INTRODUCTION

The California Current system (CCS) forms the eastern limb of the North Pacific Gyre. This complex system of opposing currents brings cool, fresh water to the tropics and exerts a strong influence on the climate of western North America (Lyle et al. 1992; Ramp et al., 1997; Norton and McLain, 1994; Dettinger et al., 1995). Physical processes such as Ekman transport, wind-stress curl, and ultimately thermohaline circulation (Hickey, 1979) enable the CCS to sense basinwide changes in ocean temperatures and wind fields. On historical time scales, this variability also affects coastal circulation, local fisheries, and rainfall along the western coast of North America (Hayward et al., 1996). On decadal time scales in the modern Pacific Ocean, periodic disturbances in basinwide wind fields, sea-surface temperatures (SST), and biological productivity are linked to El Niño/Southern Oscillation (ENSO) events that originate in equatorial regions (Chelton, 1982; Simpson, 1983; Emery and Hamilton, 1985). Typical ENSO conditions along the California margin include unusually warm SST and reduced coastal upwelling (Enfield, 1989), whereas cooler SST and strengthened coastal upwelling typify La Niña conditions. Whether these ENSO connections are analogs to glacial-interglacial cycles on longer, orbital time scales along the California margin remains an interesting but unanswered question. Improved paleoceanographic knowledge of the CCS on glacial-interglacial time scales will help to unlock the complexities between the interactions of wind fields, ice-sheet stability, and variations in the oceanic heat pump of the Pacific. Ocean Drilling Program (ODP) Site 1020 (Fig. 1), a pelagic drill site on the eastern flanks of the Gorda Ridge, lies directly under the CCS but far enough offshore to avoid influence by seasonal upwelling events (Lyle, Koizumi, Richter, et al., 1997). The excellent recovery of an undisturbed sedimentary section at this site provides an opportunity to characterize regional variations in oceanic circulation on orbital time scales.

The role of eastern boundary currents as fundamental components of basinwide oceanic circulation remains poorly understood on orbital time scales in the Pacific. One model suggests that eastern boundary currents may have strengthened during glacial times. CLIMAP Project Members (1981) reconstructed glacial SSTs throughout the oceans and saw markedly cooler water within the eastern boundary current along the coast of northwest Africa and in the Peru Chile Current. They concluded that the inferred cooling resulted from increased upwelling of subsurface water driven largely by more vigorous oceanic circulation.

The difficulty of obtaining paleoceanographic data along the California margin results largely from poor preservation of calcium carbonate (CaCO₃), an important component in almost all marine paleoreconstructions. CLIMAP Project Members (1981) used radiolarian faunal assemblages coupled with oxygen isotope (¹⁸O) stratigraphy to identify a Holocene to last glacial maximum (LGM) SST anomaly of 2°C in the North Pacific, but they characterized the entire region with only two cores. More recently, Sancetta et al. (1992) used diatom fauna, pollen, and sedimentological data from the Multitracers project and Lyle et al. (1992) to characterize the Holocene-LGM oceanographic variability of the northern CCS. They found that marine organic sediment contents (offshore) were lower during the LGM than at present, and their terrestrial pollen records north of 36°N suggested reduced upwelling during the LGM. Sabin and Pisias (1996) examined the marine-terrestrial link from a suite of cores in the latitudinal band of 35°-50°N and found that SST anomalies based on radiolarian transfer functions correlated well (warmer SST and warmer pollen-derived paleotemperatures) with terrestrial pollen records from the Pacific Northwest.

Our study uses sedimentary alkenones as indicators for overlying SST and marine productivity. As organic biomarkers, alkenones provide an excellent opportunity to characterize SSTs where traditional carbonate proxies are poorly preserved because their ratios resist sediment diagenesis (Volkman et al., 1980; Brassell et al., 1986; Prahl et al., 1993; Herbert et al., 1998). Marlowe et al. (1990) identified the coccolithophorid class Prymnesiophyceae as the exclusive photosynthetic algae that biosynthesize alkenones. Brassell et al. (1986) pioneered the use of alkenone paleothermometry and defined a link between alkenone unsaturation and overlying water temperature as the index. Subsequent laboratory culture work by Prahl and Wakeham (1987) and Prahl et al. (1988) on Emiliania huxleyi strain 55a (northeast Pacific) calibrated the index with algal growth temperatures. Since then, other workers have validated this global calibration with respect to core-top sediments and modern SST (Sikes et al., 1997; Rostek et al., 1993; Doose et al., 1997; Sonzogni et al., 1997; Herbert et al., 1998; Müller et al., 1998), and recent studies have characterized paleo-SST and productivity disturbances within the CCS on Holocene-LGM time scales (Prahl et al., 1993, 1995; Doose et al., 1997; Herbert et al., 1995, 1998). Prahl et al. (1995) and Doose et al. (1997) used -derived SSTs from core-top and LGM sediment samples and documented a warming of 3°-4°C since the LGM at 42°N. Unpublished alkenone results from our lab show that Holocene-LGM SST anomalies varied consistently along the entire California margin (23°-41°N; T.D. Herbert et al., unpubl. data), similar to those determined by most faunal studies. One study (Kennett and Venz, 1995), however, estimated an 8°C LGM-Holocene warming in the Santa Barbara Basin using coiling ratios of Neogloboquadrina pachyderma.

Detailed SST records derived from the use of alkenones along the California margin were first published by Prahl et al. (1993). They used data from the Multitracers project to relate alkenone-based SST to modern SST or photic zone temperatures and, at a site ~1°N of Site 1020, found that alkenone-based SST most closely represents winter SST or annual average temperatures at 50 m. Herbert et al. (1998) showed that along the continental margin (water depth <2 km), alkenone-derived SSTs from both Gephyrocapsa oceanica and E. huxleyi most closely relate to mean annual surface temperatures. These studies used numerous core-top samples to conclude that alkenone-derived SSTs exhibit a systematic bias of colder temperatures toward the gyre of up to ~3°C, which may reflect changes in seasonality or depth of production.

We present here an alkenone-proxy record of SST and Haptophyte algal productivity at ODP Site 1020, and we discuss our results with respect to those from other sites along the California margin (Herbert et al., 1995; Prahl et al., 1995; Sabin and Pisias, 1996; Doose et al., 1997; Ortiz et al., 1997). Site 1020 provides the opportunity to assess the scale of SST anomalies and their relationships to Haptophyte productivity and global ice volume well beyond the LGM. Our record extends over eight glacial-interglacial cycles to the Brunhes/Matuyama magnetic reversal boundary (780 ka) and thus represents one of the longest paleoceanographic records available from the California margin.