pydata / pydata/xarray

Unexpected empty spots in xarray quiver plot with hue of masked data

Open
#8,928 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug topic-plotting
Dominant language
Python
Stars
4.2k
Forks
1.4k
Avg merge
2d 15h
Merged PRs (30d)
14

Description

What happened?

When creating a quiver plot of an xarray Dataset using ds.plot.quiver, there are empty spots in the wrong places. This problem only seems to occur when a hue array is given to quiver and the data is masked, i.e., there are NaN values in the array. The empty spots are not only in the masked locations, but seem to appear in arbitrary locations.

What did you expect to happen?

Plotting the same data with the matplotlib function plt.quiver, the empty spots are not there, no matter if there is a hue (called color in matplotlib) or NaN values. This is the result I expected, and I expect that the xarray plot method gives the same result.

Minimal Complete Verifiable Example
import xarray as xr
import matplotlib.pyplot as plt

ds = xr.Dataset(coords={"x": range(20), "y": range(15)})
ds["c"] = xr.ones_like(xr.broadcast(ds.y, ds.x)[0])

# Mask out the central band
# When the next line is commented out, the problem does not occur
ds["c"] = ds.c.where((ds.x - 10)**2 > 1)

fig, axs = plt.subplots(2, 2, sharex=True, sharey=True)

# Plot the data with the xarray method
ds.plot.quiver("x", "y", "c", "c", add_guide=False, ax=axs[0, 0])  # this works as expected
ds.plot.quiver("x", "y", "c", "c", "c", add_guide=False, ax=axs[0, 1])  # this does not

# Make the same plot directly with matplotlib to see the expected result
axs[1, 0].quiver(ds.x, ds.y, ds.c, ds.c)  # this works as expected
axs[1, 1].quiver(ds.x, ds.y, ds.c, ds.c, ds.c)  # this works, too

axs[0, 0].set_title("xarray plot, no hue")
axs[0, 1].set_title("xarray plot, with hue")
axs[1, 0].set_title("manual plot, no color")
axs[1, 1].set_title("manual plot, with color")
for ax in axs.flatten():
    ax.set(xlabel="", ylabel="")

plt.show()
MVCE confirmation
  • Minimal example — the example is as focused as reasonably possible to demonstrate the underlying issue in xarray.
  • Complete example — the example is self-contained, including all data and the text of any traceback.
  • Verifiable example — the example copy & pastes into an IPython prompt or Binder notebook, returning the result.
  • New issue — a search of GitHub Issues suggests this is not a duplicate.
  • Recent environment — the issue occurs with the latest version of xarray and its dependencies.
Relevant log output

No response

Anything else we need to know?

This is the figure that is created with the example code. The problems are the empty spots in the top right panel. I expected this panel to look like the bottom right panel.
The left column shows that there are no problems when there is no hue/color argument. The quiver plot created with xarray looks just like the one created manually with matplotlib.

output

Environment

INSTALLED VERSIONS

commit: None
python: 3.8.10 (default, Nov 22 2023, 10:22:35)
[GCC 9.4.0]
python-bits: 64
OS: Linux
OS-release: 5.4.0-174-generic
machine: x86_64
processor: x86_64
byteorder: little
LC_ALL: None
LANG: en_GB.UTF-8
LOCALE: ('en_GB', 'UTF-8')
libhdf5: 1.12.0
libnetcdf: 4.7.4

xarray: 2022.12.0
pandas: 1.3.0
numpy: 1.23.5
scipy: 1.9.3
netCDF4: 1.5.8
pydap: None
h5netcdf: 0.7.1
h5py: 2.10.0
Nio: None
zarr: 2.11.3
cftime: 1.5.2
nc_time_axis: None
PseudoNetCDF: None
rasterio: None
cfgrib: 0.9.10.1
iris: None
bottleneck: 1.2.1
dask: 2022.05.0
distributed: None
matplotlib: 3.6.2
cartopy: 0.21.1
seaborn: 0.13.0
numbagg: None
fsspec: 2022.3.0
cupy: None
pint: None
sparse: None
flox: None
numpy_groupies: None
setuptools: 45.2.0
pip: 23.2.1
conda: None
pytest: 4.6.9
mypy: None
IPython: 8.12.3
sphinx: 3.5.3

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by running the minimal example using ds.plot.quiver and compare the hue-plus-NaN result with matplotlib's plt.quiver. Trace the xarray quiver plotting path where masked data and hue are handled, then add a regression test showing that empty spots occur only at masked locations and match the matplotlib result.

Written by the indexing model from the issue text.

Assessment

Tech stack
matplotlib, python
Domain
data-visualization
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
45/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.