Sohrabi et al. (2025) Establishing coastal water level magnitude-duration-frequency curves for infrastructure design

Identification

• Journal: Advances in Water Resources
• Year: 2025
• Date: 2025-11-22
• Authors: Meraj Sohrabi, Hamed Moftakhari, Hamid Moradkhani
• DOI: 10.1016/j.advwatres.2025.105179

Research Groups

• Center for Complex Hydrosystems Research, Department of Civil, Construction and Environmental Engineering, University of Alabama, Tuscaloosa, AL, USA

Short Summary

This study introduces a framework for developing water level Magnitude-Duration-Frequency (MDF) curves using global tide gauge data to address the overlooked duration aspect of extreme sea level events. It finds that the average water level difference between 1-hour and 24-hour durations for similar return periods is 0.8 meters, with differences up to 3.63 meters for a 100-year event, and highlights the impact of event selection methods on these curves.

Objective

• To introduce a framework for developing water level Magnitude-Duration-Frequency (MDF) curves to characterize the magnitude, duration, and frequency of extreme coastal water level events for improved infrastructure design and flood risk assessment, addressing the overlooked aspect of event duration.

Study Configuration

• Spatial Scale: Global, utilizing data from 217 selected coastal tide gauges.
• Temporal Scale: Hourly water level data, with selected records having at least 55 years of data. Mean Sea Level (MSL) standardized to the 1983–2001 National Tidal Datum Epoch.

Methodology and Data

• Models used:
  • Mann-Kendall (MK) test for trend analysis.
  • Linear regression for detrending sea level rise.
  • Block Maxima (BM) approach for extreme event extraction, using moving-average windows (3, 9, 12, 24 hours).
  • Peak-Over-Threshold (POT) approach for extreme event extraction, using the 99th percentile threshold and minimum 3-day event separation.
  • Generalized Extreme Value (GEV) distribution for BM method.
  • Generalized Pareto Distribution (GPD) for POT method.
• Data sources:
  • Global ESL Analysis (GESLA-V3) dataset, compiling sea level records from 36 providers.
  • 5119 tide gauge records initially, filtered down to 217 high-quality, long-term, and complete records.

Main Results

• A framework for developing water level Magnitude-Duration-Frequency (MDF) curves was introduced.
• The average water level difference between 1-hour and 24-hour durations for similar return periods is 0.8 meters.
• The maximum water level difference between 1-hour and 24-hour durations can reach up to 3.63 meters for a 100-year event.
• Differences in developed MDF curves due to event selection methods (BM vs. POT) can reach up to 0.5 meters across most regions, and up to 2 meters in certain areas like the North Sea for higher return periods.
• Shorter duration events (e.g., 3 hours) consistently show higher water levels in many locations, with 3-hour events being the most common.
• The frequency of extreme events declines significantly with increasing duration: 3-hour events occur in all studied locations, while 24-hour events occur in only 36.9% (80/217) of locations.
• MDF curves can exhibit non-monotonic relationships (decreasing, constant, or increasing slopes) unlike conventional precipitation IDF curves, influenced by local tidal conditions.
• The BM method generally provides higher water level estimates compared to the POT method, except in specific regions like Lake Ontario.
• Most stations show no clear trends in ESL event frequency, but some regions (e.g., west coast of North America, Europe, Japan) exhibit positive or negative trends.
• The U.S. Gulf Coast generally shows lower water levels across all durations and return periods, reflecting the influence of smaller tidal ranges and storm surge processes.
• Integration of MDF curves with NWS flood thresholds reveals regional variations in flood risk, highlighting vulnerability to short-duration, high-magnitude events (e.g., Charleston, SC) versus longer-duration events (e.g., Galveston, TX).

Contributions

• Introduces a novel framework for developing water level Magnitude-Duration-Frequency (MDF) curves, addressing the previously overlooked duration aspect of extreme sea level events.
• Provides a more comprehensive tool for coastal infrastructure design and flood risk assessment by integrating magnitude, duration, and frequency.
• Quantifies the significant impact of event duration on extreme water levels (e.g., 0.8 m average difference between 1-hour and 24-hour durations, up to 3.63 m for 100-year events).
• Evaluates and compares the influence of different extreme event extraction methods (Block Maxima and Peak-Over-Threshold) on MDF curves, highlighting potential discrepancies up to 2 meters.
• Offers insights into the spatial and temporal variability of extreme coastal water level durations and magnitudes globally, informing location-specific adaptation strategies.
• Demonstrates the practical application of MDF curves by integrating them with National Weather Service (NWS) flood thresholds, providing a dynamic, risk-based interpretation of coastal hazards.

Funding

• US Army Corps of Engineers (Contract W912HZ2020055)
• National Science Foundation (Award #2238000)
• The University of Alabama (financial support and article publishing charges)

Citation

@article{Sohrabi2025Establishing,
  author = {Sohrabi, Meraj and Moftakhari, Hamed and Moradkhani, Hamid},
  title = {Establishing coastal water level magnitude-duration-frequency curves for infrastructure design},
  journal = {Advances in Water Resources},
  year = {2025},
  doi = {10.1016/j.advwatres.2025.105179},
  url = {https://doi.org/10.1016/j.advwatres.2025.105179}
}