Mapping HadUK-Grid NetCDF data

[thomasattree]

Hi,
I'm fairly new to geospatial data analysis so please go easy on me.
I am attempting to add a layer to leafmap composed of the average hours of sun irradiation from the HadUK-Grid dataset.
This dataset is different to the example provided in the docs, I'm looking for a little help in attempting to understand what to include as parameters, or what sort of pre-processing is required to make the dataset compatible with the add_netcdf() api.

<xarray.Dataset>
Dimensions:                       (time: 1, projection_y_coordinate: 290,
                                   projection_x_coordinate: 180, bnds: 2)
Coordinates:
  * time                          (time) datetime64[ns] 2021-07-02T12:00:00
  * projection_y_coordinate       (projection_y_coordinate) float64 -1.975e+0...
  * projection_x_coordinate       (projection_x_coordinate) float64 -1.975e+0...
    latitude                      (projection_y_coordinate, projection_x_coordinate) float64 ...
    longitude                     (projection_y_coordinate, projection_x_coordinate) float64 ...
Dimensions without coordinates: bnds
Data variables:
    sun                           (time, projection_y_coordinate, projection_x_coordinate) float64 ...
    transverse_mercator           int32 ...
    time_bnds                     (time, bnds) datetime64[ns] ...
    projection_y_coordinate_bnds  (projection_y_coordinate, bnds) float64 ...
    projection_x_coordinate_bnds  (projection_x_coordinate, bnds) float64 ...
Attributes:
    comment:        annual resolution gridded climate observations
    creation_date:  2022-03-14T13:41:17
    frequency:      ann
    institution:    Met Office
    references:     doi: 10.1002/joc.1161
    short_name:     annual_sunshine
    source:         HadUK-Grid_v1.1.0.0
    title:          Gridded surface climate observations data for the UK
    version:        v20220310
    Conventions:    CF-1.5

I've attempted to add the file by using

m.add_netcdf(
    filename,
    lon='projection_x_coordinate', 
    lat='projection_y_coordinate',
    variables=['sun'], 
    shift_lon=False,
    layer_name='Sun',
    palette='coolwarm',
    time=0)

But this throws:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
/tmp/ipykernel_21793/1282227036.py in <cell line: 1>()
----> 1 m.add_netcdf(
      2     filename,
      3     lon='projection_x_coordinate',
      4     lat='projection_y_coordinate',
      5     variables=['sun'],

~/anaconda3/envs/python3/lib/python3.8/site-packages/leafmap/leafmap.py in add_netcdf(self, filename, variables, palette, vmin, vmax, nodata, attribution, layer_name, shift_lon, lat, lon, lev, level_index, time, **kwargs)
   1768         """
   1769 
-> 1770         tif, vars = netcdf_to_tif(
   1771             filename,
   1772             shift_lon=shift_lon,

~/anaconda3/envs/python3/lib/python3.8/site-packages/leafmap/common.py in netcdf_to_tif(filename, output, variables, shift_lon, lat, lon, lev, level_index, time, return_vars, **kwargs)
   5364 
   5365     if variables is None:
-> 5366         xds.rio.set_spatial_dims(x_dim=lon, y_dim=lat).rio.to_raster(output)
   5367     else:
   5368         xds[variables].rio.set_spatial_dims(x_dim=lon, y_dim=lat).rio.to_raster(output)

~/anaconda3/envs/python3/lib/python3.8/site-packages/rioxarray/raster_dataset.py in to_raster(self, raster_path, driver, dtype, tags, windowed, recalc_transform, lock, compute, **profile_kwargs)
    486         """
    487         variable_dim = f"band_{uuid4()}"
--> 488         data_array = self._obj.to_array(dim=variable_dim)
    489         # write data array names to raster
    490         data_array.attrs["long_name"] = data_array[variable_dim].values.tolist()

~/anaconda3/envs/python3/lib/python3.8/site-packages/xarray/core/dataset.py in to_array(self, dim, name)
   5296         data_vars = [self.variables[k] for k in self.data_vars]
   5297         broadcast_vars = broadcast_variables(*data_vars)
-> 5298         data = duck_array_ops.stack([b.data for b in broadcast_vars], axis=0)
   5299 
   5300         coords = dict(self.coords)

~/anaconda3/envs/python3/lib/python3.8/site-packages/xarray/core/duck_array_ops.py in stack(arrays, axis)
    289 def stack(arrays, axis=0):
    290     """stack() with better dtype promotion rules."""
--> 291     return _stack(as_shared_dtype(arrays), axis=axis)
    292 
    293 

~/anaconda3/envs/python3/lib/python3.8/site-packages/xarray/core/duck_array_ops.py in as_shared_dtype(scalars_or_arrays)
    189     # Note that result_type() safely gets the dtype from dask arrays without
    190     # evaluating them.
--> 191     out_type = dtypes.result_type(*arrays)
    192     return [x.astype(out_type, copy=False) for x in arrays]
    193 

~/anaconda3/envs/python3/lib/python3.8/site-packages/xarray/core/dtypes.py in result_type(*arrays_and_dtypes)
    179             return np.dtype(object)
    180 
--> 181     return np.result_type(*arrays_and_dtypes)

<__array_function__ internals> in result_type(*args, **kwargs)

TypeError: invalid type promotion

I think the issue is in the projection of the dataset, trying to use the projection variables rather than the long/lat, but I'm not sure how to apply the projection for the file.

[giswqs]

Can you provide a sample dataset for testing?

[thomasattree]

Sure, here's a sample dataset I'm attempting to map. Had to zip for GH compatibility
sun_hadukgrid_uk_5km_ann_202101-202112.nc.zip.