Using UK Met Office PP files

We fetch the input data as a sample source.

import earthkit.data as ekd

d = ekd.from_source("sample", "hadslp2.pp")
d

PP file

---

path	/var/folders/93/w0p869rx17q98wxk83gn9ys40000gn/T/earthkit-data-cgr/url-183fd439041d5a7ad2ba82b0bb1d766e42513b8d7fdfbfa3ddb18e5071abeaa4.pp
size	19.4 MiB
types	xarray, fieldlist

We can convert the data to an Xarray dataset. It is doen by using the Iris <https://scitools-iris.readthedocs.io/en/latest/>_ library under the hood. Please note, PP file support requires the scitools-iris and ncdata packages to be installed.

d.to_xarray()

/Users/cgr/venv/dev/lib/python3.13/site-packages/iris/fileformats/netcdf/saver.py:2877: IrisDeprecation: Saving to netcdf with legacy-style attribute handling for backwards compatibility.
This mode is deprecated since Iris 3.8, and will eventually be removed.
Please consider enabling the new split-attributes handling mode, by setting 'iris.FUTURE.save_split_attrs = True'.
  warn_deprecated(message)

<xarray.Dataset> Size: 20MB
Dimensions:                     (time: 1860, grid_latitude: 37,
                                 grid_longitude: 72)
Coordinates:
    time                        (time) datetime64[ns] 15kB dask.array<chunksize=(1860,), meta=np.ndarray>
    grid_latitude               (grid_latitude) float32 148B dask.array<chunksize=(37,), meta=np.ndarray>
    grid_longitude              (grid_longitude) float32 288B dask.array<chunksize=(72,), meta=np.ndarray>
Data variables:
    unknown                     (time, grid_latitude, grid_longitude) float32 20MB dask.array<chunksize=(1, 37, 72), meta=np.ndarray>
    rotated_latitude_longitude  int32 4B dask.array<chunksize=(), meta=numpy.ma.MaskedArray>
Attributes:
    Conventions:  CF-1.7

xarray.Dataset

• Dimensions:
  • time: 1860
  • grid_latitude: 37
  • grid_longitude: 72
• Coordinates: (3)
  • time
    (time)
    datetime64[ns]
    dask.array<chunksize=(1860,), meta=np.ndarray>

    axis :

    T

    standard_name :

    time

    Array Chunk Bytes 14.53 kiB 14.53 kiB Shape (1860,) (1860,) Dask graph 1 chunks in 3 graph layers Data type datetime64[ns] numpy.ndarray

  • grid_latitude
    (grid_latitude)
    float32
    dask.array<chunksize=(37,), meta=np.ndarray>

    axis :

    Y

    units :

    degrees

    standard_name :

    grid_latitude

    Array Chunk Bytes 148 B 148 B Shape (37,) (37,) Dask graph 1 chunks in 2 graph layers Data type float32 numpy.ndarray

  • grid_longitude
    (grid_longitude)
    float32
    dask.array<chunksize=(72,), meta=np.ndarray>

    axis :

    X

    units :

    degrees

    standard_name :

    grid_longitude

    Array Chunk Bytes 288 B 288 B Shape (72,) (72,) Dask graph 1 chunks in 2 graph layers Data type float32 numpy.ndarray

• Data variables: (2)
  • unknown
    (time, grid_latitude, grid_longitude)
    float32
    dask.array<chunksize=(1, 37, 72), meta=np.ndarray>

    ukmo__process_flags :

    Difference_from_another_experiment

    grid_mapping :

    rotated_latitude_longitude

    Array Chunk Bytes 18.90 MiB 10.41 kiB Shape (1860, 37, 72) (1, 37, 72) Dask graph 1860 chunks in 3722 graph layers Data type float32 numpy.ndarray

  • rotated_latitude_longitude
    ()
    int32
    dask.array<chunksize=(), meta=numpy.ma.MaskedArray>

    grid_mapping_name :

    rotated_latitude_longitude

    longitude_of_prime_meridian :

    0.0

    earth_radius :

    6371229.0

    grid_north_pole_latitude :

    0.0

    grid_north_pole_longitude :

    0.0

    north_pole_grid_longitude :

    0.0

    Array Chunk Bytes 4 B 4 B Shape () () Dask graph 1 chunks in 6 graph layers Data type int32 numpy.ma.MaskedArray

• Attributes: (1)

  Conventions :

  CF-1.7

We can also load PP data into a fieldlist.

fl = d.to_fieldlist()

/Users/cgr/venv/dev/lib/python3.13/site-packages/iris/fileformats/netcdf/saver.py:2877: IrisDeprecation: Saving to netcdf with legacy-style attribute handling for backwards compatibility.
This mode is deprecated since Iris 3.8, and will eventually be removed.
Please consider enabling the new split-attributes handling mode, by setting 'iris.FUTURE.save_split_attrs = True'.
  warn_deprecated(message)

fl.ls()

	parameter.variable	time.valid_datetime	time.base_datetime	time.step	vertical.level	vertical.level_type	ensemble.member	geography.grid_type
0	unknown	1850-01-01	1850-01-01	0 days	None	unknown	0	None
1	unknown	1850-02-01	1850-02-01	0 days	None	unknown	0	None
2	unknown	1850-03-01	1850-03-01	0 days	None	unknown	0	None
3	unknown	1850-04-01	1850-04-01	0 days	None	unknown	0	None
4	unknown	1850-05-01	1850-05-01	0 days	None	unknown	0	None
...	...	...	...	...	...	...	...	...
1855	unknown	2004-08-01	2004-08-01	0 days	None	unknown	0	None
1856	unknown	2004-09-01	2004-09-01	0 days	None	unknown	0	None
1857	unknown	2004-10-01	2004-10-01	0 days	None	unknown	0	None
1858	unknown	2004-11-01	2004-11-01	0 days	None	unknown	0	None
1859	unknown	2004-12-01	2004-12-01	0 days	None	unknown	0	None

1860 rows × 8 columns

fl[0].shape

(37, 72)

fl[0].values[:5]

array([1028.6418, 1028.6418, 1028.6418, 1028.6418, 1028.6418],
      dtype=float32)

fl[0].geography.latlons()

(array([[ 90.,  90.,  90., ...,  90.,  90.,  90.],
        [ 85.,  85.,  85., ...,  85.,  85.,  85.],
        [ 80.,  80.,  80., ...,  80.,  80.,  80.],
        ...,
        [-80., -80., -80., ..., -80., -80., -80.],
        [-85., -85., -85., ..., -85., -85., -85.],
        [-90., -90., -90., ..., -90., -90., -90.]],
       shape=(37, 72), dtype=float32),
 array([[-180., -175., -170., ...,  165.,  170.,  175.],
        [-180., -175., -170., ...,  165.,  170.,  175.],
        [-180., -175., -170., ...,  165.,  170.,  175.],
        ...,
        [-180., -175., -170., ...,  165.,  170.,  175.],
        [-180., -175., -170., ...,  165.,  170.,  175.],
        [-180., -175., -170., ...,  165.,  170.,  175.]],
       shape=(37, 72), dtype=float32))