Ice rose diagrams for probabilistic characterization of the ice drift behavior in the Beaufort Sea

The purpose of the present paper is to investigate the viability of the so-called ice rose diagram, which is a graphic tool to provide a succinct view of the ice drift information such as speed and direction of ice drift events in terms of their relative distribution at a local observation station to support the dynamic ice-structure interaction in cold regions. The ice drift data is collected by using local subsurface measurements based on acoustic Doppler current profilers (ADCP) in the Beaufort Sea during 2006-2017. Probabilistic analyses are performed in order to characterize the time-series data from these ADCP measurements. Both the ice drift speed and direction are treated as random variables for which analytical probability density functions are fitted. It is found that sea ice in the Beaufort Sea tends to drift at a lower speed during the winter season (i.e. the growth season) than during the summer season (i.e. the melting season). The Weibull distribution provides the most appropriate fit to the data of ice drift speeds during the growth season. For the melting season, a mixed distribution provides the most appropriate fit for the drift speed. Regarding directionality of the ice drift, a mixed Von Mises distribution is applied in order to represent its statistical properties. In this work, the relationship between ice drift speed and wind speed is also studied. It is found that the magnitude of the ice drift speed is approximately 2.5% of the wind speed during the winter season.