【摘要】 土壤磷素流失强度评价对于控制面源磷素流失,降低受纳水体富营养化程度极为重要。为了探索一套有效的评价方法,该文对土壤磷素流失机制进行了研究,并且在对土壤磷素循环及背景参数进行分析的基础上,应用径向基函数(radial basis function,RBF)神经网络技术建立了土壤磷素流失强度评价模型。将模型运用于江苏省土壤普查样本的算例结果显示,96.7%的学习样本拟合相对误差小于0.3,预测相对误差均在0.4以内。结果表明,模型是精确的,可以用于实际土壤磷素流失的评价。更多还原

【Abstract】 Soil phosphorus losses must be assessed to control phosphorus loss and water eutrophication.The soil phosphorus loss mechanism was studied to develop a convenient assessment method.Analysis of soil phosphorus circulation mechanism led to an assessment model based on a radial basis function network application,the model in Jiangsu Province showed that the relative fitting errors were less than 0.3 for 96.7% of the learing samples and all the relative prediction errors are less than 0.4.The results show that the model is accurate and can be used for practicable soil phosphorus applications.更多还原