SITE
SELECTION AND ANALYTICAL METHODS
This manuscript presents
the results of our study of the terrigenous component of sediment deposited
along the northern California margin—particularly Sites 1018 and 1020 (Fig.
1). Composite stratigraphic sections were developed for each site using
high-resolution shipboard measurements of gamma-ray attenuation porosity
evaluator (GRAPE), magnetic susceptibility, and color reflectance data (Lyle,
Koizumi, Richter, et al., 1997). Site 1018 is located ~75 km west of Santa Cruz,
California, on a sediment drift deposit just south of the Guide Seamount at a
water depth of 2477 meters below sea level (mbsl). This site is on top of a
sediment mound that is elevated from surrounding sediments by more then 400 m.
Overall sedimentation rates are extremely high, averaging 100 to 400 m/m.y.
during the late Pleistocene, reflecting the site's close proximity to its
sediment source in the north from Pioneer Canyon, west of San Francisco Bay
(Lyle, Koizumi, Richter, et al., 1997). Site 1018 was selected to provide a
nearshore end-member of terrigenous sediment supplied from the area draining the
central and northern California Coastal Range. Site 1020 contains a continuous
sequence of sediment deposited at rates in excess of 100 m/m.y. during the late
Pleistocene. It is located on the east flank of the Gorda Ridge on an abyssal
hill at a water depth of 3038 mbsl. To the east of Site 1020, the floor of the
Gorda Basin is covered with thick Pleistocene turbidite deposits, but the site
itself contains only a few, thin turbidite layers (these intervals were avoided
during sampling). Site 1020 is located between major sources of sediment
supplied from the Columbia River to the north and the Klamath, Eel, and Rogue
Rivers to the south. This site should enable us to evaluate the relative
importance and magnitude of various continental sources and provide a deeper
water, offshore signal of terrigenous input.
The terrigenous mineral
fraction of these samples was isolated using a series of chemical leaches that
sequentially dissolves biogenic and authigenic phases. Carbonates were removed
by treating each sample with a weak acetic acid solution. A buffered sodium
citrate and sodium hydrosulfite solution was used to remove oxides, hydroxides,
and zeolites, and warm sodium carbonate baths were used to dissolve biogenic
silica. Details of this procedure were provided by Rea and Janecek (1981). The
resulting precision of this method determined from duplicate analyses is ~±5%
relative error. Note that this method provides a segregated sample of
terrigenous material that can be analyzed further for physical, chemical, and
mineralogical parameters without concern about the dilution or contamination
that often accompanies analysis of bulk sediment or other proxy measures of the
terrigenous component.
The mass flux or mass
accumulation rate (MAR) of terrigenous material was determined by the product of
the linear sedimentation rate (LSR), dry bulk density (DBD), and weight
percentage of terrigenous material for each sample as follows:
- terrigenous MAR (g[cm2
· k.y.]-1) =
% extracted x
LSR (cm · k.y.-1) x
DBD (g · cm-3).
Dry bulk densities were
determined from shipboard GRAPE measurements using the relationship given by
Boyce (1976). Linear sedimentation rates were calculated from age models
constructed by correlation of the foraminiferal oxygen isotope stratigraphy at
each site (see Lyle et al., Chap.
11, this volume) with the SPECMAP global chronology (Imbrie et al.,
1984). At the time of publication, only preliminary oxygen isotopic correlation
age models were available for these sites. The record from Site 1018 was
complete through Stage 6 (186 ka) and from Site 1020 through Stage 9 (339 ka),
beyond which we extrapolated age and linear sedimentation rate data between two
control points provided by the top of Pseudoemiliania
lacunosa datum (460 ka) and the Brunhes magnetostratigraphic Chron
(780 ka). All data used to calculate the terrigenous MAR are provided in Table
1 and Table 2 and Figure
2. Although we show terrigenous data collected beyond ages in the oxygen
isotope age model, we have excluded these from our discussion and interpretation
until we are more confident in our age assignments.
Detailed grain-size
measurements of the isolated terrigenous component were made using an electronic
particle size analyzer (Coulter Counter multisizer). This device measures the
spherical volume equivalent of each particle counted as it is drawn through an
electrical field. Mean grain size and standard deviation (sorting) data were
calculated for each sample from a total size distribution based on a count of
150,000 particles divided into 256 size channels in the range of 1-30 µm.
Precision of data is generally within ±0.5 µm based on replicate analyses.
The clay mineralogy of the
terrigenous component was determined on a subset of samples using X-ray
diffraction pattern analysis. Samples were centrifuged to segregate the
<2-µm fraction and X-rayed from 2°
to 30° 2
at 45kV and 40 mA using the XDS Scintag 2000 X-ray diffractometer at the
University of Rhode Island. Peak areas for each phase were normalized to an
internal standard (10% talc) and converted to weight percentages according to
the weighting factors given by Heath and Pisias (1979). The mineral phases
quantified were smectite (17Å; 001, expanded by ethylene glycol), illite (10Å;
001), chlorite (7Å; 002), and kaolinite (7Å; 001). The 7Å peak is divided
between kaolinite and chlorite in proportion to the relative intensities of
their 3.58Å (002) and 3.54Å (004) peaks, respectively (Biscaye, 1965).
