2003-2020年中国1km高分辨率大气水分指数集合
收藏资源简介：
近地表大气水分是一个关键的环境和水文气候变量，对自然和人类系统具有重大影响。然而，对于精细研究来说，高分辨率的水份数据严重缺乏。在这里，我们开发了2003-2020年中国第一个 1 km 高空间分辨率的月度水分指数采集数据集 （HiMIC-Monthly）。HiMIC-Monthly 由光梯度增强机算法 （LightGBM） 生成，该算法基于对 2,419 个气象站和多个协变量的观测，包括地表温度、蒸气压、土地覆盖、不透水表面比例、人口密度和地形。该集合包括六个常用的水分指数，可以从不同角度对水分状况进行精细评估。结果表明，HiMIC-Monthly 数据集具有良好的性能，6 个水分指数的 R2 值均超过 0.96，均方根误差和平均绝对误差值在合理范围内。该数据集与各种时空范围内的原位观测表现出高度一致性，具有广泛的适用性和较强的可靠性。
Near-surface atmospheric moisture is a critical environmental and hydrometeorological variable with significant impacts on natural and human systems. However, high-resolution moisture data are severely lacking for fine-scale research. Herein, we developed the first 1 km high-spatial-resolution monthly moisture index dataset (HiMIC-Monthly) for China covering the period from 2003 to 2020. HiMIC-Monthly was generated using the Light Gradient Boosting Machine (LightGBM) algorithm, which is based on observations from 2,419 meteorological stations and multiple covariates, including land surface temperature, vapor pressure, land cover, impervious surface fraction, population density, and topography. This dataset includes six commonly used moisture indices, which enable fine-scale assessment of moisture conditions from multiple perspectives. The results demonstrate that HiMIC-Monthly exhibits excellent performance: the R² values of all six moisture indices exceed 0.96, and the root mean square error (RMSE) and mean absolute error (MAE) values fall within reasonable ranges. This dataset shows high consistency with in-situ observations across various spatiotemporal scales, demonstrating broad applicability and strong reliability.
