Ocean Drilling Program, United States

We analyzed downhole logging and core petrophysical measurements from Holes 1095B and 1096C of Ocean Drilling Program Leg 178 using wavelet transforms to assess the existence of sedimentary cycles. Both holes are located in the same drift deposit ( approximately 67 degrees S, 78 degrees W) on the continental rise off the Pacific side of the West Antarctic Peninsula, with Hole 1095B being more distal. At both drill sites, nearly continuous sedimentary sections >570 m thick were recovered, which together span roughly the past 9.6 Ma. To properly assess the significance of apparent cyclicity obtained from the wavelet transforms of the log and core measurements, we also analyzed the December insolation signal at 67 degrees S and its truncated form, in which the signal is set to 0 if the true insolation value is <510 W/m (super 2) . Direct comparison can be made by displaying all the results as depth-wavenumber maps, which is done for the insolation signal by converting time to depth using the known sedimentation rates at Sites 1095 and 1096. We find that the depth-wavenumber maps can be quite complicated to interpret over intervals in which the sedimentation rates vary abruptly, such as for Site 1095, where the rates range from approximately 20 to 110 m/m.y. Thus, accurately estimating the dominant periods, even for the insolation signal where the Milankovitch periods are known to be present, is difficult, though not impossible. Because the sedimentation rates are not known exactly and have uncertainties, any dominant period may be offset from its true value. Even with these caveats, several of the data sets have cyclicity. In some cases (e.g., the uranium natural gamma logs for Hole 1096C), the dominant wavenumbers appear to correspond closely to the 100-, 41-, and 23-k.y. Milankovitch periods. Generally, other periods dominate over certain depth intervals, of which periods of 30-35, 50-55, and 74-78 k.y. appear to be more common. In other data sets, the dominant wavenumbers and periods vary with depth, which could indicate significant transitions in cyclicity, provided that the sedimentation rates are accurate. Two of these transitions occur at approximately 2.5 and 3.5-3.8 Ma, ages that appear to be contemporaneous with paleoclimate transitions determined by prior studies.