Sea ice and icebergs
WP2: Monitoring Sea Ice and Icebergs
The decrease of sea ice extent and thickness in the Arctic over the last decades, especially during summer months, is closely linked to climate change. The expectation of an easier access to Arctic regions has triggered considerations concerning the expansion of marine transport and the possibilities to exploit natural resources in the Arctic Ocean. Hence, satellite monitoring of the Arctic sea ice and icebergs is inevitable for evaluations of chances and risks.
At CIRFA we develop methods for information retrieval from satellite images which are tested by the Norwegian Ice Service to complement manual analysis of the images. Together, we are working towards a 24/7 automatic sea ice forecasting for the ocean and the Norwegian fjords.

Image credit: Alex Perz / Unsplash


We develop and improve remote sensing methodologies and algorithms to automatically map Arctic sea-ice conditions, and provide enhanced methods for detecting icebergs. Besides synthetic aperture radar (SAR) as key data source, we also use optical and thermal satellite images, passive microwave and altimeter data.
Our studies focus on semi- and fully automated sea ice classification, retrieval of sea ice drift and deformation, observing temporal and spatial variations of sea ice conditions, and on finding optimal algorithms and radar configurations for monitoring icebergs.
The methods used include multivariate statistical analysis, anomaly detection, constant false alarm rate (CFAR) approaches, and machine learning/deep learning techniques. For validation, we use data from field measurements and generate a data base of SAR and optical images acquired over sea ice and icebergs at different seasons and meteorological conditions.
Research tasks
Development and improvement of (semi-)automated algorithms for sea ice type classification suitable for operational applications
Integration of data from new field measurement techniques (e.g. drones) into algorithm development
Retrieval and modelling of ice drift and deformation for short-term forecasting of ice conditions
Comparison and improvement of algorithms for detecting icebergs in open water and in sea ice
Team members
Professor II and WP 2 leader
WP 2 co-leader and leader of UiT`s Earth Observation group
Remote Sensing, Signal Processing, Image Processing, Neural Networks
Satellite laser and radar altimetry to study the physical properties of polar sea ice and oceans
Multi-frequency (C- and L-band) SAR sea ice classification
Sea ice classification with AI methods
Radar polarimetry for sea ice applications
Multimodal Integrated Remote Sensing for Arctic Sea Ice monitoring
Automated Classification of Sea Ice Types in SAR Imagery
Sea ice deformation and its impact on the sea ice mass balance, SiDRIFT project
Cross-platform application of a sea ice classification method for detecting deformed ice, SiDRIFT project
Multi-temporal remote sensing of dynamic Arctic phenomena
Mapping and Modeling of Iceberg Occurrences in the Barents Sea