Please note that this session was withdrawn and is no longer available in the respective programme. This withdrawal might have been the result of a merge with another session.
GI2.5 | Data-Driven Research in Earth Science
EDI
Data-Driven Research in Earth Science
Convener: Jie Dodo XuECSECS | Co-conveners: J ZhangZhou, Miguel Bernecker, Guillaume Siron, Maurizio Petrelli
The latest trends and developments in computer science accompanied by an ever-growing amount of analytical and experimental data, often measured on (semi-)automatized setups, almost inevitably requires pipelined data reduction and processing schemes to address these otherwise immensely time-consuming tasks. Such tools can be represented by code snippets, scripts, spreadsheets or a combination of those, and sometimes even full-fledged programs, in order to optimise workflows. Availability of a great variety of algorithms powered by statistical learners can nowadays be used to study and interpret highly complex geological problems. With the developments of computer science, new applications, and ideas would always pop out. However, the network between researchers working with big and complex data sets needs to be built up and all efforts across various fields of geosciences need to be communicated. In our session, we provide a spotlight for presentations about:

• Analytical data reduction and processing pipelines (e.g., data correction, standardisation, error-propagation)
• Data mining, machine learning, deep learning and artificial intelligence applications in geochemistry
• Newly developed tools and improvements of existing software that facilitate our research
• New development in computational petrology, whether they are new software, algorithms or new thermodynamic databases
• Visualization and visual analytics of big and high-dimensional datasets
• Emerging big data paradigms, such as datacubes

The presented work can be written in any programming language and be in any stage of their development. In particular, open-sourced research following the FAIR (Findable, Accessible, Interoperable, Reusable) principles is encouraged. We aim to raise awareness of what is already out there, to facilitate the sharing of new techniques and ideas, and to foster collaborations for more possibilities.