CRIMA

The main objective of the CRIMA project is to develop a platform that implements an innovative methodological approach for querying and analyzing Big Earth Data, assisted by Artificial Intelligence, capable of integrating heterogeneous domain knowledge, and apply it to environmental monitoring and assessment of risks and adaptation processes to climate change.

The project also has the following objectives:

• Integrate and harmonize different sources of Big Earth Data: in particular multidimensional data cubes from Earth Observation platforms, data from in-situ sensors and ensembles of weather and climate models.
• Use cutting-edge technologies: The project involves the use of Ontop, a cutting-edge and technologically mature Virtual Knowledge Graph (VKG) system, and openEO, a platform for accessing Earth Observation (EO) data cubes. VKGs offer a flexible, graph-based data model that can accommodate and integrate virtually any type of data source, from RDBMS to unstructured data. This allows to handle dynamic and large data such as EO data. OpenEO is emerging as a standard for abstracting access to satellite observations in the optical, SAR and hyperspectral domains, as well as climate, meteorological and hydrological data.
• Apply innovative conceptual models: the impact chains, developed by EURAC to support climate risk assessment, are graphical and conceptual models accepted as community standards to describe the relevant causal components of climate risks and their mutual relationships, thus capturing the knowledge of experts and relevant scientific literature. This allows to strengthen the cross-functional role of domain knowledge and data science.
• Include non-technical users: The goal is to make the platform accessible also to those who do not have the necessary IT skills to query the data through query languages ​​such as SPARQL, through the use of Natural Language Processing (NLP) and Large Language Models (LLM) approaches.
• Demonstrate in application scenarios: The project activities will be demonstrated and validated in two risk analysis scenarios due to extreme hydrometeorological events and prolonged drought episodes and temperature anomalies.

Activities in 2025

WP1 Project Management

Initial setup of the communication structures and reporting for the project using MS teams and cohemon portal of the province.

WP 2

The project will start with requirement analysis for the technical integration and a more detailed application scenario that allows to drive and test the progress on the technical side with relevant topics from our local stakeholders. This activity includes collectively all project partners and will run from beginning of January 2025 to end of April 2025.

WP3

In parallel Ontopic will look into the coupling of the VKG with natural language based query interfaces. This activity starting in April will continue for the remainder of the first and second project year until the end of 2026.

WP 4

After this initial phase the first development period starts. On the data processing side university of Bolzano and Eurac will look into the coupling of the VKG with openEO.

WP 5

Eurac will then also start working on the detailed ontologies of the climate risk scenarios defined in the application together with the local stake holders.