Tutorial

Intake-thredds provides an interface that combines functionality from siphon and intake to retrieve data from THREDDS data servers. This tutorial provides an introduction to the API and features of intake-thredds. Let’s begin by importing intake.

import intake

Loading a catalog

You can load data from a THREDDS catalog by providing the URL to a valid THREDDS catalog:

cat_url = 'https://psl.noaa.gov/thredds/catalog/Datasets/noaa.ersst/catalog.xml'
catalog = intake.open_thredds_cat(cat_url, name='noaa-ersst-catalog')
catalog

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Cell In[2], line 2
      1 cat_url = 'https://psl.noaa.gov/thredds/catalog/Datasets/noaa.ersst/catalog.xml'
----> 2 catalog = intake.open_thredds_cat(cat_url, name='noaa-ersst-catalog')
      3 catalog

TypeError: ThreddsCatalog.__new__() got an unexpected keyword argument 'name'

Using the catalog

Once you’ve loaded a catalog, you can display its contents by iterating over its entries:

list(catalog)

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[3], line 1
----> 1 list(catalog)

NameError: name 'catalog' is not defined

Once you’ve identified a dataset of interest, you can access it as follows:

source = catalog['err.mnmean.v3.nc']
print(source)

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[4], line 1
----> 1 source = catalog['err.mnmean.v3.nc']
      2 print(source)

NameError: name 'catalog' is not defined

Loading a dataset

To load a dataset of interest, you can use the to_dask() method which is available on a source object:

%%time
ds = source().to_dask()
ds

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
File <timed exec>:1

NameError: name 'source' is not defined

The to_dask() reads only metadata needed to construct an xarray.Dataset. The actual data are streamed over the network when computation routines are invoked on the dataset. By default, intake-thredds uses chunks={} to load the dataset with dask using a single chunk for all arrays. You can use a different chunking scheme by prividing a custom value of chunks before calling .to_dask():

%%time
# Use a custom chunking scheme
ds = source(chunks={'time': 100, 'lon': 90}).to_dask()
ds

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
File <timed exec>:2

NameError: name 'source' is not defined

Working with nested catalogs

In some scenarious, a THREDDS catalog can reference another THREDDS catalog. This results into a nested structure consisting of a parent catalog and children catalogs:

cat_url = 'https://psl.noaa.gov/thredds/catalog.xml'
catalog = intake.open_thredds_cat(cat_url)
list(catalog)

[]

print(list(catalog['Datasets']))

---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
Cell In[8], line 1
----> 1 print(list(catalog['Datasets']))

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/intake/readers/mixins.py:41, in PipelineMixin.__getitem__(self, item)
     36 outtype = self.output_instance
     37 if "Catalog" in outtype:
     38     # a better way to mark this condition, perhaps the datatype's structure?
     39     # TODO: this prevents from doing a transform/convert on a cat, so must use
     40     #  .transform for that
---> 41     return self.read()[item]
     42 if isinstance(self, Pipeline):
     43     return self.with_step((GetItem, (item,), {}), out_instance=outtype)

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/intake/readers/entry.py:534, in Catalog.__getitem__(self, item)
    532     return item.to_data(user_parameters=ups, **(kw or {}))
    533 else:
--> 534     raise KeyError(item)

KeyError: 0

print(list(catalog['Datasets']['ncep.reanalysis.dailyavgs']))

---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
Cell In[9], line 1
----> 1 print(list(catalog['Datasets']['ncep.reanalysis.dailyavgs']))

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/intake/readers/mixins.py:41, in PipelineMixin.__getitem__(self, item)
     36 outtype = self.output_instance
     37 if "Catalog" in outtype:
     38     # a better way to mark this condition, perhaps the datatype's structure?
     39     # TODO: this prevents from doing a transform/convert on a cat, so must use
     40     #  .transform for that
---> 41     return self.read()[item]
     42 if isinstance(self, Pipeline):
     43     return self.with_step((GetItem, (item,), {}), out_instance=outtype)

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/intake/readers/entry.py:534, in Catalog.__getitem__(self, item)
    532     return item.to_data(user_parameters=ups, **(kw or {}))
    533 else:
--> 534     raise KeyError(item)

KeyError: 0

To load data from such a nested catalog, intake-thredds provides a special source object THREDDSMergedSource accessible via the .open_thredds_merged() function. The inputs for this function consists of:

• url: top level URL of the THREDDS catalog

• path: a list of paths for child catalogs to descend down. The paths can include glob characters (*, ?). These glob characters are used for matching.

source = intake.open_thredds_merged(
    cat_url, path=['Datasets', 'ncep.reanalysis.dailyavgs', 'surface', 'air*sig995*194*.nc']
)
print(source)

<intake_thredds.source.THREDDSMergedSource object at 0x7f62552e3510>

To load the data into an xarray Dataset, you can invoke the .to_dask() method. Internally, THREDDSMergedSource does the following:

• descend down the given paths and collect all available datasets.

• load each dataset in a dataset.

• combine all loaded datasets into a single dataset.

%%time
ds = source.to_dask()
ds

Dataset(s):   0%|                                        | 0/2 [00:00<?, ?it/s]

/home/docs/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/pydap/handlers/dap.py:123: UserWarning: PyDAP was unable to determine the DAP protocol defaulting to DAP2 which is consider legacy and may result in slower responses. For more, see go to https://www.opendap.org/faq-page.
  _warnings.warn(

Dataset(s):  50%|████████████████                | 1/2 [00:01<00:01,  1.94s/it]

/home/docs/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/pydap/handlers/dap.py:123: UserWarning: PyDAP was unable to determine the DAP protocol defaulting to DAP2 which is consider legacy and may result in slower responses. For more, see go to https://www.opendap.org/faq-page.
  _warnings.warn(

Dataset(s): 100%|████████████████████████████████| 2/2 [00:03<00:00,  1.71s/it]

Dataset(s): 100%|████████████████████████████████| 2/2 [00:03<00:00,  1.74s/it]

CPU times: user 1.14 s, sys: 105 ms, total: 1.24 s
Wall time: 4.58 s

<xarray.Dataset> Size: 31MB
Dimensions:    (time: 731, lat: 73, lon: 144, nbnds: 2)
Coordinates:
  * lon        (lon) float32 576B 0.0 2.5 5.0 7.5 ... 350.0 352.5 355.0 357.5
  * time       (time) datetime64[ns] 6kB 1948-01-01 1948-01-02 ... 1949-12-31
  * lat        (lat) float32 292B 90.0 87.5 85.0 82.5 ... -85.0 -87.5 -90.0
Dimensions without coordinates: nbnds
Data variables:
    air        (time, lat, lon) float32 31MB dask.array<chunksize=(366, 73, 144), meta=np.ndarray>
    time_bnds  (time, nbnds) float64 12kB dask.array<chunksize=(366, 2), meta=np.ndarray>
Attributes:
    Conventions:                     COARDS
    title:                           mean daily NMC reanalysis (1948)
    description:                     Data is from NMC initialized reanalysis\...
    platform:                        Model
    history:                         created 99/05/11 by Hoop (netCDF2.3)\nCo...
    dataset_title:                   NCEP-NCAR Reanalysis 1
    References:                      http://www.psl.noaa.gov/data/gridded/dat...
    _NCProperties:                   version=2,netcdf=4.6.3,hdf5=1.10.5
    DODS_EXTRA.Unlimited_Dimension:  time

xarray.Dataset

• Dimensions:
  • time: 731
  • lat: 73
  • lon: 144
  • nbnds: 2
• Coordinates: (3)
  • lon
    (lon)
    float32
    0.0 2.5 5.0 ... 352.5 355.0 357.5

    units :

    degrees_east

    long_name :

    Longitude

    actual_range :

    [0.0, 357.5]

    standard_name :

    longitude

    axis :

    X

    array([  0. ,   2.5,   5. ,   7.5,  10. ,  12.5,  15. ,  17.5,  20. ,  22.5,
            25. ,  27.5,  30. ,  32.5,  35. ,  37.5,  40. ,  42.5,  45. ,  47.5,
            50. ,  52.5,  55. ,  57.5,  60. ,  62.5,  65. ,  67.5,  70. ,  72.5,
            75. ,  77.5,  80. ,  82.5,  85. ,  87.5,  90. ,  92.5,  95. ,  97.5,
           100. , 102.5, 105. , 107.5, 110. , 112.5, 115. , 117.5, 120. , 122.5,
           125. , 127.5, 130. , 132.5, 135. , 137.5, 140. , 142.5, 145. , 147.5,
           150. , 152.5, 155. , 157.5, 160. , 162.5, 165. , 167.5, 170. , 172.5,
           175. , 177.5, 180. , 182.5, 185. , 187.5, 190. , 192.5, 195. , 197.5,
           200. , 202.5, 205. , 207.5, 210. , 212.5, 215. , 217.5, 220. , 222.5,
           225. , 227.5, 230. , 232.5, 235. , 237.5, 240. , 242.5, 245. , 247.5,
           250. , 252.5, 255. , 257.5, 260. , 262.5, 265. , 267.5, 270. , 272.5,
           275. , 277.5, 280. , 282.5, 285. , 287.5, 290. , 292.5, 295. , 297.5,
           300. , 302.5, 305. , 307.5, 310. , 312.5, 315. , 317.5, 320. , 322.5,
           325. , 327.5, 330. , 332.5, 335. , 337.5, 340. , 342.5, 345. , 347.5,
           350. , 352.5, 355. , 357.5], dtype=float32)

  • time
    (time)
    datetime64[ns]
    1948-01-01 ... 1949-12-31

    long_name :

    Time

    delta_t :

    0000-00-01 00:00:00

    avg_period :

    0000-00-01 00:00:00

    standard_name :

    time

    axis :

    T

    actual_range :

    [1297320.0, 1306080.0]

    coordinate_defines :

    start

    _ChunkSizes :

    1

    array(['1948-01-01T00:00:00.000000000', '1948-01-02T00:00:00.000000000',
           '1948-01-03T00:00:00.000000000', ..., '1949-12-29T00:00:00.000000000',
           '1949-12-30T00:00:00.000000000', '1949-12-31T00:00:00.000000000'],
          shape=(731,), dtype='datetime64[ns]')

  • lat
    (lat)
    float32
    90.0 87.5 85.0 ... -87.5 -90.0

    units :

    degrees_north

    actual_range :

    [90.0, -90.0]

    long_name :

    Latitude

    standard_name :

    latitude

    axis :

    Y

    array([ 90. ,  87.5,  85. ,  82.5,  80. ,  77.5,  75. ,  72.5,  70. ,  67.5,
            65. ,  62.5,  60. ,  57.5,  55. ,  52.5,  50. ,  47.5,  45. ,  42.5,
            40. ,  37.5,  35. ,  32.5,  30. ,  27.5,  25. ,  22.5,  20. ,  17.5,
            15. ,  12.5,  10. ,   7.5,   5. ,   2.5,   0. ,  -2.5,  -5. ,  -7.5,
           -10. , -12.5, -15. , -17.5, -20. , -22.5, -25. , -27.5, -30. , -32.5,
           -35. , -37.5, -40. , -42.5, -45. , -47.5, -50. , -52.5, -55. , -57.5,
           -60. , -62.5, -65. , -67.5, -70. , -72.5, -75. , -77.5, -80. , -82.5,
           -85. , -87.5, -90. ], dtype=float32)

• Data variables: (2)
  • air
    (time, lat, lon)
    float32
    dask.array<chunksize=(366, 73, 144), meta=np.ndarray>

    long_name :

    mean Daily Air temperature at sigma level 995

    units :

    degK

    precision :

    2

    least_significant_digit :

    1

    GRIB_id :

    11

    GRIB_name :

    TMP

    var_desc :

    Air temperature

    dataset :

    NCEP Reanalysis Daily Averages

    statistic :

    Mean

    parent_stat :

    Individual Obs

    actual_range :

    [188.29001, 315.3]

    valid_range :

    [185.16, 331.16]

    level_desc :

    0.995 sigma

    _ChunkSizes :

    [1, 73, 144]

    Array Chunk Bytes 29.31 MiB 14.68 MiB Shape (731, 73, 144) (366, 73, 144) Dask graph 2 chunks in 5 graph layers Data type float32 numpy.ndarray

  • time_bnds
    (time, nbnds)
    float64
    dask.array<chunksize=(366, 2), meta=np.ndarray>

    _ChunkSizes :

    [1, 2]

    Array Chunk Bytes 11.42 kiB 5.72 kiB Shape (731, 2) (366, 2) Dask graph 2 chunks in 5 graph layers Data type float64 numpy.ndarray

• Indexes: (3)
  • lon
    PandasIndex

    PandasIndex(Index([  0.0,   2.5,   5.0,   7.5,  10.0,  12.5,  15.0,  17.5,  20.0,  22.5,
           ...
           335.0, 337.5, 340.0, 342.5, 345.0, 347.5, 350.0, 352.5, 355.0, 357.5],
          dtype='float32', name='lon', length=144))

  • time
    PandasIndex

    PandasIndex(DatetimeIndex(['1948-01-01', '1948-01-02', '1948-01-03', '1948-01-04',
                   '1948-01-05', '1948-01-06', '1948-01-07', '1948-01-08',
                   '1948-01-09', '1948-01-10',
                   ...
                   '1949-12-22', '1949-12-23', '1949-12-24', '1949-12-25',
                   '1949-12-26', '1949-12-27', '1949-12-28', '1949-12-29',
                   '1949-12-30', '1949-12-31'],
                  dtype='datetime64[ns]', name='time', length=731, freq=None))

  • lat
    PandasIndex

    PandasIndex(Index([ 90.0,  87.5,  85.0,  82.5,  80.0,  77.5,  75.0,  72.5,  70.0,  67.5,
            65.0,  62.5,  60.0,  57.5,  55.0,  52.5,  50.0,  47.5,  45.0,  42.5,
            40.0,  37.5,  35.0,  32.5,  30.0,  27.5,  25.0,  22.5,  20.0,  17.5,
            15.0,  12.5,  10.0,   7.5,   5.0,   2.5,   0.0,  -2.5,  -5.0,  -7.5,
           -10.0, -12.5, -15.0, -17.5, -20.0, -22.5, -25.0, -27.5, -30.0, -32.5,
           -35.0, -37.5, -40.0, -42.5, -45.0, -47.5, -50.0, -52.5, -55.0, -57.5,
           -60.0, -62.5, -65.0, -67.5, -70.0, -72.5, -75.0, -77.5, -80.0, -82.5,
           -85.0, -87.5, -90.0],
          dtype='float32', name='lat'))

• Attributes: (9)

  Conventions :

  COARDS

  title :

  mean daily NMC reanalysis (1948)

  description :

  Data is from NMC initialized reanalysis (4x/day). These are the 0.9950 sigma level values.

  platform :

  Model

  history :

  created 99/05/11 by Hoop (netCDF2.3) Converted to chunked, deflated non-packed NetCDF4 2014/09

  dataset_title :

  NCEP-NCAR Reanalysis 1

  References :

  http://www.psl.noaa.gov/data/gridded/data.ncep.reanalysis.html

  _NCProperties :

  version=2,netcdf=4.6.3,hdf5=1.10.5

  DODS_EXTRA.Unlimited_Dimension :

  time

Caching

Under the hood intake-thredds uses the driver='opendap' from intake-xarray by default. You can also choose driver='netcdf', which in combination with fsspec caches files by appending simplecache:: to the url, see fsspec docs.

import os

import fsspec

# specify caching location, where to store files to with their original names
fsspec.config.conf['simplecache'] = {'cache_storage': 'my_caching_folder', 'same_names': True}

cat_url = 'https://psl.noaa.gov/thredds/catalog.xml'
source = intake.open_thredds_merged(
    f'simplecache::{cat_url}',
    path=['Datasets', 'ncep.reanalysis.dailyavgs', 'surface', 'air.sig995.194*.nc'],
    driver='netcdf',  # specify netcdf driver to open HTTPServer
)
print(source)

<intake_thredds.source.THREDDSMergedSource object at 0x7f625513b110>

%time ds = source.to_dask()

---------------------------------------------------------------------------
InvalidSchema                             Traceback (most recent call last)
File <timed exec>:1

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/intake_thredds/source.py:83, in THREDDSMergedSource.read(self, xarray_kwargs)
     81 part = self.path[i]
     82 if '*' not in part and '?' not in part:
---> 83     cat = cat[part].read(make=self.driver[-3:])
     84 else:
     85     break

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/intake/readers/readers.py:121, in BaseReader.read(self, *args, **kwargs)
    119 kw.update(kwargs)
    120 args = kw.pop("args", ()) or args
--> 121 return self._read(*args, **kw)

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/intake/readers/catalogs.py:423, in THREDDSCatalogReader._read(self, data, make, **kwargs)
    419 from siphon.catalog import TDSCatalog
    421 from intake.readers.readers import XArrayDatasetReader
--> 423 thr = TDSCatalog(data.url)
    424 cat = THREDDSCatalog(metadata=thr.metadata)
    425 for r in thr.catalog_refs.values():

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/siphon/catalog.py:288, in TDSCatalog.__init__(self, catalog_url)
    285 self.session = session_manager.create_session()
    287 # get catalog.xml file
--> 288 resp = self.session.get(catalog_url)
    289 resp.raise_for_status()
    291 # top level server url

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/requests/sessions.py:602, in Session.get(self, url, **kwargs)
    594 r"""Sends a GET request. Returns :class:`Response` object.
    595 
    596 :param url: URL for the new :class:`Request` object.
    597 :param \*\*kwargs: Optional arguments that ``request`` takes.
    598 :rtype: requests.Response
    599 """
    601 kwargs.setdefault("allow_redirects", True)
--> 602 return self.request("GET", url, **kwargs)

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/requests/sessions.py:589, in Session.request(self, method, url, params, data, headers, cookies, files, auth, timeout, allow_redirects, proxies, hooks, stream, verify, cert, json)
    584 send_kwargs = {
    585     "timeout": timeout,
    586     "allow_redirects": allow_redirects,
    587 }
    588 send_kwargs.update(settings)
--> 589 resp = self.send(prep, **send_kwargs)
    591 return resp

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/requests/sessions.py:697, in Session.send(self, request, **kwargs)
    694 hooks = request.hooks
    696 # Get the appropriate adapter to use
--> 697 adapter = self.get_adapter(url=request.url)
    699 # Start time (approximately) of the request
    700 start = preferred_clock()

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/requests/sessions.py:792, in Session.get_adapter(self, url)
    789         return adapter
    791 # Nothing matches :-/
--> 792 raise InvalidSchema(f"No connection adapters were found for {url!r}")

InvalidSchema: No connection adapters were found for 'simplecache::https://psl.noaa.gov/thredds/catalog/Datasets/catalog.xml'

assert os.path.exists('my_caching_folder/air.sig995.1949.nc')

---------------------------------------------------------------------------
AssertionError                            Traceback (most recent call last)
Cell In[14], line 1
----> 1 assert os.path.exists('my_caching_folder/air.sig995.1949.nc')

AssertionError: 

# after caching very fast
%time ds = source.to_dask()

---------------------------------------------------------------------------
InvalidSchema                             Traceback (most recent call last)
File <timed exec>:1

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/intake_thredds/source.py:83, in THREDDSMergedSource.read(self, xarray_kwargs)
     81 part = self.path[i]
     82 if '*' not in part and '?' not in part:
---> 83     cat = cat[part].read(make=self.driver[-3:])
     84 else:
     85     break

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/intake/readers/readers.py:121, in BaseReader.read(self, *args, **kwargs)
    119 kw.update(kwargs)
    120 args = kw.pop("args", ()) or args
--> 121 return self._read(*args, **kw)

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/intake/readers/catalogs.py:423, in THREDDSCatalogReader._read(self, data, make, **kwargs)
    419 from siphon.catalog import TDSCatalog
    421 from intake.readers.readers import XArrayDatasetReader
--> 423 thr = TDSCatalog(data.url)
    424 cat = THREDDSCatalog(metadata=thr.metadata)
    425 for r in thr.catalog_refs.values():

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/siphon/catalog.py:288, in TDSCatalog.__init__(self, catalog_url)
    285 self.session = session_manager.create_session()
    287 # get catalog.xml file
--> 288 resp = self.session.get(catalog_url)
    289 resp.raise_for_status()
    291 # top level server url

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/requests/sessions.py:602, in Session.get(self, url, **kwargs)
    594 r"""Sends a GET request. Returns :class:`Response` object.
    595 
    596 :param url: URL for the new :class:`Request` object.
    597 :param \*\*kwargs: Optional arguments that ``request`` takes.
    598 :rtype: requests.Response
    599 """
    601 kwargs.setdefault("allow_redirects", True)
--> 602 return self.request("GET", url, **kwargs)

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/requests/sessions.py:589, in Session.request(self, method, url, params, data, headers, cookies, files, auth, timeout, allow_redirects, proxies, hooks, stream, verify, cert, json)
    584 send_kwargs = {
    585     "timeout": timeout,
    586     "allow_redirects": allow_redirects,
    587 }
    588 send_kwargs.update(settings)
--> 589 resp = self.send(prep, **send_kwargs)
    591 return resp

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/requests/sessions.py:697, in Session.send(self, request, **kwargs)
    694 hooks = request.hooks
    696 # Get the appropriate adapter to use
--> 697 adapter = self.get_adapter(url=request.url)
    699 # Start time (approximately) of the request
    700 start = preferred_clock()

File ~/checkouts/readthedocs.org/user_builds/intake-thredds/conda/latest/lib/python3.11/site-packages/requests/sessions.py:792, in Session.get_adapter(self, url)
    789         return adapter
    791 # Nothing matches :-/
--> 792 raise InvalidSchema(f"No connection adapters were found for {url!r}")

InvalidSchema: No connection adapters were found for 'simplecache::https://psl.noaa.gov/thredds/catalog/Datasets/catalog.xml'

Multi-file concat_kwargs with cfgrib engine

Another example demonstrating how to use caching consists of reading in ensemble members in parallel for GEFS. The example below reads in 21 ensemble members for a single timestep. It also demonstrates usage of xarray_kwargs which are passed on to xarray for opening the files, which in this cases uses the cfgrib engine:

cat_url = 'https://www.ncei.noaa.gov/thredds/catalog/model-gefs-003/202008/20200831/catalog.xml'
source = intake.open_thredds_merged(
    f'simplecache::{cat_url}',
    path=["NCEP gens-a Grid 3 Member-Forecast *-372 for 2020-08-31 00:00*"],
    driver="netcdf",
    concat_kwargs={"dim": "number"},
    xarray_kwargs=dict(
        engine="cfgrib",
        backend_kwargs=dict(
            filter_by_keys={"typeOfLevel": "heightAboveGround", "cfVarName": "t2m"}
        ),
    ),
)
source.to_dask()

Dataset(s):   0%|                                       | 0/21 [00:00<?, ?it/s]

Dataset(s):   5%|█▍                             | 1/21 [00:00<00:16,  1.23it/s]

Dataset(s):  10%|██▉                            | 2/21 [00:01<00:19,  1.00s/it]

Dataset(s):  14%|████▍                          | 3/21 [00:03<00:18,  1.04s/it]

Dataset(s):  19%|█████▉                         | 4/21 [00:03<00:15,  1.07it/s]

Dataset(s):  24%|███████▍                       | 5/21 [00:04<00:13,  1.15it/s]

Dataset(s):  29%|████████▊                      | 6/21 [00:06<00:19,  1.28s/it]

Dataset(s):  33%|██████████▎                    | 7/21 [00:07<00:15,  1.10s/it]

Dataset(s):  38%|███████████▊                   | 8/21 [00:08<00:12,  1.00it/s]

Dataset(s):  43%|█████████████▎                 | 9/21 [00:08<00:10,  1.10it/s]

Dataset(s):  48%|██████████████▎               | 10/21 [00:09<00:09,  1.16it/s]

Dataset(s):  52%|███████████████▋              | 11/21 [00:10<00:08,  1.19it/s]

Dataset(s):  57%|█████████████████▏            | 12/21 [00:11<00:07,  1.23it/s]

Dataset(s):  62%|██████████████████▌           | 13/21 [00:11<00:06,  1.25it/s]

Dataset(s):  67%|████████████████████          | 14/21 [00:12<00:05,  1.25it/s]

Dataset(s):  71%|█████████████████████▍        | 15/21 [00:13<00:04,  1.28it/s]

Dataset(s):  76%|██████████████████████▊       | 16/21 [00:14<00:03,  1.29it/s]

Dataset(s):  81%|████████████████████████▎     | 17/21 [00:14<00:03,  1.30it/s]

Dataset(s):  86%|█████████████████████████▋    | 18/21 [00:15<00:02,  1.27it/s]

Dataset(s):  90%|███████████████████████████▏  | 19/21 [00:16<00:01,  1.26it/s]

Dataset(s):  95%|████████████████████████████▌ | 20/21 [00:17<00:00,  1.27it/s]

Dataset(s): 100%|██████████████████████████████| 21/21 [00:18<00:00,  1.24it/s]

Dataset(s): 100%|██████████████████████████████| 21/21 [00:18<00:00,  1.15it/s]

<xarray.Dataset> Size: 0B
Dimensions:  ()
Data variables:
    *empty*
Attributes:
    Conventions:  CF-1.7
    history:      2025-01-15T14:16 GRIB to CDM+CF via cfgrib-0.9.15.0/ecCodes...

xarray.Dataset

• Dimensions:
• Coordinates: (0)
• Data variables: (0)
• Indexes: (0)
• Attributes: (2)

  Conventions :

  CF-1.7

  history :

  2025-01-15T14:16 GRIB to CDM+CF via cfgrib-0.9.15.0/ecCodes-2.39.0 with {"source": "N/A", "filter_by_keys": {"typeOfLevel": "heightAboveGround", "cfVarName": "t2m"}, "encode_cf": ["parameter", "time", "geography", "vertical"]}