Total Nitrogen And Phosphorus Content Estimation Models Of Summer Maize Leaves Using Hyperspectral Remote Sensing

Efficient and accurate diagnosis in plant nutrition is one of the hot topics research forplant nutrition researchers. Physical parameters and chemical parameters of the plant can beget from remote sensing data by the hyperspectral remote sensing technology, andbiochemical composition of leaves also can be estimated accurate, efficientv from thereflectance of crop spectral. The plants need mass nitrogen and phosphorus during grwoth anddevelopment which has significant impact on crop growth and yield quality, so reasonable andtimely fertilization management is important. This research take summer corns and a series oftwo years field experiments as the object in Guanzhong plain; hyperspectral remote sensingtechnology, biochemical tests and statistical to analysis the dynamic relationship between thenitrogen, phosphorus, chlorophyll and canopy hyperspectral reflectance which reflect thesummer corn treated by different nitrogen and phosphorus. The purpose of the research is tofind accurate, efficient and stable fitted model between hyperspectral reflectance and cornnutritional.1. Based on the treatments of five nitrogen fertilizer application amounts and twocultivars of summer maize, crop canopy spectral reflectance and total nitrogen content ofmaize leaves were measured at the jointing stage, huge bellbottom stage, tasseling stage,silking stage and milk stage. The canopy spectral reflectance in470,550,620and720nmwavelength of hyperspectral remote sensing were chosen to establish liner and nonlinearregression relationship between leaf total nitrogen content and canopy spectral parameters foreach cultivar, which includes original spectral reflectance, first order differential transform,and part of hyperspectral characteristic parameters. Three models with high coefficients and Fvalues of each cultivar at each growth stage were chosen to verify root mean square error andrelative error with the second year data of spectral reflectance and total nitrogen content oftwo cultivars separately. The smallest root mean square error and relative error models weretaken as the best models. The results show that: at the jointing stage, huge bellbottom stage, tasseling stage, silking stage and milk stage of maize,spectrum parameter for the best fittingregression relationship with leaf total nitrogen content was R720, DR720, SDb, DR550andDR550, and get best model to estimate total nitrogen content of maize leaf based on abovebest spectrum parameter of hyperspectral remote sensing in five growth stages isY=5.129e^-2.317x, Y=3.421-10.010x-477802.331x³, Y=4.070-2.304x-52.177x²,Y=-0.468-0.528lnx,Y=-2.390-0.793lnx.2. Based on the treatments of four phosphorus fertilizer application amounts and twocultivars of summer maize, crop canopy spectral reflectance and total phosphorus content ofmaize leaves were measured at the jointing stage, huge bellbottom stage, silking stage andmilk stage. The canopy spectral reflectance in540,720,740and850nm wavelength ofhyperspectral remote sensing were chosen to establish liner and nonlinear regressionrelationship between leaf total phosphorus content and canopy spectral parameters for eachcultivar, which includes original spectral reflectance, first order differential transform, andnormalization reflectance. After the selection, verification, the results show at the hugebellbottom stage, silking stage and milk stage of maize,spectrum parameter for the bestfitting regression relationship with leaf total nitrogen content was DR540, NR740and R850;and get best model to estimate total phosphorus content of maize leaf based on above bestspectrum parameter of hyperspectral remote sensing in three growth stages isY=15.469+1.844lnx,Y=-63.752+3071.931x-34285x²å'ŒY=-13.326+54.092x-43.123x².3. Based on the treatments nitrogen(five) and phosphorus(four) fertilizer applicationamounts and two cultivars of summer maize, crop canopy spectral reflectance and chlorophyllcontent of maize leaves were measured at the jointing stage, huge bellbottom stage, silkingstage and milk stage. The canopy spectral reflectance in540,560,600and760nmwavelength of hyperspectral remote sensing were chosen to establish liner and nonlinearregression relationship between leaf chlorophyll content and canopy spectral parameters forboth cultivar, which includes original spectral reflectance, first order differential transform,and vegetation indexs. After the selection, verification, the results show at the jointing stage,huge bellbottom stage, silking stage and milk stage of maize,spectrum parameter for the bestfitting regression relationship with leaf total nitrogen content was R760, PSDNa, GNDVI,Rch; and get best model to estimate total phosphorus content of maize leaf based on abovebest spectrum parameter of hyperspectral remote sensing in four growth stages isY=10.115-27.525x+24.799x², Y=1.713+4.118x-7.798x²+3.685x³,Y=4.701-11.623x+15.051x²and Y=3.513-6.543x+67.755x²-154.672x³.