Ages from the sites with single age dates (Site 1225, 1226, 1230, and 1231) suggest a typical oceanic pattern of precessional Milankovitch marine glacial–interglacial bedding (cf. Hodell, 1993) but with a clearly increasing terrestrial influence from west to east. Sediments are dominantly pelagic marine microfossil oozes but with variable amounts of clay and other terrestrial components. Sedimentation rates revealed by the 14C dates are not considered unusual.
The Peru continental margin is a shelf area dominated by upwelling and bottom-flowing countercurrents over at least the last 40 k.y. These currents prevent supply of oxygen to the bottom waters, as manifested by the absence of bioturbation and finely laminated sediments and additionally result in a number of localized sedimentation hiatuses or unconformities (De Vries and Schrader, 1981), particularly in water depths between 250 and 400 m (Reinhardt et al., 2002). The localized nature of unconformities is demonstrated by the absence of correlation between northernmost Site 1227 and southern Sites 1228 and 1229. In the north, Holocene and latest Pleistocene sediments are absent, and a well-developed late Pleistocene early deglaciation sequence accumulated at the considerable rate of 265 cm/k.y. for at least 1500 yr. The commencement of the rapid sedimentation at ~17.2 ka corresponds almost exactly with the onset of El Niņo-Southern Oscillation activity proposed by Rein et al. (2003). The same time interval at Sites 1228 and 1229 is represented either by absent sections or very low sedimentation rates. Our interpretation is that initial melting of high-Andean glaciers after the LGM resulted in a flood of material and possibly nutrients to the shelf, but deposition on the shelf was not uniform at this time. Clearly, in some areas, particularly nearer the shoreline, sediment bypass or erosion took place.
In contrast, Holocene sedimentation is present at both Sites 1228 and 1229 but is absent from Site 1227. Within the Holocene, however, early to middle Holocene sedimentation (10.0–2.8 ka) was either very slow or missing. This observation corresponds with the findings of other authors that the Peruvian margin and hinterland was undergoing drought, at least during the period between 10.0 and 5 ka (Rein et al., 2003; Moy et al., 2002; Sandweiss et al., 1996), possibly as far back as 15.0 ka (Hebbeln et al., 2002). Most authors attribute this drought to the absence or reduction of El Niņo during this period, resulting in much reduced precipitation along the northern South American hinterland.
The onset of more rapid sedimentation from 2.8 ka coincides with the timing of marked climate changes reported from Chile by Van Geel et al. (2000), which they interpreted to be a result of solar-induced changes to atmospheric circulation. At this stage, we cannot determine with confidence whether the high sedimentation rates along the Peru margin during the early part of deglaciation were enhanced by the onset of El Niņo or were purely a result of glacial outwash, although Skilbeck et al. (2004) suggested the presence of El Niņo cyclicity in these sediments. Our data do support a pattern whereby high sedimentation rates existed from at least 17.2–15.7 ka, were reduced between ~12.0 and 2.8 ka, and then accelerated again over the past 2800 yr.