Aquifer sampling using ball check-valves system: what about sample representativeness?
Résumé
Deep aquifer monitoring is a challenge in terms of representativeness of the sample. Such monitoring is often done when pumping is
not adequate due to: 1) too deep water table level, 2) no willing to add disturbance to the aquifer (low recharge rate) or 3) too difficult
management at surface of discharge waters. Deep sampling may rely on the use of sampling bottles but these have a fixed volume
and need to lower and to remove the bottle each time a sample is needed.
Ball check-valves and pressure gas drive samplers have been developed to overcome this bias such as the US U-tube system. Here
we present investigations with a system developed by the BRGM (GazOGaz, patent FR1259214), relying on similar functioning but
having more flexibility in the use. Such a system allows the collection of a virtually unlimited volume of water at any depth in the water
column.
Nevertheless, there is a frequent question about the use that may be done of waters collected using deep sampling. Site managers or
public authorities wonder about the comparisons that can be made with samples obtained using classical pumping methodology.
The present survey has been designed in that objective. By carefully describing physico-chemical parameters along the water column
and by performing both in-situ deep sampling and pumping at fixed depth, characteristics of waters existing in 2 boreholes located
near Castres (Tarn, France) are investigated in details with specific focus on the dissolved elements. Conclusions on how deep
sampling may be used and useful for aquifer characterization are presented together with the limits of applicability of such a sampling
procedure.
Domaines
Sciences de la Terre
Origine : Fichiers produits par l'(les) auteur(s)