Erosion Beneath the Laurentide Ice Sheet, and its Role in Pleistocene Ice Age Dynamics

劳伦泰冰盖下的侵蚀及其在更新世冰期动力学中的作用冰下侵蚀和沉积动力学影响大型冰盖的大小、稳定性和气候敏感性。这些过程可能调节大规模的涌动行为，引发气候和海平面的快速变化（MacAyeal, 1993a,b），并可能决定第四纪冰期的周期性（Clark and Pollard, 1998）。我们直观地将北方大陆的冲刷景观与冰下侵蚀联系在一起，然而对更新世冰盖侵蚀的速率、时间和空间模式的估计是相互矛盾或模糊的。冰下侵蚀很难研究，因为：(i)所涉及的过程发生在大冰原之下。和所有受侵蚀的地貌一样，变化的记录随着地表的消失而不断地被抹去；虽然更新世冰川沉积物的厚沉积物，其中包含有关侵蚀条件的信息，在以前的冰原边缘保存下来，但它们被零星地保存下来，难以确定年代。我们建议使用渗透同位素10 Be和26 Al来研究Laurentide冰盖的侵蚀历史。首先，我们将使用“埋藏测年”技术来确定前冰盖南缘的冰川沉积序列（Granger et Al., 1997）。当沉积物暴露在地球表面附近的宇宙辐射中时，在沉积石英颗粒的晶体结构中产生了宇宙成因的10 Be和26 Al。一旦沉积物被掩埋，这些同位素就会以不同的速率衰变，它们的比值提供了埋藏时间的衡量标准。我们已经将这项技术用于与土石互层的冰川河流砂的试验测量中，为以前未定年的冰川沉积物提供了地层上一致的中更新世包层年龄。其次，我们将利用积聚在土壤中并且高度集中在深度风化地形表面附近的大气产生的10 Be来研究冰期前的风化层对Laurentide耕作的贡献。我们推测，在第四纪之前覆盖加拿大地盾的深层风化层一定含有大量的10铍，远远超过第四纪间冰期克拉通上可能积累的含量。通过测量更新世期间被Laurentide冰盖侵蚀和沉积的沉积物的10 Be含量，我们将跟踪冰前风化层的移除，并评估冰盖对硬床条件开始的响应。我们的初步测量表明，更新世中期的岩石含有大量的“流星”10 Be。威斯康辛州的浓度要低两个数量级。在更新世中期，劳伦泰德冰盖似乎在“开采”一层风化严重的表层物质，但这一来源现在已经基本上耗尽了。我们将把这些方法应用于明尼苏达州西南部的耕地和互层冰川河流沉积物序列，以及南达科他州和内布拉斯加州钻探岩心中年代确定的灰层所覆盖的早更新世耕地序列。这一组合将提供更新世期间被劳伦泰德冰盖侵蚀的物质样本。该油田位于明尼苏达州西南部，在露头和钻芯中至少有9个独立的采油单元。在该地区广泛的前期工作，并且易于获得地表和地下样本，使其成为本研究的理想选择。

EROSION BENEATH THE LAURENTIDE ICE SHEET, AND ITS ROLE INPLEISTOCENE ICE AGE DYNAMICSSubglacial erosion and sediment dynamics influence the size, stability, and climatic sensitivity of large ice sheets. These processes may regulate large-scale surging behavior, initiating rapid shifts in climate and sea level (MacAyeal, 1993a,b), and perhaps dictate the periodicity of the Quaternary ice ages (Clark and Pollard, 1998). We intuitivly associate the scoured landscapes of the northern continents with subglacial erosion, yet estimates of the rates, timing and spatial pattern of erosion by the Pleistocene ice sheets are conflicting or ambiguous. Subglacial erosion is difficult to study because: (i) The processes involved take place beneath large ice sheets. (ii) As in all eroding landscapes, the record of change is continually effaced as the surface is removed, and (iii) Although thick deposits of Pleistocene glacial sediment, which contain information about erosional conditions, survive around the margins of former ice sheets, they are patchily preserved and difficult to date.We propose to study the history of erosion by the Laurentide ice sheet, using thecosmogenic isotopes 10 Be and 26 Al. First, we will date glacial sedimentary sequences at the southern margin of the former ice sheet using the technique of "burial dating" (Granger et al., 1997). Cosmogenic 10 Be and 26 Al are produced within the crystal structure of sedimentary quartz grains while the sediment is exposed to cosmic radiation near the Earth's surface. Once the sediment is buried, these isotopes decay at different rates and their ratio provides a measure of the burial time. We have used this technique in trial measurements on glaciofluvial sands interbedded with till, providing stratigraphically consistent mid-Pleistocene bracketing ages on previously undated glacial deposits. Second, we will use atmospherically-produced 10 Be, which accumulates in soils and is highly concentrated near the surface of deeply weathered terrains, to investigate the contribution of pre-glacial regolith to Laurentide tills. We presume the deep regolith that covered the Canadian Shield prior to the Quaternary must have contained an enormous inventory of 10 Be, far exceeding what could have accumulated on the craton during Quaternary interglacial periods. By measuring the 10 Be content of sediments eroded and deposited by the Laurentide ice sheet throughout the Pleistocene, we will track the removal of pre-glacial regolith and evaluate the ice sheet's response to the onset of hard-bed conditions. Our initial measurements show that mid-Pleistocene tills contain enormous concentrations of 'meteoric' 10 Be. Concentrations in Wisconsin till are two orders of magnitude lower. It appears that the Laurentide ice sheet was 'mining' a cover of heavily-weathered surficial material in the middle Pleistocene, but that this source has now been largely exhausted.We will apply these methods to well-studied sequences of till and interbedded glaciofluvial sediments in southwestern Minnesota, and early Pleistocene tills bracketed by well-dated ash layers in drill cores from South Dakota and Nebraska. This combination will provide samples of material eroded by the Laurentide Ice Sheet throughout the Pleistocene. The field area in southwestern Minnesota provides access to at least nine separate till units in outcrop and drill core. Extensive prior work in the area, and easy access to both surface and subsurface samples make it ideal for this study.