Answers are grounded in this paper and cite their sources.
Relative unsaturated hydraulic conductivity (Krw) and water retention are key soil hydraulic properties that control water and solute transport. Typically, Krw is predicted from the water retention curve (WRC) using models such as van Genuchten–Mualem (vGM), because its direct experimental determination is demanding. X-ray computed tomography (XCT) combined with pore-scale modeling provides an alternative approach by explicitly representing the resolved pore geometry and connectivity. Here, pore-network models (PNMs) were extracted from four digital subvolumes obtained from two physical cores representing the Ah and B horizons of a Haplic Greyzem. PNM-derived WRCs and relative hydraulic conductivity curves were compared with the corresponding vGM descriptions. Because direct measurements of unsaturated hydraulic conductivity were unavailable, the conductivity comparison was used as a diagnostic model intercomparison rather than as an experimental validation. For WRCs, discrepancies were quantified at common water saturations using bias, MAE, and RMSE in log10(Pc) space and were evaluated separately for low (Sw<0.2), intermediate (0.2≤Sw<0.8), and near-saturation (Sw≥0.8) ranges. An illustrative reduction of the smallest pore-throat radii decreased the low-saturation RMSE by 45.9%–62.1% and the intermediate-saturation RMSE by 9.8%–54.1%, but reduced the near-saturation RMSE by only 1.7%–5.0%. Thus, unresolved small pore throats substantially contribute to discrepancies at low and intermediate saturation, whereas the near-saturation discrepancy cannot be explained by image-resolution limitations alone. These results demonstrate both the potential and the present limitations of single-scale XCT-based pore-network modeling of soil hydraulic properties. Quantitative prediction across the full saturation range will require multiscale structural information, improved representation of unresolved porosity, and validation against direct measurements of unsaturated hydraulic conductivity.
No takes yet. Share an insight, caveat, or question.
Gerke et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: