Klaho et al. (2026) A hybrid drought index prediction framework under climate change scenarios by equipping index fusion with an individual artificial intelligence method

Identification

• Journal: Scientific Reports
• Year: 2026
• Date: 2026-09-13
• Authors: Mohamad Haytham Klaho, Ramtin Moeini, Mohammadali Alijanian
• DOI: 10.1038/s41598-026-71545-8

Research Groups

• Department of Water Resources Engineering, University of Tehran
• Khuzestan Water and Electricity Organization
• Iranian Meteorological Organization

Short Summary

This study proposes a novel framework for predicting hybrid droughts in the Marun basin, Iran, using an index-fusion-based approach that combines multiple standard drought indices with individual artificial intelligence methods. The results show that the proposed framework can improve predictive accuracy by up to 2.4% under climate change scenarios.

Objective

• Investigate the effectiveness of a hybrid drought index prediction framework in capturing the complex relationships between meteorological, hydrological, and socioeconomic variables.
• Evaluate the performance of individual artificial intelligence methods (MLP, RF, SVR) and fusion-based approaches (MLP and RF) for predicting drought conditions under climate change scenarios.

Study Configuration

• Spatial Scale: Regional scale (Marun basin, Iran)
• Temporal Scale: Historical period (2002-2021), future period (2022-2040)

Methodology and Data

• Models used:
  • Shannon Entropy (SE) method for hybrid drought index calculation
  • Copula functions (C) for multivariate distribution modeling
  • Principal Component Analysis (PCA) for dimensionality reduction
  • Individual Artificial Intelligence (IAI) methods: Multi-Layer Perceptron (MLP), Support Vector Regression (SVR), Random Forest (RF)
• Data sources: Ground observed data, climate model outputs (NCEP, CanESM5, NorESM2)

Main Results

• The proposed framework can improve predictive accuracy by up to 2.4% under climate change scenarios.
• The Copula-Hybrid Drought Index (CHDI) performs best among the hybrid indices for predicting drought conditions.
• The fusion-based approach using MLP and RF methods outperforms individual IAI models in predicting drought conditions.

Contributions

• This study provides a novel framework for predicting hybrid droughts that incorporates multiple standard drought indices with individual artificial intelligence methods.
• The results highlight the importance of considering climate change scenarios when predicting drought conditions.

Funding

• Iranian Ministry of Science, Research and Technology (MSRT)
• University of Tehran Research Council

Citation

@article{Klaho2026hybrid,
  author = {Klaho, Mohamad Haytham and Moeini, Ramtin and Alijanian, Mohammadali},
  title = {A hybrid drought index prediction framework under climate change scenarios by equipping index fusion with an individual artificial intelligence method},
  journal = {Scientific Reports},
  year = {2026},
  doi = {10.1038/s41598-026-71545-8},
  url = {https://doi.org/10.1038/s41598-026-71545-8}
}