python - 使用 xarray 为变量分配新坐标

Question

我有一个 .nc 文件在 xarray 中作为数据集打开，具有以下结构:

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 ...

此文件存在问题，变量(n2o、co2 等)的坐标不正确，与它们关联的纬度和经度错误。我想将“纬度”和“经度”数据变量(包含正确的值)分配为其他变量(n2o、co2 等)的坐标，以替换“lat”和“lon”。我不确定这是否可能。

“纬度”变量如下所示:

<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

ds.co2.lat 维度如下所示:

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

我试过:

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

ds.co2['latitude'] = newlat

这似乎没有做任何事情 - 我不确定下一步该做什么。

Answer 1

您可能正在寻找 rename 和 set_coords 方法。您所看到的是 xarray 无法确定纬度/经度是坐标变量。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 ...