Climate AI Nordics

Climate AI Nordics

Nordic network for collaborations around tackling climate change using AI and machine learning.

Featured member, November 2025: Joakim Bruslund Haurum

Hi Joakim! Tell us about yourself and your work.

Hi! My name is Joakim Bruslund Haurum and I recently started as Assistant Professor at the totally new Vejle Campus of the University of Southern Denmark (SDU) and the Pioneer Centre for AI. I am a Computer Vision researcher specializing in expert tasks such as infrastructure inspections and biodiversity monitoring. I love conducting all types of research from “pure” research where I dive into the specific of niche algorithms for Vision Transformers to more applied research where our work has a very clear path to impact in “the real world” such as detecting defects in sewer pipes.

Most recently I have been especially interested in how to deal with hierarchal data and limited labels in the context of biodiversity monitoring. I also regularly co-organize various scientific workshop in Denmark and top-tier conferences, most recently the Workshop on AI for Climate and Conservation at EurIPS (happening next week!), which spun out through funding from the Pioneer Centre together with Nico Lang and Ankit Kariryaa.Through these project I have had the pleasure of visiting leading researchers in at the Universitat de Barcelona in Spain, the Vector Institute for AI in Canada joining the BIOSCAN project, and most recently the University of Edinburgh in the UK, all of whom I still collaborate with.

Any opportunities or collaborations you’re looking for at the moment?

Since both my Assistant Professorship and the SDU Vejle campus are just 4 months old I am actively looking into potential collaborations for future research projects. So if you have an interest in how to deal with the messy data related to expert tasks I would love to chat!

Where can people reach you?

Feel free to reach out to me on any of the following channels!

You May Also Enjoy

Climate AI Nordics Newsletter, October 2025

Published:

In our October newsletter, we celebrate one year of Climate AI Nordics with a look back at our community's key milestones and achievements. This issue also features a new paper on urban change detection, details on upcoming workshops at EurIPS 2025, and several new job opportunities.

Junior researcher in machine learning

Published:

RISE is hiring a junior researcher in machine learning to develop AI-based tools for creating smarter, more sustainable cities and to help build a greener future.

Multi-resolution learning with neural operators and long short-term memory neural networks

Event date:

Webinar with Katarzyna Michalowska, SINTEF. In real-world applications, collecting large amounts of high-resolution data is rarely practical. The cost of specialized equipment, time-intensive measurements, and expensive high-fidelity simulations often means that high-resolution samples are limited. Meanwhile, low-resolution data are far easier to obtain and often exist in abundance, creating a common scenario: plentiful coarse data but limited fine data. Standard neural networks, which require fixed-resolution inputs, cannot exploit this imbalance and typically perform poorly when generalizing across resolutions. Deep operator networks, or DeepONets offer a distinct advantage over standard neural networks through a property known as discretization invariance, enabling learning across varying data resolutions. However, DeepONets alone do not effectively capture long-term temporal dependencies, limiting their performance on problems involving long time horizons. In this talk, we present a framework that addresses both challenges: multi-resolution learning and long-time horizon modelling. We achieve this by extending DeepONets with long short-term memory networks (LSTMs) and introducing a multi-stage training procedure that leverages data at multiple resolutions. This hybrid architecture first learns global dynamics from abundant low-resolution data and then fine-tunes on limited high-resolution samples, capturing both multi-resolution structure and temporal dependencies. In tests on nonlinear dynamical systems, our multi-resolution DON-LSTM achieves lower generalization error and requires fewer high-resolution samples than standard DeepONet or LSTM models. Our results demonstrate that the proposed approach is well-suited for real-world scenarios where high-resolution data are limited, highlighting its potential for practical applications across science and engineering.

Self-supervised pre-training for glacier calving front extraction from synthetic aperture radar imagery

Event date:

Webinar with Nora Gourmelon, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU). The factors influencing melt at glacier fronts facing the ocean remain an active area of research. To better understand these factors, it is essential to monitor changes in glacier calving fronts. Due to the importance of weather-dependent and seasonal variations, Synthetic Aperture Radar (SAR) is the preferred imaging method for such monitoring. In recent years, deep learning models have been developed to automate the extraction of calving front positions, however, their performance on SAR data remains suboptimal. Limited labeled data and high variability in SAR images hinder traditional supervised learning. Foundation models pre-trained on large, diverse datasets could provide robust feature representations that require minimal labeled data for fine-tuning on specific SAR tasks. However, in preliminary experiments, we found that the domain gap is too large for the task of extracting calving fronts from SAR imagery, and foundation models fine-tuned on the "CAlving Fronts and where to Find thEm" (CaFFe) benchmark dataset performed subpar. Therefore, we compiled an unlabeled dataset of Sentinel-1 SAR image sequences of Arctic glaciers, each associated with a single Sentinel-2 optical reference image. Using this dataset, we developed a novel multi-modal self-supervised pre-training strategy and applied it to pre-train a hybrid CNN-transformer model. Fine-tuning on the CaFFe benchmark showed that the pre-trained model outperforms its non-pre-trained counterpart, enabling more robust calving front segmentation and demonstrating the potential of data-efficient learning for SAR imagery.