Understanding contemporary change in the West Antarctic ice sheet

最近对南极冰盖的卫星观测显示，在过去十年左右的时间里，南极冰盖发生了巨大的变化。可以看到两种主要的变化。第一次发生在阿蒙森海海岸附近，影响了该地区的几条冰流，如松岛和斯韦茨冰川。冰流是由大约40公里宽、数百公里长的快速流动的冰（通常为10米/年）组成的河流，它们被突然的剪切边缘与邻近的缓慢流动的冰（通常为10米/年）分开。在这些冰流中，冰似乎正以每年几米的速度变薄。另一种类型的变化是在简单海岸的内陆深处发现的，那里的一条冰流正在变厚，其他冰流则显示出横向迁移的迹象。其他证据（如埋藏的裂缝）表明，该地区冰流的流动非常不稳定，容易偶尔停止。南极洲的气温非常低，几乎没有表面融化，所以冰厚的变化很可能是由冰的水平流动的变化引起的，如果冰的水平流动减缓，冰就会变厚，如果冰的水平流动加速，冰就会变薄。研究人员认为，上面强调的两个观测结果中的第一个可能是由南极洲周围海水变暖引起的。这导致浮冰架底部的融化增加，因此浮冰架变薄并倾向于（通过浮力）漂浮更多。这反过来又减少了这些冰块在冰下地形的高峰和低谷上流动时所经历的摩擦力。净效应是冰架及其上游冰流加速，因此变薄。这种类型的过程已被视为当代气候变化的一个指标。在我们知道海洋变暖的原因之前（如果它确实存在的话），我们将无法将这种变薄归因于自然或人为原因。人们认为，单纯海岸冰流的奇怪行为并不是因为海洋的变化而发生的。这是因为该地区的海岸受到巨大的罗斯冰架的保护，该地区的水温非常低。对这一地区变化的观测可能反映了冰流的内部变化，并经常将其与“天气”进行类比。人们认为冰流本身就不稳定，容易出现涌动和停滞期，就像较小的山谷冰川一样。这种行为可能是由冰流下水流的变化引起的，水流的变化影响了冰蒸汽的流动，因为它润滑了冰底部的任何沉积物。因此，水流的变化会使冰流受到更大的摩擦而停滞不前。因此，观测到的这两种类型的变化都与冰流的动力学有关。在这个项目中，我们希望通过构建一个具有足够精细分辨率的冰盖数值模型来了解这种行为，以捕捉单个冰流和冰架的形状。这意味着我们需要开发一种方法，对整个冰盖进行粗网格计算，对单个冰流和冰架进行嵌套的细网格计算。为了捕捉上述行为，我们还必须开发新过程的模型，例如通过冰块的应力传递，底部的水流以及水与下面沉积物的柔软性之间的相互作用。我们还必须把对冰盖的卫星观测结合起来，以建立一个精确的目前冰盖流动的模型。一旦完成，该模型将用于评估海洋温度变化对冰盖的长期影响。它最终将提供一个工具，帮助我们预测南极洲对未来全球海平面的贡献。

Recent satellite observations of the Antarctic ice sheet show dramatic changes over the last decade or so. Two main types of change are seen. The first happens near the coast of the Amundsen Sea and affects several ice streams in the area, such as Pine Island and Thwaites Glaciers. Ice streams are rivers of fast-flowing (up to 1 km/yr) ice that are approximately 40 km wide and several hundred kilometres long, they are separated from the neighbouring slow-flowing (typically 10 m/yr) ice by abrupt shear margins. In these ice streams, the ice appears to be thinning at the rate of several metres per year. The other type of change is found deeper inland on the Siple Coast where one ice stream is thickenning and others show signs of lateral migration. Other evidence (such as buried crevasses) suggest that the flow of the ice streams in this area is very erratic and prone to the occasional shutdown. Air temperatures are so cold in Antarctica that there is very little surface melt and so changes in ice thickness are most likely caused by changes in the horizontal flow of ice, which can lead to thicker ice if the flow slows, or to thinning ice if it accelerates. Researchers believe that the first of the two observations highlighted above may be caused by warming ocean waters around Antarctica. This leads to increased melt from the underside of floating ice shelves, which therefore thin and tend (through buoyancy) to float more. This, in turn, reduces the amount of friction these ice masses experience as they flow over peaks and troughs in the subglacial topography. The net effect is that the ice shelves and their upstream ice streams accelerate and therefore thin. This type of process has been taken as an indicator of contemporary climate change. Until we know the cause of the oceanic warming (if it indeed exists), we will not be able to attribute this thinning to natural or anthropogenic causes. The strange behaviour of the ice streams along the Siple Coast is not thought to happen because of changes in the oceans. This is because the coast in this area is protected by the huge Ross ice shelf and water temperatures in the area are extremely cold. The observations of change in this area could be a reflection of the internal variability of ice flow and the analogy to 'weather' is often drawn. Ice streams are thought to be inherently unstable and prone to surges and periods of stagnation, like their smaller counterparts the valley glaciers. This behaviour may be caused by changes in the flow of water under the ice streams, which affects ice-steam flow because it lubricates any sediments at the base of the ice. Changes in water flow can therefore cause an ice stream to experience more friction and to stagnate. Both of the types of change that have been observed are therefore associated with the dynamics of ice streams. In this project, we want to understand this behaviour by constructing a numerical model of the ice sheet which has sufiiciently fine resolution to capture the the shapes of individual ice streams and ice shelves. This means that we will need to develop a method of doing calculations on a coarse grid for the whole of the ice sheet and on nested, finer grids for individual ice streams and shelves. In order to capture the behaviour described above, we will also have to develop models of new processes such as the transmission of stresses through an ice mass, the flow of water at its base and the interaction between this water and the softness of the underlying sediments. We will also have to integrate satellite observations of the ice sheet to produce an accurate model of its present-day flow. Once complete, the model will be used to assess the longer-term effects of changing ocean temperatures on the ice sheet. It will ultimately provide a tool to help us predict what Antarctica's contribution to future global sea level will be.