首页 | 期刊介绍 | 编委会 | 道德声明 | 投稿指南 | 常用下载 | 过刊浏览 | In English
引用本文:
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  下载PDF阅读器  关闭
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 121次   下载 0  
分享到: 微信 更多
基于WorldView-2和GF-2遥感影像的赵述岛礁坪底质变化研究
万佳馨1, 任广波2, 马毅2
1.中国石油大学(华东);2.自然资源部第一海洋研究所
摘要:
珊瑚礁生态系统是海洋中生物多样性和初级生产力极高的生态系统,在自然和人为因素的双重压力下,其健康状况面临巨大的威胁。本文以西沙群岛海域赵述岛礁坪为研究区域,应用2010年的WorldView-2和2015年的GF-2两期高分辨率卫星遥感影像数据,结合现场水下珊瑚礁照片和视频资料,建立了赵述岛礁坪底质类型遥感分类体系,分别构建了针对WorldView-2遥感影像和GF-2遥感影像的珊瑚礁底质分类决策树模型,并通过对两期遥感影像的底质类型提取结果的变化分析,完成了赵述岛礁坪退化区域的提取。结果表明:1)WorldView-2影像的分类决策树具有91%的总体分类精度和0.89的Kappa系数,GF-2影像的分类决策树具有更高的分类精度;2)6年期间研究区大约1/4的珊瑚礁分布区发生了退化;3)最突出的自然变化是礁前区大面积的珊瑚丛生区转变成藻脊,结合相关资料认为东北季风、热带气旋和温度上升的综合影响是驱动因素;4)赵述岛西侧的藻脊、珊瑚沉积区大面积转化为珊瑚砂和深水区域,原因是航道和码头的开发活动。
关键词:  珊瑚礁  遥感  决策树  退化
DOI:10.11759/hykx20190103001
分类号:
基金项目:国家重点研发计划项目(2017YFC1405600);国家自然科学基金 No.51839002;高分海岸带遥感监测与应用示范
Study on substrate changes of Zhaoshu reef flat based on WorldView-2 and GF-2 remote sensing images
wanjiaxin1, renguangbo2, mayi2
1.China University of Petroleum;2.FIRST INSTITUTE OF OCEANOGRAPHY, MNR
Abstract:
Coral reef ecosystem is an ecosystem with high biodiversity and primary productivity in the ocean. Under the dual pressures of natural and human factors, coral reef ecosystem is facing a huge threat of sickness and degradation. In this paper, Zhaoshu Island coral reef flat in Xisha Island waters was taken as the research area. Based on the high-resolution WorldView-2 and GF-2 satellite remote sensing image data in 2010 and 2015, combined with underwater coral reef photographs and video data, a remote sensing classification system for Zhaoshu Island reef flat substrate types was established. The decision tree model of coral reef substrate classification based on WorldView-2 and GF-2 remote sensing image was established. The change of coral reef substrate type in Zhaoshu Island was analyzed by two-stage change detection, and the degradation region was analyzed. The results show that: (1) the classification decision tree of the WorldView-2 image has 91% overall classification accuracy and 0.89 Kappa coefficient, and the classification decision tree of the GF-2 image has higher classification accuracy; 2) about 1/4 of the coral reef distribution area in the study area has degraded during six years; 3) the most prominent natural change is the large area of the reef-clumping-area to algal ridge in the front of the reef flat. Combining with relevant data, it is believed that the driving factors are the effects of northeast monsoon, tropical cyclone and temperature rise. 4) algal ridge and reef-depositional-area on the west side of Zhaoshu Island have been transformed into coral sand and deep water areas in a large area due to the development of waterways and wharfs.
Key words:  coral reef  remote sensing  decision tree  degradation
版权所有 《海洋科学》编辑部 Copyright©2008 All Rights Reserved
主管单位:中国科学院 主办单位:中国科学院海洋研究所
地址:青岛市南海路七号  邮编:266071  电话:0532-82898755  E-mail:marinesciences@qdio.ac.cn
技术支持:北京勤云科技发展有限公司