xarray ODIM backend#

In this example, we read ODIM_H5 (HDF5) data files using the xarray odim backend.

[1]:
import glob
import os
import wradlib as wrl
import warnings

warnings.filterwarnings("ignore")
import matplotlib.pyplot as pl
import numpy as np
import xarray as xr

try:
    get_ipython().run_line_magic("matplotlib inline")
except:
    pl.ion()
from wradlib.io import open_odim_dataset
/home/runner/micromamba-root/envs/wradlib-tests/lib/python3.11/site-packages/tqdm/auto.py:22: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html
  from .autonotebook import tqdm as notebook_tqdm

Load ODIM_H5 Volume Data#

[2]:
fpath = "hdf5/knmi_polar_volume.h5"
f = wrl.util.get_wradlib_data_file(fpath)
vol = wrl.io.open_odim_dataset(f)
Downloading file 'hdf5/knmi_polar_volume.h5' from 'https://github.com/wradlib/wradlib-data/raw/pooch/data/hdf5/knmi_polar_volume.h5' to '/home/runner/work/wradlib/wradlib/wradlib-data'.

Inspect RadarVolume#

[3]:
display(vol)
<wradlib.RadarVolume>
Dimension(s): (sweep: 14)
Elevation(s): (0.3, 0.4, 0.8, 1.1, 2.0, 3.0, 4.5, 6.0, 8.0, 10.0, 12.0, 15.0, 20.0, 25.0)

Inspect root group#

The sweep dimension contains the number of scans in this radar volume. Further the dataset consists of variables (location coordinates, time_coverage) and attributes (Conventions, metadata).

[4]:
vol.root
[4]:
<xarray.Dataset>
Dimensions:              (sweep: 14)
Dimensions without coordinates: sweep
Data variables:
    volume_number        int64 0
    platform_type        <U5 'fixed'
    instrument_type      <U5 'radar'
    primary_axis         <U6 'axis_z'
    time_coverage_start  <U20 '2011-06-10T11:40:02Z'
    time_coverage_end    <U20 '2011-06-10 11:43:55Z'
    latitude             float32 ...
    longitude            float32 ...
    altitude             float32 ...
    sweep_group_name     (sweep) <U8 'sweep_0' 'sweep_1' ... 'sweep_13'
    sweep_fixed_angle    (sweep) float32 0.3 0.4 0.8 1.1 ... 12.0 15.0 20.0 25.0
Attributes:
    version:          None
    title:            None
    institution:      None
    references:       None
    source:           None
    history:          None
    comment:          im/exported using wradlib
    instrument_name:  None
    fixed_angle:      0.3

Inspect sweep group(s)#

The sweep-groups can be accessed via their respective keys. The dimensions consist of range and time with added coordinates azimuth, elevation, range and time. There will be variables like radar moments (DBZH etc.) and sweep-dependend metadata (like fixed_angle, sweep_mode etc.).

[5]:
display(vol[0])
<xarray.Dataset>
Dimensions:            (azimuth: 360, range: 320)
Coordinates:
  * azimuth            (azimuth) float32 0.5 1.5 2.5 3.5 ... 357.5 358.5 359.5
    elevation          (azimuth) float32 ...
    rtime              (azimuth) datetime64[ns] 2011-06-10T11:40:17.361118208...
  * range              (range) float32 500.0 1.5e+03 ... 3.185e+05 3.195e+05
    sweep_mode         <U20 ...
    longitude          float32 ...
    latitude           float32 ...
    altitude           float32 ...
    time               datetime64[ns] 2011-06-10T11:40:02.027777792
Data variables:
    DBZH               (azimuth, range) float32 ...
    sweep_number       int64 ...
    prt_mode           <U7 ...
    follow_mode        <U7 ...
    sweep_fixed_angle  float32 0.3
Attributes:
    fixed_angle:  0.3

Goereferencing#

[6]:
swp = vol[0].copy().pipe(wrl.georef.georeference_dataset)

Plotting#

[7]:
swp.DBZH.plot.pcolormesh(x="x", y="y")
pl.gca().set_aspect("equal")
../../_images/notebooks_fileio_wradlib_odim_backend_14_0.png
[8]:
fig = pl.figure(figsize=(10, 10))
swp.DBZH.wradlib.plot_ppi(proj="cg", fig=fig)
[8]:
<matplotlib.collections.QuadMesh at 0x7fa79317b0d0>
../../_images/notebooks_fileio_wradlib_odim_backend_15_1.png
[9]:
import cartopy
import cartopy.crs as ccrs
import cartopy.feature as cfeature

map_trans = ccrs.AzimuthalEquidistant(
    central_latitude=swp.latitude.values, central_longitude=swp.longitude.values
)
[10]:
map_proj = ccrs.AzimuthalEquidistant(
    central_latitude=swp.latitude.values, central_longitude=swp.longitude.values
)
pm = swp.DBZH.wradlib.plot_ppi(proj=map_proj)
ax = pl.gca()
ax.gridlines(crs=map_proj)
print(ax)
< GeoAxes: +proj=aeqd +ellps=WGS84 +lon_0=4.7899699211120605 +lat_0=52.953338623046875 +x_0=0.0 +y_0=0.0 +no_defs +type=crs >
../../_images/notebooks_fileio_wradlib_odim_backend_17_1.png
[11]:
map_proj = ccrs.Mercator(central_longitude=swp.longitude.values)
fig = pl.figure(figsize=(10, 8))
ax = fig.add_subplot(111, projection=map_proj)
pm = swp.DBZH.wradlib.plot_ppi(ax=ax)
ax.gridlines(draw_labels=True)
[11]:
<cartopy.mpl.gridliner.Gridliner at 0x7fa78ac4ab10>
../../_images/notebooks_fileio_wradlib_odim_backend_18_1.png
[12]:
import cartopy.feature as cfeature


def plot_borders(ax):
    borders = cfeature.NaturalEarthFeature(
        category="physical", name="coastline", scale="10m", facecolor="none"
    )
    ax.add_feature(borders, edgecolor="black", lw=2, zorder=4)


map_proj = ccrs.Mercator(central_longitude=swp.longitude.values)
fig = pl.figure(figsize=(10, 8))
ax = fig.add_subplot(111, projection=map_proj)

DBZH = swp.DBZH
pm = DBZH.where(DBZH > 0).wradlib.plot_ppi(ax=ax)
plot_borders(ax)
ax.gridlines(draw_labels=True)
[12]:
<cartopy.mpl.gridliner.Gridliner at 0x7fa78adef290>
../../_images/notebooks_fileio_wradlib_odim_backend_19_1.png
[13]:
import matplotlib.path as mpath

theta = np.linspace(0, 2 * np.pi, 100)
center, radius = [0.5, 0.5], 0.5
verts = np.vstack([np.sin(theta), np.cos(theta)]).T
circle = mpath.Path(verts * radius + center)

map_proj = ccrs.AzimuthalEquidistant(
    central_latitude=swp.latitude.values,
    central_longitude=swp.longitude.values,
)
fig = pl.figure(figsize=(10, 8))
ax = fig.add_subplot(111, projection=map_proj)
ax.set_boundary(circle, transform=ax.transAxes)

pm = swp.DBZH.wradlib.plot_ppi(proj=map_proj, ax=ax)
ax = pl.gca()
ax.gridlines(crs=map_proj)
[13]:
<cartopy.mpl.gridliner.Gridliner at 0x7fa78b025610>
../../_images/notebooks_fileio_wradlib_odim_backend_20_1.png
[14]:
fig = pl.figure(figsize=(10, 8))
proj = ccrs.AzimuthalEquidistant(
    central_latitude=swp.latitude.values, central_longitude=swp.longitude.values
)
ax = fig.add_subplot(111, projection=proj)
pm = swp.DBZH.wradlib.plot_ppi(ax=ax)
ax.gridlines()
[14]:
<cartopy.mpl.gridliner.Gridliner at 0x7fa78afa3bd0>
../../_images/notebooks_fileio_wradlib_odim_backend_21_1.png
[15]:
swp.DBZH.wradlib.plot_ppi()
[15]:
<matplotlib.collections.QuadMesh at 0x7fa7886a74d0>
../../_images/notebooks_fileio_wradlib_odim_backend_22_1.png

Inspect radar moments#

The DataArrays can be accessed by key or by attribute. Each DataArray has dimensions and coordinates of it’s parent dataset. There are attributes connected which are defined by ODIM_H5 standard.

[16]:
display(swp.DBZH)
<xarray.DataArray 'DBZH' (azimuth: 360, range: 320)>
array([[ 22. ,  17. ,  -8. , ..., -31.5, -31.5, -31.5],
       [ 24. ,  24.5,  -9. , ..., -31.5, -31.5, -31.5],
       [ 35.5,  42. ,  12. , ..., -31.5, -31.5, -31.5],
       ...,
       [ 23. ,  14. , -13. , ..., -31.5, -31.5, -31.5],
       [ 23. ,  14. ,  -9. , ..., -31.5, -31.5, -31.5],
       [ 22. ,  18.5, -11.5, ..., -31.5, -31.5, -31.5]], dtype=float32)
Coordinates: (12/15)
  * azimuth     (azimuth) float32 0.5 1.5 2.5 3.5 ... 356.5 357.5 358.5 359.5
    elevation   (azimuth) float32 0.3 0.3 0.3 0.3 0.3 ... 0.3 0.3 0.3 0.3 0.3
    rtime       (azimuth) datetime64[ns] 2011-06-10T11:40:17.361118208 ... 20...
  * range       (range) float32 500.0 1.5e+03 2.5e+03 ... 3.185e+05 3.195e+05
    sweep_mode  <U20 'azimuth_surveillance'
    longitude   float32 4.79
    ...          ...
    x           (azimuth, range) float32 4.363 13.09 ... -2.777e+03 -2.786e+03
    y           (azimuth, range) float32 500.0 1.5e+03 ... 3.183e+05 3.193e+05
    z           (azimuth, range) float32 53.0 58.0 64.0 ... 7.691e+03 7.734e+03
    gr          (azimuth, range) float32 500.0 1.5e+03 ... 3.183e+05 3.193e+05
    rays        (azimuth, range) float32 0.5 0.5 0.5 0.5 ... 359.5 359.5 359.5
    bins        (azimuth, range) float32 500.0 1.5e+03 ... 3.185e+05 3.195e+05
Attributes:
    _Undetect:      0.0
    long_name:      Equivalent reflectivity factor H
    units:          dBZ
    standard_name:  radar_equivalent_reflectivity_factor_h

Create simple plot#

Using xarray features a simple plot can be created like this. Note the sortby('rtime') method, which sorts the radials by time.

[17]:
swp.DBZH.sortby("rtime").plot(x="range", y="rtime", add_labels=False)
[17]:
<matplotlib.collections.QuadMesh at 0x7fa793112050>
../../_images/notebooks_fileio_wradlib_odim_backend_26_1.png
[18]:
fig = pl.figure(figsize=(5, 5))
pm = swp.DBZH.wradlib.plot_ppi(proj={"latmin": 33e3}, fig=fig)
../../_images/notebooks_fileio_wradlib_odim_backend_27_0.png

Mask some values#

[19]:
swp["DBZH"] = swp["DBZH"].where(swp["DBZH"] >= 0)
swp["DBZH"].plot()
[19]:
<matplotlib.collections.QuadMesh at 0x7fa788555690>
../../_images/notebooks_fileio_wradlib_odim_backend_29_1.png

Export to ODIM and CfRadial2#

[20]:
vol.to_odim("knmi_odim.h5")
[21]:
vol.to_cfradial2("knmi_odim_as_cfradial2.nc")

Import again#

[22]:
vola = wrl.io.open_odim_dataset("knmi_odim.h5")
[23]:
volb = wrl.io.open_cfradial2_dataset("knmi_odim_as_cfradial2.nc")

Check equality#

[24]:
xr.testing.assert_allclose(vol.root, vola.root)
xr.testing.assert_equal(vol[0], vola[0])
xr.testing.assert_allclose(vol.root, volb.root)
xr.testing.assert_equal(vol[0], volb[0])
xr.testing.assert_allclose(vola.root, volb.root)
xr.testing.assert_equal(vola[0], volb[0])

More ODIM loading mechanisms#

Use xr.open_dataset to retrieve explicit group#

Warning

Since \(\omega radlib\) version 1.18 the xarray backend engines for polar radar data have been renamed and prepended with wradlib- (eg. odim -> wradlib-odim). This was necessary to avoid clashes with the new xradar-package, which will eventually replace the wradlib engines. Users have to make sure to check which engine to use for their use-case when using xarray.open_dataset. Users might install and test xradar, and check if it is already robust enough for their use-cases (by using xradar’s engine="odim".

Since \(\omega radlib\) version 1.19 the xarray backend engines for polar radar data have been deprecated. The functionality is kept until wradlib version 2.0, when the backend-code will be removed completely. wradlib is importing that functionality from xradar-package whenever and wherever necessary.

Below we use a compatibility layer in wradlib to give users the chance to adapt their code. The first minimal change is that for every backend the group-layout is conforming to the CfRadial-standard naming scheme (sweep_0, sweep_1, etc.).

Below you can inspect the main differences of the wradlib compatibility layer and the plain xradar implementation.

use wradlib compatibility layer#

[25]:
swp_a = xr.open_dataset(
    f, engine="wradlib-odim", group="sweep_13", backend_kwargs=dict(reindex_angle=False)
)
display(swp_a)
<xarray.Dataset>
Dimensions:            (azimuth: 360, range: 240)
Coordinates:
  * azimuth            (azimuth) float32 0.5 1.5 2.5 3.5 ... 357.5 358.5 359.5
    elevation          (azimuth) float32 ...
    rtime              (azimuth) datetime64[ns] ...
  * range              (range) float32 250.0 750.0 ... 1.192e+05 1.198e+05
    sweep_mode         <U20 ...
    longitude          float32 ...
    latitude           float32 ...
    altitude           float32 ...
    time               datetime64[ns] ...
Data variables:
    DBZH               (azimuth, range) float32 ...
    sweep_number       int64 ...
    prt_mode           <U7 ...
    follow_mode        <U7 ...
    sweep_fixed_angle  float32 ...
Attributes:
    fixed_angle:  25.0

use xradar backend#

[26]:
swp_b = xr.open_dataset(
    f, engine="odim", group="sweep_13", backend_kwargs=dict(reindex_angle=False)
)
display(swp_b)
<xarray.Dataset>
Dimensions:            (azimuth: 360, range: 240)
Coordinates:
  * azimuth            (azimuth) float32 0.5 1.5 2.5 3.5 ... 357.5 358.5 359.5
    elevation          (azimuth) float32 ...
    time               (azimuth) datetime64[ns] ...
  * range              (range) float32 250.0 750.0 ... 1.192e+05 1.198e+05
    longitude          float32 ...
    latitude           float32 ...
    altitude           float32 ...
Data variables:
    DBZH               (azimuth, range) float32 ...
    sweep_mode         <U20 ...
    sweep_number       int64 ...
    prt_mode           <U7 ...
    follow_mode        <U7 ...
    sweep_fixed_angle  float32 ...

Use xr.open_mfdataset to retrieve timeseries of explicit group#

[27]:
flist = ["hdf5/71_20181220_060628.pvol.h5", "hdf5/71_20181220_061228.pvol.h5"]
flist = [wrl.util.get_wradlib_data_file(f) for f in flist]
ts = xr.open_mfdataset(
    flist, engine="wradlib-odim", concat_dim="time", combine="nested", group="sweep_0"
)
display(ts)
Downloading file 'hdf5/71_20181220_060628.pvol.h5' from 'https://github.com/wradlib/wradlib-data/raw/pooch/data/hdf5/71_20181220_060628.pvol.h5' to '/home/runner/work/wradlib/wradlib/wradlib-data'.
Downloading file 'hdf5/71_20181220_061228.pvol.h5' from 'https://github.com/wradlib/wradlib-data/raw/pooch/data/hdf5/71_20181220_061228.pvol.h5' to '/home/runner/work/wradlib/wradlib/wradlib-data'.
<xarray.Dataset>
Dimensions:            (time: 2, azimuth: 360, range: 1200)
Coordinates:
  * azimuth            (azimuth) float32 0.5 1.5 2.5 3.5 ... 357.5 358.5 359.5
    elevation          (azimuth) float32 dask.array<chunksize=(360,), meta=np.ndarray>
    rtime              (time, azimuth) datetime64[ns] 2018-12-20T06:06:50.112...
  * range              (range) float32 125.0 375.0 625.0 ... 2.996e+05 2.999e+05
    sweep_mode         <U20 'azimuth_surveillance'
    longitude          float64 151.2
    latitude           float64 -33.7
    altitude           float64 195.0
  * time               (time) datetime64[ns] 2018-12-20T06:06:28.040277760 20...
Data variables: (12/16)
    DBZH               (time, azimuth, range) float32 dask.array<chunksize=(1, 360, 1200), meta=np.ndarray>
    DBZH_CLEAN         (time, azimuth, range) float32 dask.array<chunksize=(1, 360, 1200), meta=np.ndarray>
    VRADDH             (time, azimuth, range) float32 dask.array<chunksize=(1, 360, 1200), meta=np.ndarray>
    VRADH              (time, azimuth, range) float32 dask.array<chunksize=(1, 360, 1200), meta=np.ndarray>
    WRADH              (time, azimuth, range) float32 dask.array<chunksize=(1, 360, 1200), meta=np.ndarray>
    TH                 (time, azimuth, range) float32 dask.array<chunksize=(1, 360, 1200), meta=np.ndarray>
    ...                 ...
    SNRH               (time, azimuth, range) float32 dask.array<chunksize=(1, 360, 1200), meta=np.ndarray>
    CLASS              (time, azimuth, range) int8 dask.array<chunksize=(1, 360, 1200), meta=np.ndarray>
    sweep_number       (time) int64 0 0
    prt_mode           (time) <U7 'not_set' 'not_set'
    follow_mode        (time) <U7 'not_set' 'not_set'
    sweep_fixed_angle  (time) float64 0.5 0.5
Attributes:
    fixed_angle:  0.5

Use wrl.io.open_odim_mfdataset to retrieve volume timeseries#

[28]:
fpath = os.path.join(wrl.util.get_wradlib_data_path(), "hdf5/71*.h5")
f = glob.glob(fpath)
ts = wrl.io.open_odim_mfdataset(f)
display(ts)
100%|██████████| 14/14 [00:04<00:00,  3.24it/s]
[29]:
display(ts[0])
<xarray.Dataset>
Dimensions:            (time: 2, azimuth: 360, range: 1200)
Coordinates:
  * azimuth            (azimuth) float32 0.5 1.5 2.5 3.5 ... 357.5 358.5 359.5
    elevation          (azimuth) float32 dask.array<chunksize=(360,), meta=np.ndarray>
    rtime              (time, azimuth) datetime64[ns] 2018-12-20T06:06:50.112...
  * range              (range) float32 125.0 375.0 625.0 ... 2.996e+05 2.999e+05
    sweep_mode         <U20 'azimuth_surveillance'
    longitude          float64 151.2
    latitude           float64 -33.7
    altitude           float64 195.0
  * time               (time) datetime64[ns] 2018-12-20T06:06:28.040277760 20...
Data variables: (12/16)
    DBZH               (time, azimuth, range) float32 dask.array<chunksize=(1, 360, 1200), meta=np.ndarray>
    DBZH_CLEAN         (time, azimuth, range) float32 dask.array<chunksize=(1, 360, 1200), meta=np.ndarray>
    VRADDH             (time, azimuth, range) float32 dask.array<chunksize=(1, 360, 1200), meta=np.ndarray>
    VRADH              (time, azimuth, range) float32 dask.array<chunksize=(1, 360, 1200), meta=np.ndarray>
    WRADH              (time, azimuth, range) float32 dask.array<chunksize=(1, 360, 1200), meta=np.ndarray>
    TH                 (time, azimuth, range) float32 dask.array<chunksize=(1, 360, 1200), meta=np.ndarray>
    ...                 ...
    SNRH               (time, azimuth, range) float32 dask.array<chunksize=(1, 360, 1200), meta=np.ndarray>
    CLASS              (time, azimuth, range) int8 dask.array<chunksize=(1, 360, 1200), meta=np.ndarray>
    sweep_number       (time) int64 0 0
    prt_mode           (time) <U7 'not_set' 'not_set'
    follow_mode        (time) <U7 'not_set' 'not_set'
    sweep_fixed_angle  (time) float64 0.5 0.5
Attributes:
    fixed_angle:  0.5
[ ]: