论文部分内容阅读
通过对一定区域内耕地产量和土壤养分含量进行统计分析,建立统计分区模型(Statistic Division Fertilization Model,简称SDFM),并依据模型参数划分高、中、低产量区间和土壤养分含量区间。在区域内进行多点实验,获取它们的区域适宜施肥量。然后在区域适宜施肥量的基础上,利用决策点实际产量和土壤养分值进行调整,得到最终施肥推荐结果,从而对施肥量进行决策。以农安县2006~2011年产量和养分采样点数据为原始数据,利用目标产量养分平衡法和统计分区施肥模型进行了决策对比。结果显示:统计分区模型决策结果出现异常值的概率只有2%,而目标产量法对氮肥、磷肥和钾肥的决策结果异常值出现概率分别为45%,47%和47%。可见统计分区施肥很大程度地提高了决策结果的精度,可在县级以下的测土配方施肥中进行推广。
Statistic Division Fertilization Model (SDFM) was established by statistical analysis of cultivated land yield and soil nutrient content in a certain area, and the high, middle and low yield intervals and soil nutrient content interval were divided according to the model parameters. In the region to carry out more experiments to obtain the appropriate amount of fertilizer for their region. Then based on the suitable fertilization amount in the region, the actual output of decision point and the value of soil nutrient value are adjusted to get the final recommendation result of fertilization, so as to make the decision of fertilization amount. Taking the data of production and nutrient sampling points from 2006 to 2011 in Nong’an County as the raw data, the decision-making comparison was made by using the target yield nutrient balance method and the statistical partition fertilization model. The results showed that the probability of outliers in the statistical partition model was only 2%, while the probability of outliers in the target yielding method was 45%, 47% and 47% respectively. Visible statistical partition fertilization greatly improves the accuracy of decision-making results, can be in the soil testing and fertilizer below the county level to promote.