光谱学与光谱分析, 2017, 37 (5): 1482, 网络出版: 2017-06-20  

土壤表层湿度影响下冬小麦晚霜冻害及冠层光谱检测

Using Hyperspectral Data for Detecting Late Frost Injury to Winter Wheat under Different Topsoil Moistures
作者单位
1 农业部农业环境重点实验室, 中国农业科学院农业环境与可持续发展研究所, 北京 100081
2 商丘市农林科学院小麦研究所, 河南 商丘 476000
摘要
为探究土壤表层湿度影响下冬小麦冠层光谱反射率响应晚霜冻害的特征, 并检验敏感光谱波段预测小麦产量变化的能力, 于2013和2014年小麦拔节期, 分别设置了表层土壤含水量为<10%(干)、 10%~20%(中)和 >20%(湿)的三个湿度处理的冻前试验, 并在低温室内进行降温处理。 分析了不同土壤表层湿度下受冻冬小麦的穗数、 穗粒数、 千粒重、 单株产量、 冠层光谱反射率及其一阶微分值的差异, 对冬小麦冻害产量变化率和高光谱特征参量进行了相关分析和一元线性拟合。 结果表明: 冬小麦穗粒数和单株产量总体上均随土壤表层湿度的降低而呈减少态势, 在土壤表层干处理条件下冻害对冬小麦产量造成的影响最为显著(p<0.05); 在绿峰(523 nm附近)、 黄边(571 nm附近)、 红边(732 nm附近)和近红外台的两个水分吸收带(952和1 145 nm附近), 干+冻害处理冬小麦冠层反射率的一阶微分值与中+冻害、 湿+冻害处理的差值明显; 剥离了土壤表层湿度对光谱的影响后, 冬小麦冠层反射率一阶微分差值在以570 nm为中心的黄边区域和以710 nm为中心的红边区域对干、 中和湿梯度处理下晚霜冻害响应的差异明显; 两年试验中的黄边面积(SDy)和570 nm处一阶微分值(d570)均与冻害产量变化率达到显著正相关(p<0.05), 说明黄边区域的高光谱特征参量可用于检测土壤表层湿度影响下的冬小麦晚霜冻害程度。 本研究可为土壤表层湿度和晚霜冻害叠加影响下冬小麦产量变化预测方法的探讨提供参考。
Abstract
To explore the characteristics of hyperspectral reflectance responding to late frost damage in winter wheat canopy under different topsoil moisture levels and test abilities of sensitive wavelengths to predict extents of yield change, respectively, pot experiments with three levels of topsoil moisture of Dry (<10%), Moderate (10%~20%) and Wet (>20%) were conducted at the jointing stage of winter wheat in the year of 2013 and 2014. After topsoil moisture treatments, frosting experiments were carried out in a specified cold climate chamber. This study analyzed differences of the ear number per plant, kernel number per ear, 1000-kernel weight, yield per plant and canopy spectral reflectance and corresponding first derivative reflectance of the frosted winter wheat canopy under different topsoil moisture levels. Correlativity and linear fitting were made for hyperspectral parameters and yield change rate of the frosted winter wheat. The results indicate that (1) kernel number per ear and yield per plant generally have a reducing trend with the decrease of topsoil moistures. Late frost injuries showed the most significant (p<0.05) influences on winter wheat yield under conditions of the treatment Dry. (2) In green peak region (near 523 nm), yellow edge region (near 571 nm), red edge region (near 732 nm) and two water absorption bands (around 952 nm and 1 145 nm) of the near-infrared region, the first derivative reflectance of winter wheat canopy was more different under conditions of the treatment Dry + Frost than the treatment Moderate + Frost and Wet + Frost. (3) After removing the influences of different topsoil moistures on spectrum reflectance, difference curves of the first derivative reflectance responding to late frost damages under the treatments Dry, Moderate and Wet were quite different mainly in the yellow edge region (centered at 570 nm) and the red edge region (centered at 710 nm). (4) In the two years experiments, yellow edge area (SDy) and first derivative value at 570 nm (d570) were significantly (p<0.05) and positively correlated with yield change rate (YCR) of the frosted winter wheat, respectively, which shows that hyperspectral parameters in the yellow edge region can be used to detect the differences of late frost damage due to the different topsoil moistures impacts. This study will provide a basis for prediction of the variations in winter wheat yield under the stacking effects of different topsoil moistures and late frost damages using hyperspectral reflectance parameters .

孟雷, 武永峰, 胡新, 吕国华, 任德超, 宋吉青. 土壤表层湿度影响下冬小麦晚霜冻害及冠层光谱检测[J]. 光谱学与光谱分析, 2017, 37(5): 1482. MENG Lei, WU Yong-feng, HU Xin, L Guo-hua, REN De-chao, SONG Ji-qing. Using Hyperspectral Data for Detecting Late Frost Injury to Winter Wheat under Different Topsoil Moistures[J]. Spectroscopy and Spectral Analysis, 2017, 37(5): 1482.

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!