Skip to main content

Leveraging the Impact of Copernicus Data and Data Science for the Energy Sector

Point of contact
Frédéric Adragna
Centre National d'Etudes Spatiales
18, avenue Edoaurd BELIN
31400 Toulouse
Phone: +33-6-07716915

Large amount of data is generated by the Copernicus program that has an untapped potential for many sectors of the European industry. Among these, the energy sector could benefit from the numerous kinds of information based on earth observations. The large amount and variety of data also represents an obstacle to their efficient use by stakeholders. Indeed, they have not always the expertise and infrastructure to assimilate EO data in their operational processes. There is therefore a need to translate the large amount of data available in Copernicus in a form that can directly tackle the industry need. In this Action, we will focus on the growing need of the energy sector for solar forecast and the potential of data science to make the required EO data transformation.

CNES coordinates this action and will select the most appropriate consortium to ensure the success of this action; in that perspective, CNES would opt for an entity that already has:

  • collected a large number of identified requests from users for these services,
  • a strong network constituted of eminent researchers,
  • and as the Technology Readiness Level of this service is already high (7), this entity must also have several prototypes available to minimize risk and ensure success of the Action.

The first phase of the project consists of a co-design activity where identified key stakeholders will be involved in selecting the relevant Copernicus data (e.g.  CAMS aerosol data, C3S energy information…) and defining the specifications of the forecast to be integrated in their operational processes. The co-design will also address the definition of key performance indicators (KPI) for assessing the performance of the forecasts. In the second phase, the relevant Copernicus dataset will be synthetized in a user-oriented information using machine learning techniques. In the end, an existing prototype service will be adapted to match the end-user needs identified through the co-design. The prototype service is deployed on the SoDa platform, where 135 million requests are made each year. This will leverage the impact of our work to the community. Finally, the feedback phase will consist in evaluating the performances of the developed services under consideration of the performance indicators defined in the first phase. A collection of further needs of the industry for future improvements will also be conducted.

 Outputs and Results

  • Results of the co-design phase on the specification of the forecast and the KPI
  • Online solar forecast service
  • Validation report
  • Publications in international journals
  • Communication of the results in conference