Search

The timing of past glaciation across the Tian Shan provides a proxy for past climate change in this critical area. Correlating glacial stages across the region is difficult but cosmogenic exposure ages have considerable potential. A drawback is the large observed scatter in 10Be surface exposure data. To quantify the robustness of the dating, we compile, recalculate, and perform statistical analyses on sets of 10Be surface exposure ages from 25 moraines, consisting of 114 new and previously published ages. We assess boulder age scatter by dividing boulder groups into quality classes and rejecting boulder groups of poor quality. This allows us to distinguish and correlate robustly dated glacier limits, resulting in a more conservative chronology than advanced in previous publications. Our analysis shows that only one regional glacial stage can be reliably correlated across the Tian Shan, with glacier expansions occurring between 15 and 28 ka during marine oxygen isotope stage (MIS) 2. However, there are examples of older more extensive indicators of glacial stages between MIS 3 and MIS 6. Paleoglacier extent during MIS 2 was mainly restricted to valley glaciation. Local deviations occur: in the central Kyrgyz Tian Shan paleoglaciers were more extensive and we propose that the topographic context explains this pattern. Correlation between glacial stages prior to late MIS 2 is less reliable, because of the low number of samples and/or the poor resolution of the dating. With the current resolution and spatial coverage of robustly-dated glacier limits we advise that paleoclimatic implications for the Tian Shan glacial chronology beyond MIS 2 are speculative and that continued work toward robust glacial chronologies is needed to resolve questions regarding drivers of past glaciation in the Tian Shan and Central Asia.

Constraining the timing of past ice-sheet change is important for assessing the cryospheric expression of climate change and improving our understanding of ice sheet dynamics. Geochronology used to construct past ice-sheet reconstructions, however, can be ineffective in polar environments where ice sheets were polythermal and left varying imprints on landscapes. Cosmogenic-nuclide exposure dating, for example, is especially hampered by the lack of ice-sheet erosion and resultant cosmogenic nuclide inheritance. Here, we apply in situ cosmogenic 10Be, 14C and 26Al methods to decipher various elements of the Laurentide Ice Sheet history of north-central Baffin Island. A clearly defined erosion boundary across the landscape reveals the transition in basal ice-sheet conditions as ice flow became channelized into northern Baffin Island fiords. 10Be and 26Al concentrations indicate that the boundary represents a juxtaposition of sliding, erosive ice and cold-bedded ice that preserved ancient bedrock that has not been significantly impacted by the ice sheet in perhaps one to two million years.We combine 10Be measurements from ice-sculpted bedrock with measurements of in situ14C, which has no inheritance due to its quick decay during ice-sheet cover, to determine the local timing of deglaciation. The average 10Be and in situ14C ages for upland deglaciation in north-central Baffin Island are 7.7 ± 0.9 and 8.4 ± 1.4 ka, respectively. Finally, in situ14C measurements from surfaces being uncovered by present-day retreat of small ice caps mantling uplands within the study area have concentrations too low to be compatible with continuous post-glacial exposure. These samples require shielding by ice for a significant portion of the Holocene, and more burial than during the Little Ice Age alone. Simple exposure-burial modeling suggests that 2400–2900 yr of total ice cover during Neoglaciation is required to explain measured in situ14C inventories. Combined, multiple cosmogenic nucli