python - Assign new coordinates to a variable using xarray

Question

I have a .nc file opened as a dataset in xarray with the following structure:

ds

<xarray.Dataset>
Dimensions:    (lat: 733, lon: 720, time: 204)
Coordinates:
  * time       (time) int16 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 ...
Dimensions without coordinates: lat , lon
Data variables :
    latitude   (lat) float32 56.0 55.9917 55.9833 55.975 55.9667 55.9583 ...
    longitude  (lon) float32 -11.0 -10.9917 -10.9833 -10.975 -10.9667 ...
    n2o        (lat, lon, time) float64 nan nan nan nan nan nan nan nan nan ...
    co2        (lat, lon, time) float64 nan nan nan nan nan nan nan nan nan ...
    no3        (lat, lon, time) float64 nan nan nan nan nan nan nan nan nan ...
    ch4        (lat, lon, time) float64 nan nan nan nan nan nan nan nan nan ...
    soc        (lat, lon, time) float64 nan nan nan nan nan nan nan nan nan ...

There is an issue with this file, the variables (n2o, co2 etc.) have incorrect coordinates, the lat and lon associated with them are wrong. I would like to assign the 'latitude' and 'longitude' data variables (which contain the correct values) as the coordinates of the other variables (n2o, co2 etc.), to replace 'lat' and 'lon'. I am not sure if this is possible.

The 'latitude' variable looks like this:

<xarray.Variable (lat: 733)>
array([ 56.      ,  55.991665,  55.98333 , ...,  49.915367,  49.907032,
    49.898697], dtype=float32)
Attributes:
    units:      degrees of latitude North to South in 30 arc seconds steps
    long_name:  latitude

The ds.co2.lat dimension looks like this:

<xarray.DataArray 'lat' (lat: 733)>
array([  0,   1,   2, ..., 730, 731, 732], dtype=int64)
Dimensions without coordinates: lat

I have tried:

newlat = ds.variables['latitude'][:]

ds.co2['latitude'] = newlat

This didn't seem to do anything - I am not sure what to do next.

标签： pythonnetcdfpython-xarray

Answer 1

您可能正在寻找renameandset_coords方法。您所看到的是 xarray 无法确定纬度/经度是否是坐标变量。解决方案是首先重命名纬度/经度变量，然后将这些变量关联为坐标。

In [10]: ds = xr.Dataset({'a': xr.Variable(('lon', 'lat'), np.random.random((4, 5))), 'b': xr.Variable(('lon', 'lat'),np.random.random((4, 5))), 'longitude':
    ...:  xr.Variable('lon', [1, 2, 3, 4]), 'latitude': xr.Variable('lat', [1, 2, 3, 4, 5])})

In [11]: ds
Out[11]:
<xarray.Dataset>
Dimensions:    (lat: 5, lon: 4)
Dimensions without coordinates: lat, lon
Data variables:
    a          (lon, lat) float64 0.8694 0.5929 0.04661 0.9571 0.1814 ...
    b          (lon, lat) float64 0.8893 0.5772 0.8457 0.8337 0.7966 0.1619 ...
    longitude  (lon) int64 1 2 3 4
    latitude   (lat) int64 1 2 3 4 5

In [12]: ds.rename({'latitude': 'lat', 'longitude': 'lon'}).set_coords(['lon', 'lat'])
Out[12]:
<xarray.Dataset>
Dimensions:  (lat: 5, lon: 4)
Coordinates:
  * lon      (lon) int64 1 2 3 4
  * lat      (lat) int64 1 2 3 4 5
Data variables:
    a        (lon, lat) float64 0.8694 0.5929 0.04661 0.9571 0.1814 0.01788 ...
    b        (lon, lat) float64 0.8893 0.5772 0.8457 0.8337 0.7966 0.1619 ...