Türk et al. (2026) Catchment transit time variability with different SAS function parameterizations for the unsaturated zone and groundwater

Identification

• Journal: Hydrology and earth system sciences
• Year: 2026
• Date: 2026-02-20
• Authors: Hatice Türk, Ansgar Kahmen, Markus Hrachowitz, Peter Strauss, Günter Blöschl, Michael Stockinger
• DOI: 10.5194/hess-30-1053-2026

Research Groups

• BOKU University, Institute of Soil Physics and Rural Water Management, Department of Landscape, Water and Infrastructure, Vienna, Austria
• Department of Water Management, Faculty of Civil Engineering and Geosciences, Delft University of Technology, Delft, the Netherlands
• Institute for Land and Water Management Research, Federal Agency for Water Management, Petzenkirchen, Austria
• Vienna University of Technology, Institute of Hydraulic Engineering and Water Resources Management, Vienna, Austria

Short Summary

This study investigated whether stable water isotope (δ2H) measurements in streamflow can effectively represent preferential flow in the unsaturated zone and groundwater using StorAge Selection (SAS) functions. It found that δ2H data are sensitive to preferential flow in the unsaturated zone but insufficient to constrain groundwater preferential flow due to the damping effect of large passive groundwater storage.

Objective

• To test if stable water isotope (δ2H) measurements in streamflow can implicitly represent preferential flow in the unsaturated root zone and groundwater aquifer within a conceptual catchment-scale transport model.
• To quantify the influence of preferential flow representation on transit time distributions (TTDs).
• To assess if tracer data variability is sufficient to characterize preferential groundwater flow using different StorAge Selection (SAS) function parameterizations.
• To examine how groundwater mixing assumptions and passive storage volumes affect tracer signal fit and TTD estimations.

Study Configuration

• Spatial Scale: Two headwater catchments: Hydrological Open Air Laboratory (HOAL), Austria (66 hectares), and Wüstebach catchment, Germany (38.5 hectares).
• Temporal Scale: HOAL: October 2013–2019 (calibration: October 2014–2019); Wüstebach: October 2009–2013 (calibration: October 2010–2013). Daily time steps for modeling, transit times tracked up to 1000 days.

Methodology and Data

• Models used: Process-based tracer transport model (Türk et al., 2025) based on the DYNAMITE (DYNAmic MIxing Tank) framework (Hrachowitz et al., 2014), integrating StorAge Selection (SAS) functions (Rinaldo et al., 2015) parameterized with beta distributions.
• Data sources: Long-term daily hydro-meteorological data (precipitation, streamflow) and stable hydrogen isotope (δ2H) measurements in precipitation and streamflow. Data collected from HOAL (Austria) and Wüstebach (Germany) catchments using rain gauges, flumes/weirs, and automatic/grab samplers. Isotopic analysis performed using cavity ring-down spectroscopy.

Main Results

• Streamflow δ2H data were sufficiently sensitive to characterize preferential flow in the unsaturated root zone, with calibrated SAS shape parameters (α0) reflecting distinct flow activation mechanisms in HOAL (0.14, strong young-water preference, precipitation intensity and soil wetness dependent) and Wüstebach (0.98, slight young-water preference, soil wetness dependent).
• Streamflow δ2H signals showed limited sensitivity to variations in groundwater StorAge Selection (SAS) function parameterizations, yielding consistently strong correlations (Spearman r: 0.54–0.60 for HOAL, 0.71–0.76 for Wüstebach) across different groundwater SAS shapes.
• Despite similar tracer signal fits, different groundwater SAS function parameterizations resulted in considerable uncertainty in Transit Time Distribution (TTD) estimates for older water (100 days < T < 1000 days), with a variability of approximately ±20% for HOAL and ±23% for Wüstebach in the fraction of streamflow older than 100 days.
• Increasing passive groundwater storage volumes significantly dampened streamflow δ2H variability and shifted TTDs towards older ages, requiring passive storage volumes orders of magnitude larger than active storage (e.g., 500 mm in HOAL, 5000 mm in Wüstebach) to match observed attenuation.
• The majority of streamflow was relatively old: 13% younger than 1000 days in HOAL (15% wet, 10% dry) and 27% younger than 1000 days in Wüstebach (35% wet, 20% dry).

Contributions

• Systematically evaluated the sensitivity of streamflow δ2H signals and inferred transit times to various StorAge Selection (SAS) function parameterizations for both the unsaturated root zone and groundwater compartments within a conceptual catchment-scale transport model.
• Quantitatively demonstrated that streamflow stable water isotope data are sufficient to identify and characterize preferential flow in the unsaturated zone but insufficient to constrain preferential groundwater flow in catchments with substantial passive groundwater storage.
• Quantified the significant uncertainty (±20% to ±23% for T > 100 days) in groundwater transit time distribution (TTD) estimates introduced by different groundwater SAS function parameterizations and passive storage volumes, even when tracer signal fits are comparable.
• Underscored the critical role of passive groundwater storage as a dominant control on catchment memory, masking young-water contributions and delaying solute transport, and highlighted the need for complementary data to reduce uncertainties in groundwater TTDs.

Funding

• Austrian Science Fund (FWF – Österreichischer Wissenschaftsfonds) [Grant No. 10.55776/P34666].
• Doctoral School “Human River Systems in the 21st Century (HR21)” of the BOKU University, Vienna.

Citation

@article{Türk2026Catchment,
  author = {Türk, Hatice and Kahmen, Ansgar and Hrachowitz, Markus and Strauss, Peter and Blöschl, Günter and Stockinger, Michael},
  title = {Catchment transit time variability with different SAS function parameterizations for the unsaturated zone and groundwater},
  journal = {Hydrology and earth system sciences},
  year = {2026},
  doi = {10.5194/hess-30-1053-2026},
  url = {https://doi.org/10.5194/hess-30-1053-2026}
}