畦灌土壤入渗参数估算的线性回归法

土壤入渗参数是地面灌溉设计的基本参数。针对现有估算方法存在计算工作量较大、精度不稳定的问题进行了研究，提出了畦灌土壤入渗参数的改进计算公式。在70m的畦田长度上等间距布置了8个水位传感器以观测地表水深沿畦长的分布，以及该分布随时间的变化。将畦灌地表水深与离畦首的距离之间的关系按指数函数进行非线性最小二乘拟合，并根据水量平衡原理提出了估算土壤入渗参数的线性回归法，避开了难以定值的地表储水形状系数。实例计算表明，地表水深与离畦首的距离之间呈现较好的指数函数关系，线性回归法估算土壤入渗参数的计算工作量较小，计算精度较高。

Linear regression approach for estimating soil infiltration parameters of border irrigation

A modified approach for estimating the infiltration parameters of border irrigation is presented. Eight water-level probes are uniformly installed along a 70m border to measure the spatial and temporal variation of surface water depth. The relationships between surface-water depth and distance from water inlet are fitted by a nonlinear least-square method, and then, based on the principle of water balance a linear regression method is proposed for estimating the soil infiltration parameters of border irrigation. In this way the surface storage shape coefficient, which is not easy to be determined, is not needed anymore. The results of application indicate that surface-water depth and distance from water inlet are well fitted with an exponential function, and the proposed linear regression approach can reduce computation time and improve the precision of estimation.