Hi,

I'm having a little trouble with real time Aqau processing using MODIS_L0_to_L1A_GEO.csh.  I'm calling it as follows from within a c-shell script:

modis_L0_to_L1A_GEO.csh $PDSFILE

From my understanding, calling the script this way should allow it to see if there are definitive attitude and ephemeris files available from ftp://oceans.gsfc.nasa.gov/MODISA/ATTEPH/, and if there are not it should fall back to using real-time data that is also tries to download.  However, in my case, perhaps because I'm processing the Aqua files in real-time, just after we receive the pass, the script fails to find either definitive or real time files.  And this stops my processing dead in its tracks.

I've attached output from my script below (sorry for the length).

It doesn't seem possible for the processing to go any further if neither real-time or definitive files are available - am I correct about this?  When should I expect real-time files to become available after a pass?

Thanks very much!
Ryan

Determining required definitive attitude and ephemeris files...
/home/xtuser/seadas/etc/modis_definitive_atteph.csh aqua 2005306164043 2005306165117

Checking local hard disk for required definitive attitude and ephemeris files...
Missing on hard disk: PM1ATTNR.P2005306.1600.*
Missing on hard disk: PM1EPHND.P2005306.1200.*

Attempting to download missing files...

*** TO CANCEL THESE TRANSFERS (AND ABORT GEOLOCATION PROCESSING) PRESS CTRL-\ ***
*** WARNING: IF YOU ABORT, MAKE SURE TO REMOVE THE PARTIALLY DOWNLOADED FILES ***

/home/xtuser/seadas/util/bin/wget --passive-ftp --tries=2 -q \
     -P/home/xtuser/seadas/data/modis/atteph/2005/306 \
     ftp://oceans.gsfc.nasa.gov/MODISA/ATTEPH/2005/306/PM1ATTNR.P2005306.1600.*

*** It appears the ftp site may not be hosting the requested definitive attitude file:
*** ftp://oceans.gsfc.nasa.gov/MODISA/ATTEPH/2005/306/PM1ATTNR.P2005306.1600.*

*Failed to determine/retrieve required definitive attitude and ephemeris files*
*Falling back to real-time attitude/ephemeris*

Determining required real-time attitude and ephemeris files...
/home/xtuser/seadas/etc/modis_predicted_atteph.csh 2005306164043 2005306165117

Checking local hard disk for required real-time attitude and ephemeris files...
Missing on hard disk: Predicted attitude file corresponding to pass start time.
Missing on hard disk: Predicted attitude file corresponding to pass stop time.
Missing on hard disk: Predicted ephemeris file corresponding to pass start time.
Missing on hard disk: Predicted ephemeris file corresponding to pass stop time.

Attempting to download missing files...

*** TO CANCEL THESE TRANSFERS (AND ABORT GEOLOCATION PROCESSING) PRESS CTRL-\ ***
*** WARNING: IF YOU ABORT, MAKE SURE TO REMOVE THE PARTIALLY DOWNLOADED FILES ***

*** No real-time attitude file corresponding to pass start time exists on ftp site. Exiting.

*Failed to determine/retrieve required real-time attitude and ephemeris files*

The required attitude and ephemeris files are not available!
Processing cannot proceed. Exiting.

Hey Ryan, yes your understanding of the way the script works is completely correct. For now I've assumed that each ground station is using its own GBAD processor to create real-time attitude and ephemeris files for each pass (using the L0 PDS file output from RT-STPS). RT-STPS and the GBAD processor are available at http://directreadout.gsfc.nasa.gov/index.cfm?section=downloads. We were considering the possibility of incorporating the GBAD processor into MODISL1DB and SeaDAS, but since most DB stations already have the software we felt it was unnecessary.

So to answer your first question, if you are trying to do real-time processing I'd suggest using the GBAD processor to create the real-time attitude and ephemeris files yourself, and specify these files on the command-line e.g.:

modis_L0_to_L1A_GEO.csh $PDSFILE -a1 attitude_file1 -a2 attitude_file2 -e1 ephemeris_file1 -e2 ephemeris_file2

And when should you expect real-time files to become available on our ftp server? NOAA pushes this data to us every orbit, so normally the data would be available within 2 hours of your pass, barring a backup at NOAA or with our production system.

Also remember that if you reprocess your MODIS data when the definitive att/eph files become available, your geolocation will be improved, though usually only slightly.

Ok - great.  Thanks for the clarification!

Ryan

Hi,
just a question about this problem of the definitive ATT and EPH. I'm processing L0 Aqua files using the temporarily ATT and EPH. Is it correct or the output L1A files have some problems?

thanks a lot
Simone

Simone,

There will be no problem with the L1A file.  The GEO file may be very slightly different depending on whether you use the definitive or the predicted attitude and ephemeris data, but the difference is really so insignificant that it can be considered negligible.

Mark

Thank you very much.
However, I'm trying to use GBAD processor to create real-time attitude and ephemeris files but I'm still having some problems.
I'm trying to obtain ATT and EPH from a L0 file downloaded from ftp://oceans.gsfc.nasa.gov but the GBAD processor responds: "Packet File Empty". Can I obtain the ATT and EPH with GBAD processor from this L0 files? or should I do something before?
Can you help me?

thanks a lot

s.colella,

I'm assuming you are using Aqua L0 files?  If so, I don't believe the GBAD processor will give you any predicted attitude or ephemeris files because Aqua does not have that data embedded into it internally like the Terra L0 files.  That's probably why it isn't giving you anything.  You can get attitude/ephemeris files from here:

ftp://oceans.gsfc.nasa.gov/MODISA/ATTEPH/

Regards, Mark

ok. thank you very much

Just to make sure you're not creating work for yourself, do you know that as long as you have an internet connection, you don't need to worry about manually retrieving the attitude and ephemeris files. Our software will do all of this automatically in the GUI or on the command-line. For example, to process the most recent Level-0 file on our site (MOD00.P2007275.0755_1.PDS) you could simply execute the following command directly on the  Unix command-line:

modis_L0_to_L1A_GEO.csh  MOD00.P2007275.0755_1.PDS

The processing code will automatically look for definitive attitude and ephemeris on your hard disk and then on our ftp site. If no definitive is available as is the case with NRT data, the code will then fallback to use real-time attitude and ephemeris, first checking your hard disk and then downloading the files from our site... so you shouldn't need to do any of this manually.

The GBAD processor needs a separate Level 0 file containing just the GBAD packets (APID 957). Examples of these files can be found at

ftp://is.sci.gsfc.nasa.gov/gsfcdata/aqua/gbad/
P1540957*01.PDS

The corresponding MODIS Level 0 files (APID 64) are available at

ftp://is.sci.gsfc.nasa.gov/gsfcdata/aqua/modis/level0
P1540064*01.PDS

Cheers,
Liam.

Regarding the difference in Aqua MODIS geolocation when using definitive vs. predicted attitude and ephemeris:

Normally the difference is less than 100 meters in the final geolocation product. The Rapid Response project tracks the difference every day, as shown on this website.

However, sometimes the prediction accuracy degrades faster than normal, and the error can increase to several hundred meters, as shown in this example.

If your MODIS application only uses 1000 meter resolution data, then these differences are probably not important. However if you are using the MODIS 250 meter resolution bands to detect changes over time, especially over land surfaces, then you should reprocess your data when the definitive attitude and ephemeris data become available.

Cheers,
Liam.