This special issue of Catena contains twelve papers originally presented at a symposium entitled Reconstruction and Climatic Implications of Quaternary Paleosols and Paleosol Sequences, which was held at the XIV International Congress of INQUA (International Union for Quaternary Research) in Berlin 1995, and was organized by INQUA Commission VI (Paleopedology). The purpose of the symposium was review methods for recognizing Quaternary paleosols and for reconstructing paleoclimates from their properties.

Paleosols are soils formed in past geological periods, which (a) have been preserved by burial beneath younger sediments, or (b) have remained at the land surface but contain relict features indicating different conditions of climate of vegetation from those of the present. Because well-developed soils take fairly long periods of time to form and require stable land surface conditions, implying little or deposition during development, buried paleosols in sequences of Quaternary sediments provide evidence of major breaks in deposition as of environmental conditions during those breaks. However, recognizing paleosols which may have been truncated by erosion before burial of modified by diagenetic processes after burial is often problematic. Also even when they have been identified, problems arise in utilizing pedological features for reliable paleoclimatic interpretation. All the papers in this issue address problems of paleosol recognition and interpretation, and most demonstrate how paleosol studies can lead to detailed reconstructions of Quaternary climatic change at local, regional of international scales.

The most useful paleosols are those forming time-transgressive chronosequences without historical overlap (Vreeken,1975). These are stacked sequences of paleosols separated by intervening unaltered sediments, which together represent segments of Quaternary history encompassing several episodes of climatic change. In particular, sequences of paleosols formed in the thick loess deposits in various parts of the world provide paleoclimatic records for much or all of the Quaternary period that are at least as detailed as those obtained from deep ocean sediments. The two papers by Bronger et al. demonstrate this using mineralogical (thin section) analyses of closely spaced samples through two sequences exposed in Tadjikistan.

Paleosol sequences are now also recognized in volcanic, alluvial and slope deposits. Examples are described in the papers by Watanabe et al., Yamskikh and Kleber and Gusev, respectively. In the volcanic ash sequences of Japan. Watanabe et al. use analyses of organic carbon and ratios of elements subject to mobilization by weathering to identify buried soils and reconstruct local temperature and vegetation changes during the Holocene. In alluvial soil sequences on a low terrace of the River Yenisei in central Siberia, Yamskikh has found that Holocene environmental changes are best recorded in particular components of the organic matter in buried soils. Using mineralogical and micromorphological analyses of the Moshaysk district of Russia, Kleber and Gusev recognize regional sequences of Quaternary slope deposits, which together with intervening episodes of soil development also provide a history of changing environmental conditions.

Paleosols are less frequently preserved in glacial sequences, because they were often removed by repeated glacial erosion. However, Olsen's paper on Norway shows that truncated paleosols may still be preserved locally in strongly glaciated regions, and are very useful for dating and correlated the glacial deposits. He emphasizes the value of magnetic susceptibility measurements for identifying paleosols in this environment. Stremme also considers buried interglacial soils developed in tills, and shows that in lowland Europe they can be correlated with soils developed loess sequences and thus establish a clear succession of interglacials for the Middle and Late Pleistocene throughout the continent.

Another useful type of paleosol sequence is the post-incisive chronosequence (Vreeken,1975),which consists of surface soils formed on a succession of river terraces or other deposits of similar lithology but different ages. Vidic compares a chronosequence formed on a succession of glacial outwash terraces in the Ljubljana Basin, Slovenia with others formed on river terraces in parts of the USA. By calculating soil development indices she shows that rates of profile development decrease with time, usually as logarithmic function, and are influenced mainly by rainfall. However, the indices are an imprecise basis for dating and correlating the deposits of the Ljubljana Basin, principally because of the large variability of soil properties on each surface. From micromorphological studies of two post-incisive chronosequences in Libya, Cremaschi and Trombino show that in the Late Tertiary, Middle Pleistocene and Early to Middle Holocene there were warm humid periods in region that is currently hyperarid.

Individual features of paleosols often provide a clear indication of past climatic conditions. The most common are the wide range of frost effects in soils, the genesis and environmental significance of which are reviewed in detail by Van Vliet-Lanoe using laboratory studies as well as field observations in arctic areas. One particular feature of cryoarid soils in northeast Asia, stratified carbonate pendants found on the underside of stones, are used by Pustovoytov to reconstruct changes in rainfall during the Holocene.

The exact dating of soil development periods is often very difficult. Ideally the length of period should be defined be dates for the beginning of soil development and (for buried soils) its termination by burial. Radiocarbon assay is the commonest dating method used for soils formed in the Holocene and later part of the last major cold stage, but when applied to soil humus it gives only an approximate date sometime during the soil development period. This is because the turnover of organic matter in soils (incorporation of litter followed by oxidation to carbon dioxide which then returns to the atmosphere) occurs at various rates according to climate and other factors. In many parts of eastern Europe Chernozems were formed under semi-arid steppe grassland in the Early Holocene, but have since been changed to Luvisols and Podzoluvisols formed in more humid woodland conditions. In parts this region Alexandrovskiy and Chichagova have used the ages of burial mounds built in various archeological periods of the Holocene to date this change in soils preserved beneath the mounds .Comparison of radiocarbon dates for the buried soils and for surface soils close to the mounds then indicates changes in rates of humus turnover as a result of the Mid-Holocene climatic change.

The breadth of subjects covered by these papers is typical of modern paleopedology. Using an increasing repertoire of field and laboratory techniques, paleosols are now being recognized ever more widely in situations where they were previously unsuspected. This has inevitably led to demands from Quaternary scientists for more precise stratigraphic, geomorphic, paleoclimatic and archeological interpretations, and as paleopedologists attempt to meet these demands a continued expansion and broadening of paleosols studies seems likely.

Reference

Vreeken,W.J. 1975. Principal kinds of chronosequence and their significance in soil history. J.Soil Sci. 26. 378-394. VOLUME 34 NOS. 1-2