Mosaic

Use this Forum to find information on, or ask a question about, NASA Earth Science data.
Post Reply
cortivo
Posts: 21
Joined: Thu Mar 20, 2014 9:42 am America/New_York
Answers: 0

Mosaic

by cortivo » Fri Sep 29, 2017 2:06 pm America/New_York

Hi.
The problem is: (step by step)

1) Images: A2003001173500.L2_LAC_OC.nc A2003005185000.L2_LAC_OC.nc A2003001191000.L2_LAC_OC.nc  A2003006175000.L2_LAC_OC.nc  A2003006175500.L2_LAC_OC.nc A2003002164000.L2_LAC_OC.nc  A2003007165500.L2_LAC_OC.nc A2003002181500.L2_LAC_OC.nc  A2003007170000.L2_LAC_OC.nc A2003002182000.L2_LAC_OC.nc  A2003007183500.L2_LAC_OC.nc A2003003172000.L2_LAC_OC.nc  A2003008174000.L2_LAC_OC.nc A2003003172500.L2_LAC_OC.nc 
A2003003190000.L2_LAC_OC.nc  A2003004180500.L2_LAC_OC.nc A2003005171000.L2_LAC_OC.nc A2003005184500.L2_LAC_OC.nc

2) Area: [37S 40S] x [60W 54W] (Argentina coast), pixel size square by 0.05° (the default option in seadas GUI)

There are differences in the chlor_a and chl_ocx when I perform the mosaic with Seadas GUI and by command line.

The command line used: gpt.sh -q 12 -e mosaico.xml -Ssource=`cat files.txt` -t mosaic.L3_OC.nc -f netCDF4-CF

files.txt is a text file that contains the name of the L2 files that must be used to create the mosaic

Am I doing something wrong?

Another question. I noted that where there are NO DATA seadas puts 0(zero) instead NaN...
How can I change it?

attachment 2

attachment 3


attachment 5

attachment 6

Filters:

OB SeaDAS - knowles
Subject Matter Expert
Subject Matter Expert
Posts: 285
Joined: Mon Apr 07, 2008 4:40 pm America/New_York
Answers: 0

Mosaic

by OB SeaDAS - knowles » Mon Oct 02, 2017 3:23 pm America/New_York

I get a match when running gpt and GUI mode (I just tested the first 6 files from your list)

It does look like this part of your command line isn't working as you are expecting:
-Ssource=`cat files.txt`

Try this command line instead so all the files get sent in directly on the command line:
gpt.sh -q 12 -e mosaico.xml `cat files.txt` -t mosaic.L3_OC.nc -f netCDF4-CF

Regarding the NaN question.  The concept of NaNs is triggered off of the count bands (for instance chlor_a is tied to chlor_a_count).  Using a valid pixel expression of "chlor_a_count > 0" for the chlor_a band is how this is handled in SeaDAS. 

Another way to merge level2 files is with the OCSSW components of SeaDAS (l2bin followed by l3mapgen).

Danny

cortivo
Posts: 21
Joined: Thu Mar 20, 2014 9:42 am America/New_York
Answers: 0

Mosaic

by cortivo » Mon Oct 02, 2017 3:52 pm America/New_York

Hi.
Thanks for your answer.
Using your command line suggestion it works!

Post Reply