[GRASS-user] Cannot read NetCDF with 'semi_major_axis' or 'inverse_flattening'

[Ken Mankoff]

Hi Stefan,

Thanks for the hint. I often use '-o'. In this case after discussion with
the data provider it turns out that it was not EPSG:3413 but instead some
custom projection, even though part of the NetCDF file reported EPSG:3413
in the metadata.

Unfortunately r.import does not work. I get the errors below. Given the
0..3..6..9... it appears to import something, but no raster exists when the
command finishes.

r.import input=NetCDF:file.nc:surface output=surface
WARNING: Datum <unnamed> not recognised by GRASS and no parameters found
WARNING: Projection of dataset does not appear to match current location.

Location PROJ_INFO is:
name: WGS 84 / NSIDC Sea Ice Polar Stereographic North
datum: wgs84
ellps: wgs84
proj: stere
lat_0: 90
lat_ts: 70
lon_0: -45
x_0: 0
y_0: 0
no_defs: defined
init: EPSG:3413

Dataset PROJ_INFO is:
name: unnamed
a: 6378273
es: 0.006693883000000017
proj: stere
lat_0: 90
lat_ts: 70
lon_0: -45
x_0: 0
y_0: 0
no_defs: defined

ERROR: datum

WARNING: Datum <unnamed> not recognised by GRASS and no parameters found
WARNING: Datum <unnamed> not recognised by GRASS and no parameters found
Importing raster map <surface>...
0..3..6..9..12..15..18..21..24..27..30..33..36..39..42..45..48..51..54..57..60..63..66..69..72..75..78..81..84..87..90..93..96..99..100
ERROR: Unable to open element file <windows> for
<tmp.r.import.3257873 at PERMANENT>
ERROR: Unable to get reprojected map extent


  -k.



On Sat, Jul 24, 2021 at 2:19 PM Stefan Blumentrath <
Stefan.Blumentrath at nina.no> wrote:

> Hei Ken,
>
> Did you try the -o flag ("Override projection check (use current
> location's projection)
> Assume that the dataset has same projection as the current location")?
> https://grass.osgeo.org/grass78/manuals/r.external.html
> As the error message suggests?
>
> If you are sure that projections in fact match, a mismatch in CRS
> description can be overridden that way, without modifying the input file.
>
> Alternatively, you can use r.import to reproject while converting to GRASS
> format or you create a new location...
>
> Cheers
> Stefan
>
>
> -----Original Message-----
> From: grass-user <grass-user-bounces at lists.osgeo.org> On Behalf Of Ken
> Mankoff
> Sent: fredag 23. juli 2021 19:33
> To: GRASS user list <grass-user at lists.osgeo.org>
> Subject: [GRASS-user] Cannot read NetCDF with 'semi_major_axis' or
> 'inverse_flattening'
>
> Hello list,
>
> I have been given an updated NetCDF file - everything is the same as the
> old version except there are two new attributes in the 'mapping' variable:
>
>         char mapping ;
>                 mapping:geoid = "eigen-6c4" ;
>                 mapping:grid_mapping_name = "polar_stereographic" ;
>                 mapping:latitude_of_projection_origin = 90. ;
>                 mapping:standard_parallel = 70. ;
>                 mapping:straight_vertical_longitude_from_pole = -45. ;
>                 mapping:semi_major_axis = 6378273. ;
>                 mapping:inverse_flattening = 298.27940504282 ;
>                 mapping:false_easting = 0. ;
>                 mapping:false_northing = 0. ;
>                 mapping:_Storage = "contiguous" ;
>
>
> Specifically:
>                 mapping:semi_major_axis = 6378273. ;
>                 mapping:inverse_flattening = 298.27940504282 ;
>
> When I try to load this file with:
>
> r.external source=netCDF:file.nc:surface output=surface
>
> I get this error:
>
> WARNING: Datum <unnamed> not recognised by GRASS and no parameters found
> ERROR: Projection of dataset does not appear to match current location.
>
> Location PROJ_INFO is:
> name: WGS 84 / NSIDC Sea Ice Polar Stereographic North
> datum: wgs84
> ellps: wgs84
> proj: stere
> lat_0: 90
> lat_ts: 70
> lon_0: -45
> x_0: 0
> y_0: 0
> no_defs: defined
> init: EPSG:3413
>
> Dataset PROJ_INFO is:
> name: unnamed
> a: 6378273
> es: 0.006693883000000017
> proj: stere
> lat_0: 90
> lat_ts: 70
> lon_0: -45
> x_0: 0
> y_0: 0
> no_defs: defined
>
> ERROR: datum
>
> In case of no significant differences in the projection definitions, use
> the -o flag to ignore them and use current location definition.
> Consider generating a new location from the input dataset using the
> 'location' parameter.
>
>
> If I remove thos two attributes (semi_major_axis and inverse_flattening)
> with:
>
> ncatted -O -h -a semi_major_axis,mapping,d,l,6378273 -a
> inverse_flattening,mapping,d,f,298.27940504282 file.nc foo.nc
>
> And then try to read in the data, it works.
>
> Is there some reason that those two attributes make it difficult to load
> the data in GRASS? Is there a better work-around than removing them? The
> documented projection of the data has not changed, only some of the
> metadata.
>
> Versions:
> gdal 3.0.4
> grass 7.8.2
>
> Thank you,
>
>    -k.
>
> _______________________________________________
> grass-user mailing list
> grass-user at lists.osgeo.org
>
> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.osgeo.org%2Fmailman%2Flistinfo%2Fgrass-user&data=04%7C01%7C%7C900b853648c545edd6fd08d94dffe4cb%7C6cef373021314901831055b3abf02c73%7C0%7C0%7C637626583719044920%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=wWKyBuAKwbuobSGVQ6q97VXj1OIiCLSN9NlbXDLw7tc%3D&reserved=0
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/grass-user/attachments/20210725/2ebd4dcd/attachment.html>