Establishing a luminescence chronology for a palaeosol-loess profile at Tokaj (Hungary): a comparison of quartz OSL and polymineral IRSL

Ann-Kathrin Schatz1, Jan-Pieter Buylaert2, Andrew Murray3, Thomas Stevens4, Thomas Scholten1
1 Soil Science & Geomorphology Group, University of Tübingen
2 Radiation Research Division, Risø National Laboratory for Sustainable Energy, Risø DTU, 4000-Roskilde, Denmark
3 Nordic Laboratory for Luminescence Dating, Department of Earth Sciences, University of Aarhus, Risø DTU, 4000-Roskilde, Denmark
4 Centre for Quaternary Research, Department of Geography, Royal Holloway, University of London, Egham, Surrey TW20 0EX, UK

P 2.18 in Loess and terrestrial archives

The loess-palaeosol sequences in SE Europe are among the longest terrestrial records in Europe and provide important information on local and regional palaeoclimate and palaeoenvironment during the Pleistocene. For the majority of these deposits, independent chronologies have not yet been established, and reliable correlations between different sections, as well as detailed interpretations of proxy records are not possible. In this study we compare OSL ages of fine-grained (4-11µm) quartz with polymineral low-temperature IRSL (50 °C, IR50) and post-IR elevated-temperature IRSL ages (290 °C, pIRIR290) from the Tokaj loess section in NE Hungary. This Late Pleistocene, 15 m long section consists of three loess layers and two intercalated palaeosols; independent age control is fragmentary and based on uncalibrated radiocarbon ages from an adjacent section. Our study establishes the first complete luminescence chronology for this well-studied loess profile.


Preheat plateau test show a drop in quartz OSL De between 160°C and 240 °C, but above 240 °C a clear De plateau is present. Quartz OSL SAR is shown to be generally appropriate for these samples, as indicated by satisfactory luminescence characteristics (low IRSL, bright OSL, recycling ratios ~1, low recuperation), but a satisfactory dose recovery result was only obtained when a large dose was added to a sample without any prior optical or thermal pre-treatment. This resulted in a mean dose recovery ratio of 1.04±0.05 (n=6) after subtracting the natural from the natural dose from the measured dose.


The pIRIR290 SAR protocol also resulted in acceptable dose recovery results (1.08 ± 0.01) when a large dose was added to the natural dose. Bleaching experiments suggest a detectable, non-bleachable residual pIRIR dose of 10 ± 4 Gy. As expected, the mean fading rate (g-value) of the pIRIR290 (1.07 ± 0.11 %/decade) is significantly lower than that of the IR50 signal (2.3 ± 0.2%/decade; n=31, 9 samples), confirming that the pIRIR290 signal is more stable than the IR50 signal. Since the pIRIR290 still shows measurable fading, standard fading correction was carried out for both IR50 and pIRIR290 ages and corrected and uncorrected ages are compared.

Fading correction improves the IR50 results substantially, but even after correction the younger samples still underestimate compared to standard quartz OSL ages; IR50 ages are therefore rejected. Uncorrected pIRIR290 ages and quartz OSL ages agree well and are consistent with independent age control. This seems to confirm that pIRIR290 ages do not need fading correction.

Using the uncorrected pIRIR290 and OSL ages we present the first complete chronology for the Tokaj loess profile. In the context of the Hungarian Late Quaternary geology, our results indicate that the major part of the Tokaj loess has been deposited during MIS3 (60-24 ka), with periods of soil formation between about 35 and 25 ka and during the onset of MIS3 (≥ 58 ka). Our results also indicate episodic deposition of loess and varying, non-linear sedimentation rates. Proxy analyses based on the traditional concept of continuous deposition must be reconsidered in the light of our new data.

last modified 2012-09-11