摘要
阐明青藏高原湿地的时空变化规律和驱动因素,对理解青藏高原湿地对气候变化的响应及保护其生态系统和生物多样性至关重要.由于湿地边界的不确定性、遥感数据光谱和纹理特征的复杂性,湿地的智能识别监测面临许多技术挑战.本研究提出了一种面向遥感场景分类的深度学习框架,实现了基于Landsat遥感图像的青藏高原湿地智能提取.结果表明,近30年(1990~2019年)青藏高原湿地面积增加了31.2%,其中青藏高原中部,即羌塘高原、柴达木盆地和三江源地区的湿地面积增长最明显.气温升高引起的多年冻土融化是该地区湿地面积增长的主要驱动力.因厄尔尼诺而导致的气温和降水异常进一步加剧了湿地面积的增长.
The Tibetan Plateau(TP),known as the“Third Pole”of the Earth and“Asian Water Tower”,is the magnifier of global climate change and birthplace of many large rivers in Asia.There are unique alpine wetlands on the TP,accounting for 20%of Chinese wetlands area,and the lakes alone constitute half of the national lake area.Wetlands are critical to human survival and development as one of the three major ecosystems[1].
作者
李洋
侯正阳
张立强
宋长青
朴世龙
林金泰
彭书时
方克艳
杨静
渠瀛
王跃宾
李景文
李柔静
姚昕
Yang Li;Zhengyang Hou;Liqiang Zhang;Changqing Song;Shilong Piao;Jintai Lin;Shushi Peng;Keyan Fang;Jing Yang;Ying Qu;Yuebin Wang;Jingwen Li;Roujing Li;Xin Yao(State Key Laboratory of Remote Sensing Science,Faculty of Geographical Science,Beijing Normal University,Beijing 100875,China;College of Urban and Environmental Sciences,Peking University,Beijing 100871,China;Laboratory for Climate and Ocean-Atmosphere Studies,Department of Atmospheric and Oceanic Sciences,School of Physics,Peking University,Beijing 100871,China;College of Geographical Sciences,Fujian Normal University,Fuzhou 350007,China;Guangxi Key Laboratory of Spatial Information and Geomatics,Guilin University of Technology,Guilin 541006,China)
基金
supported by the Second Tibetan Plateau Scientific Expedition and Research Program(2019QZKK0608 and 2019QZKK0604)。
