Study On The Monitorition By RS For Crop Growth Condition And Spatial Variability On The Filed Scale

The index of crop growth monitoring by RS have closely relation with crop yield. Meanwhile, the growth monitoring becomes the precondition of the yield estimation, which is important for the macro control of food. In the usual monitoring, we mainly analysis contemporaneous RS images among different years in order to reflect the relative difference of the crop growth in areas. This was the semi-quantitative method and it did not deeply study by combing the soil spatial variability which influenced the growth difference. For this, this study was developed in the dissertation.The experiment was carried out on the National Precision Agricultural Research Basement of Xiao Tangshan, Beijing from 2005 to 2006. The winter wheat was the mainly f study object in the experiment. The relation among the soil nutrient,the spatial variability of the crop growth and the satellite image was studied by combining the high-resolution Quickbird image in filed scale and agronomic parameter on the ground based on many methods such as the remote sensing technology,traditional statistics and geo-statistics. The study referred to many aspects: the establishment of the inversion model for the growth parameter and its validation, the relation of the soil nutrient and growth spatial variability, the precision production and management based on the growth variability. The contents and results are summarized as below:(1) The various vegetation index was selected which were caculated from canopy spectra in actual measurement. The linear regression model was established for the vegetation index and LAI from which the LAI. could be inversed. And then, the simple linear regression model based on NDVI was the best model, and the fitting precision reached to 0.7588 after the validation. At last, this best model was chosen to inverse the image and the figure for the distribution of the LAI could be got. Therefore, the crop growth could be monitored in large area.(2) The spatial variation of soil nutrients was studied by the geo-statistics. The presented study tried to calculate the NDVI (Normalized Difference Vegitation Index) values from the Quickbird imagery as the indicator of wheat growth. After the semivariogram analysis, the Kriging method was used to interpolate the spatial distribution map for each soil nutrient, then Kappa coefficients was calculated to evaluate the consistency between soil nutrients and NDVI. And then, the relation between the soil nutrients and the spatial variability of the wheat growth could be got. So the soil AP and OM contents were the determinative factors for wheat growth in the studied field.(3) On the basis of the spatial variability, the study took much data to divide the management zone. The data were as follows: the spectral index got from Quickbird imagery, soil sampling data of the study area ,the yield data in the present year. The result showed that the homogeneous degree of the soil nutrient was increased, the variability of spectral index and the yield were reduced in the management zone divided by soil data. As a result, the spatial variability of the soil nutrient and spectrual data should be taken into consideration when dividing the management zone. We could get satisfactory division result by using the soil nutrient...