Al-Awad et al. (2026) A Hierarchical Artificial Intelligence Framework for the Inverse Calibration of Spatially Distributed Manning’s Roughness Coefficients in HEC-RAS Models

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Hydrology
• Year: 2026
• Date: 2026-09-10
• Authors: Khabeer Al-Awad, Layth Abdulameer, Mahmoud Saleh Al-Khafaji, Aysar Tuama Al-Awadi, Ahmed N. Al-Dujaili, Anmar Dulaimi, Luís Filipe Almeida Bernardo, Hugo Alexandre Silva Pinto
• DOI: 10.3390/hydrology13090244

Research Groups

• Department of Civil Engineering, University of Baghdad
• Institute of Water Resources Management, Ministry of Water Resources, Iraq

Short Summary

This study proposes a hierarchical artificial intelligence framework for the inverse calibration of spatially distributed Manning’s roughness coefficients in one-dimensional hydraulic models, achieving improved accuracy and reduced uncertainty in river hydraulic simulations.

Objective

• To develop an efficient method for calibrating Manning’s roughness coefficients across different channel zones using a deep learning framework.

Study Configuration

• Spatial Scale: 48 km reach of the Tigris River in Baghdad, Iraq.
• Temporal Scale: Not specified (presumably steady-state conditions).

Methodology and Data

• Models used: HEC-RAS hydraulic model, Gradient Boosting Regression, Random Forest, Multi-Layer Perceptron, three-layer neural network with Differential Evolution optimisation and cubic spline interpolation.
• Data sources: Measured cross-sections of the Tigris River.

Main Results

• The deep learning framework achieved a 96.6% reduction in root mean square error (from 1.202 to 0.041 m), with R2 = 0.992 and negligible bias (−0.004 m).
• Conventional machine learning models produced spatially variable Manning’s roughness distributions, with the calibrated main-channel roughness being 34.0–57.5% higher than the corresponding bank values.

Contributions

• The proposed framework provides an effective approach for calibrating spatially distributed roughness coefficients in one-dimensional hydraulic models, improving river hydraulic simulations and supporting future applications to flood modelling.
• The study demonstrates the potential of deep learning techniques for inverse calibration problems in hydrology.

Funding

• Not specified.

Citation

@article{AlAwad2026Hierarchical,
  author = {Al-Awad, Khabeer and Abdulameer, Layth and Al-Khafaji, Mahmoud Saleh and Al-Awadi, Aysar Tuama and Al-Dujaili, Ahmed N. and Dulaimi, Anmar and Bernardo, Luís Filipe Almeida and Pinto, Hugo Alexandre Silva},
  title = {A Hierarchical Artificial Intelligence Framework for the Inverse Calibration of Spatially Distributed Manning’s Roughness Coefficients in HEC-RAS Models},
  journal = {Hydrology},
  year = {2026},
  doi = {10.3390/hydrology13090244},
  url = {https://doi.org/10.3390/hydrology13090244}
}