Climate AI Nordics launches official YouTube channel

🔴 We’re Live!

As Climate AI Nordics just celebrated its first anniversary, we are now thrilled to announce the launch of our official YouTube channel!

This past year has been filled with incredible milestones, from our first annual workshop to a diverse series of expert webinars. Now, we are making these resources accessible to everyone. Whether you are a researcher, a practitioner, or simply an enthusiast, our new channel is your hub for the latest discussions at the intersection of AI, climate, and conservation.

🎥 Now Streaming: The Nordic Workshop on AI for Climate Change

Did you miss our first annual workshop in Gothenburg on May 13, 2025? Or perhaps you want to revisit a specific presentation?

We have released the recordings from this full-day event, co-arranged with our partners at CLIMES. You can now watch leading researchers from across the region discuss how AI is revolutionizing fields such as: biodiversity monitoring, atmospheric chemistry, soundscape analysis, and Earth observation.

Direct link to 2025 Workshop Playlist

🧠 An amazing collection of webinars

Beyond the workshop, our channel already features a rich archive of our webinars. These sessions have served as a meeting place for deep dives into specific technical challenges and solutions in the intersection of AI and climate action.

Explore talks from the past year, including:

Peter Dueben (ECMWF) on the shift from physics-based forecasts to foundation models in Earth system modelling.
Lily Xu (Columbia University) on real-world AI deployment for wildlife protection using reinforcement learning.
Joel Oskarsson (Linköping University) on graph-based approaches to weather prediction.
Benjamin Cretois (NINA) on using acoustic sensors for large-scale biodiversity monitoring.
Sigrid Passano Hellan (NORCE/Bjerknes Centre) on making ML more accessible through Bayesian optimisation.
…and many more contributions from experts at King’s College London, University of Florida, AfriClimate AI, RISE, ETH Zurich, MIT, Lund University, and many more.

This YouTube channel is just one more way we are connecting the Nordic AI/Climate community. As we look forward to the upcoming AICC workshop at EurIPS 2025 in Copenhagen, we invite you to subscribe, watch, and share these resources.

👉 Click here to visit the channel and subscribe!

A special thank you to RISE Research Institutes of Sweden for sponsoring our work, and to all the speakers who have made this first year a resounding success.

Share on

LinkedIn BlueSky

Featured paper: Separating the Albedo-Reducing Effect of Snow Impurities Using Machine Learning

Published: 2025-11-25

This study presents an AI-driven inversion framework that uses a deep-learning emulator of a radiative transfer model to estimate snow albedo and quantify the impact of light-absorbing particles (LAPs) such as dust, black carbon, and algae. Trained on 5.8 million simulations, the method accelerates computations by 30× while maintaining accuracy, enabling large-scale mapping from drone or satellite data. Applied to field observations, it revealed strong but variable LAP effects on albedo and complex interactions between multiple impurities, offering a powerful tool for improving melt forecasts and climate projections.

Featured member, November 2025: Joakim Bruslund Haurum

Published: 2025-11-25

Joakim Bruslund Haurum, an Assistant Professor at the University of Southern Denmark’s new Vejle Campus and the Pioneer Centre for AI, is the featured member for November 2025. He is a Computer Vision researcher specializing in expert tasks, balancing pure research on algorithms with applied work in infrastructure inspection and biodiversity monitoring. Beyond his studies on hierarchical data, Joakim engages with the scientific community (and with Climate AI Nordics) by co-organizing events like the AI for Climate and Conservation workshop at EurIPS. He is currently seeking collaborations with others interested in solving challenges related to messy data in expert domains.

TESSERA: precomputed FAIR global pixel embeddings for earth representation and analysis

Event date: 2025-12-11

Webinar with Zhengpeng (Frank) Feng, Cambridge. Petabytes of satellite Earth Observation (EO) data are freely available and can address critical global challenges. However, EO data quality is poor due to clouds and variable lighting conditions. To address this, practitioners typically use compositing, but this critically removes the temporal phenological signal. Moreover, supervised machine learning to map composited pixels to task-specific classes requires accurately labelled data that are rarely available. We present TESSERA, a pixel-oriented foundation model for EO time series that creates 128-dimensional latent embeddings requiring only a few labels for task-specific training to achieve state-of-the-art performance across diverse complex tasks. TESSERA uses two encoders that combine optical data with synthetic aperture radar backscatter coefficients at 10m resolution, creating embeddings fused with a multilayer perceptron to generate annual global embedding maps. TESSERA closely matches or outperforms state-of-the-art task-specific models and other foundation models across five diverse downstream tasks. It is unprecedented in ease of use, scale, and accuracy: no other open foundation model provides precomputed outputs with global, annual coverage at 10m resolution.