Monitoring Of Wheat Growth And Nitrogen Nutrition In Hetao Irrigation Area Based On UAV Multispectral Technology

In precision agriculture,multispectral technology had become an important tool for evaluating soil productivity and analyzing plant health due to its advantages of convenience,high efficiency,fast speed and no damage to plants.It is widely used in crop growth,water and nutrition status,yield and quality monitoring and other aspects.In this study,spring wheat in Hetao irrigation area was selected as the research object,and the changes of leaf SPAD value,plant nitrogen content,dry matter mass and yield of wheat under different nitrogen application rates were analyzed.Spectral indices such as GNDVI,LCI,NDRE,NDVI and OSAVI were used.The prediction models of single vegetation index and multivariate vegetation index based on UAV multispectral vegetation index,including SPAD value,nitrogen content,dry matter mass,yield and nitrogen nutrient index（NNI）,were established in order to provide basis for nitrogen nutrient diagnosis and precise fertilization of wheat in Hetao irrigation area.The main research conclusions are as follows:（1）SPAD values of flag leaves of wheat under different nitrogen application rates increased first and then decreased with the growth process,and reached the maximum value at flowering stage.Plant nitrogen content（NCP）decreased gradually with the growth process and reached the lowest value at maturity stage.Above-ground dry matter accumulation（DW）increased rapidly after jointing and accumulated slowly at filling stage until reaching the maximum at maturity stage.SPAD,NCP and DW of flag leaves of wheat were firstly increased and then decreased with the increase of nitrogen application rate.The number of ears,grain number per ear,1000-grain weight and yield of wheat increased with the increase of nitrogen application rate.When the total nitrogen application rate（pure N）reached more than 260 kg/hm²,wheat yield and its component factors did not increase significantly.（2）The prediction model of four key growth stages of spring wheat in Hetao Irrigation area was established by stepwise regression analysis using UAV multi-spectral images.The SPAD value of flag leaves at heading stage of wheat had the highest prediction accuracy,and the model was SPAD=61.897*LCI+21.900（R²=0.998）.The prediction accuracy of nitrogen content（NCP）in filling stage was the highest,as follows:NCP=0.790-16.074*LCI+23.820*NDRE（R²=0.894）;The prediction accuracy of dry matter（DW）at heading stage was the highest,and the model was DW=80.529*NDRE-95.044*OSAVI+22.897（R²=0.991）.The best time for wheat yield prediction was heading stage,and the prediction model was Y=30117.366*NDRE-2366.573（R²=0.973）.（3）The equation of nitrogen dilution curve based on above-ground dry matter reestablishment of wheat was N=3.544W^-0.2444（R²=0.6039）.The NNI of wheat increased significantly with the increase of nitrogen application,which decreased first and then increased with the growth process,and decreased to less than 1 at heading and flowering stage.The most accurate prediction period of wheat nitrogen nutrition index（NNI）was heading stage,and the prediction model was NNI=4.254*NDRE-2.927*NDVI+1.818（R²=0.996）.It was concluded that 375 kg/hm²was applied with diammonium phosphate（18%N）and 315 kg/hm²urea（46%N）at tillering stage in Hetao irrigation area,which could basically meet the nitrogen demands of spring wheat during the whole growth stage.