Dataset Open Access
A Bias-Corrected HighResMIP Dataset for Impact Assessment Studies
Yakubu, Fuseini;
Böhner, Jürgen;
Bouwer, Laurens M.;
ul Hasson, Shabeh
Data curator(s)
Sadikni, RemonEditor(s)
Kern, Stefan
Abstract: The BC-HiRMIP dataset provides globally consist, comprehensive bias-corrected climate datat of High Resolution Model Intercomparison Project (HighResMIP) experiments at daily temporal and 0.5° spatial resolution, covering the period 1979-2050 for hist-1950 and highres-future scenario. Across four global climate models (MPI-ESM1-2-XR, EC-Earth3P-HR, CNRM-CM6-1-HR, HadGEM3-GC31-HM), BC-HiRMIP includes up to 11 essential meteorological variables (temperature suite, precipitation, humidity, radiation, wind, and pressure), spanning equilibrium climate sensitivities of 2.99-5.62°C. For more information see "BC-HiRMIP.pdf" and "Data Description-BC-HiRMIP.pdf"
The data are available in netCDF file format per model, run (historical or future), and variable.
TableOfContents: daily mean near-surface air temperature (tas); daily minimum near-surface air temperature (tasmin), daily maximum near-surface air temperature (tasmax); daily sum of precipitation (pr); daily sum of precipitation in form of snow (prsn)*; daily mean surface downwelling shortwave radiation (rsds). daily mean surface downwelling longwave radiation (rlds); daily mean near-surface wind speed (sfcWind); daily mean near-surface relative humidity (hurs); daily mean near-surface specific humidity (huss)**; daily mean surface air pressure (ps)**
*) Only CNRM-CM6-1-HR and MPI-ESM1-2-XR
**) Only EC-Earth3P-HR and HadGEM3-GC31-HM
TechnicalInfo: dimension: 720 columns x 360 rows; temporalExtent_startDate_hist-1950: 1979-01-01 00:00:00; temporalExtent_endDate_hist-1950: 2014-12-31 23:59:59; temporalDuration_hist-1950: 36; temporalDurationUnit_hist-1950: a; temporalExtent_startDate_highres-future: 2015-01-01 00:00:00; temporalExtent_endDate_highres-future: 2050-12-31 23:59:59; temporalDuration_highres-future: 36; temporalDurationUnit_highres-future: a; temporalResolution: 1; temporalResolutionUnit: d; spatialResolution: 0.5; spatialResolutionUnit: degree; horizontalResolutionXdirection: 0.5; horizontalResolutionXdirectionUnit: degree; horizontalResolutionYdirection: 0.25; horizontalResolutionYdirectionUnit: degree; verticalResolution: none; verticalResolutionUnit: none
Methods: The ISIMIP3BASD v2.5 bias correction method (see Lange [2019; 2021]) was applied to adjust systematic biases while preserving the climate change signals. This parametric quantile mapping approach:
• Corrects biases across all percentiles of variable distributions
• Preserves trends in these percentiles
• Applies variable-specific treatments (e.g., handling drizzle issues for precipitation)
• Maintains physical relationships between variables (particularly for temperature variables)
using the W5E5 V2.0 (Lange et al., 2021) observational reference dataset at 0.5 degree resolution. The four climate models participating in the HighResMIP experiment that are used here are:
- MPI-ESM1-2-XR
- EC-Earth3P-HR
- CNRM-CM6-1-HR
- HadGEM3-GC31-HM
The historical runs "hist-1950" begin 1979-01-01 and end 2014-12-31. The future projection runs "highres-future" begin 2015-01-01, end 2050-12-31, and follow the ssp5-8.5 scenario pathway. The dataset was remapped from their native grids resolution to match the reference dataset grid using CDO: conservative remapping for flux variables (precipitation and radiation) and bilinear remapping for other continuous variables.
The routines (python) used to create and work with the data sets are available from this web page as well: Bias_CorrectonScript.pr and DeriveHUSS.py
Quality: The bias-correction method used is of high quality and has been used for similar datasets multiple times. The performance of the bias correction in replicating the same distribution as that of the W5E5 v2.0 dataset was evaluated for diverse areas located in five different climate types after Köppen-Geiger with convincing results, illustrating the skill of the bias-correction method especially with respect to the distribution of values across the expected range of variation (see also the document BC-HiRMIP.pdf)
Units: K; K; K; kg m-2 s-1; kg m-2 s-1; W m-2; W m-2; m s-1; percent; kg kg-1; pa
GeoLocation: westBoundCoordinate: -180.0; westBoundCoordinateUnit: degrees East; eastBoundCoordinate: 180.0; eastBoundCoordinateUnit: degrees East; southBoundCoordinate: -90.0; southBoundCoordinateUnit: degrees North; northBoundCoordinate: 90.0; northBoundCoordinateUnit: degrees North
Size: CNRM-CM6-1-HR: 203.9 GByte, MPI-ESM1-2-XR: 204.2 GByte, EC-Earth3P-HR: 249.9 GByte, HadGCM3-GC31-HM: 250.1 GByte
Format: netCDF
DataSources: See the file "Data Description-BC-HiRMIP.pdf"
Contact: fuseini.yakubu (at) uni-hamburg.de; shabeh.hasson (at) uni-hamburg.de
- Publication date:
- February 23, 2026
- DOI:
-
- Keyword(s):
- HighResMip Climate Modeling Bias Correction Air Temperature Near Surface Humidity Precipitation Amount Near Surface Wind Speed Shortwave Radiation Longwave Radiation HAREME GERICS ICDC
- Related identifiers:
- Cites:
10.5281/zenodo.4686991, 10.5194/gmd-12-3055-2019, 10.48364/ISIMIP.342217 - Communities:
- License (for files):
- Creative Commons Attribution 4.0 International