Pore-Water Pressures Associated with Clogging of Soil Pipes: Numerical Analysis of Laboratory Experiments

Clogging of soil pipes due to excessive internal erosion has been hypothesized to cause extreme erosion events such as landslides, debris flows, and gullies, but confirmation of this phenomenon has been lacking. Laboratory and field measurements have failed to measure pore-water pressures within pipes and models of pipe flow have not addressed internal erosion or pipe clogging. The objective of this study was to model laboratory experiments of pipe flow in which clogging was observed in order to understand the clogging process. Richards' equation was used to model pipe flow, with the soil pipe represented as a highly conductive, low-retention porous medium. The modeling used two contrasting boundary conditions, constant flux (CF) and constant head (CH), to quantify pressure buildups due to pipe clogging and differences in simulated pressures between the two imposed boundary conditions. Unique to these simulations was inclusion of pipe enlargement with time due to internal erosion, representation of partially full flow conditions, and inclusion of pipe clogging. Both CF and CH boundary conditions confirmed the concept of pressure buildup as a result of pipe clogging. Pressure jumps of around 54 m for CF and 18 cm for CH occurred in <0.1 s, while soil water pressures 4 cm radially outward from the pipe had not responded. These findings demonstrate the need to measure pressures within soil pipes due to hydraulic nonequilibrium between the pipe and soil matrix. Pore water pressures within the pipe below the clog rapidly (<0.25 s) drained to unsaturated conditions, indicating the ability of soil pipes to drain hillslopes and rapidly recover when clogs are flushed from the soil pipe. These dynamic processes need to be incorporated into stability models to properly model hillslope processes.