CORDEX Domains

CORDEX Regional Climate Model Data on Single Levels - Domain Reference

Full details of dataset: https://cds.climate.copernicus.eu/datasets/projections-cordex-domains-single-levels?tab=overview

CORDEX (Coordinated Regional Climate Downscaling Experiment) defines specific regional domains for climate projections. Each domain covers a geographic region at one or more resolutions.

Available CORDEX Domains

Domain Code	Region Name	Bounding Box (min_lon, min_lat, max_lon, max_lat)	Resolution	Approx. Grid Spacing
AFR-44	Africa	-24.64, -45.76, 60.28, 42.24	0.44°	~50 km
AFR-22	Africa	-24.64, -45.76, 60.28, 42.24	0.22°	~25 km
ANT-44	Antarctica	-180.0, -89.5, 180.0, -60.0	0.44°	~50 km
ARC-44	Arctic	-180.0, 60.0, 180.0, 90.0	0.44°	~50 km
AUS-44	Australasia	89.5, -52.36, 179.99, 12.21	0.44°	~50 km
AUS-22	Australasia	89.5, -52.36, 179.99, 12.21	0.22°	~25 km
CAM-44	Central America	-122.0, -19.76, -59.52, 34.24	0.44°	~50 km
CAM-22	Central America	-122.0, -19.76, -59.52, 34.24	0.22°	~25 km
CAS-44	Central Asia	34.0, 18.0, 115.0, 70.0	0.44°	~50 km
CAS-22	Central Asia	34.0, 18.0, 115.0, 70.0	0.22°	~25 km
EAS-44	East Asia	65.0, -15.0, 155.0, 65.0	0.44°	~50 km
EAS-22	East Asia	65.0, -15.0, 155.0, 65.0	0.22°	~25 km
EUR-44	Europe	-44.0, 22.0, 65.0, 72.0	0.44°	~50 km
EUR-22	Europe	-44.0, 22.0, 65.0, 72.0	0.22°	~25 km
EUR-11	Europe	-44.0, 22.0, 65.0, 72.0	0.11°	~12.5 km
MED-44	Mediterranean	-10.0, 30.0, 40.0, 48.0	0.44°	~50 km
MED-22	Mediterranean	-10.0, 30.0, 40.0, 48.0	0.22°	~25 km
MNA-44	Middle East & North Africa	-25.0, 0.0, 75.0, 50.0	0.44°	~50 km
MNA-22	Middle East & North Africa	-25.0, 0.0, 75.0, 50.0	0.22°	~25 km
NAM-44	North America	-172.0, 12.0, -35.0, 76.0	0.44°	~50 km
NAM-22	North America	-172.0, 12.0, -35.0, 76.0	0.22°	~25 km
SAM-44	South America	-93.0, -56.0, -25.0, 18.0	0.44°	~50 km
SAM-22	South America	-93.0, -56.0, -25.0, 18.0	0.22°	~25 km
WAS-44	South Asia (West Asia)	20.0, -15.0, 115.0, 45.0	0.44°	~50 km
WAS-22	South Asia (West Asia)	20.0, -15.0, 115.0, 45.0	0.22°	~25 km
SEA-44	Southeast Asia	89.0, -15.0, 146.0, 27.0	0.44°	~50 km
SEA-22	Southeast Asia	89.0, -15.0, 146.0, 27.0	0.22°	~25 km

Domain Resolution Notes

0.44° (~50 km): Standard resolution, available for all domains
0.22° (~25 km): High resolution, available for most domains
0.11° (~12.5 km): Very high resolution, currently only available for Europe (EUR-11)

Query Parameters

The CDS connector supports additional query parameters to customize CORDEX data retrieval. These parameters can be passed via the query_params dictionary in the get_data() method.

Default Values

When downloading CORDEX data, the following default parameters are automatically applied:

Parameter	Default Value	Description
`experiment`	`"historical"`	Type of CORDEX experiment
`horizontal_resolution`	`"0_44_degree_x_0_44_degree"`	Spatial resolution (~50 km)
`temporal_resolution`	`"daily_mean"`	Temporal aggregation
`ensemble_member`	`"r1i1p1"`	Ensemble member identifier
`format`	`"netcdf"`	Output file format
`start_year`	Derived from `date_start`	Start year for data
`end_year`	Derived from `date_end`	End year for data

Valid Parameter Values

`experiment`

Type of CORDEX experiment:

"historical" (default) - Historical period (typically 1950-2005)
"evaluation" - Evaluation runs driven by ERA-Interim reanalysis (typically 1980-2010)
"rcp_2_6" - RCP 2.6 scenario (low emissions)
"rcp_4_5" - RCP 4.5 scenario (medium emissions)
"rcp_8_5" - RCP 8.5 scenario (high emissions)

`horizontal_resolution`

Spatial resolution of the model grid:

"0_44_degree_x_0_44_degree" (default) - ~50 km grid spacing
"0_22_degree_x_0_22_degree" - ~25 km grid spacing (high resolution)
"0_11_degree_x_0_11_degree" - ~12.5 km grid spacing (very high resolution, EUR-11 only)

`temporal_resolution`

Temporal aggregation of the data:

"daily_mean" (default) - Daily mean values
"3_hourly" - 3-hour intervals
"6_hourly" - 6-hour intervals
"monthly_mean" - Monthly mean values
"fixed" - Time-invariant fields (e.g., orography, land mask)

`ensemble_member`

Ensemble member identifier (format: rXiYpZ):

"r1i1p1" (default) - First realization, first initialization, first physics
"r0i0p0" - Used for evaluation runs
"r12i1p1", "r2i1p1", "r3i1p1" - Additional ensemble members
r = realization (different initial conditions)
i = initialization method
p = physics parameterization

`gcm_model`

Global Climate Model providing boundary conditions (examples):

"mohc_hadgem2_es" - Met Office Hadley Centre HadGEM2-ES
"mpi_m_mpi_esm_lr" - Max Planck Institute MPI-ESM-LR
"ichec_ec_earth" - EC-EARTH consortium model
"cnrm_cerfacs_cm5" - CNRM-CERFACS CNRM-CM5
"ncc_noresm1_m" - Norwegian Climate Centre NorESM1-M
"era_interim" - ERA-Interim reanalysis (for evaluation runs)
Many others available

`rcm_model`

Regional Climate Model (examples):

"clmcom_clm_cclm4_8_17" - CLM-Community CCLM4-8-17
"smhi_rca4" - SMHI RCA4
"knmi_racmo22t" - KNMI RACMO22T
"gerics_remo2009" - GERICS REMO2009
"dmi_hirham5" - DMI HIRHAM5
Many others available

Usage Examples

Basic Usage - Download with Defaults

from terrakit import DataConnector

# Initialize CDS connector
dc = DataConnector(connector_type="climate_data_store")

# Download CORDEX data with default parameters
# (historical experiment, 0.44° resolution, daily_mean, r1i1p1 ensemble)
data = dc.connector.get_data(
    data_collection_name="projections-cordex-domains-single-levels",
    date_start="2000-01-01",
    date_end="2000-12-31",
    bbox=[-10, 35, 30, 60],  # Europe region (automatically mapped to EUR-44)
    bands=["2m_air_temperature", "mean_precipitation_flux"]
)

print(f"Data shape: {data.shape}")
print(f"Dimensions: {data.dims}")

Override Default Parameters with query_params

from terrakit import DataConnector

dc = DataConnector(connector_type="climate_data_store")

# Example 1: Get RCP 8.5 scenario data (future projections)
data_rcp85 = dc.connector.get_data(
    data_collection_name="projections-cordex-domains-single-levels",
    date_start="2050-01-01",
    date_end="2050-12-31",
    bbox=[-10, 35, 30, 60],
    bands=["2m_air_temperature"],
    query_params={"experiment": "rcp_8_5"}
)

# Example 2: Get high-resolution data (0.22°)
data_hires = dc.connector.get_data(
    data_collection_name="projections-cordex-domains-single-levels",
    date_start="2000-01-01",
    date_end="2000-12-31",
    bbox=[-10, 35, 30, 60],
    bands=["2m_air_temperature"],
    query_params={"horizontal_resolution": "0_22_degree_x_0_22_degree"}
)

# Example 3: Get 3-hourly data instead of daily mean
data_3hourly = dc.connector.get_data(
    data_collection_name="projections-cordex-domains-single-levels",
    date_start="2000-01-01",
    date_end="2000-01-07",
    bbox=[-10, 35, 30, 60],
    bands=["10m_wind_speed"],
    query_params={"temporal_resolution": "3_hourly"}
)

# Example 4: Specify GCM and RCM models
data_custom_models = dc.connector.get_data(
    data_collection_name="projections-cordex-domains-single-levels",
    date_start="2000-01-01",
    date_end="2000-12-31",
    bbox=[-10, 35, 30, 60],
    bands=["2m_air_temperature", "mean_precipitation_flux"],
    query_params={
        "gcm_model": "mohc_hadgem2_es",
        "rcm_model": "knmi_racmo22t"
    }
)

# Example 5: Multiple parameter overrides for future scenario
data_future = dc.connector.get_data(
    data_collection_name="projections-cordex-domains-single-levels",
    date_start="2080-01-01",
    date_end="2080-12-31",
    bbox=[-10, 35, 30, 60],
    bands=["2m_air_temperature", "mean_precipitation_flux"],
    query_params={
        "experiment": "rcp_4_5",
        "horizontal_resolution": "0_22_degree_x_0_22_degree",
        "temporal_resolution": "monthly_mean",
        "ensemble_member": "r12i1p1",
        "gcm_model": "ichec_ec_earth",
        "rcm_model": "smhi_rca4"
    }
)

Working with Different CORDEX Domains

from terrakit import DataConnector

dc = DataConnector(connector_type="climate_data_store")

# List all available CORDEX domains
domains = dc.connector.list_cordex_domains()
print(f"Available domains: {list(domains.keys())}")

# Get information for a specific domain
eur_info = dc.connector.get_cordex_domain_info("EUR-44")
print(f"Europe domain: {eur_info}")

# Download data for different regions
# North America
data_nam = dc.connector.get_data(
    data_collection_name="projections-cordex-domains-single-levels",
    date_start="2000-01-01",
    date_end="2000-12-31",
    bbox=[-120, 25, -70, 50],  # Automatically mapped to NAM-44
    bands=["2m_air_temperature"]
)

# Africa
data_afr = dc.connector.get_data(
    data_collection_name="projections-cordex-domains-single-levels",
    date_start="2000-01-01",
    date_end="2000-12-31",
    bbox=[10, -10, 40, 20],  # Automatically mapped to AFR-44
    bands=["2m_air_temperature"]
)

Comparing Historical and Future Scenarios

from terrakit import DataConnector

dc = DataConnector(connector_type="climate_data_store")

# Get historical baseline (1980-2010)
data_historical = dc.connector.get_data(
    data_collection_name="projections-cordex-domains-single-levels",
    date_start="1990-01-01",
    date_end="1990-12-31",
    bbox=[-10, 35, 30, 60],
    bands=["2m_air_temperature"],
    query_params={"experiment": "historical"}
)

# Get future projection under RCP 8.5 (2070-2100)
data_future_rcp85 = dc.connector.get_data(
    data_collection_name="projections-cordex-domains-single-levels",
    date_start="2090-01-01",
    date_end="2090-12-31",
    bbox=[-10, 35, 30, 60],
    bands=["2m_air_temperature"],
    query_params={"experiment": "rcp_8_5"}
)

# Compare the two periods
print(f"Historical mean: {data_historical.mean().values}")
print(f"Future RCP 8.5 mean: {data_future_rcp85.mean().values}")
print(f"Temperature change: {data_future_rcp85.mean().values - data_historical.mean().values} K")

Notes

Domain Selection: TerraKit automatically maps your bounding box to the appropriate CORDEX domain
Model Availability: Not all GCM-RCM combinations are available for all domains, experiments, and time periods
Resolution: Higher resolutions (0.22°, 0.11°) may not be available for all domains
Time Periods:
Historical: typically 1950-2005
Evaluation: typically 1980-2010
Scenarios (RCP): typically 2006-2100
Ensemble Members: Different ensemble members represent uncertainty in initial conditions and model physics