diff --git a/doc/sphinx/source/input.rst b/doc/sphinx/source/input.rst
index a0c0db504b..c2febd1688 100644
--- a/doc/sphinx/source/input.rst
+++ b/doc/sphinx/source/input.rst
@@ -301,6 +301,8 @@ A list of the datasets for which a CMORizers is available is provided in the fol
 +------------------------------+------------------------------------------------------------------------------------------------------+------+-----------------+
 | ESACCI-WATERVAPOUR           | prw (Amon)                                                                                           |   3  | Python          |
 +------------------------------+------------------------------------------------------------------------------------------------------+------+-----------------+
+| ESDC                         | tas, tasmax, tasmin (Amon)                                                                           |   2  | Python          |
++------------------------------+------------------------------------------------------------------------------------------------------+------+-----------------+
 | ESRL                         | co2s (Amon)                                                                                          |   2  | NCL             |
 +------------------------------+------------------------------------------------------------------------------------------------------+------+-----------------+
 | FLUXCOM                      | gpp (Lmon)                                                                                           |   3  | Python          |
diff --git a/environment.yml b/environment.yml
index 066a641793..7f97fae73f 100644
--- a/environment.yml
+++ b/environment.yml
@@ -11,6 +11,7 @@ channels:
 dependencies:
   - pip!=21.3
   - python>=3.8
+  - aiohttp
   - cartopy
   - cdo>=1.9.7
   - cdsapi
@@ -62,6 +63,7 @@ dependencies:
   - xesmf==0.3.0
   - xgboost>1.6.1  # github.com/ESMValGroup/ESMValTool/issues/2779
   - xlsxwriter
+  - zarr
   # Python packages needed for testing
   - flake8
   - pytest >=3.9,!=6.0.0rc1,!=6.0.0
diff --git a/environment_osx.yml b/environment_osx.yml
index a9cc65c3a8..b58b8a6050 100644
--- a/environment_osx.yml
+++ b/environment_osx.yml
@@ -11,6 +11,7 @@ channels:
 dependencies:
   - pip!=21.3
   - python>=3.8
+  - aiohttp
   - cartopy
   - cdo>=1.9.7
   - cdsapi
@@ -62,6 +63,7 @@ dependencies:
   - xesmf==0.3.0
   - xgboost>1.6.1  # github.com/ESMValGroup/ESMValTool/issues/2779
   - xlsxwriter
+  - zarr
   # Python packages needed for testing
   - flake8
   - pytest >=3.9,!=6.0.0rc1,!=6.0.0
diff --git a/esmvaltool/cmorizers/data/cmor_config/ESDC.yml b/esmvaltool/cmorizers/data/cmor_config/ESDC.yml
new file mode 100644
index 0000000000..7fc4a77ff7
--- /dev/null
+++ b/esmvaltool/cmorizers/data/cmor_config/ESDC.yml
@@ -0,0 +1,26 @@
+---
+filename: 'esdc-8d-{grid}-{chunking}-{version}.zarr'
+
+attributes:
+  project_id: OBS6
+  dataset_id: ESDC
+  version: 3.0.1
+  tier: 2
+  grid: 0.25deg
+  chunking: 1x720x1440
+  # chunking: 256x128x128
+  modeling_realm: reanaly
+  source: http://data.rsc4earth.de/EarthSystemDataCube/
+  reference: 'esdc'
+  comment: ''
+
+variables:
+  tas:
+    mip: Amon
+    raw: air_temperature_2m
+  tasmax:
+    mip: Amon
+    raw: max_air_temperature_2m
+  tasmin:
+    mip: Amon
+    raw: min_air_temperature_2m
diff --git a/esmvaltool/cmorizers/data/datasets.yml b/esmvaltool/cmorizers/data/datasets.yml
index ca0abfa5dc..096fd7b693 100644
--- a/esmvaltool/cmorizers/data/datasets.yml
+++ b/esmvaltool/cmorizers/data/datasets.yml
@@ -524,6 +524,18 @@ datasets:
                     data/tcwv/dataset3_1/CDR-*/...
         All files need to be in one directory, not in yearly subdirectories.
 
+  ESDC:
+    tier: 2
+    source: http://data.rsc4earth.de/EarthSystemDataCube/
+    last_access: 2023-01-26
+    info: |
+      It is not necessary to download the data, as the cmorizer script can access
+      it directly from the cloud if it is not available locally.
+
+      To download a dataset, the dataset folder can be explored on the source
+      website, and downloaded using wget:
+        ```wget -m -nH -np -R "index.html*" http://data.rsc4earth.de/EarthSystemDataCube/v3.0.1/```
+
   ESRL:
     tier: 2
     source: http://www.esrl.noaa.gov/gmd/dv/data/index.php
diff --git a/esmvaltool/cmorizers/data/formatters/datasets/esdc.py b/esmvaltool/cmorizers/data/formatters/datasets/esdc.py
new file mode 100644
index 0000000000..bf473f53be
--- /dev/null
+++ b/esmvaltool/cmorizers/data/formatters/datasets/esdc.py
@@ -0,0 +1,149 @@
+"""ESMValTool CMORizer for Earth System Data Cube data.
+
+Tier
+    Tier 2: other freely-available dataset.
+
+Source
+    http://data.rsc4earth.de/EarthSystemDataCube/
+
+Last access
+    20230126
+
+Download and processing instructions
+    It is not necessary to download the data, as the cmorizer script can access
+    it directly from the cloud if it is not available locally.
+
+    To download a dataset, the dataset folder can be explored on the source
+    website, and downloaded using wget:
+        ```wget -m -nH -np -R "index.html*" http://data.rsc4earth.de/EarthSystemDataCube/v3.0.1/```
+"""  # noqa: E501
+import logging
+from copy import deepcopy
+from pathlib import Path
+
+import cf_units
+import iris.std_names
+import xarray as xr
+from esmvalcore.preprocessor import monthly_statistics
+
+from esmvaltool.cmorizers.data import utilities as utils
+
+logger = logging.getLogger(__name__)
+
+
+def _fix_cube(var, cube, cfg):
+    """General fixes for all cubes."""
+    cmor_info = cfg['cmor_table'].get_variable(var['mip'], var['short_name'])
+
+    # Set correct names
+    cube.var_name = cmor_info.short_name
+    if cmor_info.standard_name:
+        cube.standard_name = cmor_info.standard_name
+    cube.long_name = cmor_info.long_name
+
+    # Set calendar to gregorian instead of proleptic gregorian
+    old_unit = cube.coord('time').units
+    if old_unit.calendar == 'proleptic_gregorian':
+        logger.info("Converting time units to gregorian")
+        cube.coord('time').units = cf_units.Unit(old_unit.origin,
+                                                 calendar='gregorian')
+    utils.fix_coords(cube)
+    cube.convert_units(cmor_info.units)
+    if 'height2m' in cmor_info.dimensions:
+        utils.add_height2m(cube)
+    # Conversion from 8-d to monthly frequency
+    cube = monthly_statistics(cube, operator="mean")
+
+    # Fix metadata
+    attrs = cfg['attributes']
+    attrs['mip'] = var['mip']
+    utils.fix_var_metadata(cube, cmor_info)
+    utils.set_global_atts(cube, attrs)
+
+    return cube
+
+
+def _open_zarr(path):
+    """Open zarr dataset."""
+    logger.info('Opening zarr in "%s"', path)
+    try:
+        zarr_dataset = xr.open_dataset(path, engine='zarr')
+        return zarr_dataset
+    except KeyError as exception:
+        # Happens when the zarr folder is missing metadata, e.g. when
+        # it is a zarr array instead of a zarr dataset.
+        logger.error('Could not open zarr dataset "%s": "KeyError: %s"', path,
+                     exception)
+        raise exception
+
+
+def _extract_variable(zarr_path, var, cfg, out_dir):
+    """Open and cmorize cube."""
+    attributes = deepcopy(cfg['attributes'])
+    all_attributes = {
+        **attributes,
+        **var
+    }  # add the mip to the other attributes
+    raw_name = var['raw']
+    zarr_dataset = _open_zarr(zarr_path)
+    cube_xr = zarr_dataset[raw_name]
+
+    # Invalid standard names must be removed before converting to iris
+    standard_name = cube_xr.attrs.get('standard_name', None)
+    if (standard_name is not None
+            and standard_name not in iris.std_names.STD_NAMES):
+        del cube_xr.attrs['standard_name']
+        logger.info('Removed invalid standard name "%s".', standard_name)
+
+    cube_iris = cube_xr.to_iris()
+    cube = _fix_cube(var, cube_iris, cfg)
+
+    utils.save_variable(cube=cube,
+                        var=var['short_name'],
+                        outdir=out_dir,
+                        attrs=all_attributes,
+                        unlimited_dimensions=['time'])
+
+
+def cmorization(in_dir, out_dir, cfg, cfg_user, start_date, end_date):
+    """Cmorize the dataset."""
+    if start_date:
+        logger.warning('start_date set to "%s", but will be ignored',
+                       start_date)
+    if end_date:
+        logger.warning('end_date set to "%s", but will be ignored', end_date)
+
+    attributes = cfg['attributes']
+    variables = cfg['variables']
+    version = attributes['version']
+    filename_pattern = cfg['filename'].format(grid=attributes['grid'],
+                                              chunking=attributes['chunking'],
+                                              version=version)
+
+    local_path = Path(in_dir)
+    in_files = list(local_path.glob(filename_pattern))
+    logger.debug('Pattern %s matched: %s', Path(local_path, filename_pattern),
+                 in_files)
+
+    if len(in_files) > 1:
+        logger.warning(
+            'Pattern has matched "%i" files, '
+            'but only the first one will be used.', len(in_files))
+        logger.warning('The following files will be ignored.: "%s"',
+                       in_files[1:])
+        zarr_path = in_files[0]
+    elif len(in_files) == 0:
+        logger.info(
+            'No local matches for pattern "%s", '
+            'attempting connection to the cloud.',
+            Path(local_path, filename_pattern))
+        if '*' in filename_pattern:
+            logger.warning(
+                'Detected a wildcard character in path (*), '
+                'online connection to \"%s\" may not work', filename_pattern)
+        zarr_path = f'{attributes["source"]}/v{version}/{filename_pattern}'
+
+    for short_name, var in variables.items():
+        if 'short_name' not in var:
+            var['short_name'] = short_name
+        _extract_variable(zarr_path, var, cfg, out_dir)
diff --git a/esmvaltool/recipes/examples/recipe_check_obs.yml b/esmvaltool/recipes/examples/recipe_check_obs.yml
index 31b96ed704..78d3b37038 100644
--- a/esmvaltool/recipes/examples/recipe_check_obs.yml
+++ b/esmvaltool/recipes/examples/recipe_check_obs.yml
@@ -288,6 +288,18 @@ diagnostics:
     scripts: null
 
 
+  ESDC:
+    description: ESDC check
+    variables:
+      tas:
+      tasmin:
+      tasmax:
+    additional_datasets:
+      - {dataset: ESDC, project: OBS6, mip: Amon, tier: 2,
+         type: reanaly, version: 3.0.1,
+         start_year: 1979, end_year: 2021}
+    scripts: null
+
   ESRL:
     description: ESRL check
     variables:
diff --git a/esmvaltool/references/esdc.bibtex b/esmvaltool/references/esdc.bibtex
new file mode 100644
index 0000000000..887cb1f2bf
--- /dev/null
+++ b/esmvaltool/references/esdc.bibtex
@@ -0,0 +1,11 @@
+@article{esdc,
+    doi = {10.5194/esd-11-201-2020},
+    url = {https://esd.copernicus.org/articles/11/201/2020/},
+    year = {2020},
+    volume = {11},
+    number = {1},
+    pages = {201--234},
+    author = {Mahecha, M. D. and Gans, F. and Brandt, G. and Christiansen, R. and Cornell, S. E. and Fomferra, N. and Kraemer, G. and Peters, J. and Bodesheim, P. and Camps-Valls, G. and Donges, J. F. and Dorigo, W. and Estupinan-Suarez, L. M. and Gutierrez-Velez, V. H. and Gutwin, M. and Jung, M. and Londo\~no, M. C. and Miralles, D. G. and Papastefanou, P. and Reichstein, M.},
+    title = {Earth system data cubes unravel global multivariate dynamics},
+    journal = {Earth System Dynamics}
+}
diff --git a/setup.py b/setup.py
index bf0c37d176..4d59862c54 100755
--- a/setup.py
+++ b/setup.py
@@ -20,6 +20,7 @@
     # Installation dependencies
     # Use with pip install . to install from source
     'install': [
+        'aiohttp',
         'cartopy',
         'cdo',
         'cdsapi',
@@ -65,6 +66,7 @@
         'xesmf==0.3.0',
         'xgboost>1.6.1',  # github.com/ESMValGroup/ESMValTool/issues/2779
         'xlsxwriter',
+        'zarr',
     ],
     # Test dependencies
     # Execute `pip install .[test]` once and the use `pytest` to run tests