alextc
alextc

Reputation: 3515

xarray- add a new dimension to a DataArray and assign coordinates to the dimension

I have a DataArray object named test. It contains a variable named FFDI 90TH PERCENTILEand latitude and longitude dimensions.

import numpy as np
import pandas as pd
import xarray as xr

print(test)

<xarray.DataArray 'FFDI 90TH PERCENTILE' (latitude: 106, longitude: 193)>
array([[ 2.699949,  2.699277,  2.677113, ...,  3.353225,  3.381503,  3.392549],
       [ 2.7     ,  2.704608,  2.70228 , ...,  3.422083,  3.435692,  3.465664],
       [ 2.720069,  2.71194 ,  2.711843, ...,  3.5     ,  3.5     ,  3.501185],
       ...,
       [34.863322, 34.825574, 34.694171, ...,  8.599811,  8.50329 ,  8.815733],
       [34.728609, 35.180146, 35.203714, ...,  8.164053,  8.01015 ,  7.94335 ],
       [34.654186, 34.865241, 34.987067, ...,  7.814975,  7.644326,  7.925   ]])
Coordinates:
  * latitude   (latitude) float32 -39.2 -39.149525 ... -33.950478 -33.9
  * longitude  (longitude) float32 140.8 140.84792 140.89584 ... 149.95209 150.0

I have the following times elements:

times = pd.date_range("1972/12/01","2017/12/01",freq='D',closed='left')
time_da = xr.DataArray(times, [('time', times)])

I would like to add a new dimension and call it time; and assign times as above to the time dimension as coordinates. So that the new test DataArray would look like:

<xarray.DataArray 'FFDI 90TH PERCENTILE' (time: 16436, latitude: 106, longitude: 193)>

I have done the following attempts with assign_coords and expand_dims. Neither of them worked.

One:

test_assigned = test.assign_coords({'time': times.values})
TypeError: assign_coords() takes 1 positional argument but 2 were given

Two:

test_assigned = test.assign_coords(time=times.values)
ValueError: cannot add coordinates with new dimensions to a DataArray

Upvotes: 1

Views: 4120

Answers (1)

spencerkclark
spencerkclark

Reputation: 2097

This is a use-case for expand_dims, which will expand out the array along the new dimension and assign a coordinate to it if provided:

result = test.expand_dims(time=time_da)

Upvotes: 2

Related Questions