TanDEM-X (TerraSAR-X add-on for Digital Elevation Measurement) is the first Synthetic Aperture Radar (SAR) mission using close formation flying for bistatic SAR interferometry. The primary goal of the mission is to generate a global digital elevation model (DEM) with 2 m height precision and 10 m ground resolution from the configurable SAR interferometer with space baselines of a few hundred meters. As a key mission requirement for the interferometric SAR processing, the relative position, or baseline vector, of the two satellites must be determined with an accuracy of 1 mm from GPS measurements collected by the onboard receivers. Given the high importance of the accurate baseline solutions for the interferometric SAR processing, independent baseline solutions are routinely generated by both GFZ and the German Space Operations Center (DLR/GSOC) within the TanDEM-X project. For a further independent performance assessment, the TanDEM-X baselines were additionally computed by the Astronomical Institute of the University of Bern (AIUB) on a best effort basis to perform an inter-agency comparison of dual- and single-frequency baseline solutions (Jäggi et al., 2012).
Figure 1 shows the daily standard deviations (STDs) of the inter-agency baseline comparisons for January 2011 based on dual- and single-frequency GPS carrier phase observations, respectively. Empty bars indicate the statistics for entire 24 h arcs (including maneuver periods) and characterize the capability of the three agencies to handle the frequent TanDEM-X formation-keeping maneuvers. Filled bars, on the other hand, exclude time intervals for each day starting 20 min before the first maneuver and ending 20 min after the second maneuver, which allows to assess the consistency outside the maneuver periods.
Excluding maneuver time periods, Figure 1 (left) shows an almost constant consistency level between AIUB and DLR with STDs of about 0.5, 0.9, and 1.0 mm in the radial, along-track, and out-of-plane components, respectively. Apart from the out-of-plane component, the comparisons between GFZ and the other agencies reveal a slightly larger variability for the radial and along-track component, e.g., about 0.24 instead of 0.09 mm variability for the along-track component, but with almost identical overall STDs. Figure 1 (right) shows a further improvement when using single-frequency data, e.g., STDs of about 0.3, 0.4, and 0.8 mm in the radial, along-track, and out-of-plane components between AIUB and DLR. When including maneuver time periods in the comparison of the solutions, a significant increase of the STD (up to a factor of two over 24 h) can be observed for the baseline difference in the radial and along-track direction.
Figure 2 show biases of about 1 mm at maximum between all agencies for dual-frequency solutions. Smallest biases of 0.1–0.2 mm occur for the radial direction. Tight relative constraints imposed on the empirical accelerations in the radial direction ensure a similar leveling for all baseline solutions. A consistent force modeling is more important to keep biases between the different solutions small than to significantly improve the STDs.
Jäggi, A., Montenbruck, O., Moon, Y., Wermuth, M., König, R., Michalak, G., Bock, H., Bodenmann, D. (2012): Inter-agency comparison of TanDEM-X baseline solutions. Advances in Space Research, 50(2): 260-271, doi: 10.1016/j.asr.2012.03.027.