Estimation of Sea-ice Concentration along the Northern Sea Route Using Sentinel-1 Data

Abstract

The usage of the northeast and northwest passages in the Arctic Sea has increased much over the years. The Polar Code adopted by International Maritime Organization (IMO) for ships operation in polar waters is expected to enter into force on 1st January 2017. For the navigation safety in the polar areas, a real-time service of updated sea-ice distribution map is essential during traveling long distance. Sentinel-1 constellation of two satellites (Sentinel-1A and Sentinel-1B) is capable of providing C-band Synthetic Aperture Radar (SAR) images of the Arctic region up to 28 times a day (175 orbits in 12 days for one satellite) with varied spatial resolutions from 5m to 100m within 3 hours. Therefore, for automatic detection of sea-ice, Sennitel-1 data were acquired from June 1st 2016 to June 30th 2016 which is the period of sea-ice melting. In this study, Level-1 Ground Range Detected (GRD) data were used after radiometric and geometric calibrations. First, the ocean region was separated from land region using General Bathymetric Chart of the Oceans (GEBCO 2014) data. After the land masking, satellite image quality were enhanced by power-law scaling technique to distribute between sea-ice and ocean. Daily SIC distribution was produced along the Northern Sea Route (NSR) and the results were verified by comparing with a AMSR L2 SIC data which was generated using dynamic composite algorithm developed by Korea Institute ofn 1st January 2017. For the navigation safety in the polar areas, a real-time service of updated sea-ice distribution map is essential during traveling long distance. Sentinel-1 constellation of two satellites (Sentinel-1A and Sentinel-1B) is capable of providing C-band Synthetic Aperture Radar (SAR) images of the Arctic region up to 28 times a day (175 orbits in 12 days for one satellite) with varied spatial resolutions from 5m to 100m within 3 hours. Therefore, for automatic detection of sea-ice, Sennitel-1 data were acquired from June 1st 2016 to June 30th 2016 which is the period of sea-ice melting. In this study, Level-1 Ground Range Detected (GRD) data were used after radiometric and geometric calibrations. First, the ocean region was separated from land region using General Bathymetric Chart of the Oceans (GEBCO 2014) data. After the land masking, satellite image quality were enhanced by power-law scaling technique to distribute between sea-ice and ocean. Daily SIC distribution was produced along the Northern Sea Route (NSR) and the results were verified by comparing with a AMSR L2 SIC data which was generated using dynamic composite algorithm developed by Korea Institute of