高CO_(2)浓度下冬小麦的高光谱特征及其与叶面积指数和SPAD值的反演  

Hyperspectral characteristics and leaf area index(LAI)and SPAD value inversion of winter wheat under elevated CO_(2) concentration

作　　者：蔡瑶 缪宇轩 吴浩 王丹[1] CAI Yao;MIAO Yuxuan;WU Hao;WANG Dan(School of Applied Meteorology, Nanjing University of Information Science & Technology, Nanjing 210044, China)

机构地区：[1]南京信息工程大学应用气象学院,江苏南京210044

出　　处：《浙江农业学报》2022年第3期582-589,共8页Acta Agriculturae Zhejiangensis

基　　金：国家自然科学基金(31500503,31770485);江苏省自然科学基金(BK20150894)。

摘　　要：以冬小麦为研究对象,利用开顶式气室试验,开展以环境CO_(2)浓度为对照(CK)和比CK处理的CO_(2)浓度高200μmol·mol^(-1)(T)处理的试验,测定冬小麦主要生育期冠层光谱反射率、叶面积指数(LAI)和SPAD值,分析LAI、SPAD值与原始光谱反射率、光谱特征参数的相关性,并探究最优回归反演模型。结果表明,高CO_(2)浓度处理提高冬小麦孕穗-抽穗期和灌浆期的LAI和SPAD值,影响光谱反射率大小,但不改变光谱反射率曲线波形。CO_(2)浓度升高导致红边位置先红移再蓝移。用红黄边面积比(x)估算小麦LAI(y)的回归模型最优,回归方程为y=3.96×10^(-3)x^(2)-8.60×10^(-2)x+1.93,在验证集上的决定系数和均方根误差分别为0.66和0.42。用红边振幅(x)估算小麦SPAD值(y)的回归模型最优,回归方程为y=-5.151×10^(4)x^(2)+2.883×10^(3)x+33.83,在验证集上的决定系数和均方根误差分别为0.63和4.61。In the present study,winter wheat was selected as the research object,and an open-top chamber experiment was carried out with ambient CO_(2)as control(CK)and plus 200μmol·mol^(-1)CO_(2)on the basis of CK as treatment(T).The canopy spectral reflectance,leaf area index(LAI)and SPAD value of winter wheat at main growth stages were measured.The correlation within LAI,SPAD value,original spectral reflectance and spectral characteristic parameters were analyzed,and the optimal estimation models were established.The results showed that the elevated CO_(2) concentration increased LAI and SPAD value of winter wheat at booting-heading stage and filling stage.The elevated CO_(2) concentration did not change the spectral reflectance curve,but changed the reflectance.The elevated CO_(2)concentration induced the red edge position red shift first and then blue shift.Red and yellow edge area ratio(x)was the most suitable spectral parameter to estimate LAI(y)of wheat,and the regression equation was y=3.96×10^(-3)x^(2)-8.60×10^(-2)x+1.93,of which the coefficient of determination(R^(2))and the root mean square error(RMSE)on validation set were 0.66 and 0.42,respectively.Red edge amplitude(x)was the most suitable spectral parameter to estimate SPAD value(y)of wheat,and the regression equation was y=-5.151×10^(4)x^(2)+2.883×10^(3)x+33.83,of which the R^(2) and RMSE on validation set were 0.63 and 4.61,respectively.

关 键 词：高二氧化碳 高光谱 叶绿素含量 叶面积指数 冬小麦 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置] S512.11[自动化与计算机技术—控制科学与工程]