Evaluating the rate of leakage through defects in a geomembrane

Geomembranes are impermeable barriers when they remain intact. However, there is potential for damage to occur during and after installation. The development of defects in a geomembrane is critical when evaluating a barrier system for quality control and quality assurance purposes. Tears and punctures are common defect types that have been evaluated extensively in literature. The majority of experimental procedures to date used small-scale permeameters to develop empirical equations for flow through a defect. Thus it is difficult to firmly conclude the expected flow rate through defects in field-like conditions. The main objective of this research project was to evaluate flow through a geomembrane defect using a large-scale constant head apparatus. A large-scale constant head apparatus was developed to allow for the measurement of flow through a geomembrane defect of known dimensions when overlain by soil. This research only evaluated flow rate through a defect with sand overlying the geomembrane. A zero pressure boundary condition was maintained by allowing the water to flow freely for collection, thus representing a system similar to that of a secondary drainage layer. The specific objectives of the research were to: i) compile a database of experimental results obtained through the use of the large scale apparatus for various defect shapes and sizes, geomembrane thicknesses, and applied slopes; ii) develop 3D numerical model representative of the experimental geomembrane flow system to evaluate the underlying phenomena; iii) assess the effects of a geomembrane thickness and slope on the flow rate through the selected geomembrane defect; and iv) to provide correction factors for the selected defects based on experimental and numerical results. The results of the physical model for circular defects were compared to both the applicable empirical solution and finite element models (FEM). A 2-dimensional axisymmetric FEM was developed as an approximation to the experimental results for circular defects without a horizontal hydraulic gradient or slope. A 3-dimensional FEM was developed for all geomembrane defects tested experimentally. Currently there are no empirical equations proposed for geomembrane systems with elongated defects or defects on a sloped geomembrane surface. The experimental flow rates were most similar to the 3D FEM results, followed by the 2D FEM results, then Bonaparte et al.’s (1989) empirical estimations, respectively. Both of the finite element models and empirical estimations overestimated experimental flow rates through circular defects. The 3D FEM models resulted in high localised hydraulic gradients immediately surrounding the defect. These findings support the assumption that the inconsistencies between the experimental results with the empirical and numerical estimations are due to high velocity, non-Darcian flow properties occurring immediately around the defect. FEMs cannot examine the flow regimes occurring within the pore-spaces in the material immediately near the defect. Further examination is required to explicitly determine that the overestimation of flow rates between the experimental results and the empirical and numerical estimations is the result of high velocity, non-Darcian flow effects