Figure legends

Figure 1: Map of the Can Vila catchment, showing the main land cover types and the location of the instruments and sampling sites used for this study.
Figure 2: Isotopic ratios of all the precipitation samples compressed into a single year period. Non-effective points represent the samples that were tentatively discarded as evapotranspired after the recharge model. Lines represent the seasonal variations according to the best fits of Eq(4).
Figure 3: Isotopic ratios of mobile soil and ground waters samples compressed into a single year period. Lines represent the seasonal variations according to the best fits of Eq(4), the dashed line showing that the amplitude A was not significantly different from 0 for VP01.
Figure 4: Variation of the isotopic ratios of ground water with water table level at VP05. The grey line separates the shallow and deep water levels analysed in Table 4.
Figure 5: Stream discharge (5-minute step) and isotopic ratios of stream waters during the recorded period.
Figure 6: δ 2H versus δ 18O plot of the precipitation, soil and stream waters, along with the Local Meteoric Water Line.
Figure 7: Variation in time-weighted young water fraction at the Can Vila catchment with increasing quantiles of the flow duration curve. The curve represents Eq. (8), using parameters obtained by fitting Eq. (9) to all the stream water δ18O isotope values. Maximum sampled discharge was 226 mm d-1. Vertical bars represent standard errors. (1) denotes a dimensionless variable. Reproduced from Gallart et al . (2019).
Figure 8: Time threshold for the definition of the young water fractions smaller than 1 shown in Figure 7, after Eqs. (6) and (7). Bars represent 90% confidence intervals. (1) denotes a dimensionless variable.
Figure 9: Flow-weighted young water fractions (\(\mathbf{F}_{\mathbf{\text{yw}}}^{\mathbf{*}}\)) and discharge sensitivities of young water fraction (Sd ) for 6 one-year windows and the full record available at Can Vila. Q represents discharge. Bars represent standard errors. (1) denotes a dimensionless variable.
Figure 10: comparison between relative cumulated time, flow and\(\mathbf{F}_{\mathbf{\text{yw}}}^{\mathbf{*}}\) simulated applying Eq. (8) to the 5-minute step flow record, as well as the relative cumulated flows for the dynamic and weekly sampling records. The 1:1 line is shown as a reference for a uniform distribution.