Moradi et al. (2026) Evaluating Design Peak Flow Estimation Methods

Identification

• Journal: Water Resources Management
• Year: 2026
• Date: 2026-04-01
• Authors: Soheyl Moradi, Andrea Petroselli, Matej Vojtek
• DOI: 10.1007/s11269-026-04620-4

Research Groups

• Institute of Geography, Slovak Academy of Sciences, Bratislava, Slovakia
• Department of Agriculture and Forest Sciences (DAFNE), Tuscia University, Viterbo, Italy
• Department of Geography, Geoinformatics and Regional Development, Faculty of Natural Sciences and Informatics, Constantine the Philosopher University in Nitra, Nitra, Slovakia

Short Summary

This study evaluates four hydrological approaches for design peak flow estimation across five sub-basins of the Gidra River in Slovakia for return periods of 2–1000 years. It finds that the STORAGE + COSMO4SUB (S + C) model, particularly with the Pearson Type III distribution, provides more consistent and balanced results compared to empirical methods, especially in ungauged basins, though uncertainties remain for extreme events.

Objective

• To comprehensively compare and evaluate four peak flow estimation techniques (Fuller’s empirical formula, Rational method, EBA4SUB, and STORAGE + COSMO4SUB) in terms of accuracy, data requirements, computational effort, and practical applicability for small and ungauged catchments in Slovakia.

Study Configuration

• Spatial Scale: Gidra River basin (~200 km²) in western Slovakia, including five sub-basins: Častá (4.6 km²), Štefanová (10.1 km²), Píla (31.4 km²), Budmerice (64.8 km²), and Pavlice (109.5 km²).
• Temporal Scale: Design peak flows for return periods (T) of 2, 5, 10, 20, 50, 100, and 1000 years. Rainfall data from 1981–2023. The S + C model generated 500-year synthetic rainfall series at 15-minute resolution.

Methodology and Data

• Models used:
  • Fuller’s empirical formula
  • Rational method
  • EBA4SUB (Event-Based Approach for Small and Ungauged Basins) framework, incorporating CN4GA (Curve Number for Green-Ampt) and WFIUH (Width-Function Based Instantaneous Unit Hydrograph).
  • STORAGE + COSMO4SUB (S + C) model (Continuous Simulation Model for Small and Ungauged Basins), incorporating the STORAGE (STOchastic RAinfall GEnerator) model for rainfall simulation, continuous CN4GA, and WFIUH.
  • Statistical distributions: Gamma, Pearson Type III, and Gumbel for flood frequency analysis.
• Data sources:
  • Topography: Resampled Digital Elevation Model (DEM) with 20 m resolution (derived from original DMR3.5 with 10 m resolution from Geodetic and Cartographic Institute). Used for watershed delineation, area, channel length, elevation difference, and slope.
  • Soil: Bonited Soil-Ecological Units (BPEJ) from the Soil Science and Conservation Research Institute in Bratislava, and forest soil units from the National Forest Centre in Zvolen. Used for Green-Ampt parameters (saturated hydraulic conductivity, matric pressure head difference, change in soil water content).
  • Land Cover: Basic Data Base for the Geographic Information System (ZBGIS) for 2023, adapted to CORINE Land Cover (CLC) legend (20 m resolution). Used to assign runoff coefficients (Ct) and Curve Number (CN) values.
  • Rainfall: Daily records from Častá and Cífer rainfall stations (Slovak Hydrometeorological Institute - SHMI) for 1981–2023. Depth–duration–frequency (DDF) curves constructed using regional hydrological data.
  • Observed Data: Continuous data from a hydrological station in the Píla sub-basin (established 1961) and official SHMI design values for Píla.

Main Results

• Píla Sub-Basin (Gauged):
  • All methods deviated from official design flows (6–44.5 m³/s).
  • Empirical approaches (Rational method, Fuller’s formula) systematically overestimated peak flows; Fuller’s formula showed up to threefold overestimation at long return periods.
  • EBA4SUB under Antecedent Moisture Condition II (AMC II) severely underestimated peak flows for short return periods (often near zero for T ≤ 10 years), while under AMC III, it provided estimates closer to official values for short return periods.
  • The S + C model provided closer agreement, generally within a factor of two of official values. The Pearson Type III distribution within S + C showed the most consistent agreement for return periods up to 100 years.
• Ungauged Sub-Basins (Častá, Štefanová, Budmerice, Pavlice):
  • Estimated peak flows spanned a wide range (near 0 to 479 m³/s).
  • Inter-method variability (Coefficient of Variation at T = 1000 years) decreased from 0.65 in small basins to 0.55 in larger ones.
  • Empirical methods (Rational, Fuller) scaled largely linearly with basin area, potentially oversimplifying hydrological processes.
  • EBA4SUB under AMC II consistently underestimated flows, especially in steep, small catchments, while AMC III produced more realistic estimates in larger basins.
  • The S + C model yielded the most stable and physically coherent results across basin scales, with Pearson Type III offering a balanced compromise between physical realism and statistical flexibility for low to moderate return periods.

Contributions

• Provides a comprehensive, side-by-side comparative analysis of four distinct peak flow estimation techniques (Fuller, Rational, EBA4SUB, COSMO4SUB) applied consistently within the Gidra basin, addressing a notable gap in the Slovak hydrological literature.
• Offers novel insights into the strengths and limitations of empirical versus physically-based models for design peak flow estimation, particularly for small and ungauged catchments in Slovakia and similar regions globally.
• Evaluates model performance not only against observed data in a gauged sub-basin but also assesses internal consistency, scaling behavior, and sensitivity to basin characteristics in ungauged areas.
• Highlights the importance of conservative parameterization and supports the use of probabilistic, process-informed approaches (like S + C with Pearson Type III) over purely empirical methods for flood hazard assessment in data-limited contexts.

Funding

• EU NextGenerationEU through the Recovery and Resilience Plan for Slovakia under project No. 09I03-03-V03-00085.
• The Ministry of Education, Science, Research and Sport of the Slovak Republic in cooperation with Centre for Scientific and Technical Information of the Slovak Republic.

Citation

@article{Moradi2026Evaluating,
  author = {Moradi, Soheyl and Petroselli, Andrea and Vojtek, Matej},
  title = {Evaluating Design Peak Flow Estimation Methods},
  journal = {Water Resources Management},
  year = {2026},
  doi = {10.1007/s11269-026-04620-4},
  url = {https://doi.org/10.1007/s11269-026-04620-4}
}