农田基础环境信息空间变异性分析

The analysis of the spatial variability of farmland basic environment infor-mation

运用 SPSS软件 ,通过对上海精准农业实验区内采集的各项农田基础环境信息数据进行处理 ,计算出各项农田环境的相应统计学参数 ,并分析它们在不同田块上的空间变异性。在此基础上 ,得出结论为 :有机质、全氮、速效氮和 Mn的空间变异性较小 ,全氮和速效氮空间变异性较小的原因可能是均匀地大量施用氮肥造成的 ;速效钾、B与 Fe的空间变异性幅度与产量基本一致 ;而大量养分要素全磷、速效磷和微量元素 Cu、Zn的在田间的变异性较大 ,它们可能是造成产量存在明显的空间变异性的主要原因。

Although the Chinese government has applied a series of policies to control her population, it is still increasing and will only reach the climax by the year of 2050^([1]). In order to support her large population, China has to introduce new technologies into agriculture in order to increase its productivity without adversely affecting the rural environment. A basic solution is precision agriculture, which tries to apply the precisely needed quantity of chemical inputs such as of fertilizers to the farmland. The required quantities depend on the spatial variability of the basic environmental factors (for example soil characteristics) and should finally lead to “prescriptions” for modern farmland management. This may be the appropriate way for different countries to increase agriculture productivity without harming the environment. Therefore, it is important to study precision farming, which is based on the use of modern high-tech information management, biology and engineering. The inherent soil nutrient factors always have an effect on the yield. \;Experiments, carried out have shown that in different farmlands, the productivity may have a variability up to 100%^([5]). It is necessary for scientists to study the spatial variability of farmland environmental parameter. With the support of modern high-tech such as GIS, GPS, RS, computer aided cartography and analysis and artificial intelligence, it has become possible for scientists to study the spatial information accurately and find out those factors that have the greatest impacts on farmland productivity and its environment. In this paper, the author tries to analyze the data collected on an experimental base with the software SPSS and work out their statistics indices. The analysis includes arithmetic means (which shows the central tendency of observations cluster), kurtosis (which shows the extent of observations cluster around a central point), standard deviation (which shows the dispersion around means), skewness (which shows the symmetry of distribution) and dispersion index (=standard deviation/means), etc. Based on the experimental data, the variation of basic farmland parameter and their correlation with the variation of paddy yield could be found out. The completed research work has shown that the spatial variability of organic matter, total nitrogen, quick result nitrogen and Mn in the study area shows only a very limited variation (the little spatial variability of total nitrogen and quick result nitrogen could eventually be caused by equable application of a great deal of nitrogenous fertilizer). The spatial variability of quick result potassium, B and Fe corresponds well with the paddy yield. The spatial variability of paddy yield in the experiment field is probably caused by the spatial distribution variability of salinity, abundant element Phosphor and microelements, Cu and Zn.