Alimohammadi et al. (2025) Reliability analysis of dam spillway capacity under extreme flood uncertainty: Boostan Dam, Iran

Identification

• Journal: Journal of Hydrology Regional Studies
• Year: 2025
• Date: 2025-11-20
• Authors: Saeed Alimohammadi, Masoumeh Behrouz, Hamid Ebrahimi, Mohammad Reza Majdzadeh Tabatabai
• DOI: 10.1016/j.ejrh.2025.102948

Research Groups

• Civil Engineering, Civil, Water and Environmental Engineering Faculty, Shahid Beheshti University, Tehran, Iran

Short Summary

This study developed an integrated analytical-numerical framework to quantify uncertainties in Probable Maximum Precipitation (PMP), Probable Maximum Flood (PMF), and peak discharge quantiles for the Boostan Dam, Iran, and assessed spillway reliability. It found that the existing spillway has negligible reliability for PMF and requires significant widening (50-80 m) to achieve 95% reliability for 1,000-10,000 year floods.

Objective

• To advance probabilistic dam safety assessment by explicitly integrating uncertainty quantification of Probable Maximum Precipitation (PMP) and Probable Maximum Flood (PMF) with spillway reliability analysis.
• To provide dam designers and planners with probabilistic, risk-informed guidance for spillway design and operation under extreme hydrological uncertainties.

Study Configuration

• Spatial Scale: Boostan Dam, Gorganrud River Basin, Golestan Province, Iran, with a watershed area of 1587 km². The analysis considered 5 rain-gauge stations and multiple sub-basins.
• Temporal Scale: 47 years of maximum daily rainfall and peak discharge records from the Tamar station at the dam inlet.

Methodology and Data

• Models used: Frequency Factor Equation (FFE), Generalized Extreme Value (GEV) distribution, Gamma distribution, HEC-SSP (for peak discharge quantile estimation), Monte Carlo–Latin Hypercube Sampling (MCS-LHS) (for scenario generation and uncertainty propagation), HEC-HMS (for rainfall-runoff modeling and hydrograph simulation), Load–Resistance Interference Method (LRIM) (for spillway reliability assessment).
• Data sources: Maximum daily rainfall from rain-gauge stations, peak-discharge records from a river gauge station, river and basin characteristics derived from topographic and land-use maps, and dam and spillway characteristics.

Main Results

• The Hershfield method significantly underestimates PMP by 37.5 % (sample estimate of 112 mm versus an expected PMP of 154.3 mm) and PMF by 65 %.
• The expected PMP value was calculated as 154.3 mm with a standard deviation of 29.4 mm, resulting in a 95 % confidence interval of [95.2 mm, 212.8 mm].
• The Probable Maximum Flood (PMF) was found to span a wide credible interval of [1330 m³/s, 6295 m³/s] when propagating PMP uncertainty and hydrologic-model parameter variability.
• The existing spillway, with a width of 40 m and a capacity of 785 m³/s, offers negligible reliability (<1 %) for the PMF event.
• To achieve 95 % reliability for floods with a 1,000-year return period, the spillway width must be increased to approximately 50 m.
• To achieve 95 % reliability for floods with a 10,000-year return period, the spillway width must be increased to approximately 80 m.
• For the 2,000-year design flood mandated by national guidelines, the current spillway provides 85 % reliability, requiring an increase to 70 m width for 95 % reliability.
• Even with a significant enlargement to a 120 m spillway width, the PMF reliability remains low at 13 %.
• The Logarithmic Safety Factor (W3) was identified as the most suitable performance function for reliability calculations due to its minimum absolute skewness coefficient.

Contributions

• Developed and applied an integrated analytical–numerical framework that extends the Frequency Factor Equation to Generalized Extreme Value (GEV) and Gamma distributions, coupled with Monte Carlo–Latin Hypercube Sampling (MCS-LHS) for comprehensive uncertainty quantification of PMP/PMF.
• Bridged the gap between theoretical uncertainty models and practical dam engineering by applying this advanced framework to a real-world case study (Boostan Dam, Iran), providing quantitative, probabilistic reliability assessments.
• Demonstrated the inadequacy of traditional deterministic "single-value" PMP/PMF estimates and highlighted the critical need for a paradigm shift towards reliability-based design practices in dam safety assessment.
• Provided actionable, risk-informed guidance for spillway design and operation, including specific recommendations for spillway width adjustments and a dual-spillway strategy for the Boostan Dam.
• Contributed to advancing dam safety regulations by advocating for the incorporation of uncertainty bands, confidence intervals, and formal reliability assessments in national guidelines.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Citation

@article{Alimohammadi2025Reliability,
  author = {Alimohammadi, Saeed and Behrouz, Masoumeh and Ebrahimi, Hamid and Tabatabai, Mohammad Reza Majdzadeh},
  title = {Reliability analysis of dam spillway capacity under extreme flood uncertainty: Boostan Dam, Iran},
  journal = {Journal of Hydrology Regional Studies},
  year = {2025},
  doi = {10.1016/j.ejrh.2025.102948},
  url = {https://doi.org/10.1016/j.ejrh.2025.102948}
}