Ellahi et al. (2026) A Novel Integrated Standardized Index for Drought Assessment of Homogeneous Regions

Identification

• Journal: Environmental Earth Sciences
• Year: 2026
• Date: 2026-03-30
• Authors: RegionsAsad Ellahi, Hamza Amin, Ijaz Hussain, Jianyi Lin, Hanen Louati, Mhassen. E. E. Dalam, Mohammed M. A. Almazah
• DOI: 10.1007/s12665-026-12880-x

Research Groups

• Department of Statistics, Quaid-i-Azam University, Islamabad, Pakistan
• Wah Medical College, JWMC Office, National University of Medical Sciences, Rawalpindi, Pakistan
• Nanjing University of Information Science & Technology (NUIST), Beijing, China
• State Key Laboratory for Ecological Security of Regions and Cities, Institute of Urban Environment, Chinese Academy of Sciences, Beijing, China
• Mathematics Department, Faculty of Science, Northern Border University, Arar, Saudi Arabia
• Department of Mathematics, College of Sciences, King Khalid University, Abha, Saudi Arabia

Short Summary

This study introduces the Regionally Integrated Standardized Drought Index (RISDI), a novel framework integrating Model-Based Clustering, Principal Component Analysis, and K-Component Gaussian Mixture Distribution, to efficiently and accurately assess multi-regional drought severity by reducing data complexity while remaining sensitive to local changes.

Objective

• To develop and validate a comprehensive framework for efficient and accurate multi-regional drought assessment, introducing the Regionally Integrated Standardized Drought Index (RISDI) to reduce data complexity while preserving local drought signals.

Study Configuration

• Spatial Scale: 42 districts of Punjab Province, Pakistan, covering approximately 205,344 square kilometers.
• Temporal Scale: Monthly-averaged precipitation data from January 1981 to December 2022 (42 years).

Methodology and Data

• Models used: Model-Based Clustering (MBC) with Bayesian Information Criterion (BIC), Principal Component Analysis (PCA), K-Component Gaussian Mixture Distribution (K-CGMD), Expectation-Maximization (EM) algorithm, Weighted Mean approach, Standardized Precipitation Index (SPI) for comparative assessment.
• Data sources: Monthly-averaged precipitation data from NASA’s POWER Data Access Viewer.

Main Results

• Nine optimal regional clusters were identified using MBC, with a BIC value of -187,122.
• Clusters 1 to 4 showed lower mean, median, and lower-quartile precipitation with higher kurtosis, indicating generally dry conditions with occasional heavy rainfall.
• Clusters 5 to 9 exhibited higher mean, median, and upper quantiles, reflecting wetter and more consistent rainfall regimes.
• The 12-component Gaussian mixture distribution provided the best fit for standardizing aggregated precipitation data across all clusters, accurately capturing variability and density patterns.
• RISDI demonstrated high correlations, comparable standard deviations, and low-centered Root Mean Square Error (RMSE) values when compared to individual district SPIs within each cluster, confirming its reliability and representativeness.
• RISDI effectively reduces the complexity of large precipitation datasets while remaining sensitive to local changes, efficiently assessing multi-regional drought severity and patterns.

Contributions

• Introduces a novel, integrated, and standardized drought index (RISDI) for multi-regional drought assessment.
• Addresses information loss and overlapping representation issues inherent in existing approaches by retaining contributions from all locations within a cluster through a variance-informed weighting mechanism.
• Preserves intra-cluster heterogeneity by integrating contributions from all locations via a statistically grounded PCA-based weighting scheme.
• Uniquely combines Model-Based Clustering, variance-informed regional integration, and K-Component Gaussian Mixture Distribution standardization to flexibly model non-normal precipitation patterns.
• Provides an efficient dimensionality reduction method that remains sensitive to local drought signals.
• The proposed framework is generalizable and can be extended to other hydroclimatic variables (e.g., soil moisture, streamflow) for comprehensive multi-regional drought monitoring.

Funding

• Deanship of Research and Graduate Studies at King Khalid University (Large Research Project, grant number RGP.2/145/46).
• Deanship of Scientific Research at Northern Border University, Arar, Saudi Arabia (Project number NBU-FFR-2026-2920-02).

Citation

@article{Ellahi2026Novel,
  author = {Ellahi, RegionsAsad and Amin, Hamza and Hussain, Ijaz and Lin, Jianyi and Louati, Hanen and Dalam, Mhassen. E. E. and Almazah, Mohammed M. A.},
  title = {A Novel Integrated Standardized Index for Drought Assessment of Homogeneous Regions},
  journal = {Environmental Earth Sciences},
  year = {2026},
  doi = {10.1007/s12665-026-12880-x},
  url = {https://doi.org/10.1007/s12665-026-12880-x}
}