An upgraded high-precision gridded precipitation dataset for the Chinese mainland considering spatial autocorrelation and covariates

Abstract

Precipitation is a critical driver of the water cycle, profoundly influencing water resources, agricultural productivity, and natural disasters. However, existing gridded precipitation datasets exhibit markable deficiencies in capturing the spatial autocorrelation and associated environmental and climatic influences – here referred to collectively as precipitation-related covariates – that limit their accuracy, particularly in regions with sparse meteorological stations. To address these challenges, this study proposes a completely new gridded precipitation generation scheme that integrates long-term daily observations from 3746 gauges with 11 key precipitation-related covariates. Building upon the improved inverse distance weighting interpolation method used in our previous dataset CHM_PRE V1, we integrated a machine learning algorithm – light gradient boosting machine (LGBM) – to incorporate precipitation-related covariates in a data-driven manner. This integration allows for a more comprehensive characterization of precipitation patterns, jointly capturing spatial autocorrelation and covariate-based variability. By this novel scheme, a new high-precision, long-term, daily gridded precipitation dataset for the Chinese mainland (CHM_PRE V2) was developed. Validation against 63 397 high-density gauges demonstrated that CHM_PRE V2 significantly outperforms existing gridded precipitation datasets. Specifically, it achieves a mean absolute error of 1.48 mm d−1 and a Kling-Gupta efficiency of 0.88, representing improvements of 12.84 % and 12.86 %, respectively, compared to the previously optimal dataset. Regarding precipitation event detection, CHM_PRE V2 achieved a Heidke skill score of 0.68 and a false alarm ratio of 0.24, surpassing the previously optimal dataset by 17.24 % and 29.17 %, respectively. These results demonstrate that CHM_PRE V2 markedly enhances precipitation measurement accuracy, reduces overestimation of precipitation events, and provides a reliable foundation for hydrological modeling and climate assessments. This dataset features a resolution of 0.1°, spans from 1960 to 2023, and will be updated annually. Free access to the dataset can be found at https://doi.org/10.11888/Atmos.tpdc.300523 (Hu and Miao, 2025).