Interpreting tracer breakthrough tailing in a conduit-dominated karstic aquifer

Breakthrough tailing has been observed during dye-tracing recovery tests in the Norville aquifer system (chalk), France. Karst-conduit flow and transport parameters were assessed using two different interpretative methods: the linear graphical method and the Chatwin method (implemented in the Qtracer2 program). The linear graphical method was used to model the observed tailing effects, which was explained by a second smaller delayed breakthrough curve. By comparing the results of tracer-test interpretation for the two methods, it was possible to relate the area of this second curve to the importance of turbulent flow in spring discharge. The more turbulent the flow, the less important the contribution of the second breakthrough curve and the tailing effect. The observed tailing could possibly be controlled by hydrodynamics to a greater extent than usually expected, the tailing effects being mostly attributed to diffusion phenomena. Tailing effects were expected to increase with discharge and the piezometric level, which would have resulted in overpressure in conduits, fissure flooding, etc. Instead, breakthrough tailing tended to disappear with increasing aquifer discharge, which would support the hypothesis of there being mostly hydrodynamic-controlled tailing effects instead of matrix- or fissure-diffusion.