Effect of matrix preferential flow on seepage based on runoff and infiltration numerical models

Seepage in unsaturated soil slopes under variable rainfall is a critical issue in both geotechnical and hydraulic engineering. Previous research has primarily focused on either hydrological modeling or stress analysis using the single permeability model (SPM). However, the interaction between pore w...

Full description

Bibliographic Details
Main Author: Han Xinkai
Format: Thesis
Language:English
Published: Universiti Teknologi Malaysia 2026
Subjects:
Online Access:https://utmik.utm.my/handle/123456789/190864
Abstract Abstract here
Description
Summary:Seepage in unsaturated soil slopes under variable rainfall is a critical issue in both geotechnical and hydraulic engineering. Previous research has primarily focused on either hydrological modeling or stress analysis using the single permeability model (SPM). However, the interaction between pore water pressure (PWP) and effective stress is a coupled phenomenon that is often overlooked, particularly in unsaturated soils. Furthermore, the complexity of water transport via preferential flow (PF) in heterogeneous soils remains inadequately understood. To address this gap, this study introduces a coupled SPM to examine the interactions between PWP and effective stress. Additionally, a dual permeability model (DPM) based on an integrated saturated–unsaturated subsurface flow framework was developed to more accurately simulate seepage behavior by coupling it with a dual permeability with runoff model (DPWRM), recognizing the critical role of surface runoff in the overall seepage water balance. Initially, a series of laboratory tests were conducted to characterize the soil properties to be used in numerical simulations. These tests included compaction, permeability, shear strength, and construction of soil-water characteristic curve (SWCC) to define the relationship between soil suction and water content. Mathematical models of the SPM, DPM, and DPWRM were developed to account for the interrelationship between PWP and effective stress. The obtained experimental data were used as input parameters for numerical simulations performed using COMSOL Multiphysics, to ensure realistic modeling of seepage phenomena. The numerical analysis using SPM revealed that precipitation primarily infiltrates the uppermost soil layer before draining toward the base of the slope. The PWP variation was most significant near the surface, especially under different rainfall intensities. A comparison of effective saturation between matrix flow (MF) alone and a combination of MF (2 mm soil) with PF (2 mm sand) showed that PF enhances soil permeability and drainage capacity in the SPM. This study also quantified rainfall-induced seepage in DPM by analyzing the temporal and spatial influences of PF on hydrological mechanisms. Water mainly flowed from the MF to the PF domain. Compared to DPM, SPM showed a gradual PWP increase at the slope top. The PWP in the MF rose gradually, with a delayed rise at the slope top. The PWP in the PF domain rose rapidly but quickly reached a constant value in the slope-top area. Compared to the SPM model, the DPM model offered a more accurate and dynamic prediction of pore water pressure during seepage simulations under complex rainfall and heterogeneous soil conditions. The runoff velocity increased rapidly during the first 95 minutes, then stabilized at 0.323 m/s, with a cumulative volume of 5.11 m³ after 180 minutes in the DPWRM. PWP demonstrated a pronounced rise near the surface in both the crest and middle slope regions in DPWRM, followed by attenuated increases at deeper layers. The findings of this study offer new insights into rainfall-induced slope seepage behavior and establishes a valuable numerical framework for assessing hydrological mechanisms in geotechnical engineering applications.