Trouble reading FIDUCEO_FCDR_L15_MVIRI Easy nc files

[cbhramar]

Hi team,

I am trying to read MVIRI level 1.5 FIDUCEO files, EASY version. An example filename would be: 'FIDUCEO_FCDR_L15_MVIRI_MET7-57.0_201611110230_201611110300_EASY_v2.6_fv3.1.nc'

But I am facing issues with loading the VIS, WV, and IR channels from the file.

My code:

from satpy import Scene

scn = Scene(filenames=['FIDUCEO_FCDR_L15_MVIRI_MET7-57.0_201611110230_201611110300_EASY_v2.6_fv3.1.nc'],
	reader='mviri_l1b_fiduceo_nc')
scn.load(['VIS', 'WV', 'IR'])

This script produces the following errors:

....../lib/python3.8/site-packages/satpy/readers/mviri_l1b_fiduceo_nc.py:742: RuntimeWarning: Mean of empty slice
  sub_lonlat = np.nanmean(
....../lib/python3.8/site-packages/satpy/readers/mviri_l1b_fiduceo_nc.py:742: RuntimeWarning: Mean of empty slice
  sub_lonlat = np.nanmean(
....../lib/python3.8/site-packages/satpy/readers/mviri_l1b_fiduceo_nc.py:742: RuntimeWarning: Mean of empty slice
  sub_lonlat = np.nanmean(

Does this mean there are no values in any of the channels? I have also tried running it with an array of filenames instead of just one:

from satpy import Scene
import glob

files = glob.glob('FIDUCEO_FCDR_L15_MVIRI_MET7-57.0*.nc')
files.sort()

scn = Scene(filenames=files, reader='mviri_l1b_fiduceo_nc')
scn.load(['VIS', 'WV', 'IR'])

This script results in the same errors mentioned above multiple times.

Thanks in advance for your help.

[pnuu]

These are just warnings, not errors, and happen due to space pixels that are masked out. Have a look at the images, they should be ok.

scn.save_datasets()

[cbhramar]

Thanks a lot @pnuu!

I want to access toa_bidirectional_reflectance_vis values from these files, I believe they are stored in the channel named VIS on easy FCDR. is that correct?

Can you please also suggest the easiest method to extract the value at a given latitude/longitude, or a method to convert pixel values (x, y) to lat/lon?

Interestingly. more than 50% of the files at this particular date (2016-11-11) seem to be creating empty images on scn.save_dataset('VIS'). Is it somehow related to the uncertainty attached with retrieving the product ('u_independent_toa_bidirectional_reflectance')?

Thanks again