Mezin et al. (2025) A Comparative Study of Traditional Models and AI-Based Techniques for Hydrological Modeling

Identification

• Journal: Lecture notes in networks and systems
• Year: 2025
• Date: 2025-10-07
• Authors: Hammadi Mezin, Redouane Ezzahir
• DOI: 10.1007/978-3-032-01536-5_65

Research Groups

MAISI, National School of Applied Sciences, Ibn Zohr University, Agadir, Morocco

Short Summary

This review paper comparatively examines AI-based techniques and traditional models for hydrological modeling, identifying their strengths, limitations, and predictive accuracy across diverse geographic regions to inform future water management strategies.

Objective

• To examine and discuss previous research on the applications of AI-based techniques to improve hydrological modeling and compare them with traditional models, focusing on identifying strengths, limitations, evaluating accuracy and efficiency, and contrasting outcomes across various geographic regions and environmental conditions.

Study Configuration

• Spatial Scale: Global (reviewing studies across various geographic regions).
• Temporal Scale: Historical (reviewing previous research).

Methodology and Data

• Models used: Traditional hydrological models (e.g., SWAT, HEC-HMS, MODFLOW, MIKE SHE) and AI-based techniques (e.g., Artificial Neural Networks, Support Vector Machines, Random Forests, Deep Learning). The paper itself is a review.
• Data sources: Existing scientific literature and research papers.

Main Results

• The review identifies the strengths and limitations of both traditional and AI-based approaches in hydrological modeling.
• It evaluates the accuracy and efficiency of various predictive models.
• Outcomes are contrasted across different geographic regions and environmental conditions, providing insights into areas for future improvement.
• The study highlights the prospects for the development of AI-driven water management solutions.

Contributions

• Provides a comprehensive comparative analysis of traditional and AI-based hydrological modeling techniques.
• Offers insights into the current state of the art, identifying gaps and future research directions for improving water resource management.
• Contributes to the understanding of how AI can ensure the sustainability of water resources through enhanced predictions and identification of influencing factors.

Funding

Not specified in the provided text.

Citation

@article{Mezin2025Comparative,
  author = {Mezin, Hammadi and Ezzahir, Redouane},
  title = {A Comparative Study of Traditional Models and AI-Based Techniques for Hydrological Modeling},
  journal = {Lecture notes in networks and systems},
  year = {2025},
  doi = {10.1007/978-3-032-01536-5_65},
  url = {https://doi.org/10.1007/978-3-032-01536-5_65}
}