| Precision agriculture now becomes the topic field of new technology research. The study on spatial variability of soil characteristic and crop information is the technology support of precision agriculture. The emphasis of past research is the spatial variability of soil characteristic or crop information, but the character of crop growth is ignorable. The growth of crop is an important factor between soil characteristic and the yield of winter wheat, we should go deep into it. Under the support of the project of"211"of Jiangsu province and Jiangsu province important water conservancy laboratory, this paper took the south soil characteristic and winter wheat information as the object, particularly studied the spatial variability and the connection of them. The main conclusion was as follows.(1) The result indicated that, Under this research condition ( the area was 45m×45m, the space between gridding was 5m×5m ), soil characteristic belonged to middling variation, thereinto it, variation of the values of soil content moisture were adjacente. The winter wheat LAI of every period belonged to middling variation, Chlorophyl and height of wheat belonged to low variation, the variation of winter wheat yield were low too. Every variable had determinate variance degree, this current of distributing accorded with the normal school. The number of sampling was confirmed in small yardstick. Not only the workload was least, but also the mathematical expectation met the request. When the believe level was 95%, precision wasΔ=0.05x, the number of sampling in reason was fewer than 100 and accorded to the request. Soil characteristic and crop information had the characters of the spatial variability. Using this method could offset the shortage of using classical statistics to study on the messages of plant.(2) Disquisition indicated that, soil characteristic and wheat information had the characteristic of regionalized variable. That was to say, it not only had the characteristic of random, but also had the character of configuration. The majority of the regionalized variable accorded with the model of globular model. Some accorded with the model of linearity, there was farthing which had no model. The correlation distance of soil characteristic was smaller than crop information as a whole. It indicated that the... |
