Study On Variable-rate Fertilization In Winter Wheat Based On Multi-sources Data

Variable-rate fertilization is one of the key technologies in precision farming. The aim of this study was to develop the time-specific and time-critical method to overcome the limitations of traditional field sampling methods for variable rate fertilization. The experiment was carried out on the National Precision Agricultural Research Basement of Xiao Tangshan, Beijing from 2005 to 2006. The dissertation mainly studied the variable rate nitrogen (VRN) through extracting the winter wheat (Triticum aestivum L.) spectrum information from the hyperspectral data and SPAD reading by ASD Fro2500 spectroadiometers and SPAD meter (Chlorophyll Meter), put great emphasis on the study of canopy N dynamic distributing character and analyzed the fertilizer mechanism of VRT based on hyperspectral data, and the timely obtained information of wheat growth status, especially N status, will help us to apply fertilizer management. The study constituted the variable-rate diagnostic model based on multi-sources date , and discussed the influence of winter wheat grain yield ,quality and economical and environmental benefits, at last, compared with the advantage and disadvantage of different fertilizer pattern conditions based on multi-sources date. The results are summarized as below:Effects of variable nitrogen application based on characteristics of canopy light reflectance in wheat, the test studied the VRT using normalized different SPAD (NDSPAD) value for the first time, and based on grid-sampling of NO3-N soil nitrogen and yield in 2005, all of the them increased significantly grain yield and dry matter of winter wheat, and decreased the variable coefficients of grain yield and dry matter. The variable-rate treatment improved the grain yield output, cost proportion, and increased the economic benefits. The VRT also could improve gain quality and environmental benefits.Studies of variable nitrogen application in winter wheat based on spectral index and simulation models: Combining CERES-Wheat models and spectral index (Optimization of soil-adjusted vegetation index ,OSAVI) was used to determine the amount of nitrogen fertilizer recommended for variable rate management. This result showed that the CERES-Wheat model can be used to predict the state and biophysical variables of wheat. The grain yield and grain protein content were measured among the uniform rate treatment, No-fertilization treatment and variable rate fertilizer treatments. It indicated that variable rate fertilization increased grain yield and grain protein content, and it did reduce the variability of them. Economical benefits of fertilization in variable rate application were better than those in the uniform rate treatment, No-fertilization treatment.Set up the N fertilization optimization algorithm based on multi-sources date in Beijing agricultural production conditions. It provided a techno-support for precision agriculture variable-rate technology. The results of four VRN showed that the economic benefit of VRN based on spectral index and simulation models was best among four rate technology and balance rate technology to carry through the field manage.