Observed SIC for certain regions can vary substantially from one year to another. (a) SIC observed for September 2007, when the extent of the sea ice was lower than usual. (b) SIC observed for September 2013, when the extent of the sea ice was greater than usual. (c) Area where the observed SIC differed the most between September 2007 and September 2013 is highlighted (white), used as the ROI for gradient attribution.
Image:
Harald L. Joakimsen (UiT)

Observed SIC for certain regions can vary substantially from one year to another. (a) SIC observed for September 2007, when the extent of the sea ice was lower than usual. (b) SIC observed for September 2013, when the extent of the sea ice was greater than usual. (c) Area where the observed SIC differed the most between September 2007 and September 2013 is highlighted (white), used as the ROI for gradient attribution.

Collaborative paper with The Alan Turing Institute and the British Antarctic Survey accepted in IEEE Geoscience and Remote Sensing Letters

We are thrilled to announce that the paper "Interrogating Sea Ice Predictability With Gradients", a collaborative paper between Visual Intelligence, The Alan Turing Institute and the British Antarctic Survey, has been accepted in IEEE Geoscience and Remote Sensing Letters.

Collaborative paper with The Alan Turing Institute and the British Antarctic Survey accepted in IEEE Geoscience and Remote Sensing Letters

We are thrilled to announce that the paper "Interrogating Sea Ice Predictability With Gradients", a collaborative paper between Visual Intelligence, The Alan Turing Institute (AT) and the British Antarctic Survey (BAS), was accepted in the journal IEEE Geoscience and Remote Sensing Letters on February 14th 2024.

The paper focuses on interrogating the effect of the IceNet's, a state-of-the-art ice prediction model, input feature with a gradient-based analysis, which takes advantage of the developments within the deep learning literature to open the so-called "black box".

The authors' analysis focuses on the unusually large sea ice extent event in September 2013 and indicates that IceNet places a strong emphasis on previous observations of SIC, linear trends, and seasonal components when making predictions. They further identify which input features are most influential for the prediction and also which spatial location these measurements are particularly influential.

The authors of this particular work from the Visual Intelligence side are post doc. Luigi Luppino (UiT), PhD students Harald Lykke Joakimsen and Iver Martinsen (UiT), and Robert Jenssen (UiT). From the UK side, the authors are Scott Hosking (BAS/AT) and Andrew McDonald (BAS).

You may read the paper via IEEE Explore.

Abstract

Predicting sea ice concentration (SIC) is an important task in climate analysis. The recently proposed deep learning system IceNet is the state-of-the-art sea ice prediction model. IceNet takes high-dimensional climate simulations and observational data as input features and forecasts SIC for the next 6 months over a spatial grid over the northern hemisphere. The model has proven to be particularly good at predicting extreme sea ice events compared with previous dynamical models, but lacks interpretability. In the original IceNet paper, a permute-and-predict approach was taken for assessing feature importance. However, this approach is not capable of revealing whether a feature contributes positively or negatively to the final prediction, nor can it reveal the importance of features over the spatial grid of predictions. In this letter, we take steps to instead interrogate the effect of the IceNet input feature with a gradient-based analysis, taking advantage of developments within the deep learning literature to open the so-called black box. Our analysis focuses on the unusually large sea ice extent event in September 2013 and indicates that IceNet places a strong emphasis on previous observations of SIC, linear trends, and seasonal components when making predictions. In our analysis, we identify which input features are most influential for the prediction and also which spatial location these measurements are particularly influential.

Latest news

Call for Papers and Abstracts: NLDL 2027

April 22, 2026

The Call for Papers and Abstracts for the Northern Lights Deep Learning (NLDL) Conference 2027 is officially announced – with submission deadlines on August 7th and Mid-September 2026 respectively.

Trends in Visual Intelligence 2026

April 17, 2026

The field of Visual Intelligence is continuously transforming. Chief Research Scientist Arnt-Børre Salberg dives deeper into the current trends in the field of visual intelligence as of early 2026.

Centre-developed seismic foundation model is now open source!

April 6, 2026

The NCS model, a seismic foundation model trained on data from the Norwegian data repository for subsurface data, is now available as an open-source model, allowing anyone to download, utilize, and further develop the model.

Visual Intelligence Annual Report 2025

March 31, 2026

The Visual Intelligence Annual Report 2025, highlighting the centre's progress, activities, achieved innovations, staff, funding, and publications for 2025, is now available to read on our websites.

Visual Intelligence strengthens ties with Pioneer Centre for AI in EHR-related research

March 26, 2026

Visual Intelligence researchers contributed to the Pioneer Centre for AI workshop on Electronic Health Records research. The aim was to strengthen ties between the two centres on EHR-related research.

Nordlys: Her blir KI-studentene grillet av sin «egen» teknologi

March 24, 2026

Tre av studentene i sivilingeniør i Kunstig Intelligens ved UiT skal delta i NM i KI. Slik gikk det da de ble intervjuet ved hjelp av kunstig intelligens (News article in nordlys.no).

My Research Stay at Visual Intelligence: Rami Al-Belmpeisi

March 15, 2026

Rami Al-Belmpeisi is a PhD Research Fellow in the Visual Computing section at DTU Compute, Technical University of Denmark. He visited Visual Intelligence in Tromsø from November 2025 to February 2026.

Visual Intelligence inspires future students at the UiT Open Day

March 12, 2026

Visual Intelligence researchers came to the UiT Open Day to inform the students about UiT's study programme, and inspire them to pursue AI-related studies and career paths.

Dagsavisen: Hun lærer kunstig intelligens å forstå medisinske bilder

March 4, 2026

Elisabeth Wetzer forsker på hvordan maskiner kan lære å analysere medisinske bilder – og samtidig forstå hva legene faktisk ser etter (Norwegian news article on dagsavisen.no)

My Research Stay at Visual Intelligence: Artur Radzivil

February 12, 2026

Artur Radzivil is a PhD Research Fellow at Vilnius Gediminas Technical University. He visited Visual Intelligence in Oslo from September to November 2025.