中文摘要：

水生植物是浅水湖泊的重要类群,也是湖泊环境变化的指示物,快速监测水生植物的时空分布对湖泊生态修复和管理具有重要的指导意义.基于多时相环境星影像,构建太湖水生植物分类决策树模型,将太湖水生植物分成挺水、浮叶和沉水植物3种类别,结合沉水植物优势种生活史特征,提出了一种考虑生活史信息的太湖沉水植物优势种遥感监测方法,并对太湖7大沉水植物优势种进行时空监测.通过2013年7、8和9月野外调查样点验证,3期遥感影像分类后总的精度分别为83.04%、81.82%和85.47%,2013年太湖沉水植物优势种总体识别精度为62.20%.提出的水生植物遥感分类及沉水植物优势种识别方法为太湖管理部门开展水草打捞和生态修复提供依据和参考,同时为研究太湖沉水植物的历史变迁奠定基础.

英文摘要：

Aquatic vegetation is not only an important group / type of shallow lakes,but also an indicator of lake environment. It plays an important role in ecological restoration and management of shallow lakes. To monitor real-time spatial-temporal distributions of aquatic vegetation in this study,classification tree（ CT） model for mapping aquatic vegetation types was developed through HJ-CCD images. Aquatic vegetation of Lake Taihu was classified into emergent,floating-leaved and submerged vegetation by CT model. Living histories of dominant species of submerged vegetation were extracted from literatures and surveys. Combining with living histories and multi-temporal HJ-CCD images,a method was proposed for mapping dominant species of submerged vegetation of Lake Taihu. Based on the method,submerged vegetation in Lake Taihu was subdivided into seven dominant species. By verification of the using field survey points,the overall accuracies of CT model for mapping aquatic vegetation were 83. 04%,81. 82%and 85. 47%,respectively during three periods of the investigations on July,August and September of 2013. The overall accuracy of dominant species of submerged vegetation was 62. 20%. Therefore,CT model and the method for mapping dominant species of submerged vegetation proposed in this study could be helpful for lake management including guiding aquatic vegetation harvesting and ecological restoration. Meanwhile,it lays foundation for further research of historical changes in submerged vegetation of Lake Taihu.