There was an error in the fluorimeter data computing for tracer test 1 (Q= 20 l/s) at site 4. Indeed, due to a misinterpretation of the fluorimeter recording, some data should appear as interpolated. This interpolation was based, namely, on the comparison with other tracer test results. Figure 4 is corrected here with the interpolated data appearing as a red dotted line (top graph, see `interpolated data'). In consequence, the recovery rate (39.9%) calculated in Table 1 is an estimation based on the interpolated data. A note has been added to Table 1. Correction: Hydrogeology Journal (2018) 26:641-650 https://doi.org/10.1007/s10040-017-1671-4