ABSTRACT
One of the most common techniques used to evaluate the aquifer transport properties is the well-known two-well injection-withdrawal tracer test. A radially converging flow field is created by a pumping well. A pulse of tracer-labelled water is introduced into the groundwater through the injection well. The time evolution of the tracer concentration is monitored at the pumping well. The resulting breakthrough curve can be interpreted with analytical or numerical solutions to evaluate the transport properties. Most often, the injection is supposed to be instantaneous. Such an assumption allows to interpret the breakthrough curve as the impulse response of the aquifer-piezometer system.
As mentioned by many authors, the way the injection is conducted can differ dramatically from the usual Dirac-type pulse of tracer, due to the complex interaction between the piezometer equipment and the aquifer material. This can have a critical influence on the shape of the breakthrough curve leading at the extreme to a double-peak. The interpretation of such a result disregarding the way the injection was conducted could prove erroneous.
Different tracer tests performed in the alluvial plain of the river Meuse in Belgium have shown double-peaked breakthrough curves. One of the tracer tests was selected to be replicated under different injection conditions. The breakthrough curves did not present two peaks anymore.
After some theoretical and field considerations, the tracer test results are analysed and criticised. It is shown how important it is to conduct further investigation in order to gain a better understanding of the influence of the injection mode on the breakthrough curve.
