Abstract: The boundary zone between sea ice and open water is an area where ocean, ice, and atmosphere meet and interact. To analyze sea ice changes over long time series, it is important to identify the sea ice boundary and to obtain data on parameters such as sea ice extent, area, and ice perimeter. Passive microwave imaging is the preferred method to obtain data for long-term monitoring of sea ice changes. Based on classical statistics and random set theory, three different average methods are proposed to extract monthly sea ice extent from daily passive microwave sea ice concentration data. The impact of the differences in these models for long-term analysis of sea ice changes is discussed.

Key words: sea ice margin, averaging method, random sets, Hausdorff distance, sea ice variation