Effect of GIA models with 3D composite mantle viscosity on GRACE mass balance estimates for Antarctica
Effect of GIA models with 3D composite mantle viscosity on GRACE mass balance estimates for Antarctica
复制标题
DOI：
10.1016/j.epsl.2015.01.001
复制
发表时间：
2015-03
影响因子：
5.3
通讯作者：
W. Wal;P. Whitehouse;E. Schrama
中科院分区：
文献类型：
--
作者：
W. Wal;P. Whitehouse;E. Schrama
地震数据表明南极洲下方的地幔存在很大的粘度变化。考虑到这种变化将影响冰川等静压调整（GIA）模型的预测，该模型用于校正冰质量变化的卫星测量结果。然而，大多数用于此目的的 GIA 模型都假设地幔在粘度方面是均匀分层的。本研究的目的是估计粘度横向变化对根据重力恢复和气候实验（GRACE）数据得出的南极质量平衡估计的影响。为此，最近开发的基于地幔温度横向变化的全球 GIA 模型经过调整以适应北半球的限制，然后与 GPS 得出的南极洲抬升率进行比较。我们发现，与具有径向变化 (1D) 流变学的现有 GIA 模型相比，这些模型可以更好地拟合南极洲的 GPS 抬升率。当 3D 粘度模型与特定冰载荷历史相结合来校正 GRACE 测量时，南极洲的质量损失比之前发现的相同冰载荷历史和首选 1D 粘度剖面要小。由于使用不同的 3D 粘度合理实现而产生的质量平衡估计变化对于 ICE-5G 冰模型而言为 20 Gt/年，对于 W12a 冰模型为 16 Gt/年；这些值大于 GRACE 测量误差，但小于未知冰历史引起的变化。虽然存在可以重现使用 3D 地球模型得出的总质量平衡估计值的 1D 地球模型，但重力速率的空间模式可能会受到 3D 粘度的显着影响，而具有 1D 粘度的 GIA 模型无法重现。例如，具有 1D 粘度的模型始终预测 ICE-5G 冰模型在罗斯海的最大重力率，但是，对于三个首选 3D 模型之一，在威德尔海附近发现了最大值（对于同一冰模型）。这表明粘度的 3D 变化会影响当前隆起和重力率对冰历史时间变化的敏感性。特别是，在西南极洲发现的低粘度（< 10 19 Pa s）使得地幔对最近冰厚度的变化非常敏感。
Seismic data indicate that there are large viscosity variations in the mantle beneath Antarctica. Consideration of such variations would affect predictions of models of Glacial Isostatic Adjustment (GIA), which are used to correct satellite measurements of ice mass change. However, most GIA models used for that purpose have assumed the mantle to be uniformly stratified in terms of viscosity. The goal of this study is to estimate the effect of lateral variations in viscosity on Antarctic mass balance estimates derived from the Gravity Recovery and Climate Experiment (GRACE) data. To this end, recently-developed global GIA models based on lateral variations in mantle temperature are tuned to fit constraints in the northern hemisphere and then compared to GPS-derived uplift rates in Antarctica. We find that these models can provide a better fit to GPS uplift rates in Antarctica than existing GIA models with a radially-varying (1D) rheology. When 3D viscosity models in combination with specific ice loading histories are used to correct GRACE measurements, mass loss in Antarctica is smaller than previously found for the same ice loading histories and their preferred 1D viscosity profiles. The variation in mass balance estimates arising from using different plausible realizations of 3D viscosity amounts to 20 Gt/yr for the ICE-5G ice model and 16 Gt/yr for the W12a ice model; these values are larger than the GRACE measurement error, but smaller than the variation arising from unknown ice history. While there exist 1D Earth models that can reproduce the total mass balance estimates derived using 3D Earth models, the spatial pattern of gravity rates can be significantly affected by 3D viscosity in a way that cannot be reproduced by GIA models with 1D viscosity. As an example, models with 1D viscosity always predict maximum gravity rates in the Ross Sea for the ICE-5G ice model, however, for one of the three preferred 3D models the maximum (for the same ice model) is found near the Weddell Sea. This demonstrates that 3D variations in viscosity affect the sensitivity of present-day uplift and gravity rates to changes in the timing of the ice history. In particular, low viscosities (< 10 19 Pa s) found in West Antarctica make the mantle very sensitive to recent changes in ice thickness.