Dehghani et al. (2025) Classifying Drought Severity in Northern Iran Using Machine Learning and Integrated Climate Indices

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

Identification

• Journal: International Journal of Climatology
• Year: 2025
• Date: 2025-11-17
• Authors: Fatemeh Dehghani, Mahboubeh Molavi‐Arabshahi
• DOI: 10.1002/joc.70188

Research Groups

Not specified in the abstract.

Short Summary

This study assessed the effectiveness of machine learning models (Random Forest, AdaBoost, Decision Tree, Transformer) for drought classification in the northern Iranian provinces, finding that Random Forest consistently outperformed other models across all regions.

Objective

• To assess the effectiveness of various machine learning models (Random Forest, AdaBoost, Decision Tree, Transformer) for drought classification in the northern Iranian provinces of Golestan, Mazandaran, and Guilan.

Study Configuration

• Spatial Scale: Northern Iranian provinces of Golestan, Mazandaran, and Guilan.
• Temporal Scale: Historical climate data (specific duration not specified in the abstract).

Methodology and Data

• Models used: Random Forest (RF), AdaBoost, Decision Tree (DT), Transformer.
• Data sources: Historical climate data. Preprocessing included lagged features and statistical aggregates for tree-based models, and normalised data for the transformer model. Model performance was evaluated using F1-score, recall, accuracy, and precision.

Main Results

• Random Forest (RF) consistently outperformed other models across all regions, demonstrating superior accuracy, precision, and recall.
• AdaBoost followed closely in performance.
• Decision Tree (DT) provided moderate performance.
• The Transformer model showed limited effectiveness, particularly in Guilan and Mazandaran.
• Optimal hyperparameters were determined for each model to ensure robust evaluation.

Contributions

• Underscores the effectiveness of Random Forest in drought prediction for the studied regions.
• Highlights the regional variability in machine learning model performance for drought classification.
• Emphasises the importance of model selection and tuning for achieving reliable drought predictions.
• Offers insights into the application of machine learning techniques for drought monitoring.
• Provides a benchmark for future studies exploring advanced or hybrid models and additional climatic variables.

Funding

Not specified in the abstract.

Citation

@article{Dehghani2025Classifying,
  author = {Dehghani, Fatemeh and Molavi‐Arabshahi, Mahboubeh},
  title = {Classifying Drought Severity in Northern Iran Using Machine Learning and Integrated Climate Indices},
  journal = {International Journal of Climatology},
  year = {2025},
  doi = {10.1002/joc.70188},
  url = {https://doi.org/10.1002/joc.70188}
}