pytroll / satpy

Python package for earth-observing satellite data processing
http://satpy.readthedocs.org/en/latest/
GNU General Public License v3.0
1.05k stars 289 forks source link

iber projection lost in the North Pacific #1240

Open gerritholl opened 4 years ago

gerritholl commented 4 years ago

Describe the bug

According to its description the satpy built-in iber projection should cover "North half of the Iberian Peninsula and the Gulf of Biscay image 0 degrees". However, as defined it is somewhere in the remote North Pacific.

To Reproduce

import pyresample
ar = pyresample.load_area("/home/gholl/checkouts/satpy/satpy/etc/areas.yaml",
        "iber")
(lons, lats) = ar.get_lonlats()
print(lons.min(), lons.mean(), lons.max(), lats.min(), lats.mean(), lats.max())

Expected behavior

I expect the range of longitudes and latitudes to correspond to the described area, somewhere near the meridian, perhaps a little west of it.

Actual results

In reality the area covers longitude 166 to 180 degrees, seemingly shifted by 180 degrees compared to where it should be expected:

166.3308164211224 173.2332707794299 179.85396117722175 39.629863255964146 42.60892251305624 45.41767241192362

Environment Info:

djhoese commented 4 years ago

My guess is the description was a copy/paste error from the sve area right above iber:

https://github.com/pytroll/satpy/blob/9261a1c590dc13b70e24c7b7a765b996663d59a1/satpy/etc/areas.yaml#L55-L77

Actually...nevermind that description seems more accurate for iber than sve. I'm a little confused by the projection for iber. It looks like it should be a UTM zone but is missing the actual UTM zone number. PROJ doesn't even define a default: https://proj.org/operations/projections/utm.html