Experimental investigation of infiltration ponding in one-dimensional sand-geotextile columns

The paper describes the results of laboratory tests that were carried out to quantify the geotextile water characteristic curves of a woven and nonwoven geotextile in new and modified conditions. The geotextiles were modified by contamination with kaolin powder. The specimens were placed in 2 m high, one-dimensional sand columns and subjected to constant head water infiltration loading. Conductivity probes and tensiometers were used to record the unsaturated saturated response of the sand-geotextile layers and a control column with sand only. The test results showed a detectable delay in infiltration front advancement and ponding level, which increased with decreasing saturated permeability of the geotextile specimen. The use of geotextile water characteristic curves from in-air laboratory suction plate tests is shown to underestimate water ponding heights in the columns, which means that their use for design may be non-conservative. Another practical implication of the work reported here is that for geotextile-sand systems under surface water infiltration loading, current filtration design criteria that call for the ratio of saturated hydraulic (conductivity) permeability of the geotextile to that of the surrounding sand be at least one, may need to be strengthened. For ratios of 1 and 10, the estimated ponding height above the geotextile was estimated to be 4 and 0.1 cm, respectively. Hence, ratios of saturated hydraulic permeability greater than one may be necessary to prevent lateral flow of water above a geotextile in sand fills used in wall, slope and road applications.