Relations between the structure of storage and the transport of chemical compounds in karstic aquifers

Study of the movement of chemical compounds naturally present in the water, or which result from pollution, are examined according to the reservoir structure in karstic aquifers. Structure is represented by a simple geometrical model; slow flow takes place in blocks with a network of low-permeability cracks. The blocks are separated by highly permeable karstic conduits that allow rapid flow, and these form the aquifer drainage system. The karst studied covers 110 km 2. It is fed by an interrupted stream draining a 35 km 2 non-karstic basin, contaminated at the entry to the karst by effluents from a sewage treatment station. The underground water reappears as a resurgence with an annual average flow of approximately 1 m 3 s −1, after an apparent underground course of 8 km in the karst. Several local sources of pollution (effluent from septic tanks) contaminate the underground water during its course. Sixteen measurement operations were performed at 12 water points, between the interrupted stream and the spring. Some sampling points were at drains, and others were in the low-permeability fissured blocks. Comparison at each point of the concentrations of 14 chemical compounds gave the following results: when pollutant discharge occurs in a permeable zone, movement is rapid in the drainage network formed by the karstic conduits, and does not reach the less permeable fissured blocks which are thus protected; however, if discharge is in a low-permeability zone, the flow does not allow rapid movement of the polluted water, and this increases the pollutant concentration at the discharge. This simple pattern can be upset by a reversal of the apparent piezometric gradient between a block and a conduit during floods or pumping; this may reverse flow directions and hence modify the movement of contaminants. The study made it possible to site five boreholes whose positions in the karstic structure were unknown, showing the interest of such an approach for the forecasting of the impact of potential pollution.