摘要
我国草地面积广阔,但目前正面临着严重的退化问题,制约着我国畜牧业的持续发展,因此急需一种有效的草地退化检测的新手段。文中以坝上草原为研究对象,利用地面实测高光谱数据,运用包络线去除法和一阶微分法对坝上草原主要优势种(羊草、针茅、脚苔草、糙隐子草)和退化指示种(冷蒿、狼毒、星毛委陵菜)的光谱曲线进行了处理并提取了多种特征参数,找到了退化指示种识别的有效方法。在此基础上,运用线性混合光谱分解模型计算了退化指示种的覆盖度,经验证在误差范围内,证明线性混合分解模型在研究区内有效,可为大面积的草原退化检测提供科学依据。
The grasslands distribute broadly in China,but the serious degradation of grasslands is restricting the sustainable development of animal production. Therefore it is urgently needed an effective new method to detect the grassland degradation. Bashang region was chosen as a study object in the present article. We used the SOC710 protable spectrometer,and methods of continuum removing and first- order differential to process the spectral curve of grassland vegetation and to extract the spectral characteristics of dominant species and indicator species for degradation in Bashang region. As a result,the research found an effective method to identify indicator species for degradation. On this basis,the spectra of mixed grass community were unmixed by linear mixing models,and the proportion of all the endmembers was counted. The results were well within the margin of error.This article provided a scientific basis for the detection of large area of grassland degradation.
出处
《干旱区资源与环境》
CSSCI
CSCD
北大核心
2016年第3期133-139,共7页
Journal of Arid Land Resources and Environment
基金
河北省自然科学基金(D2013205077
D2014205070)
河北省高校重点学科建设项目资助
关键词
坝上草原
退化指示种
光谱特征
草种识别
Bashang region
indicator species for degradation
characteristics of spectra
grass species recognition
