An approach to automated mapping of the movement of arctic sea ice by statistical examination of NOAA-1 satellite infrared data is discussed. The approach is based on the empirically observed relationship that airborne or satellite radiant temperature measurements of sea ice are proportional to the thickness of the ice and have normal distributions, and that the areas beneath the distribution curves are proportional to the surficial area of the pack ice types from which they were derived. Data which show the variations of radiant temperature distributions with time of year and with cloud cover are examined. The means of radiant temperature distribution curves that correspond to multi-year ice for several different dates appear to correlate with the temperatures for equilibrium ice calculated by the Maykut-Untersteiner Thermodynamic Ice Model for these same dates. Data are also presented to show the effects of large areas of open water on the radiant temperature distributions of pack ice. The relationships, if proved empirically with more data and by further model analysis, can provide the basis for a relatively simple algorithm for sea ice mapping using satellite infrared data.

Media Info

  • Features: References;
  • Pagination: p. 243-267

Subject/Index Terms

Filing Info

  • Accession Number: 00151965
  • Record Type: Publication
  • Source Agency: Engineering Index
  • Files: TRIS
  • Created Date: Apr 13 1977 12:00AM