<p>High-spatial-resolution and long-term climate data are highly desirable for understanding climate-related natural processes. China covers a large area with a low density of weather stations in some regions, especially mountainous regions. This study describes a high-spatial-resolution (0.5’, ∼1 km) dataset of monthly temperatures (minimum, maximum, and mean TMPs) and precipitation (PRE) for the main land area of China for the period 1901–2017. The dataset was spatially downscaled from raw 30’ climatic research unit (CRU) time series data and validated using data from 745 weather stations across China. Compared to raw CRU data of low spatial resolution, the mean absolute error decreased by 0.56 °C for the TMPs and 10.1 % for PRE, the root-mean-square error decreased by 0.65 °C for the TMPs and 11.6 % for PRE, and the Nash–Sutcliffe efficiency coefficients increased from 0.83 to 0.95 for the TMPs and from 0.63 to 0.76 for PRE. Indirect validations from site-scale observations indicated that the dataset captured the climatology well, as well as the annual and seasonal monotonic trends in each climatic variable considered. We concluded that the new high-spatial-resolution dataset is sufficiently reliable for use in investigation of climate change across China. This dataset will be useful in investigations related to climate change across China. The dataset presented in this article is published in the Network Common Data Form (NetCDF) at <a href="https://doi.org/10.5281/zenodo.3114194" target="_blank">https://doi.org/10.5281/zenodo.3114194</a> for precipitation (Peng, 2019a) and <a href="https://doi.org/10.5281/zenodo.3185722" target="_blank">https://doi.org/10.5281/zenodo.3185722</a> for temperatures (Peng, 2019b). The dataset includes 156 NetCDF files compressed with zip format and one user guidance text file.</p>