Bharghavi et al. (2025) Evaluating climate change impact on drought: a comprehensive review of drought indices and future projections

Identification

• Journal: Natural Hazards
• Year: 2025
• Date: 2025-09-23
• Authors: Kandula Bharghavi, Thotli Lokeswara Reddy, Hemalatha Kapa, Penti Rajesh, Hasanapuram Sushmitha, K. Krishna Reddy
• DOI: 10.1007/s11069-025-07681-7

Research Groups

• Department of Physics, Yogi Vemana University, Kadapa, Andhra Pradesh, India
• Department of Physics, Rajiv Gandhi University of Knowledge Technologies, Idupulapaya, Vempalli, Andhra Pradesh, India
• Department of H&S (Physics), Marri Laxman Reddy Institute of Technology and Management, Domara Pocham Pally, Hyderabad, Telangana, India

Short Summary

This review systematically evaluates the performance of key drought indices across ten global regions under climate change, finding that climate change exacerbates drought conditions and that the Standardised Precipitation Evapotranspiration Index (SPEI) consistently performs well, while highlighting the increasing role of remote sensing, AI, and ML in drought monitoring and prediction.

Objective

• Critically assess the influence of climate change on drought frequency, intensity, and duration across various geographical regions.
• Evaluate the application and comparative performance of standardised drought indices.
• Analyse various drought indices, highlighting their advantages, limitations, and applicability in different climatic regions.
• Examine how climate change affects drought patterns using historical and projected climate data.
• Identify emerging trends in drought forecasting models, remote sensing technologies, and machine learning approaches.
• Provide recommendations for future research and policy development to enhance climate resilience.

Study Configuration

• Spatial Scale: Global and multi-regional, covering ten major global regions including China, Iran, India, Europe (e.g., Mediterranean, Boreal, temperate, continental), North America (e.g., Great Plains, Continental U.S.), Africa (e.g., equatorial, Sahel, South Africa, Ethiopia), and South America (e.g., Amazon, southern Chile).
• Temporal Scale: Literature review encompassing publications from 2010 to 2024 (with emphasis on 2015–2024), analyzing studies covering historical periods (e.g., 1901–2018, 1961–2020) and future climate change projections.

Methodology and Data

• Models used:
  • Drought Indices: Standardised Precipitation Index (SPI), Standardised Precipitation Evapotranspiration Index (SPEI), Palmer Drought Severity Index (PDSI), Combined Drought Indicator (CDI), Vegetation Condition Index (VCI), Temperature Condition Index (TCI), Standardised Precipitation Evapotranspiration Runoff Index (SPERI), Reconnaissance Drought Index (RDI), Streamflow Drought Index (SDI), Standardised Runoff Index (SRI), Soil Water Deficit Index (SWDI), Soil Moisture Agricultural Drought Index (SMADI), Atmospheric Water Deficit (AWD), Crop Moisture Index (CMI), Integrated Drought Condition Index (IDCI), Modified Palmer Drought Severity Index (MPDSI), Vegetation Health Index (VHI), Scaled Drought Condition Index (SDCI), Evaporative Stress Index (ESI), GRACE-based Water Storage Deficit Index (WSDI), Evapotranspiration Condition Index (ETCI), China-Z Index (CZI), Modified CZI (MCZI), Standardised Soil Water Index (SSWI), Precipitation Condition Index (PCI), Standardised Streamflow Index (SSI), Hybrid Drought Index (HDI), Multi-Index Drought (MID) model, Geographically Independent Integrated Drought Index (GIIDI), Linear Combined Index (LCI), Comprehensive Agriculture Drought Condition Indicator (CADCI).
  • Artificial Intelligence/Machine Learning: Support Vector Machine (SVM), Random Forest (RF), Gaussian Process Regression (GPR), Gradient Boosting Machine (GBM), Long Short-Term Memory (LSTM) networks, Deep Learning.
• Data sources:
  • Literature Search: Web of Science, Scopus, Google Scholar, ScienceDirect.
  • Drought Index Calculation: Satellite-derived data (e.g., MODIS, TRMM, GPM, GLDAS, GRACE, Landsat), gridded climate data (e.g., CHIRPS, GPCC, ERA5), ground-based observations (e.g., AWS), precipitation, temperature, evapotranspiration, soil moisture, vegetation health, streamflow, runoff.

Main Results

• Climate change intensifies droughts globally, increasing their frequency, intensity, and duration, with regionally variable impacts.
• No single drought index is universally applicable; selection depends on drought type, region, and data availability.
• The Standardised Precipitation Evapotranspiration Index (SPEI) consistently performs well across most global regions (average scores 0.80–0.89), while the Palmer Drought Severity Index (PDSI) shows more variable performance, particularly in tropical regions (e.g., Amazon Basin 0.72, Southern Africa 0.69).
• Remote sensing-based indices (e.g., VCI, TCI, VHI) are crucial for monitoring, especially agricultural droughts, offering high spatial and temporal resolution and addressing data scarcity.
• Artificial Intelligence (AI) and Machine Learning (ML) significantly enhance drought prediction and monitoring, improving accuracy by 15–50% and extending forecast lead times by 1–2 months, particularly with models like LSTM.
• Future projections indicate an increase in agricultural and hydrological droughts, with intensity expected to increase linearly with rising global temperatures, emphasizing the importance of multi-index assessments.
• Key challenges include uncertainties in climate projections, scale issues in drought assessment, lack of long-term observational data in some regions, and the need for standardisation in index selection and application.

Contributions

• Provides a comprehensive, systematic evaluation and comparison of seven widely used drought indices (CDI, PDSI, SPEI, SPERI, SPI, TCI, and VCI) across ten major global regions using consistent performance metrics.
• Identifies indices that perform consistently well or poorly across diverse environments and highlights region-specific strengths and limitations, offering guidance for index selection in future drought risk assessments.
• Synthesizes the evolving patterns of drought under changing climatic conditions, including mechanisms, observed trends, and future projections.
• Underscores the increasing role and advancements in remote sensing, AI, and ML techniques for enhancing drought monitoring, prediction, and early warning systems.
• Highlights critical knowledge gaps and challenges in drought assessment, such as non-stationarity, anthropogenic influences, data reliability, model transparency, and the need for standardisation.

Funding

• The authors acknowledge contributions from individuals and Yogi Vemana University. No specific project names, programs, or reference codes are provided.

Citation

@article{Bharghavi2025Evaluating,
  author = {Bharghavi, Kandula and Reddy, Thotli Lokeswara and Kapa, Hemalatha and Rajesh, Penti and Sushmitha, Hasanapuram and Reddy, K. Krishna},
  title = {Evaluating climate change impact on drought: a comprehensive review of drought indices and future projections},
  journal = {Natural Hazards},
  year = {2025},
  doi = {10.1007/s11069-025-07681-7},
  url = {https://doi.org/10.1007/s11069-025-07681-7}
}