SeaDAS inconsistency

[giovana.damiati]

Hi,

I'm using SeaDAS to analyze seawater turbidity data (using kd490), and I tried to open 2020 and 2021 data recently but it didn't work, basically nothing happens, the files just don't open. In addition, some specific points or "pins" fall on land or on water in different months, even when I use the same coordinates to the same points (I tested the same coordinates on Google Earth, and the coordinates are correct and fall on water).

Thank you,

Giovana

[OB SeaDAS - xuanyang02]

It might be helpful if you let us know what files you have trouble to open. Are they MODIS or VIIRS files? L2, L3b or L3m?

SeaDAS 8.0.0 is a beta version, we do have some known issues and the fix will be in the next release.

In the meantime, you can try SeaDAS-7.5.3: https://oceandata.sci.gsfc.nasa.gov/Sea ... ler/7.5.3/.

Here are two posts on reading L3b and L3m files for your reference:
viewtopic.php?f=7&t=2310&sid=f275b9902d ... 724f0e95b5

viewtopic.php?f=7&t=2223&p=7860&sid=f27 ... 95b5#p7860

[giovana.damiati]

They are MODIS-Aqua....And all files from the years 2020 and 2021 I have trouble to open.

[OB.DAACx - SeanBailey]

As to the apparent navigation issue, I suspect what you are seeing is missing data not that the pixel is on land in some L3 files. The L3 grid is consistent, even at monthly scales, it is quite possible for bins (and subsequently mapped pixels) to not be filled.

Sean

[giovana.damiati]

About the data I’m using, they’re actually falling on land, because the coordinates are reliable, and are the same for all the months that I'm analysing. Unfortunately there's no way to attach the images around here, as it would help to better clarify what is happening.

Giovana

[OB SeaDAS - xuanyang02]

You can attach the image here by click "Attachments" tab and "Add files" while you edit your post.

I can open L2 file A2020283185000.L2_LAC_OC.nc
but not A20202752020305.L3m_MO_KD490_Kd_490_4km.nc due to null global attribute value, which is a known issue.

[giovana.damiati]

I'm sorry, I hadn't seen the option to attach files before. But here are the images that I mentioned.
If you examine attached figure 1 (Modis-aqua monthly data for Feb/2019, using 4 km resolution), you will notice that those two coordinates fall on land. For the next month (Mar/2019, see figure 2), the very same coordinates fall on water. Using Google Earth, both coordinated fall on water (as it is expected, figure 3).

[giovana.damiati]

To xuanyang02,

I'm not aware of this problem, could you explain it to me? Is it related to version 8.0 of SeaDAS? And how could we solve this? because it’s exactly the type of file I’m using (L3m).

[OB SeaDAS - xuanyang02]

If you do "ncdump -h A20202752020305.L3m_MO_KD490_Kd_490_4km.nc", you will find there are a few global attributes with null value. For example, "ofile", "width" and etc in "input_parameters" group. There is a bug in SeaDAS 8.0.0 that reading such a file will fail. The fix is in the next release. In the meantime, you can try SeaDAS-7.5.3 to read L3m files. You can download it from https://oceandata.sci.gsfc.nasa.gov/Sea ... ler/7.5.3/

Thanks for sharing the above images. In figure 1, we just have missing data at the two pixels you pinned. It does not mean these two pixels are on land.

Bing

[giovana.damiati]

Ok, I’ll try to do what you’ve recommended. Thank you for your help!