地理科学进展 ›› 2009, Vol. 28 ›› Issue (2): 174-186.doi: 10.11820/dlkxjz.2009.02.003

• 地表过程 • 上一篇    下一篇

EOF 分析在海岸地貌与沉积学研究中的应用进展

夏非1, 张永战1, 吴蔚2   

  1. 1. 南京大学地理与海洋科学学院,南京大学海岸与海岛开发教育部重点实验室,南京210093;
    2. 南京信息工程大学大气科学学院海洋科学系,南京210044
  • 收稿日期:2008-11-01 修回日期:2009-01-01 出版日期:2009-03-25 发布日期:2009-03-25
  • 作者简介:夏非(1985-),男,江苏南京人,硕士研究生,从事海岸海洋地貌与沉积学研究.E-mail:xiafei1985@sohu.com
  • 基金资助:

    加拿大国际发展署(CIDA)资助———中加合作“中国海岸社区生态规划与环境管理”项目(S-061562-001).

Progress in Applications of the EOF Analysis in the Research of Coastal Geomorphology and Sedimentology

XIA Fei1, ZHANG Yongzhan1, WU Wei2   

  1. 1. School of Geographic and Oceanographic Sciences, Nanjing University|The Key Laboratory of Coastal and Island Development, MOE, Nanjing University, Nanjing 210093, China;
    2. Department of Oceanography, School of Atmospheric Science, Nanjing University of Information Science and Technology, Nanjing 210044, China
  • Received:2008-11-01 Revised:2009-01-01 Online:2009-03-25 Published:2009-03-25

摘要:

海岸地貌形成演化的过程和机制是当今海岸地貌与沉积学研究的核心任务。但是,对于海岸带这样高维数 非线性系统行为的认识与研究还很有限,因而预测海岸地区地貌演变与沉积物输运就变得比较困难,尤其是动力 数值模型在较大时空尺度上的模拟并不理想。数据驱动模型可以一定程度上解决这一问题,并且在数据驱动模型 构建之前,经验正交函数(EOF)分析技术能够有效提取、定量表达数据中的主要型式与它们的时空演变,以及不同 型式之间是如何相互联系的。这不但有助于构建数据驱动模型进行模拟,更有益于理解这些主控因素是如何影响 海岸地貌演化的,进而不断丰富对于海岸地貌演化的过程与机制的认识。故本文在系统整理大量国内外文献的基 础上,首先简要介绍了EOF 分析的基本原理,然后从砂质和淤泥质海岸剖面的时空变化特征、河口近岸海底冲淤 变化的时空演变特征和预测、海岸海洋地区沉积物粒度分布特征和输运等几个方面综述EOF 分析在这些研究中 的应用进展,同时比较分析研究实例中的问题,进而分析当前研究中的不足,提出应该主要从提高海岸地貌与沉积 数据的野外获取手段与时空分辨率,拓宽可供EOF 分析的时间或非时间序列的数据类型;深入挖掘EOF 分析结果 所显示出的海岸地貌与沉积现象中的物理过程与机制,以期为特定时空尺度的海岸地貌与沉积数值模拟提供有价 值的参考信息; 多与其他线性或非线性分析技术联合使用以作深入分析解释或模拟这三方面来做进一步改进,为 实现EOF 分析在海岸地貌与沉积学研究中的深入应用与发展提供借鉴。

关键词: EOF 分析, 过程和机制, 海岸地貌演变预测, 海岸地貌与沉积学, 时空变化特征

Abstract:

The processes and mechanisms of coastal geomorphological evolution are the main tasks of the current research on coastal geomorphology and sedimentology. However, for the coastal zones that are high-dimensional and non-linear systems, their behavior has not been understood well, so predicting sediment transport and coastal geomorphological evolution in these areas are difficult tasks. The physically based prediction models are not modelling well, especially at larger temporal and spatial scales. Data-driving models can resolve such problems to a certain extent, and prior to the establishment of data-driving models, the EOF analysis technique can detect and quantify dominant patterns in the data and their evolution in time and space effectively, as well as how different patterns are related to each other. Thus, it is possible to obtain valuable information on the behavior of coastal zones that may be used not only for developing data-driving models, but also for increasing the understanding of the factors governing the geomorphological evolution, so it can improve the understanding of the processes and mechanisms of coastal geomorphological evolution. Based on a great deal of domestic and oversea references, this paper firstly introduces the principle of the EOF analysis technique, then makes a review of the progress in applications of the EOF analysis in studies of coastal geomorphological and sedimentary problems as the following aspects, characteristics of temporal and spatial changes on sandy and muddy coast profiles, characteristics of temporal and spatial changes of nearshore submarine erosion/accretion and their predictions, distribution of sediments grain sizes and sediments transportation et al., it also analyzes problems and deficiencies in the study cases, and proposes that the current research work should pay more attention to the following three main aspects in order to realize the deeper applications and development of the EOF analysis in the research of coastal geomorphology and sedimentology. First, the temporal and spatial resolution of field data and collection measures must be upgraded; the data types analyzed should be broadened. Second, the processes and mechanisms of coastal geomorphological evolution which the EOF analysis has revealed must be digged deeply in order to provide valuable information for numerical simulation at a certain scale. Third, the EOF analysis technique should be used with other linear or non -linear analysis methods for further data analysis or simulations.

Key words: characteristics of temporal and spatial changes, coastal geomorphology and sedimentology, prediction of coastal geomor -phological evolution, process and mechanism, the EOF analysis