CoverageJSON: retrieving vertical profiles with Polytope

The polytope source provides access to the Polytope web services.

The following example retrieves vertical profiles on pressure levels using polytope.

import earthkit.data as ekd

location = [38.9, -9.1] # lat, lon

request = {
    "class": "od",
    "stream": "enfo",
    "type": "pf",
    "date": -1,
    "time": "0000",
    "levtype": "pl",
    "expver": 1,
    "domain": "g",
    "param": "203/133",
    "number": "1",
    "step": "0",
    "levelist": "1/to/1000",
    "feature": {
        "type": "verticalprofile",
        "points": [location],
    },
}

ds = ekd.from_source("polytope", "ecmwf-mars", request=request, stream=False, address='polytope.ecmwf.int')

The data is retrieved as CoverageJSON, which we can convert to Xarray for further processing (the conversion uses the covjsonkit package under the hood).

ds.to_xarray()

<xarray.Dataset> Size: 640B
Dimensions:    (latitude: 1, longitude: 1, number: 1, datetime: 1, time: 1,
                levelist: 22)
Coordinates:
  * latitude   (latitude) float64 8B 38.91
  * longitude  (longitude) float64 8B 350.9
  * number     (number) int64 8B 1
  * datetime   (datetime) <U20 80B '2025-10-21T00:00:00Z'
  * time       (time) int64 8B 0
  * levelist   (levelist) int64 176B 1 2 3 5 7 10 ... 500 600 700 850 925 1000
Data variables:
    q          (latitude, longitude, number, datetime, time, levelist) float64 176B ...
    o3         (latitude, longitude, number, datetime, time, levelist) float64 176B ...
Attributes:
    class:          od
    Forecast date:  2025-10-21T00:00:00Z
    domain:         g
    expver:         0001
    levelist:       1
    levtype:        pl
    number:         1
    stream:         enfo
    type:           pf

xarray.Dataset

• Dimensions:
  • latitude: 1
  • longitude: 1
  • number: 1
  • datetime: 1
  • time: 1
  • levelist: 22
• Coordinates: (6)
  • latitude
    (latitude)
    float64
    38.91

    array([38.910368])

  • longitude
    (longitude)
    float64
    350.9

    array([350.889193])

  • number
    (number)
    int64
    1

    array([1])

  • datetime
    (datetime)
    <U20
    '2025-10-21T00:00:00Z'

    array(['2025-10-21T00:00:00Z'], dtype='<U20')

  • time
    (time)
    int64
    0

    array([0])

  • levelist
    (levelist)
    int64
    1 2 3 5 7 ... 600 700 850 925 1000

    array([   1,    2,    3,    5,    7,   10,   20,   30,   50,   70,  100,  150,
            200,  250,  300,  400,  500,  600,  700,  850,  925, 1000])

• Data variables: (2)
  • q
    (latitude, longitude, number, datetime, time, levelist)
    float64
    4.374e-06 4.079e-06 ... 0.01231

    type :

    Parameter

    units :

    kg kg**-1

    long_name :

    q

    array([[[[[[4.37405072e-06, 4.07915559e-06, 3.77062770e-06,
                3.53311952e-06, 3.44785440e-06, 3.48169488e-06,
                3.37518122e-06, 3.01450564e-06, 2.87986950e-06,
                3.06835682e-06, 3.43844295e-06, 7.81938434e-06,
                2.66581774e-05, 1.15647912e-04, 3.50624323e-04,
                9.18745995e-04, 1.71685219e-03, 7.50303268e-04,
                6.17456436e-03, 1.00030899e-02, 1.16314888e-02,
                1.23100281e-02]]]]]])

  • o3
    (latitude, longitude, number, datetime, time, levelist)
    float64
    5.122e-06 9.796e-06 ... 6.814e-08

    type :

    Parameter

    units :

    kg kg**-1

    long_name :

    o3

    array([[[[[[5.12171846e-06, 9.79590550e-06, 1.19635743e-05,
                1.26918603e-05, 1.23685068e-05, 1.18499456e-05,
                9.15313376e-06, 5.98923452e-06, 3.21018797e-06,
                1.78527000e-06, 7.77188689e-07, 1.81316864e-07,
                1.08251697e-07, 1.01019396e-07, 8.85629561e-08,
                8.67730705e-08, 8.37026164e-08, 8.43447150e-08,
                7.81721496e-08, 6.56430785e-08, 6.55188686e-08,
                6.81388244e-08]]]]]])

• Indexes: (6)
  • latitude
    PandasIndex

    PandasIndex(Index([38.910368186756], dtype='float64', name='latitude'))

  • longitude
    PandasIndex

    PandasIndex(Index([350.889192886457], dtype='float64', name='longitude'))

  • number
    PandasIndex

    PandasIndex(Index([1], dtype='int64', name='number'))

  • datetime
    PandasIndex

    PandasIndex(Index(['2025-10-21T00:00:00Z'], dtype='object', name='datetime'))

  • time
    PandasIndex

    PandasIndex(Index([0], dtype='int64', name='time'))

  • levelist
    PandasIndex

    PandasIndex(Index([   1,    2,    3,    5,    7,   10,   20,   30,   50,   70,  100,  150,
            200,  250,  300,  400,  500,  600,  700,  850,  925, 1000],
          dtype='int64', name='levelist'))

• Attributes: (9)

  class :

  od

  Forecast date :

  2025-10-21T00:00:00Z

  domain :

  g

  expver :

  0001

  levelist :

  1

  levtype :

  pl

  number :

  1

  stream :

  enfo

  type :

  pf