THE MIS 2 ICE WEDGE PROBLEM IN UNGLACIATED YUKON AND THE PROBLEM OF YEDOMA DEFINITIONS

The concept of Yedoma is increasingly used to correlate the extensive fine-grained syngenetic sediments that accumulated during Pleistocene cold stages across Beringia. The Russian concept of Yedoma, as presented by Andrei Sher in the 1990s, identifies at least three uses: geomorphic (Yedoma surface), stratigraphic (Yedoma suite) and sedimentologic, referring to the ice-rich silts with large penetrating ice wedges. The latter has largely become the de facto definition for Yedoma, characterizing extensive fine-grained sediments that includes large syngenetic ice wedges that accumulated during Pleistocene cold stages. But this definition, requiring ice wedges to be present, remains problematic as we apply it to the North American context, and in particular to sites of greater relief where sedimentation is affected by hillslope and aspect effects. Late Pleistocene Yedoma deposits in the Klondike area (Yukon) show varying stratigraphic expression because of changes in surface processes coupled to past climate. During the MIS 3 interstadial, large, isotopically depleted (ca. -26 to -28 per mil d18O) syngenetic ice wedges are present within loessal sediments, while the contemporaneous pore ice is typically 2-4 per mil more enriched. The contrast between wedge and pore ice would be expected since ice wedges preferentially sample winter or spring snow melt following thermal contraction cracking in winter, while pore ice is a late season phenomenon in which waters at the base of the active layer are incorporated into the top of permafrost by freeze-back in the fall. However, during the height of the last cold stage, ice wedges are rare in the Klondike record, suggesting a lack of thermal contraction cracking. However, rare, narrow ice wedges have been found dating to MIS 2 and the corresponding isotope data from ice wedges are depleted (ca. -30 to -33 per mil d18O), and these values are similar to contemporaneous pore ice. This creates a paradox— how can pore ice be isotopically similar to wedge ice when they have different seasonalities? We now think that these differing records have underestimated the importance of aspect, and the role of perennial snow banks and ice patches that likely covered these surfaces in proximal slope positions. The presence of snow cover would reduce or inhibit thermal contraction cracking, while the release of that melt through the summer and into the fall, provides a mechanism for the depleted pore ice.