In-Depth Verification of Sentinel-1 and TerraSAR-X Geolocation Accuracy Using the Australian Corner Reflector Array

Christoph Gisinger,Helko Breit,Ulrich Balss,Adrian Schubert,Nuno Miranda,Michael Eineder,Matthew Garthwaite,David Small,Martin Willberg

doi:10.1109/tgrs.2019.2961248

Abstract

This article shows how the array of corner reflectors (CRs) in Queensland, Australia, together with highly accurate geodetic synthetic aperture radar (SAR) techniques-also called imaging geodesy-can be used to measure the absolute and relative geometric fidelity of SAR missions. We describe, in detail, the end-to-end methodology and apply it to TerraSAR-X Stripmap (SM) and ScanSAR (SC) data and to Sentinel-1 interferometric wide swath (IW) data. Geometric distortions within images that are caused by commonly used SAR processor approximations are explained, and we show how to correct them during postprocessing. Our results, supported by the analysis of 140 images across the different SAR modes and using the 40 reflectors of the array, confirm our methodology and achieve the limits predicted by theory for both Sentinel-1 and TerraSAR-X. After our corrections, the Sentinel-1 residual errors are 6 cm in range and 26 cm in azimuth, including all error sources. The findings are confirmed by the mutual independent processing carried out at University of Zurich (UZH) and German Aerospace Center (DLR). This represents an improvement of the geolocation accuracy by approximately a factor of four in range and a factor of two in azimuth compared with the standard Sentinel-1 products. The TerraSAR-X results are even better. The achieved geolocation accuracy now approaches that of the global navigation satellite system (GNSS)-based survey of the CRs positions, which highlights the potential of the end-to-end SAR methodology for imaging geodesy.

Highlights

T HE ability to accurately geolocate an synthetic aperture radar (SAR) image onto a predefined Earth model without needing ground control points is a unique feature of the SAR imaging technique known since the late 1970s and the SEASAT mission [1]
Closer inspection of the data revealed that these measurements are almost entirely related to two products acquired in December 2017; we assume some problems with these specific images even though the estimated signal-to-clutter ratios (SCRs) are in line with the other data and the atmospheric corrections are comparable to other acquisition dates in the series
If we recall the 2–4 cm accuracy of the corner reflectors (CRs) reference coordinates as well as the additional uncertainties introduced by the atmospheric path delays (2 cm for the troposphere and sub-cm for the ionosphere, see Section II-B), the accuracy of the TerraSAR-X precise science orbit, and the SCR contribution, these TerraSAR-X SM results can be interpreted as the geolocation limit presently achievable at the Australian CR array

Summary

Introduction

T HE ability to accurately geolocate an SAR image onto a predefined Earth model (e.g., a reference ellipsoid or a Digital Terrain Model) without needing ground control points is a unique feature of the SAR imaging technique known since the late 1970s and the SEASAT mission [1]. Other considerable absolute geolocation error sources, which were assumed to be negligible at the time of the SEASAT mission, are the meter-level path delays introduced by the Earth’s atmosphere and, at a submeter scale, the dynamic effects of the solid Earth. Current spaceborne SAR missions design the geometric accuracy to meet the resolution supported by the SAR sensor. They typically aim for an ALE of half the pixel spacing to ensure accurate geolocation of the SAR scenes for direct comparison with other geospatial data sets. For Sentinel-1, RADARSAT-2, TerraSAR-X, and COSMO-SkyMed, this translates into official requirements for the guaranteed geolocation ranging from 1 m up to a few meters [2]–[6]

Objectives

Results

Conclusion