32 PAGES news • Vol 18 • No 1 • April 2010 Science Highlights: Open Section El Niño/Southern Oscillation and changes in the zonal gradient of tropical Paciic sea surface temperature over the last 1.2 ka JESSICA L. CONROY 1 , J.T. OVERPECK 1,2,3 AND J.E. COLE 1,3 1 Department of Geosciences, University of Arizona, Tucson, USA; jconroy@email.arizona.edu 2 Institute of the Environment, University of Arizona, Tucson, USA; 3 Department of Atmospheric Sciences, University of Arizona, Tucson, USA Two new climate reconstructions from the tropical Paciic show large changes in the zonal gradient of sea surface temperature during the last 1.2 ka, with a much stronger zonal gradient from 1000-1300 AD. The tropical Paciic is home to the El Niño/ Southern Oscillation (ENSO), the largest source of interannual variability in the global climate system (McPhaden et al., 2006). Our understanding of this impor- tant climate mode is limited by incom- plete knowledge of the history of ENSO variability and the inability of many cli- mate models to correctly simulate aspects of tropical Paciic climate (Guilyardi et al., 2009). Recently published lake sediment records from the Galápagos archipelago (Conroy et al., 2009a) and a high-reso- lution marine sediment record from the Makassar Strait (Oppo et al., 2009) add to the emerging picture of past changes in tropical Paciic climate. These new sedi- ment records from the eastern equatorial Paciic (EEP) cold tongue and Indo-Paciic warm pool (IPWP), two key ENSO regions, indicate substantial multidecadal to cen- tennial variability in sea surface tempera- ture (SST) during the last two millennia. Here we use these proxy records and avail- able ENSO reconstructions to explore the changes in the zonal gradient of tropical Paciic SST and ENSO variability over the last 1.2 ka. The zonal gradient of tropical Paciic SST over the last 1.2 ka A sub-decadally resolved sediment re- cord from El Junco Lake (Galápagos ar- chipelago) provides a new climate record from the EEP (Conroy et al., 2009a). The strength of this record is that the proxy— changes in the ratio of tychoplanktonic to epiphytic diatoms (or, diatoms that are suspended in the water column via wind turbulence and those living attached to shoreline vegetation)—correlates signii- cantly with instrumental SST during the 20 th century (Conroy et al., 2009a). El Junco is a closed-basin lake, and changes in lake level result from changes in rainfall, which in the Galápagos are tightly coupled to EEP SST. When SST is warm, precipitation increases, lake level rises, and there are more tychoplanktonic diatoms in the lake. When lake level drops from decreased precipitation due to cooler SST, a greater fraction of the lake area is near-shore, and more epiphytic diatoms are deposited at the core site. The diatom-inferred SST indi- cates that the warmest SST in the eastern tropical Paciic in the last 1.2 ka occurred in the last 50 years (Fig. 1a). Although an increase in eutrophication in the 20 th cen- tury due to cattle grazing around the lake likely contributed to an overall increase in diatom concentration, the increase in nutrients did not alter the ratio of tycho- planktonic to epiphytic diatoms, which responds primarily to lake level. The trend in our ratio begins prior to the increase in concentration, and changes in this ratio are not correlated with diatom concentra- tion. Nutrient loading is also highest near shore, and thus we would expect to see an increase in shore species relative to open water species if there was a preferential impact on some diatom species. However, we observe a decline in the abundance of shoreline species during the last century. Several pollen time series from the El Junco core and a Galápagos coral record of ∆ 14 C (Guilderson and Schrag, 1998) agree with our hypothesis of recent SST warm- ing. However, other records, including El Junco grain size (Conroy et al., 2008) and a δD record derived from botryococcene algae (Sachs et al., 2009) show markedly diferent multidecadal to century-scale variability compared to the diatom record, even though all these proxies are hypoth- esized to be related to precipitation. The lack of agreement between diferent hy- pothesized climate indicators within the same basin likely indicates diferent factors inluencing various aspects of the lake en- vironment. For example, sand abundance is likely controlled more by the intensity of rainfall than mean rainfall, whereas the di- atom record, which responds to lake level changes, is driven by both seasonal and interannual rainfall and the balance be- tween precipitation and evaporation. Dif- ferences among the proxies highlight the need for not only a multiproxy approach when evaluating climate proxies, but also modern limnological analyses and calibra- tion with the instrumental record. To date, only the diatom record is signiicantly cor- related with instrumental climate data. A new, continuous, high-resolution reconstruction of SST from the IPWP was Figure 1: Sea surface temperature (SST) reconstructions and zonal SST gradient of the tropical Paciic over the last 1.2 ka. a) Indo Paciic Warm Pool (IPWP) SST reconstruction in red (Oppo et al., 2009), eastern equatorial Paciic (EEP) SST reconstruction in blue (Conroy et al., 2009a). Numbers in parentheses are diatom index values from Conroy et al. (2009a). b) Zonal SST gradient (°C), calculated from the combined IPWP and EEP records. Light blue curve is SST gradient calculated from instrumental SST data in the grid cells containing the proxy records (Smith et al., 2008). Horizontal line indicates long-term mean zonal SST gradient.