基于Pfafstetter规则的流域编码算法并行化方法

发布时间：2018-10-30 15:36

【摘要】：流域编码是以子流域划分进行流域相关研究的重要内容。Pfafstetter流域编码以编码唯一、顾及流域拓扑关系及编码效率高等优点而被广泛采用。本文在流域相关研究的分析范围不断增大、数据精度越来越高的需求背景下,以Pfafstetter编码为基础,对流域编码并行化方法进行研究。首先,分析了Pfafstetter编码不全面和码位不一致的问题,改进了Pfafstetter编码规则;然后,从数据并行的角度,讨论了并行计算环境下的数据划分及并行化策略,进而设计了流域编码并行算法;最后,利用长江中上游流域SRTM数据,在集群系统上对流域编码并行算法的正确性和并行性能进行了测试。实验结果表明,本文设计实现的流域编码并行算法可获取与实际较为一致的计算结果,且提高了编码计算效率,可为基于子流域划分的流域分析并行化提供参考。
[Abstract]:Watershed coding is an important part of watershed classification. Pfafstetter watershed coding is widely used to encode only, take into account the advantages of basin topology and high coding efficiency. In this paper, based on Pfafstetter coding, the parallelization method of watershed coding is studied under the background of increasing analysis scope and increasing data precision. Firstly, the problems of incomplete Pfafstetter coding and inconsistent code points are analyzed, and the rules of Pfafstetter coding are improved. Then, from the point of view of data parallelism, the data partition and parallelization strategy in parallel computing environment are discussed, and the parallel algorithm of watershed coding is designed. Finally, the correctness and parallel performance of the parallel algorithm are tested on the cluster system using the SRTM data of the upper and middle reaches of the Yangtze River. The experimental results show that the proposed parallel algorithm of watershed coding is consistent with the actual results, and improves the efficiency of coding calculation. It can be used as a reference for the parallelization of watershed analysis based on sub-watershed division.
【作者单位】：滁州学院安徽省地理信息集成应用协同创新中心;
【基金】：国家自然科学基金项目(41201415) 教育部科学技术研究重点项目(212078) 测绘遥感信息工程国家重点实验室开放基金项目(14I02) 滁州学院科研启动基金项目(2014qd028)
【分类号】：P208;TP338.6

【参考文献】