Constraining Antarctica's contribution to sea-level change: development of a new glacial isostatic adjustment model for Antarctica

南极冰盖目前正在融化，并经历着快速的动态变化。然而，它融化的速度，以及它对海平面上升的影响，目前还知之甚少。我的研究金的目的是更好地估计南极洲目前的冰量变化。这样的估计可以从卫星重力数据中得出，卫星重力数据用于从地球重力场的测量中推断冰质量的变化。然而，这些数据必须根据冰川后反弹（也称为冰川均衡调整；GIA）的持续影响进行校正。卫星重力测量不能区分冰质量的变化和固体地球的质量运动。在最后一次冰期期间，南极冰盖要大得多，额外的质量导致冰下的陆地下沉。随着冰盖缩小到现在的大小，陆地开始反弹。由于地幔的粘性，这一过程一直持续到今天。事实上，由卫星重力数据记录的GIA信号可能与由于当前冰质量变化而产生的信号具有相同的量级。因此，为了确定冰质量变化的模式，尽可能准确地约束GIA模式是至关重要的。我将解决目前用于模拟南极洲GIA的方法的两个基本问题，从而大大减少南极洲海平面贡献的不确定性。在我的研究项目的第一部分，我将模拟南极洲下面的三维地球结构。目前的南极GIA模型没有考虑到地壳厚度或地幔粘度的变化。然而，这种变化之所以如此之大，是因为东南极洲和西南极洲有着非常不同的地质历史：东南极洲下面的地壳古老而“坚硬”，因此在载荷作用下不易变形。相比之下，南极洲西部的地壳年轻而“脆弱”，因此在载荷下更容易变形。为了真实地确定回弹信号，必须考虑地球结构的空间变化，以便计算出正确的荷载响应；不这样做可以改变反弹信号高达30%。最近获得的有关南极地球结构的新数据，以及GIA模型的发展，使我的这方面的建议非常及时。我要讲的第二个问题与海平面变化有关。西南极洲是一个以海洋为基础的冰盖，这意味着它的床低于目前的海平面。为了模拟冰面积随时间的变化，从而确定GIA信号，有必要考虑海平面变化和冰动力之间的反馈。海平面直接控制着冰盖开始漂浮的位置，从而控制着冰流入海洋的位置。然而，由于GIA，海平面并不是随时间均匀变化的：在冰盖增长的地区，陆地被额外的负荷压下，海面由于冰盖引力的增加而向上拉，导致海平面上升。当冰退缩时，情况正好相反。在这两种情况下，反馈都会影响冰盖的动力学行为。目前的冰盖模型没有考虑到这种反馈，尽管它们有可能增加冰盖的稳定性：在未来海平面上升的情况下，它们的加入可能从根本上改变对冰盖行为的预测。因此，在我的项目的第二个组成部分中，我将开发一个耦合的全球地理信息系统-冰盖模型。这两个研究领域都将受益于改进的模型能力。这些模型的改进将用于得出一个更现实的南极洲GIA估计。该模型将进行校准，以适应与以前的冰面积、过去的海平面和目前的回弹有关的野外约束。然后，它将被用来校正卫星数据，以适应GIA的影响，从而确定一个更准确的南极冰质量变化图。

The Antarctic Ice Sheet is currently melting and undergoing rapid dynamic change. However, the rate at which it is melting, and hence its contribution to sea-level rise, is currently poorly known. The aim of my Fellowship is to derive a better estimate of current ice-mass change across Antarctica. Such an estimate can be derived from satellite gravity data, which are used to infer ice-mass changes from measurements of the Earth's gravity field. However, these data must be corrected for the ongoing effects of postglacial rebound (also known as glacial isostatic adjustment; GIA).Satellite gravity measurements cannot distinguish between changes in ice mass and mass movement of the solid Earth. During the last glaciation the Antarctic Ice Sheet was much larger and the extra mass caused the land beneath the ice to subside. As the ice sheet shrank towards its present size, the land began to rebound. This process - GIA - continues today due to the viscous nature of Earth's mantle. In fact, the GIA signal, as recorded by satellite gravity data, can be of the same magnitude as the signal due to current ice-mass change. Therefore, it is vital to constrain the pattern of GIA as accurately as possible in order to determine the pattern of ice-mass change.I will address two fundamental problems with the methods currently used to model GIA in Antarctica, thus significantly reducing the uncertainty on Antarctica's sea-level contribution.In the first part of my Fellowship I will model the 3D Earth structure beneath Antarctica. Current Antarctic GIA models do not account for variations in crustal thickness or mantle viscosity. However, such variations are known to be large because East and West Antarctica have very different geological histories: the crust beneath East Antarctica is old and 'stiff' and thus will not deform easily under loading. In contrast, the crust beneath West Antarctica is young and 'weak' and so deforms more easily under loading. In order to realistically determine the rebound signal, spatial variations in Earth structure must be accounted for so that the correct response to loading is calculated; not doing so can alter the rebound signal by up to 30%. The recent acquisition of new data relating to Antarctic Earth structure, and GIA model developments, make this aspect of my proposal very timely.The second problem I will address is related to sea-level change. West Antarctica is a marine-based ice sheet meaning that its bed is grounded below present sea level. In order to model changes in ice extent over time, and hence determine the GIA signal, it is necessary to consider feedbacks between sea-level change and ice dynamics. Sea level directly controls the position where an ice sheet begins to float and thus where ice is lost to the ocean. However, due to GIA, sea level does not vary uniformly over time: In areas where the ice sheet grows the land is pushed down by the extra load and the sea surface is drawn upwards due to the increased gravitational attraction of the ice sheet, resulting in an increase in sea level. The opposite happens when ice retreats. In both cases, feedbacks influence the dynamical behaviour of the ice sheet. Such feedbacks are not accounted for in current ice-sheet models, despite their potential to increase the stability of an ice sheet: Their inclusion could fundamentally alter predictions of ice-sheet behaviour under future sea-level rise scenarios. In the second component of my project I will therefore develop a coupled GIA-ice-sheet model. Both fields of research will benefit from improved model capabilities.These model improvements will be used to derive a more realistic estimate of GIA across Antarctica. The model will be calibrated to fit field constraints relating to former ice extent, past sea-levels, and present-day rebound. It will then be used to correct satellite data for the effects of GIA, and hence determine a more accurate map of current Antarctic ice-mass change.