Study On Nitrogen Nutrition Diagnosis For Cold-terra Rice Based On Airborne Multispectral Imager And SPAD

Excess or deficiency of nitrogen will have an important impact on rice growth and development,yield and grain quality.Traditional nitrogen nutrition diagnosis methods require extensive field experience,which is time-consuming and destructive.Therefore,based on airborne multi-spectrum and SPAD,nitrogen nutrition diagnosis for cold-terra rice was carried out.Rapid nondestructive nitrogen nutrition diagnosis has become a hot topic in precision agriculture research.Heilongjiang Province is an emerging rice cultivation area in China,and the rules of rice growth and fertilization under cold climate conditions are special.In this study,two main rice cultivar in the first accumulated temperature zone of Heilongjiang Province were selected for the nitrogen fertilizer experiment.Using a multi-spectral camera equipped with a drone to obtain the canopy spectral data of rice in critical periods,and using the SPAD chlorophyll meter to obtain canopy SPAD readings,and simultaneously obtain the agronomic parameters such as leaf nitrogen content,leaf dry matter weight and maturity yield.The dilution curve of the critical nitrogen concentration for cold-terra rice based on leaf was established.A nitrogen nutrition diagnosis model of SPAD index based on the critical nitrogen concentration curve was constructed,a nitrogen nutrition diagnosis model based on the canopy spectral vegetation index and agronomic parameters of the UAV platform was constructed.Diagnosis of nitrogen nutrition of rice in the field and small area is realized.The main findings are as follows:1.According to the theory of critical nitrogen concentration dilution,the critical nitrogen concentration dilution model of the main cultivar is constructed.Wuyoudao 4: Nc =1.96*LDM-0.56,R2= 0.87?p <0.01?,RMSE = 0.31,n RMSE = 13.07%;Songjing 9: Nc =1.99*LDM-0.44,R2= 0.94?p <0.01?,RMSE = 0.37,n RMSE = 15.89%,the tested model has good reliability and stability.The nitrogen nutrition index?NNI?was calculated using the constructed critical nitrogen concentration dilution curve to diagnose the nitrogen nutrition status of rice.Based on the research results,it is recommended that the rice cultivar Wuyoudao4 should be applied at a level of less than 120kg·hm-2in the first accumulatedtemperature zone of Heilongjiang Province,and that of Songjing9 should be applied in the first accumulated temperature zone of Heilongjiang Province less than 180kg·hm-2,basically consistent with the high-yield fertilization level in the field.2.A nitrogen diagnosis model based on SPAD index was constructed.At different growth stages,the SPAD readings of the top 3-4 fully unfolded leaves of the rice were measured,and the literature analysis method was used to select the SPAD value and index related to the nitrogen index,and the SPAD values of different leaf positions of the rice canopy were analyzed and compared.And the correlation between the index and the nitrogen index,a nitrogen diagnosis model based on the SPAD index was established.The results show that the normalized SPAD index?NSI4?of the top 4 fully expanded leaves is sensitive to nitrogen response,and the difference between cultivar can be effectively eliminated by normalization.A nitrogen fertilizer diagnosis model of NNI and NSI4 was established in the critical period of rice field management and nitrogen fertilizer regulation?Stem elongation-heading stage?,NNI = 0.3048e^1.3864*NSI4?R²= 0.862,p<0.01?.After testing,Wuyoudao 4: RMSE=0.15 n RMSE =14.59%;Songjing9:RMSE=0.15,n RMSE =16.67%,indicating that the constructed model can estimate the nitrogen status of rice better.3.The mini-MCA multi-spectral camera equipped with a drone was used to diagnose rice nitrogen nutrition based on the small and medium-scale region.The mini-MCA multi-spectral phase included Piexl Wrench 2 image processing and analysis software was used to obtain the NDVI of the canopy vegetation of cold rice,and Agronomic parameters to build relevant models.The results show that NDVI can be used to estimate nitrogen nutrition index?NNI?of rice leaves,and NNI is not affected by the growth period.By comprehensively comparing the evaluation indexes such as the dertermination coefficient R2,RMSE,n RMSE,the index model has the best effect.NDVI nitrogen fertilizer diagnosis model of Wuyoudao4:NNI =0.3916e^1.0809*NDVI,RMSE=0.12,n RMSE =12.43% Songjing9: NNI = 0.3325e^1.2705*NDVI,RMSE=0.10,n RMSE=10.36%,indicating that the constructed model can better estimate the nitrogen status of rice.In summary,the multi-spectral camera based on the UAV remote sensing platform Chlorophyll meter has good feasibility for dynamic monitoring of the nitrogen status of riceleaves.It can solve the problems of destructive sampling,time-consuming,and labor-intensive limitations.It is a good choice for small-medium scale application.