期刊
GEOPHYSICAL JOURNAL INTERNATIONAL
卷 196, 期 3, 页码 1431-1440出版社
OXFORD UNIV PRESS
DOI: 10.1093/gji/ggt485
关键词
Spatial analysis; Satellite geodesy; Geopotential theory
资金
- BMBF (Bundesministerium fur Bildung und Forschung)
- BfG (Bundesanstalt fur Gewasserkunde)
Global high-resolution gravity models derived from ESA's GOCE (Gravity field and steady-state Ocean Circulation Explorer) satellite mission have enabled a wealth of new applications in oceanography, geophysics and geodesy. Yet, it has been suspected for long that the traditional spherical harmonics (SHs) modelling fails to exploit the full information contained in the GOCE gradiometer measurements. Here, we demonstrate that a regional gravity retrieval strategy indeed performs favourable in a number of case studies and when compared with SH modelling, while applied to the same GOCE level 1b data. The regional strategy results in an up to a 50 per cent smaller noise level in smooth ocean areas and facilitates extracting additional signal in oceanic areas with high-frequency gravity field features. Our approach to regional gravity field determination from GOCE gradiometry employs a parametrization with radial basis functions, and it allows for regional tailoring of the regularization. This also enables us to assign more than one regularization area within a single regional solution, leading to an improved flexibility in dealing with the geographically varying roughness of the gravity field. We show that this results in a reduction of noise in areas of smooth gravity field and that it allows to extract additional information in regions with a strong high-frequency signal. Our regional solutions have been derived as refinements to the global GOCE-only model ITG-Goce02, and case studies have been carried out for the Pacific Ocean, in the area around the South Sandwich Trench in the southern Atlantic, and in the North Sea. Our results are validated against the global high-resolution gravity field model EGM2008 and against altimetry observations.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据