Beyond the Eemian: The Application of an Enhanced Elevated Temperature Infra-Red Stimulated Luminescence Measurement Protocol for Late and Middle Pleistocene Danube Loess

Thiel, C.^1,2, Buylaert, J.-P.², Murray, A. S.², Terhorst, B.³, Tsukamoto, S.¹, Frechen, M.¹

¹Leibniz Institute for Applied Geophysics (LIAG), Geochronology and Isotope Hydrology, Stilleweg 2, D-30655 Hannover, Germany

²Nordic Laboratory for Luminescence Dating, Department of Earth Sciences, University of Aarhus, Risø DTU, Denmark

³Institute of Geography, Chair of Physical Geography, University of Würzburg, Germany

Loess is a good candidate for luminescence dating, partly as a result of the prolonged aeolian transport prior to deposition, which re-sets the luminescence signal to zero, and partly because of its mineral composition, being made up mainly of quartz and feldspars; these are the most commonly used dosimeters in luminescence dating. The age range measured using standard quartz optically stimulated luminescence (OSL) of loess is usually limited by the saturation level of ~200 Gy, i.e. limited to 50-70 ka (for a dose rate of 3-4 Gy/ka). In contrast, feldspar infrared stimulated luminescence (IRSL) signals have a more extended growth curve (up to ~2000 Gy), but the practical age range is hampered by anomalous fading, which leads to significant age underestimation; a reliable correction is still not available for this phenomenon. Because of these limitations, reliable age estimates of loess based on luminescence dating have been restricted to the Late Pleistocene.

This study presents a new IRSL measurement protocol and tests its application using Danube loess from different sites in Austria. As part of our investigations into IRSL signals showing the least fading, we measured the IRSL of polymineral fine grains (4-11 μm) using a SAR protocol including a preheat and cutheat of 320°C, IR stimulation at 50°C and a subsequent post-IR stimulation at 290°C. The IRSL was detected through a Schott BG39/Corning 7-59 filter combination. At the end of each SAR measurement cycle an IR stimulation of 325°C was conducted to minimise signal carryover to the next cycle. Fading was measured for each aliquot.

Using this measurement protocol, indefinitely old samples show a natural signal in field saturation; this has not been observed before. We deduce that the post- IR signal at 290°C does not fade detectably; undesirable fading corrections, which are based on inherent assumptions, are therefore not needed. The good agreement of fading uncorrected IRSL ages with independent age control underlines these findings.

We conclude that this enhanced measurement protocol has the potential to date Middle Pleistocene loess at least up to 300 ka.

Corresponding author: Thiel, C. | FALI MAIL !!!