Intermediate Xarray DataStructures

Intermediate Xarray DataStructures#

In this lesson, we cover the basics of Xarray data structures. By the end of the lesson, we will be able to:

Understand the basic data structures in Xarray
Inspect DataArray and Dataset objects.
Select Data using Dataset Coordinates
Understand that these Datastructures generalize across fields

We’ll start by reviewing the various components of the Xarray data model, represented here visually:

https://docs.xarray.dev/en/stable/_images/dataset-diagram.png

Variable#

At the the core of everything Xarray works with Variables:

Variables hold arbitrary metadata, dimension names, and an Array of data

DataArray#

A DataArray is lowest level object you will regularly interact with. It is a collection of 1 variable containing the data, and optionally Variables associated with the dimensions (coords)

Exercise#

In this exercise the air dataset has been deconstructed into a dictionary of NumPy arrays. Please

Build a Dataset from scratch from this dictionary of NumPy arrays.

Hint: You will want to start by building DataArrays, then combining them into a Dataset.

def air_as_dictionary():
    ds = xr.tutorial.open_dataset("air_temperature")
    return {
        'lat': ds['lat'].values,
        'lon': ds['lon'].values,
        'attrs': ds.attrs,
        'time': ds['time'].values,
        'air-temp': ds['air'].values,
    }


data = air_as_dictionary()


# Your code here

Why Xarray?#

Metadata provides context and provides code that is more legible. This reduces the likelihood of errors from typos and makes analysis more intuitive and fun!

Analysis without xarray: `X(`#

# plot the first timestep
lat = ds.air.lat.data  # numpy array
lon = ds.air.lon.data  # numpy array
temp = ds.air.data  # numpy array

plt.figure()
plt.pcolormesh(lon, lat, temp[0, :, :]);

../_images/a8fb3d0e1d5a2052d43da3655abade08e74180b37b308fd13940cf6e087b9004.png

temp.mean(axis=1)  ## what did I just do? I can't tell by looking at this line.

array([[279.398 , 279.6664, 279.6612, ..., 279.9508, 280.3152, 280.6624],
       [279.0572, 279.538 , 279.7296, ..., 279.7756, 280.27  , 280.7976],
       [279.0104, 279.2808, 279.5508, ..., 279.682 , 280.1976, 280.814 ],
       ...,
       [279.63  , 279.934 , 280.534 , ..., 279.802 , 280.346 , 280.778 ],
       [279.398 , 279.666 , 280.318 , ..., 279.766 , 280.342 , 280.834 ],
       [279.27  , 279.354 , 279.882 , ..., 279.426 , 279.97  , 280.482 ]],
      shape=(2920, 53))

Analysis with xarray `=)`#

How readable is this code?

ds.air.isel(time=0).plot(x="lon");

../_images/1496805dd2dd3f67d48335e07d4fd9b73abe521766175eed1a29cf67744ce338.png

Use dimension names instead of axis numbers

ds.air.mean(dim="time").plot(x="lon")

<matplotlib.collections.QuadMesh at 0x7f95eea03950>

../_images/7a9ea94fd0d910c99c6a5a131e174591ced81e9d5602d7e28400d3fe9d6d78de.png

Other field DataStructures#

Link to next notebook: Data structures In Biology

Exercise#

Dataset of non-alignable data.

For this exercise we will be using the "precipitation" example data. This data has two parts observed rainfall (observed), and predicted rainfall, on a different lat-lon grid. The next cell calls a function that will give you two different Datasets. Try to combine them using the tools we’ve discussed thus far and observe what goes wrong.

tree = xr.tutorial.open_datatree('precipitation.nc4')
reanalysis = xr.Dataset(tree['observed'])
observed = xr.Dataset(tree['reanalysis'])

For more see the user guide, the gallery, and the tutorial material.

Intermediate Xarray DataStructures

Contents

Intermediate Xarray DataStructures#

Variable#

DataArray#

Dataset#

What’s in a Dataset?#

What’s in a DataArray?#

Name (optional)#

Named dimensions#

Coordinate variables#

Arbitrary attributes#

Underlying data#

Review#

Exercise#

Why Xarray?#

Analysis without xarray: `X(`#

Analysis with xarray `=)`#

Extracting data or “indexing”#

Label-based indexing#

Position-based indexing#

Other field DataStructures#

Exercise#

Intermediate Xarray DataStructures

Contents

Intermediate Xarray DataStructures#

Variable#

DataArray#

Dataset#

What’s in a Dataset?#

What’s in a DataArray?#

Name (optional)#

Named dimensions#

Coordinate variables#

Arbitrary attributes#

Underlying data#

Review#

Exercise#

Why Xarray?#

Analysis without xarray: X(#

Analysis with xarray =)#

Extracting data or “indexing”#

Label-based indexing#

Position-based indexing#

Other field DataStructures#

Exercise#

Analysis without xarray: `X(`#

Analysis with xarray `=)`#