Quantification of the Loess Plateau's soil hydrodynamics in relation to bulk density

Abstract

Investigating the effects of varying degrees of soil compaction on its hydrodynamic properties is still a vital step in optimizing water utilization. Furthermore, hydrodynamic parameters such as saturated hydraulic conductivity (Ks) and soil water retention characteristics (SWRC) are essential data for soil water and solute transport calculations. However, it takes a lot of time and money to get direct measurements of hydrodynamic properties. The purpose of this study was to measure how the Loess Plateau's SWRC, Ks, and soil pores were affected by five different degrees of bulk density (BD), and quantify interactions between BD, soil organic carbon (SOC), and particle size distribution (PSD) on hydrodynamic parameters using pedotransfer function (PTF). Hydrodynamic parameters were predicted using multiple linear regression (MLR), and the best models were chosen using statistical standards and compared with Rosetta3 models based on predictors % sand, silt, and clay (SSC) and SSC+BD. The results showed that increasing soil BD from 1.00 to 1.40 g cm⁻³ led to significant reductions in soil saturated water content (SSAT), quickly draining pores (QDP), and Ks. Enhances SOC content and clay from micropores under BD, and low SOC soil suffers pore collapse. MLR model-based (BD+SOC) predicted hydrodynamic parameters, and the models demonstrated that “BD+SOC” is the best combination. MLR-BD+SOC model outperformed (root mean square error [RMSE]: 0.001–0.005; and R²: 0.91–0.98) on Rosetta3 models. The Rosetta3-SSC+BD model improved predictions in low-SOC soils but underperformed in SOC-rich soils. These findings emphasize integrating BD and SOC in PTF for accurate hydrodynamic modeling, particularly in erosion-prone, heterogeneous landscapes.