Study On Estimation Model Of Water And Nitrogen Content And Yield Of Winter Wheat Based On Spectral Information

Water and fertilizer are essential factors in the growth process of winter wheat.It is of great significance to obtain the water and nitrogen status of plants in time to improve the water and nitrogen management level of crops and ensure the improvement of quality and efficiency of crops.High quality and high yield of grain is related to the national economy and the people's livelihood.Under natural conditions,irrigation and fertilization are used to monitor the yield of winter wheat in real time and accurately,so as to ensure the increase of income and yield of winter wheat.In this paper,the response of plant water content,plant nitrogen content and spectral reflectance was studied,the optimal single parameter regression model of the whole growth period,the combination model of different growth periods and the corresponding principal component regression model were constructed respectively.At the same time,based on the measured hyperspectral information of winter wheat canopy,the spectral response function based on Landsat8 OLI sensor is used to simulate the multispectral reflectance,and the water content and nitrogen content estimation model based on spectral transformation and vegetation index is constructed.On this basis,a regional yield estimation model based on the coupling of water and nitrogen is proposed,which provides a theoretical method and technical support for regional winter wheat water and nitrogen management and yield monitoring.The following main conclusions are reached:(1)Under different water treatments,the canopy reflectance of winter wheat decreased with the increase of irrigation in the visible band and increased with the increase of irrigation in the near infrared band;Under different nitrogen application levels,in the visible band,with the increase of nitrogen application,the canopy reflectance of winter wheat decreased,while in the near infrared band,the change characteristics were opposite.(2)Based on the principle of the highest correlation coefficient between the spectral transformation and the "three-edge" parameter in each growth period,a combination of plant water content monitoring model based on the spectral transformation and the "three-edge" parameter in each growth period was established.Compared with the optimal single parameter regression model in the whole growth period,the combination model in each growth period significantly improved the estimation accuracy of plant water content in each growth period;By using principal component analysis method to solve the problem of collinearity among indicators,a principal component analysis monitoring model based on comprehensive spectral information is constructed.Compared with the model combination based on spectral transformation and "three-edge" parameters,the monitoring accuracy of jointing heading stage is significantly improved,and the monitoring accuracy of other growth stages is also significantly improved.At the same time,the same model construction method was used to construct the monitoring model of plant nitrogen content in winter wheat,and the combination of principal component analysis model constructed in different growth periods was the best.(3)The corresponding spectral transformation value and vegetation index are calculated based on the simulated multispectral data of each growth period,and the plant water content monitoring model combination based on the spectral transformation and vegetation index is constructed according to the growth period.The model combination based on the vegetation index is more suitable for estimating the plant water content of winter wheat in each growth period.According to the accuracy test of the model,the coefficient R2 of jointing-heading stage,heading-filling stage,filling-maturing stage and the whole growth stage are 0.365,0.635,0.698 and 0.899,respectively.At the same time,the combination of plant nitrogen content model based on vegetation index is the best model for monitoring winter wheat plant nitrogen content.(4)Comprehensive comparison of winter wheat yield estimation models based on plant water content,plant nitrogen content,and water nitrogen coupling.According to the simulation accuracy of the model,the order from high to low is based on the water nitrogen coupling estimation model,the estimation model with plant water content as the intermediate parameter,and the estimation model with plant nitrogen content as the intermediate parameter.According to the results of water and nitrogen estimation of regional winter wheat,the objective evaluation of regional winter wheat yield was carried out.The range of winter wheat yield estimated based on Landsat8 OLI in Daxing District of Beijing was 2184-8913 kg/hm~2,with an average yield of 5717 kg/hm~2,which was close to the field measured yield.