Operational TanDEM-X DEM calibration and first validation results

详细信息    查看全文

• 作者：Astrid Gruber ; ^{Astrid.Gruber@dlr.de} ; Birgit Wessel ^{Birgit.Wessel@dlr.de} ; Martin Huber ^{Martin.Huber@dlr.de} ; Achim Roth ^{Achim.Roth@dlr.de}
• 关键词：Block adjustment ; DEM ; Calibration ; InSAR ; TanDEM-X
• 刊名：ISPRS Journal of Photogrammetry and Remote Sensing
• 出版年：2012
• 出版时间：September, 2012
• 年：2012
• 卷：73
• 期：Complete
• 页码：39-49
• 全文大小：1693 K

文摘

In June 2010, the German TanDEM-X satellite was launched. Together with its twin satellite TerraSAR-X it flies in a close formation enabling single-pass SAR interferometry. The primary goal of the TanDEM-X mission is the derivation of a global digital elevation model (DEM) with unprecedented global accuracies of 10 m in absolute and 2 m in relative height. A significant calibration effort is required to achieve this high quality world-wide. In spite of an intensive instrument calibration and a highly accurate orbit and baseline determination, some systematic height errors like offsets and tilts in the order of some meters remain in the interferometric DEMs and have to be determined and removed during the TanDEM-X DEM calibration. The objective of this article is the presentation of an approach for the estimation of correction parameters for remaining systematic height errors applicable to interferometric height models. The approach is based on a least-squares block adjustment using the elevation of ICESat GLA 14 data as ground control points and connecting points of adjacent, overlapping DEMs as tie-points. In the first part its implementation in DLR¡¯s ground segment is outlined. In the second part the approach is applied and validated for two of the first TanDEM-X DEM test sites. Therefore, independent reference data, in particular high resolution reference DEMs and GPS tracks, are used. The results show that the absolute height errors of the TanDEM-X DEM are small in these cases, mostly in the order of 1-2 m. An additional benefit of the proposed block adjustment method is that it improves the relative accuracy of adjacent DEMs.