The Polish loess-palaeosol sequences preserve a significant terrestrial record of the Quaternary climate change. In Poland, loess and loess-like formations occur in the southern part of the country, mostly in the south polish uplands,
These very interesting deposits were studied by polish researchers for the last 40 years and in the past a chronostratigraphic framework for polish loess deposits has been established through geomorphological, lithological and pedostratigraphical analysis (Jersak, 1973). On the other hand, another chronology was constructed by Maruszczak (1991); it was fully based on thermoluminescence (TL) age determinations and correlated with the ice core chronology proposed by Martinson
The aim of the current study is to establish a detailed OSL chronostratigraphy of the investigated profile to enable future comparison with those that will be available for other investigated loess profiles from Poland (Złota – Moska
Our chronostratigraphy is expected to fill a gap in the European loess research conducted in the Czech Republic, Germany and Ukraine (Fuchs
The Tyszowce loess section (λ = 23°42’45”E, φ = 50°36’30”N; Fig. 1) is located in the northern part of the Sokal Plateau-Ridge, which is a latitudinal cretaceous hump with thick (10–30 m) loess cover.
In the current study, we attempt to constrain the numerical chronology of a loess profile in Tyszowce (southeast Poland) using two different luminescence methods (OSL, post-IR IRSL) and two different types of material, namely silt-sized quartz fraction (45–63 μm) and polymineral fine grained material (4–11 μm). The profile in Tyszowce was also sampled in a continuous vertical section at close intervals of
The stratigraphy of the Tyszowce profile was presented more closely elsewhere (Maruszczak, 1974; Buraczyński and Wojtanowicz, 1975; Wojtanowicz and Buraczyński, 1978). The latest description of the section was published by Jary (2007).
The Tyszowce loess section represents Late Pleistocene loess. Similar to the paper of Jary and Ciszek (2013), the labelling system proposed by Kukla and An (1989), slightly modified by Markovic
The loess-palaeosol sequence in Tyszowce consists of five units developed in the Late Pleistocene and Holocene (Maruszczak, 1974; Buraczyński and Wojtanowicz, 1975; Wojtanowicz and Buraczyński, 1978): two polygenetic palaeosol complexes (
The formation of the
The thickness of the
The
The thickness of the
For this study, 21 samples from the 19 m thick loess profile at Tyszowce were collected. The sampling points represented the most characteristic profile sections. The procedures for sample collection for all necessary measurements can be found in Moska
The procedures include collection of samples for luminescence dating from a clean vertical section using thin-walled steel pipes from an Eijkelkamp system. At the same time, about 1 kg of loess was taken into plastic bags from around the tubes for gamma spectrometry.
In addition, six samples were collected for radiocarbon dating from the upper part of the profile. Four of them were collected from the L1L1 loess formation and two samples from the L1S1 soil formation. Each sample contained about 2 kg of raw loess or soil material. Samples for grain-size distribution, carbonate contents, organic carbon content and magnetic susceptibility were collected every 5 cm from the entire profile.
For OSL measurements, medium grains of quartz (45–63 μm) were extracted from the sediment samples using a routine treatment with 20% hydrochloric acid (HCl) and 20% hydrogen peroxide (H2O2). The quartz grains were separated using density separation with the application of sodium polytungstate solutions leaving grains of densities between 2.62 g/cm3 and 2.75 g/cm3. The grains were sieved before etching with concentrated hydrofluoric acid (HF, 40 min).
For post-IR IRSL measurements, polymineral fine grains (4–11 μm) were extracted from the sediment samples. The sediment was first treated with 20% hydrochloric acid (HCl) and 20% hydrogen peroxide (H2O2). Subsequently the sediment was rinsed in deionized water. After drying, the sediment was suspended in alcohol and grains of desired diameter range were extracted by sedimentation. First, larger grains were deposited by leaving the suspension in a 7 cm high column of alcohol for 11 min. The suspension was then transferred to another test tube where the desired range was obtained by leaving the suspension for further 11 min. This step was repeated 4 times. After that, the solution from above the material settled during 12 hours was decanted to obtain the required fraction. The clean polymineral grains were suspended in 50 ml of acetone and 2 ml of the suspension was pipetted into flat-bottom tubes with stainless steel discs placed in them. The grains settled on the disc surface and after evaporation of the acetone, discs with a monolayer of fine polymineral grains were obtained.
All OSL measurements were performed using an automated Daybreak 2200 TL/OSL reader (Bortolot, 2000). This reader uses blue diodes (470 ± 4 nm) delivering about 60 mW/cm2 at sample position and a 6 mm Hoya U-340 filter was used for the OSL detection. Laboratory irradiations were performed using a calibrated 90Sr/90Y beta source mounted onto the reader delivering a dose rate of
For the coarse and medium grains quartz fraction, equivalent doses were determined using the single-aliquot regenerative-dose (SAR) protocol (Murray and Wintle, 2000). The OSL SAR protocol which was used in our measurements contained the following steps listed in Table 1.
Steps used in protocols which were used for determined equivalent doses. For the medium grain quartz fraction the single-aliquot regenerative-dose (SAR) protocol (Murray and Wintle, 2000) was used. For the polymineral fine grains samples a post-IR IRSL protocol (Thiel et al., 2011) was used.Step OSL SAR protocol post-IR IRSL protocol 1 Irradiation with the regenerative beta dose Irradiation with the regenerative beta dose 2 Preheat at the temperature 260°C for 10 s Preheat at the temperature 320°C for 60 s 3 Blue light stimulation at the temperature 125°C for 100 s IR stimulation at the temperature 50°C for 200 s 4 Irradiation with the test dose IR stimulation at the temperature 290°C for 200 s 5 Cut-heat at the temperature 220°C Irradiation with the test dose 6 Blue light stimulation at the temperature 125°C for 100 s Preheat at the temperature 320°C for 60 s 7 IR stimulation at the temperature 50°C for 200 s 8 IR stimulation at the temperature 290°C for 200 s 9 IR stimulation at the temperature 325°C for 100 s
Intensities measured in steps 4 and 8 were used for equivalent dose determination. For equivalent dose calculation, the first second of the OSL decay curve was used and the background signal was estimated using its last 10 seconds. The ages were calculated using the Central Age Model (CAM) (Galbraith
Sample names in the current study, specific activities of natural radionuclides, dose rates, calculated ages for all investigated fractions using CAM model and laboratory codes.Depth (m) Sample name U-238 (Bq/kg) Th-232 (Bq/kg) K-40 (Bq/kg) Dose rate 45–63 μm (Gy/ka) De 45–63 μm (Gy) OSL age 45–63 μm (ka) Laboratory code Dose rate 4–11 μm (Gy/ka) De 4–11 μm (Gy) OSL age 4–11 μm (ka) Laboratory code 1.5 Tysz21 20.2±0.4 24.4±0.3 420±11 2.12±0.15 36.9±3.1 17.3±1.4 GdTL2430 2.67±0.15 49.3±1.5 18.4±1.2 GdTL2409 2.2 Tysz20 20.5±0.4 23.8±0.4 396±11 2.04±0.14 36.2±1.1 17.7±1.4 GdTL2431 2.58±0.15 47.6±1.6 18.4±1.2 GdTL2410 2.7 Tysz19 22.4±0.4 26.1±0.5 412±11 2.14±0.15 37.6±1.1 17.4±1.4 GdTL2432 2.73±0.15 52.1±1.6 19.0±1.1 GdTL2411 3.6 Tysz18 19.3±0.3 23.2±0.5 365±11 1.90±0.13 35.1±1.0 18.4±1.4 GdTL2433 2.41±0.20 53.9±1.6 22.2±1.4 GdTL2412 5.2 Tysz17 25.1±0.4 30.5±0.6 479±13 2.42±0.18 37.8±1.1 15.4±1.3 GdTL2434 3.10±0.19 55.9±1.7 18.0±1.1 GdTL2413 6.0 Tysz16 25.0±0.5 31.1±0.6 501±14 2.48±0.18 39.5±1.1 15.8±1.3 GdTL2435 3.17±0.18 58.8±1.7 18.5±1.1 GdTL2414 6.9 Tysz15 25.7±0.5 31.6±0.7 493±14 2.48±0.18 38.2±1.2 15.3±1.3 GdTL2436 3.18±0.18 58.3±1.8 18.3±1.1 GdTL2415 8.5 Tysz14 26.2±0.4 32.3±0.6 520±14 2.56±018 45.9±1.5 17.8±1.4 GdTL2437 3.28±0.19 59.0±1.8 18.0±1.1 GdTL2416 8.9 Tysz13 26.5±0.5 32.3±0.6 507±14 2.53±0.18 43.4±1.5 17.1±1.3 GdTL2438 3.25±0.19 60.9±2.0 18.7±1.2 GdTL2417 10.0 Tysz12 27.5±0.5 34.6±0.7 516±14 2.60±0.18 47.5±1.5 18.2±1.3 GdTL2439 3.36±0.19 65.2±2.0 19.3±1.2 GdTL2418 10.7 Tysz11 26.8±0.5 34.1±0.6 571±16 2.73±0.19 52.2±2.0 19.1±1.4 GdTL2440 3.49±0.20 72.7±2.1 20.0±1.2 GdTL2419 11.9 Tysz10 28.4±0.5 35.6±0.7 535±14 2.69±0.20 57.5±2.3 21.2±1.8 GdTL2441 3.47±0.20 71.9±2.0 20.7±1.3 GdTL2420 13.0 Tysz9 29.2±0.5 35.9±0.6 552±15 2.75±0.20 54.7±1.7 19.8±1.7 GdTL2442 3.54±0.20 84.1±2.2 23.7±1.5 GdTL2421 13.8 Tysz8 27.1±0.5 35.4±0.6 560±15 2.72±0.20 56.8±2.4 20.7±1.7 GdTL2443 3.49±0.20 83.1±2.7 23.7±1.5 GdTL2422 14.8 Tysz7 26.6±0.4 39.4±0.7 560±15 2.77±0.20 77.0±3.0 27.6±2.2 GdTL2444 3.57±0.20 92.3±2.5 25.8±1.5 GdTL2423 15.4 Tysz6 24.6±0.4 38.7±0.6 477±13 2.49±0.18 100.0±3.0 40.0±3.1 GdTL2445 3.25±0.19 150.5±4.0 46.1±2.9 GdTL2424 15.6 Tysz5 27.1±0.4 36.7±0.7 533±14 2.67±0.20 115.2±3.5 42.9±3.5 GdTL2446 3.46±0.19 192.5±4.5 55.5±3.2 GdTL2425 16.4 Tysz4 26.6±0.4 36.9±0.6 545±15 2.69±0.20 146.9±5.0 54.4±4.4 GdTL2447 3.46±0.18 196.3±3.3 56.7±3.3 GdTL2426 17.5 Tysz3 22.7±0.4 35.1±0.6 448±12 2.31±0.17 172.2±4.5 74.2±5.8 GdTL2448 3.00±0.18 236.3±5.0 78.5±4.9 GdTL2427 17.8 Tysz2 23.8±0.4 34.3±0.6 474±13 2.39±0.17 183.4±3.5 76.4±5.6 GdTL2449 3.09±0.18 250.0±6.0 80.7±4.9 GdTL2428 18.4 Tysz1 12.8±0.2 19.7±0.3 325±8 1.54±0.15 186.8±5.0 121.3±12.3 GdTL2450 1.93±0.15 365.0±8.0 188±15 GdTL2429
A preheat plateau test was performed for two quartz samples (Tysz_6 and Tysz_12) to establish the most appropriate preheat temperature. The preheat temperatures were varied from 200°C to 300°C in 20°C steps. No systematic variation in
Thiel
For the SAR protocol applied to quartz the dose response curves were built up to 400 Gy for the oldest samples and were fitted to a single saturating exponential function to determine Des. For those measurements three regenerative beta dose points were selected that bracketed the expected value of equivalent dose following test measurements on three aliquots were used (see supplementary material). Such a method of determining the equivalent dose was chosen due to the large number of samples. Such data can be fitted by a single exponential function, however the value of D0 increases with growing De values as recently pointed out by Timar-Gabor
For two oldest samples, post-IR IRSL protocol where applied for doses up to almost 2000 Gy and were fitted with a double saturating exponential function to determine De. Similar experiment was used to create full growth curves (up to 2000 Gy) for 2 oldest samples for both methods. Examples for full growth curve Tysz_1 are shown in Fig. 3, together with typical decay curves for each of the signals. Growth curves in this case were fitted with a double saturating exponential function to determine De.
In the laboratory, all OSL samples were dried. High-resolution gamma spectrometry using an HPGe detector manufactured by Canberra was carried out in order to determine the content of U, Th and K in the samples. The measurements were performed on 800 g samples placed in Marinelli beakers. The samples were stored for about 3 weeks to ensure secondary equilibrium between gaseous 222Rn and 226Ra in the 238U decay chain before measurements. The spectra were collected over a period of 24 hours. The activities of the isotopes present in the sediment were determined using IAEA standards RGU, RGTh, RGK after subtraction of the detector background. Dose rates were calculated using the conversion factors of Guerin
Radiocarbon dates were obtained for loess and soil samples. The search for best suitable radiocarbon materials took place at the site. Unfortunately, it was not possible to find charcoal or snail shells, so we decided to collect about 2 kg of material for each sample and try to extract humic acids in the laboratory. For all samples, 14C age determination of the humic acid fraction was possible. Humic acids from geological or archaeological samples are always assessed as a second-choice material for 14C dating. It is assumed that the 14C ages may be affected by the presence of humic acids originating from other (younger) organic material, e.g. from soil horizons located above a sample (Wild
14C ages of loess and soil. The age was calibrated using OxCal program v4.2, (Bronk Ramsey 2009; Southon et al., 2013).No. Lab. No. Sample name Age 14C (BP) Range of calendar (calibrated) age 68% confidence level Range of calendar (calibrated) age 95% confidence level 1 GdA-3133 TYSZ-1 25110 ± 115 29320BP (68.2%) 28974BP 29486BP (95.4%) 28831BP 2 GdA-3134 TYSZ-2 25420 ± 120 29674BP (68.2%) 29319BP 29888BP (95.4%) 29109BP 3 GdA-3135 TYSZ-3 17870 ± 70 21790BP (68.2%) 21546BP 21877BP (95.4%) 21416BP 4 GdA-3136 TYSZ-4 16880 ± 75 20485BP (68.2%) 20245BP 20577BP (95.4%) 20114BP 5 GdA-3137 TYSZ-5 16380 ± 75 19886BP (68.2%) 19638BP 20004BP (95.4%) 19554BP 6 GdA-3138 TYSZ-6 15710 ± 70 19031BP (68.2%) 18861BP 19150BP (95.4%) 18794BP 7 GdA-3139 TYSZ-7 9455 ± 40 10745BP (58.2%) 10652BP 11059BP (2.0%) 11034BP 10622BP (10.0%) 10601BP 10995BP (1.3%) 10977BP 10787BP (92.1%) 10575BP
Grain size distribution was determined using a laser diffractometer Mastersizer 2000 (manufactured by Malvern, England). Before the measurement organic matter was removed by H2O2 and, next, carbonates using a 10% HCl solution. For better dispersion, sodium hexametaphosphate (calgon) was added to the solution before measurement. Summary of the amount of grains of different fractions can be found in Fig. 2 where the percentages of the individual investigated fractions (1–4 μm, 8–16 μm, 31–63 μm, and larger than 63 μm) are shown. Investigations of aeolian particle dynamics have found that the coarse grain population, or silt fraction, is generally transported by surface winds in short suspension episodes (Pye, 1987). This coarse aeolian population accumulates to form thick deposits in adjacent downwind areas. Conversely, the fine grain population, or clay fraction, once off the ground, can be dispersed over a wide altitudinal band. It is mainly transported by upper level flow, and is deposited far from the source areas (Pye, 1987, 1995).
Recent studies of the magnetic properties of the loess palaeosol sequences have demonstrated the potential of magnetic susceptibility as a climatic proxy (Liu
Carbonates were analyzed using the Scheibler method and the organic substance using the Tiurin method (Tyurin, 1935).
The calcium carbonate content in the S0 soil (modern soil) is typical for well-developed soils,
In the investigated profile, the L1L1 cover reaches almost 14 meters indicating that in this area the climate conditions were favourable for loess accumulation. We can notice that the L1L1 cover in Złota, which is located about 200 km to the west, is only about 6–7 meters thick (Moska
Other radiocarbon ages range from 19 to 21 ka agreeing very well with the luminescence results. Because of the large thickness of this unit, 14 samples were collected for luminescence dating (Tysz_8 – Tysz_21), and 28 luminescence ages were obtained. It is very characteristic that according to luminescence results, the loess cover in the L1L1 unit was created during a 3–5 ka period, which means that the calculated loess sedimentation rate for this unit is more than 3 mm per year which seem to be very high compared to about 1 mm per year calculated for different European loess profiles (Frechen
The L1S1 soil unit is the most important fossil soil which has been identified in many locations where loess deposition occurred in Western and Eastern Europe. In Tyszowce, this complex of soil reaches 1 meter of thickness (Fig. 2). The magnetic susceptibility (MS) signal does not show the typical features of soils, it is weak with no characteristic peak in the signal. The upper part of the L1S1 unit is almost completely free from carbonates. The two 14C results were obtained also from humic acid extracts from the bulk material (Tysz-1 and Tysz-2). Both results (29.5 ± 0.2 and 29.1 ± 0.2 cal. ka BP) correlate very well with those obtained for the corresponding unit in the Złota section (Moska
Below the L1S1 pedocomplex, the lower loess unit was deposited probably during the Lower Pleniweichselian (Jary, 2007). The thickness of the L1L2 unit is only 100 cm which is much less than that of the L1L1 unit. The L1L2 unit is also more than 2 times thinner than in central Poland in the Złota profile (Moska
The calcium carbonate content in the S1 soil is typical for a well-developed soil,
In the bottom part of the profile, a clearly visible soil occurs. In the lowest and the oldest part of the investigated loess profile, a sequence of a fossil soil and a well-developed illuvial horizon was formed. The S1 unit is 1.5 m thick and in the central part of this unit we can observe the highest value of MS as a characteristics peak which is observed for all well-developed soils. The amount of CaCO3 is close to zero in the entire unit. Characteristic for the bottom part of this unit is an incredibly high content of sand grains exceeding 40%, also suggesting that this sand material originates from the neighbourhood. Six luminescence results obtained for this unit (samples Tysz_1, Tysz_2, Tysz_3) show that the ages increase with depth. The results obtained for quartz (Fig. 4) span the period between 75 ka and 123 ka which fits well with the typical period of S1 soil formation,
The loess-palaeosol sequence in Tyszowce is one of the most important sites for loess research in Poland as it is representative not only for the whole eastern part of the country but also demonstrates the importance of this site as a representative record of the Late Pleistocene climate and palaeoenvironment in this part of Europe. This is due to the presence of clearly separated five stratigraphic units with well-preserved features. The units characterize this section from the sedimentological and pedological point of view and demonstrate the continuity of their sedimentation, which is the foundation for building loess stratigraphy and its use for correlation with loess in the Ukraine and Western Europe. The MS variation displays a high degree of similarity to the enviromagnetic records observed in other European Late Pleistocene loess sections. The results obtained in this study are in concordance with grain size distribution and magnetic susceptibility variations agreeing very well with stratigraphic boundaries. The two employed luminescence dating methods yield mostly consistent results with differences visible only for the two soil units L1S1 and S1. During the loess accumulation periods, the results are in agreement within uncertainties. Similarly, consistent results were obtained for the other investigated loess profiles. Contrary to indications in literature we do not see the age underestimation in quartz at least until 70 ka (e.g. Wintle and Adamiec, 2017).
We can observe high consistency of luminescence results between Tyszowce and Biały Kościół (Moska
The 14C results correlate very well with the luminescence results only one 14C result seems to be underestimated (Tysz_7). In addition, the 14C results obtained for the L1S1 soil complex correlate very well with results from Nussloch (Germany, Antoine
The L1S1 soil unit is clearly visible and humic horizons of gley soil is dated here to about 40 ± 3 ka, which is in agreement with loess deposits in other locations in Europe and Poland.
The S1 pedocomplex is well developed and luminescence results for quartz are within the geological boundaries. The oldest sample Tysz_1 is a kind of mystery because of the significant change of the dose rate. Post-IR IRSL ages using the polymineral fine grain fraction are very similar to results obtained from quartz, only two post-IR IRSL results obtained for samples which were collected from soils (Tysz_5 and Tysz_1) are older than for quartz fraction and also older than expected ages from the geological point of view. The results presented in this study yield a robust chronology of the evolution of loess deposits in western Poland and, in conjunction with similar research carried out for other loess profiles in Poland, it will allow to present a comprehensive picture of their chronological evolution.