Pixel extraction from many SENTINEL-3 products in SNAP

[daviguima]

Thanks so much for all the tutorials and also for the xml template @marpet

through some trial and error I’ve managed to get snappy up and running system-wide alongside GDAL:

Ps. I’ve also had this SEVERE message in the console but after some searching I saw this guy saying it was ok:

Warning message while resampling the Sentinel-2 data

No, you don’t have to worry about those warning messages, it is caused by some evolutions on the product format, but it is read properly by SNAP. On the other hand, the SEVERE message about LD_LIBRARY_PATH is neither problematic (see https://github.com/senbox-org/snap-engine/issues/80)

So…
Is there a way of making the gpt/gpf or snappy to simulate this behavior of the GUI?:

Region of interest (ROI)

Hi Carlo, right-click in the view and select “Export Mask Pixels” The mask severs as a ROI. You can create masks by many different ways. You can load a shapefile, draw polygon or use Band Maths. Marco

I did it in SNAP Desktop GUI and got this file: S3A_OL_2_WFR____20190309T141223_20190309T141523_20190310T211622_0179_042_167_3060_MAR_O_NT_002_manacapuru_Mask.txt (3.1 MB)

@marpet

Unfortunately, this approach does not export only the pixels covered by your shapefile but always the bounding box. Especially when exporting to CSV/ASCII format the amount of data is huge.

And I didn’t tested it to see where the lon/lat pixel data will fall on a map, was planning on doing it later today, but… My real question is: does that mean that the pixels that I extracted this way in the above file are actually everything inside the bounding box rectangle of the shapefile or only the ones actually inside the shapefile?

Sorry to keep bothering you guys with this, you already did such a great job by making all these great tools and data freely available to us.

Thanks again and keep safe!
David