光谱学与光谱分析, 2019, 39 (7): 2264, 网络出版: 2019-07-23  

倒伏胁迫对水稻可视茎叶穗比率的影响及光谱响应解析

Analysis of Effect and Spectral Response of Lodging Stress on the Ratio of Visible Stem, Leaf and Panicle in Rice
作者单位
1 湖北大学资源环境学院, 湖北 武汉 430062
2 农业部农业遥感机理与定量遥感重点实验室, 北京农业信息技术研究中心, 北京 100097
摘要
倒伏胁迫下作物的冠层光谱响应机理解析, 是大范围作物倒伏灾情遥感监测的重要基础。 倒伏胁迫直接改变了遥感光谱探测视场内的可视茎叶穗比率, 通过解析冠层光谱与可视茎叶穗比率间的关系, 探索不同强度的倒伏胁迫下水稻可视茎叶穗组分变化规律及其与冠层光谱响应规律, 为大范围作物倒伏灾情遥感监测提供理论支持。 以2017年江苏省兴化市、 大丰区的实发倒伏水稻为研究对象, 在野外观测实验的支持下, 分析不同倒伏强度的倒伏水稻冠层光谱变化规律, 并对不同倒伏强度下的冠层可视茎叶穗比率与倒伏角度进行相关性分析, 筛选能有效表征倒伏强度的敏感农学参数, 采用灰色关联分析法构建倒伏水稻冠层光谱指标与敏感农学参数之间的响应模型, 实现水稻倒伏灾情的光谱诊断, 并利用野外实测样本评价诊断精度。 研究结果表明, 随着倒伏强度的加大, 冠层光谱表现出规律性变化, 红光波段与近红外波段响应较为明显, “红边”位置明显“蓝移”, 且“红边”振幅与“红边”面积增大, 说明红光波段和近红外波段对水稻倒伏胁迫强度较为敏感; 冠层可视叶茎比存在随倒伏强度增加而减少的规律, 其相关性可达0.715, 说明倒伏后的水稻冠层可视叶茎比对于倒伏强度有着较好的表征能力; 通过对可视叶茎比与冠层高光谱反射率进行相关性分析, 分别于红光波段和近红外波段内筛选出698与1 132 nm作为敏感波段, 进而计算特征植被指数; 利用灰色关联分析构建了基于特征植被指数的水稻可视叶茎比光谱响应模型, 检验样本的决定系数为0.635, 以可视叶茎比预测结果进行倒伏灾情等级划分的精度达到82%。 因此, 倒伏发生后水稻冠层的茎、 叶、 穗等组分在光谱探测器视场中的贡献比例发生了规律性改变, 茎、 叶、 穗本身光谱反射率差异和视场内比率差异直接反映于倒伏水稻冠层光谱差异, 其中可视叶茎比能有效表征受倒伏胁迫的水稻群体结构变化, 与倒伏强度具有较好的响应关系, 不同倒伏强度的可视叶茎比与水稻冠层光谱之间的响应规律可以有效区分倒伏灾情等级, 有助于为区域尺度的水稻倒伏灾情遥感监测提供先验知识。
Abstract
The analysis of canopy spectral response mechanism of crop lodging stress is an important basis for remote sensing monitoring of large-scale crop lodging disasters. Lodging stress directly change the ratio of visual stem, leaf and panicle in remote sensing spectrum detection field of view. By analyzing the relationship between canopy spectra and the ratio of visual stem, leaf and panicle, this paper explores the change regulation of visual stem, leaf and panicle components and spectral response of rice canopy under different intensities of lodging stress, and provides theoretical support for remote sensing monitoring of large-scale crop lodging disaster. Taking the real lodging rice in Xinghua City and Dafeng District of Jiangsu Province in 2017 as the research object, with the support of field observation experiment , the rule of canopy spectral variation of lodging rice with different lodging intensities was analyzed, and the correlation between the ratio of canopy visual stem, leaf and panicle and lodging angle under different lodging intensity was analyzed, and parameters of sensitive agronomy that can effectively represent the lodging intensity was screened. A response model between rice canopy spectral indices and sensitive agronomic parameters was constructed by grey relational analysis to realize the spectrum diagnosis of rice lodging disaster, and field-measured samples were used to evaluate the diagnostic accuracy. The results showed that with the increase of lodging strength, the canopy spectra showed regular changes, red-band and near-infrared band response was more obvious, “Red edge” position is obviously “blue shift”, and “red edge” amplitude and “red edge” area increase, it shows that the red-band and near-infrared band on rice lodging stress intensity is more sensitive. The correlation of the canopy visual leaf-stalk ratio and lodging strength decreased with the increase of lodging strength, which was more than 0.715, indicating that the visible leaf stem ratio of canopy was better in characterizing the lodging strength. Through correlation analysis between visual leaf-stem ratio and hyperspectral reflectance, 698 and 1 132 nm in the red and near-infrared bands were respectively selected as the sensitive bands, and then the characteristic vegetation index was calculated. The spectral response model of rice visual leaf-stem ratio based on characteristic vegetation index was constructed by using grey correlation analysis, and the determining factor for the test sample was 0.635, and the precision of the classification of the disaster level with the visual leaf-stem ratio inversion result reached 82%. Therefore, the contribution proportion of stem, leaf and panicle in the canopy of rice in the field of spectral detectors was changed regularly after lodging. The difference of spectral reflectance and the ratio of apparent field in the Miho of stem, leaf and panicle is directly reflected in the spectral difference of lodging rice canopy. While visual leaf-stem ratio can effectively characterize the population structure change of rice under lodging stress, which has a good response relationship with the lodging intensity. The response law of visual leaf-stem ratio and rice canopy spectrum of different lodging intensity can effectively distinguish the lodging intensity, which will help provide a prior knowledge for remote sensing monitoring of rice lodging at the regional scale.

谢新锐, 顾晓鹤, 林丽群, 杨贵军, 张丽妍. 倒伏胁迫对水稻可视茎叶穗比率的影响及光谱响应解析[J]. 光谱学与光谱分析, 2019, 39(7): 2264. XIE Xin-rui, GU Xiao-he, LIN Li-qun, YANG Gui-jun, ZHANG Li-yan. Analysis of Effect and Spectral Response of Lodging Stress on the Ratio of Visible Stem, Leaf and Panicle in Rice[J]. Spectroscopy and Spectral Analysis, 2019, 39(7): 2264.

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!