1.  Dataset information

Dataset name: CHM_Wind

Summary: CHM_Wind, an innovative and comprehensive long-term daily Mean Wind speed (Wind) dataset with a spatial resolution of 0.1° and data collected from 1961 to 2025 in mainland China. 

Content of the data set:

(1)  Metadata for CHM_Wind.docx: This document provides detailed information about the dataset.

(2) CHM_Wind_XXXX.zip: Compressed file containing gridded daily mean wind speed (Wind) fields at 0.1° spatial resolution for the period 1961–2025. To facilitate downloading, the dataset has been divided into multiple NetCDF files, each covering a subset of the 1961–2025 period.

2.  Brief calculation introduction:

The dataset utilizes high-density meteorological station data and follows a structured framework based on the China Hydro-Meteorology dataset (CHM; https://zenodo.org/communities/chm/records?q=&l=list&p=1&s=10). To ensure data integrity, rigorous quality control was applied and missing values were removed to facilitate reliable interpolation.

Daily wind speed was interpolated to a 0.1° spatial resolution, consistent with the spatial resolution of CHM_Drought and CHM_PRE. Unlike the temperature interpolation based on angular distance weighting, wind speed was interpolated using Empirical Bayesian Kriging (EBK) Regression Prediction (https://pro.arcgis.com/en/pro-app/3.4/help/analysis/geostatistical-analyst/what-is-ebk-regression-prediction-.htm). 

3.  Data advantage characteristics:

EBK Regression Prediction combines Empirical Bayesian Kriging with regression analysis and incorporates explanatory raster variables that are known to influence the target variable, thereby improving prediction accuracy relative to either regression or kriging alone. In this study, elevation was used as an explanatory raster to account for the strong topographic control on wind speed. By integrating elevation information and local regression-kriging models, this method provides a robust framework for generating spatially continuous wind-speed fields over complex terrain. 

4.  Details of the variables in the file

Each NetCDF file contains the following four variables:

(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) Tmean: Daily mean temperature (time, lat, lon).

5.  Authors and contacts

Qi Zhang (qizhang@qhnu.edu.cn)

Chiyuan Miao (miaocy@bnu.edu.cn)

Jinlong Hu (hujl98@mail.bnu.edu.cn)