AL-26/BE-10 BURIAL DATING OF EARLY AND MIDDLE PLEISTOCENE GLACIAL/INTERGLACIAL SEQUENCES AROUND THE LAURENTIDE ICE SHEET

Although thick sequences of glacial and interglacial sediment are present around the margins of former ice sheets, they are difficult to correlate or relate to other records of Pleistocene climate change due to the lack of appropriate techniques for determining the age of glacial deposits older than 50,000 yr. In an effort to address this difficulty, we are using the technique of "burial dating" to determine the age of early and middle Pleistocene glacial deposits near the southern margin of the Laurentide Ice Sheet, in Minnesota, South Dakota, Nebraska, and Iowa, USA. This technique relies on the fact that the two cosmic-ray-produced nuclides Al-26 and Be-10 are produced at a known ratio (Al-26/Be-10=6.1) in sedimentary quartz grains undergoing surface exposure. Upon burial of the sediments, production ceases and the nuclides decay at different rates. Thus, if sedimentary deposits are buried by thick glacial sediments, the Al-26/Be-10 ratio reflects the time of burial and hence the age of the ice advance. Initially, we measured Al-26 and Be-10 in interglacial fluvial sediments recovered in drillcore in southern Minnesota and adjacent South Dakota. However, we conclude from these measurements and from measurements on modern fluvial sediments in glaciated regions that fluvial sediments may incorporate previously buried material whose Al-26/Be-10 ratio has not been reset to the production ratio, thus yielding ages that are too old. In order to avoid this problem, we have now made Al-26/Be-10 measurements on quartz recovered from interglacial paleosols buried by glacial tills. In addition, we have attempted to test the technique on sedimentary sequences whose age is constrained by one or more of the three "Pearlette-series" volcanic ashes that occur in the Pleistocene section in central North America. Results from modern and ancient paleosols indicate that well-developed paleosols appear likely to have the expected surface production ratio of Al-26 to Be-10 at the time of burial, and are well suited to this dating technique. We conclude that understanding the stratigraphic context of material to be dated is very important to obtaining correct numerical ages using this technique, and we report numerical ages for a number of advances of the Laurentide Ice Sheet that buried paleosols beneath thick till deposits between 0.75 and 1.5 Myr ago.