Running hot and cold in the eastern equatorial Pacific

Understanding paleoceanography of the eastern equatorial Pacific is of particular importance, given its propensity for high variability on all timescales including the seasonal cycle, the interannual El Niño–Southern Oscillation (ENSO), and longer-term changes, each with possible global consequences (Cane and Clement, 1999). For example, tropical Pacific warming imposed in atmospheric climate models can shift the mass balance for northern hemisphere ice sheets from positive (ice growth) to negative (net melting), implying that the tropical oceans could provide a trigger for deglaciation (Hostetler et al., 2006). In spite of the best efforts of many paleoceanographers, however, the region remains enigmatic. Some have considered ice-age climate in the region to be similar to a warm ‘‘El Niño’’ state (Koutavas et al., 2002; Stott et al., 2002; Koutavas and Lynch-Stieglitz, 2003), while others inferred a cold ‘‘La Niña’’ state (Andreasen et al., 2001; Beaufort et al., 2001; Martı́nez et al., 2003), and still others call on ice-age linkages to higher latitudes that may reflect neither of these ENSO states (Pisias and Mix, 1997; Feldberg and Mix, 2003). In this issue, David Lea and colleagues present a welcome new data set; the highest-resolution ice-age temperature data yet available from the eastern equatorial Pacific region based on Mg/Ca in planktonic foraminifera. This proxy has become popular over the past few years, with much of the credit going to Lea and colleagues for their efforts at developing culture-based temperature calibrations (e.g. Lea et al., 1999) and for documenting potential biases such as dissolution (Dekens et al., 2002) and contamination (e.g. Lea et al., 2005). Based on these and other published Mg/Ca data, glacial–interglacial temperature changes in the tropical Pacific are relatively uniform, with ice age cooling of about 3 1C relative to modern conditions both north and south of the equatorial front, and in the eastern and western Pacific. If this geographic uniformity is correct, it is difficult to call on oceanographic mechanisms such as stronger upwelling or advection to explain the cooling. The Mg/Ca data suggest that at the end of the last ice age, warming of the tropics began about 19 ka BP, preceding the major deglacial transition by several thousand years. Lea et al. (2000, this issue) note that the Mg/Ca temperature changes mimic changes in atmospheric carbon