中国250米灌溉耕地分布数据集(2000-2020)
收藏资源简介:
灌溉耕地分布是开展生态、水文和气候研究的关键数据,并在水土资源管理中具有特别重要的地位。通过半自动机器学习模型,融合多源遥感数据(包括耕地分布、植被指数、水稻田分布)、灌溉统计和调查数据,以及灌溉适宜性分析,生成了中国逐年、250米灌溉耕地分布图(CIrrMap250)。利用2万个参考样本和高分辨率灌溉取水数据,对灌溉耕地分布数据的精度进行评估。结果显示,CIrrMap250在2000年、2010年和2020年的总体精度为0.79-0.88,优于现有的同类产品。
The distribution of irrigated cropland is critical data for ecological, hydrological and climate research, and holds particular importance in water and soil resource management. A semi-automated machine learning model was employed to integrate multi-source remote sensing data (including cropland distribution, vegetation indices, and paddy field distribution), irrigation statistics and survey data, as well as irrigation suitability analysis, to generate the annual 250-meter resolution irrigated cropland distribution map of China (CIrrMap250). The accuracy of the irrigated cropland distribution data was evaluated using 20,000 reference samples and high-resolution irrigation water intake data. The results demonstrate that the overall accuracy of CIrrMap250 ranges from 0.79 to 0.88 for the years 2000, 2010 and 2020, outperforming existing similar products.




