基于复杂网络的海洋涡旋移动特征研究——以南海为例

发布时间：2018-06-13 17:03

本文选题：复杂网络 + 区域划分　；参考：《海洋学报》2017年07期

【摘要】：海洋涡旋作为一种快速连续变化的海洋现象,如何分析和挖掘其移动特征成为当前海洋涡旋定量研究的重点。本文引入空间数据挖掘的社区网络划分方法,将涡旋过程看作复杂的移动网络,对涡旋移动的聚集性特征进行探索和分析。首先,以网格为统计单元对1992-2011年近20年南海海洋涡旋移动数据进行组织,基于图论模型构建了涡旋瞬时移动(TP),涡旋移动起止点(OD),涡旋最小描述距离的特征点移动网(MDL)和涡旋过程移动再生数据(RSP)4种状态的海洋涡旋的移动网络图;其次,采用基于快速模块度优化的区域划分方法分别得到4种状态下涡旋移动的聚集性区域;最后,利用弦图对区域内和区域间涡旋移动规律进行了可视化分析,发现海洋涡旋的RSP数据能够弥补原始涡旋移动数据在区域划分方法中呈现的数量不足的问题,能够在足够数据量的情况下,有效地发现从起点到终点的主要移动通道和涡旋移动的聚集性区域,这些区域反映了南海涡旋从其产生、发展到结束整个演化过程的聚集性特征。
[Abstract]:As a rapid and continuous ocean phenomenon, the ocean vortex is a fast and continuous ocean phenomenon. How to analyze and excavate its movement features is the key point of the current quantitative study of the oceanic vortex. This paper introduces the method of community network division of spatial data mining, regards the vortex process as a complex mobile network, and explores and analyzes the characteristics of the aggregation of vortex movement. First, using the grid as the statistical unit to organize the data of the South China Sea Ocean vortex movement in the last 1992-2011 years, based on the graph theory, the moving network of the vortex instantaneous movement (TP), the scroll moving start stop (OD), the feature point mobile network (MDL) and the vortex process moving regenerative data (RSP) are constructed. Secondly, the region partition method based on fast modularity optimization is used to obtain the aggregation regions of vortex movement in 4 states. Finally, the visualization analysis of vortex movement between regions and regions is carried out by using string graph. It is found that the RSP data of the ocean vortex can make up for the original vortex moving data in the region division method. The problem of insufficient quantity can effectively discover the main moving channels from the starting point to the end point and the aggregation region of vortex movement, which reflect the aggregation characteristics of the South China Sea vortex from its generation to the end of the whole evolution process.
【作者单位】：中国科学院地理科学与资源研究所资源与环境信息系统国家重点实验室;中国科学院大学;
【基金】：国家自然科学基金项目(41371378,41421001)
【分类号】：P731.2

【参考文献】