论文部分内容阅读
当土壤转换函数应用于土壤水力性质估计时,对于预测值的不确定性往往容易被忽视。为了有针对性地提出减少这种不确定性的方法和措施,提高土壤转换函数的实际应用能力,以两种现有的土壤转换函数(Vereecken和HYPRES模型)为例,将其应用于山东省平度市土壤饱和导水率的空间预测,并利用拉丁超立方抽样(LHS)方法对预测结果的不确定性进行了分析。结果表明,饱和导水率空间预测的不确定性主要来源于土壤基本性质的空间插值误差和土壤转换函数自身的预测误差。当Vereecken模型应用于饱和导水率空间预测时,预测结果的不确定性主要由土壤基本性质空间插值误差所决定,土壤转换函数预测误差的影响较小,而HYPRES模型则是受二者的双重影响。
When the soil transfer function is applied to the estimation of soil hydraulic properties, the uncertainty about the predicted value tends to be easily neglected. In order to point out the methods and measures to reduce this uncertainty and improve the practical application ability of soil transfer function, two existing soil transfer functions (Vereecken and HYPRES models) are used as examples to apply them to Shandong Province The spatial prediction of saturated hydraulic conductivity of soil in Pingdu City was also analyzed. The uncertainty of prediction results was analyzed by Latin Hypercube Sampling (LHS). The results show that the uncertainty of the spatial prediction of saturated hydraulic conductivity mainly comes from the spatial interpolation error of the basic properties of the soil and the prediction error of the soil transfer function itself. When the Vereecken model is applied to predict the saturated hydraulic conductivity, the uncertainty of the prediction results is mainly determined by the spatial interpolation error of the basic properties of the soil, and the influence of the prediction error of the soil transfer function is small, while the HYPRES model is affected by the dual influences.