAL-CIRCULATION MODEL

High-resolution isotopic records from remote tropical and temperate sites may provide new insights into recent climatic variability, but the interpretation of these records requires a thorough understanding of the controls on interannual variability in the isotopic content of precipitation. To investigate the relationship between isotopic variability and interannual climate change, we use the isotopic tracer version of the Goddard Institute for Space Studies (GISS) general circulation model (GCM). We generate interannual-type climate variability by specifying sea surface temperature (SST) anomalies observed during extreme cool (1955-1956) and warm (1982-1983) phases of the El Nino/Southern Oscillation (ENSO) system during January and July. In response to SST anomalies, the GCM simulates many tropical and extratropical climatic features typical of ENSO variability. Across the tropics, simulated isotopic anomalies correlate significantly with rainfall amount. At higher latitudes, altered pressure distributions set up surface wind changes that generate significant regional temperature anomalies. Extratropical deltaO-18 anomalies reach +/- 4-5 parts per thousand but in most regions do not correlate with temperature anomalies. We examine four regions from which interannual paleoclimatic and/or modem isotopic records are available: the tropical western Pacific, tropical South America, western-central Europe and the Tibetan Plateau. In the Pacific warm pool, the GCM quantitatively reproduces the observed inverse relationship between the amount and isotopic content of rainfall, which provides the basis for coral paleoclimatic reconstruction of ENSO-related rainfall anomalies. In tropical South America, isotopic anomalies also correlate strongly with precipitation amount, consistent with the hypothesis that air mass stability over the Amazon basin controls deltaO-18 variability downstream at the Quelccaya ice core site. In Europe, interannual temperature changes correlate positively with deltaO-18 anomalies, quantitatively consistent with long-term observations and supporting temperature-based interpretations of paleoclimatic records. On the Tibetan Plateau, isotopic variations do not correlate with local climatic anomalies nor with changes in oceanic vapor sources, suggesting a complex climatic signal in Tibetan ice cores possibly mediated by continental processes. In contrast to our interannual results, last glacial maximum (18 ka BP) simulations yield significant correlation between temperature and isotopic anomalies. The model thus suggests distinct modes of isotopic response to simulated climate variability that depend on the scale and processes of climate change. Interannual temperature changes often result from time-varying advective changes that can confound the relationship between temperature and isotopic variations, via changes in vapor sources and transport. The near-global temperature change associated with glacial-interglacial variations yields a more consistent pattern of isotopic variability because (1) the effects of large-scale cooling are integrated over the entire pathway of vapor transport and (2) the degree of cooling causes an isotopic response that exceeds the level of variability produced by advective changes.