Xarray: slicing coordinates returns an empty dimension

Question

community!

While working with outputs of climate simulation models I've faced some strange xarray behavior - slicing by latitude returns an empty array.

xr.open_mfdataset(nc_list[1])

Output


Dimensions:        (bnds: 2, time: 1128, lev: 13, lat: 360, lon: 720)
Coordinates:
  * bnds           (bnds) float64 0.0 1.0
  * time           (time) float64 4.14e+03 4.141e+03 ... 5.266e+03 5.267e+03
  * lev            (lev) float64 1.0 2.0 3.0 4.0 5.0 ... 9.0 10.0 11.0 12.0 13.0
  * lat            (lat) float64 89.75 89.25 88.75 ... -88.75 -89.25 -89.75
  * lon            (lon) float64 -179.8 -179.2 -178.8 ... 178.8 179.2 179.8
Data variables:
    depth          (time, lev, lat, lon) float64 ...
    depth_bnds     (bnds, time, lev, lat, lon) float64 ...
    soilmoistfroz  (time, lev, lat, lon) float32 ...
Attributes: (12/15)

Slicing by longtitude works fine:

xr.open_mfdataset(nc_list[1]).sel(lon=slice(0, 100))

Output


Dimensions:        (bnds: 2, time: 1128, lev: 13, lat: 360, lon: 200)
Coordinates:
  * bnds           (bnds) float64 0.0 1.0
  * time           (time) float64 4.14e+03 4.141e+03 ... 5.266e+03 5.267e+03
  * lev            (lev) float64 1.0 2.0 3.0 4.0 5.0 ... 9.0 10.0 11.0 12.0 13.0
  * lat            (lat) float64 89.75 89.25 88.75 ... -88.75 -89.25 -89.75
  * lon            (lon) float64 0.25 0.75 1.25 1.75 ... 98.25 98.75 99.25 99.75
Data variables:
    depth          (time, lev, lat, lon) float64 ...
    depth_bnds     (bnds, time, lev, lat, lon) float64 ...
    soilmoistfroz  (time, lev, lat, lon) float32 ...
Attributes: (12/15)

However, filtering by latitude returns an empty list:

xr.open_mfdataset(nc_list[1]).sel(lat=slice(0, 100))

Output


Dimensions:        (bnds: 2, time: 1128, lev: 13, lat: 0, lon: 720)
Coordinates:
  * bnds           (bnds) float64 0.0 1.0
  * time           (time) float64 4.14e+03 4.141e+03 ... 5.266e+03 5.267e+03
  * lev            (lev) float64 1.0 2.0 3.0 4.0 5.0 ... 9.0 10.0 11.0 12.0 13.0
  * lat            (lat) float64 
  * lon            (lon) float64 -179.8 -179.2 -178.8 ... 178.8 179.2 179.8
Data variables:
    depth          (time, lev, lat, lon) float64 ...
    depth_bnds     (bnds, time, lev, lat, lon) float64 ...
    soilmoistfroz  (time, lev, lat, lon) float32 ...
Attributes: (12/15)

Can you tell me what I’m doing wrong?

SOLVED

The order of the coordinate values matters, thus slicing by latitude should be performed with

xr.open_dataset(path, decode_times=False).sel(lat=slice(100, 0))

Xarray: slicing coordinates returns an empty dimension

0 Answers