satpy - SEVIRI night fog RGB recipe issue

97 views
Skip to first unread message

Pedro Miranda

unread,
Jun 24, 2020, 8:44:07 PM6/24/20
to pytroll
Hello. I'm from the Portuguese Air Force Meteorology Centre and we're starting to use satpy to generate images from MSG. It's a wonderful tool!
However, I've noticed that night fog RGB appearance was weird, and different from images generated in other sources, namely the green component - with some exceptions, low clouds were not showing up in a greenish tone as supposed.

So, I looked into in /etc/composites/seviri.yaml and in /etc/enhacements/generic.yaml.  The night_fog composition is not quite as the recipe from Eumetsat.

In generic.yaml, the range for the green component (0 to 6K) is different from EUMETSAT recommendation (0 to 10K), and a gamma of 2 is applied, when Eumetsat recommends gamma 1.
However, this wasn't the main cause for the lack of green.
In seviri.yaml, the IR3.9 is CO2 corrected, but for this it would be necessary to "tune" the temperature ranges for the green color down (maybe -5K to 5 K?). Anyway, EUMETSAT does not recommend any kind of CO2 correction here.
This was the main reason for the lack of green!

Changed both on my local version to match EUMETSAT recipe and now it looks good.

Maybe these changes should be included on satpy. I confess I'm still getting used to GitHub, so I'm not yet sure on how to suggest this changes directly there.

Best regards,

Pedro Miranda

Raspaud Martin

unread,
Jun 26, 2020, 3:08:41 AM6/26/20
to pyt...@googlegroups.com
Hi Pedro, 

Thank your for bringing this issue to our attention! And nice detective work you did here :) 

You are indeed right that the night fog composite in SatPy doesn't follow the night microphysics recipe from the document you mention, I think it's an old recipe that somehow made its way to satpy through legacy code, sorry for that. 
However, we do have a night_microphysics composite that you could try out and that should follows the recipe you linked to. 

Keep us updated! 

Best regards, 
Martin 
--
You received this message because you are subscribed to the Google Groups "pytroll" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pytroll+u...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/pytroll/8205b6dc-f8d6-4a62-b016-3e3090a0d605o%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages