CRU TS v4.10¶

Climatic Research Unit gridded Time Series (CRU TS) version 4.10

Available

At a glance¶

Note on period: v4.10 covers 1901-01 to 2025-12; v4.09 on CEDA covers 1901-01 to 2024-12. No date has been announced for the next version. As of 2026-10-12 the last month of data is December 2025.

Overview¶

CRU TS provides month-by-month variations in climate from 1901 to the recent past on a high-resolution (0.5° × 0.5°) land grid, interpolated from monthly observations at more than 4,000 weather stations worldwide. Ten variables are provided together — mean, maximum and minimum temperature, diurnal temperature range, precipitation, wet day frequency, vapour pressure, cloud cover, frost day frequency and potential evapotranspiration — and the dataset has long been used in climate impact, ecological, agricultural and hydrological studies and for model evaluation. Grid cells without nearby stations are filled with the climatology, so there are no gaps, but each file also gives the number of stations used for interpolation (stn).

Available variables¶

Variable names and units as in the CEDA dataset records. Each netCDF file holds one variable plus stn, the number of stations used in the interpolation (integer, 0–8; not in pet); the ASCII version has separate stn files. From v4.08 most variables also come with interpolation quality measures MAE and MAEA (*.mae.gz, *.maea.gz). TMP, DTR and PRE are primary variables built from observations only; VAP, WET and CLD are secondary variables that blend in synthetic values derived from other variables where observations are sparse; TMN, TMX, FRS and PET are derived from other variables.

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

Temperature (monthly means)¶

Precipitation and moisture (monthly totals and day counts)¶

Cloud, frost and evapotranspiration¶

The full variable list is in the official documentation.

How to get the data¶

• CEDA Archive (CRU TS archive) — Web download (free CEDA account login)
  • The official archive, and CRU's preferred route. v4.09 has 586 files (about 14 GB) under data.ceda.ac.uk/badc/cru/data/cru_ts/cru_ts_4.09/. As of 2026-10-12 v4.10 is not yet on CEDA.
• CRU website — CRU TS v4.10 — Direct HTTPS download (no login) No login needed
  • A local copy, to be used only when access via CEDA is not possible or the version is not yet on CEDA. Each variable folder has netCDF (*.dat.nc.gz) and ASCII files plus stn, mae and maea files, as decadal blocks and for the full period (1901–2025). Read Release_Notes_CRU_TS_4.10.txt first. Country averages (CRU CY) are available alongside.
• CRU high-resolution gridded datasets (version index) — Web documentation
  • Starting page for earlier version folders (cru_ts_4.09/, cru_ts_4.08/, …) and the Google Earth interface.

Example code¶

Downloads the full-period mean temperature netCDF (gzip-compressed) from the CRU website and computes annual means over the Korean Peninsula. The file name follows the 'cru_ts4.xx.1901.YYYY.var.dat.nc.gz' pattern used up to v4.08; check the actual name in the folder listing.

import gzip, shutil, urllib.request
import xarray as xr

base = "https://crudata.uea.ac.uk/cru/data/hrg/cru_ts_4.10/cruts.2604091129.v4.10/tmp/"
name = "cru_ts4.10.1901.2025.tmp.dat.nc"
urllib.request.urlretrieve(base + name + ".gz", name + ".gz")
with gzip.open(name + ".gz", "rb") as fi, open(name, "wb") as fo:
    shutil.copyfileobj(fi, fo)

ds = xr.open_dataset(name)
kor = ds.sel(lat=slice(33, 39), lon=slice(124, 132))
tmp = kor["tmp"].where(kor["stn"] > 0)   # drop cells filled with climatology (no stations)
print(tmp.groupby("time.year").mean().mean(["lat", "lon"]).sel(year=slice(1991, 2025)))

Practical tips¶

• There are no gaps because cells without stations are filled with the 1961–1990 climatology (zero anomaly). For trend or variability analyses, exclude cells with stn = 0 or report their fraction — especially in the early 20th century and in sparsely observed regions.
• tmx and tmn are monthly averages of daily maxima and minima. Use daily data for extremes such as heatwaves and cold spells.
• Past values change slightly with every version (new stations, quality control, interpolation changes). 'Cohort' station selection was introduced for TMP and PRE in v4.08 and extended to DTR in v4.09, while PRE reverted to the previous method in v4.09. Do not mix versions, and state the version in publications.
• The most recent months rely mainly on CLIMAT reports and have fewer stations. For v4.09, CRU states that some December 2024 observations came from raw (not QC'd) DWD CLIMAT data.
• VAP, WET and CLD include synthetic values derived from temperature and diurnal range where observations are sparse, and FRS and PET are entirely derived. Long-term trends in these variables may partly reflect temperature trends.
• Terms differ between v4.09 and earlier on CEDA and v4.10 on the CRU website (OGL v3.0 → ODbL). If you redistribute derived data, check the attribution and Share-Alike conditions of v4.10.
• Superseded decadal files may not be archived once a new version appears. Keep a copy of the files you used for reproducibility.

Notices and change history¶

References for your paper¶

Dataset citation¶

CRU TS v4.09 (CEDA dataset record — no DOI)

@misc{cruts409,
  author    = {{University of East Anglia Climatic Research Unit} and Harris, I. C. and Jones, P. D. and Osborn, T.},
  title     = {{CRU TS4.09}: Climatic Research Unit ({CRU}) Time-Series ({TS}) version 4.09 of high-resolution gridded data of month-by-month variation in climate ({Jan.} 1901- {Dec.} 2024)},
  year      = {2025},
  publisher = {NERC EDS Centre for Environmental Data Analysis},
  url       = {https://catalogue.ceda.ac.uk/uuid/9cf07e92afaa405da4f40b6733f362d3}
}

CRU TS v4.10 (CRU website — not yet on CEDA; no version DOI yet)

@misc{cruts410,
  author       = {{Climatic Research Unit, University of East Anglia}},
  title        = {{CRU-TS} v4.10: high-resolution gridded monthly climate data (1901--2025)},
  year         = {2026},
  howpublished = {\url{https://crudata.uea.ac.uk/cru/data/hrg/cru_ts_4.10/}},
  note         = {Accessed DD Month YYYY}
}

Reference papers¶

CRU TS v4 reference paper (requested for all v4.xx versions) — DOI 10.1038/s41597-020-0453-3

@article{harris2020cruts,
  author  = {Harris, Ian and Osborn, Timothy J. and Jones, Phil and Lister, David},
  title   = {Version 4 of the {CRU TS} monthly high-resolution gridded multivariate climate dataset},
  journal = {Scientific Data},
  volume  = {7},
  number  = {1},
  pages   = {109},
  year    = {2020},
  doi     = {10.1038/s41597-020-0453-3}
}

Acknowledgements¶

CRU asks users to name the dataset as CRU-TS, ideally with its version (e.g. CRU-TS v4.10), and to cite Harris et al. (2020). The attribution for v4.10 is "Climatic Research Unit, University of East Anglia"; for v4.09 and earlier it is "Climatic Research Unit (University of East Anglia) and Met Office". If you downloaded from CEDA, use the citation given in the CEDA record. The dataset DOI in the paper (10.5285/10d3e3640f004c578403419aac167d82) refers to v4.03.

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¶

• CRU high-resolution gridded datasets (CRU TS overview)
• CRU TS v4.10 page
• CRU TS v4.10 Release Notes
• CEDA — CRU TS v4 collection
• CEDA — CRU TS4.09 dataset record
• Harris et al. (2020), Scientific Data