Phenology-based Spartina alterniflora mapping in coastal wetland of the Yangtze Estuary using time series of GaoFen satellite no. 1 wide field of view imagery | |
Ai, Jinquan1,2; Gao, Wei1,2,3,4; Gao, Zhiqiang5; Shi, Runhe1,2,3,4; Zhang, Chao1,2,3 | |
刊名 | JOURNAL OF APPLIED REMOTE SENSING |
2017-05-24 | |
卷号 | 11 |
关键词 | phenology-based mapping normalized difference vegetation index time series classification support vector machine invasive plant species training sample sizes |
ISSN号 | 1931-3195 |
英文摘要 | Spartina alterniflora is an aggressive invasive plant species that replaces native species, changes the structure and function of the ecosystem across coastal wetlands in China, and is thus a major conservation concern. Mapping the spread of its invasion is a necessary first step for the implementation of effective ecological management strategies. The performance of a phenology-based approach for S. alterniflora mapping is explored in the coastal wetland of the Yangtze Estuary using a time series of GaoFen satellite no. 1 wide field of view camera (GF-1 WFV) imagery. First, a time series of the normalized difference vegetation index (NDVI) was constructed to evaluate the phenology of S. alterniflora. Two phenological stages (the senescence stage from November to mid-December and the green-up stage from late April to May) were determined as important for S. alterniflora detection in the study area based on NDVI temporal profiles, spectral reflectance curves of S. alterniflora and its coexistent species, and field surveys. Three phenology feature sets representing three major phenology-based detection strategies were then compared to map S. alterniflora: (1) the single-date imagery acquired within the optimal phenological window, (2) the multitemporal imagery, including four images from the two important phenological windows, and (3) the monthly NDVI time series imagery. Support vector machines and maximum likelihood classifiers were applied on each phenology feature set at different training sample sizes. For all phenology feature sets, the overall results were produced consistently with high mapping accuracies under sufficient training samples sizes, although significantly improved classification accuracies (10%) were obtained when the monthly NDVI time series imagery was employed. The optimal single-date imagery had the lowest accuracies of all detection strategies. The multitemporal analysis demonstrated little reduction in the overall accuracy compared with the use of monthly NDVI time series imagery. These results show the importance of considering the phenological stage for image selection for mapping S. alterniflora using GF-1 WFV imagery. Furthermore, in light of the better tradeoff between the number of images and classification accuracy when using multitemporal GF-1 WFV imagery, we suggest using multitemporal imagery acquired at appropriate phenological windows for S. alterniflora mapping at regional scales. (C) 2017 Society of Photo-Optical Instrumentation Engineers (SPIE) |
研究领域[WOS] | Environmental Sciences & Ecology ; Remote Sensing ; Imaging Science & Photographic Technology |
关键词[WOS] | SUPPORT VECTOR MACHINES ; SALT-MARSH ; INVASIVE PLANT ; CHINA ; CLASSIFICATION ; VEGETATION ; COVER ; ABANDONMENT ; ECOSYSTEM ; DATES |
收录类别 | SCI |
语种 | 英语 |
WOS记录号 | WOS:000402812300002 |
内容类型 | 期刊论文 |
源URL | [http://ir.yic.ac.cn/handle/133337/22529] |
专题 | 烟台海岸带研究所_海岸带信息集成与综合管理实验室 烟台海岸带研究所_中科院海岸带环境过程与生态修复重点实验室 |
作者单位 | 1.East China Normal Univ, Key Lab Geog Informat Sci, Minist Educ, Shanghai, Peoples R China 2.East China Normal Univ, Sch Geog Sci, Shanghai, Peoples R China 3.East China Normal Univ, Joint Lab Environm Remote Sensing & Data Assimila, Shanghai, Peoples R China 4.East China Normal Univ & Colorado State Univ, Joint Res Inst New Energy & Environm, Shanghai, Peoples R China 5.Chinese Acad Sci, Yantai Inst Coastal Zone Res, Yantai, Peoples R China |
推荐引用方式 GB/T 7714 | Ai, Jinquan,Gao, Wei,Gao, Zhiqiang,et al. Phenology-based Spartina alterniflora mapping in coastal wetland of the Yangtze Estuary using time series of GaoFen satellite no. 1 wide field of view imagery[J]. JOURNAL OF APPLIED REMOTE SENSING,2017,11. |
APA | Ai, Jinquan,Gao, Wei,Gao, Zhiqiang,Shi, Runhe,&Zhang, Chao.(2017).Phenology-based Spartina alterniflora mapping in coastal wetland of the Yangtze Estuary using time series of GaoFen satellite no. 1 wide field of view imagery.JOURNAL OF APPLIED REMOTE SENSING,11. |
MLA | Ai, Jinquan,et al."Phenology-based Spartina alterniflora mapping in coastal wetland of the Yangtze Estuary using time series of GaoFen satellite no. 1 wide field of view imagery".JOURNAL OF APPLIED REMOTE SENSING 11(2017). |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论