Search

Karst aquifers are known for their wide distribution of water transfer velocities. From this observation, a multiple geochemical tracer approach seems to be particularly well suited to provide a significant assessment of groundwater flows, but the choice of adapted tracers is essential. In this study, several common tracers in karst aquifers such as physicochemical parameters, major ions, stable isotopes, and δ 13 C to more specific tracers such as dating tracers – 14 C, 3 H, 3 H– 3 He, CFC-12, SF 6 and 85 Kr, and 39 Ar – were used, in a fractured karstic carbonated aquifer located in Burgundy (France). The information carried by each tracer and the best sampling strategy are compared on the basis of geochemical monitoring done during several recharge events and over longer time periods (months to years). This study’s results demonstrate that at the seasonal and recharge event time scale, the variability of concentrations is low for most tracers due to the broad spectrum of groundwater mixings. The tracers used traditionally for the study of karst aquifers, i.e., physicochemical parameters and major ions, efficiently describe hydrological processes such as the direct and differed recharge, but require being monitored at short time steps during recharge events to be maximized. From stable isotopes, tritium, and Cl − contents, the proportion of the fast direct recharge by the largest porosity was estimated using a binary mixing model. The use of tracers such as CFC-12, SF 6 , and 85 Kr in karst aquifers provides additional information, notably an estimation of apparent age, but they require good preliminary knowledge of the karst system to interpret the results suitably. The CFC-12 and SF 6 methods efficiently determine the apparent age of baseflow, but it is preferable to sample the groundwater during the recharge event. Furthermore, these methods are based on different assumptions such as regional enrichment in atmospheric SF 6 , excess air, and flow models among others. 85 Kr and 39 Ar concentrations can potentially provide a more direct estimation of groundwater residence time. Conversely, the 3 H– 3 He method is inefficient in the karst aquifer for dating due to 3 He degassing. [ABSTRACT FROM AUTHOR]