A phase 1 investigation into the development of an end-to-end workflow to perform automated calibration and validation checks on Synthetic Aperture Radar data using Australian corner reflector sites.
The Challenge
The Commonwealth Scientific and Industrial Research Organisation (CSIRO) AusCalVal Facility aims to provide national support for calibrating and validating Earth Observation sensor data and derived products in Australia. As part of this initiative, the CSIRO is working to develop Australia’s capacity to automate the processing and analysis of satellite-acquired Synthetic Aperture Radar (SAR) data to perform calibration and validation over Australian corner reflector sites.
Currently there are no end-to-end open-source workflows to conduct this type of analysis but there are existing open-source python libraries that could be implemented into an all-encompassing workflow that would aim to pull imagery from a repository, run the analysis, pushes the results to be stored that can then be accessed through a user-friendly web-based interface.
Partners
This project was conducted in collaboration with the CSIRO.
The Solution
This project was a Phase 1 study conducted in two parts. Component one aimed to develop a workflow using open-source Python libraries and Sentinel-1 SAR data to run a single automated check and generation results for calibrating Sentinel-1 data against a single Queensland Corner Reflector Array. The second component of work saw the preparation of a report summarising learnings from Component 1, as well as outlining the high-level architecture required to develop an end-to-end workflow with recommendations. This involved outlining and providing options for the processing environments that could be implemented, storage and database requirements, servers, and the web application.
Impact
The Australian Earth Observation from Space Road map highlights data quality assurance and integrity monitoring as an important focus point over the next nine years. The development of an end-to-end workflow to perform automated calibration and validation checks on SAR data would be a step forward in how this data is assessed in Australia and contribute to meeting the activities outlined in the roadmap. It will assist in providing a consistent method of calibration and validation for Sentinel-1 SAR imagery, that is open-source and gives a framework that could be expanded to assess other SAR datasets in the future and potentially applied globally.
Contact
To learn more, contact FrontierSI at contact@frontiersi.com.au or connect with Space Technical Lead, Jia-Urnn Lee, at jlee@frontiersi.com.au.