基于多源数据的黄海绿潮遥感监测研究

	基于多源数据的黄海绿潮遥感监测研究
	郑翔宇1,2
学位类型	硕士
	2017-06
学位授予单位	中国科学院大学
学位授予地点	北京
其他摘要	近些年来，黄海海域每年都会暴发不同规模的绿潮（大型藻类——浒苔）灾害，对当地养殖业、旅游业、交通运输业、海洋生态环境等造成了严重的危害。针对绿潮暴发持续时间长、规模大、位置不固定的特点，利用遥感手段进行监测显得尤为重要，但目前监测所用的卫星遥感数据，或空间分辨率过低，监测精度得不到保障，或时间分辨率过低，监测的时间序列跨度过大。因此，为了弥补这类不足，本文以南黄海地区为研究区，综合使用多种卫星遥感数据（GF-1 WFV、HJ-1A/1B CCD、CBERS-04 WFI、Landsat-7 ETM+、Landsta-8 OLI、MODIS SST-8day）以及无人机和船测数据，结合遥感和GIS技术对2014至2016年黄海绿潮进行了监测，并在此基础上进行了相关研究。本文的主要研究内容为：（1）通过对同一影像分别采用不同的大气校正方法进行校正，并对多个变化量的进行统计与分析，确定当使用NDVI指数提取绿潮时，效果最好的大气校正方法。（2）分析2014年至2016年黄海绿潮的时空分布特征；对不同数据源的监测结果进行对比。（3）以2016年的黄海海表温度和绿潮监测结果为例，分析两者之间的相关性；根据本文的研究结果，从遥感角度研究绿潮灾害的防控策略。研究结果表明：（1）当采用NDVI阈值法提取绿潮信息时，COST大气校正后的影像绿潮提取效果最好，FLAASH、6S大气校正依次次之，但COST大气校正在其它绿潮提取算法中的适应性仍待进一步考证。（2）宏观上看，2014至2016年黄海绿潮的时空分布特征基本一致，四月底至五月初绿潮初生于江苏辐射沙脊群，五月份不断生长并持续向北推移，六月份开始暴发，7月份绿潮开始消亡，至八月份绿潮灾害基本结束；通过对多种数据源的监测结果进行对比，空间分辨率带来的混合像元效应是产生监测误差的主要原因。（3）黄海海表温度与绿潮的暴发具有很大的相关性；为了防控绿潮灾害，首先从长远来看，要从源头降低海水富营养化程度，对于近期暴发的绿潮，可采取前置打捞，及时预警的策略，降低绿潮带来的危害。综上分析，本研究利用多源数据对黄海绿潮进行动态监测，并比较了监测结果，提升了监测精度和置信度，具有一定的创新性。另外，本研究从影响绿潮提取的大气校正因素、时空分布特征、监测结果对比、温度因子与绿潮暴发的相关性、防控策略等方面对黄海绿潮展开研究，一定程度上丰富了对绿潮的认识，对防治绿潮、减少损失具有重要的现实意义。 ; Different scales of green tide (Macroalgae-Ulva prolifera) disaster have run rampantly in the Yellow Sea annually in recent years and causing huge losses to local aquaculture, tourism, transportation, marine ecological environment. In view of the characteristics of long duration, large scale and fixed position, it is very important to use remote sensing to monitor green tide. However, the current remote sensing data which is used for monitoring has all kinds of limitations: either the spatial resolution is too low, the monitoring accuracy can’t be guaranteed, or the time resolution is too low, the time series of monitoring is too sparse. Therefore, in order to make up for this deficiency, this paper takes the Yellow Sea as the research area, comprehensive using a variety of satellite remote sensing data (GF-1 WFV, HJ-1A/1B CCD, CBERS-04 WFI, Landsat-7 ETM+, Landsta-8 OLI, MODIS SST-8day), as well as UAV data and ship-measured data to monitor the green tide disaster in the Yellow Sea from 2014 to 2016. On this basis, some relevant researches are also carried on in this paper. The main research contents are listed as follows: (1) By using different atmospheric correction methods to correct the same image, calculating and analyzing a number of statistics, the best atmospheric correction method is confirmed when extracting green tide information by NDVI index. (2) Analyze the temporal and spatial distribution characteristics of green tide disaster in the Yellow Sea from 2014 to 2016; Compare the monitoring results from different data sources. (3) By taking the sea surface temperature data of the Yellow Sea and the monitoring results in 2016 as an example to analyze their correlation; Discuss the prevention and control strategy of green tide disaster from the perspective of remote sensing. The results prove that: (1) When NDVI threshold method is used to extract the green tide information, the extraction effect is the best after the image is corrected by COST method, and followed by FLAASH and 6S correction methods. But the adaptability of COST atmospheric correction in other green tide extraction algorithms is still need further investigation. (2) From a macro point of view, the spatial and temporal distribution of the green tide in the Yellow Sea is basically the same in the three years. It firstly appeared in the radial submarine sand ridges system off Jiangsu Province from late April to early May, and then continuously grew and drifted to northward, broke out in June and went extinct in July, and finally ended in August. By comparing the monitoring results of different data sources, the mixed pixel effect caused by spatial resolution is the main reason of monitoring errors. (3) There is a significant correlation between the outbreak of the green tide and the SST of the Yellow Sea; In order to prevent and control the green tide, in the long run, we should reduce the degree of eutrophication of the sea from the source, for the recent outbreak of green tide, we can take strategy of pre salvage and timely warning to reduce the disaster caused by green tide. In summary, by using multi-resource data, the dynamic monitoring of the green tide in the Yellow Sea is given and the monitoring results are compared in this paper. It improves the monitoring accuracy and confidence, therefore this study has a certain innovation in the field. Besides, this paper conduct research from aspects of atmospheric correction factors affecting the extraction of green tide information, spatial and temporal distribution characteristics, comparison of monitoring results, correlation between temperature factors and outbreak of green tide, prevention and control strategy. To some extent, it enriches the understanding of the green tide, and be of important practical significance for preventing and controlling the green tide disaster and reducing the loss.
语种	中文
文献类型	学位论文
条目标识符	http://ir.yic.ac.cn/handle/133337/22004
专题	中国科学院烟台海岸带研究所知识产出_学位论文
作者单位	1.中国科学院烟台海岸带研究所 2.中国科学院大学
第一作者单位	中国科学院烟台海岸带研究所
推荐引用方式 GB/T 7714	郑翔宇. 基于多源数据的黄海绿潮遥感监测研究[D]. 北京. 中国科学院大学,2017.

条目包含的文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
基于多源数据的黄海绿潮遥感监测研究.pd（6169KB）	学位论文		开放获取	CC BY-NC-SA	请求全文