Objectives and expected impact

The Copernicus Earth Observation programme of the European Union is believed to be a game changer for both science and the industry, providing free and open data at large scale and high quality, constituting a fundamental paradigm change in Earth Observation. Today, Copernicus is producing several terabytes of data every day, while every product is downloaded on average 10 times. However, the availability of the sheer volume of Copernicus data outstrips our capacity to extract meaningful information from them. The EO community needs technology enablers from the ICT field to propel the development of entirely new applications at scale.

DeepCube – “Explainable AI pipelines for big Copernicus data” – is a three-year project, funded by the Horizon 2020 programme of the European Union under the topic “Big data technologies and Artificial Intelligence for Copernicus”. The project aims to unlock the potential of Copernicus data, leveraging on advances in the fields of Artificial Intelligence and Semantic Web. The ultimate goal of DeepCube is to address new and ambitious problems that imply high environmental and societal impact, enhance our understanding of Earth’s processes that are correlated with Climate Change, and feasibly generate high business value.

To achieve this, the DeepCube Consortium will combine mature and new ICT technologies, such as the Earth System Data Cube, the Semantic Cube, the Hopsworks platform for distributed Deep Learning, and a state-of-the-art visualization tool tailored for linked Copernicus data, and integrate them to deliver an open and interoperable platform that can be deployed in several cloud infrastructures and High-Performance Computing, including the cloud-based platforms providing centralized access to Copernicus data, known as DIAS (Data and Information Access Services). These tools will then be used to develop novel Deep Learning pipelines to extract value from big Copernicus data.

The DeepCube technologies will be showcased in six Use Cases (UCs):

  • Forecasting of localized extreme drought and heat impacts in Africa (UC1)
  • Climate induced migration in Africa (UC2)
  • Fire hazard short-term forecasting in the Mediterranean (UC3)
  • Global volcanic unrest detection and alerting (UC4a)
  • Deformation trend change detection for critical infrastructure monitoring (UC4b)
  • Copernicus services for sustainable and environmentally-friendly tourism (UC5).

DeepCube develops architectures that extend to non-conventional data and problems settings, such as Interferometric SAR, social network data, and industrial data, and introduces a novel hybrid modeling paradigm for data-driven AI models that respect physical laws, opening up the Deep Learning black box through Explainable AI and Causality.