统计分区施肥系统模型的建立及应用

发布时间：2019-02-13 03:56

【摘要】：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。
[Abstract]:Based on the statistical analysis of cultivated land yield and soil nutrient content in a certain area, a statistical zoning model called SDFM), was established and divided into high, middle and low yield areas and soil nutrient content intervals according to the parameters of the model. A multi-point experiment was carried out in the region to obtain the appropriate fertilization rate in the region. Then, on the basis of the suitable fertilization rate in the region, the actual yield and soil nutrient value of the decision point are adjusted, and the final fertilization recommendation result is obtained, and then the decision on the fertilization amount is made. Based on the data of yield and nutrient sampling point in Nong'an County in 2006 and 2011, the decision making was made by using the nutrient balance method of target yield and the statistical fertilization model. The results show that the probability of abnormal value in the decision results of statistical zoning model is only 2, while the probability of abnormal value of decision results of target yield method for nitrogen fertilizer, phosphate fertilizer and potash fertilizer is 4547% and 47%, respectively. It can be seen that statistical zonal fertilization greatly improves the precision of decision-making results and can be popularized in the soil testing formula fertilization below the county level.
【作者单位】：吉林农业大学资源与环境学院;
【基金】：国家自然基金项目(40671124)资助
【分类号】：S147

【相似文献】