Space pixels masking optional in ahi_hsd but not hrit_jma reader

[storm-345016]

Hi,

At the moment it is possible to use a kwarg with the ahi_hsd reader to disable the masking of the space pixels (mask_space).

The same feature doesn't exist when working from hrit; for example this prevents seeing the Moon in some recent Himawari-9 imagery that comes through EUMETCast in the HRIT format.

Is this a possible feature to carry across? I've tried knocking it out with my limited Python abilities in the reader but to no avail.

Many thanks :)

[djhoese]

This might be a question for @simonrp84 who I think added the HSD support originally. I'm not sure why it wasn't also done for the HRIT.

Here's the line that would need a if self._mask_space: added:

satpy/satpy/readers/hrit_jma.py

Line 391 in 2dba225

res = self._mask_space(self.calibrate(res, key["calibration"]))

But we'd also need tests and the proper handling of the keyword argument in the __init__ method.

[simonrp84]

Sorry, this one probably isn't for me. I've never used the HRIT data and don't have access to it. I may be able to help but someone else is probably in a better position to do so...

[djhoese]

But did you add the kwarg functionality to the HSD reader?

[pnuu]

I'm quite certain that the space pixels are already masked out before Eumetcast dissemination.

[storm-345016]

I'm quite certain that the space pixels are already masked out before Eumetcast dissemination.

They do seem to be in what is disseminated (if somewhat surprisingly for bandwidth); see the following:

https://groups.io/g/MSG-1/message/34915

[simonrp84]

But did you add the kwarg functionality to the HSD reader?

No, I didn't.

[djhoese]

Ah sorry. Ok so I guess this was originally added by @sfinkens in #524. There are a couple related issues. I guess in my opinion if masking is needed to match HSD (and isn't pre-applied by EUMETSAT) then having a flag to match HSD is welcome. Whoever wants to implement it feel free to make a PR (with tests).

Edit: A quick check if the masking is needed would be to comment out the line I mentioned above and see if there is any useful data in the space pixels.

[sfinkens]

I'll put that on my list for the PCW (in two weeks)