The study introduces HiMIC-Monthly, a high-resolution (1 km × 1 km) monthly atmospheric moisture index dataset for China from 2003 to 2020. This dataset is the first of its kind to provide multiple moisture indices (six in total) at such a fine spatial resolution, addressing the lack of detailed moisture data for fine-scale studies. The dataset is generated using the Light Gradient Boosting Machine (LightGBM) algorithm, which predicts moisture indices based on observations from 2,419 weather stations and multiple covariates, including land surface temperature, vapor pressure, land cover, impervious surface proportion, population density, and topography. The performance of the dataset is evaluated using three metrics: R², RMSE, and MAE, all of which show high accuracy, with R² values exceeding 0.96 and RMSE and MAE values within reasonable ranges. The dataset exhibits high consistency with in situ observations across various spatial and temporal scales, demonstrating broad applicability and reliability. The study also compares the HiMIC-Monthly dataset with an existing coarse-resolution dataset (CMFD), showing superior performance in capturing spatial variations, particularly in urban areas. The HiMIC-Monthly dataset has significant potential applications in urban climate, environmental science, ecosystems, and public health, providing valuable insights into human heat stress, disease spread, urban dry/wet island effects, plant growth, and crop yield assessment.The study introduces HiMIC-Monthly, a high-resolution (1 km × 1 km) monthly atmospheric moisture index dataset for China from 2003 to 2020. This dataset is the first of its kind to provide multiple moisture indices (six in total) at such a fine spatial resolution, addressing the lack of detailed moisture data for fine-scale studies. The dataset is generated using the Light Gradient Boosting Machine (LightGBM) algorithm, which predicts moisture indices based on observations from 2,419 weather stations and multiple covariates, including land surface temperature, vapor pressure, land cover, impervious surface proportion, population density, and topography. The performance of the dataset is evaluated using three metrics: R², RMSE, and MAE, all of which show high accuracy, with R² values exceeding 0.96 and RMSE and MAE values within reasonable ranges. The dataset exhibits high consistency with in situ observations across various spatial and temporal scales, demonstrating broad applicability and reliability. The study also compares the HiMIC-Monthly dataset with an existing coarse-resolution dataset (CMFD), showing superior performance in capturing spatial variations, particularly in urban areas. The HiMIC-Monthly dataset has significant potential applications in urban climate, environmental science, ecosystems, and public health, providing valuable insights into human heat stress, disease spread, urban dry/wet island effects, plant growth, and crop yield assessment.