尚伟涛^1,2,3, 高志强^1,2, 姜晓鹏^1,2,3, 田信鹏^1,2, 郭少方⁴
1.中国科学院 海岸带环境过程与生态修复重点实验室, 中国科学院 烟台海岸带研究所, 山东 烟台 264003;2.山东省海岸带环境过程重点实验室, 中国科学院 烟台海岸带研究所, 山东 烟台 264003;3.中国科学院大学, 北京 100049;4.烟台市科技情报研究所, 山东 烟台 264003

摘要:
浒苔在近岸搁浅后会破坏海岸景观，干扰水上运动，给滨海旅游业造成严重影响。本文使用无人机搭载的多光谱和可见光传感器对山东半岛的海阳、乳山和文登的三个海滩搁浅的浒苔进行航拍监测，并结合地物光谱测量数据，分别选择归一化植被指数（NDVI）、差值植被指数（DVI）和虚拟基线高度浮藻指数（VB-FAH）对海滩搁浅浒苔与岸边植被及非植被（海水、沙滩）进行识别评估，并分别估算了三个研究区搁浅浒苔的生物量。研究结果表明：NDVI可以识别植被和非植被，但无法区分潮间带上部和潮间带下部分布的浒苔；DVI和VB-FAH对植被和非植被的区分度不高，但对不同分布的搁浅浒苔具有一定的区分度，其中，DVI对潮间带上部和潮间带下部分布浒苔的识别能力优于VB-FAH。因此，通过对岸边植被进行腌膜，利用DVI构建海滩搁浅浒苔生物量估算模型，实现了海滩搁浅浒苔生物量的估算。海阳、乳山和文登三个海滩搁浅浒苔的生物量分别为1 468 t、745 t和5 034 t，本文提出的方法可以为搁浅浒苔的清理和资源合理分配提供技术支持。
关键词:无人机遥感绿潮搁浅生物量多光谱
DOI：10.11759/hykx20200708001
分类号:P76
基金项目:国家自然科学基金项目（41876107）；山东省联合基金项目（U1706219）；中国科学院海洋大科学研究中心重点部署项目（COMS2019J02）；中国科学院前沿科学重点研究计划（ZDBS-LY-7010）；中国科学院海洋生态与环境科学重点实验室（中国科学院海洋研究所）开放基金资助（KLMEES202005）；国家重点研发计划“蓝色粮仓科技创新”项目（2019YFD0900705）

Estimation of green tide stranded biomass in the Yellow Sea based on unmanned aerial vehicle remote sensing
SHANG Wei-tao^1,2,3, GAO Zhi-qiang^1,2, JIANG Xiao-peng^1,2,3, TIAN Xin-peng^1,2, GUO Shao-fang⁴
1.CAS Key Laboratory of Coastal Environmental Processes and Ecological Remediation, Yantai Institute of Coastal Zone Research, Chinese Academy of Sciences, Yantai 264003, China;2.Shandong Key Laboratory of Coastal Environmental Processes, Yantai Institute of Coastal Zone Research, Chinese Academy of Sciences, Yantai 264003, China;3.University of Chinese Academy of Sciences, Beijing 100049, China;4.Yantai Institute of S&T Information, Yantai 264003, China
Abstract:
Ulva prolifera stranded near a shore damages the coastal landscape, interferes with water sports, and seriously affects coastal tourism. In this study, the multispectral and visible light sensors carried using an unmanned aerial vehicle (UAV) were used to monitor the stranded Ulva prolifera on the beaches of Haiyang, Rushan, and Wendeng in the Shandong Peninsula. The normalized difference vegetation index (NDVI), difference vegetation index (DVI), and virtual baseline height floating algae index (VB-FAH) were selected to identify and evaluate the beach-stranded Ulva prolifera and shore vegetation and nonvegetation (sea water and sand), respectively, and the biomass of the stranded Ulva prolifera in three research areas was estimated. The results showed that the NDVI can identify vegetation and nonvegetation but it cannot identify the Ulva prolifera distributed in the upper and lower tidal zones. DVI and VB-FAH cannot distinguish between vegetation and nonvegetation but they can identify Ulva prolifera distributed in the upper and lower tidal zones. In these zones, DVI was superior to the VB-FAH in identifying the distribution of Ulva prolifera. Therefore, the biomass estimation model of the beach-stranded Ulva prolifera was established by masking shore vegetation; and using DVI, the biomass of the beach-stranded Ulva prolifera was estimated. The biomass of the stranded Ulva prolifera was 1468, 745, and 5034 t in the beaches of Haiyang, Rushan, and Wendeng. The method proposed in this study can provide technical support for the cleaning up of the stranded Ulva prolifera and rational use of resources.
Key words:UAV remote sensinggreen tidestranded biomassmultispectral





PDF全文下载地址:

点我下载PDF