Operational Sea Ice Mapping in the Arctic Sea Route Using Satellite Data

Abstract

For research and monitoring of sea ice in the Arctic region, satellite data obtained from an optical sensor, radiometer, and synthetic aperture radar (SAR) was used to response the Polar Code enacted on January 1, 2017, by the International Maritime Organization. We have worked in the fields of sea ice monitoring and its system development (Yang et al., 2016 Yang et al., 2018 Hong, Bae, and Yang, 2018). At the ISRS presentation, we would like to introduce the results in the Arctic Sea. In this system, sea ice concentration (SIC) from Advanced Microwave Scanning Radiometer 2 (AMSR2) has been reproduced to serve the information over 5 times a day using a dynamic composite technique (Yang et al., 2016). It became to receive in near real-time but still low-resolution. Hence, two additional techniques have recently been adapted to obtain detailed sea ice information. First is the fusion of SIC and sea ice extent acquired from Moderate Resolution Imaging Spectroradiometer (MODIS) of 1 km resolution (Yang et al., 2018), and second is the discrimination of sea ice area from the high-resolution SAR image using support vector machine which is one of the techniques of machine learning (Hong, Bae, and Yang, 2018). In this paper, the sea ice information acquired by three kinds of method is discussed in an operational perspective. Maritime Organization. We have worked in the fields of sea ice monitoring and its system development (Yang et al., 2016 Yang et al., 2018 Hong, Bae, and Yang, 2018). At the ISRS presentation, we would like to introduce the results in the Arctic Sea. In this system, sea ice concentration (SIC) from Advanced Microwave Scanning Radiometer 2 (AMSR2) has been reproduced to serve the information over 5 times a day using a dynamic composite technique (Yang et al., 2016). It became to receive in near real-time but still low-resolution. Hence, two additional techniques have recently been adapted to obtain detailed sea ice information. First is the fusion of SIC and sea ice extent acquired from Moderate Resolution Imaging Spectroradiometer (MODIS) of 1 km resolution (Yang et al., 2018), and second is the discrimination of sea ice area from the high-resolution SAR image using support vector machine which is one of the techniques of machine learning (Hong, Bae, and Yang, 2018). In this paper, the sea ice information acquired by three kinds of method is discussed in an operational perspective.