一种基于TIN的多尺度流域河网提取算法

Research on algorithm for extracting multi-scale drainage network based on TIN

多尺度水文模拟是水文研究的重要领域,多尺度的流域河网是多尺度水文模拟的基础,而现有基于TIN提取算法存在河道定义单一、未处理平坦区域及不支持多尺度等不足。提出一种基于TIN的多尺度流域河网提取算法,定义谷线河道、重心—重心河道和重心—谷线河道等流域河网特征;利用回溯的方法确定平坦区域三角形的流向,进一步确定平坦区域的河道;定义基于TIN数据的汇水面积的概念,用以实现空间多尺度流域河网的提取。设计河网的二叉树拓扑结构和编码方案及算法,用于构建流域河网的空间拓扑结构为水文研究提供接口。研究及实验结果表明,提出的算法提取出的流域河网与地形数据中的实际河网吻合。

Multi-scale hydrologic modeling is an important area of hydrological research.Multi-scale river basin is the basis of multi-scale hydrological modeling.While existing extraction algorithm which based on TIN has some shortages such as the definition of river singly,treating only for non-flat areas and un-supporting for multi-scale river basin.This thesis presents a method for extracting multi-scale drainage network based on TIN,defines some features of river basin such as the line,the river valley,the center of gravity-the center of gravity and center of gravity river-the river valley line;determines the flow of the triangular flat areas by using the method of flat back,then determines the river of flat areas further;introduces the concept of watershed area based on TIN to extract multi-scale drainage network.It proposes a binary tree topology with a coding schema to represent the drainage network,providing interfaces for digital hydrology research.The result indicates that the drainage network extracted by this method is basically consistent with the actual river.