From Pixels to Insights: Python for Earth Observation - ELENI TOKMAKTSI

[EuroPython 2026 - S2 on 2026-07-16] 🎤 *From Pixels to Insights: Python for Earth Observation by ELENI TOKMAKTSI* 🔗 https://ep2026.europython.eu/session/from-pixels-to-insights-python-for-earth-observation 📝 Abstract: Due to climate change, *large wildfires* continue affecting all continents of the Earth, leading to forest loss and exacerbating environmental impacts. Even more, current changing weather conditions associated with global warming will further increase fire danger to a global extent. Investigating these phenomena in a data-driven manner helps in decision making in pre-intervention and post-restoration following wildfire events. *Satellites* orbiting Earth can offer critical insights by capturing detailed images of the planet’s surface. Utilizing this data, environmental scientists perform detailed assessments to monitor an ecosystem’s loss and post-restoration progress. However, data from satellites arrive in the form of millions of raw pixels and turning them into analysis ready products is a *time-consuming* task full of *multi-step* and *error-prone* processes. This talk introduces an end-to-end Python workflow to automate the processing of *Earth Observation (EO) data.* The presentation walks through pixels to insights showcasing a real example and highlighting both the power of automation in the field of remote sensing and the importance of EO data in *climate change monitoring.* After the session, the audience will understand how utilizing open Python packages such as *Rasterio* and *NumPy* with openly distributed data from the [European Space Agency’s Sentinel-2 mission](https://dataspace.copernicus.eu/data-collections/copernicus-sentinel-missions/sentinel-2) can lead us quickly into crucial and spatially meaningful information. Although the demonstration focuses on wildfire events, the automated workflow is broadly applicable to other environmental fields such as land cover change detection, hydrological assessments and coastal studies. *No prior knowledge is required*! The *goal* is not to delve into complex mathematical equations, but to showcase the power of automation in remote sensing with Python. --- This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License: https://creativecommons.org/licenses/by-nc-sa/4.0/