Customer: Centre National d’Etudes Spatiales (CNES)

Programme: SWOT

Supply Chain: CNES > CS Group SPACE

Context

CS Group responsabilities for Surfwater, dynamic monitoring of water surfaces are as follows:

• Specification, design, development, maintenance
• Calibration and validation

The features are as follows:

• Time series processing
• Near real-time product delivery for continuous water surface monitoring (new SW product for each S1/S2 product made available) and evolution assessment.
• Monthly synthesis product: water occurrence map
• Combined processing of Sentinel 1 and 2 data for both high-frequency monitoring and heterogeneous information for better water occurrence detection.
• NRT processing based on ‘incremental methodology’ using previous NRT product collectively with last Sentinel data instead of reprocessing the whole temporal series
• Both desktop and HPC applications sharing the same algorithm implementation.

Project implementation

The project objectives are as follows:

• Specify, design and assess a water surface detection processing chain.
• Make use of Sentinel-1 and Sentinel-2 data in order to provide continental water surface evolution monitoring.

The processes for carrying out the project are:

• Agile

Technical characteristics

The solution key points are as follows:

• Continuous benchmarking and quality assessment using ‘Continuous Integration’ approach
• Automatic performance report generation
• Calibration / Validation based on quality metrics and high quality reference data over various well selected test sites for robustness evaluation

The main technologies used in this project are:

Domain	Technology(ies)
Hardware environment(s)	Linux 64 bits, HPC, workstation
Operating System(s)	Red Hat 7
Programming language(s)	Python (v3.x), C++, Java, Markdown
Interoperability (protocols, format, APIs)	OpenSearch Geo & Time, WMS, WPS, WCS, CLI, SDK Python, REST Interfaces
Production software (IDE, DEVOPS etc.)	Pylint, PBS Pro (parallel processing on HPC)