SeaWiFS GPS outage recovery

Use this Forum to find information on, or ask a question about, NASA Earth Science data.
Post Reply
Posts: 97
Joined: Fri Jan 30, 2004 9:11 am America/New_York
Answers: 0

SeaWiFS GPS outage recovery

by genecarlfeldman » Tue Mar 28, 2006 3:09 pm America/New_York

The message below from Fred Patt summarizes the current status of the SeaWiFS data that was collected during the recent outage of the onboard GPS. This issue inly effects the data between 3 - 26 March 2006.

The GPS outage on the OrbView-2 spacecraft ended at about 17:30 UTC on Sunday, March 26. The outage commenced on March 3, so the total period of operation without GPS is over 24 days.

The lack of GPS data has had a significant effect on SeaWiFS navigation during this period. The two sources of error are: the lack of accurate orbit information in the absence of GPS data, and the errors in the spacecraft time code, which uses the GPS output as its reference. The former source can be largely corrected by using the valid GPS data after tracking was restored to correct the propagation errors in the orbit model used in the SeaWiFS navigation software. This correction was made to the elements.dat file yesterday, and the renavigation of the affected Level-1A data is in progress.

The determination of the time code errors is a much more challenging problem. The errors changed regularly during the GPS outage. The time code glitches that occurred during GAC data collection can be analyzed, but information about glitches during the back orbit is not readily available. The time code errors will need to be estimated after the initial reprocessing of the data, by a combination of automated assessment (using island targets) and review of imagery using the Quality Control tools. Once these estimates have been made, they can be implemented in the processing system. More than one iteration may be required, and for some periods it may not be possible to recover navigation to acceptable accuracy.

The process is expected to require one to two weeks. We hope to have at least preliminary corrections by early next week.


Post Reply