首页 | 本学科首页   官方微博 | 高级检索  
     

联合星载光学和SAR影像的漳江口红树林与互花米草遥感监测
引用本文:董迪,曾纪胜,魏征,严金辉. 联合星载光学和SAR影像的漳江口红树林与互花米草遥感监测[J]. 热带海洋学报, 2020, 39(2): 107-117. DOI: 10.11978/2019063
作者姓名:董迪  曾纪胜  魏征  严金辉
作者单位:自然资源部南海局南海规划与环境研究院, 广东 广州 510300
基金项目:地理国情监测国家测绘地理信息局重点实验室开放基金(2017NGCM08);广东省自然科学基金-博士启动(2018A030310032);广东省海洋遥感重点实验室开放基金(2017B030301005-LORS1806)
摘    要:文章以福建省漳江口国家级红树林自然保护区为研究区, 提出了一种联合星载光学和合成孔径雷达(SAR)影像的红树林与互花米草分布的自动提取算法, 提取研究区内红树林和互花米草的空间分布。本研究选择2016年、2017年和2018年各一景低潮时的Sentinel-2 A光学影像数据, 获取植被和其他地物的光谱和纹理信息。算法首先基于归一化植被指数、增强型植被指数、地表水分指数以及数字高程模型的相关规则计算红树林和互花米草的潜在分布区; 通过随机森林分类算法区分红树林和互花米草, 2016年、2017年和2018年影像的分类总体精度和Kappa系数分别为98.53%和0.980、96.52%和0.952、98.71%和0.978, 分类效果良好; 使用当年所有Sentinel-1 A/B的SAR影像获得研究区的常年海水分布范围, 使用与海水交界的判据, 实现红树林、互花米草提取范围的优化。研究表明, 研究区2016年、2017年和2018年红树林总面积分别为56.85hm2、59.88hm2和58.61hm2, 互花米草总面积分别为109.23hm2、124.00hm2和142.39hm2, 与前人的研究成果在红树林和互花米草的空间分布和面积量级上有较好的一致性。

关 键 词:红树林  互花米草  遥感  分层分类  随机森林分类  漳江口  
收稿时间:2019-07-09
修稿时间:2019-10-12

Integrating spaceborne optical and SAR imagery for monitoring mangroves and Spartina alterniflora in Zhangjiang Estuary
Di DONG,Jisheng ZENG,Zheng WEI,Jinhui YAN. Integrating spaceborne optical and SAR imagery for monitoring mangroves and Spartina alterniflora in Zhangjiang Estuary[J]. Journal of Tropical Oceanography, 2020, 39(2): 107-117. DOI: 10.11978/2019063
Authors:Di DONG  Jisheng ZENG  Zheng WEI  Jinhui YAN
Affiliation:South China Sea Institute of Planning and Environmental Research, South China Sea Bureau of Ministry of Natural Resources, Guangzhou 510300, China
Abstract:Mangroves are an important type of coastal wetlands with ecological, environmental, economic, and cultural values. Spartina alterniflora is an invasive alien plant, threatening mangroves in China. The competition between Spartina alterniflora and mangroves is an important ecological issue along the southeast coast of China. Accurate monitoring of Spartina alterniflora and mangroves with remote sensing is of great significance for scientific protection of mangrove ecosystems. We propose a new method to monitor Spartina alterniflora and mangroves, integrating Sentinel-1 SAR and Sentinel-2 optical imagery. The Yunxiao National Nature Reserve of Mangroves, located in Zhangjiang Estuary, Fujian, China, is chosen as the study area. We select one Sentinel-2A image at low tide in 2016, 2017 and 2018 to obtain the spectral and texture information of vegetation and other objects. The new method comprises of three steps: 1) use rules related to NDVI, EVI, LSWI, and DEM to get the potential masks of Spartina alterniflora and mangroves; 2) use random forest classification method to distinguish Spartina alterniflora and mangroves further; and 3) use all Sentinel-1 A/B images in that year to get the estuarine yearlong seawater body, and use the criterion interaction with sea water to refine the detected Spartina alterniflora and mangroves. The random forest classifier is found suitable for mapping wetlands with overall accuracy of 98.53%, 96.52% and 98.71%, and Kappa coefficients of 0.980, 0.952 and 0.978 in 2016, 2017 and 2018, respectively. The total areas of the detected Spartina alterniflora and mangroves in the study region are 109.23 and 56.85 hm 2in 2016, 124.00 and 59.88 hm2 in 2017, and 142.39 and 58.61 hm2 in 2018, respectively, consistent with previous research results in terms of spatial distribution and magnitude of the area.
Keywords:mangroves  Spartina alterniflora  remote sensing  hierarchical classification  random forest classification  Zhangjiang Estuary  
本文献已被 CNKI 等数据库收录!
点击此处可从《热带海洋学报》浏览原始摘要信息
点击此处可从《热带海洋学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号