Abstract. Alpine streams are particularly valuable for downstream water resources and for ecosystem conservation. However, the details of where and when water is stored and released in the heterogeneous mountain environment are rarely known. The use of physico-chemical flow path tracers is particularly challenging due to the temporary accumulation and storage of water in the form of snow and ice. Alternatively, biological tracers might complement information on flow and storage of water, especially as the different microhabitats in Alpine aquatic systems are inhabited by characteristic organismal communities. In this study, we explored the potential of particles of environmental DNA found in the water (eDNA) to characterize hydrological flow paths and connectivity in an Alpine catchment in Switzerland. Between March and September 2017, we sampled water at multiple time points at 11 sites distributed over the 13.4 km² Vallon de Nant catchment for genetic species information based on naturally occurring eDNA. The sites correspond to three different water source types and habitats (main channel, tributaries, and springs).

Comparison of typical hydrological tracers and eDNA with temporal evolution of streamflow revealed that in the main channel and in the tributaries, the change in streamflow, dq/dt, is strongly correlated with biological richness. In springs, electrical conductivity was found to have a positive but not as strong correlation with biological richness. At the catchment scale, our results show that biological richness as indicated by the diversity detected by eDNA samples. When streamflow is increasing, transport of additional, and probably terrestrial, DNA into water storage or flow compartments is occurring. Such processes include overbank flow, stream network expansion and retraction, and hyporheic exchange. In general, our results highlight the importance of considering the at-site sampling habitat in combination with upstream connected habitats to understand how streams integrate eDNA over a catchment and to interpret spatially distributed eDNA samples, both for hydrological and biodiversity assessments. We identify next steps to be addressed to use eDNA as an independent tracer of Alpine water sources and we provide recommendations for future observation of eDNA in Alpine stream ecosystems.