I have a "dataarray" which I am trying to reproject it using rioxarray. However after reprojection when I use xarray.to_netcdf, the saved file is a dataset where the "spatial_ref" coordinate converted to a variable. I am not sure if that is xarray or rioxarray.reprojection causes this behavior. Here are some codes that show the issue:
import xarray as xr
import rioxarray
from pyproj import CRS
lst = xr.open_dataset("lst.nc") # File which carries the original CRS
luc = xr.open_rasterio("luc.tif") # File with the desired projection system
file_to_reproject = xr.open_dataarray("myfile.nc") # File to reproject
cc_lst = CRS.from_cf(lst.crs.attrs) # Get the CRS
cc_luc = luc.rio.crs
file_to_reproject = file_to_reproject.rio.write_crs(cc_lst) # write crs
file_reprojected= file_to_reproject.rio.reproject(cc_luc) #reproject the file
print(file_reprojected)
<xarray.DataArray (season: 4, y: 4343, x: 4172)>
array([[[nan, nan, nan, ..., nan, nan, nan],
[nan, nan, nan, ..., nan, nan, nan]]])
Coordinates:
* x (x) float64 -3.382e+06 -3.381e+06 ... 4.817e+05 4.826e+05
* y (y) float64 5.361e+06 5.36e+06 ... 1.338e+06 1.337e+06
* season (season) object 'DJF' 'JJA' 'MAM' 'SON'
spatial_ref int64 0
# however after saving the file and reloading it I get a dataset where
# spatial_ref turned into variable
file_reprojected.to_netcdf("file_reprojected.nc")
ds= xr.open_dataset("file_reprojected.nc")
>>> print(ds)
<xarray.Dataset>
Dimensions: (season: 4, x: 4172, y: 4343)
Coordinates:
* x (x) float64 -3.382e+06 ... 4.826e+05
* y (y) float64 5.361e+06 5.36e+06 ... 1.337e+06
* season (season) object 'DJF' 'JJA' 'MAM' 'SON'
Data variables:
spatial_ref int64 ...
__xarray_dataarray_variable__ (season, y, x) float64 ...
# Here is the content of spatial_ref
>>> print(ds["spatial_ref"])
<xarray.DataArray 'spatial_ref' ()>
array(0)
Attributes: (12/19)
crs_wkt: PROJCS["unknown",GEOGCS["unknown",DATUM["...
semi_major_axis: 6378137.0
semi_minor_axis: 6356752.314140356
inverse_flattening: 298.257222101
reference_ellipsoid_name: GRS 1980
longitude_of_prime_meridian: 0.0
... ...
As you can see after reprojecting the dataarray and saving it the file turns into dataset with spatial_ref as a new variable.
I am just interested in georeferenced dataarray. How can I fix this?
Edit1
I think it might have something with this (here) issue
Edit2
I forgot to mention when I import the reprojected file is has no crs:
ds= xr.open_dataset("file_reprojected.nc")
ds.rio.crs # Retruns nothing
The issue was I did not set decode_coords="all" when I used xarray.open_dataarray. With the following everything looks ok:
file_to_reproject = xr.open_dataarray("myfile.nc",decode_coords="all")