Total Nitrogen Content Of Soil In Panjin Paddy Field Based On Hyperspectral Data Model Research Sensing

Rice is a food crop that Chinese people rely on for their livelihoods.In recent years,due to various reasons,the rice planting industry has experienced a decline in planting area and a slow increase in output.To solve this problem,farmers often use excessive nitrogen fertilizers.To achieve the purpose of increasing yield,this will make a large surplus of nitrogen in the soil,which will not only attenuate the efficiency and utilization of nitrogen fertilizer,but also bring significant pressure on the environment.Although two-thirds of the nitrogen in rice comes from the soil,which plays an important role in the growth and development of crops,insufficient fertility or excessive fertilization has always been one of the important reasons for soil failure and environmental pollution.Therefore,monitoring the total nitrogen content of paddy soil is a prerequisite for accurate scientific fertilization.In this paper,in cooperation with the Liaoning Province Saline-Alkaline Land Utilization Research Institute,the ASD ground object spectrometer is used for field spectrum acquisition and soil sample sampling.The measured data and GF-1 satellite imagery are used as the basic data to construct the whole ground through spectral transformation and spectral index transformation.The nitrogen estimation model and the construction of a total nitrogen inversion model by synergistic analysis of resampling hyperspectral data and satellite data can obtain a total nitrogen content model of paddy soil in Panjin area that meets different needs and provide a basis for reasonable quantitative fertilization of paddy soil.Taking the research area of Panjin City as the research area,the above scientific methods are used to explore the best estimation methods for soil monitoring total nitrogen from different monitoring methods.The main conclusions are as follows:（1）According to the characteristics of the spectral curve,the reflectivity increases with the increase of total nitrogen,and the correlation is obviously improved by differential transformation and index,and the band with high correlation with soil measured total nitrogen is 427 nm、487 nm、492 nm、502 nm、506 nm、705 nm、843 nm、844 nm、886 nm.The distribution of soil total nitrogen content in Panjin area is higher in the west,and the higher in the west is near the Liaohe River system.The distance from the entrance to the sea is also one of the influencing factors of the spatial distribution of soil total nitrogen.（2）The correlation coefficient between the reciprocal second order differential transformation of reflectivity logarithm and the measured soil total nitrogen content is the most obvious,reaching 0.71.The difference spectral index model of the second order differential of reflectivity logarithm R²=0.723.The selected band is the best DI_{（697,705）}and DI_{（505,521）},fitting effect after band change.（3）By resampling the hyperspectral data to the satellite band to construct the inversion model,the determination coefficient of the inversion model R²=0.566,and the model prediction value and the measured value are well fitted,which can be used to retrieve the total nitrogen in a large range of satellites.