Walter Knöpfle

GeMoS-a system for the geocoding and mosaicking of interferometric digital elevation models

The task of mapping and monitoring the Earths surface covers various application fields. Usually this 3-dimensional problem is split into a 2-dimensional planar description of the Earths surface in maps supplemented with the height information, stored as separate digital elevation models (DEMs). However there is a lack of suitable global height information. DLRs Remote Sensing Data Center (DFD) implemented a production chain for the operational generation of DEMs. It currently offers this service on the basis of SAR data from the ERS tandem missions. Starting in late 1999 it will be used for the processing of the X-band data of the Shuttle Radar Topography Mission (SRTM). The system will also be adapted for the generation of interferometric products delivered from the ASAR sensor of ESAs ERS follow on system ENVISAT. Part of the interferometric processing chain is the Geocoding and Mosaicking System (GeMoS). It enables the conversion of the unwrapped phase to height values, their geocoding into a variety of map projections, as well as the composition of individual InSAR DEMs to a large area, quality enhanced elevation product. An adjustment procedure improves the imaging parameters and as a consequence the product quality. The present paper briefly describes the system, its functionalities and the implemented approaches.

Operational interferometric SAR products

Today, SAR interferometry plays an important role in the observation of the Earth. It enables the generation of digital elevation models. Differential interferometry allows the determination of uplifting and subsidence and coherence is utilized for land use classification purposes. Most of todays radar satellites deliver data suitable for SAR interferometry. New missions will provide continuation of this capability. DLRs German Remote Sensing Data Center is implementing an operational SAR interferometry processing system, currently supporting ERS SAR data. It will be further developed for ESAs ERS follow on system ENVISAT as well as for the SRTM mission (Shuttle Radar Topography Mapper). The main objective is the operational interferometric derivation of DEMs to fill a global elevation data base. These DEMs are utilized for the terrain correction of amplitude images. Interferograms as base products for, for example, differential interferometry as well as coherence maps will be available for the repeat pass interferometry missions.

Analysis of SRTM interferometric X-band data: first results

During February 11/sup th/-22/sup nd/, 2000, the Shuttle Radar Topography Mission (SRTM) imaged the Earth with the first spaceborne single-pass SAR interferometer. The acquired data set will greatly enhance the knowledge of the global topography. At DLR the German Remote Sensing Data Center will process the X-band SAR data set to a nearly global homogenous digital elevation model (DEM). This work summarizes the first experiences with the SAR data.

Quality assessment of InSAR-derived DEMs generated with ERS tandem data

Provides a quality assessment of digital elevation models generated by means of SAR interferometry and the use of ERS tandem data. A total of 8 scene pairs have been investigated, acquired at different dates but covering the same site to allow a study of the different parameters such as temporal decorrelation, baseline, and varying incidence angles. The ERS tandem configuration implies some characteristics that are examined, concerning in particular the system parameters. Moreover, peculiarities detected during the interferometric processing are reported. The achieved results are compared to an existing reference DEM. Additionally, the InSAR-derived DEMs are compared to each other to study SAR-specific effects.

Filtering of interferometric SRTM X-SAR data

In February 2000, the Shuttle Radar Topography Mission acquired a global digital elevation model (DEM) within only 11 days. During the years 2000 and 2001, extensive testing and calibration activities followed at DLR. One of the questions during that phase was, which kind of interferogram filtering should be performed. We analyzed different types of filters from the perspective of an operational DEM processing chain. Our paper describes the filters investigated and the arguments that finally led us to the decision to implement a relatively simple Gaussian smoothing in the time domain. This filter performed best in many disciplines and produced the smallest artifacts.

SRTM X-SAR DEM of Europe-Results and algorithmic improvements

In February 2000, the Shuttle Radar Topography Mission acquired a global digital elevation model (DEM) within only 11 days. During the years 2000 and 2001 extensive testing and calibration activities primarily based on ocean surface data took place at DLR. The calibration phase is now finished and the quality of the products analyzed so far met all specifications. The routine processing of the German X-band data at DLR began in November 2001, starting with Europe. Now, that hundreds of image frames over Europe are being processed and mosaicked to a continent-wide DEM, the largescale system stability becomes visible. Our paper reports the experiences from mosaicking based on self-consistency tests and on comparison with independent reference data. While the overall quality is promising, there are still some outliers to be investigated and compensated. The calibration concept based on the use of ocean data will be presented and discussed. Adaptations and improvements based on self-consistency will be presented, that allow a more robust and automated processing.

The German processing/archiving and distribution system for X-SAR/SRTM

The generation of a world wide digital elevation model requires a complex processing system. Furthermore a robot maintained archiving system is required. Catalog interfaces for the users have to be provided. The authors consider the X-SAR SRTM (Shuttle Radar Topography Mission).

Optimization of the tie-pointing procedure for the terrain correction of SAR data

Usually tie-points are manually derived from map sheets. This time consuming process requires a significant part of the overall terrain correction process. Additionally it hampers the automation of the SAR-orthorectification. Therefore an autocorrelation method was implemented for DFDs operational SAR geocoding system GEOS. It considers tie-point chips as reference. Further improvements can be achieved by considering a bundle block adjustment. The advantages are the reduction of tie-points required per scene and the ability to span areas without any reference for geolocation information. GEOS is currently supplemented by such a technique.

ERS‐1 SAR geocoding system as link between spaceborne and earth reference data

The operational European ground segment of the European Remote Sensing Satellite ERS‐1 supports geocoded SAR image data generated by the German Processing and Archiving Facility (D–PAF) of DLR in Oberpfaffenhofen near Munich. Geocoding of spaceborne satellite data means to establish an easy and user friendly link between the satellite measurements and Geo‐Information Systems (GIS) as well as the possibility to intercompare SAR data with other geocoded satellite based information. Although the near to fully automatic geocoding system is based on precisely known satellite house‐keeping data, it strongly depends on georeference information at least for the validation of the data and the verification of its results. Such reference information are large scale topographic maps, which are still the unique global source for earth based information and Digital Elevation Models. The later are necessary to correct SAR data for disturbing height induced geometric and radiometric defects. Additionally automatically ge...