变化“水深”环境下鄱阳湖湿地植被优势群落时空格局遥感研究

The first objective of this project is to develop a robust remote sensing classification method to discriminate the major vegetation communities in Poyang Lake wetland, with the special features in spectrum, phenology or living conditions between different vegetation specifies. Then, the spatial-temporal distributions and cross-community transitions can be documented using multi-sourced long term remotely sensed imagery. The second task is to develop a model to estimate the “geometric water depth” for each location of the lake, where remote sensed inundation area, gauged water level and bathymetry will be used. Additionally, an algorithm will also be developed to derive the diffuse attenuation coefficients and euphotic depth of the lake water, which will be based on the full knowledge of the concentrations of the optical-sensitive water components and the associated inherent optical properties. As such, the long term data of the “optical water depth” for Poyang Lake can be obtained. Then, the influence of optical depth on the distribution of the wetland dominate specifies will be examined, where the results are based on similar “geometric water depth”. Likewise, with the data from the “same” “optical water depth”, the response of vegetation to different “geometric water depth” will also be studied. Finally, the ultimate objective is to understand the driven mechanisms of the wetland transitions (both temporally and spatially) in Poyang Lake, with the combined information of both “geometric” and “optical” water depth in Poyang Lake, where the factors of climate changes and other human activities will also be considered. This study is expected to provide important methodologies and scientific data for the environmental and ecological conservations of the Poyang wetland, which also has implications for other similar wetlands.

利用不同植被类型在光谱、物候及生境特征上的差异，构建鄱阳湖湿地植被优势群落高精度遥感分类算法，在多源长时序遥感数据的支持下，获取湖泊湿地植被不同群落演替的时空动态信息。建立水面范围-水位-湖底地形相结合的湖泊水体几何深度空间化计算模型，获取鄱阳湖长时序动态的实际水深淹没特征。基于水体不同光敏要素浓度与固有光学特性，发展湖泊水体漫衰减系数及真光层深度的遥感定量反演算法，获取鄱阳湖水体光学深度的时空动态特征。以长时序数据为基础，通过约束条件下的单因子分析方法，分别探讨不同几何或光学水深条件下，湖泊湿地优势植被群落的空间差异。综合考虑气象、人为扰动因素的影响作用，耦合几何水深与光学水深的时空变化特征，解析变化“水深”环境对湖泊湿地植被群落演变的潜在影响机制。本研究拟为湖泊湿地生态环境保护提供方法基础与科学数据，具有显著的科学意义与应用价值。

本项目利用了不同植被类型在光谱、物候及生境特征上的差异，构建了鄱阳湖湿地植被优势群落高精度遥感分类算法，在多源长时序遥感数据的支持下，获取鄱阳湖及长江中下游湖泊湿地植被不同群落演替的时空动态信息。基于水体不同光敏要素浓度与固有光学特性，发展湖泊典型水质参数（悬浮泥沙、透明度、叶绿素等）的遥感定量反演算法，获取鄱阳湖及长江中下游湖泊群的水质参数信息。以长时序数据为基础，通过约束条件下的单因子分析方法，分别探讨不同几何或光学水深条件下，湖泊湿地优势植被群落的空间差异。综合考虑气象、人为扰动因素的影响作用，解析了变化“水深”环境对湖泊湿地植被群落演变及水质参数变化的潜在影响机制。