ИСТИНА

A methodology of yedoma study was developed more than 30 years ago (Vasil’chuk, 1982); it includes detailed isotope horizontal and vertical sampling of large syngenetic ice wedges, and radiocarbon dating of surrounding sediments and of organic microinclusions directly in the wedges. The revision of the ratio of 18O and winter temperature has been done using new data obtained over the past 20 years, including new 2015 ice wedge research data of A. Maslakov at Kotelny Island and Chukotka and J.Vasil'chuk at Yamal Peninsula. A good correspondence of the new data and previously derived relations is provided. Archive of the yedoma sections aged from 40 to 12 ka significantly supplemented with data obtained during Russian-German almost 20 years of research (Meyer et al, 2002, 2015; Schirrmeister, 2002, 2009, 2013; Opel et al., 2011) and Russian expeditions (Streletskaya et al., 2009-2015, Oblogov, 2015) and detailed work of international team at Duvanny Yar (Murton at al., 2015), and other works of the last decade. Syngenetic ice wedges are a direct indicator of the existence of ‘cold’ permafrost (French, 2012), with mean annual ground temperatures not above −1 to −3°C (Vasil’chuk, 2013). Winter air temperature is the main factor controlling the permafrost ground temperature in northern Eurasia. The duration of winters is about 8–10 months in these regions, summer duration is only 2–4 months; hence, even a two-or threefold increase in total summer temperatures would not significantly influence annual temperatures. Large syngenetic ice wedges are widespread in the Russian permafrost zone. Syngenetic ice wedges accumulate in such a way that ice becomes vertically stratified. The most reliable information was obtained from uniformly thick ice-wedge systems, and from small buried ice wedges formed simultaneously. Although δ18O data from large and small wedges tend to be similar, the larger wedges provide a more complete palaeoclimatic record. The main advantage of sampling small ice wedges is that their age is more accurately known; they are similar to, or slightly less than, the age of the host sediments. However, small ice wedges are rare, often located at the same depth, sometimes isolated within tiered ice-wedge systems. As such, they may have an anomalous δ18O composition, reflecting the supply of stagnant bog water or water from other non-atmospheric sources (Vasil’chuk, 1992). The δ18O composition of ice-wedge ice is a function of the isotopic signatures of the contributing sources, the proportions contributed from each of them, and the isotopic changes (by mixing, evaporation, or fractionation) during either freezing or by diffusion after freezing. The parameter most strongly related to the δ18O composition of ice-wedge ice is winter air temperature. In order to establish this relationship, the data on winter air temperature and the δ18O composition have been compared with modern ice wedges <100 years old, each comprising 8–12 elementary ice veins, in different regions of the Eurasian permafrost zone (Vasil’chuk, 1990, 1992, 2006, 2013). The objectives is reconstruction of mean winter tempratures. Vertical sampling of ice wedges is favoured over horizontal sampling because with the latter it is impossible to establish the exact sequence of ice-wedge formation. Vertical sample spacing was typically 50–100 cm. This allows the oxygen isotope curves to be placed at a chronological scale. The reconstruction of mean winter and January palaeotemperatures is based on the methodology and relationship established by Vasil’chuk (1992), and revised according to new isotope data in 2015. δ18O values in modern syngenetic ice wedges (δ18Oiw) show a strong empirical relationship with winter air temperatures. These relationships are expressed in the following simplified regression equation (Vasil’chuk 1990, 1992): tmean January = 1.5 δ18Oiw (±3°C). where tmean January is mean January temperature of the period of modern ice-wedge formation during last 60–100 years; δ18O is oxygen isotope composition of ice-wedge ice formed during last 60–100 years As re-deposition of organic material is common in permafrost (Vasil’chuk, 2013), 14C dates should be carefully evaluated, especially those beyond the range of radiocarbon dating; these usually correspond to re-worked organic material in yedoma. Therefore, the youngest 14C date from the data set in a particular horizon is most likely the closest to the actual time of accumulation and freezing of the yedoma sediment. In arctic conditions wind-transported organic material may have an age that is clearly different from the time of formation of the wedge. Connecting the age of the surrounding sediment to the formation of an ice wedge is also far from straightforward. When dating is obtained from interpolation, the time window may not be so narrow (4 ka). As in Siberia today, the winter climatic conditions were stable for a long period 40-12 ka BP; therefore, the exact limits of the time window are not very significant for the accuracy of the palaeoreconstruction. Based on the principle of the choice of the youngest 14C date, we selected for this paper 7 horizons (time duration 4 ka) formed approximately during 40-36, 36-32, 32-28, 28-24, 24-20, 20-16 and 16-12 ka BP in every yedoma outcrops. Vasil’chuk Y. K. 1982: Regularities of development of engineering geology conditions in the north of Western Siberia in the Holocene. English summary (27 pp.) of PhD dissertation in Geology and Mineralogy, Moscow State University (in Russian). Nauka publ., Moscow (305 pp.). Vasil’chuk, Y. K. 1992: Oxygen isotope composition of ground ice application to paleogeocryological reconstructions). Volume 1, 420 pp., Volume 2, 264 pp. Theoretical Problems Department, Russian Academy of Sciences and Lomonosov’s Moscow University Publications, Moscow (in Russian). Vasil’chuk, Y. K. 2013: Syngenetic ice wedges: cyclical formation, radiocarbon age and stable-isotope records. Permafrost and Periglacial Processes 24, 82–93.