Seminar #4: ‘AI for Safe(r) Transportation in the Nordic Arctic: Satellite-Assisted Road-Tranportation in Lapland and AI-Generated Ice-Charts Around Greenland

WhenTuesday 02 February 2021, 14:00–16:00 (Central European Time) on Zoom.
Chair Rasmus Gjedssø Bertelsen (UiT The Arctic University of Norway)
PresentersMatilde Brandt Kreiner (DMI)
Timo Sukuvaara (FMI)
AbstractThis seminar looks at two cases of AI in Nordic Arctic contexts. All five Nordic countries are Arctic states from the Barents Region in the East to the West Nordic Region of Greenland, Iceland and Faroe Islands in the West. Northern Norway, Sweden and Finland has a population of about 1.6 million people, with terrestrial characteristics. The West Nordic Region has a population of almost half a million, being a sparsely populated and largely maritime region. Arctic conditions in both regions influence safe transportation whether on road, sea or by air.

This seminar will present research, development and innovation around satellite and AI for safe road-transportation at the Finnish Meteorological Institute Arctic station at Sodankylä in Finnish Lapland and Danish Meteorological Institute work on AI-assisted generation of ice-charts for Greenland from satellite data. There appears to be significant potential for AI supporting Nordic Arctic societies. Northern Norway, Sweden and Finland have well-developed higher education and research as well as infrastructure. Greenland struggles with low formal human capital and poor infrastructure. Data may be insufficient in sparsely populated regions. Some data and technologies have dual-use potential in a region of geostrategic importance, which complicates access to data and collaboration with Russia (and China).
QuestionsWhat unique challenges do data-poor environments present for the development of AI, such as the rural Arctic areas of the Nordic countries? How can or should different Nordic countries collaborate on the implementation of AI to avoid conflicts?

How can AI support sparsely populated Nordic (Arctic) communities – compared to urban centers? What are the differences in AI for southern urban and northern sparsely populated Nordic regions? Is there sufficient data in sparsely populated settings to develop and apply AI? How can sparsely populated Nordic (Arctic) regions absorb, contribute to and benefit from AI?

Both the Barents and the West Nordic region are geostrategically situated between the US and Russia – and increasingly China. How is the dual-use potential of AI-relevant data and technology to be handled, for example with regards to weather and ice-conditions? What forms of collaborations are likely to develop in this region throughout the coming decade?
Recommended textsMalmgren-Hansen, D., L. T. Pedersen, A. Aasbjerg Nielsen , M. Brandt Kreiner, R. Saldo, H. Skriver, J. Lavelle, J. Buus-Hinkler, & K. Harnvig Krane. “A Convolutional Neural Network Architecture for Sentinel-1 and AMSR2 Data Fusion.” IEEE Transactions on Geoscience and Remote Sensing (July 2020).
Sawyer, G., C. Oligschläger & N. Khabarov. “A Case Study Navigation through sea-ice in Greenland,” ESA Sentinels Benefits Study (March 2019).
Sukuvaara, T., K. Mäenpää, D. Stepanova & V. Karsisto, “Vehicular Networking Road Weather Information System Tailored for Arctic Winter Conditions,” International Journal of Communication Networks and Information Security (IJCNIS) Vol. 12, No. 2, August 2020.