Ocean Color Forum - Not logged in
I have noticed that the coastline overlay on some or our SeaWiFS HRPT products acquired at Tromso, Norway (processed to Level 1A using SeaDAS 6.1 on a Linux machine) often stops (e.g. the continents are not always complete) in far northern latitudes, most notably around Greenland, Iceland and Svalbard.
Could this behaviour be a consequence of the shifting ice-sheets and an inability to consistently retrieve accurate geolocation information in that area or something more sinister, prehaps to do with an error in our acquisition and/or processing?
Thanks in advance for any help.
Thanks for the advice.
Having cross-referenced the problem with the forum post that you recommended, I can confirm that the elements.dat file is updated daily and there is only one version of SeaDAS installed on our machine.
I understood that the Timing Anomaly file is particular to SeaWiFS GAC/LAC products, so is not relevant to HRPT processing?? Nevertheless, I input the file into the processing stream and it appeared to make no difference to this specific problem.
Our processing completes successfully but in some products (like the example attached - circled in green) the coastline around Greenland/Iceland/Svalbard is incomplete - the remainder of the coastline across the rest of the product appears nominal. I considered that it could be possible that the dynamic nature of the ice-sheets at such far northern latitudes could be the cause but would I be correct to assume, based on your response, that this is not a familiar scenario in SeaWiFS HRPT products acquired in the far northern hemisphere and I should further investigate our processing stream for errors?
Thanks again for the assistance.
The problem has nothing to do with the dynamic nature of the ice sheet. The coastline database used by SeaDAS is fixed (i.e. unchanging).
I'll ask our navigation specialist if he has any ideas. As for the time anomalies, these affect ALL SeaWiFS data - HRPT is just a direct broadcast
version of LAC.
Alexa, could you provide a few dates on which you have seen this problem?
Some recent examples:
02/10/2010 (DoY 275) Orbit 70115
29/09/2010 (DoY 272) Orbit 70072
21/09/2010 (DoY 264) Orbit 69957
20/09/2010 (DoY 263) Orbit 69943
There appears to be valid geolocation information throughout the products but the coastline overlay is incomplete around Greenland/Iceland/Svalbard to varying degrees.
Thanks for the help
I have not been able to reproduce the problem with the GAC data (all we have for that region), but it does not cover the edges of the HRPT image where you are seeing the problem. If you could provide one of your files (e.g., on an FTP site), we may have more luck figuring this out.
Apologies for the delayed reply.
I shall provide you with the FTP details via email.