›› 2013, Vol. 21 ›› Issue (6): 1059-1064.DOI: 10.11733/j.issn.1007-0435.2013.06.004

Previous Articles     Next Articles

Application of Hyperspectral Remote Sensing in the Biomass Monitor of Two Grassland Types in Xilinhot

ZHAO Feng-jie1,2, WU Hui-hui1,2, LIU Zhao-yang1,2, QIN Xing-hu1,2, WANG Guang-jun1,2, ZHANG Ze-hua1,2   

  1. 1. Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Key Laboratory for Biological Control of Chinese Academy of Agricultural Sciences (CAAS), Beijing 100081, China;
    2. Station of Pest Investigation in Rangeland, Ministry of Agriculture, Institute of Plant Protection, Chinese Academic of Agricultural Sciences (CAAS), Xilinhot, Inner Mongolia 026000, China
  • Received:2013-04-16 Revised:2013-05-17 Online:2013-12-15 Published:2013-11-27

高光谱遥感在锡林浩特2种草地类型生产力监测中的应用

赵凤杰1,2, 吴惠惠1,2, 刘朝阳1,2, 秦兴虎1,2, 王广君1,2, 张泽华1,2   

  1. 1. 中国农业科学院植物保护研究所 中国农业科学院生物防治重点开放实验室, 北京 100081;
    2. 农业部锡林郭勒草原有害生物科学观测实验站, 内蒙古 锡林浩特 026000
  • 通讯作者: 王广君
  • 作者简介:赵凤杰(1987-),女,山东东营人,硕士研究生,研究方向为草地遥感,E-mail:grzhaofengjie@163.com
  • 基金资助:
    公益性行业(农业)科研专项经费(201003079);现代农业产业技术体系建设专项资金(CARS-35-07)资助

Abstract: To establish the biomass model of two grassland types in Xilinhot, the reflectance spectra and biomass were measured by spectrometer in both grazing district and ungrazing district. Two grassland types were Leymus chinensis and forb grassland type and Stipa krylovii with Caragana microphylia grassland type. Reflectance spectra and biomass of tested grassland were recorded, and the relationships were analyzed by simple regression. Results indicated that the models of biomass and reflectance spectra were different between tested grasslands. The biomass models of L. chinensis grassland type and S. krylovii with C. microphylia grassland type in grazing district were y=326.81x-19.994, R=0.8612 and y=209.18x+11.435, R=0.9442; while the biomass models of L. chinensis grassland type and S. krylovii with C. microphylia grassland type in ungrazing district were y=614.15x-119.28; R=0.9992 and y=602.32x-148.08, R=0.9356, respectively. The x represented vegetation indices NDVI, and y represented vegetation biomass. Therefore, the correlation models of reflectance spectra and biomass in different grassland types were established, and both precision and accuracy of these models were improved remarkably by prolonging the duration of investigation and hyperspectral remote sensing in this experiment.

Key words: Hyperspectral remote sensing, Biomass model, Leymus chinensis and forb grassland type, Stipa krylovii with Caragana microphylia grassland type

摘要: 为构建锡林浩特2种草地类型的反射光谱特征值与生物量关系模型,使用地物波谱仪对放牧区、禁牧区的羊草(Leymus chinensis)+杂类草型草地和具锦鸡儿(Caragana microphylia)的克氏针茅(Stipa krylovii)型草地进行植被反射光谱与生物量关系的研究,使用简单回归对二者间的关系进行模拟。结果表明:不同草地类型的反射光谱特征值与生物量反演模型不同:放牧区羊草+杂类草型是y=326.81x-19.994,R=0.8612;放牧区具锦鸡儿的克氏针茅型是y=209.18x+11.435,R=0.9442;禁牧区羊草+杂类草型是y=614.15x-119.28,R=0.9992;禁牧区具锦鸡儿的克氏针茅型是y=602.32x-148.08,R=0.9356。其中x为归一化植被指数NDVI,y为草地生物量。通过研究高光谱遥感反射光谱和生物量之间的关系,并延长调查持续时间,建立了二者之间的相关模型,使模型在精度和准确性方面得到明显提高。

关键词: 高光谱遥感, 生物量模型, 羊草+杂类草型草地, 具锦鸡儿的针茅型草地

CLC Number: