Lessons from the AMS 14C and OSL/IRSL-dating of the Dunaszekcső loess record, Hungary

Újvári, Gábor and Molnár, Mihály and Novothny, Á. and Kovács, J. (2014) Lessons from the AMS 14C and OSL/IRSL-dating of the Dunaszekcső loess record, Hungary. In: Loessfest'14 and 7th Loess Seminar, 2014.09.07-08., Wroclaw, Lengyelország.

Preview

Text abstract_ujvari_etal_LOESSFEST'14.pdf Download (205kB) | Preview

Abstract

Reliable chronologies are prerequisites of appropriate proxy interpretations from terrestrial archives of Quaternary climate and environmental change. Loess records may provide a wealth of paleoenvironmental information, yet they are usually poorly dated. This mostly means low resolution dating of loess profiles and also imprecise chronologies, i.e. age-depth models that have uncertainties of millennial magnitude. This prevents us from addressing issues like synchroneity of abrupt climatic/environmental events on millennial time scales. Two different means of dating are commonly applied for loess sequences: luminescence and radiocarbon dating. Major problems are low precision of luminescence ages and the general lack of organic macrofossils (e.g. charcoal) in loess that can reliably be dated using 14C. Other datable phases in loess are mollusc shells, rhizoliths and organic matter. Evidences are growing that rhizoliths are unreliable phases for 14C-dating and organic matter 14C ages are often seriously compromised by rejuvenation in loess sequences. Also mollusc shells are often regarded as unreliable material for 14C-dating, as they may incorporate 14C-deficient (or dead) carbon from the local carbonate-rich substrate during shell formation, thereby producing anomalously old ages by up to 3000 years. In this study an attempt has been made to address some of the dating issues and problems mentioned above by triple-dating (AMS 14C and OSL/IRSL) of the Dunaszekcső loess-paleosol sequence (South-Hungary). While the OSL/IRSL techniques directly date the sediment (quartz and K-feldspar grains) and provide burial ages, radiocarbon yield ages from phases like organic matter, mollusc shells and rhizoliths and determines the time elapsed since the living system was last in equilibrium with atmospheric 14C and became closed after burial. As revealed in this study all loess rhizoliths sampled at three different depths (4.00 m: 9744-10156 2σ age range in cal yr BP, 5.00 m: 8013-8167 cal yr BP and 6.00 m: 9534-9686 cal yr BP) yield Holocene ages, so absolute ages cannot be gained this way for loess deposition. As charcoals are widely accepted as phases yielding very reliable 14C ages, mollusc shell 14C ages were tested against charcoal ages. Here we focused on molluscs with smaller (< 10 mm) shells as some evidence exists that some species do not incorporate dead carbon into their shells or at least in low amounts. Our results demonstrate that Succinella oblonga and Vitrea crystallina yield statistically indistinguishable ages (2σ age ranges: 29990-30830 and 29600-30530 cal yr BP) when compared with the charcoal 14C age (29960-30780 cal yr BP, depth 8.20 m), and others like Clausilia sp. and Chondrula tridens give slightly older ages than the charcoals and show larger age anomalies (500-900 14C yr). Compared to the charcoal ages at 8.20-8.25 m depth, the post-IR IRSL225 age of 28520±1120 yr (2σ age range: 26280-30760 yr) from a depth of 7.75 m match quite well the charcoal ages (Dsz-Ch1, 2σ: 29960-30780 cal yr BP and Dsz-Ch2, 2σ: 29350-30150 cal yr BP). At the same time, the post-IR OSL approach seems to slightly underestimate (2σ: 20640-26960 yr), while the post-IR IRSL290 overestimate (2σ: 30260-37100 yr) the expected/true age of deposition at the respective depth (7.75 m). At a depth of 4.00 m, slight underestimation of mollusc AMS 14C ages (Trochulus hispidus, 2σ: 22370-22740 cal yr BP, Arianta arbustorum, 2σ: 24470-25120 cal yr BP) by post-IR OSL (2σ: 17140-21980 yr) and a moderate to significant overestimation by OSL (2σ: 26760-33800 yr) and post-IR IRSL290 (2σ: 27660-35740 yr) has been recognized. Again, the post-IR IRSL225 age (2σ: 23180-26900 yr) lies the closest to the AMS 14C ages. To decide which technique, AMS 14C or OSL/IRSL yields more accurate ages is not possible without independent absolute chronological data based on another method. Yet, we think that the consistent 14C ages of charcoals and small molluscs (two phases having very different origin and genesis) suggest that these ages are reliable and may reflect the real age of sedimentation. Clearly, the precision of 14C ages are an order of magnitude better (calibrated 2σ age ranges 500-800 yr) than the luminescence ages (2σ age ranges: 3700-7900 yr) and this may be another reason for creating age-depth models based purely on 14C ages, if high precision is needed. The use of a mixture of ages (i.e. 14C and OSL/IRSL) seems to counterproductive in this respect and we suggest to separate the results of the two techniques in modeling. OSL/IRSL-based age models are useful in checking the accuracy of 14C-based chronologies for the last 50 ka and vice versa and proxy interpretations should be tested against both 14C and OSL/IRSL age models independently.

Actions (login required)

Edit Item