Like discussed yesterday I would create an issue on the S-band processing. It seems some of these radars are processed with the wrong wave length. This is a plot of a timeseries of the wavelength for FR/NIM over time as reported in the vp files (note I have only downloaded data for the migration seasons):
This is a s-band radar the correct wavelength is thus about 10 cm ( http://eumetnet.eu/wp-content/themes/aeron-child/observations-programme/current-activities/opera/database/OPERA_Database/index.html ), compared to the 5 cm wave length for c-band.
Some how this thus changed to the correct value over time, I assume by the correct meta data being included.
I guess it would be best to reprocess this data to the correct values if possible. Otherwise at least a warning to users would be good.
Note currently I only here report results for FR/NIM but I assume it is more common to the other s-band radars. Just reading all vp files is kind of slow, so for large datasets I would maybe check the first file for each day.
Like discussed yesterday I would create an issue on the S-band processing. It seems some of these radars are processed with the wrong wave length. This is a plot of a timeseries of the wavelength for
FR/NIMover time as reported in the vp files (note I have only downloaded data for the migration seasons):This is a s-band radar the correct wavelength is thus about 10 cm ( http://eumetnet.eu/wp-content/themes/aeron-child/observations-programme/current-activities/opera/database/OPERA_Database/index.html ), compared to the 5 cm wave length for c-band.
Some how this thus changed to the correct value over time, I assume by the correct meta data being included.
I guess it would be best to reprocess this data to the correct values if possible. Otherwise at least a warning to users would be good.
Note currently I only here report results for
FR/NIMbut I assume it is more common to the other s-band radars. Just reading all vp files is kind of slow, so for large datasets I would maybe check the first file for each day.