Back to search

SSF-Svalbard Science Forum

FEATURE DETECTION IN GLACIAL CHANNELS USING MACHINE LEARNING, RIS ID 11603

Awarded: NOK 83,091

For my master thesis I will be using machine learning for detecting morphological features to improve geometric reconstructions of glacial channels based on data collected from drifters. The data will be gathered from channels on the Kongsvegen glacier during a field campaign in 2021. There are two drifter platforms which will be deployed, one using GNSS enabled surface drifters, measuring position and velocity, and the second submersible drifters measuring pressure, acceleration and magnetic fields. The drifters will be deployed by hand into channels and be captured furter downstream. The project is motivated by the need for an increased spatial and temporal scale of these kind of measurement as well as the need for large amounts of reference data for training machine learning algorithms. The project will build upon, and refine, previous work done on the drifter platform in 2018, 2019 and 2020. In order to reduce the cost of logistics and the environmental impact of the project there will be cooperation with a project monitoring the glacier lake outburst flood of Setevatnet (RIS ID 11572), which is a lake that drains through the subglacial system of Kongsvegen.

Funding scheme:

SSF-Svalbard Science Forum