HR: 16:40h
AN: H34A-02    [Abstracts]
TI: Validation of Isotope Data in the World of Interdisciplinary Science: A Cautionary Tale
AU: * Bullen, T
EM: tdbullen@usgs.gov
AF: WRD/U.S. Geological Survey, MS 420, 345 Middlefield Rd., Menlo Park, CA 94025, United States
AU: Skulan, J
EM: jlskulan@geology.wisc.edu
AF: Dept. of Geology and Geophysics, University of Wisconsin-Madison, Madison, WI , United States
AU: Anbar, A
EM: aanbar@exchange.asu.edu
AF: Dept. of Chemistry and Biochemistry, Arizona State University, Tempe, AZ , United States
AB: Isotope geoscientists, as well as their colleagues in the geologic, hydrologic and biologic science communities, generally have great confidence in both the precision and accuracy of their isotopic data. Moreover, isotope geoscientists are increasingly realizing the potential of their isotope tracing techniques to be used in applied studies in other research disciplines, such as clinical medicine and forensic science. However, researchers in these other disciplines may not have the same confidence in the quality of isotope data as applied to their field of study, due to either lack of familiarity with the techniques or cultural differences in research approach. For example, in a recent attempt to publish a manuscript in which we described the use of Ca stable isotopes in urine to assess the effects of bedrest on bone loss in human subjects, we were confronted by a reviewer who challenged the validity of our isotope data as measured in the urine matrix. In response to this reviewer's concerns, we developed a supplemental study to demonstrate that we were indeed obtaining `the right answer` on the urine samples with the double-spiking, thermal ionization mass spectrometry technique used for the study. First, using the two urine samples from the dataset that had the lightest and heaviest Ca isotope compositions observed in the study (i.e., δ⁴⁴Ca of -1.72‰ and +0.36‰ relative to seawater), we generated a series of mixtures of the lighter sample with seawater (δ⁴⁴Ca = 0‰) and the heavier sample with NIST SRM 915a (δ⁴⁴Ca = -2.01‰). In each case, the amount of standard added to each sample was minute, thus maintaining the urine matrix. Measured Ca isotope compositions agreed with those calculated assuming a simple mixing relationship, confirming the lack of a matrix effect on the accuracy of the data. Second, in order to test the robustness of the double spiking approach, we analyzed the two urine samples under optimally-spiked, under-spiked and over-spiked conditions and found no significant difference in Ca isotope ratios reported by the spike subtraction algorithm. Third, during processing of the two urine samples that had been optimally spiked, we collected Ca off the cation exchange resin purification columns in two fractions, the first 50% eluted being heavy and the second 50% eluted being light due to isotopic fractionation along the pathway through the resin matrix. In each case, the measured Ca isotope compositions of each fraction were identical, implying that the urine matrix did not impact the separation procedure. In the end, these results assuaged the reviewer`s concerns, and this experience demonstrates the challenges isotope geoscientists face as they move further into the realm of applied interdisciplinary science.
DE: 1041 Stable isotope geochemistry (0454, 4870)
DE: 1094 Instruments and techniques
DE: 9820 Techniques applicable in three or more fields
SC: Hydrology [H]
MN: 2007 Joint Assembly