| 引用本文: | 蔺亚玲,李相虎,谭志强,宋炎炎,徐铖宇.基于遥感时空融合的鄱阳湖洪泛湿地植物群落动态变化特征.湖泊科学,2023,35(4):1408-1422. DOI:10.18307/2023.0434 |
| Lin Yaling,Li Xianghu,Tan Zhiqiang,Song Yanyan,Xu Chengyu.Dynamic characteristics of vegetation communities in the floodplain wetland of Lake Poyang based on spatio-temporal fusion of remote sensing data. J. Lake Sci.2023,35(4):1408-1422. DOI:10.18307/2023.0434 |
|
| |
|
|
| 本文已被：浏览 1420次 下载 1506次 |
码上扫一扫！ |
|
|
| 基于遥感时空融合的鄱阳湖洪泛湿地植物群落动态变化特征 |
|
蔺亚玲1,2, 李相虎1,2, 谭志强1, 宋炎炎1,2, 徐铖宇1,2
|
|
1.中国科学院南京地理与湖泊研究所, 中国科学院流域地理学重点实验室, 南京 210008;2.中国科学院大学, 北京 100049
|
|
| 摘要: |
| 湿地植物群落结构及其动态变化特征是反映湿地生态系统健康状况的重要指标,对预测湿地生态系统演变方向和生态恢复至关重要。本文基于融合Landsat和MODIS数据获得的30 m、8 d分辨率的NDVI数据集,结合物候特征构建决策树分类方法,解译了2000-2020年鄱阳湖洪泛湿地植被空间分布格局,分析不同植物群落时空动态特征及转移变化过程。结果表明:遥感时空融合为鄱阳湖湿地植被精细化分类提供了良好的数据基础,基于物候特征的决策树分类方法将研究区解译为水体、泥滩、浮叶植物群落、蓼子草(Polygonum hydropiper)-虉草(Phalaris arundinacea)群落、苔草(Carex cinerascens)群落、芦荻(Phragmites australis-Triarrhena lutarioriparia)群落、蒌蒿(Artemisia selengensis)群落和林地8种景观,总体分类精度达89.36%。从全湖范围来看,湿地植被总面积虽有波动,但整体呈显著增加趋势,其中蓼子草-虉草群落的增加最为明显,而分布最广的苔草群落先增加后有微弱减小,次优势种的芦荻群落面积在300~500 km2之间波动;从不同区域的对比来看,苔草群落在北部和东部湖区的变化趋势较明显,而次优势种的分布存在空间差异,北部和东部湖区主要为蓼子草-虉草群落,南部湖区以芦荻群落为主。此外,在2000-2020年间,鄱阳湖湿地景观演变趋势呈现出从水体→泥滩→低滩地植被(苔草群落、蓼子草-虉草群落)到苔草群落与芦荻群落相互转化的过渡性特征,整体表现出高滩地的芦苇、南荻等植被向低滩地扩张,而低滩地的苔草、虉草、蓼子草等向湖心扩张的态势。 |
| 关键词: 鄱阳湖湿地 植物群落 时空动态 转移变化 数据融合 |
| DOI：10.18307/2023.0434 |
| 分类号: |
| 基金项目:国家重点研发计划项目(2018YFE0206400)和国家自然科学基金项目(41871093,42171104)联合资助。 |
|
| Dynamic characteristics of vegetation communities in the floodplain wetland of Lake Poyang based on spatio-temporal fusion of remote sensing data |
|
Lin Yaling1,2, Li Xianghu1,2, Tan Zhiqiang1, Song Yanyan1,2, Xu Chengyu1,2
|
|
1.Key Laboratory of Watershed Geographic Sciences, Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences, Nanjing 210008, P. R. China;2.University of Chinese Academy of Sciences, Beijing 100049, P. R. China
|
| Abstract: |
| The structure and dynamic characteristics of wetland vegetation communities are important indicators reflecting the health of the wetland ecosystem, which are essential for predicting the direction of development and ecological restoration of the wetland ecosystem. Based on the 30-m and 8-day resolution NDVI dataset obtained by fusing Landsat and MODIS data, we adopted a phenology-based decision tree approach to map the spatial distribution patterns of different vegetation communities in the floodplain wetland of Lake Poyang from 2000 to 2020, and analysed the spatiotemporal dynamics and transformation processes of different vegetation communities. The results indicated that the spatio-temporal adaptive fusion model for NDVI products (STAFFN) could provide a powerful database for fine classification of wetland vegetation in Lake Poyang. The study area was classified into eight landscapes using a phenology-based decision tree approach: water, mudflat, floating aquatic macrophyte community, Polygonum hydropiper-Phalaris arundinacea community, Carex community, Phragmites australis-Triarrhena lutarioriparia community, Artemisia selengensis community and forest. The overall classification accuracy was 89.36%. Within the whole lake, the total vegetation cover fluctuated, but showed a significant increasing trend during the observation period. Among the different vegetation communities, the Polygonum hydropiper-Phalaris arundinacea community showed the most significant increase, while the Carex community, which is the most widespread, first increased and then decreased slightly, and the Phragmites australis-Triarrhena lutarioriparia community, which is the subdominant species, fluctuated between 300 and 500 km2. Comparison of the sub-lake regions showed that the trend in the Carex community was relatively clear in the northern and eastern regions. There were spatial differences in the distribution of the subdominant species. The northern and eastern regions were dominated by the Polygonum hydropiper-Phalaris arundinacea community, while the southern region was dominated by the Phragmites australis-Triarrhena lutarioriparia community. In addition, the landscape evolution trend of Poyang Lake wetland during 2000-2020 showed a transitional characteristic from water → mudflat → lower flat vegetation (Carex community and Polygonum hydropiper-Phalaris arundinacea community) to mutual transformation of Carex community and Phragmites australis-Triarrhena lutarioriparia community. The overall situation was that the vegetation (e.g. Phragmites australis, Triarrhena lutarioriparia, etc.) on the higher level expanded towards the lower level and the Carex, Phalaris arundinacea and Polygonum hydropiper, etc. on the lower level expanded towards the centre of the lake. |
| Key words: Lake Poyang wetland vegetation community spatiotemporal dynamics transformation data fusion |
|
|
|
|
