Data underlying the publication: The annual dynamic dataset of high-resolution crop water use in China from 1991 to 2019

Here, we produced annual blue and green water use for 15 main crops with a resolution of 1 km in China.Initially, we estimated the daily crop blue and green water use at site scale by incorporating more localized input parameters using dynamic water balance model. Subsequently, we developed a spatial prediction of crop water requirements based on random forest model. The resulting maps showed high correlation with locally observed values at field stations, statistics, and exhibited some strengths compared with existing datasets that covering various scales.

本研究生成了覆盖中国全域、空间分辨率为1千米的15种主要作物年度蓝水（Blue Water）与绿水（Green Water）消耗量数据集。首先，本研究通过动态水平衡模型（Dynamic Water Balance Model）融入更多本地化输入参数，在站点尺度上估算了作物逐日蓝水与绿水消耗量。随后，本研究基于随机森林模型（Random Forest Model）构建了作物需水量的空间预测结果。最终生成的空间图谱与野外台站、统计资料中的本地实测值具有高度相关性，且相较于覆盖多尺度的现有数据集，展现出一定优势。