Predicting soil interpedal macroporosity and hydraulic conductivity dynamics: A model for integrating laser-scanned profile imagery with soil moisture sensor data

The size and spatial distribution of soil pores control the infiltration, percolation, and retention of water within a pedon. These distributions are often represented within hydrologic flux equations as static hydraulic properties such as saturated hydraulic conductivity and water retention parameters. However, the assumption that these hydraulic properties are static does not adequately represent the potentially rapid response of highly-structured soil to moisture variability-induced shrink-swell processes.  We use a recently-developed, high-resolution (180 um) laser imaging technique to capture structural macropore data and derive a function that relates interpedal, planar macropore width to matrix water content. Subsequently, we develop an expression for transient hydraulic conductivity that accounts for dynamic macropore geometries and propose a method for partitioning total soil water content obtained from in situ sensor data into matrix and macropore water content. The model was applied to a soil profile in northeastern Kansas where intact soil monoliths had been imaged to quantify soil macorpore properties and continuous soil water content data were collected at multiple depths. Model-predicted macropore width showed significant sensitivity to matrix water content. Rainfall events that followed periods of low soil moisture were predicted to allow water to fill macropores - created by the shrinkage of soil structural units - which significantly and rapidly increased unsaturated hydraulic conductivity. This model offers a means by which to monitor and characterize the dynamic hydraulic properties of soils susceptible to shrink-swell processes that impact hydrologic partitioning and preferential flow.

Methods
We used soil moisture sensor data, measured soil physical properties (particle-size distribution, bulk density, water retention, and coefficient of linear extensibility), and macropore data generated from multistripe-laser triangulation scanned images of intact soil monoliths taken from 3 horizons of an agricultural soil at the Konza Prairie Biological Station near Manhattan, KS, USA to test a newly developed theory that (1) links properties of soil macropores obtained at one moisture state to time series of soil moisture such that macropore properties below the surface can be predicted through time at any moisture state; (2) partitions soil water content into macropore and matrix water contents; and (3) predicts both saturated and unsaturated soil hydraulic conductivity in a dynamic dual porosity system. Soil moisture data were obtained from installed sensors (ECH2O 5TM, METER Group, Pullman, WA) at a depths of 10, 40, and 120 cm and recorded on a data logger (CR1000X, Campbell Scientific, Logan, UT). Soil properties were measured by the USDA-NRCS Kellogg Laboratory using standard procedures (Soil Survey Staff, 2022) [Soil Survey Staff. 2022. Kellogg Soil Survey Laboratory methods manual. Soil Survey Investigations Report No. 42, Version 6.0. U.S. Department of Agriculture, Natural Resources Conservation Service.]. Coefficient of linear extensibility was measured using a modified procedure from Schafer and Singer (1976) [Schafer, W. M., & Singer, M. J. 1976. New method of measuring shrink-swell potential using soil pastes. Soil Science Society of America Journal, 40 (5), 805-806. doi: 10.2136/sssaj1976.03615995004000050050x]. Macropore data were generated following Eck et al. (2013) [Eck, D. V., Hirmas, D. R., & Gimenez, D. 2013. Quantifying soil structure from field excavation walls using multistripe laser triangulation scanning. Soil Science Society of America Journal, 77 (4), 1319-1328. doi:10.2136/sssaj2012.0421]. Analyses were conducted in R 4.4.2 (R Core Team, 2024) [R Core Team. 2024. R: A language and environment for statistical computing [Computer software manual]. Vienna, Austria. Retrieved from https://www.R-project.org/].