Characterization and measurement of the hydraulic properties of unsaturated porous media is still a challenge in natural environments, although exact knowledge of the soil's hydraulic properties [unsaturated soil hydraulic conductivity (K), volumetric water content (θ), soil matric head (h)] is crucial for solving many soil, hydrological, and environmental issues. The main purpose of this study was to establish a modified multistep outflow technique that facilitates laboratory operations and reduces time and costs. We also focused on inverse parameter optimization of the Mualem–van Genuchten (MVG) model with laboratory‐provided data for three fine‐grained soils. This modified version was assessed by inverse modeling of the MVG model and the generalized reduced gradient algorithm. The results represent the best estimate of the adjusted parameters and the very low uncertainty associated with the fitted values. The soil hydraulic functions were simulated well up to pressures of 1,500 kPa and their uncertainties were extremely low. The R² values were 88 to 97 and 81 to 96% for the soil water characteristics and the unsaturated hydraulic conductivity functions, respectively. The simultaneous optimization of the nonlinear hydraulic functions from a single transient flow experiment agreed well with their observed data for each soil examined. Besides, by eliminating the use of tensiometers and having no independent measurements of saturated soil hydraulic conductivity and saturated volumetric water content, the modified version was faster and easier than the conventional method. The proposed protocol appears to be suited for the vadose zone flow simulation of fine‐grained soils.

中文翻译：

---

通过改进的多步流出对细粒土壤的逆流模拟，模拟渗流区流过程

尽管对土壤水力特性的确切知识[不饱和土壤水力传导率（K），体积水含量（θ），土壤基质水头（h）的精确了解，在自然环境中表征和测量不饱和多孔介质的水力特性仍然是一个挑战。）]对于解决许多土壤，水文和环境问题至关重要。这项研究的主要目的是建立一种改进的多步流出技术，以促进实验室操作并减少时间和成本。我们还将重点放在Mualem-van Genuchten（MVG）模型的逆参数优化上，该模型具有实验室提供的三种细粒土壤的数据。通过MVG模型的逆建模和广义归一化梯度算法评估了此修改版本。结果代表对调整后的参数的最佳估计，并且与拟合值相关的不确定性非常低。在高达1,500 kPa的压力下对土壤水力功能进行了很好的模拟，其不确定性极低。的- [R ²土壤水分特性和非饱和导水率函数的数值分别为88％至97％和81％至96％。来自单个瞬态流动实验的非线性水力功能的同时优化与每种土壤的观测数据非常吻合。此外，由于取消了张力计的使用，并且没有独立测量饱和土壤的水力传导率和饱和体积水含量，因此修改后的版本比传统方法更快，更容易。拟议的协议似乎适合于细颗粒土壤渗流区流模拟。