基于多源遥感数据的湿地植物分类和生物量反演研究进展

Advance in Classification and Biomass Estimation of Plants in Wetlands based on Multi-source Remote Sensing Data

湿地植物是湿地生态系统的重要组成部分,是指示湿地生态系统演化的重要指标。利用遥感技术,可以在区域尺度上对湿地植物进行多时段综合观测,识别湿地植物类型,并对植物生物量进行反演,定量估算湿地生态系统资源,因此,遥感技术是评价湿地生态系统健康的重要手段。但是,利用单一种类的遥感数据源的反演,在反演精度和提取湿地植物信息的丰富度上存在缺陷,而利用多源遥感数据提取湿地植物信息的精度更高,应用前景广阔。重点从遥感数据源、多源遥感融合技术、湿地植物遥感分类方法、多源遥感数据识别湿地植物类型的应用、湿地植物生物量反演模型以及应用等方面,阐述多源遥感技术在湿地植物分类和生物量反演研究中的主要进展,讨论多源遥感技术在湿地研究中的优势和不足,展望其未来发展趋势。

Plants in the wetlands are not only one of important parts of wetland ecosystem, but also a signifi- cant indicator of wetland ecosystem evolution. In recent decades, both quantity and quality of global wetlands have been decreasing at a higher speed caused by global warming and human activities. As an effective tool of earth observation, remote sensing technology makes multi-period and extensive observation in the wetlands possible on regional scale. Classification and biomass estimation of plants in the wetlands by using remote sensing technology could help us to assess wetland resources and evaluate the health status of the wetlands quantitatively. However, many studies have revealed that estimation based on single type of remote sensing da- ta does not perform well in accuracy and can＇ t acquire more fine information about the plants in the wetlands. Compared with estimation by single remote sensing data, information extraction based multi-source remote sensing data has more tremendous application prospects in the observation field of plants in the wetlands. This paper summarized applications of multi-source remote sensing to the classification and biomass estimation of the plants in the wetlands from following several aspects： types of remote sensing data applied to the wet- lands; multi-source remote sensing data and their fusion technology; the methods of classification on the plants in the wetlands based on remote sensing technology; the applications on identifying the types of the plants in the wetlands based on multi-source remote sensing data; the models of estimating biomass of the plants in the wetlands using multi-source remote sensing data. The advantages and disadvantages of multi-source remote sensing data and their future development in this field were also discussed.