Study On Correlation Model Between Hyperspectral And Agronomic Parameters Of Summer Corn With Variable Rate Fertilization

In this paper,summer corn was used as the experimental object.Based on the field design and laboratory tests,the spectral changes of corn under differentnitrogen application rates and different growth stages were analyzed by using the frontier technology of hyperspectral remote sensing.The suitable spectral parameters and vegetation index of nitrogen level and growth index were determined.The correlation between physiological and biochemical parameters of summer corn and canopy spectrum was established.A hyperspectral prediction model of shoot dry biomass,plant nitrogen content,leaf chlorophyll content,leaf area index and yield of summer corn was established.It provides data and technical support for monitoring the growth of summer corn,guiding the field management of summer corn and promoting the development of precision agriculture.The results were as follows:?1?The spectral reflectivity of summer corn has a reflection peak in the range of550 nm,called"green peak",and an absorption valley in the range of 680 nm,called"red valley",and the reflectivity curve increases rapidly after 760 nm.After reaching the near infrared range,it tends to stabilize and form a high reflectivity platform.The spectral reflectance of leaves decreased gradually from bell-filling stage to filling stage.There was no significant difference in reflectivity in the band of?350-760nm?,but the reflectivity in the band of?760-1100nm?was higher under low nitrogen treatment.The tendency of low reflectivity under high nitrogen treatment.?2?From the small bell stage to the filling stage of summer corn,the total nitrogen content of the plant decreased gradually during the whole growth period.Leaf chlorophyll content,aboveground dry biomass and leaf area index all showed an increasing trend.At the same growth stage,the agronomic parameters increased with the increase of N application rate,and reached the maximum N₄-N₅ level at the N₃level and decreased at N₄-N₅ level.?3?Under the same nitrogen application rate at different growth stages,the total nitrogen content of plants decreased with the growth period,and the other three indexes showed an increasing trend.?4?On the basis of previous studies,two kinds of vegetation indices,NDVI and RVI,red valley and green peak were selected to establish chlorophyll content SPAD value,above ground dry biomass,plant total nitrogen content,leaf area index.The regression model of the measured output data shows a good correlation.The hyperspectral prediction model of maize yield combined with the model of yield and different nitrogen application rate can establish the relationship model between different nitrogen application amount and spectral data,which can effectively guide maize production.