img

官方微信

高级检索

中国沙漠 ›› 2007, Vol. 27 ›› Issue (3): 412-418.

• 生物土壤与生态 • 上一篇    下一篇

基于ETM+影像的草地退化评价模型研究——以西藏自治区那曲县为例

李辉霞1, 刘淑珍2   

  1. 1.佛山科学技术学院资源环境系, 广东 佛山 528000; 2.中国科学院成都山地灾害与环境研究所, 四川 成都 610041
  • 收稿日期:2006-06-19 修回日期:2006-08-08 出版日期:2007-05-20 发布日期:2007-05-20

Research on Grassland Degradation Assessment Model based on ETM+Image—A Case Study in Naqu County of Tibet

LI Hui-xia1, LIU Shu-zhen2   

  1. 1.Department of Resource and Environment, Foshan University, Foshan 528000, China; 2.Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, Chengdu 610041, China
  • Received:2006-06-19 Revised:2006-08-08 Online:2007-05-20 Published:2007-05-20

摘要: 在实地调查和样方测定基础上,选用建群种植株高度、草地植被盖度和草地生物量作为草地退化地面评价体系中的单一评价指标,并通过加权综合三个单一评价指标,构建草地退化地面综合评价指标;在分析退化草地光谱特征的基础上,从ETM+影像的波段数据中派生出草地退化的遥感评价指标。通过相关分析的方法,选出最能反映草地退化趋势、最适于用于线性拟合的地面评价指标(地面综合评价指标)和遥感评价指标(TM4/TM5),采用线性回归技术构建草地退化遥感评价模型,并通过计算确定系数(R2)、均方根差(RMSE)和相对误差对模型精度做出评价。确定主要草地类型的不同退化等级标准,完成草地退化图的编制。通过分析草地退化地面评价指标与遥感评价指标之间的关系,探讨了西藏北部草地退化的遥感评价模型,为科学、快速评价草地退化提供一种新思路。

关键词: ETM+影像, 草地退化, 遥感模型

Abstract: First, the height, coverage and biomass is respectively selected as the single index of ground assessment of grassland degradation based on field survey and systematic random samples, and the ground integrated assessment index is calculated by integrating the height, coverage and biomass. Meanwhile, the remote sensing indices are derived from ETM+ image based on the spectral value of degraded grassland. Second, according to the results of correlation analysis between ground assessment indices and RS assessment indices, the ground integrated assessment index and the spectral index of TM4/TM5 are chosen as the best variables to build the grassland degradation model with linear regression. And the decision coefficient (R2), error of mean square root (RMSE) and relative error are calculated to test the precision of the assessment model. Last, the standards of different degraded level of main grassland types are defined and the grassland degradation map is drawn. In this paper, the grassland degradation model of north Tibet is studied based on the relation between ground assessment indices and remote sensing assessment indices, which provides a new idea for scientific and quick assessment of grassland degradation.

Key words: ETM+image, grassland degradation, remote sensing model

中图分类号: