Approach

“AI for cold regions” (AI-CORE) addresses two urgent research challenges (understanding the impact of climate change on the cryosphere; applying/developing AI methods to process versatile and extensive input data) and combines the expertise of Germany’s leading experts in these fields to overcome them. Permafrost degradation, glacier/ice sheet shrinkage, sea ice recession, and decreasing spring snowpack have been observed and are projected to continue their decline in the future, leading to changes in runoff, rising sea level, increased coastal erosions, changes in the planetary radiation budget, and impacts on flora, fauna, and human livelihoods and economies. In AI-CORE, four geo-scientific use cases were defined where cryosphere research will be linked with the development and implementation of AI. This will allow the processing and analysis of the vastly available data originating from in situ, modelling, and earth observation with state-of-the-art methods. Finally, by implementing these methods on the distributed processing infrastructures of the project consortium, “best-case” solutions will be identified for applying AI methods to geo-scientific research questions.

Geo-Scientific Use Cases (UC)

• UC1: AI for change pattern identification of outlet glaciers in Greenland
• UC2: AI for object identification in permafrost areas
• UC3: AI for edge detection of calving fronts of glaciers/ice shelves in Antarctica based on variable input layers
• UC4: Firn line detection and monitoring: The glacier mass balance indicator