利用精确的田间实验资料对几个常用根系吸水模型的评价与改进

本文利用大型蒸渗仪测得的作物腾发量、中子水分仪观测的土壤水分和准确测定的根系密度分布资料，对常用的几个宏观的权重因子类的根系吸水模型-Molz-Remson(1970)模型，Feddes(1978)模型，Selim-Iskan-dar(1978)模型以及作者对上述模型进行修正所得的几个根系吸水模型进行了验证和评价；利用修正的Feddes模型的计算结果对根系从不同土层吸水的分布进行了分析。结果表明，Molz-Remson(1970)模型、Feddes模型以及Selim-Iskandar模型模拟根系吸水所得的土壤水分剖面与实测值之间存在比较严重的偏差；利用Feddes模型中的土壤水势影响函数(Feddes reduction function)对Molz-Remson模型和Selim-Iskandar模型进行修正后结果没有得到改善；利用根系密度函数对Feddes模型进行修正后，计算结果与实测值吻合很好，总体偏差由修正前的24.7%降低为5.7%.

The evaluation of water uptake models by using precise field observation data

In this paper, the macroscopic water uptake models widely used in modeling the water flow in soil\|plant system are evaluated by using the precisely obtained evaptranspiration and soil water content data from the lysimeter and the root distribution data obtained in the wheat field nearby. Meanwhile, the Molz\|Remson model and Selim\|Iskandar model are modified with the Feddes reduction function, and the Feddes model is modified with the root length density to account for the role of root distribution. Those models are incorporated into the soil water flow equations as extraction term. The flow equations are numerically solved with the measured evapotranspiration as input. The predicted and measured soil water content profiles are compared to evaluate the validity of the extraction term models. Results show that:(1)the average deviation of the molz\|Remson(linear)model, Feddes model and Selim\|Iskandar model got as high as 20%, and the Molz\|Remson model 13%;(2)modifications made to the Molz\|Remson model and Iskandar model did not achieve any improvement to the model behavior, but that to the Feddes model achieved a great success in lifting its prediction ability.