Sorptivity function and water repellency determined by miniaturized infiltration data for characterizing soil water capture under different agricultural managements

Abstract

Improving soil water capture in the current climatic context is essential for agricultural production. Water sorptivity (S) as function of soil water content, together with water repellency characterization are two key hydraulic properties describing soil water capture, but also remaining poorly understood. The objectives of this work were: i- to determine the S under different soil water contents (θ) under different agricultural management; ii- to determine the soil water repellency at different soil water contents using the Beerkan Estimation of Soil Transfer algorithm (BEST) adapted for water repellent soils (BEST-WR), evaluating its performance as compared to the standard method of water repellency index (RI). Miniaturized tension infiltration experiments (1 cm infiltration radius) were conducted on undisturbed soil samples across a range of moisture contents under two contrasting agricultural managements (bare fallow, BF; cover fallow, CF). From miniaturized infiltration data, Sorptivity was determined using two different approaches, through the Leeds-Harrison et al. method (SLH) and using the BEST-WR algorithm (SBEST), which accounts for water repellency via an exponential scaling factor (αWR). Results showed that S was highest in dry conditions, remained relatively constant until approximately 0.25 m³m⁻³, and then decreased towards saturation. Regarding the effects of agricultural management, CF significantly increased S across the entire moisture range compared to BF, enhancing the soil's water capture capacity. Both treatments exhibited slight soil water repellency, but the inclusion of cover crops shifted the soil water content of maximum repellency toward higher values (0.34 m³m⁻³ for CF vs. 0.25 m³m⁻³ for BF), suggesting that management modifies the threshold of wettability rather than the overall degree of repellency. Furthermore, the parameter αWR obtained from the BEST-WR algorithm proved to be a sensitive indicator for differentiating hydrophobic behavior, particularly in dry conditions. Our results show that the use of miniaturized infiltration experiments with the adapted methodology of Leeds-Harrison et al. could be used for the S(θ) function determination in a relatively simple way on undisturbed soil samples. The determination of this relationship can be useful to understand the complex soil water dynamics under unsaturated conditions, complementing the basic soil functions.