Residence times for surface waters in altered granite area : example of the 'Furan' water catchment, east French Massif Central

Full text: The topic of this study is to understand the behaviour of surface waters in a small catchment area (ca. 3300 ha - altitudes ranging from 550 m to 1300 m). Six springs located in the southern part of the French Massif Central have been studied for two years. The 'Furan' watershed composed of metric arena overlays in a granitic and gneissic basement. Oxygen and hydrogen stable isotope ratios were determined routinely. Tritium, carbon isotopes as well as alkalinity were punctually determined. Cations and anions were quantified using ICP-AES and ion chromatography. Further more, to better understand stable isotopes catchment's behavior, three years of daily rainwater sampling provide a well define local meteoric water line (collection at 400 m asl). Moreover, two other LMWL were established, at 770 m and 1100 m to quantify spatial and altitudinal changes. Cations and anions content of springs are enriched regarding precipitation compositions (few mg/L), and the chemical content refers to low temperature reactions. Total dissolved ions range between 30 mg/L and 130 mg/L. Few alkalinity values were measured, from 1 to 7 mg/L as bicarbonate alkalinity, but there is no clear evidence that they match real values, sulphate content being higher. Ionic mass balance, based on pe redox equilibrium, were made using PhreeQC and appears to be good for half of the springs (< -10%). The other half gives in first approach bad values, leading to break in the ammonium ion. Such NH4+ content might be related to agricultural impact. Inverse modelling suggests interaction with minerals such as kaolinite, chlorite, K-feldspar, underlining reactions between percolating water and altered granites. From rainwater isotopic records, altitudinal variations are too small to modify both slope and D-excess of the LMWL. So the monthly mean of three years record is taken as representative of local rains and leads to δD=7.4δ18O+3.8 (r2=0.96). During the year 2003 occurred a 40% decrease of pluviometry / mean annual pluviometry: high isotopes values recorded for these summer months lead to a depletion of -0.4 for the slope and 3.5 per mille for D-excess. This may affect stable isotopes values for very near surface aquifers, and was slightly detected regarding sampled springs. Stable isotopes ratios measured on those springs show no clear annual cycle: winter months are not always associated with depleted ratios. Maximum variations are ± 1 per mille for δ18O vs SMOW and ± 10 per mille δD vs SMOW. Using isotopes ratios, these springs show an alignment along lines parallel to the LMWL, showing that they experiment little altitude and temperature gradient modifications along their flow path and that they infiltrate at heights ranging between 100 to 300 m above their outlet. Modelling residence times in such aquifers was estimated using FlowPC and O, H isotopes ratios. Two ways were examined: first taking the real precipitation input and second using elevated springs percolating downward. The first approach provides no good fit, despite an estimation of the evapotranspiration effect was taken into account (using Turc's formula). Soil effects (PCO2) on d18O isotopic composition seem to have first order importance (δ13CDIC from -16 to -22 per mille PDB, clearly related to C3 plant composition) and are in progress to improve input parameters. The second set of calculation eliminates such hypothesis and provides better results: transit time of ca. 2 months using dispersion model and 20% of aquifer stagnant water (efficiency > 75%), in agreement with the unique 2003 drought for 3 months. Coupling isotopic geochemistry and major ion chemistry appears to be a good way to access surficial fluid circulation in small water catchment. Recharge areas can be located at c.a. 200 to 300 m above spring discharge, and times of aquifer recharge can be estimated for each spring. The percolation of rainwater into soil and subsurface aquifers leads to poor-content waters, that experiment little chemical interactions with surrounding rocks (granitic bearing) before discharge. (author)