A strontium isotopic study of mineral and surface waters from the Cézallier (Massif Central, France): implications for mixing processes in areas of disseminated emergences of mineral waters

Surface and mineral waters from a geothermal area (Cézallier) in the centre of the Massif Central, France, were analysed for Sr isotopes, Sr and major elements in order to use Sr isotopes to investigate the discharge of a mineralised water area, especially in the case of disseminate mineralised water emergences. The 87 Sr 86 Sr ratios of mineralised waters ranged from 0.715541 to 0.716773. The variation in 87 Sr 86 Sr values was small for the 5 mineral springs sampled over two periods (5 × 10 −6–14 × 10 −6). Regarding the reproductivity of the 87 Sr 86 Sr ratio measurements, no differences can be seen and the 87 Sr 86 Sr ratio of mineral waters can be considered to be constant with time. The 87 Sr 86 Sr ratios of surface waters collected along the Zagat River ranged from 0.705303 to 0.715091. The lowest values were always observed in the headwaters. The difference in the 87 Sr 86 Sr ratios between the two sampling periods was ∼ 22 × 10 −4 and can be related to the hydrological situation (low or high flow). The 87 Sr 86 Sr ratios of surface waters collected in the drain ranged from 0.715072 to 0.716585. The difference in the 87 Sr 86 Sr ratios between the two sampling periods is ∼ 11 × 10 −4 for the drain headwaters and ∼ 13 × 10 −5 at the outlet of this area. Likewise, these variations can also be related to the hydrological situation. Plots of 87 Sr 86 Sr vs. 1 Sr ratios show two distinct binary mixing trends and the existence of at least four end-members. A mixing parameter f giving the proportions of the two components can be estimated for assumed 87 Sr 86 Sr ratio and Sr content values of the pure end-members. An estimate of the mineral water discharge can be calculated using the results of the mixing model and the measured flow. In the drain area, the mixing model gives an estimate ∼ 0.5–1 l s −1, or 10–13% of mineral water in the total flow. A similar calculation was done for the river where the mineral water flow was ∼ 0.96 l s −1 (3.5 m 3 h −1 100 m downstream from the bridge in April 1994, and 6 m 3 h −1 20 m downstream from the bridge in September 1994.