地理研究 ›› 2013, Vol. 32 ›› Issue (5): 965-973.doi: 10.11821/yj2013050018

• 信息技术 • 上一篇    下一篇

基于分层策略和模糊推理的北方石质山区土壤厚度制图——以滦平县虎什哈流域为例

李润奎1, 彭明1, 河野泰之2, 张洪勋1, 宋现锋1,2   

  1. 1. 中国科学院大学, 北京100049;
    2. 日本京都大学东南亚研究所, 日本京都606-8501
  • 收稿日期:2012-04-10 修回日期:2013-01-20 出版日期:2013-05-10 发布日期:2013-05-10
  • 通讯作者: 宋现锋(1969-),男,博士,教授,研究方向为GIS与环境应用。E-mail:song.osgeo@gmail.com E-mail:song.osgeo@gmail.com
  • 作者简介:李润奎(1983-),男,博士,讲师,主要从事GIS与环境模拟研究。E-mail:lirunkui@gmail.com
  • 基金资助:

    日本京都大学“2011 CSEAS Joint Research and Publication Program”项目;国家自然科学基金项目(41201038,41023010和40901191);国家基础研究计划“973”项目(2010CB428801和2010CB428804);中国博士后科学基金项目(2012M510588)

Soil depth mapping in lithoidal mountainous area using stratified strategy and fuzzy logic:A case study in Hushiha watershed,North China

LI Runkui1, PENG Ming1, Yasuyuki Kono2, ZHANG Hongxun1, SONG Xianfeng1,2   

  1. 1. University of Chinese Academy of Sciences, Beijing 100049, China;
    2. Center for Southeast Asian Studies, Kyoto University, Kyoto 606-8501, Japan
  • Received:2012-04-10 Revised:2013-01-20 Online:2013-05-10 Published:2013-05-10

摘要: 土壤厚度信息对土壤碳储量估算、水文-生态过程模拟等有着重要影响,但我国北方石质山区缺乏现势性好、分辨率高的土壤厚度分布数据及其快速获取方法。基于北方石质山区土壤厚度的分布特点,先将地貌信息和植被类型叠加生成制图单元,再根据每类单元的特征,针对性地选取地形参数和地表覆被参数作为指示土壤厚度空间分布的环境变量,结合少量土壤样点,依据模糊推理模型,构建了基于分层制图策略的土壤厚度空间推测方法。以河北省滦平县虎什哈流域为例,利用从ASTER GDEMs提取的地形变量和从我国环境减灾卫星影像(HJ-A)提取的地表覆被信息对土壤厚度进行了推测。结果表明,该方法精度较高,是利用免费地形和遥感数据进行低成本快速土壤制图的实用途径。

关键词: 土壤厚度, 潮河流域, 模糊推理, 环境卫星, ASTER GDEMs

Abstract: Spatial variation of soil depth has significant impacts on soil carbon storage estimation and eco-hydrological modeling.However,we are lack of detailed information of soil depth and its mapping approaches in the lithoidal mountainous areas of North China. This paper presents a fuzzy soil inference model for soil depth prediction in such a mountainous basin based on stratified strategy,using topographic attributes from terrain analysis,land cover attributes derived from remote sensing images,and a small number of sample points collected in field work.Two categorical environmental variables,landform and land cover,are firstly used for map intersection to produce basic soil mapping units.Other terrain and land cover environmental variables were selected separately to predict spatial distribution of soil depth within each unit with the fuzzy soil inference model.A case study in Hushiha of North China showed that the proposed method improved spatial detail of soil depth compared with the traditional soil map of Heibei Province.The proposed method for mapping soil depth takes full advantages of free ASTER GDEMs and remote sensing images from HJ-A of China,which makes it possible to map soil depth quickly using GIS and geo-data under fuzzy logic with less labour consuming works.

Key words: soil depth, Chaohe watershed, fuzzy logic, HJ-A images, ASTER GDEMs