Disgrifiad
Accurate measurements of equivalent dose (De) underpin the use of luminescence methods for retrospective dosimetry and dating. In the last 20 years new protocols have been introduced for measurement of quartz and of feldspar, almost entirely based upon single aliquot approaches. The ability to make replicate measurements of De using different aliquots derived from a single sample made it possible to examine the distribution of De values. Many researchers have studied the spread of these De values and either used the pattern to make inferences about the sample (such as the extent to which it was bleached at deposition, the impact of microdosimetry, physical mixing or bioturbation), or have applied statistical models to extract an appropriate value for age determination.A consistent theme of research throughout the last 20 years has been a debate about the extent to which variations in De can be attributed to differences in the trapped charge population in the sample (i.e. the dose) and the extent to which the differences arise from aspects of the protocol or the instrument used for measurement. Instrumental causes of variability in De are already known (e.g. the inhomogeneity of some beta sources used for irradiation, Ballarini et al. (2006)). However, other instrumental sources are not well documented and are often simply grouped under the collective heading “Instrumental uncertainty” (Duller et al. 2000; Thomsen et al. 2005).
This presentation describes one source of this instrumental uncertainty, and demonstrates that the magnitude of this effect varies between different readers. A simple method of assessing the magnitude of this effect is described, coupled with a means of correction to remove this source of uncertainty from datasets. The variability observed is difficult to detect as part of routine instrument operation, but will impact both multiple grain and single grain datasets if uncorrected.
Cyfnod | 11 Medi 2018 |
---|---|
Teitl y digwyddiad | UK Luminescence and Electron Spin Resonance Dating conference |
Math o ddigwyddiad | Cynhadledd |
Lleoliad | Sheffield, Teyrnas Unedig Prydain Fawr a Gogledd IwerddonDangos ar fap |
Graddau amlygrwydd | Rhyngwladol |