Ali et al. (2025) Development and application of a novel drought index for regional drought assessment: a case study from Pakistan

Identification

• Journal: Stochastic Environmental Research and Risk Assessment
• Year: 2025
• Date: 2025-10-09
• Authors: Farman Ali, Dong Su, Jing‐Cheng Han, Yuefei Huang, Alina Mukhtar, Zulfiqar Ali, Muhammad Ahmad, Shafeeq Ur Rahman
• DOI: 10.1007/s00477-025-03107-9

Research Groups

• College of Civil and Transportation Engineering, Shenzhen University, Shenzhen, China
• Water Science and Environmental Engineering Research Centre, College of Chemistry and Environmental Engineering, Shenzhen University, Shenzhen, China
• Key Laboratory of Coastal Urban Resilient Infrastructures (Shenzhen University), Ministry of Education, Shenzhen, China
• Laboratory of Ecological Protection and High Quality Development in the Upper Yellow River, School of Civil Engineering and Water Resources, Qinghai University, Xining, China
• College of Statistical Sciences, University of the Punjab, Lahore, Pakistan

Short Summary

This study introduces the Inter-Het Regional Drought Index (IHRDI), a novel drought assessment tool for Pakistan that integrates precipitation data interdependence and heterogeneity using Bayesian networks and square deviation. The IHRDI demonstrates superior performance over existing indices in capturing regional drought conditions and reveals increasing drought trends across most regions of Pakistan.

Objective

• To develop a novel regional drought index (IHRDI) based on an integrated weighting scheme that accounts for both interdependence and heterogeneity among meteorological stations to minimize bias in precipitation data aggregation.
• To improve regional drought monitoring and assessment by simultaneously considering the spatial and temporal distribution of precipitation.
• To provide a more reliable and regionally representative drought index, overcoming the limitations of single-station indices and addressing spatial variability across stations.

Study Configuration

• Spatial Scale: Five homogeneous regions across Pakistan, covering an area of 796,096 square kilometers. Precipitation values were extracted from 40 station locations using gridded data at 0.5° × 0.5° spatial resolution.
• Temporal Scale: 1981–2021 (41 years) precipitation time series, analyzed at multiple timescales ranging from 1 month to 48 months (T1 to T48).

Methodology and Data

• Models used:
  • Inter-Het Regional Drought Index (IHRDI) - novel index developed in this study.
  • Bayesian network model - utilized to assign interdependence-based weights.
  • Square deviation approach - employed for the computation of heterogeneity-based weights.
  • K-Component Gaussian Mixture Distributions (K-CGMDs) - used to model precipitation time series and calculate cumulative density functions.
  • Standardized Precipitation Index (SPI) - used for comparative evaluation.
  • Seasonally Combinative Regional Drought Index (SCRDI) - used for comparative evaluation.
  • Mann–Kendall test and Sen’s slope statistics - applied for trend analysis of drought conditions.
• Data sources: Gridded climate data from the Climate Research Unit (CRU) at 0.5° × 0.5° spatial resolution, archived from the Climate Change Knowledge Portal (CCKP).

Main Results

• The IHRDI, based on an integrated weighting scheme combining Bayesian network-derived interdependence weights and square deviation-derived heterogeneity weights, effectively addresses issues of under- and over-representation in precipitation data.
• Comparative evaluation showed that IHRDI consistently exhibits a higher correlation with SPI than SCRDI across most regions and time scales, demonstrating superior performance in capturing regional drought conditions.
• Drought quantification using IHRDI revealed significant variability in drought severity and duration across regions, with both characteristics worsening at longer time scales. Maximum drought severity reached 38.346 and duration up to 69 months in some regions (e.g., Sindh, KPK).
• Trend analysis using the Mann–Kendall test indicated statistically significant increasing trends in drought conditions across Sindh, Punjab, Khyber Pakhtunkhwa (KPK), and Northern Areas for all time scales.
• Conversely, Baluchistan showed a significant decreasing trend in drought conditions, particularly from the 6-month time scale onwards, suggesting a decreasing risk of drought in that specific region.

Contributions

• Development of a novel regional drought index (IHRDI) that uniquely integrates both interdependence (using Bayesian networks) and heterogeneity (using square deviation) among meteorological stations, leading to a more precise and representative regional drought assessment.
• Introduction of an integrated weighting scheme that minimizes bias from station proximity and varying precipitation patterns, improving upon traditional equal or simple unequal weighting methods in regional drought analysis.
• Comprehensive comparative evaluation demonstrating the superior performance and stability of IHRDI against widely used indices like SPI and SCRDI in capturing regional drought conditions.
• Detailed quantification of drought severity, duration, and trends across five homogeneous regions in Pakistan, offering crucial insights for regional water resource management and climate adaptation policies.

Funding

• National Natural Science Foundation of China (Nos. 52479019 & 51809007)
• China National Key R&D Program (No. 2024YFC3013303)
• Shenzhen Talent Research Start-up Fund

Citation

@article{Ali2025Development,
  author = {Ali, Farman and Su, Dong and Han, Jing‐Cheng and Huang, Yuefei and Mukhtar, Alina and Ali, Zulfiqar and Ahmad, Muhammad and Rahman, Shafeeq Ur},
  title = {Development and application of a novel drought index for regional drought assessment: a case study from Pakistan},
  journal = {Stochastic Environmental Research and Risk Assessment},
  year = {2025},
  doi = {10.1007/s00477-025-03107-9},
  url = {https://doi.org/10.1007/s00477-025-03107-9}
}