LABORATORY SORPTIVE AND NON-REACTIVE TRACER TESTS IN A HIGHLY HETEROGENEOUS SYNTHETIC AQUIFER

Tracer test techniques are a valuable tool frequently used in the field to estimate flow and solute transport aquifer parameters. By monitoring concentrations with time of a solute that is injected at a known location (or locations) into the aquifer it is possible to infer aquifer parameters. However, practitioners of this technique encounter several difficulties. Parameters describing the migration of solutes in the subsurface, namely the dispersion and the retardation coefficients, are scale-dependent, questioning the validity of the classical sorptive-advection-dispersion equation at the field-scale. The mechanisms governing the migration of sorptive solutes in the subsurface at scales larger than the laboratory-scale does not seem to be fully understood yet. For this reason, several large-scale field tracer tests described in the literature have been aimed to gain a better understanding of the underlying sorptive and mixing processes governing solute transport in the field. Though observations from this type of field experiments have enormously improved our knowledge on solute transport in natural field settings, they are hindered by undesirable uncontrolled processes that makes the analysis much more complex. In order to have a better control of all the variables that influence the sorptive and mixing processes governing solute transport, several tracer tests were conducted in a three-dimensional sand tank packed with a highly heterogeneous medium. The heterogeneous medium was constructed by distributing five different types of silica sand in a regular rectangular mesh through out 30-layers. At the laboratory-scale, hydraulic, mixing and sorptive properties of all the systems formed by combining any tracer chemical specie and any sand type are well characterized such that the sorptive and hydraulic spatial variations in the tank are well known. Experiments were designed to evaluate differences in the upscaling behavior of dispersion and retardation coefficients between forced-gradient tracer tests involving a small source and uniform-gradient tracer tests involving a large source. In the uniform-flow case, special attention was paid to the behavior of retardation as a function of distance from the source.