Transforms from the same file are different

Asked Oct 12 '21 at 23:35

Active Oct 13 '21 at 00:30

Viewed 62 times

Reading a NetCDF file in two ways:

src = rasterio.open(f'NetCDF:{url}:Rad', 'r') and doing src.transform gives:

Affine(2004.017315487541, 0.0, -5434894.700982174, 0.0, -2004.017315487541, 5434894.700982174)

Whereas reading the same file as

ncfile = xarray.open_dataset(str(url), engine='netcdf4', decode_coords="all") and doing ncfile['Rad'].rio.transform() gives:

Affine(5.599999810285109e-05, 0.0, -0.15187199485493216, 0.0, -5.599999810285109e-05, 0.15187199485493216)

Why is it so?

edited Oct 13 '21 at 00:30

Vince

asked Oct 12 '21 at 23:35

Coddy

Did you try rioxarray.open_rasterio? – snowman2 Oct 12 '21 at 23:47
Are you able to provide the file? – snowman2 Oct 12 '21 at 23:47
rio.open_rasterio(f'NetCDF:{url}:Rad', 'r').rio.transform() gives the same as 1 – Coddy Oct 12 '21 at 23:51
@snowman2 here you go https://drive.google.com/file/d/1MLJKvrcTaxbfJ7gRd-H4kUibOIdlYtAk/view?usp=sharing – Coddy Oct 12 '21 at 23:56
Does this answer your question? How do I add projection to this NetCDF file? (Satellite) – snowman2 Oct 13 '21 at 02:17
I looked at the file, and it seems to be a GOES 16 netCDF file similar to the one mentioned above. I think GDAL handles some conversions automatically for you. – snowman2 Oct 13 '21 at 02:18
but I am using rasterio, xarray and rioxarray – Coddy Oct 13 '21 at 02:26
I don't think it answers the question. – Coddy Oct 13 '21 at 02:26
rioxarray uses rasterio, and rasterio is a python wrapper for GDAL. – snowman2 Oct 13 '21 at 02:39
So why two different values then? – Coddy Oct 13 '21 at 02:40
I believe the netCDF driver in GDAL handles the conversion in the issue above for you. – snowman2 Oct 13 '21 at 02:57

0 Answers0