Machine-learning for mapping permafrost landscape dynamics

Speaker: Ingmar Nitze, AWI

The Arctic is warming at an unprecedented pace, which leads to rapid landscape changes, particularly in the terrestrial Arctic which is largely underlain by permafrost (perennially frozen soils). These landscape dynamics have strong impacts on local (geohazards, infrastructure) to global (climate system) scales. Here we will present an overview of current studies and advances in mapping landscape dynamics in Arctic permafrost using remote sensing (satellites, aerial, UAV) imagery and machine/deep-learning. We will particularly dive into the data scientific challenges, such as data acquisition, diversity of input data, data quality, spatial relations, and temporal dynamics, which are typical for geospatial data and particularly polar science.