Sea ice classification on different spatial scales for operational and scientific use
With the possibility of acquiring satellite radar images at different spatial resolutions and swath widths (i. e. at different scales) there is a need for selecting the optimal image product for the analysis of sea ice conditions. Both resolution and areal coverage influence the discrimination (and...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Book Part |
| Language: | unknown |
| Published: |
European Space Agency (ESA)
2013
|
| Subjects: | |
| Online Access: | https://epic.awi.de/id/eprint/33756/ https://epic.awi.de/id/eprint/33756/1/846659dierking.pdf http://www.spacebooks-online.com/product_info.php?products_id=17572 https://hdl.handle.net/10013/epic.44280 https://hdl.handle.net/10013/epic.44280.d001 |
| Summary: | With the possibility of acquiring satellite radar images at different spatial resolutions and swath widths (i. e. at different scales) there is a need for selecting the optimal image product for the analysis of sea ice conditions. Both resolution and areal coverage influence the discrimination (and definition) of different sea ice zones and structures. A wide coverage is usually linked with a relatively coarse spatial resolution, whereas high-resolution products are restricted to narrow swath widths. Examples are presented which demonstrate pros and cons when balancing both parameters for the analysis of sea ice conditions. It is shown that X- and C-band images reveal similar signature variations (except over thin ice) and are hence well suited for combining images taken at different scales. Cases were observed for which strongly averaged intensity images are beneficial for discriminating different ice zones. For making full use of texture analysis, higher resolution products are preferable. |
|---|