PlumX Metrics
Embed PlumX Metrics

SVM-Based Sea Ice Extent Retrieval Using Multisource Scatterometer Measurements

Remote Sensing, ISSN: 2072-4292, Vol: 15, Issue: 6
2023
  • 3
    Citations
  • 0
    Usage
  • 3
    Captures
  • 1
    Mentions
  • 0
    Social Media
Metric Options:   Counts1 Year3 Year

Metrics Details

  • Citations
    3
    • Citation Indexes
      3
  • Captures
    3
  • Mentions
    1
    • News Mentions
      1
      • 1

Most Recent News

Data on Support Vector Machines Published by a Researcher at Nanjing University of Information Science and Technology (NUIST) (SVM-Based Sea Ice Extent Retrieval Using Multisource Scatterometer Measurements)

2023 APR 04 (NewsRx) -- By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News -- Investigators publish new report

Article Description

This study aims to explore the joint usage of multisource scatterometer measurements in polar sea water and ice discrimination. All radar backscatter measurements from current operating satellite scatterometers are considered, including the C-band ASCAT scatterometer on board the MetOp series satellites, the Ku-band scatterometer on board the HY-2B satellite (HSCAT), and the Ku-band scatterometer on board the CFOSAT satellite (CSCAT). By performing seven experiments that use the same support vector machine (SVM) classifier method but with different input data, we find that the SVM model with all available HSCAT, CSCAT, and ASCAT scatterometer data as inputs gives the best performance. In addition to the SVM outputs, we employ the image erosion/dilation techniques and area growth method to reduce misclassifications of sea water and ice. The sea ice extent obtained in this study shows a good agreement with the National Snow and Ice Data Center (NSIDC) sea ice concentration data from the years 2019 to 2021. More specifically, the sea ice areas are closer to the sea ice areas calculated using 15% as the threshold for NSIDC sea ice concentration data in both Arctic and Antarctic. The sea ice edges acquired by the multisource scatterometer show a close correlation with sea ice edges from the Sentinel-1 Synthetic Aperture Radar (SAR) images. In addition, we found that the coverage of multisource scatterometer data in a half-day is usually above 97%, and more importantly, the sea ice areas obtained on the basis of half-day and daily multisource scatterometer data are very close to each other. The presented work can serve as guidance on the usage of all available scatterometer measurements in sea ice monitoring.

Bibliographic Details

Provide Feedback

Have ideas for a new metric? Would you like to see something else here?Let us know