净套作下大豆叶绿素密度动态及光谱估测研究  

Estimating chlorophyll density of soybean under monoculture and relay intercropping by hyperspectral remote sensing

作　　者：黄山[1] 杨峰[1] 张勇 王锐[1] 邓俊才[1] 雍太文[1] 刘卫国[1] 杨文钰[1] 

机构地区：[1]四川农业大学农学院,农业部西南作物生理生态与耕作重点实验室,四川成都611130

出　　处：《西北农林科技大学学报（自然科学版）》2016年第5期29-37,共9页Journal of Northwest A&F University(Natural Science Edition)

基　　金：国家自然科学基金项目(31571615);农业部公益性行业科研专项(201203096)

摘　　要：【目的】利用高光谱数据构建一种适合于净作和套作群体的大豆叶绿素密度光谱估测通用模型,为监测以叶绿素密度为参数的净、套作大豆长势提供支撑。【方法】以南豆12为供试大豆品种,设计净作和套作2种种植模式,测定净、套作下大豆不同生育时期叶绿素密度及其冠层光谱特征,对通过原始、全波段导数光谱和多波段组合计算的植被指数与叶绿素密度进行相关性分析,比较不同高光谱特征参数与叶绿素密度之间的关系,借助高光谱遥感分析方法构建适合于净、套作下大豆叶绿素密度的估测模型。【结果】净作和套作大豆叶绿素密度在整个生育期呈先上升后下降的单峰曲线,在结荚期达到最大值,其中净作比套作平均高5.5%。同样,大豆冠层光谱反射率在近红外波段（700~1 000nm）也呈现先增后降趋势,结荚期最高,达到70%,并在结荚期之前红边位置出现红移现象。原始光谱自由组合的比值植被指数RVI（507/697）与叶绿素密度的相关性最高,相关系数大于0.962（P〈0.01）。对净、套作下大豆叶绿素密度模型进行比较分析,发现二次模型能够较好地估测净、套作下大豆叶绿素密度（R2〉0.75,RMSE=0.25）。【结论】用比值植被指数RVI（507/697）构建的叶绿素密度二次估测模型,能够较好地对净、套作下大豆叶绿素密度进行估测。【Objective】Hyperspectral data were used to build a chlorophyll density model,which can may provide basis for soybean growth monitoring under monoculture and intercropping systems.【Method】In this research,chlorophyll density and canopy spectral characteristics of soybean at different growth stages were analyzed under monoculture and intercropping system using Nandou 12 as species.The correlation between original spectrum,derivative spectrum,vegetation index,combination of more bands and chlorophyll density was also analyzed.The model for estimating chlorophyll density was built by comparing the relationship between hyperspectral feature parameters and chlorophyll density.【Result】The chlorophyll density in the whole growth period displayed an trend that increased first and then decreased under mono-culture and intercropping system.The maximum value of chlorophyll density appeared in the pod stage,and the monoculture system was 5.5% higher than relay intercropping system.Similar results were found in the canopy spectral reflectance of soybean in the near infrared region（700-1 000nm）with the highest reflectance spectral value of up to 70%in the pod stage.In addition,the position of red edge appeared a red shift with the increasing growth stage before the pod stage.The ratio vegetation index（507/697）had the highest correlation with chlorophyll density（r0.962,P〈0.01）compared with others.Compared to other models,the quadratic function model could estimate the canopy chlorophyll density better with the determination coefficient of0.75 and RMSEof 0.25.【Conclusion】The quadratic function model built with vegetation index RVI（507/697）could be used to estimate soybean chlorophyll density under monoculture and intercropping system.

关 键 词：叶绿素密度 光谱估测模型 冠层光谱 光谱反射率 光谱参数 

分 类 号：S127[农业科学—农业基础科学]