Altarhouni (2026) Predicting Sandstorms and Drought Using Artificial Intelligence Techniques Based on Climate Data: A Case Study of Jifarah Plain, Libya

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

Identification

• Journal: Tobruk University Journal of Engineering Sciences
• Year: 2026
• Date: 2026-08-30
• Authors: Lutfiyah Abraham Mohammed Altarhouni
• DOI: 10.64516/tujes.v7i2.81

Research Groups

Not explicitly mentioned (Data provided by the Libyan National Meteorological Center).

Short Summary

This study develops an AI-based framework to predict drought severity and sandstorm occurrence in Libya's Jifarah Plain using climate data from 1985 to 2022. The results demonstrate that a weighted Ensemble model provides the highest predictive accuracy for these climate hazards.

Objective

• To develop and evaluate an artificial-intelligence framework for the joint spatio-temporal prediction of drought severity and sandstorm occurrence in the Jifarah Plain, Libya.

Study Configuration

• Spatial Scale: Regional (Jifarah Plain, Libya), focusing on three meteorological stations: Tripoli Airport, Al-Aziziyah, and Al-Asa.
• Temporal Scale: 1985–2022 (Monthly resolution).

Methodology and Data

• Models used: Random Forest (RF), Long Short-Term Memory (LSTM), and a weighted Ensemble model.
• Data sources: Ground observations from the Libyan National Meteorological Center (LNMC) and NASA POWER reanalysis data.

Main Results

• Model Performance: The Ensemble model achieved the highest accuracy with $R^2 = 0.955$ and $\text{RMSE} = 0.057$, surpassing RF ($R^2 = 0.888$, $\text{RMSE} = 0.089$) and LSTM ($R^2 = 0.838$, $\text{RMSE} = 0.115$).
• Climate Trends: A marginal warming trend of $+0.014\text{ °C/year}$ was identified ($p = 0.054$, $R^2 = 0.10$).
• Hazard Timing: The period of maximum combined risk for sandstorms and drought is between March and June.

Contributions

• Proposes a reproducible AI framework adaptable to other arid regions with sparse ground-based monitoring networks.
• Introduces the Sandstorm Occurrence Probability Index (SOPI) and drought-risk layers to support early-warning systems and climate-adaptation planning.

Funding

Not mentioned in the provided text.

Citation

@article{Altarhouni2026Predicting,
  author = {Altarhouni, Lutfiyah Abraham Mohammed},
  title = {Predicting Sandstorms and Drought Using Artificial Intelligence Techniques Based on Climate Data: A Case Study of Jifarah Plain, Libya},
  journal = {Tobruk University Journal of Engineering Sciences},
  year = {2026},
  doi = {10.64516/tujes.v7i2.81},
  url = {https://doi.org/10.64516/tujes.v7i2.81}
}