Gao et al. (2025) Quantifying controls on rapid and delayed runoff response in double-peak hydrographs using ensemble rainfall-runoff analysis (ERRA)

Identification

• Journal: Hydrology and earth system sciences
• Year: 2025
• Date: 2025-11-19
• Authors: Huibin Gao, Laurent Pfister, James W. Kirchner
• DOI: 10.5194/hess-29-6529-2025

Research Groups

• Department of Environmental Systems Science, ETH Zurich, 8092 Zürich, Switzerland
• Swiss Federal Research Institute WSL, 8903 Birmensdorf, Switzerland
• Environmental Sensing and Modelling Research Unit, Luxembourg Institute of Science and Technology (LIST), 4422 Belvaux, Luxembourg
• University of Luxembourg, Faculty of Science, Technology, and Medicine, 4365 Esch-sur-Alzette, Luxembourg
• Department of Earth and Planetary Science, University of California, Berkeley, CA, 94720, USA

Short Summary

This study quantifies the controls of precipitation intensity and antecedent wetness on double-peak runoff generation in the Weierbach catchment using Ensemble Rainfall-Runoff Analysis (ERRA), demonstrating distinct thresholds for the rapid near-surface and delayed groundwater-mediated runoff components.

Objective

• To quantify how precipitation intensity and antecedent wetness influence groundwater recharge and double-peak runoff generation in the Weierbach catchment.
• To quantify the coupling between precipitation, groundwater recharge, and streamflow, including their temporal responses and how individual effects of correlated inputs differ.
• To determine how double-peak runoff response and its distinct peaks vary with changes in precipitation intensity and antecedent wetness conditions.

Study Configuration

• Spatial Scale: Weierbach experimental catchment (0.45 km²), a forested headwater catchment in Luxembourg, ranging from 450 to 500 m in elevation.
• Temporal Scale: September 2014 to December 2019 (5 years) with hourly time steps for all variables.

Methodology and Data

• Models used: Ensemble Rainfall-Runoff Analysis (ERRA; Kirchner, 2024a), a data-driven, model-independent, nonparametric approach combining least-squares deconvolution with de-mixing techniques and broken-stick regression.
• Data sources: High-frequency hydro-meteorological measurements from the Weierbach catchment (Hissler et al., 2020), including:
  • Precipitation (P) from Holtz rainfall monitoring station.
  • Water table depth (WTD) from three wells (GW2, GW3, GW5).
  • Volumetric soil water content (VWC) from five sites at 10, 20, 40, and 60 cm depths.
  • Streamflow (Q) at the catchment outlet.
  • Groundwater recharge (GR) calculated from WTD fluctuations and a drainable porosity of 10%.

Main Results

• The overall runoff response distribution (RRDP) exhibits a double-peak pattern: a sharp first peak (0.0063 h⁻¹) within the first hour and a broader, delayed second peak (0.0020 h⁻¹) peaking at approximately 37 hours after rainfall.
• The total runoff response can be quantitatively explained as the sum of two distinct pathways: a rapid near-surface pathway (direct precipitation to streamflow) dominating the first peak, and a delayed groundwater-mediated pathway (precipitation to groundwater recharge to streamflow) dominating the second peak.
• Under dry conditions (antecedent water table depth > 1.66 m), runoff response shows a single peak that increases nonlinearly with precipitation intensity, particularly above 4 mm h⁻¹. No clear second peak occurs.
• Under wet conditions (antecedent water table depth ≤ 1.66 m), double-peak hydrographs emerge. The first peak increases nonlinearly with precipitation intensity (especially above 2 mm h⁻¹), and the second peak becomes higher, narrower, and earlier with increasing precipitation intensity.
• Runoff response is negligible at precipitation intensities below approximately 0.8 mm h⁻¹ under wet conditions and 1.5 mm h⁻¹ under dry conditions.
• Groundwater recharge occurs even under dry conditions (at about half the rate of wet conditions), but its translation into streamflow is significantly less efficient, suggesting that groundwater loss to evapotranspiration and/or limited connectivity, rather than lack of recharge, explains the absence of a second peak in dry conditions.

Contributions

• Provides a novel quantitative framework using Ensemble Rainfall-Runoff Analysis (ERRA) to de-mix and quantify the contributions of near-surface and groundwater-mediated pathways to double-peak hydrographs.
• Quantifies the nonlinear and nonstationary controls of precipitation intensity and antecedent wetness on both the rapid and delayed runoff components, identifying specific thresholds for their initiation and amplification.
• Offers a data-driven, model-independent approach that avoids arbitrary hydrograph separation, enhancing the understanding of complex runoff generation processes.
• Contributes to resolving the "old water paradox" by clarifying the mechanisms and conditions under which stored pre-event water is released to streamflow, particularly through the delayed second peak.

Funding

Not explicitly mentioned in the provided text.

Citation

@article{Gao2025Quantifying,
  author = {Gao, Huibin and Pfister, Laurent and Kirchner, James W.},
  title = {Quantifying controls on rapid and delayed runoff response in double-peak hydrographs using ensemble rainfall-runoff analysis (ERRA)},
  journal = {Hydrology and earth system sciences},
  year = {2025},
  doi = {10.5194/hess-29-6529-2025},
  url = {https://doi.org/10.5194/hess-29-6529-2025}
}