Impact of Supraglacial Lakes on Ice-Shelf Stability

位于南极洲浮冰架顶部的融水湖可能导致了过去二十年来南极洲冰川冰盖的巨大变化。例如，2002年，位于南极半岛东侧的面积为1，600平方公里的拉森B冰架破裂成数千座小冰山，随后由于在过去十年中在其表面形成了2，000多个融水湖而漂走。我们的研究项目解决了南极冰架上形成表面湖泊的原因，以及这些表面湖泊随后如何促成可能导致冰架破裂的力量，如拉森B。我们的项目主要集中在作出精确的全球定位系统（GPS）测量冰架弯曲的填充和排水的表面湖的麦克默多冰架。观测到的垂直位移（几十厘米量级）对湖泊填充的反应将用于校准和测试计算机模拟模型，这些模型将更普遍地预测冰架对地表湖泊的反应，并在未来各种气候条件下进行预测。我们的项目将每小时测量一次冰架的垂直运动（使用GPS测量仪器）以及靠近麦默多站机场的一个表面湖周围的温度和阳光条件（导致融化）。在这一初步的数据收集工作之后，将进行计算机模拟和其他更多的理论分析，以确定所选择的麦克默多冰架表面湖作为继续研究的实地实验室的适用性。最终，这项研究将有助于了解将气候变化与海平面上升联系起来的冰川学过程。研究的成功结果将使冰川学家能够更好地评估促进或削弱南极冰架在控制南极洲内部冰转移到海洋中的影响的过程。该项目将开展两项推广活动：（1）在网上张贴一份实地活动期刊;（2）为国际冰川学会建立一个开放的冰川学教学和推广网络共享网站，拟议的项目力求通过实验验证一个冰架不稳定性理论，该理论是为解释2002年拉森B冰架的爆炸性破裂而提出的。该理论认为，浮冰架上冰上湖的填充和排水在冰内引起足够的挠曲应力，以（a）引起起源于冰架底部/表面的向上/向下传播的裂缝，（B）将冰架切割成宽度往往小于冰厚度一半的碎片。狭窄宽度的重要性在于，它们促进了冰架碎片在破碎过程中的倾覆。这种倾覆释放了大量的重力势能（相当于拉森B冰架的数千千吨TNT），从而促进了拉森B事件的爆炸性。通过实验验证冰架破裂的表面湖泊机制的观测动机是基于这样一个事实，即在拉森B冰架破裂前的十年里，在拉森冰架上形成了2,000个表面湖泊，这些湖泊被观察到（通过卫星图像）在开始破裂的前一天以协调的方式排水。现场-该项目的观测部分将侧重于麦克默多冰架上的一个冰上湖，那里的夏季表面持续融化。该湖将在一个临时的现场季节进行研究，以确定周围的地表溪流和海岸线是否与重型建筑设备梳理将允许地表水人工鼓励填补湖泊。如果成功地鼓励开发，麦克默多冰架表面湖将允许测量关键的冰架弯曲和应力变量，而不必访问位于南极洲其他地方的冰架表面湖的后勤要求更高。 在为期6周的临时野外季节收集的数据包括：确定地表湖如何生长和填充的能量和水平衡参数，以及对冰盖垂直弯曲的各种全球定位系统测量，以响应地表湖内所含的不断变化的融水负荷。这些数据将用于（1）限制粘弹性弯曲和冰架可能断裂的计算机模型，以响应湖中融水负荷的增加，和（2）确定继续研究麦克默多冰架上的初期表面融水湖特征是否为限制更多-南极洲温暖地区其他冰架上的表面融水湖的一般行为。 受实地项目获得的观测数据约束的计算机模型将为冰架的能量和水平衡模型提供一般信息，并能够更准确地预测南极内陆冰周围不断变化的冰架状况。该项目将创建有史以来第一个基于地面的观测结果，有助于产生能够预测冰架粘弹性和断裂行为的模型，以应对冰上湖泊的演变，包括由于能量平衡效应引起的缓慢变化，以及由于填充和排水引起的快速变化。

Meltwater lakes that sit on top of Antarctica's floating ice shelves have likely contributed to the dramatic changes seen in Antarctica's glacial ice cover over the past two decades. In 2002, the 1,600-square-kilometer Larsen B Ice Shelf located on the Eastern side of the Antarctic Peninsula, for example, broke into thousands of small icebergs, which subsequently floated away as a result of the formation of more than 2,000 meltwater lakes on its surface over the prior decade. Our research project addresses the reasons why surface lakes form on Antarctic ice shelves and how these surface lakes subsequently contribute to the forces that may contribute to ice-shelf breakup like that of the Larsen B. Our project focuses primarily on making precise global positioning system (GPS) measurements of ice-shelf bending in response to the filling and draining of a surface lake on the McMurdo Ice Shelf. The observed vertical displacements (on the order of tens of centimeters) in response to lake filling will be used to calibrate and test computer simulation models that predict the response of ice shelves to surface lakes more generally and in a variety of future climate conditions. Our project will make hourly measurements of both vertical ice-shelf movements (using GPS surveying instruments) and of temperature and sunlight conditions (that drive melting) around a surface lake located close to the McMurdo Station airfield. Following this initial data-gathering effort, computer simulations and other more theoretical analysis will be undertaken to determine the suitability of the chosen McMurdo Ice Shelf surface lake as a field-laboratory for continued study. Ultimately, the research will contribute to understanding of the glaciological processes that link climate change to rising sea level. A successful outcome of the research will allow glaciologists to better assess the processes that promote or erode the influence Antarctic ice shelves have in controlling the transfer of ice from the interior of Antarctica into the ocean. The project will undertake two outreach activities: (1) web-posting of a field-activity journal and (2) establishing an open-access glaciological teaching and outreach web-sharing site for the International Glaciological Society.The proposed project seeks to experimentally verify a theory of ice-shelf instability proposed to explain the explosive break-up of Larsen B Ice Shelf in 2002. This theory holds that the filling and draining of supraglacial lakes on floating ice shelves induces sufficient flexure stress within the ice to (a) induce upward/downward propagating fractures originating at the base/surface of the ice shelf that (b) dissect the ice shelf into fragments that tend to have widths less than about half the ice thickness. The significance of narrow widths is that they promote capsize of the ice-shelf fragments during the break-up process. This capsize releases large amounts of gravitational potential energy (comparable to thousands of kilotons of TNT for the Larsen B Ice Shelf) thereby promoting explosiveness of the Larsen B event. The observational motivation for experimentally verifying the surface-lake mechanism for ice-shelf breakup is based on the fact that 2,000 surface lakes developed on the Larsen B Ice Shelf in the decade prior to its break up, and that these lakes were observed (via satellite imagery) to drain in a coordinated fashion during the day prior to the initiation of the break up.The field-observation component of the project will focus on a supraglacial lake on the McMurdo Ice Shelf where there is persistent summer season surface melting. The lake will be studied during a single provisional field season to determine whether grooming of surrounding surface streams and shorelines with heavy construction equipment will allow surface water to be manually encouraged to fill the lake. If successfully encouraged to develop, the McMurdo Ice Shelf surface lake will allow measurements of key ice-shelf flexure and stress variables needed to develop the theory of ice-shelf surface lakes without having to access the much more logistically demanding surface lakes of ice-shelves located elsewhere in Antarctica. Data to be gathered during the 6-week provisional field season include: energy- and water-balance parameters determining how the surface lake grows and fills, and various global positioning system measurements of the vertical bending of the ice sheet in response to the changing meltwater load contained within the surface lake. These data will be used to (1) constrain a computer model of viscoelastic flexure and possible fracture of the ice shelf in response to the increasing load of meltwater in the lake, and (2) determine whether continued study of the incipient surface-meltwater lake features on the McMurdo Ice Shelf provides a promising avenue for constraining the more-general behavior of surface meltwater lakes on other ice shelves located in warmer parts of Antarctica. Computer models constrained by the observational data obtained from the field project will inform energy- and water-balance models of ice shelves in general, and allow more accurate forecasts of changing ice-shelf conditions surrounding the inland ice of Antarctica. The project will create the first-ever ground-based observations useful for spawning the development of models capable of predicting viscoelastic and fracture behavior of ice shelves in response to supraglacial lake evolution, including slow changes due to energy balance effects, as well as fast changes due to filling and draining.