Veettil et al. (2026) Artificial intelligence applications in drought quantification and impact assessment on hydrological and agricultural indicators

Identification

• Journal: Elsevier eBooks
• Year: 2026
• Date: 2026-01-01
• Authors: Anoop Valiya Veettil, Ali Fares, Ripendra Awal, Md Symum Islam
• DOI: 10.1016/b978-0-443-44625-2.00007-2

Research Groups

• Cooperative Agricultural Research Center, College of Agriculture, Food, and Natural Resources, Prairie View A&M University, Prairie View, TX, United States
• College of Agriculture, Food, and Natural Resources, Prairie View A&M University, Prairie View, TX, United States

Short Summary

This chapter reviews the application of artificial intelligence (AI) methods for robust quantification, near-real-time monitoring, and high-resolution spatial classification of various drought types and their impacts on hydrological and agricultural indicators. It highlights AI's role in improving drought prediction, monitoring, and linking severity to outcomes for better management.

Objective

• To review and synthesize artificial intelligence (AI) applications for quantifying different drought types (meteorological, agricultural, hydrological, socioeconomic) and assessing their impacts on key indicators (streamflow, groundwater, reservoir storage, soil moisture, vegetation status), focusing on prediction, near-real-time monitoring, and high-resolution spatial classification.

Study Configuration

• Spatial Scale: Global to field-scale, with a focus on high-resolution spatial classification and field-scale mapping.
• Temporal Scale: Near-real-time monitoring, subseasonal forecasts, and earlier detection of onset and flash drought.

Methodology and Data

• Models used: Random forest, support vector machine, deep learning methods (convolutional neural networks, long-short term memory architectures), and hybrid frameworks combining physics-based models with AI.
• Data sources: Multisource weather data, soil data, satellite data (e.g., Soil Moisture Active Passive (SMAP), Moderate Resolution Imaging Spectroradiometer (MODIS), Sentinel-2), and standardized drought indices.

Main Results

• AI methods significantly improve drought prediction, enable near-real-time monitoring, and provide high-resolution spatial classification.
• Operational gains include earlier detection of flash droughts, subseasonal category change forecasts, and field-scale maps of vegetation stress.
• AI-driven interpretability tools expose key drought drivers such as precipitation deficits, evaporative demand, and antecedent storage.
• AI effectively links drought severity to hydrological responses and agricultural outcomes, supporting risk-based triggers for water allocation, irrigation scheduling, and contingency planning.

Contributions

• Provides a comprehensive review and synthesis of AI applications for drought quantification and impact assessment across various drought types and indicators.
• Highlights operational benefits of AI in drought management, including improved prediction, monitoring, and impact assessment.
• Identifies key challenges in AI application for drought, such as data sparsity, transferability, uncertainty quantification, and privacy.
• Proposes a practical framework that combines physics-based models with AI for scalable early warning and resilient drought management.

Funding

• Not specified in the provided text.

Citation

@article{Veettil2026Artificial,
  author = {Veettil, Anoop Valiya and Fares, Ali and Awal, Ripendra and Islam, Md Symum},
  title = {Artificial intelligence applications in drought quantification and impact assessment on hydrological and agricultural indicators},
  journal = {Elsevier eBooks},
  year = {2026},
  doi = {10.1016/b978-0-443-44625-2.00007-2},
  url = {https://doi.org/10.1016/b978-0-443-44625-2.00007-2}
}