【作者】 李行；

【导师】 周云轩；

【作者基本信息】 华东师范大学 ， 自然地理学， 2010， 博士

【摘要】 近年来,长江三角洲在流域筑坝拦沙和海岸带人类活动的影响下开始出现退化迹象。随着全球气候变暖导致的海平面的进一步上升,长江三角洲的海岸侵蚀问题将会对社会经济发展构成潜在的威胁。鉴于该区域在我国社会经济发展中的重要性,有必要开展海岸侵蚀灾害管理决策支持技术的研究,以防患于未然。基于这一考虑,本文主要利用遥感和GIS技术对长江三角洲海岸侵蚀决策支持系统若干关键技术进行了研究,为进一步建立业务化运行的决策支持系统提供理论和技术保障。主要研究内容包括4个方面：(1)在系统讨论了基于GIS的岸线变化分析方法的基础上,考虑到岸线演变与海岸地形间的相关性,就淤泥质非平直岸线提出了基于地形梯度的正交断面方法。以崇明东滩鸟类国家级自然保护区为例,对该方法进行了验证。我们选择1987年至2006年间的6景TM遥感影像,以植被线作为岸线指标,在eCognition和ArcGIS软件的支持下,解译出了对应年份的数字岸线,并利用所提出的正交断面方法进行分析。结果显示：崇明东滩以东南角节点为界,分为南侧的侵蚀岸段和其余的淤涨岸段。总体淤涨速率呈减慢的趋势。最大侵蚀速率为22.0m/a,最大淤涨速率为247.2m/a。北侧自东旺沙水闸向东约4km长的岸段存在明显的冲淤交替现象,但总体上呈现缓慢的淤涨趋势。此外,两个岸线演变受抑制区段都位于岛影缓流区的边界。(2)长江是全球范围内受水利工程影响最为严重的河流之一。入海泥沙通量自1960年代以来就持续降低,已经导致了长江口水下三角洲的局部退化。崇明东滩作为长江口少有的自然淤涨潮滩且处于长江河口的中心位置,其岸线的演变必然受到了流域来沙减少趋势的影响,而且崇明东滩对流域来沙变化的响应模式将对长江河口水下三角洲的演变具有一定的指示意义。我们选择1987、1990、1995、1998、2002、2006、2008七个年份的TM遥感数据,利用所提出的正交断面法计算了崇明东滩的岸线变化率,将其与大通站泥沙通量相关联,间接地研究了崇明东滩岸线变化与流域大型水利工程之间的关系。分析显示,崇明东滩的岸线变化率与大通站泥沙通量之间具有明显的相关性(P<0.03)。进一步计算表明,对于崇明东滩的淤涨变化而言,大通站泥沙通量的临界值约为0.24/0.29亿吨。根据其他研究者的预测结果,我们计算发现,在2003-2005年间三峡大坝导致崇明东滩岸线变化速率平均降低了15.3m/a,相当于1987年以来岸线变化速率总降低量的15%。同时,南水北调工程潜在的影响约为三峡大坝的5-10%。崇明东滩在未来的20-50年内将达到其淤涨极限,届时侵蚀将占主导地位。考虑到其它在建和规划中的大型水利工程项目,及其可能的级联效应,这一周期将会大大缩短。(3)沿岸输沙率是海岸工程中最重要的物理量之一,与岸滩演变、港口航道回淤、以及岸滩对海岸工程的地貌响应等许多问题有关。传统的沿岸输沙率估算方法所需要的水动力学参数很难精确获取,而且针对淤泥质海岸的研究较少。为此,我们以南汇边滩南槽岸段为例,研究了在GIS环境下对一线模型进行修正进而估算沿岸输沙率的方法。主要步骤包括：(a)以海图数据中的0m等深线作为岸线指标计算岸线变化速率；(b)针对试验区的特点,设计计算单元,并在此基础上对一线模型进行合理的修正；(c)将岸线变化速率、闭合深度等参数代入修正后的模型,反算出沿岸输沙率。与地形法的计算结果比较显示,估算结果的总体相对误差为8.77%,证明了所提出方法的有效性。(4)根据长江三角洲海岸侵蚀的复杂性,建立了适用于长江三角洲的海岸侵蚀决策支持框架模型,该框架由人与计算机系统构成,其中计算机系统主要包括海岸侵蚀综合数据库、基于GIS的海岸侵蚀风险评价模型、可视化工具集和情景生成器4个主要部分。在该框架下,重点讨论了基于GIS的海岸侵蚀风险评价模型。首先,选择行政单元、人口密度、底质类型、海岸特征4种自然和社会因子对研究区域的岸线进行分割,得到相对同质的评价单元。其次,利用DPSIR (Driving forces, Pressure, State, Impact, Response)指标框架模型分析了与长江三角洲海岸侵蚀密切相关的自然和社会因素,建立了适合于长江三角洲实际情况的海岸侵蚀风险评价指标体系,该指标体系包括10个脆弱性因子(近岸高程、岸滩坡度、沿岸输沙率、岸线变化率、平均潮差、有效波高、相对海平面变化、潮滩宽度、潮滩植被类型、潮滩植被带宽度)和3个影响因子(人口密度、主要土地利用类型、重点生态区域)。然后,利用本文所讨论的方法和相关研究成果对各因子进行量化分级。并引入层次分析法(Analytic Hierarchy Process,AHP)结合专家知识计算了各评价因子的权重。进而,针对脆弱性因子和影响因子分别计算得到脆弱性指标和灾害指标。将两个指标加权平均得到风险指标。评价结果表明,宝山区岸段、浦东新区五号沟至三甲港岸段和金山区金山咀以西岸段风险水平最高；浦东新区的其余岸段、奉贤区至金山区金山咀岸段、横沙通道沿岸、崇明岛西南角正对白茆沙的岸段以及江苏启东市连兴港以西正对顾园沙的岸段为高风险水平；启东港沿岸、崇明岛南岸、长兴岛大部分岸段和横沙岛北岸、太仓市沿岸以及南汇区大治河口以南岸段为中等风险水平岸段；崇明东滩、九段沙以及北支沿岸的部分岸段风险水平最低；其余岸段的为低风险岸段。更多还原

【Abstract】 In the past, the huge amount of sediment load from the Yangtze River contributed to the successive accretion of the Yangtze Delta. However, contemporarily global warming and the construction of large hydraulic engineering works in the catchment and its estuary significantly decelerate the trend, and the delta degradation has been observed since 2003. Considering the combined influence of the planning dams and the South-to-North Water Diversion Project, the anticipated coastal erosion problem will be even worse. Since the Yangtze Delta is one of the most important economic regions in China, a long-term coastal spatial planning will be an inevitable issue for local governments. While, most coasts in the Yangtze Delta are typically muddy that are characterized by extremely complex land-sea interaction. In view of the complexity and uncertainty of human and natural dimension under the circumstances of global change, the need for integrating human and computer intelligence necessitates a Decision Support system (DSS) for the risk assessment of coastal erosion. In this thesis, we specially focused on some key techniques for the DSS. It mainly includes four aspects:(1) we develop an orthogonal transect scheme based on topographic gradient to trace the realistic changes of non-straight shoreline, and furthermore investigate the muddy non-straight shoreline dynamics of Chongming Dongtan in Shanghai using mapping based shoreline evolution models. Also, a shoreline interpolation was conducted in order to make the orthogonal transects trace the more realistic shoreline evolution. Based on the eCognition object-oriented image segmentation method, the shorelines of Dongtan are extracted from 6 scenes of Landsat-TM satellite images of 1987,1990,1995,1998,2003, and 2006, respectively. The results derived from the orthogonal transect scheme indicate that the shorelines of Dongtan are divided into the eroding segment in the south and accreting segment in the east and north. The eroding shoreline segments basically maintain unaltered and the accreting shoreline segments continue to expand by gradually decreasing speed. The maximum accretion rate and the maximum erosion rate are respectively 247.2m/a and 22.0m/a. The "suppressed" shoreline segments lie in the region of the boundaries of the island shadow. The results obtained appear coherent to the true situation, and can be utilized as a reference of decision making for coastal management.(2) Few rivers in the world are influenced so markedly by hydraulic engineering works as Yangtze River. An effective decrease in sediment discharge to the estuary has been induced since 1960s. As one of the only few naturally accreting tidal flats in Yangtze estuary, Chongming Dongtan would definitely modify its shoreline dynamics in response to the decrease. Based on Landsat-TM satellite images of 1987,1990, 1995,1998,2002,2006, and 2008, the proposed orthogonal transect method is used to obtain the shoreline change rate. According to the calculated average shoreline change rate, we divide the shoreline of Dongtan into four segments, erosion segment, rapid accretion segment, stabilization segment, and general accretion segment. They have different response patterns to the reduced sediment discharge. Furthermore, we provide a quantitative assessment of the impact of sediment discharge by regression analysis. The results show that, during 2003-2005, the Three Gorges Dam (TGD) and possible cascade effects contributed about 15% of total decrease of average shoreline change rate since 1987, while the anticipated impact of the South-to-North Water Diversion Schemes would be 5-10% of the TGD. And a critical sediment discharge at Datong Station, about 0.24/0.29×108 tons/year, controlling the transition of accretion to erosion in Dongtan is found. Without actions taken, Dongtan would reach the limits to growth within coming 20-50 years or a shorter span of time.(3) A new integration approach of beach evolution model and GIS was developed to evaluate the longshore sediment transport rate. By partitioning calculation units and modifying one-line model to meet complex physical setting, the integrated scheme was established. Firstly, the shoreline change rates are derived from EPR method based on GIS, where the shoreline indicator is selected as Om-isobath from digitized nautical charts. Then, the closure depth was obtained using the available information about the research results in literatures and the bathymetry of a specific coast. Finally, the longshore sediment transport rates can be calculated inversely in calculation units by substituting those values of the shoreline change rates and the closure depth into the modified one-line model. The estimated longshore sediment transport rates would include the accumulated effect of episodic events, sediment runoff transport, cross-shore sediment transport, and sediment exchange between flats and channels, etc., which were usually difficult to model. In experiments, the longshore sediment transport rates are computed for a portion of Nanhui Beach of Shanghai, respectively, by the proposed approach and topographic method. The results present that the total relative error is 8.77%. And some suggestions are also given to improve the proposed approach.(4) We present a Decision Support Framework (DSF) for the risk assessment of coastal erosion in consideration of the potential management problems and challenges for economic development in the coastal zone of the Yangtze Delta. The framework consists of four major components:integrated database, GIS-based risk assessment models, visualization toolkit and scenarios generator. Especially, we developed a GIS-based risk assessment model for the muddy coasts of Yangtze Delta. Firstly, four primary indicators including administration unit, population density, geological types and coastal characteristics were used to segment the shorelines into the relatively homogeneous units. Then, the DPSIR (Driving forces, Pressure, State, Impact, Response) framework was used to analysis physical and socioeconomic factors asscociated to the coastal erosion of the Yangtze Delta, and constructed an indicator system that includes 10 vulnerability indicators, i.e., coastal elevation, coastal slope, longshore sediment transport rates, shoreline change rates, tidal range, significant wave height, relative sea level rise, intertidal width, intertidal vegetation type and intertidal vegetation zone width, and 3 impact indicators, i.e., population density, land use type and ecological hotspots. Subsequently, depending on the nature of each of these variables, they were assigned ranks ranging from 1 to 5, with 1 representing minimum vulnerability/hazard and 5 representing maximum vulnerability/hazard. Before evaluating the vulnerability and hazard, these variables were weighted based on their relative importance in determining the coastal erosion vulnerability and hazard. An Analytic Hierarchy Process (AHP) was employed for the variables weightings. Finally, the Vulnerability Index (VI) and Hazard Index (HI) were evaluated based on a linear model. And the Risk Index (RI) was calculated by a simple weighted average of VI and HI. The assessment results show the validity of the approach. Accordingly, the DSF will make the specialized data and information more accessible to managers and other stakeholders, and has an extensive capability to facilitate communication and synergetic work between humans and computers. In this way, the proposed DSF is expected to make manager make more scientific decision.更多还原