Lower to middle Weichselian pedogenesis and palaeoclimate in Central Europe using combined micromorphology and geochemistry: the loess-paleosol sequence of Alsheim (Mainz Basin, Germany)

Highlights

• Weichselian humus zones having spongy microstructure can be regarded as Chernozems.

• The degree of fineness is useful for differentiation of loess deposits in one region.

• MAT for Weichselian Interstadials is 8.9 °C for Bwb horizons and 9.6 °C for Ahb horizons.

• MAP is <500 mm for Weichselian ISs, MAP is 300–400 mm for periods of loess deposition.

Abstract

Lower to middle Weichselian loess, loess derivatives and buried soils of the loess-paleosol sequence Alsheim (Central Europe) were characterised by particle size distribution, geochemical and micromorphological data. High rates of sedimentation with alternating phases of relocation are the main cause for a much less differentiation into Middle and Upper Weichselian loess-paleosol units of the Alsheim loess-paleosol sequence compared to other loess-paleosol sequences (e.g. Nussloch near Heidelberg), whereas the Lower Weichselian has distinct phases of pedogenesis resulting in Ah, Bw and Btw horizons.

To distinguish between different loess deposits locally and intraregional, the degree of fineness is an easily applicable and suitable tool, though inappropriate for interregional comparisons. The chemical index of alteration (CIA) is low (<50 = no weathering) for loess deposits in the Alsheim loess-paleosol sequence, which is in contrast to the worldwide compiled loess samples with CIA values ranging from >53 to <70 (Gallet et al., 1998). The highest weathering was detectable for Btw horizons with CIA values >70.

A direct quantitative estimation of mean annual palaeotemperature and mean annual palaeoprecipitation can be provided by calculations based on geochemistry of soil horizons and sediments. The present mean annual precipitation (MAP) in the Mainz Basin is 789 mm. In contrast, palaeoprecipitation data suggest a 150 mm higher amount for the Last Interglacial (Btw horizon), a much lower amount of around 300–400 mm MAPP (periods of loess and sandy loess deposition) and a MAPP of <500 mm for Weichselian Interstadials (humus zones and Bw horizons). The calculated mean annual palaeotemperature (MAPT) for Interstadials with 8.9 °C for Bw horizons or with 9.6 °C for humus zones (or to 2 K lower, considering the relation of the present MAT of the Mainz Basin with the MAT of Germany) seems to be a good approximation of the MAPT for Lower and Middle Weichselian Interstadials. A MAPT of 8.7 °C (or 6.7 °C) for Stadials (loess and sandy loess samples) is higher than other temperature estimations for Weichselian Stadials in Europe.

Micromorphology shows compacted granular structure and moderately to strongly developed pedality as characteristic properties for aquatic loess, whereas channel microstructure with no pedality is typical for loess deposits. Spongy microstructure suggests a classification of the Lower Weichselian Mosbach Humus Zones as Chernozems. The Eemian paleosol (Btw horizon in Als III) has only weak clay illuviation, characteristic for drier regions in Europe.

Palaeoclimate and soil formation of the Last Glacial–Interglacial cycle calculated from geochemistry and micromorphological data are in good accordance with other proxy data in Central Europe. This suggests that paleoclimate reconstruction based on palaeopedological analyses could be successfully implemented in Europe. Such data may provide a useful alternative to other proxies for correlating European records.

Introduction

Loess-paleosol sequences are excellent terrestrial archives for Late Cenozoic climate change (e.g. Catt, 1991; Frechen et al., 2001; Kemp, 2001; Porter, 2001; Schatz et al., 2011). Whereas the broader application of luminescence dating techniques led to a deeper decipherment of loess-paleosol sequences (e.g. Roberts, 2008; Kadereit et al., 2010; Thiel et al., 2011), detailed micromorphological-sedimentological (e.g. Kemp et al., 1994; Xiao et al., 1995; Kühn et al., 2006a) and palaeopedological studies contributed to an improved understanding of past climate fingerprints in loess and corresponding environments (e.g. Kemp et al., 2006; Mason et al., 2008; Haesaerts et al., 2010; Marković et al., 2011; Zech et al., 2011).

Against the background of the fact that modern soils at the today's land surface have formed under present climate conditions, buried soils being in a similar developmental stage and having the same soil properties are most likely indicative to comparable climate conditions of the past (Catt, 1991). Nevertheless it has to be taken into account that not all paleosol properties necessarily reflect environmental conditions comparable with today (Kemp, 2001; Kühn et al., 2006b). If buried paleosols are not welded or accretionary they can be used as proxies for palaeoclimate conditions (e.g. Buggle et al., 2009; Sheldon and Tabor, 2009; Suchodoletz et al., 2009; Dreibrodt et al., 2010a; Pietsch and Kühn, 2012).

Many regions in loess areas have been settled since ancient times and are characterised by severe soil erosion caused by centuries of agriculture or viticulture (e.g. Lang, 1997; Kalis et al., 2003; Mäckel et al., 2003; Ruddiman et al., 2008; Fuller et al., 2011). As a result calcaric Regosols on slopes and Anthrosols or colluvial deposits on footslopes and in valley floors are widely spread (Leopold and Völkel, 2007; Dreibrodt et al., 2010b). These soils, however, are hardly climate-indicative, because they result predominantly from soil erosion and are therefore not suitable for a comparison with buried in situ paleosols. Without having the opportunity of a comparison of paleosols with modern soils within an area, only proxies – developed elsewhere – or directly taken from the palaeosols can yield information about the palaeoenvironment.

Aside from a precise characterisation of petrographic and elemental properties, geochemistry offers the possibility to get indicatory values on palaeoprecipitation and palaeotemperature from paleosol horizons (e.g. Sheldon and Tabor, 2009). Combined with micromorphology it provides essential information about weathering intensity, pedogenic processes, relocation processes and palaeoenvironmental conditions.

Micromorphological features of paleosols reflect soil forming process such as clay illuviation, calcification, decalcification, redoximorphosis or relocation, and hence reflect environmental conditions under which these soils formed (e.g. Bronger et al., 1994; Kemp, 1998; Fedoroff et al., 2010). Micromorphology gives additional information about the intensity of soil forming processes, i.e. about soil developmental stages and their succession (Pietsch and Kühn, 2009) and it helps to refine the pedostratigraphy of loess-paleosol sequences (Kühn et al., 2006a).

In this paper we focus on particle size distribution, geochemical and micromorphological data characterising particularly lower Weichselian loess and loess derivatives as well as weathering intensities of buried Bw, Btw and Ah (humus zones, Chernozem) soil horizons of a loess-paleosol sequence situated in Central Europe. By our knowledge, the use of geochemical data of distinct buried soil horizons in terms of palaeoenvironmental proxies is the first attempt to calculate palaeotemperature and palaeoprecipitation in Europe for the Late Cenozoic.