Pyschik et al. (2026) Detecting the occurrence of preferential flow in soils with stable water isotopes

Identification

• Journal: Hydrology and earth system sciences
• Year: 2026
• Date: 2026-01-30
• Authors: Jonas Pyschik, Markus Weiler
• DOI: 10.5194/hess-30-485-2026

Research Groups

• Department of Hydrology, University of Freiburg, Freiburg, Germany

Short Summary

This study develops a novel, non-invasive method using stable water isotope profiles from soil cores to detect and characterize preferential flow pathways across hillslope and catchment scales, revealing significant spatial and vertical variability in flow features.

Objective

• Develop and apply a novel, non-invasive method using stable water isotope depth profiles to identify the occurrence and spatial variability of vertical and lateral preferential flow pathways in soils at hillslope and catchment scales.

Study Configuration

• Spatial Scale: Four first-order catchments in Germany (Sauerland, Black Forest, Ore Mountains) and Austria (Tyrolean Alps), covering diverse topographical, climatic, and land use conditions. Approximately 100 soil cores (6 cm diameter, 1–3 m deep) were drilled per catchment, totaling 393 profiles.
• Temporal Scale: Soil sampling campaigns were conducted seasonally across 2022–2023: July 2022 and March 2023 (Ore Mountains), May 2022 and August 2022 (Black Forest), June/July 2022 and October 2022 (Sauerland), and September 2022 (Tyrolean Alps). Samples were analyzed within four weeks of collection.

Methodology and Data

• Models used:
  • 3rd-degree polynomial regression (for individual soil profiles).
  • k-means clustering (from R caret package) for grouping polynomial coefficients.
  • LOESS regression (for generating reference profiles).
  • Two end-member mixing model (from R simmr package).
  • Logistic regression analysis (for spatial analysis of topographic and land use factors).
• Data sources:
  • Soil cores collected using an electrical auger, with samples taken at 10–20 cm depth intervals.
  • Stable water isotope analysis (δ²H and δ¹⁸O) of soil water using the direct vapor equilibration method and laser spectrometers (Picarro LX-i2130 and LX-i2120).
  • Digital Elevation Models (DEM, European Space Agency, 2024) with 30 m resolution for elevation, slope, and aspect.
  • Satellite-based land use data (ESA WorldCover 10 m 2020 V100, Zanaga et al., 2021).

Main Results

• Preferential flow (PF) was identified in 63 out of 393 total soil profiles (16%).
• The alpine catchment (Tyrolean Alps) exhibited the highest number of PF profiles (42), often with multiple PF depths (mean: 2.32 depths per identified profile).
• PF was predominantly detected at depths of 50–65 cm, but also frequently below 1 m.
• Mixing model results indicated that identified preferential flow consisted of approximately 65% new event water (topsoil signature) and 35% older pre-event water (reference signature), with no significant differences across catchments or sampling times.
• Spatial analysis revealed a statistically significant relationship (p=0.044) between PF occurrence and cosine-transformed aspect, suggesting north-facing slopes were more prone to PF, primarily driven by the alpine catchment. Other topographic and land use variables showed no significant correlations.
• A non-significant trend indicated shallower preferential flowpaths in forested areas compared to grasslands within the alpine catchment.

Contributions

• Introduces a novel, non-invasive, and cost-effective method for detecting preferential flow at catchment scales using stable water isotope depth profiles, overcoming limitations of traditional, more invasive techniques.
• Develops a robust approach for reconstructing "reference profiles" of matrix flow seasonality from observed soil water isotope measurements, even without long-term precipitation isotope data.
• Provides empirical evidence of the widespread occurrence and significant spatial and vertical variability of preferential flow pathways across diverse hydrological settings, particularly highlighting the role of pronounced isotopic seasonality in detection.
• Quantifies the mixing proportions of event and pre-event water within preferential flow, offering insights into the hydrological connectivity between surface infiltration and deeper soil layers.

Funding

• Deutsche Forschungsgemeinschaft (project no. 453746323) through the research unit FOR 5288: "Fast and Invisible: Conquering Subsurface Stormflow through an Interdisciplinary Multi-Site Approach".

Citation

@article{Pyschik2026Detecting,
  author = {Pyschik, Jonas and Weiler, Markus},
  title = {Detecting the occurrence of preferential flow in soils with stable water isotopes},
  journal = {Hydrology and earth system sciences},
  year = {2026},
  doi = {10.5194/hess-30-485-2026},
  url = {https://doi.org/10.5194/hess-30-485-2026}
}