首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于GF-1 WFV影像的黄河口湿地植被盖度估测及分析
引用本文:王建步,张杰,马毅,任广波.基于GF-1 WFV影像的黄河口湿地植被盖度估测及分析[J].海洋学报,2018,40(6):40-50.
作者姓名:王建步  张杰  马毅  任广波
作者单位:国家海洋局第一海洋研究所, 山东 青岛 266061
基金项目:海洋公益性行业科研专项经费项目(201205008-01);高分海岸带遥感监测与应用示范项目。
摘    要:植被盖度是植被生长状况的重要定量指标,而目前开展的植被盖度遥感估测工作主要集中在陆地区域,对河口湿地植被盖度遥感估测工作比较少见。本文基于国产GF-1 WFV遥感影像开展了黄河口湿地植被盖度估算,并结合植被类型、土壤盐度和植被指数分布状况开展了植被盖度分布特征分析,得到如下结论:(1)基于GF-1 WFV卫星影像的NDVI、SRI、SAVI、MSAVI和DVI 5种植被指数,分别建立植被盖度估测模型,其中基于NDVI、SRI、MSAVI和DVI 4种植被指数建立的多变量线性回归模型估测精度最好,其决定系数(R2)最大,为0.904,均方根误差RMSE最小,为0.14;(2)植被盖度估算模型的精度与植被盖度本身有一定的关系,其中各植被盖度回归模型中,盖度大于0.8时估算精度要优于盖度小于0.6的区域,RMSE最大相差0.04;(3)以潮滩碱蓬和潮滩芦苇为主的植被覆盖区属于低植被盖度区,盖度位于0.03~0.5,盐度在1.5 g/L左右;芦苇草甸、互花米草和柽柳灌丛植被覆盖区属于高植被盖度区,盖度位于0.8~1.0,其中芦苇草甸土壤盐度小于1.2 g/L,柽柳灌丛土壤盐度在1.4~2.0 g/L,在高盖度植被区混生有中等盖度的植被,盖度在0.5~0.8,土壤盐度在1.8 g/L左右。

关 键 词:GF-1WFV影像    黄河口湿地    植被盖度    植被指数
收稿时间:2017/9/15 0:00:00
修稿时间:2017/10/16 0:00:00

Vegetation fraction coverage estimation and analysis of the Yellow River Estuary wetland based on GF-1 WFV satellite image
Wang Jianbu,Zhang Jie,Ma Yi and Ren Guangbo.Vegetation fraction coverage estimation and analysis of the Yellow River Estuary wetland based on GF-1 WFV satellite image[J].Acta Oceanologica Sinica (in Chinese),2018,40(6):40-50.
Authors:Wang Jianbu  Zhang Jie  Ma Yi and Ren Guangbo
Institution:The First Institute of Oceanography, State Oceanic Administration, Qingdao 266061, China
Abstract:Vegetation fraction coverage (VFC) is an important quantitative indicator of the vegetation growth. At present, remote sensing estimation work of VFC has been mainly implemented in land areas but seldom in estuary wetland. In this paper, we carried out VFC estimation of the Yellow River Estuary wetland based on homemade GF-1 WFV satellite image, and developed the analysis of VFC distribution characteristics based on vegetation type, soil salinity and vegetation index. The main conclusions we drew from this study are:(1) According to GF-1 WFV satellite image, VFC estimation model was built based on five vegetation index, including NDVI, SRI, SAVI, MSAVI and DVI. The largest determination coefficient R2 (0.904) and the smallest root-mean-square error RMSE (0.14) were obtained from the multivariate linear regression model, which was the best model of all, built upon NDVI, SRI, MSAVI and DVI. (2) Estimation precision of VFC estimation model was found depending on the value of VFC. The estimation precision was higher in areas with a VFC value larger than 0.8, compared to areas with a VFC value smaller than 0.6 and the maximum difference for RMSE is 0.04. (3) VFC areas that mainly occupied with suaeda and tidal flat phragmites were considered to be low VFC areas, which had VFC values in the range of 0.03 to 0.5, and salt salinities were around 1.5 g/L. Phragmites meadow, spartina and tamarix chinesis shrub occupied areas belonged to high VFC areas with VFC values varied from 0.8 to 1.0. The salt salinity of phragmites meadow was smaller than 1.2 g/L, and that of tamarix chinesis shrub was in between 1.4 and 2.0 g/L. Medium VFC vegetation with VFC values ranging from 0.5 to 0.8 was found in high VFC areas and possessed salt salinity around 1.8 g/L. (4) Among all studied areas, low and high VFC areas accounted for 25.1% and 20.2%, respectively, while medium VFC areas only took up 8.3% in proportion.
Keywords:GF-1 WFV satellite image  Yellow River Estuary wetland  vegetation fraction coverage  vegetation index
本文献已被 CNKI 等数据库收录!
点击此处可从《海洋学报》浏览原始摘要信息
点击此处可从《海洋学报》下载免费的PDF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