Repeated falling-head method for in situ measurements of saturated hydraulic conductivity using a single cylinder

•A method to determine the field-saturated hydraulic conductivity, Kfs, is proposed.•HYDRUS (2D/3D) was used to evaluate the dimensionless shape factor.•Repeating the falling-head test eliminates the effect of initial soil wetness.•Kfs can be obtained without requiring any soil-specific parameters.•Kfs obtained with the repeated falling head test agrees well with common methods. A simple, repeated falling-head (RFH) method, which uses an inexpensive portable cylinder to quickly determine field-saturated hydraulic conductivity (Kfs), is presented. A cylinder of radius r, inserted vertically from the soil surface to a depth d, is used as a water supply tank, through which water infiltrates into the soil. The falling-head test is repeated without allowing the cylinder to be drained entirely to eliminate the effects of the initial soil moisture conditions. Either change in the water level (H) in the water supply cylinder or strong linearity between the infiltration flux (qs) and H can be used to determine Kfs directly or by multiplying the slope with an empirical ring-installation scaling length, Lg; thus, no soil-dependent variable is necessary. To obtain the Lg value for various conditions, we carried out numerical experiments using HYDRUS (2D/3D) with a recently implemented reservoir boundary condition for six different soil textures, ranging from coarse soil (e.g., sand) to fine soil (e.g., silty clay loam), and two different ranges of water levels, 15 ≤ H ≤ 25 cm and 2 ≤ H ≤ 6 cm. Numerical results showed that the Lg has a strong linear dependence on d and r, where Lg increases as d and r increase. We propose a new linear model to determine Lg with optimized coefficients for each H range. Numerical results showed no significant differences in the estimated Kfs after the second repetition, confirming that repeating the falling head test twice is sufficient. The newly obtained Lg values were then applied to the experimental data obtained at multiple fields having different soil textures. The Kfs values determined using the RFH method agreed well with those measured using the field two-ponding depths steady-state method and/or the FH method in the lab. These results demonstrate the reliability of the proposed RFH method.