Representation::croppedRepresentation() not implemented for RegularGrid

MiNeEU commented 1 year ago

What happened?

When I try to process grb in Lambert it is not possible to crop it

filename = 'CLSTEMPERATURE_00.grb' f = mv.read(filename) f = mv.read(data=f, area=[47,15,50,23])

Is it possible to implement it? Thanks

Python - ERROR - 20231005.095534 - Line 0 in 'PythonScript': Exception: Serious Bug: Representation::croppedRepresentation() not implemented for RegularGrid[x=LocalConfiguration[root={type => linear , start => -1.28924e+06 , end => 1.23076e+06 , N => 561 , endpoint => true}],y=LocalConfiguration[root={type => linear , start => -827936 , end => 1.14306e+06 , N => 439 , endpoint => true}],projection=LocalConfiguration[root={type => lambert_conformal_conic , longitude0 => 17 , latitude0 => 46.244 , latitude1 => 46.244 , latitude2 => 46.244}],firstPointBottomLeft=0,bbox=BoundingBox[north=56.4659,west=-3.18266,south=37.759,east=36.3038]] Python - ERROR - 20231005.095534 - Line 0 in 'PythonScript': MIR: Serious Bug: Representation::croppedRepresentation() not implemented for RegularGrid[x=LocalConfiguration[root={type => linear , start => -1.28924e+06 , end => 1.23076e+06 , N => 561 , endpoint => true}],y=LocalConfiguration[root={type => linear , start => -827936 , end => 1.14306e+06 , N => 439 , endpoint => true}],projection=LocalConfiguration[root={type => lambert_conformal_conic , longitude0 => 17 , latitude0 => 46.244 , latitude1 => 46.244 , latitude2 => 46.244}],firstPointBottomLeft=0,bbox=BoundingBox[north=56.4659,west=-3.18266,south=37.759,east=36.3038]] Python - ERROR - 20231005.095534 - Line 0 in 'PythonScript': Interpolation failed (-2) Python - ERROR - 20231005.095534 - Line 0 in 'PythonScript': Retrieve-> Error code: -2 Traceback (most recent call last):

What are the steps to reproduce the bug?

When I try to process grb in Lambert it is not possible to crop it

CLSTEMPERATURE_00.grb

filename = 'CLSTEMPERATURE_00.grb' f = mv.read(filename) f = mv.read(data=f, area=[47,15,50,23])

Is it possible to implement it? Thanks

Version

1.15.0

Platform (OS and architecture)

Linux x86_64

Relevant log output

Python - ERROR  - 20231005.095534 - Line 0 in 'PythonScript': Exception: Serious Bug: Representation::croppedRepresentation() not implemented for RegularGrid[x=LocalConfiguration[root={type => linear , start => -1.28924e+06 , end => 1.23076e+06 , N => 561 , endpoint => true}],y=LocalConfiguration[root={type => linear , start => -827936 , end => 1.14306e+06 , N => 439 , endpoint => true}],projection=LocalConfiguration[root={type => lambert_conformal_conic , longitude0 => 17 , latitude0 => 46.244 , latitude1 => 46.244 , latitude2 => 46.244}],firstPointBottomLeft=0,bbox=BoundingBox[north=56.4659,west=-3.18266,south=37.759,east=36.3038]]
Python - ERROR  - 20231005.095534 - Line 0 in 'PythonScript': MIR: Serious Bug: Representation::croppedRepresentation() not implemented for RegularGrid[x=LocalConfiguration[root={type => linear , start => -1.28924e+06 , end => 1.23076e+06 , N => 561 , endpoint => true}],y=LocalConfiguration[root={type => linear , start => -827936 , end => 1.14306e+06 , N => 439 , endpoint => true}],projection=LocalConfiguration[root={type => lambert_conformal_conic , longitude0 => 17 , latitude0 => 46.244 , latitude1 => 46.244 , latitude2 => 46.244}],firstPointBottomLeft=0,bbox=BoundingBox[north=56.4659,west=-3.18266,south=37.759,east=36.3038]]
Python - ERROR  - 20231005.095534 - Line 0 in 'PythonScript': Interpolation failed (-2)
Python - ERROR  - 20231005.095534 - Line 0 in 'PythonScript': Retrieve-> Error code: -2
Traceback (most recent call last):
  File "test_final_griborg_mask_area.py", line 46, in <module>
    f = mv.read(data=f, area=[47,15,50,23])
  File "/users/hips001/bin/Python-hips/lib/python3.7/site-packages/metview/bindings.py", line 1278, in wrapped
    return value_from_metview(val)
  File "/users/hips001/bin/Python-hips/lib/python3.7/site-packages/metview/bindings.py", line 1232, in value_from_metview
    raise retval
Exception: Metview error: Exception: Serious Bug: Representation::croppedRepresentation() not implemented for RegularGrid[x=LocalConfiguration[root={type => linear , start => -1.28924e+06 , end => 1.23076e+06 , N => 561 , endpoint => true}],y=LocalConfiguration[root={type => linear , start => -827936 , end => 1.14306e+06 , N => 439 , endpoint => true}],projection=LocalConfiguration[root={type => lambert_conformal_conic , longitude0 => 17 , latitude0 => 46.244 , latitude1 => 46.244 , latitude2 => 46.244}],firstPointBottomLeft=0,bbox=BoundingBox[north=56.4659,west=-3.18266,south=37.759,east=36.3038]]

Accompanying data

https://aladin.nowcasting.eu/shmu/alaro/oper/CLSTEMPERATURE_00.grb

Organisation

SHMU

iainrussell commented 1 year ago

Hi @MiNeEU,

Many thanks for the report. I can confirm that we do not support the direct cropping of these Lambert Conformal grids. You do have a couple of options though, depending on what you want to do with the data next.

1) You can change the grid to be regular lat/lon within that read command, e.g.

f = mv.read(data=f, area=[47,15,50,23], grid=[0.1,0.1])

This will crop and interpolate onto the regular grid in one step. Your data will now be on a regular grid - the finer the grid you choose (I used 0.1 degrees in this example, but you can change it), the closer it will be to the original data.

2) You could convert to Geopoints if you don't mind the data becoming 'scattered' points. e.g.

geo = mv.grib_to_geo(geopoints_format = "x_y_v", data = f)
inside_mask = geo.mask([47,15,50,23])
inside = geo.filter(inside_mask)

I hope this helps!

Cheers, Iain

MiNeEU commented 1 year ago

Hi Iain, many thanks for your quick replay. I go to try it. Have a great day. Kind regards Michal Nestiak, NWP SHMU

On Fri, 6 Oct 2023, 13:07 Iain Russell, @.***> wrote:

Hi @MiNeEU https://github.com/MiNeEU,

Many thanks for the report. I can confirm that we do not support the direct cropping of these Lambert Conformal grids. You do have a couple of options though, depending on what you want to do with the data next.

You can change the grid to be regular lat/lon within that read command, e.g.

f = mv.read(data=f, area=[47,15,50,23], grid=[0.1,0.1])

This will crop and interpolate onto the regular grid in one step. Your data will now be on a regular grid - the finer the grid you choose (I used 0.1 degrees in this example, but you can change it), the closer it will be to the original data.

You could convert to Geopoints if you don't mind the data becoming 'scattered' points. e.g.

geo = mv.grib_to_geo(geopoints_format = "x_y_v", data = f) inside_mask = geo.mask([47,15,50,23]) inside = geo.filter(inside_mask)

I hope this helps!

Cheers, Iain

— Reply to this email directly, view it on GitHub https://github.com/ecmwf/metview-python/issues/52#issuecomment-1750419944, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABTJVZ3P4KY7OKQC6XVFN7DX57Q5XAVCNFSM6AAAAAA5UAHRH2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONJQGQYTSOJUGQ . You are receiving this because you were mentioned.Message ID: @.***>

ecmwf / metview-python