2026-10-08, 16:15–17:00 (Europe/Madrid), Room 18
There is no single "correct" way to build a geospatial data pipeline. The tools and methods used often depend on the specific project, the size of the data, and the available infrastructure. This workshop presents one possible approach to building a pipeline that is both reproducible and maintainable. Rather than prescribing a definitive standard, we will demonstrate a functional workflow using open-source tools to solve common problems. The goal is to provide a starting point and discussion where participants can learn and reflect on other tools according to their own needs.
Why we chose Kedro (and what it is) (15 min): We will introduce the basic concept of Kedro: a framework that forces a clean separation between "how you process data" (nodes) and "where your data is stored" (the data catalog). We will explain why we moved away from standalone scripts to this structured approach to help with collaboration and long-term maintenance.
How to build a data pipeline (15 min)
Pain points and open discussion (15 min)
Open-Earth-Monitor Cyberinfrastructure (Grant agreement ID: 101059548)
Please provide URL that you plan to use to distribute your materials (if available). –TBD
Biel is a scientist at Vizzuality dedicated to advancing projects that both mitigate and deepen understanding of environmental and climate crises. He leverages his expertise in Python to acquire, clean, and assess datasets for quality, supporting the successful delivery of Vizzuality’s initiatives. With a strong academic background in environmental science, data science, remote sensing, and GIS from the University of Barcelona, he also brings practical experience from his work in software development at an agrotech company focused on regenerative agriculture.