摘要
遥感大数据对赋能水利新质生产力发展具有显著作用,尤其在提升水资源调查、监测与管理效率方面展现出强大潜力。青藏高原被称为“亚洲水塔”,湖泊河流众多,对气候变化也最为敏感,然而由于地理位置特殊、地形环境复杂等因素,青藏高原水资源分布的精细化认知和监测一直是水利工作面临的一项挑战。针对青藏高原地区缺乏高空间分辨率陆表水体信息问题,提出分级弱监督信息挖掘策略,生成了青藏高原地区2 m分辨率陆表水体空间分布信息产品。经评估,产品总体精度为91.36%,用户精度为85.39%,生产者精度为91.71%,显示出较高的准确性和可靠性。研究成果不仅有助于深化对青藏高原地区水资源分布状况的认识,也展示出遥感大数据在复杂地形区域的应用潜力和价值,为水资源管理、生态保护以及区域可持续发展提供有力的数据支撑和决策依据。
Remote sensing big data plays a significant role in empowering the development of new productive forces in water resources,especially in enhancing the efficiency of water resources investigation,monitoring,and management.The Tibetan Plateau,known as the“Asian Water Tower”,boasts numerous lakes and rivers and is highly sensitive to climate change.However,due to its unique geographical location and complex terrain,finegrained understanding and monitoring of water resources distribution in the Tibetan Plateau have always been a challenge for water management.To address the lack of high spatial resolution surface water information in the Tibetan Plateau region,a hierarchical weak supervision information mining strategy was proposed,generating a spatial distribution information product of surface water at a resolution of 2 meters.Evaluation results show that the overall accuracy of the product is 91.36%,with a user accuracy of 85.39%and a producer accuracy of 91.71%,demonstrating high accuracy and reliability.The research findings not only contribute to a deeper understanding of the distribution of water resources in the Tibetan Plateau region but also showcase the potential and value of remote sensing big data applications in complex terrain areas,providing robust data support and decision-making basis for water resource management,ecological protection,and regional sustainable development.
作者
何国金
郑铠沅
杨瑞清
尹然宇
刘慧婵
陈慧玲
彭雪丽
彭燕
王桂周
龙腾飞
陈彦琳
陶浩翔
洪方舟
He Guojin;Zheng Kaiyuan;Yang Ruiqing;Yin Ranyu;Liu Huichan;Chen Huiling;Peng Xueli;Peng Yan;Wang Guizhou;Long Tengfei;Chen Yanlin;Tao Haoxiang;Hong Fangzhou
出处
《中国水利》
2024年第11期56-66,共11页
China Water Resources
基金
第二次青藏高原综合科学考察研究专题项目“生态资产与生态系统服务评估”(2019QZKK0307)。
关键词
陆表水体
青藏高原
国产高分卫星
弱监督
语义分割
surface water
Tibetan Plateau
domestic high-resolution satellite
weak supervision
semantic segmentation