草地学报

草地学报 ›› 2020, Vol. 28 ›› Issue (3): 675-683.DOI: 10.11733/j.issn.1007-0435.2020.03.011

• 研究论文 • 上一篇    下一篇

草坪草叶绿素高光谱估测模型的建立及Pb2+胁迫下的反演研究

纪童1,2, 王波3, 王占军4, 杨军银1,2, 李强1,2, 刘志刚1,2, 关文昊1,2, 何国兴1,2, 潘冬荣5, 柳小妮1,2   

  1. 1. 甘肃农业大学草业学院, 甘肃 兰州 730070;
    2. 草业生态系统教育部重点实验室(甘肃农业大学), 甘肃 兰州 730070;
    3. 盐池县草原实验站, 宁夏 盐池 751506;
    4. 宁夏农科院荒漠化治理研究所, 宁夏 银川 750002;
    5. 甘肃省草原技术推广总站, 甘肃 兰州 730000
  • 收稿日期:2019-11-04 修回日期:2020-02-11 出版日期:2020-06-15 发布日期:2020-05-30
  • 通讯作者: 柳小妮
  • 作者简介:纪童(1995-),男,河北唐山人,硕士研究生,主要从事草原学研究,E-mail:986462673@qq.com;
  • 基金资助:
    国家自然科学基金（31160475）；甘肃省知识产权计划“18ZC1LA010”；甘肃省农业财政项目（201647）共同资助

Hyperspectral-based Estimation on the Chlorophyll Content of Turfgrass

JI Tong1,2, WANG Bo3, WANG Zhan-jun4, YANG Jun-ying1,2, LI Qiang1,2, LIU Zhi-gang1,2, GUAN Wen-hao1,2, HE Guo-xing1,2, PAN Dong-rong5, LIU Xiao-ni1,2   

  1. 1. College of Pratacultural Science, Gansu Agricultural University, Lanzhou, Gansu Province 730070, China;
    2. Key Laboratory of Grassland Ecosystem, Ministry of Education/Pratacultural Engineering Laboratory of Gansu Province, Lanzhou, Gansu Province 730070, China;
    3. Grassland Experiment Station of Yanchi, Yanchi, Ningxia 751506, China;
    4. Institute of desertification control, Ningxia academy of agricultural sciences, Yinchuan, Ningxia 750002, China;
    5. Grassland Technique Extension Station of Gansu Province, Lanzhou, Gansu Province 730070, China
  • Received:2019-11-04 Revised:2020-02-11 Online:2020-06-15 Published:2020-05-30

PDF

摘要： 本研究以3种常用草坪草种为材料，通过盆栽试验测定了生长旺盛期草坪草冠层成像光谱数据和叶绿素相对含量（Soil and plant analyzer develotrnent，SPAD），通过Person相关系数分析了1/SPAD与19个植被指数的相关性，筛选与叶绿素相关性较高的植被指数进行主成分分析，建立了1/SPAD估测模型，并利用所建模型反演了Pb2+胁迫下3种草坪草SPAD的变化。结果表明：植被指数NDVI670、VARI、PSRI、ARVI、RGI和GI与1/SPAD极显著相关，可作为主成分分析原始变量；主成分分析得到的2个主成分，可清楚的区分"红象"高羊茅与"肯塔基"早熟禾；多元逐步回归模型y1/SPAD=-0.117Z1+0.062Z2+0.041（R2=0.763，RMSE=0.01），总体估测精度为0.9248，说明利用主成分分析进行草坪草叶绿素的估测效果较佳；最优模型反演发现，除低浓度Pb2+（500 mg·L-1）胁迫显著促进了"红象"高羊茅叶绿素的合成外，3种草坪草的SPAD均随Pb2+浓度增加极显著下降。

关键词: 高光谱反演, 草坪草, 叶绿素, 植被指数, 模型, Pb胁迫

Abstract: Using three common turfgrass species as materials,through pot experiments,the turfgrass canopy imaging spectrum data and relative chlorophyll SPAD (Soil and plant analyzer develotrnent) were measured during the vigorous growth period,and 1/SPAD and 19 were analyzed by the Person correlation coefficient. Correlation of vegetation index,vegetative index with high correlation with chlorophyll was selected for principal component analysis,1/SPAD estimation model was established,and the model was used to invert the changes of SPAD of three turfgrasses under Pb2+stress. The results showed that:the vegetation indices NDVI670,VARI,PSRI,ARVI,RGI,and GI were extremely significantly related to 1/SPAD and can be used as the primary variables for the principal component analysis;The two principal components obtained by the principal component analysis can be clearly distinguished "Red Elephant" tall fescue and Kentucky Kentucky bluegrass;Multiple stepwise regression model y1/SPAD=-0.117Z1+0.062Z2+0.041 (R2=0.763,RMSE=0.01),the overall estimation accuracy is 0.9256,indicating that the principal component analysis is a better method to estimate the chlorophyll of turfgrass;The optimal model inversion found that the addition of low concentrations of Pb2 + (500 mg·L-1) stress significantly promoted the chlorophyll synthesis of "Red Elephant" In addition,the SPAD of all three turfgrasses decreased significantly with the increase of Pb2+ concentration.

Key words: The inversion hyperspectral, Turfgrass, Chlorophyll, Vegetation indices, Model, The stress of Pb

中图分类号: