A luminescence-based chronology for the Harletz loess sequence, Bulgaria Johanna Lomax, Markus Fuchs, Pierre Antoine, Denis-Didier Rousseau, France Lagroix, Christine Hatté, Samuel Taylor, Jessica Till, Maxime Debret, Olivier Moine, et al. To cite this version: Johanna Lomax, Markus Fuchs, Pierre Antoine, Denis-Didier Rousseau, France Lagroix, et al.. A luminescence-based chronology for the Harletz loess sequence, Bulgaria. Boreas, Wiley, 2019, 48 (1), pp.179-194. 10.1111/bor.12348. hal-02352640 HAL Id: hal-02352640 https://hal.archives-ouvertes.fr/hal-02352640 Submitted on 1 Apr 2021 HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. ___________________________________ This is not the published version of the article / Þetta er ekki útgefna útgáfa greinarinnar Author(s)/Höf.: Johanna Lomax, Markus Fuchs, Pierre Antoine, Denis-didier Rousseau, France Lagroix, Christine Hatté, Samuel Neil Taylor, Jessica Lynn Till, Olivier Moine And Diana Jordanova Title/Titill: A luminescence based chronology for the Harletz loess sequence, Bulgaria Year/Útgáfuár: 2018 Version/Útgáfa: Pre-print (Óritrýnt handrit) Please cite the original version: Vinsamlega vísið til útgefnu greinarinnar: Lomax, J., Fuchs, M., Antoine, P., Rousseau, D.-D., Lagroix, F., Hatté, C., . Jordanova, D. (2019). A luminescence-based chronology for the Harletz loess sequence, Bulgaria. Boreas, 48(1), 179-194. doi:10.1111/bor.12348 Rights/Réttur: Copyright © 1999-2019 John Wiley & Sons, Inc. All rights reserved Page 1 of 41 Boreas 1 1 2 3 4 1 A luminescence based chronology for the Harletz loess sequence, Bulgaria 5 6 2 JOHANNA LOMAX, MARKUS FUCHS, PIERRE ANTOINE, DENIS-DIDIER ROUSSEAU, 7 3 FRANCE LAGROIX, CHRISTINE HATTÉ, SAMUEL NEIL TAYLOR, JESSICA LYNN TILL, 8 4 MAXIME DEBRET, OLIVIER MOINE AND DIANA JORDANOVA 9 10 5 Lomax, J., Fuchs, M., Antoine, P., Rousseau, D.-D., Lagroix, F., Hatté, C., Taylor, S. N., Till, J. L., 11 6 Debret, M., Moine, O. & Jordanova, D.: A luminescence based chronology of the Harletz loess 12 7 sequence, Bulgaria. Boreas…. 13 8 14 9 15 16 10 The Harletz loess-paleosol-sequence is located in north-western Bulgaria and represents an important 17 11 link between well studied loess sequences in eastern Romania and further sites to the west of the 18 19 12 Carpathians (e.g. Serbia Forand Hungary). Review The aim of this study Only is to establish a chronostratigraphy of the 20 13 deposits, using various methods of luminescence dating, together with basic stratigraphic field 21 22 14 observations as well as magnetic properties. Luminescence dating was carried out using the quartz fine 23 15 grain fraction and a SAR protocol, and the feldspar coarse grain fraction, applying the MET-pIRIR 24 25 16 protocol. Due to underestimation of the quartz fine grain fraction in the lower parts of the sequence, 26 17 the resulting chronology is mainly based on the feldspar ages, which are derived from the stimulation 27 28 18 temperature at 150 °C. A comparison with nearby sequences from Serbia, Hungary and Romania, and 29 19 interpretations obtained through the stratigraphic and sedimentological signature of the sequence 30 31 20 supports the established chronology. Our data suggest that the prominent paleosol (soil complex) in 32 21 the upper quarter of the sequence was formed during MIS 5. It would follow that large parts of last 33 34 22 glacial loess overlying this paleosol were probably eroded, and that the thick loess accumulation 35 23 underlying this soil complex can be allocated to the penultimate glacial (MIS 6). A prominent MIS 6 36 37 24 tephra, which has been reported from other sequences in the area, is also present at Harletz. 38 39 25 40 41 26 Johanna Lomax ([email protected]) and Markus Fuchs, Department of 42 27 Geography, Justus-Liebig-University Giessen, 35390 Giessen, Germany; Pierre Antoine and Olivier 43 28 Moine, UMR CNRS-Univ. Paris 1-UPEC 8591, Laboratoire de Géographie Physique, 92195 Meudon, 44 29 France; Denis-Didier Rousseau, Ecole Normale Supérieure de Paris, Laboratoire de Météorologie 45 30 Dynamique, UMR CNRS 8539, 75231 Paris, France and Lamont–Doherty Earth Observatory of 46 31 Columbia University, Palisades, NY 10964, USA; France Lagroix, Samuel Neil Taylor and Jessica 47 32 Lynn Till, Institut de Physique du Globe de Paris, Sorbonne Paris Cité, Univ Paris Diderot, UMR 48 33 7154 CNRS, 75005 Paris, France; Christine Hatté, Laboratoire des Sciences du Climat et de 49 34 l’Environnement, UMR 8212 CEA-CNRS-UVSQ, Université Paris-Saclay, 91198 Gif-sur-Yvette, 50 35 France; Jessica Lynn Till, Institute of Earth Sciences, University of Iceland, 101 Reykjavik, Iceland; 51 36 Maxime Debret, UFR Sciences et Techniques, Université de Rouen Normandie, 76000 Rouen, France; 52 37 Diana Jordanova, National Institute of Geophysics, Geodesy and Geography, Bulgarian Academy of 53 38 Sciences, Sofia, Bulgaria; received 8th January 2018, accepted 21th August 2018. 54 39 55 40 56 41 57 58 59 60 Boreas Page 2 of 41 2 1 2 3 4 1 Loess-paleosol-sequences are important pedo-sedimentary archives, allowing to reconstruct 5 6 2 palaeoenvironmental parameters such as variations in both wind directions and speeds, aridity, 7 8 3 sediment supply, and vegetation cover but also temperature and precipitation. This information is only 9 10 4 useful, when a detailed chronology is provided along with the palaeoenvironmental information 11 12 5 released from the available proxies. In order to establish a numerical chronology in loess deposits, 13 14 6 basically two methods are available. One is radiocarbon dating, which has been successfully applied 15 16 7 for example to mollusc shells at Dunaszeckcső (Hungary) (Újvári et al. 2014) and to earthworm 17 18 8 calcite granules at Nussloch (Germany) (Moine et al. 2017). But since radiocarbon dating has an upper 19 For Review Only 20 9 age limit of around 45 ka and can thus only date Last Glacial loess, the standard method for dating 21 22 10 loess, especially for a timeframe beyond 45 ka, are luminescence techniques. 23 24 11 25 In many sedimentary contexts, optically stimulated luminescence (OSL) dating of quartz is 26 12 preferred over infrared stimulated luminescence (IRSL) dating of the feldspar fraction, because the 27 28 13 quartz signal is assumed to be athermally stable, as opposed to feldspar, which shows anomalous 29 30 14 fading of the signal, which can lead to significant age underestimation. Loess deposits though usually 31 32 15 have high dose rates of around 3 Gy ka-1, which makes quartz problematic with respect to the upper 33 34 16 age limit. A range of studies show that above a stored paleodose of around 150-200 Gy the quartz 35 36 17 signal approaches saturation (e.g. Chapot et al. 2012; Timar-Gabor et al. 2015a). Transferred to loess 37 38 18 with its typical dose rate, this corresponds to an upper age limit of around 50-70 ka. Beyond this 39 40 19 range, quartz ages can still be achieved, mainly because laboratory generated growth curves can grow 41 42 20 up to doses of more than 500 Gy (e.g. Chapot et al. 2012), but such OSL ages can be underestimated 43 44 21 (e.g. Lai 2010; Lowick & Preusser 2011; Chapot et al. 2012; Timar-Gabor et al. 2015a). 45 46 47 22 A further problem when dating the quartz fraction may be different ages derived from different 48 49 23 grain sizes. This was shown by various studies in the Lower Danube region and Carpathian Basin, in 50 51 24 which contrasting luminescence ages were determined when dating the fine (4-11 µm) or coarse (63- 52 53 25 125 µm) grain fraction of quartz in age ranges >40 ka (Timar-Gabor et al. 2011, 2015b; Constantin et 54 55 26 al. 2012, 2014). In these studies, the coarse grain ages appear to provide the more reliable ages, and 56 57 27 the fine grain ages appear to be underestimated. 58 59 60 Page 3 of 41 Boreas 3 1 2 3 4 1 Furthermore, a range of studies in the recent past has shown that the thermal lifetime of the quartz 5 6 2 luminescence signal may be limited in some regions of the world (e.g. Lai & Fan 2013; Lowick & 7 8 3 Valla 2018). Next to saturation issues of the luminescence signal, this could be a further reason for age 9 10 4 underestimates when dating older (>100 ka) samples using quartz. However, the thermal stability of 11 12 5 the quartz luminescence signal seems to be regionally different. For example, the quartz fine grain 13 14 6 signal of Romanian loess samples has been shown to be thermally stable enough to theoretically 15 16 7 obtain ages up to 20 Ma (Timar-Gabor et al. 2017). However, the true maximum age limit is much 17 18 8 lower, due to the limited number of electron traps. 19 For Review Only 20 21 9 Feldspars IRSL signals saturate at much higher doses than quartz signals. Therefore, in order to 22 23 10 extend the age range when dating sedimentary archives, the past decade has thus seen massive 24 11 25 improvements in feldspar dating, through the application of elevated temperature IRSL measurements 26 12 (e.g.