Scales of Water Retention Dynamics Observed in Eroded Luvisols from an Arable Postglacial Soil Landscape

Abstract

Core Ideas Hysteretic, seasonal, and interannual soil water retention dynamics were identified. Hydraulic functions for each drying and wetting event were described separately. Shorter term dynamics indicated soil structural and wettability changes. Longer term dynamics reflected soil management and erosion history. For each timescale, different processes control water retention dynamics. Soil water retention is frequently described by unique main drying curves measured in the laboratory on intact soil cores. In the field, however, soil pore structure changes as a result of swelling and shrinkage, wetting and drying, or tillage operations. For erosion‐affected arable soils characterized by truncated profiles, water retention dynamics could be even more complex. The objective of this study was to separate shorter term hysteretic from longer term seasonal dynamics in field‐measured water retention data of eroded Luvisols. Soil water content and matric potential data were from tensiometers and time‐domain reflectometry sensors of six lysimeter soil monoliths from two sites. For 2012 through 2014, drying and wetting periods were identified and fitted with the van Genuchten (VG) retention function. The results confirmed that drying water retention curves were steeper than those obtained in the laboratory. Steepness increased and fitted VG parameter values of saturated water content decreased within seasons, indicating that rewetting rates successively declined after each dry–wet cycle. Drying water retention curves returned to a similar level in each spring except for soils of transferred monoliths, where drying water retention tended to increase. Results suggested that water retention dynamics of eroded Luvisols was affected by continual incorporation of subsoil material in the Ap horizon and crop management practices in addition to hysteresis and seasonal dynamics. The disentangling of dry–wet cycles from time series of soil water content and matric potential was found useful for identifying the processes responsible for water retention dynamics and could eventually help improve flow and transport models.