无人机多光谱遥感反演花蕾期棉花光合参数研究

Investigation on Photosynthetic Parameters of Cotton during Budding Period by Multi-spectral Remote Sensing of Unmanned Aerial Vehicle

光合作用对作物的生长发育、干物质的积累以及产量的形成起着至关重要的作用。为探讨遥感技术反演作物冠层光合参数的可行性,以无人机作为遥感平台,搭载6波段多光谱相机,通过采集棉花花蕾期不同时刻(09:00、11:00、13:00、15:00、17:00)冠层多光谱遥感图像,提取其冠层光谱反射率信息,并同步测定棉花冠层叶片的净光合速率(Pn)、蒸腾速率(Tr)、气孔导度(Gs)和胞间二氧化碳浓度(Ci)等光合参数。通过对4种光合参数和6波段光谱反射率进行相关性分析,并分别使用一元线性回归和主成分回归、岭回归、偏最小二乘回归等多元回归方法,建立不同光合参数在不同时刻的反演模型。结果表明:净光合速率(Pn)、蒸腾速率(Tr)、气孔导度(Gs)和胞间二氧化碳浓度(Ci)的最优反演模型分别为13:00的基于蓝光波段反射率的一元线性模型,15:00的基于红光波段反射率的一元线性模型,15:00的岭回归模型和15:00的基于红光波段反射率的一元线性模型,模型的决定系数R2均在0.5以上,验证相对误差RE均小于9%。该研究可为大范围监测作物的光合作用提供一定的参考。

Photosynthesis plays a vital role in crop growth,dry mater accumulation and yield formation.How to monitor it quickly and widely is still a problem so far.Taking the unmanned aerial vehicle(UAV)as the remote sensing platform,and a multispectral camera with six bands was mounted.To explore the feasibility of retrieving crop canopy photosynthetic parameters by using remote sensing technology,the cotton in budding period were studied.The camera was used to capture the image of cotton canopy at different times in one day(09:00,11:00,13:00,15:00 and 17:00),of which the reflectance information was extracted.The parameters of cotton photosynthetic(net photosynthetic rate(P n),stomatal conductance(G s),intercellular carbon dioxide concentration(C i)and transpiration rate(T r))were measured at the moment when the UAV was landed.Through the correlation analysis of the four photosynthetic parameters and the six-band reflectance,the retrieving model of different photosynthetic parameters at different times was established by univariate linear regression,principal component regression(PCR),ridge regression(RR)and partial least-squares regression(PLSR),respectively.The results showed that the best retrieving models of net photosynthetic rate(P n),transpiration rate(T r),stomatal conductance(G s)and intercellular carbon dioxide concentration(C i)were the univariate linear model based on the reflectance of the blue light band at 13:00,the univariate linear model based on the reflectance of the red light band at 15:00,the ridge regression model at 15:00 and the univariate linear model based on the red light band at 15:00,respectively.The decision coefficients(R 2)of the models were more than 0.5,and the relative errors(RE)were less than 9%.The research result can provide a certain reference for monitoring the photosynthesis of crops in a large scale.