利用GRACE重力與ICESat測高衛星資料估算南極大陸之冰下地形及地殼厚度

Abstract

本研究結合GRACE (Gravity Recovery and Climate Experiment) 重力與 ICESat (Ice, Cloud, and land Elevation Satellite) 測高衛星資料，提供一種新的方式可用以估計南極大陸的冰下地形及地殼厚度。南極大陸為世界第五大洲，大約98%的面積為冰雪所覆蓋，地處偏遠加上氣候惡劣，使得當地的測量工作相當不易，只有一些鑽井、震測及地面與航空重力測量不均勻的分布在大陸上，為了解決資料覆蓋率不足與不均勻的問題，我們嘗試使用衛星資料克服。GRACE重力衛星資料能顯示質量分布，而ICESat測高衛星則提供了高解析度的地表地形，將兩者與BEDPMAP冰層模型和其他資料加以整合，建立一初始模型，視南極地區之重力異常的貢獻主要來自三個部分：冰層，橢球面以上之陸地與莫荷起伏面，並假設模型為均質無側向密度變化，加以正演計算由此三層所引起之重力效應，並將結果與GRACE觀測重力比較，再根據兩者之間的殘差值，利用試誤法調整初始模型，直到模型的計算重力值與觀測重力相吻合。我們分別考慮了均衡與非均衡狀態下的重力異常，結果顯示，南極大陸相當接近均衡狀態，且地殼厚度大致介於17∼50公里，冰層厚度大約介於0∼4300公尺。由於重力逆推問題存在著非唯一性，因此，我們的估算結果只是在合理的假設範圍內，其中一種可能的解。

We propose a new method combining GRACE (Gravity Recovery and Climate Experiment) gravity and ICESat (Ice, Cloud, and land Elevation Satellite) altimetry data to estimate the land topography for Antarctica. Antarctica is the fifth-largest continent in the world and about 98% of Antarctica is covered by ice, where in-situ measurements are difficult. Some experimental airborne radar and ground-based radar data have revealed very limited land topography beneath heavy ice sheet. To estimate the land topography for the full coverage of Antarctica, we combine GRACE data that indicate the mass distribution, with data of ICESat laser altimetry that provide high-resolution mapping of ice topography. Our approach is actually based on some geological constraints: assuming uniform densities of the land and ice considering the Airy-type isostasy. In the beginning we construct an initial model for the ice thickness and land topography based on the BEDMAP ice thickness and ICESat data. Thereafter we forward compute the model’s gravity field and compare with the GRACE observed data. Our initial model undergoes the adjustments to improve the fit between modeled results and the observed data. The final results show that the crust thickness is between 17 to 50 kilometer and the ice thickness is between 0 to 4300 m. And final examination is done by comparing our results with previous but sparse observations of ice thickness to reconfirm the reliability of our results. As the gravitational inversion problem is non-unique, our estimating result is just one of all possibilities constrained by available data in optimal way.