Discussion of “Laboratory Modeling of the Mechanisms of Piping Erosion Initiation” by Mandie S. Fleshman and John D. Rice

Abstract

Evaluating piping in cohesionless soils remains an important and difficult problem when determining the safety of water-retaining structures. The more published piping research testing the better, especially when the researchers carefully describe some of the test- ing details as the authors have done and as Fleshman (2012) did more completely. They have presented a laboratory study of the effects on sand of saturated, vertical, upward water flow to a hori- zontal exit surface. They tested a variety of sands in a 51-mm diameter, 127-mm high cylindrical plexiglass sample holder and observed sand behavior while increasing the vertical seepage gra- dient. However, they also reported heaving, boiling, and quicksand behavior at vertical gradients higher, and, therefore, nonconserva- tive, versus Terzaghi’s critical gradient icr [authors’ Eq. (1)]. Eq. (1) expresses the static equilibrium of vertical forces during vertical-up seepage; namely the buoyant weight of sand = γ 0V versus the seep- age force = icrγwV wherein V = volume. Herein the reader will find brief discussions of the special nature of the piping investigated, the use of larger quicksand tanks to obtain similar information, and the range of vertical gradients that produce sand boils in the field. Sand boils indicate near-surface piping. Finally, this discusser believes that sample side shear, the obstruction of the three pore pressure probes to water and particle movement, and the resulting arching in the authors’ tests caused their higher gradients. Fleshman (2012, p. 17–18 and Fig. 4) demonstrated this in some test results not included in the authors’ paper.