CHM_PRE_SSP, a new bias-corrected daily CMIP6 precipitation projection dataset over the Chinese mainland

1. Description Dataset name: CHM_PRE_SSP CHM_PRE_SSP is a long-term, bias-corrected precipitation projection dataset developed for the Chinese mainland. It contains historical simulations from 1961 to 2014 and future projections from 2015 to 2100 under three Shared Socioeconomic Pathway scenarios: SSP1-2.6, SSP2-4.5, and SSP5-8.5. The dataset was produced from 28 Coupled Model Intercomparison Project Phase 6 (CMIP6) global climate models (GCMs) and is provided at a spatial resolution of 0.1° and at daily, monthly, and annual temporal resolutions. The CHM_PRE V2 daily gridded precipitation dataset was used as the observational reference during the historical period. CHM_PRE V2 was developed from long-term observations at 3,746 meteorological stations and 11 precipitation-related covariates, using an improved inverse distance weighting method combined with the machine learning algorithm. Singularity Stochastic Removal (SSR) was applied to address the large number of tied zero-precipitation values and improve the representation of precipitation occurrence. Quantile Delta Mapping (QDM) was then used to correct precipitation distributions while preserving the relative changes projected by the GCMs. The correction was performed independently for each calendar month and grid cell. Independent validation showed that CHM_PRE_SSP substantially outperformed the raw CMIP6 simulations in representing annual precipitation amount, wet-day frequency, precipitation extremes, and dry- and wet-spell duration. For example, the mean absolute error (MAE) in annual precipitation was reduced by 89.1%, while the MAEs of wet-day fraction, RR95p, and wet-spell length were reduced by 96.3%, 77.7%, and 85.2%, respectively. CHM_PRE_SSP is expected to support climate, hydrological, and ecological change assessments and to provide high-resolution meteorological forcing for a wide range of environmental and Earth system models 2. Content of the dataset The dataset contains bias-corrected precipitation simulations from the following 28 CMIP6 GCMs: ACCESS-CM2, BCC-CSM2-MR, CAMS-CSM1-0, CanESM5, CESM2, CESM2-WACCM, CMCC-CM2-SR5, CMCC-ESM2, CNRM-CM6-1, CNRM-ESM2-1, EC-Earth3, EC-Earth3-Veg, EC-Earth3-Veg-LR, GFDL-ESM4, HadGEM3-GC31-LL, IITM-ESM, INM-CM4-8, INM-CM5-0, IPSL-CM6A-LR, KACE-1-0-G, KIOST-ESM, MIROC6, MPI-ESM1-2-HR, MPI-ESM1-2-LR, MRI-ESM2-0, NESM3, NorESM2-MM, and UKESM1-0-LL. The released products include: 1. Daily historical simulations: bias-corrected daily precipitation for 1961–2014, provided in NetCDF format.2. Daily future projections: bias-corrected daily precipitation for 2015–2100 under SSP1-2.6, SSP2-4.5, and SSP5-8.5, provided in NetCDF format. Two source-model simulations end in 2099 because of differences in the original CMIP6 archives.3. Monthly products: monthly total precipitation derived from the daily data, provided in NetCDF and GeoTIFF formats.4. Annual products: annual total precipitation derived from the daily data, provided in NetCDF and GeoTIFF formats. The dataset is organized into multiple files. Each file contains data from one GCM for one experiment or future scenario. Filenames identify the variable, temporal resolution, GCM, experiment or scenario, ensemble member, grid label, bias-correction status, study domain, spatial resolution, and covered period. For example, this file `pr_day_BCC-CSM2-MR_ssp126_r1i1p1f1_gn_BC_China_0d1deg_20150101_21001231.nc` contains bias-corrected daily precipitation from BCC-CSM2-MR under SSP1-2.6 for 1 January 2015 to 31 December 2100. The complete CHM_PRE_SSP collection has a total storage volume of approximately 437 GB. Because Zenodo provides a default storage quota of 50 GB per record, the collection is distributed across 13 linked Zenodo records. These records are components of the same CHM_PRE_SSP dataset and should be used together according to the required temporal resolution, file format, GCM, and future scenario. This record is Part 1 of the collection and contains only the annual and monthly precipitation products in NetCDF format. It includes historical simulations for 1961–2014 and future projections under SSP1-2.6, SSP2-4.5, and SSP5-8.5 for all 28 GCMs. The 13 Zenodo records are organised as follows: Annual and monthly precipitation in NetCDF format — all models and experiments (this record). Contains annual and monthly total precipitation for all 28 GCMs, including the historical period and the SSP1-2.6, SSP2-4.5, and SSP5-8.5 projections. DOI: 10.5281/zenodo.21535095 Annual and monthly precipitation in GeoTIFF format — all models and experiments. Contains the same annual and monthly total precipitation products as Part 1, provided in GeoTIFF format for use in GIS and spatial-analysis software. DOI: 10.5281/zenodo.21535105 Daily historical precipitation in NetCDF format — Models A–G. Contains bias-corrected daily precipitation for 1961–2014 from the 14 GCMs in the Models A–G group. DOI: 10.5281/zenodo.21535153 Daily historical precipitation in NetCDF format — Models H–U. Contains bias-corrected daily precipitation for 1961–2014 from the 14 GCMs in the Models H–U group. DOI: 10.5281/zenodo.21535166 Daily SSP1-2.6 precipitation in NetCDF format — Models A–C. Contains bias-corrected daily precipitation projections for 2015–2100 under SSP1-2.6 from the 10 GCMs in the Models A–C group. DOI: 10.5281/zenodo.21535170 Daily SSP1-2.6 precipitation in NetCDF format — Models E–I. Contains bias-corrected daily precipitation projections for 2015–2100 under SSP1-2.6 from the nine GCMs in the Models E–I group. DOI: 10.5281/zenodo.21535173 Daily SSP1-2.6 precipitation in NetCDF format — Models K–U. Contains bias-corrected daily precipitation projections for 2015–2100 under SSP1-2.6 from the nine GCMs in the Models K–U group. DOI: 10.5281/zenodo.21535182 Daily SSP2-4.5 precipitation in NetCDF format — Models A–C. Contains bias-corrected daily precipitation projections for 2015–2100 under SSP2-4.5 from the 10 GCMs in the Models A–C group. DOI: 10.5281/zenodo.21535188 Daily SSP2-4.5 precipitation in NetCDF format — Models E–I. Contains bias-corrected daily precipitation projections for 2015–2100 under SSP2-4.5 from the nine GCMs in the Models E–I group. DOI: 10.5281/zenodo.21535208 Daily SSP2-4.5 precipitation in NetCDF format — Models K–U. Contains bias-corrected daily precipitation projections for 2015–2100 under SSP2-4.5 from the nine GCMs in the Models K–U group. DOI: 10.5281/zenodo.21535224 Daily SSP5-8.5 precipitation in NetCDF format — Models A–C. Contains bias-corrected daily precipitation projections for 2015–2100 under SSP5-8.5 from the 10 GCMs in the Models A–C group. DOI: 10.5281/zenodo.21535228 Daily SSP5-8.5 precipitation in NetCDF format — Models E–I. Contains bias-corrected daily precipitation projections for 2015–2100 under SSP5-8.5 from the nine GCMs in the Models E–I group. DOI: 10.5281/zenodo.21535239 Daily SSP5-8.5 precipitation in NetCDF format — Models K–U. Contains bias-corrected daily precipitation projections for 2015–2100 under SSP5-8.5 from the nine GCMs in the Models K–U group. DOI: 10.5281/zenodo.21535248 3. Variables and file information Each NetCDF file contains the following dimensions and data variable: (1) lat: latitude dimension, measured in degrees (°). (2) lon: longitude dimension, measured in degrees (°). (3) time: time dimension, measured in days since January 1, 1961. (4) pr: bias-corrected precipitation with dimensions `(time, lat, lon)`. For daily products, precipitation is expressed in mm/day. Monthly and annual products contain accumulated precipitation totals for the corresponding month or year and are expressed in mm. Grid cells outside the Chinese-mainland study area are stored as missing values. GeoTIFF products contain the same monthly or annual precipitation totals on the 0.1° geographic grid. 4. Resolution and data range Historical period: 1 January 1961–31 December 2014 Future period: 1 January 2015–31 December 2100; two GCM simulations end in 2099 Spatial resolution: 0.1° × 0.1° Temporal resolutions: daily, monthly, and annual Spatial coverage: 18°N–54°N, 72°E–136°E (as detailed in the table below) North:54°N West:72°E East:136°E South:18°N 5. Example of use The NetCDF files can be opened with software supporting the NetCDF format, including Python, R, MATLAB, CDO, NCO, and Panoply. GeoTIFF files can be opened with standard GIS software such as QGIS and ArcGIS. Example in Python: from pathlib import Path import xarray as xr # Open a single NetCDF file path_nc_daily = Path("pr_day_BCC-CSM2-MR_ssp126_r1i1p1f1_gn_BC_China_0d1deg_20150101_21001231.nc") ds_daily = xr.open_dataset(path_nc_daily) # Open multiple NetCDF files as one dataset. monthly_files = sorted(Path("monthly").glob("*.nc")) ds_monthly = xr.open_mfdataset( monthly_files, chunks="auto", ) 6. References 1. Hu, J., Miao, C., Fan, X., Ji, J., Su, J., & Wu, Y. CHM_PRE_SSP, a new bias-corrected daily CMIP6 precipitation projection dataset over the Chinese mainland. Scientific Data (Submitted).2. Hu, J., Miao, C., Su, J., Zhang, Q., Gou, J., & Sun, Q. (2025). An upgraded high-precision gridded precipitation dataset for the Chinese mainland considering spatial autocorrelation and covariates. Earth System Science Data, 17(8), 3987–4004. https://doi.org/10.5194/essd-17-3987-2025. 7. Authors and contacts Jinlong Hu ([e-mail]) Chiyuan Miao ([e-mail])