A comparison of undisturbed soil cores with restructured columns using a new one-step outflow model

Abstract

Undisturbed soil cores may be best for studying field processes in the laboratory, but core variability can make them difficult to use. The goal of this study was to develop a method to make packed columns that have one-step outflow curves similar to those of undisturbed cores. Undisturbed 15.2-cm-long cores and bulk soil were obtained from a sandy loam permanent pasture (Pasture) and a sandy clay loam uncultivated field (Springhill). Ordinary-packed columns were made by compressing bulk soil to the average core bulk density, and layered columns were made by duplicating the average bulk densities of three core layers. Outflow was recorded from pressure head of −7.6 to −133 cm from newly packed columns and after drying to produce shrinkage cracks. A new outflow model that included initial water content (θ_i), final water content (θ_f), and time to reach 63.2% outflow (Τ) was fit to the data by adjusting a time exponent β (average r² = 0.995). Column parameters were compared with cores by equivalence testing. Drying reduced θ_i and θ_f, possibly due to shrinkage cracks that increased drainable porosity. Columns of the Pasture soil had significantly greater Τ than natural cores, possibly due to lack of root-formed pores in the columns. Overall, layered and dried columns were most similar to natural cores and less variable, and cumulative one-step outflow curves from undisturbed cores and packed columns can be quantified with a simplified four-parameter model. These methods and results may help make packed soil columns more representative of undisturbed soil.