测绘通报 ›› 2020, Vol. 0 ›› Issue (2): 90-95.doi: 10.13474/j.cnki.11-2246.2020.0051

• 技术交流 • 上一篇    下一篇

融合多尺度影像数据的杞麓湖流域景观格局分析

白少云1, 刘斌1, 余珮珩2, 陈敏2, 潘佳威2, 陈奕云2,3   

  1. 1. 云南省水利水电勘测设计研究院, 云南 昆明 650021;
    2. 武汉大学资源与环境科学学院, 湖北 武汉 430079;
    3. 土壤与农业可持续发展国家重点实验室, 江苏 南京 210008
  • 收稿日期:2019-10-12 出版日期:2020-02-25 发布日期:2020-03-04
  • 通讯作者: 余珮珩。E-mail:peiheng@whu.edu.cn E-mail:peiheng@whu.edu.cn
  • 作者简介:白少云(1971-),男,教授级高级工程师,从事水利水电勘测、大型精密工程测量方法的研究及应用。E-mail:baishaoyun_on@sina.com
  • 基金资助:
    国家高分辨率对地观测系统重大科技专项(89-Y40-G19-9001-18/20-03);云南省院士工作站建设专项(2015IC013);国家自然科学基金(41771440)

Analysis of the Qilu Lake watershed landscape pattern based on multi-scale image data

BAI Shaoyun1, LIU Bin1, YU Peiheng2, CHEN Min2, PAN Jiawei2, CHEN Yiyun2,3   

  1. 1. Yunnan Survey and Design Institute of Water Conservancy and Hydropower, Kunming 650021, China;
    2. School of Resource and Environmental Science, Wuhan University, Wuhan 430079, China;
    3. State Key Laboratory of Soil and Sustainable Agriculture, Chinese Academy of Sciences, Nanjing 210008, China
  • Received:2019-10-12 Online:2020-02-25 Published:2020-03-04

摘要: 景观格局是景观斑块秩序和规律的具体体现,分析景观格局特征能够在空间结构上推进流域生态保护红线的划定和管控工作,维持可持续发展的底线。基于不同遥感数据源测算的景观格局可能存在计算结果差异,以及对景观格局描述不一致的问题。本文以杞麓湖流域为研究区,选取最新的Sentinel-2、GF-1和Landsat 8卫星遥感影像,基于斑块尺度、类型尺度和景观尺度的7个景观指数,从景观类型分布和整体格局两个方面探讨了杞麓湖流域景观格局现状,并分析不同数据源对景观格局评价结果的异同。结果表明:①杞麓湖流域的景观格局以林地和耕地为基底,零碎分散着建设用地和其他用地;②不同卫星遥感影像对流域景观格局的描述存在差异,影像分辨率与斑块数目、最大斑块指数、形状指数、分离度指数存在显著的正相关关系,而对于测度整体的香农多样性指数、香农均匀度指数、优势度指数的影响相对较小。

关键词: 多尺度影像数据, 景观格局, 杞麓湖流域, 最大斑块指数, 形状指数

Abstract: Landscape pattern is the concrete embodiment of landscape patch order and rule. Analysis of landscape pattern characteristics can promote the delimitation and management of the red line of watershed ecological protection in spatial structure. In addition, there may be differences in calculation results and inconsistent description of landscape pattern based on different remote sensing observation data sources. Taking Qilu Lake watershed as the research area, this paper uses the latest satellite remote sensing images of Entinel-2, GF-1 and Landsat 8 to calculate seven landscape indices of patch scale, type scale and landscape scale, discusses the current situation of landscape pattern in Qilu Lake watershed from two aspects including landscape type distribution and overall pattern, and analyses the similarities and differences of landscape pattern evaluation from different data sources. The results show that:① The landscape pattern of Qilu lake watershed is dominated by woodland and arable land, scattered with construction land and other lands. ② Different satellite remote sensing images have different descriptions of the landscape pattern of the watershed, and the image resolution has a significant positive correlation with Number of Patches, Largest Patch Index, Landscape Shape Index and Splitting Index and low correlation with Shannon's Diversity Index, Shannon's Evenness Index and Dominance Index.

Key words: multi-scale image data, landscape pattern, Qilu Lake watershed, largest patch index, shape index

中图分类号: