基于高光谱遥感的雅鲁藏布江源区草地退化及其对全球变化响应敏感性研究

Alpine ecosystem of source region of Yarlung Zangbo plays a vital role in maintaining ecological security of its middle and lower reaches with great population and its delta in the Bay of Bengal. The eco-enviromental quality deteriorated evidently in recent years, traditional multispectral remote sesning is incompetent for extracting information of grassland dagradation accurately. Utilizing hyperspectral remote sensing and novel field imaging spectrometer system, this study will concentrate on the analysis of spectral characteristics of grassland features, plant community and the corresponding spectral pattern of vegetation succession during the process of degradation based on the spectral library with degradation information, develop hyperspectral information model of monitoring grassland degradation. The senstivity of different ecosystems such as alpine steppe, alpine meadow and alpine marsh responding to global change will be demonstrated and analyzed by inspecting and comparing the differences of spectral response and spatial pattern of vegetation-soil features during the process of degradation among different ecosystems with or without human-induced influences. This study will not only provide spatial information and scientific support for the planning of national key regions with important ecofunctions in source region and construct ecological safe shelter zone in Tibet, but also lay the foundation for application of remote sesing especially hyperspectral remote sensing in assessing and monitoring ecological environment of Qinghai-Tibet plateau.

雅江源区高寒脆弱草地生态系统直接关系着中下游西藏经济发达地区和孟加拉湾三角洲地区的生态安全，近年来源区生态环境退化明显，传统多光谱遥感因其信息量有限制约着草地退化信息的准确提取。本项目针对雅江源区生态环境遥感监测研究的缺失，利用高光谱卫星遥感和新型地面成像光谱系统，在构建草地退化光谱知识库的基础上，分析退化过程中典型地物、植物群落组成及其不同演替阶段的光谱特征，揭示源区草地退化光谱响应模式；构建源区草地退化高光谱遥感监测信息模型，获取源区草地退化分布现状；并根据源区放牧特点，通过对比有无人类活动干扰和不同类型的植被-土壤退化光谱响应和空间分布特征的差异，揭示源区草地退化对全球变化响应的敏感性。本项目的实施将为雅江源区国家级生态功能保护区规划、西藏生态安全屏障建设等方面提供空间信息和理论支持，促进遥感技术、特别是高光谱遥感技术在西藏高寒生态环境监测中的应用。

雅鲁藏布江源区高寒脆弱草地生态系统直接关系着中下游西藏经济发达地区和孟加拉湾三角洲地区的生态安全，近年来源区生态环境退化明显，传统多光谱遥感因其信息量有限制约着草地退化信息的准确提取。本项目建立了目前已知规模最大数据最为详实的高寒草地光谱库，分析了源区典型高寒草甸、高寒草原和高寒荒漠的建群种、优势种和草地混合光谱特征，揭示了草地退化过程中光谱响应模式；基于高光谱指数、光谱解混、通用模式分解方法，优化和发展了高寒草地覆盖度等地表参数提取方法；揭示了草地退化过程中植被盖度、生物多样性、优势种变化和草地类型变化特征，基于机器学习选择了高寒草地退化遥感监测的敏感波段，基于光谱分析和高光谱遥感建立了源区草地退化遥感监测模型；基于遥感影像和时间序列植被指数，分析了源区草地退化时空变化特征及其驱动机制，初步评估了不同地形条件下不同类型草地退化对气候变化的敏感性。该项目主要开展了西藏高寒草地遥感监测的基础工作，为今后进一步应用遥感研究高寒草地生态系统奠定了基础，促进了遥感特别是高光谱遥感在西藏高原生态环境监测中的应用。