News articles

Success story: TU Wien’s repository hosts 2.73 TB of satellite data described in Nature's Scientific Data article

Nature's Scientific Data published recently an article describing a worldwide satellite image dataset hosted by the institutional research data repository of TU Wien.

Satellite image zoomed into the area around the confluence of the Amazon and Tapajós streams, in central Pará in Brazil.  Different colours reveal streams and branches (dark blue), dense rainforest (green) and settlements (yellow).

Detail of Copernicus Sentinel-1 Global Backscatter Model (S1GBM) in Brazil l Contains modified Copernicus Sentinel-1 data (2016-2017) / processed by TU Wien

TU Wien’s research data repository, opens an external URL in a new window helps the remote sensing scientists of the GEO Department at TU Wien, opens an external URL in a new window to comply with the FAIR principles, opens an external URL in a new window, by providing free access to these detail-rich global mosaics, describing our planet as perceived by a radar - that is sensing with microwaves - with a 10m pixel sampling, totalling a volume of 2.73 TB.

The scientists from the GEO Department at TU Wien processed and subsequently aggregated 500,000 individual satellite images from the Copernicus Sentinel-1 SAR satellite mission to a set of global and harmonised mosaics. In respect to well-known optical imagery, the obtained radar signals depict a new source of information, measuring ground variables from another physical perspective, and revealing new properties on vegetation and land cover.

A close look at the Earth’s surface

Earth’s land surface is covered by a rich variety of soils, vegetation, human land uses, and many more features forming our planet’s multifaceted landscape. Being an every-day business of many researchers and users in satellite remote sensing, the characterisation of land-cover and -use lays the foundation for applications, for instance in agriculture, administration, climate change, or hazard management.

In contrast to well-known optical imagery from multispectral sensors, the high-resolution C-band SAR-sensors on-board the Copernicus Sentinel-1A/B satellites operate in the microwave spectrum independent of sunlight and undisturbed by clouds, responding to different physical processes.

The Sentinel-1 Global Backscatter Model (S1GBM) describes Earth’s surface for the period 2016-17 by the mean C-band radar cross section in VV- and VH-polarisation at a 10 m pixel sampling. In the frame of an ESA project, Copernicus Sentinel-1 input data and intermediate products (totalling a volume of 1.1 PB) was processed and quality-curated by the team at TU Wien, supported by resources of Vienna Scientific Cluster (VSC), opens an external URL in a new window and Earth Observation Data Centre for Water Resources (EODC), opens an external URL in a new window.

The overall mosaic quality excels (the few) existing datasets, with minimised imprinting from orbit discontinuities and incidence angle normalisation in large parts of the world. Supporting the design and verification of upcoming radar sensors, the obtained S1GBM data potentially also serve land cover classification and determination of vegetation and soil states, as well as water body mapping. Bernhard Bauer-Marschallinger, from TU Wien, states, “All the efforts were worth it, not only for its applications. Every time I zoom into the mosaic, I’m astonished of the clear image, for instance, when resolving the capillary structure of the Amazon’s water bodies and its shorelines. Our planet is quite beautiful as perceived by SAR”.

Data Access

Developers from the broader user community are invited to exploit this novel data resource and to integrate Copernicus Sentinel-1 GBM parameters in models for various variables of land cover, soil composition, or vegetation structure.

Data access is provided openly via the repository page, opens an external URL in a new window, following the FAIR-principles on digital data, opens an external URL in a new window.

For more details on generation, methods, and an in-depth dataset analysis please read our recent article in Scientific Data, opens an external URL in a new window. In this publication – as an example of the Copernicus Sentinel-1 GBM's potential use – the mapping of permanent water bodies is demonstrated and evaluated against the Global Surface Water (GSW) benchmark.

Contact

TU Wien
Center for Research Data Management
Favoritenstraße 16 (top floor), 1040 Vienna

research.data@tuwien.ac.at

Twitter: @RDMTUWien