Lei et al. (2026) Discharge-based classifications of spatio-temporal patterns of potentially gaining and losing subcatchments in the Bode River catchment, Central Germany

Identification

• Journal: Journal of Hydrology Regional Studies
• Year: 2026
• Date: 2026-01-22
• Authors: Chao Lei, Stefano Basso, Matthew J. Cohen, Andréas Musolff, Larisa Tarasova, Christian Schmidt
• DOI: 10.1016/j.ejrh.2026.103161

Research Groups

• Department of Hydrogeology, Helmholtz Centre for Environmental Research — UFZ, Leipzig, Germany
• Department of Geography and Social Anthropology, Norwegian University of Science and Technology (NTNU), Trondheim, Norway
• School of Forest, Fisheries, and Geomatics Sciences, University of Florida, Gainesville, FL, USA
• Department Catchment Hydrology, Helmholtz Centre for Environmental Research — UFZ, Halle (Saale), Germany

Short Summary

This study introduces a transferable discharge-based approach, Relative Specific Discharge (RSD), to classify spatio-temporal patterns of potentially gaining and losing subcatchments in the Bode River catchment, Central Germany. The method effectively identifies high-yield (potentially gaining) and low-yield (potentially losing) hotspots, demonstrating greater robustness than water balance methods, especially for smaller subcatchments, and reveals dynamic seasonal and annual transitions in water yield.

Objective

• To classify each gauged subcatchment as high-yield (potentially gaining hotspots) or low-yield (potentially losing hotspots) by comparing its specific discharge to the catchment-wide average.
• To analyze temporal variability by examining annual and seasonal patterns of water yield under varying climatic conditions.
• To validate the newly proposed Relative Specific Discharge (RSD) approach by comparing its classifications with those derived from the well-established water balance-based Effective Catchment Index (ECI).
• To evaluate uncertainties, focusing on the influence of actual evapotranspiration (AET)-related uncertainties on the water balance approach and on the impacts of discharge-related uncertainties in the specific-discharge-based method.

Study Configuration

• Spatial Scale: Bode River catchment in Central Germany, with a total area of approximately 3.3 × 10⁹ m². The study focused on 18 gauged subcatchments, ranging in area from 4.00 × 10⁶ m² to 7.799 × 10⁸ m². Elevation within the catchment ranges from 16 m to 1142 m above sea level.
• Temporal Scale: Long-term (15-year mean discharge from November 2006 to October 2021), annual (individual mean annual discharges for 15 hydrological years), and monthly (mean monthly discharges averaged over the 15-year period).

Methodology and Data

• Models used:
  • Relative Specific Discharge (RSD) approach: A conceptual, discharge-based method that classifies subcatchments as high-yield (RSD > 1) or low-yield (0 < RSD ≤ 1) by comparing their specific discharge (dQ/dA) to the specific discharge of the entire catchment (Q/A)out.
  • Effective Catchment Index (ECI): A water balance-based approach (ECI = log(dQ / d(P-AET))) used for validation, classifying subcatchments as gaining (ECI > 0) or losing (ECI < 0).
  • Hierarchical clustering: Applied using the Nearest Point Algorithm (Euclidean distance) to classify temporal (annual and seasonal) yield patterns across subcatchments.
  • Mesoscale Hydrological Model (mHM): Used to generate an additional actual evapotranspiration (AET) dataset for uncertainty analysis.
• Data sources:
  • Discharge: Daily discharge data for 18 gauging stations, obtained from the State Office of Flood Protection and Water Management (LHW) Saxony-Anhalt (https://gld.lhw-sachsen-anhalt.de/).
  • Topography: Drainage area, mean elevation, and mean slope derived from the EU-DEM v1.1 at a 25 m resolution (European Union, Copernicus Land Monitoring Service, European Environment Agency, 2016).
  • Land Cover: Percentages of agricultural and forested/seminatural areas based on CORINE Land Cover for 2018.
  • Precipitation (P): DWD-HYRAS2 (Daily, 1 km × 1 km spatial resolution).
  • Actual Evapotranspiration (AET): TerraClimate (Monthly, 1/24th degree ≈ 13.6 km² spatial resolution), GLEAM (Monthly, 0.25° × 0.25° ≈ 487 km² spatial resolution), FLDAS (Monthly, 0.1° × 0.1° ≈ 78 km² spatial resolution), and mHM model (Daily, 4 km × 4 km spatial resolution).

Main Results

• Long-term Classification: Over the 15-year period, the RSD approach identified 6 low-yield and 12 high-yield subcatchments in the Bode River catchment. High-yield subcatchments are predominantly found in mountainous areas, showing a strong positive correlation with mean elevation (Spearman rank correlation ρ = 0.91, p < 0.05) and mean slope (ρ = 0.60, p < 0.05). Conversely, lowland agricultural subcatchments are mainly low-yield.
• Validation and Robustness: The RSD classification aligns with the ECI method for 15 out of 18 subcatchments. Discrepancies between the two methods are more likely when relative water inputs to a subcatchment exceed approximately 5.54 times the catchment mean. The RSD method demonstrated greater robustness in classification under uncertainty, with only 2 out of 18 subcatchments showing ambiguous classifications, compared to 9 out of 18 for the ECI method, particularly for smaller subcatchments.
• Annual Variations: Analysis of annual RSD revealed three distinct clusters. High-yield mountainous subcatchments showed increased water yield relative to the entire catchment during drier hydrological conditions (e.g., 2019-2020), indicating greater resilience to prolonged dry periods. Lowland subcatchments exhibited RSD patterns consistent with declining catchment-wide discharge.
• Seasonal Variations: Two general types of seasonal changes in RSD were identified. Most subcatchments showed stronger low-yield conditions (potential water losses) around August and stronger high-yield conditions (potential water gains) in winter. In contrast, a smaller group of four subcatchments displayed an opposite pattern, with RSD peaking around August, indicating higher discharge generation relative to the catchment average during summer months. Subcatchments with long-term averaged RSD values between 0.46 and 1.5 tend to switch between high-yield and low-yield conditions seasonally.

Contributions

• Developed and validated a parsimonious and transferable conceptual approach (Relative Specific Discharge, RSD) that uses only discharge data and drainage area to detect spatio-temporal variability in high-yield and low-yield conditions within nested regional catchments.
• Provided a robust alternative to water balance methods for identifying potentially gaining and losing hotspots, particularly for small subcatchments where precipitation and evapotranspiration data uncertainties significantly impact traditional approaches.
• Offered novel insights into the dynamic temporal (annual and seasonal) variability of water yield and potentially gaining/losing conditions, which are often not captured by static water balance methods.
• Demonstrated the resilience of high-yield mountainous subcatchments to prolonged dry conditions and identified contrasting seasonal patterns of water yield between mountainous and lowland areas.
• Established a proof-of-concept for a readily applicable tool for low-flow management and understanding water yield responses in gauged regional catchments, leveraging widely available hydrometric data.

Funding

• Helmholtz International Research School ‘Trajectories Towards Water Security’ (TRACER) (TRACER, Grant HIRS-0017).

Citation

@article{Lei2026Dischargebased,
  author = {Lei, Chao and Basso, Stefano and Cohen, Matthew J. and Musolff, Andréas and Tarasova, Larisa and Schmidt, Christian},
  title = {Discharge-based classifications of spatio-temporal patterns of potentially gaining and losing subcatchments in the Bode River catchment, Central Germany},
  journal = {Journal of Hydrology Regional Studies},
  year = {2026},
  doi = {10.1016/j.ejrh.2026.103161},
  url = {https://doi.org/10.1016/j.ejrh.2026.103161}
}