CDOM products

[msantiago]

Hi,

I've been analyzing L1 to L3 MODIS Aqua ocean color data for CDOM products. My study area is in the Pacific Arctic Region (e.g, Bering and Chukchi seas). However, after I do the analysis in Python and SeaDAS, the following message is given by Python:
______
-W- /Users/seadas/ocssw/build/src/l2bin/dataday.c line 275: No valid latitudes found
Consider QC fail for file: /Volumes/SantiagoRS/Aqua/test_july2017/l1-l2output/A2017202233500.L2_SST

...look at the file though, as I might be lying...

====> running l3bin binary <=======

ascii_list is :  /Volumes/SantiagoRS/Aqua/test_july2017/l1-l2output/l2bin/ascii_bl2_list.txt
Does file exist?
True
L3BIN 4.20 (Apr 27 2016 10:27:28)
0 input files
_________

I have never seen the message "no valid latitudes found" before. So I went ahead and downloaded the ocean color L1 data for my study area again in case I had downloaded the data at different latitudes. I still get the same message, though. I have downloaded and analyzed data in July from previous years as recently as last night, and the png output files are created with no problem. So I'm inclined to believe that there may be a problem with the July 2017 data? 

Does anyone have any suggestions?

Any help will be appreciated!

Melishia Santiago

[gnwiii]

Your description of the problem does not provide enough detail for anyone trying to help to be able to duplicate the problem on another system.   How to report bugs effectively is also good advice when asking for help.

Another post discusses a similar problem with l2bin which occurred with a user-written shell script but not with the SeaDAS GUI.

It isn't clear from your post whether you also get the failure using the SeaDAS GUI or running the individual commands in a terminal.

The standard level-2 SST file, "works for me", using l2bin:
$ l2bin infile=A2017202233500.L2_LAC_SST.nc ofile=A2017202233500.L3b_LAC_SST.nc l3bprod=ALL resolve=4
L2BIN 4.0.9 (Apr 27 2016 10:22:37)
INTERP parameter disabled.
Single HDF input
Averaging: standard
Resolution: 4
Max Qual Allowed: 2
prod_avg:
flagusemask: 107599675
required: 0
NetCDF4 input file
total number of bins: 23761676
1970001 2038018 270
krow:     0 out of   4320  (-90.00 to -78.75) Wed Sep  6 15:12:54 2017
[...]
krow:  3510 out of   4320  ( 56.25 to  67.50) Wed Sep  6 15:12:54 2017
# bins in row group : 1097902
# filled bins       :   18981

krow:  3780 out of   4320  ( 67.50 to  78.75) Wed Sep  6 15:12:56 2017
# bins in row group :  676089
# filled bins       :    2147

krow:  4050 out of   4320  ( 78.75 to  90.00) Wed Sep  6 15:12:57 2017
krow: 4320 Wed Sep  6 15:12:57 2017
total_filled_bins: 21128
Writing Global Attributes
Freeing Dynamic Memory
Freeing L2 arrays


and then:

$ l3bin in=A2017202233500.L3b_LAC_SST.nc out=A2017202233500.L3b2_LAC_SST.nc out_parm=sst
L3BIN 4.20 (Apr 27 2016 10:27:28)
Single HDF input
irow:     0 of     4320 Wed Sep  6 15:22:40 2017
[...]
irow:  4000 of     4320 Wed Sep  6 15:22:41 2017

[msantiago]

Thanks you for your reply. Yes, the problem occurred with a shell script. I get the failure using the individual commands in a terminal. Also, I'm using a Mac running OSX 10.12, Python 2.7, and SeaDAS v7.4.

The MODIS Aqua files are L1 to process them to L3. I want to analyze CDOM index and SST products for the following area: latlon = 62S, -175W, 72N, -165E in July 2017.

I looked at the other post suggested below, and is it possible that I need to manually update the luts? 

Thanks again,
Melishia

[OB.DAACx - SeanBailey]

The warning message from l2bin indicated there was an issue with this file:
/Volumes/SantiagoRS/Aqua/test_july2017/l1-l2output/A2017202233500.L2_SST

Have you looked at that file?

BTW, l2bin did not fail, it just did not use that file in the binning process because it did not find sufficient geolocation information.

The next step (l3bin) did not work because there were zero files to process...or so it says:

> ====> running l3bin binary <=======
>
> ascii_list is :  /Volumes/SantiagoRS/Aqua/test_july2017/l1-l2output/l2bin/ascii_bl2_list.txt
> Does file exist?
> True
> L3BIN 4.20 (Apr 27 2016 10:27:28)
> 0 input files

This may be because the l2bin process did not produce an output.  The first step is to verify the inputs...

Sean

[msantiago]

Thanks Sean. Yes, I looked at that file in SeaDAS. But the following SeaDAS error pops up on my screen:

"An exception occurred:
Type: java.util.concurrent.ExecutionException
Message:java.lang.IllegalArgumentException: numElems is less than one

I also checked the other inputs in SeaDAS, and the same error occurs.

Thanks,
Melishia

[OB.DAACx - SeanBailey]

Well, that's the problem then.  The L2 files are either corrupt or have too few pixels.  Both cases will be a problem for the binner.

[msantiago]

Thanks Sean! Do I have any other options to analyze CDOM algorithms besides L3 data? Also, I understand my study area (e.g., Bering and Chukchi seas) is mostly cloudy during summer months. However, do you have any plans to reprocess the L2 so the data are accurate?

[OB.DAACx - SeanBailey]

There is nothing wrong with the products we produced, as the binner has no issues with them.  The files you are passing to the binner have a problem. 
You need to either  figure out why they are suspect or remove the suspect files from your binning process.

Sean

[msantiago]

Yes, that makes sense. I will remove the suspect files from my binning process. Thanks so much for your help!

Melishia