近海开发影响下底栖生境演变的声学观测与评估研究

	近海开发影响下底栖生境演变的声学观测与评估研究
	李东 1,2
学位类型	博士
导师	张华 ; 唐诚
	2017-05
学位授予单位	中国科学院大学
学位授予地点	北京
关键词	围填海人工鱼礁多波束底栖生境自动分类
摘要	近海开发已成为沿海地区进行空间资源扩展的主要途径。大规模沿海开发活动如围填海工程、海洋牧场建设等在促进了沿海地区社会经济发展的同时，对海洋底栖生境也产生了显著影响。声学方法改变了传统调查方法效率不高、获取信息不全的缺点，为底栖生境调查提供了一种新的技术方法。多波束声呐系统是当代海洋勘测中一项高新技术产品，国内将其应用于近海底栖生境观测、评估的研究还比较少见；声学数据结合分类算法可以实现底栖生境的识别与评估，但目前还缺乏人工鱼礁区底栖生境自动化分类方法的相关研究。本论文在总结多波束数据处理方法的基础上设计了一种底栖生境自动化制图方法，实现人工鱼礁区底质类型的识别；利用高分辨率、高精度的多波束声呐系统，分析人工鱼礁区、围填海区微地形地貌特征。主要研究成果如下： 1、设计了一种底栖生境自动分类方法，并将其应用于人工鱼礁区。结果表明：（1）多波束数据量大，包含信息丰富，通过数据挖掘可以提取反映海底底质属性的多种声学变量（如坡度slope、底栖位置指数BPI等）；（2）PCA能够在保留主要信息的同时，将多种声学变量实现数据压缩，减少数据冗余（特征值大于1的前3个PCs的累积频率近80%）；（3）ISODATA非监督分类算法与人工解译方法相比，处理速度快效率高，可以消除相邻声相边界的不确定性，确保分类结果的正确性，大大降低操作人员的偏差，节省人力成本。这一自动化智能分类方法在山东近海大规模人工鱼礁区取得了良好效果，实现了鱼礁区三种基本底质（沙质海底、泥质海底及人工鱼礁）的分类识别。底质取样、水下摄像验证结果与自动化分类结果保持一致，进一步证明了该方法的有效性。人工鱼礁区底栖生境的自动化分类技术可以将复杂的多波束数据转化为可视化的专题图，能够使用户快速获取感兴趣区的海底底质信息，满足鱼礁管理的需求，具有较强的实际意义。该底栖生境制图方法主要在GIS平台下完成，可以实现进一步的空间分析与算法改进。 2、利用多波束系统结合地理空间分析方法实现了大规模人工鱼礁区底栖生境的监测与评估。分析结果显示：（1）采用多波束声呐系统能够快速获取鱼礁的形态、分布及其周围海底的微地形地貌信息，比传统调查方法（如水下探摸）准确度高、效率高、成本低；（2）借助地理信息系统地形分析技术可以提取鱼礁区地形特征变量，坡度、曲率、粗糙度、地形耐用指数以及地形起伏度的高值区均出现在礁石分布区域，能够将礁石从平坦海底区分开来；（3）由于自身重力及水动力作用，礁石发生沉降现象（沉降深度达0.45 m），鱼礁的存在使底层流速减弱、流向发生变化，其周围出现特有的冲淤地形，多波束水深数据可以定量分析投石后引起的海底地形变化特征，为人工鱼礁建设评估工作提供数据支持；（4）不同类型的人工鱼礁，其构造不同，形成的局部流场也不同，为底栖生物创造了不同的底栖环境，实地验证发现水泥预制件礁体比山石鱼礁的聚鱼效果更为明显。高分辨率的多波束数据不仅可以反映出海底鱼礁的宏观信息（分布、规模、占地面积等），还可以提供鱼礁的微观信息（三维结构、大小、形状及沉降深度等），达到人工鱼礁精准监测的目的。 3、结合历史海图与多波束数据，在GIS平台下构建龙口湾水下DEM，分析了龙口湾近50年的形态演变过程。结果显示：1960s到2010s半个世纪以来，龙口湾水下地形地貌发生了巨大改变，研究区水域面积减少了13.5 km²。具体来说：（1）1960s~1990s，龙口湾自然演变，受人类活动影响较小；（2）1990s~2000s，龙口港扩建加之航道清淤，龙口湾形态发生了明显变化，龙口浅滩面积缩小，平均水深从7.56 m增加到8.16 m；（3）2000s~2010s，龙口港的进一步扩展及离岸人工岛的建设使龙口湾海底地形发生了剧烈改变，龙口浅滩基本消失，人工岛北部水下地形复杂、海底较为破碎，陆地面积从23.10 km²增长到29.35 km²。近几十年频繁的海岸活动（包括港口扩建、航道清淤及人工岛建设）已成为决定龙口湾形态变化的主导因素。高分辨率的多波束数据显示，人工岛吹填工程抽取沉积物后，在海底留下了大量取土坑，平均水深加深了2.34 m，使海底地形地貌发生了极大改变，对海底底栖生境造成了严重破坏。本研究中设计的鱼礁区底栖生境自动化分类方法，是海底声学探测领域的一次尝试，有助于人工鱼礁的科学管理，具有较强的现实意义；利用多波束声呐手段结合地理空间分析方法获取人工鱼礁区、围填海区微地形地貌特征，为监测近海人类活动影响下的底栖生境演变提供了一种新的技术手段。
其他摘要	Offshore development has become the main way to expand spatial resources in the coastal areas. Large-scale exploitation activities such as reclamation and marine ranching construction have promoted economic and social development, but they also had a significant impact on the marine benthic habitat. Acoustic methods, which provide a new technique for benthic habitat survey have overcome the shortcomings of traditional means that are inefficient and provide incomplete information. Multibeam echo sounder system (MBES) is a high-tech product in modern marine survey. It is rare to apply multibeam sonar for offshore benthic habitat observation and evaluation in our country. Benthic habitat can be identified and classified by acoustic data combined with classification algorithm. However, there is a lack of research on the acoustic mapping and classification for benthic habitat in artificial reef (AR) water. In this paper, a method of automated mapping for benthic habitat was developed and applied to identify the habitat types in AR area. The micro-topography features of the AR area and reclamation area were analyzed by using MBES with high-resolution and high-precision. The main results of this paper are as follows: (1) An automated classification method is designed and tested to process bathymetric and backscatter data from MBES and transform the variables into simple, easily visualized maps. A suite of acoustic derived datasets was processed using ArcGIS^TM for further characterization of benthic habitats. These variables are valid indicators of substrate type and well represent variation in seafloor characteristics. To reduce the redundancy in acoustic variables, a principal component analysis (PCA) is used to condense the highly collinear dataset. An acoustic benthic map of bottom sediments is classified using an iterative self-organizing data analysis technique (ISODATA). The ISODATA clustering method can greatly reduce the likelihood of operator bias, reduce the time and cost of data interpretation, and ensure the consistency of the classification results. This method is implemented in a GIS environment, which allows further spatial analyses and algorithm development. The main benefit of ISODATA is that it combines multiple variables into a simplified comprehensive overview of the interested area. The approach is tested with MBES surveys in a 1.15 km² fish farm with a high density of ARs off the Yantai coast in northern China. Using this method, 3 basic benthic habitats (sandy bottom, muddy sediments, and ARs) are distinguished. The results of the classification are validated using sediment samples and underwater surveys. Our study shows that the use of MBES is an effective method for acoustic mapping and classification of ARs. (2) The benthic habitat in large-scale AR was monitored and evaluated by using MBES combined with geospatial analysis. The results showed that the multibeam sonar system can obtain detailed bathymetric data of ARs to identify their location, form and other information. The topographic variables such as slope, curvature, rugosity, terrain ruggedness index (TRI) and topographic relief (TR) were extracted by using terrain analysis tool to examine the ARs topographic features. The variables with high-value distributed around the ARs, and this can distinguish ARs from the flat seabed. ARs reduced the underlying flow rate and changed the current direction. Because of the AR’s gravity and hydrodynamic effect, the seabed terrain changed significantly (depth from 5 to 10 m) and there were erosion (sinking depth about 0.45 m) and deposition appearance around ARs. Multibeam bathymetry data can be used to quantitatively analyze the changed characteristics of submarine terrain caused by ARs deployment and provide necessary data to evaluate the AR effect. Different kinds of ARs with distinctive structure formed different local flow and supplied diverse benthic habitat. Field surveys found the biomass around concrete reef was much larger than that of stones reef. It is possible to provide a new technique to analyze the micro topography changes caused by ARs and this could contribute to ARs examination and assessment work which has a strong practical significance. (3) Analysis of a series of historical bathymetric surveys has revealed large changes in morphology from 1960s to 2010s in Longkou Bay, China. Water depths were extracted from digitized admiralty charts to explore the accretion-erosion characteristics in a Geographical Information System (GIS) environment, providing quantitative estimates of morphological changes. Multibeam swath mapping were used to analyze the geomorphologic features caused by the construction of artificial islands. Results illustrated that the shoreline and bathymetry of Longkou Bay changed dramatically in recent decades. The subaqueous area decreased by 13.5 km² in the study area during the last 50 years. From 1960s to 1990s, the evolution of Longkou Bay was mainly governed by natural processes with a patchy distribution of deposition and erosion, and there were few signs of being related to large-scale human activities. During the period of 1990s to 2010s, intensive coastal developments including large port engineering projects, channel dredging and artificial islands construction became the main processes affecting morphological changes in the Longkou Bay. The high-resolution bathymetric results near the artificial island showed that the seafloor was dredged at many sites and the mean water depth increased over 2 m, leaving large areas of borrow pits. The sudden change of the underwater topography will lead to the destruction of local benthic habitat and effective measures need to be taken to protect and remediate heavily disturbed subaqueous environment. In this study, the automated classification approach for benthic habitat in AR water we designed was an attempt in the field of seabed acoustic exploration and filled the gap that there was no application in AR area by using acoustic mapping method. The MBES combined with geospatial analysis was used to obtain the micro topography features in the AR area and reclamation area. This provided a new technique for monitoring the evolution of offshore benthic habitat which is largely influenced by large-scale human activities.
文献类型	学位论文
条目标识符	http://ir.yic.ac.cn/handle/133337/22001
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	李东. 近海开发影响下底栖生境演变的声学观测与评估研究[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
????????????????????（22342KB）	学位论文		开放获取	CC BY-NC-SA	请求全文