ERA5-Land¶

ERA5-Land (global land reanalysis at about 9 km, forced by ERA5 meteorological fields)

Available

At a glance¶

Note on period: The consolidated dataset covers January 1950 to 2–3 months before the present; ERA5-Land-T (about 5-day latency) fills the remaining period and is replaced by final data about two months later. On 2026-10-12 the CDS hourly data extended to 2026-10-05 and the monthly means to 2026-08. Forecast (accumulated) parameters are missing for 1 January 1950 00 UTC.

Overview¶

ERA5-Land is a replay of the land component of the ERA5 climate reanalysis, produced by ECMWF for C3S at about 9 km resolution with the H-TESSEL land surface model. It runs without data assimilation of its own, forced by ERA5 atmospheric fields (with lapse-rate correction of temperature, humidity and pressure), and provides hourly land variables — soil moisture, soil temperature, snow, runoff, evaporation — at higher resolution than ERA5. It is widely used for flood and drought, hydrological and agricultural studies, and is downloaded from the CDS on a 0.1° grid.

Available variables¶

Main variables from the parameter tables (Tables 1–3) of the official ERA5-Land documentation. Names are GRIB short names; when downloaded from the CDS as NetCDF, some are stored under different names (e.g. 2t→t2m, 2d→d2m). There are fewer atmospheric variables than in ERA5 and no pressure-level data. Forcing variables such as sp, u10, v10, tp, sf, ssrd and strd are ERA5 values interpolated to the ERA5-Land grid.

instantaneous mean accumulated maximum minimum static — instantaneous: value at the time stamp · mean/accumulated/maximum/minimum: over the indicated interval · static: does not change with time

Surface state (instantaneous)¶

Soil (4 layers, instantaneous)¶

Accumulations (from 00 UTC of the day to the validity time)¶

Invariant fields¶

The full variable list is in the official documentation.

How to get the data¶

• Copernicus Climate Data Store (CDS) — hourly data — Web download / CDS API (cdsapi)
  • Requires an ECMWF account, and you must accept the dataset licence before API requests succeed. Default format is GRIB; NetCDF can be selected.
• CDS — monthly averaged data — Web download / CDS API
  • Two products — monthly averaged reanalysis (per-day units) and monthly averaged reanalysis by hour of day.
• CDS — post-processed daily statistics — Web download / CDS API
  • Daily mean, maximum and minimum are computed from the hourly data on request; you can specify the time zone (UTC offset).
• CDS — time-series (ARCO) — Web download / CDS API
  • For quickly retrieving long series at a single point. Returns the grid point nearest the requested location.

Example code¶

Put the url (https://cds.climate.copernicus.eu/api) and your API key in ~/.cdsapirc before running. The daily precipitation total is the 00:00 value of the following day.

# pip install --upgrade cdsapi
import cdsapi

client = cdsapi.Client()
client.retrieve(
    "reanalysis-era5-land",
    {
        "variable": ["2m_temperature", "total_precipitation"],
        "year": "2024",
        "month": "07",
        "day": ["01", "02"],
        "time": ["00:00", "06:00", "12:00", "18:00"],
        "area": [39, 124, 33, 132],  # North, West, South, East (around the Korean Peninsula)
        "data_format": "netcdf",
        "download_format": "unarchived",
    },
    "era5land_korea_20240701.nc",
)
# Precipitation total for 2024-07-01 = tp at 2024-07-02 00:00 (units m; ×1000 for mm)

Practical tips¶

• The accumulation convention differs from ERA5. Reusing ERA5 code that sums hourly tp inflates daily totals several times over; for ERA5-Land, use the next day's 00 UTC value or differences between consecutive steps.
• The incorrect accumulated variables in the monthly averaged reanalysis for September 2022 – February 2024 were never replaced. For that period, use the 00:00 value of the 'by hour of day' product or compute monthly totals yourself from the hourly data.
• 1950–1978 was forced by the preliminary ERA5 back-extension (ERA5.P) and not rerun, so parts of it disagree with the final ERA5 (warm/dry anomalies reported in central Africa and South America in some years). Be careful with trend analyses that include the pre-1979 period.
• Forcing variables such as precipitation, radiation and wind are interpolated from ERA5 — higher resolution, but no new information. They are nearly identical to ERA5; the value of ERA5-Land lies in land-surface variables such as soil moisture, snow and runoff.
• Snow values over the East Antarctic ice sheet are unrealistically low, and glacier points carry a fixed 10 m snow depth. Do not use ERA5-Land snow for polar or glacier studies.
• No uncertainty (ensemble) information is provided. The documentation suggests using the ERA5 EDA uncertainty as an approximation if needed.

Notices and change history¶

References for your paper¶

Dataset citation¶

Hourly data (CDS) — DOI 10.24381/cds.e2161bac

@misc{era5land_hourly,
  author    = {Mu{\~n}oz Sabater, J.},
  title     = {{ERA5-Land} hourly data from 1950 to present},
  year      = {2019},
  publisher = {Copernicus Climate Change Service (C3S) Climate Data Store (CDS)},
  doi       = {10.24381/cds.e2161bac},
  note      = {Accessed on DD-MMM-YYYY}
}

Monthly averaged data (CDS) — DOI 10.24381/cds.68d2bb30

@misc{era5land_monthly,
  author    = {Mu{\~n}oz Sabater, J.},
  title     = {{ERA5-Land} monthly averaged data from 1950 to present},
  year      = {2019},
  publisher = {Copernicus Climate Change Service (C3S) Climate Data Store (CDS)},
  doi       = {10.24381/cds.68d2bb30},
  note      = {Accessed on DD-MMM-YYYY}
}

Post-processed daily statistics (CDS) — DOI 10.24381/cds.e9c9c792

@misc{era5land_daily_stats,
  author    = {{Copernicus Climate Change Service}},
  title     = {{ERA5-Land} post-processed daily statistics from 1950 to present},
  year      = {2024},
  publisher = {ECMWF},
  doi       = {10.24381/cds.e9c9c792},
  note      = {Accessed on DD-MMM-YYYY}
}

Hourly time-series (CDS) — DOI 10.24381/ee82e357

@misc{era5land_timeseries,
  author    = {{Copernicus Climate Change Service}},
  title     = {{ERA5} Land hourly time-series data from 1950 to present},
  year      = {2025},
  publisher = {ECMWF},
  doi       = {10.24381/ee82e357},
  note      = {Accessed on DD-MMM-YYYY}
}

Reference papers¶

ERA5-Land reference paper — DOI 10.5194/essd-13-4349-2021

@article{munozsabater2021era5land,
  author  = {Mu{\~n}oz-Sabater, Joaqu{\'i}n and Dutra, Emanuel and Agust{\'i}-Panareda, Anna and Albergel, Cl{\'e}ment and Arduini, Gabriele and Balsamo, Gianpaolo and Boussetta, Souhail and Choulga, Margarita and Harrigan, Shaun and Hersbach, Hans and Martens, Brecht and Miralles, Diego G. and Piles, Mar{\'i}a and Rodr{\'i}guez-Fern{\'a}ndez, Nemesio J. and Zsoter, Ervin and Buontempo, Carlo and Th{\'e}paut, Jean-No{\"e}l},
  title   = {{ERA5-Land}: a state-of-the-art global reanalysis dataset for land applications},
  journal = {Earth System Science Data},
  volume  = {13},
  number  = {9},
  pages   = {4349--4383},
  year    = {2021},
  doi     = {10.5194/essd-13-4349-2021}
}

Acknowledgements¶

Cite the CDS catalogue entry (DOI and access date) for each dataset you actually used (hourly, monthly, daily statistics, time-series), and give clear and visible attribution to the Copernicus programme and the C3S data. DataCite lists 'Copernicus Climate Change Service' as the creator of the hourly and monthly datasets, so check the citation guidance on each CDS dataset page for the exact recommended wording.

Information on the datasets used in this study was compiled with the help of the Korean Climate Data Navigator (KCDN) (Sim, 2026; https://doi.org/10.5281/zenodo.23300923).

Links¶

• CDS — ERA5-Land hourly data page
• ERA5-Land: data documentation (parameter tables, accumulation conventions, known issues)
• ECMWF Forum — C3S Announcements
• ECMWF — ERA5-Land