Automated Processing of LIDAR Point Clouds for Sea Ice Analysis

Abstract The number of ships equipped with advanced sensors has grown significantly, with some now including LiDAR systems. This could be particularly useful during navigation in ice, where assessing severity of sea ice conditions is crucial for safe path planning. While a LiDAR system can be used to create multimodal maps of the surrounding sea ice conditions, there are currently no established methods to process this information into practical and usable parameters, leaving much of its potential underutilized. We address this gap by presenting an algorithm that can automatically extract important sea ice parameters such as total ice concentration, freeboard, ridge frequency, and degree of ice deformation. The algorithm is tested on a dataset collected with a shipborne instrumentation system mounted on RV Kronprins Haakon during a summer campaign in the Fram Strait in 2024. We demostrate applicability of the algorithm using digital multimodal ice maps created from a shipborne sea ice scanning system [1] as an example.