地球信息科学学报 ›› 2016, Vol. 18 ›› Issue (10): 1384-1398.doi: 10.3724/SP.J.1047.2016.01384

• • 上一篇    下一篇

尼泊尔土地覆被遥感制图及其空间格局分析

曹小敏1,2(), 李爱农1,**(), 雷光斌1,2, 谭剑波1,2, 张正健1, 严冬1,2, 谢瀚1,3, 张帅旗1,4, 杨勇帅1,2, 孙明江1,4   

  1. 1. 中国科学院水利部成都山地灾害与环境研究所,成都 610041
    2. 中国科学院大学,北京 100049
    3. 西南交通大学,成都 610031
    4. 成都信息工程大学,成都 610225
  • 收稿日期:2015-03-16 修回日期:2015-04-30 出版日期:2016-10-25 发布日期:2016-10-25
  • 通讯作者: 李爱农 E-mail:llcxm24@163.com;ainongli@imde.ac.cn
  • 作者简介:

    作者简介:曹小敏(1990-),女,硕士生,主要从事土地覆被遥感制图与变化检测研究。E-mail: llcxm24@163.com

  • 基金资助:
    中国科学院国际合作局对外合作重点项目(GJHZ201320);中国科学院重点部署项目课题(KZZD-EW-08-01);中国科学院战略性先导科技专项子课题(XDA05050105)

Land Cover Mapping and Spatial Pattern Analysis with Remote Sensing in Nepal

CAO Xiaomin1,2(), LI Ainong1,*(), LEI Guangbin1,2, TAN Jianbo1,2, ZHANG Zhengjian1, YAN Dong1,2, XIE Han1,3, ZHANG Shuaiqi1,4, YANG Yongshuai1,2, SUN Mingjiang1,4   

  1. 1. Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, Chengdu 610041, China
    2. University of ChineseAcademy of Sciences, Beijing 100049, China
    3. Southwest Jiaotong University, Chengdu 610031, China
    4. Chengdu University of Information Technology, Chengdu 610225, China
  • Received:2015-03-16 Revised:2015-04-30 Online:2016-10-25 Published:2016-10-25
  • Contact: LI Ainong E-mail:llcxm24@163.com;ainongli@imde.ac.cn

摘要:

尼泊尔作为兴都库什-喜马拉雅区域的重要组成部分,是一个从平原到山地再到高原的典型过渡地带,具有多样且复杂的土地覆被类型。开展尼泊尔国家尺度的土地覆被研究,对该国的国土资源管理、生态环境保护具有重要的科学和实践意义,同时也能为中国倡导的“一带一路”国际区域经济合作战略提供基础数据保障。本文选用Landsat TM影像为数据源,基于面向对象与决策树相结合的土地覆被遥感制图方法,生产了尼泊尔全境2010年土地覆被产品(NepalCover-2010),该产品包括8个一级类和32个二级类。同时,本文基于Google Earth高分辨率影像获取验证样本开展了NepalCover-2010产品的精度验证工作,并进一步分析了该国土地覆被类型的数量结构特征和空间格局特征与地形、气象要素间的关系。研究结果表明:① NepalCover-2010产品一级类总体分类精度达94.83%,Kappa系数为0.94;二级类的总体分类精度达87.17%,Kappa系数为0.85,能够准确地反映尼泊尔土地覆被类型的空间分布格局,是同类土地覆被产品中精度最好的。② 林地是尼泊尔最主要的土地覆被类型,其面积约占尼泊尔国土面积的41%。耕地次之,其面积占比约为25%,其中水田和旱地的面积分布比例约为2:3。③ 地形和气象要素对尼泊尔的土地覆被类型空间分布格局具有显著影响,自南向北随着地势的抬升,各土地覆被类型出现顺序表现出水田、常绿阔叶林、旱地、常绿阔叶灌木林、常绿针叶林、草原、稀疏植被和冰川/永久积雪的垂直地带性特征。

关键词: 一带一路, 尼泊尔, 土地覆被, 空间格局, 遥感

Abstract:

Nepal, located in the core region of the Hindu-Kush-Himalayan region, has complicated and diverse land cover classes. Meanwhile, its topography shows a typical transition pattern from plain to mountain, and then to plateau. Land cover mapping of Nepal not only has a great scientific and practical significance to its land resource management and eco-environmental protection, but also contributes greatly to the basic data collection in this area, which will support China’s international regional economic cooperation strategy of “the Belt and Road Initiative”. In this paper, the land cover product of Nepal in 2010 (hereinafter referred to as the NepalCover-2010) was produced from Landsat TM image data sources, based on the combined cartographic method of object-oriented methods and decision trees. The product contains 8 primary categories and 32 secondary categories. Besides, the accuracy validation of the NepalCover-2010 was performed using samples obtained from high-resolution Google Earth imagery. The quantitative structural characteristics of Nepal and the relationships between the spatial distribution of typical land cover classes and the topographical and meteorological elements were analyzed. The results demonstrated that: (1) the overall accuracies of the primary and secondary categories for the NepalCover-2010 are 94.83% and 87.17% respectively. The Kappa coefficients are 0.94 and 0.85 respectively. The NepalCover-2010 can reflect the spatial distribution characteristics of land covers in Nepal. The comparison between the NepalCover-2010 with similar land cover products indicated that the NepalCover-2010 has the best accuracy. (2) Forest is the major land cover class in Nepal, and it accounts for 41% of the total land area of Nepal. Cultivated land accounts for 25% and the area ratio of paddy field to dry land is about 2:3. (3) The topographical and meteorological elements impact greatly on the spatial distribution of land cover classes. The spatial distribution of each land cover class possesses the characteristics of vertical zonality, and the appearance pattern of the typical land cover classes from the south to the north is paddy fields, evergreen broadleaf forests, dry lands, evergreen broadleaf shrubs, evergreen needleleaf forests, steppes, sparse vegetation, snow and ice, along with the increase in altitude.

Key words: the Belt and Road Initiative, Nepal, land cover, spatial pattern, remote sensing