5.	ACCELERATOR MASS SPECTROMETRY (AMS)

5.1	Paleoclimate studies using AMS radiocarbon {EMBED EQUATION |} dating of 
pollen from lake sediments and peat 	deposits
	T.A. Brown,* G.W. Farwell and P.M. Grootes†
	We have continued our development and refinement of techniques for the 
isolation and AMS {EMBED EQUATION |} dating of essentially pure pollen fractions
 from lake sediment and peat samples under partial funding through an NSF grant 
under the Paleoclimate from Arctic Lakes and Estuaries (PALE) Initiative of the 
ARCSS Program (Grant No. ATM91-23963).  We have applied these techniques to a 
number of cores taken from lakes and bogs in Washington, British Columbia, and 
Alaska.  
	Some of our results were summarized in last year's Annual Report.1  A 
comprehensive description of the work and a discussion of the results are given 
in the Ph.D. dissertation of Thomas A. Brown.2  A paper describing the principal
 features of our {EMBED EQUATION |} AMS system and its performance (absolute 
accuracy, ± 0.5% or   {SYMBOL 177 \f "MT Symbol"} 40 years) has now been 
published.3
	Radiocarbon dating of organic material from lake sediments and peat 
deposits has been used for decades in paleoclimate studies.  During the past 
several years, AMS measurements have increasingly supplanted the traditional 
{EMBED EQUATION |}-counting method of {EMBED EQUATION |} dating since they offer
 greatly increased sensitivity (sub-milligram samples can be dated) and, in many
 instances, greater accuracy; they are also very much faster.  However, the 
typical "bulk carbon" sample preparation techniques used for both AMS and {EMBED
 EQUATION |}-counting leave unanswered the question of just what is being dated,
 a grave disadvantage in palynological and other paleoclimate studies.  In 
contrast, the extraction procedures2 that have been developed here typically 
produce purified pollen samples which can be clearly identified under the 
microscope; thus, an unambiguous proxy climate indicator --pollen-- is dated, 
and nothing else.  Additionally, our results to date2,4 show that the extraction
 and dating of pollen fractions eliminates "hard water effects" as sources of 
dating errors (sometimes thousands of years) and demonstrate that significant 
age differences can exist between pollen and macrofossils at the same level in a
 sediment core.
	Recent results from the application of our pollen extraction/AMS dating 
procedures to a low-organic-content Arctic lake sediment core2 were useful in 
rectifying an apparent age-depth reversal in the results obtained from "bulk 
carbon" samples from the core.  They also demonstrated a clear need for careful 
screening of prepared samples, through microscopic examination, for suitability 
in obtaining valid and consistent radiocarbon dates for palynological and other 
paleoclimate studies.
	With the recent departures of Thomas A. Brown and Pieter M. Grootes it 
has become unfeasible to continue the program of radiocarbon AMS measurements 
here in the Nuclear Physics Laboratory.  The NSF-PALE paleoclimate work will 
continue, however, as a collaborative effort, with the AMS measurements to be 
carried out at the Lawrence Livermore National Laboratory, Livermore, 
California.
*Now at Center for Accelerator Mass Spectrometry, L-397, Lawrence Livermore 
National Laboratory, Livermore, CA, USA 94551.
†Now at C-14 Leibnitz Labor, Leibnitzstrasse 19, Christian Albrechts 
Universität, 24118 Kiel, Germany.
1Nuclear Physics Laboratory Annual Report, University of Washington (1994) pp. 
30-31.
2Thomas A. Brown, Ph.D. Dissertation (Geophysics), University of Washington 
(1994).
3T.A. Brown, G.W. Farwell, and P.M. Grootes, Proceedings of the 6th 
International Conference on Accelerator Mass Spectrometry (1993), Nucl. Instrum.
 Methods B 92, 16 (1994).
4T.A. Brown, G.W. Farwell, and P.M. Grootes, presented at the 15th International
 Radiocarbon Conference, August 15-19, 1994, University of Glasgow, Scotland.


35