USGS visual identity mark and link to main Web site

Core OL-92 from Owens Lake, southeast California

AMS radiocarbon dates on sediments from Owens Lake Drill Hole OL-92

James L. Bischoff
U.S. Geological Survey, Menlo Park, California
Thomas W. Stafford Jr.
INSTAAR, University of Colorado, Boulder, Colorado
Meyer Rubin
U.S. Geological Survey, Reston, Virginia

Contents:

Introduction

 The Owens River system consisted of a chain of pluvial lakes occupying a succession of closed basins in southeastern California. The lakes were supplied primarily by the Owens River which drains the eastern side of the Sierra Nevada, and during historic times Owens Lake has been the terminus of the system. During extreme dry periods (interglacials), Owens Lake was saline, alkaline, and biologically highly-productive. During extreme wet periods (glacials), in contrast, the lake was flushed, overflowing with fresh water, and relatively unproductive. To obtain a record of variations in runoff, three 7.6 -cm diameter cores were drilled in Owens Lake in the late spring of 1992. The three cores extend from 0.00 to 7.16m (OL-923), 5.49 to 61.37m (OL-92-1) and 61.26 to 322.86m (OL-92-2), together representing about 800 kyr of sedimentation (Smith, 1993). We report here on results of radiocarbon analyses of carbonates and humates from the upper 31m of the section to place the upper part of the core in temporal context. Our results indicate coherent and linear progression of dates with depth down to 23m and 30 kyr. Scatter of results below this depth indicate that the practical limit of radiocarbon dating in this core is about 30 kyrs.

Stratigraphy and Sampling procedures

The Bishop Tuff, well-dated at 759 kyrs (Sarna and Pringle, 1992) was encountered at 309m in the core (Sarna et al, 1993) indicating an overall average sedimentation rate of about 40 cm/kyr. Correcting for less compaction in the upper part of the core, we estimated the practical limit of about 35 kyr for radiocarbon dating occurs at about 30m depth, and limited our sampling down core to this level. Samples were taken from cores 1 and 3 of OL-92 series. OL-92-1 was drilled by rotary drilling down to 61m. Because the top 5.5m were disturbed by the drilling, OL-92-3, essentially a pushed-in gravity core, was taken immediately adjacent to the drilling pad to provide a high resolution sampling of the top-most 7m of the section. This latter core (OL-92-3)was driven from a back hoe-excavated 3.5m pit to a total depth of 7.16m below the surface. The upper-most stratigraphy as exposed in the pit and in the two cores (Smith, 1993) is as follows (present surface = 0): 0-0.94m artificial fill; 0.94-1.3m historic salt bed; 1.32m - 5.16m oolitic sand; -5.16m to bottom of core silty clay with occasional thin sand beds. The salt bed was deposited between 1912 and 1921 as a consequence of the artificial desiccation of the lake due to the diversion of the Owens River by the City of Los Angeles (Smith, 1993). Thus, the natural section in the time range of interest is characterized by a 3.8m bed of oolitic sand, presumably forming up to historic time, overlaying silty clays. The contact between the oolitic sands and silty clays is sharp and abrupt and represents the most striking lithologic change in the entire core. In fact, oolitic sands are absent throughout the rest of the section.

We took 8 samples of the oolites from depths between 3.72 and 5.12m from OL-92-3, and 12 samples of silty clay from depths between 5.23 and 7.11m from OL-92-3, and from depths between 7.21 and 31.13m from OL-92-1. Correlation of beds between OL-92-1 and OL-92-3 based on lithology and on measured depth was straightforward and unambiguous. Each sample consisted of about 50 cc of wet sediment and represented about 3 cm of section. In addition, we took two samples of silty clay sediment from considerably deeper in the core, beyond the limit of radiocarbon dating in order to asses the 14C background and contamination limits for dating this material. These samples were taken at depths of 36.1m (ca. 50 kyrs) and 55.1m (ca. 80 kyrs).

Extractions and target preparation

The humate fraction of the organic material was extracted from the silty clays by suspending the samples in 200 mls of 1N NaOH solution in capped polyethylene bottles held in a 70°C oven overnight. The coffee-colored supernate was then separated from the residual solids by filtering. The supernate was then titrated to pH 5 with 3N HCl and the resulting precipitate ("humate fraction") was collected on filter paper, rinsed with distilled water and air-dried. Air-dried humate yields ranged from 12 to 200 mg, more-or-less in proportion to the bulk organic content of each sediment sample. Humates were combusted to CO2 in sealed evacuated silica tubes using CuO, Cu and Ag.

Oolite samples were cleaned, weighed, and acidified with 10% HCl in vacuum. The resulting CO2 was dried by passing through an oxygen flow- combustion train, using hot platinum as a catalyst and trapping water and SO2. The CO2 gas volumes for oolites and humates were measured and admitted to graphitizers designed for AMS target preparation (Vogel et al., 1985). These small-volume units were charged with 400 Torr of CO2 and 900 Torr of H2. Heating overnight in the presence of iron and zinc catalysts at 675°C produced sufficient graphite to be pressed into targets for the Lawrence Livermore accelerator. All reactions were carried to completion to eliminate isotope fractionation due to processing. Replicate targets were made of 4 oolite samples and one humate sample evaluate contamination during target preparation and to evaluate precision.

For the two "infinitely old" control samples, targets were prepared of the "total organic", "humate" and "humin" fractions to determine the viability of the CuO-Cu-Ag procedure described above. For the total- organic fraction, a decalcified sediment-sample was combusted. The humate fraction was the NaOH extractable fraction from another aliquot, separated out as above, and the humin fraction was the solid residue after NaOH extraction. Graphite targets were prepared as above. Targets were analyzed at the Center for Accelerator Mass Spectroscopy (CAMS) of the Lawrence Livermore Laboratory by John Southon.

Results

Results for the samples from the top 31m are given in Table 1 and Figure 1. Results from the two deeper samples are shown in Table 2. Dates are based on the Libby half-life (5568 yr), using an assumed d13C of -5 o/oo for the oolites and -15 o/oo for the organic fractions. No attempt has been made to calibrate the results to absolute years (e.g. Bard, et al., 1990). Counting errors on individual samples are generally on the order of ±2% or less. Replicates on the 5 samples for which duplicate targets were prepared agreed within counting error for two of the oolite samples (4.8m and 5.02 m) and between 5 and 6% for the other two oolite samples and the single humate (3.82 m, 4.29m and 31.13m). Results on the "infinite" samples indicate a probable limiting age of between 30 and 35 kyrs (Table 2). For the 36m sample, three replicate analyses of the total- organic fraction yielded dates from 25 to 27 kyr, while the corresponding humate yielded 30 kyr, and the humin 32 kyr. Similarly, the total organics for the 55m sample yielded dates from 35.0 to 35.7 kyrs, the humate 38, and the humin 37. Targets made from infinitely old calcite yielded dates of 43.5 kyr while that of infinitely old coal yielded 35 kyr (unpublished data of U.S.G.S. Reston radiocarbon laboratory), indicating contamination levels introduced during sample processing of carbonates and organic matter and the ideal dating limits. The results from the 55m sample are at this limit, whereas those at 36m are considerable younger. Thus, we conclude that the practical limit for dating humates from the Owens core is about 30 kyr, about the same as encountered in radiocarbon analyses of organic material extracted from other lake sediments (i.e., Robinson et al, 1988; Thompson et al, 1990). This somewhat low limit is probably due to the relative ease of contamination of the plastic and porous-wet sediment by the circulating drilling mud that entrains and slurries younger sediments from the side wall of the hole in the upper parts of the section.

Results on samples younger than the limit of 30 kyrs are relatively coherent, and represent two linear trends with an apparent hiatus between 5100 and 8300 yrs bp (Table 1, Fig. 1). The twelve dates on the oolites, which span only 1.25m of section (3.72 to 5.02 m) define a linear trend which projects to a zero age at -1.5m depth. This depth is essentially that of the base of the historic salt layer (-1.3 m), a result which adds considerable confidence to the dates. The magnitude of the reservoir effect on the dates (low initial 14C/C ratios in lake water), based on the detailed analysis of Benson (1993) for nearby Walker Lake, should be on the order of 200 years and certainly less than 500 yrs.

The dates on the humate extracts likewise result in a linear trend, except for the three samples from -12.97 to -15.32 m. Reexamination of this interval in the core indicated clearly that this section is slumped. The character of this slumped sediment in texture, mottling, color, and bedding is identical to that found above in the section between 8 to 9m where the dates would plot exactly on the trend. The deepest samples at 23.27m and at 31.13m all give dates of 30 to 32 kyrs bp, essentially at the practical limit as indicated by the deeper control samples. Thus, we are left with a coherent trend in the humates beginning at the base of the oolites at 8200 yrs to a date of 25,400 yrs at 23.27 m. The apparent sedimentation rate of the oolites is 70 cm/kyr and that of the silty clay is 83 cm/kyr. The contact between the oolites and the silty clays is abrupt, the bedding apparently conformable, and there is no evidence of pedogenesis at the top of the silty clay. We interpret the offset between the two trends as a disconformity, possibly caused by a period of complete desiccation of the lake followed by wind deflation of pre-existing sediments, or, alternatively, a sublacustrine slumping away of part of the section. The linear trend of the humate dates projects to zero age at +2 m. Thus, because the true zero-age of the section is at -1.3 meters (base of the salt), the amount of missing section represented by the disconformity is about 3.3m.

Deeper in the core, there is no evidence of a significant change in sedimentological conditions as might be expected for the Pleistocene/ Holocene transition, between 10,000 to 14,000 yrs. Rather the significant change in lake conditions seems to have occurred only after 8200 yrs.

Tables

  1. AMS Radiocarbon dates from carbonates and organic matter from sediments from drill hole OL-92, Owens Lake, CA.
    The samples were analyzed at the Center for Accelerator Mass Spectroscopy (CAMS) of the Lawrence Livermore Laboratory by John Southon. Column headings used in the table are listed below: 
        depth = depth in meters
    material = oolites or humate
    CAMS = Sample identification number used by CAMS
    Lab = laboratory that prepared the sample, one of
        B = Institute of Arctic and Alpine Research
        R = USGS, Reston
    14C age = 14C age in years before present
    error = estimate of uncertainty in the 14C age
  2. AMS Radiocarbon dates on various organic fractions from two sediment samples from drill hole OL-92, Owens Lake, taken from depths below radiocarbon dating limit (Extractions and preparation of graphite targets carried out at Institute of Arctic and Alpine Research 14C laboratory. Samples were analyzed at the Center for Accelerator Mass Spectroscopy (CAMS) of the Lawrence Livermore Laboratory by John Southon.) Under "material", total refers to total organics in the sample.

    Depth 36m, approximate age 50 kyr before present: 

        material  CAMS  Lab  14C age±error
    total     6923   B     27470±110
    total     6924   B     26580±110
    total     7562   B     25370±290
    humate    6928   B     29550±190
    humin     6926   B     31760±180
    Depth 55m, approximate age 80 kyr before present: 
        material  CAMS  Lab  14C age±error
    total     6921   B     35030±150
    total     7563   B     35190±990
    total     6922   B     36730±130
    humate    6927   B     38430±170
    humin     6925   B     36730±240

Figures
  1. AMS radiocarbon dates on sediments (oolites and humate extractions) from Owens Lake Drill Hole OL-92.

References

U.S. Department of Interior, U.S. Geological Survey
URL of this page: https://pubs.usgs.gov/openfile/of93-683/5-chronology/3-radio/ams14c.html
Maintained by: Eastern Publications Group Web Team
Last modified: 03.01.01 (krw)