Technical Note: two-component Electrical Conductivity-based hydrograph separaTion employing an EXPonential mixing model (EXPECT) provides reliable high temporal resolution young water fraction estimates in three small Swiss catchments

Abstract. The young water fraction represents the fraction of water molecules in a stream that have entered the catchment relatively recently, typically within 2–3 months. It can be reliably estimated in spatially heterogeneous and nonstationary catchments from the amplitude ratio of seasonal isotope (δ¹⁸O or δ²H) cycles of streamwater and precipitation, respectively. It has been found that young water fractions increase with discharge, thus reflecting increased direct runoff with wetter catchment conditions. This so-called discharge sensitivity of the young water fraction (S^*_d) can be useful for describing and comparing catchments’ hydrological behaviour; however, the estimation of S^*_d can be highly uncertain and unreliable when the streamwater isotope data are sparse and don’t capture the entire flow regime.

Here, we present a new method that can increase the temporal resolution of the young water fraction estimates, and thus better constrain the estimation of S^*_d. Our so-called EXPECT method is built upon three key assumptions: 1) the two-component hydrograph separation technique can be used to obtain the portion of young water and old water in a stream by considering EC as a proxy of the water age, 2) the EC value of the young water endmember (EC_yw) is lower than that of the old water endmember (EC_ow), and 3) the mixing of young water and old water fractions is described assuming an exponential decay of electrical conductivity with increasing young water fraction. We calibrate the two endmembers, EC_yw and EC_ow, by constraining the time-weighted and flow-weighted average young water fraction achieved with hydrograph separation to be equal to the same quantities obtained from seasonal isotope cycles.

We test the EXPECT method with data from three small experimental catchments in the Swiss Alptal valley by using two different temporal resolutions of Q and EC data: sampling-resolution (i.e., we only consider Q and EC measurements during dates of isotope sampling) and daily-resolution. By leveraging high-resolution and low-cost EC measurements and bi-weekly isotope data, the EXPECT method has provided reliable young water fraction estimates at bi-weekly and daily resolution, from which S^*_d could be determined with higher accuracy compared to the existing method that uses only bi-weekly isotope data. For proper use of the EXPECT method, we further highlight its main limitations that may vary in their relevance depending on the characteristics of the catchments under study.