CGS01 - A High Accuracy Daily Precipitation Gridded Dataset (2000-2024) across China

The precipitation dataset CGS01 was produced by a dual thin-plate spline fusion method combining GPM satellite observationsand ground gauge observations, aiming to quantify daily precipitation distribution in regions with sparse and unevenly distributedgauges as well as alpine areas. The CGS01 is generated using only 27% of the gauges adopted in previous studies across China. It achieves evidently higher accuracythan existing products, with comparable reliability in alpine areas to that in plain regions. Validated against observed runoff data, the watershed water balance analysis error is only 10.9%. The national average error of annual precipitation is 2%. The correlation coefficient, KGE, POD, and MAE of daily precipitation are 0.80, 0.72, 0.91, and 1.69 mm/d. Furthermore, CGS01 can reasonably reflect the controlling effects of topography and wind direction on precipitation spatial patterns in high mountain and canyon areas without in-situ observations.The method distinguishes between systematic and random errors of GPM satellite precipitation data and corrects them using targeted separate methods. A novel systematic error correction method different from conventional approaches is proposed.1. The last field separated by underscores ("_") in the file name represents the year and month.2. Each TIFF file is a multi-band file, where Band 1 corresponds to daily precipitation on the first day, and subsequent bands follow in sequence.3. Precipitation unit: mm/d.4. Spatial resolution: 0.1 degrees.