Yield Prediction And Construction Of Critical Nitrogen Concentration Model For Maize Under Mulched Drip Irrigation

Because of its reliability.Critical nitrogen concentration(Nc)dilution curve is used as an unique tool in crop biology for effectively diagnosing N nutrition status.Faced with over-inputted N fertilize.imprecise field management and difficult dynamic regulation in maize production,in Ningxia Yellow River Irrigation Area,China,from 2019 to 2020,an experiment involved field N fertilizer levels(Pure N:0,120,240 and 360kg/hm2)was conducted in typical irrigation areas of Ningxia.According to the principle of Nc dilution,the Nc dilution curve of mulched drip irrigation for maize was drawn based on aboveground biomass(AGB),Leaf dry matter(LDM),Leaf Area Index(LAI)and Canopy cover(CC).Further,the N nutrition index(NNI)and N accumulated deficit(NAD)models were combined to dynamically diagnose and evaluate the N nutrition status of maize.Meanwhile,the maize grain yield was predicted with the Nc dilution curve established by CC.The main results are as follows:(i)AGB has a power-function relationship with Nc,maximum N concentration(Nmax)and minimum N concentration(Nmin)of maize,and the determination coefficients(R2)were 0.88,0.96 and 0.89,respectively.The model test indicators’s root mean square error(RMSE)and normalized root mean square error(n-RMSE)were 0.17,9.09%,respectively,showing a good stability.By NNI and NAD modeling,it was determined that 240 kg/hm2 was the optimal N application rate for maize under mulched drip irrigation.(ii)The maize Nc model based on LDM was divided into two sections:when LDM<1.15 t/hm2,Nc=3.21%;when LDM≥1.15t/hm2,Nc=3.31LDM-0.21.The evaluation index,RMSE of the fitted model,was 0.15,and the n-RMSE was 6.71%,showing a high accuracy of model.Based on the linear plus platform model and the NNI model,the optimal N application rate for maize was between 222.71～240kg/hm2,and the maximum yield of maize estimated by the two methods was almost similar.There was an extremely significant positive correlation between NNI and yield during different growth phrase.Meanwhile,NNI was negatively correlated with N’s agronomic utilization efficiency(NAE),fertilizer recovery efficiency(NRE)and its fertilizer physiological efficiency(NPE)while little significantly correlated with N’s partial factor productivity(NPEP).(iii)The Nc dilution curve(Nc=3.56LAI-0.36)of maize from jointing to silking stage based on LAI could exactely distinguish the data of N-limited treatment and non-N-limited treatment in the independent verification experiment,and the model was fairly stable.According to NNI and Critical nitrogen uptake(Nuptc)deduced by Nc dilution curve,the optimal N application rate of maize was 240 kg/hm2.During critical growth stage of maize,the relationship between N’s integral nutrition index(INNI)and its relative yield(RY)could predict the yield properly.(iv)The dynamic relationship between maize’s CC and AGB,and that between LAI and N’s aboveground accumulation(ANA)can both be described by an exponential function model.The determination coefficients for CC-based models of Nc,Nmax,and Nmin were 0.92,0.84,and 0.87,respectively.The model verification parameters,RMSE and n-RMSE were 0.24 and 11.75%,respectively.Based on CC calculation,it was estimated with NNI and NAD models that the optimal N application rate for maize was 240 kg/hm2,which was consistent with the appropriate N application rate recommended by AGB and LAI.(v)The relationship among NNI,NAD and RY at different growth stages of maize can be described by a linear plus platform model,and during belling stage(V12)and tasseling stage(VT)the highest R2 could be found out.The model verification results showed that the average R2 of maize V12 and VT were both higher than 0.944,but the n-RMSE of the same stages were both less than 10%,which further confirmed the stability of the model at maize’s critical growth stages such as V12 and VT.The significant relationship among NNI,NAD and RY properly eluicidated the variation of RY under N-limited and non-N-limited growth conditions,thus providing a theoretical basis for the prediction of maize yield under mulched drip irrigation in Ningxia Yellow River Irrigation Area,China.