Manipulating Dimensions (Data Resolution)#
Sometimes we need to change the resolution of our data. We might need to look at inferred values between dimension (grid) spaces or change the dimension spacing completely (for instance to add another variable). Learning goals:
Interpolate data to new coordinates
import numpy as np
import pandas as pd
import xarray as xr
np.random.seed(0)
arr = xr.DataArray(
np.random.randn(4, 6),
dims=("x", "y"),
coords={
"x": [-3.2, 2.1, 5.3, 6.5],
"y": pd.date_range("2009-01-05", periods=6, freq="M"),
},
)
arr
/tmp/ipykernel_4639/39313188.py:6: FutureWarning: 'M' is deprecated and will be removed in a future version, please use 'ME' instead.
"y": pd.date_range("2009-01-05", periods=6, freq="M"),
<xarray.DataArray (x: 4, y: 6)> Size: 192B
array([[ 1.76405235, 0.40015721, 0.97873798, 2.2408932 , 1.86755799,
-0.97727788],
[ 0.95008842, -0.15135721, -0.10321885, 0.4105985 , 0.14404357,
1.45427351],
[ 0.76103773, 0.12167502, 0.44386323, 0.33367433, 1.49407907,
-0.20515826],
[ 0.3130677 , -0.85409574, -2.55298982, 0.6536186 , 0.8644362 ,
-0.74216502]])
Coordinates:
* x (x) float64 32B -3.2 2.1 5.3 6.5
* y (y) datetime64[ns] 48B 2009-01-31 2009-02-28 ... 2009-06-30Interpolation#
If we want to look at values between the current grid cells (interpolation), we
can do that with interp (requires scipy):
arr.interp(
x=np.linspace(2, 6, 10),
y=pd.date_range("2009-04-01", "2009-04-30", freq="D"),
)
Show code cell output
<xarray.DataArray (x: 10, y: 30)> Size: 2kB
array([[-6.52066740e-02, -4.76087761e-02, -3.00108783e-02,
-1.24129804e-02, 5.18491747e-03, 2.27828153e-02,
4.03807132e-02, 5.79786111e-02, 7.55765089e-02,
9.31744068e-02, 1.10772305e-01, 1.28370203e-01,
1.45968100e-01, 1.63565998e-01, 1.81163896e-01,
1.98761794e-01, 2.16359692e-01, 2.33957590e-01,
2.51555488e-01, 2.69153385e-01, 2.86751283e-01,
3.04349181e-01, 3.21947079e-01, 3.39544977e-01,
3.57142875e-01, 3.74740773e-01, 3.92338671e-01,
4.09936568e-01, 4.27534466e-01, 4.45132364e-01],
[-2.94432103e-02, -1.45548765e-02, 3.33457251e-04,
1.52217910e-02, 3.01101248e-02, 4.49984586e-02,
5.98867924e-02, 7.47751261e-02, 8.96634599e-02,
1.04551794e-01, 1.19440127e-01, 1.34328461e-01,
1.49216795e-01, 1.64105129e-01, 1.78993463e-01,
1.93881796e-01, 2.08770130e-01, 2.23658464e-01,
2.38546798e-01, 2.53435131e-01, 2.68323465e-01,
2.83211799e-01, 2.98100133e-01, 3.12988467e-01,
3.27876800e-01, 3.42765134e-01, 3.57653468e-01,
3.72541802e-01, 3.87430135e-01, 4.02318469e-01],
...
[-1.74483269e-01, -1.54611067e-01, -1.34738864e-01,
-1.14866662e-01, -9.49944589e-02, -7.51222563e-02,
-5.52500537e-02, -3.53778511e-02, -1.55056484e-02,
4.36655420e-03, 2.42387568e-02, 4.41109594e-02,
6.39831621e-02, 8.38553647e-02, 1.03727567e-01,
1.23599770e-01, 1.43471973e-01, 1.63344175e-01,
1.83216378e-01, 2.03088580e-01, 2.22960783e-01,
2.42832986e-01, 2.62705188e-01, 2.82577391e-01,
3.02449594e-01, 3.22321796e-01, 3.42193999e-01,
3.62066201e-01, 3.81938404e-01, 4.01810607e-01],
[-1.24348073e+00, -1.18266041e+00, -1.12184009e+00,
-1.06101978e+00, -1.00019946e+00, -9.39379140e-01,
-8.78558822e-01, -8.17738504e-01, -7.56918187e-01,
-6.96097869e-01, -6.35277552e-01, -5.74457234e-01,
-5.13636916e-01, -4.52816599e-01, -3.91996281e-01,
-3.31175963e-01, -2.70355646e-01, -2.09535328e-01,
-1.48715010e-01, -8.78946927e-02, -2.70743751e-02,
3.37459426e-02, 9.45662602e-02, 1.55386578e-01,
2.16206896e-01, 2.77027213e-01, 3.37847531e-01,
3.98667848e-01, 4.59488166e-01, 5.20308484e-01]])
Coordinates:
* x (x) float64 80B 2.0 2.444 2.889 3.333 ... 4.667 5.111 5.556 6.0
* y (y) datetime64[ns] 240B 2009-04-01 2009-04-02 ... 2009-04-30when trying to extrapolate, the resulting values will be nan.
If we already have a object with the desired coordinates, we can use
interp_like:
other = xr.DataArray(
dims=("x", "y"),
coords={
"x": np.linspace(2, 4, 10),
"y": pd.date_range("2009-04-01", "2009-04-30", freq="D"),
},
)
arr.interp_like(other)
Show code cell output
<xarray.DataArray (x: 10, y: 30)> Size: 2kB
array([[-6.52066740e-02, -4.76087761e-02, -3.00108783e-02,
-1.24129804e-02, 5.18491747e-03, 2.27828153e-02,
4.03807132e-02, 5.79786111e-02, 7.55765089e-02,
9.31744068e-02, 1.10772305e-01, 1.28370203e-01,
1.45968100e-01, 1.63565998e-01, 1.81163896e-01,
1.98761794e-01, 2.16359692e-01, 2.33957590e-01,
2.51555488e-01, 2.69153385e-01, 2.86751283e-01,
3.04349181e-01, 3.21947079e-01, 3.39544977e-01,
3.57142875e-01, 3.74740773e-01, 3.92338671e-01,
4.09936568e-01, 4.27534466e-01, 4.45132364e-01],
[-6.59905628e-02, -4.96577701e-02, -3.33249774e-02,
-1.69921847e-02, -6.59391953e-04, 1.56734008e-02,
3.20061935e-02, 4.83389862e-02, 6.46717789e-02,
8.10045716e-02, 9.73373643e-02, 1.13670157e-01,
1.30002950e-01, 1.46335742e-01, 1.62668535e-01,
1.79001328e-01, 1.95334121e-01, 2.11666913e-01,
2.27999706e-01, 2.44332499e-01, 2.60665291e-01,
2.76998084e-01, 2.93330877e-01, 3.09663670e-01,
3.25996462e-01, 3.42329255e-01, 3.58662048e-01,
3.74994840e-01, 3.91327633e-01, 4.07660426e-01],
...
[ 1.89840905e-01, 1.96062485e-01, 2.02284065e-01,
2.08505645e-01, 2.14727225e-01, 2.20948806e-01,
2.27170386e-01, 2.33391966e-01, 2.39613546e-01,
2.45835126e-01, 2.52056706e-01, 2.58278287e-01,
2.64499867e-01, 2.70721447e-01, 2.76943027e-01,
2.83164607e-01, 2.89386188e-01, 2.95607768e-01,
3.01829348e-01, 3.08050928e-01, 3.14272508e-01,
3.20494088e-01, 3.26715669e-01, 3.32937249e-01,
3.39158829e-01, 3.45380409e-01, 3.51601989e-01,
3.57823569e-01, 3.64045150e-01, 3.70266730e-01],
[ 2.26388257e-01, 2.31165378e-01, 2.35942500e-01,
2.40719621e-01, 2.45496742e-01, 2.50273863e-01,
2.55050985e-01, 2.59828106e-01, 2.64605227e-01,
2.69382348e-01, 2.74159470e-01, 2.78936591e-01,
2.83713712e-01, 2.88490833e-01, 2.93267955e-01,
2.98045076e-01, 3.02822197e-01, 3.07599318e-01,
3.12376440e-01, 3.17153561e-01, 3.21930682e-01,
3.26707803e-01, 3.31484925e-01, 3.36262046e-01,
3.41039167e-01, 3.45816288e-01, 3.50593410e-01,
3.55370531e-01, 3.60147652e-01, 3.64924773e-01]])
Coordinates:
* x (x) float64 80B 2.0 2.222 2.444 2.667 ... 3.333 3.556 3.778 4.0
* y (y) datetime64[ns] 240B 2009-04-01 2009-04-02 ... 2009-04-30