Agricultural drought: A comprehensive review of research trends, monitoring indices, modeling frameworks, knowledge gaps, and future pathways

Study region: Agricultural drought (AD) poses persistent risks to global food systems, with severe consequences for climate-sensitive regions such as South Asia, China, Africa, and drought-prone agroecosystems worldwide. These regions experience increasing soil-moisture deficits, vegetation stress, crop-yield losses, and heightened socio-economic vulnerability. This review, although global in scope, gives particular emphasis to drought-affected agricultural landscapes, including semi-arid, monsoon-dependent, and rain-fed farming regions where drought impacts are most pronounced. Study focus: This study provides a broad assessment of AD research spanning 1960–2024 through scientometric analysis (2294 Scopus publications) combined with a systematic literature review. It synthesizes global trends in drought indices, satellite-based monitoring, machine learning (ML) and artificial intelligence (AI) applications, hydrological and crop modeling frameworks, socio-economic vulnerability assessments, and early warning systems. The review evaluates major methodological developments—ranging from optical, thermal, and microwave remote sensing to composite drought indicators, advanced ML techniques and hybrid models. It also examines critical gaps in data quality, model interpretability, sensor limitations, multi-source data fusion, and the lack of region-specific calibration. New hydrological insights: This review highlights several emerging directions: (i) increasing reliance on multi-sensor satellite data for soil moisture, evapotranspiration, and vegetation-health diagnostics; (ii) the growing importance of explainable AI, graph neural networks, and digital-twin frameworks for capturing spatiotemporal drought propagation; (iii) improved understanding of lag relationships between meteorological triggers and agricultural impacts; and (iv) recognition of socio-ecological feedbacks as fundamental components of drought risk. Findings reveal that AD research is rapidly transitioning toward integrated, data-rich, and context-specific monitoring systems. Future priorities include enhanced calibration standards, cross-regional model transferability, integration of socio-economic vulnerability, sensor–model fusion, and real-time drought advisory systems to support adaptation and policy planning.