Variability in Coupled Ocean-Atmosphere Models of the Pre-Industrial and Modern Climate

Proxies from the geological record are an invaluable source of climate information, and have led to profound insights about the variability of the climate system. On long (orbital) timescales, the time-averaging of proxies arguably removes most of the higher frequency (centennial and shorter) unforced signal. However, as paleoclimatologists construct paleoclimate records (e.g., isotopic concentrations of foraminiferal calcite from drill cores) with limited spatial coverage but increasing time-resolution (annual to decadal), the interpretation of the proxy record becomes more complicated. It is difficult to separate the forced portion of the high-resolution signal from that arising through natural (intrinsic) variability of the ocean-atmosphere system. A high-resolution isotopic record of foraminiferal calcite from the Santa Barbara Basin is currently being constructed by P. Gomez. As a first step toward deconstructing the forced and natural variability in this record, two long (2000 year) simulations have been completed of the pre-industrial and modern climate using a coupled ocean-atmosphere model (the Fast Ocean-Atmosphere Model). These simulations will be analyzed using common statistical methods to identify spatiotemporal patterns of sea-surface temperature at annual, decadal, and centennial time scales. Pre-industrial patterns of variability will be compared with existing Late Holocene paleoclimate records to determine whether their signals rise above that of naturally occurring variability. Finally, pre-industrial and modern simulations will be compared to determine how anthropogenic influences have modified internal modes of variability.