LOESSFEST'09 | Aug. 31st – Sept. 3rd, 2009 |Novi Sad-Serbia

Insights into the Mid- & Late Pleistocene Paleoenvironment of the Ukraine – The Loess-Paleosol Sequence Stary Kaydaky

Buggle, B.1, Gerasimenko, N.2, Hambach, U.3, Glaser, B.1, Zöller, L.3

1Soil Physics Department, University of Bayreuth, D-95440 Bayreuth, Germany

2Earth Science and Geomorphology Department, Тarasa Shevchenko National University of Kyiv, Ukraine

3Chair of Geomorphology, University of Bayreuth, D-95440 Bayreuth, Germany

Introduction

Quaternary scientists consider two different stratigraphic schemes for Ukrainian loess-paleosol sequences. Key stone in this discussion is the chronostratigraphic position of the Kaydaky pedocomplex (MIS 5e vs. MIS 7). The objective of our study was firstly to clarify the chronostratigraphy of this site by a rock magnetic approach. Secondly we applied geochemical weathering indices and a soil color index for hematite formation (rubification) to achieve a (semi-quantitative) characterization of pedogenic processes. Together with palynological studies on characteristic horizons, we intended to gain further insight into the Mid- and Late Pleistocene paleoenvironment of the Ukraine.

Results/Conclusions

The findings of our chronostratigraphic work (Buggle et al., 2009, doi:10.1016/j. quaint.2008.07.013) support that the Kaydaky pedocomplex is a formation of MIS 5e. We could identify pattern in the magnetic susceptibility record, which characteristically correlates to records from the Vyazivok site (Ukraine)” having a well established chronology and to a marine δ18O proxy of the global ice volume. Based on this chronological placement of the Kaydaky unit, we were able to suggest a chronostratigraphy for the upper and lower part of the Stary Kaydaky section. The oldest studied pedocomplex is addressed to MIS 13-15.

In most of the Mid-Pleistocene parts of the section, the weathering record strongly suffers from multiple pedogenic overprint since loess layers were affected by erosion and are very thin and decalcified. However, the pollen composition of the stratigraphic units as well as the rubification record still provides valuable paleoenvironmental information. The rubification index is generally higher in the lowermost pedomembers (fossil forest or forest-steppe soils) of each interglacial pedocomplex, indicating higher hematite content. Lower values of the rubification index are recorded for the younger pedocomplexes (fossil steppe soils) of each interglacial, reflecting increasing humus accumulation. A similar pattern was revealed from the pollen records, showing highest percentages of pollen from arboreal and broad-leaved species in the lower parts of interglacial pedocomplexes and highest frequencies of xerophytes in the upper horizons (and in the loess). These data give evidence that the warmest and probably also most humid phases occurred at the first half of warm stage of pedogenesis . This climate optimum phase is followed by a distinct shift to probably cooler and dryer more steppe-like climate conditions in the later phase of warm stages. Unfortunately, a hiatus prevents the identification of similar patterns from the MIS 11 pedocomplex.

While the climate deterioration in Eurasia during the last warm stage (MIS 5e, vs. MIS 5a and c) is a widely studied phenomenon, suitable terrestrial archives for studying such trends also in Mid-Pleistocene warm stages are less common.

Concluding from the paleosol development of the last major five warm stages, the paleoenvironmental archive of Stary Kaydaky suggests that the partition of these warm stages in a climate optimum phase at the beginning followed by a second period with probably prevailing cooler and more arid (steppe-like) conditions was most strongly expressed in MIS 9. Also, pollen data suggest that wettest climate conditions occurred in the lower paleosol of the MIS 9 pedocomplex.

Regarding the major periods of loess formation (MIS 2. MIS 6, MIS 8, MIS 10, MIS 12), the pollen spectra indicates a decrease of arboreal vegetation. Only pollen of pine, alder, birch and rarely willow are represented indicating cool environments. Yet, loess forming conditions were different in the older (MIS 8, MIS 10, MIS 12) and younger loess units (MIS 2, MIS 6). In the older units steppe bushes (Rhamnus, Crataegus) still occurred and willow and alder were spread, probably growing in nearby gullies. In younger loesses arboreal vegetation almost disappeared. However arcto-boreal birch species could be found indicating cold climate. Within the herbal association, the role of xerophytes (Chenopodiaceae and Artemisia) strongly increased in the younger loesses indicating dry steppic conditions.

We cannot exclude that similar climatic conditions are simply not preserved in the older loess units due to erosion. Yet, the trend of increasing aridity during the Midpleistocene cold periods, which is revealed from the Stary Kaydaky section, is in agreement to findings from other paleoenvironmental archives of the region.

Corresponding author: Björn Buggle | Buggle@uni-bayreuth.de