Abstract
El Niño-Southern Oscillation (ENSO) has a significant impact on global agricultural systems in tropical regions, where rainfed rice production is highly vulnerable to climatic extremes, including droughts and floods. This systematic review synthesizes findings from two decades of research to examine the effects of ENSO phases-El Niño and La Niña-on cereal crop yields, with a focus on rainfed rice in Thailand. The study also evaluates the role of artificial intelligence (AI) in predicting ENSO-induced impacts on crop productivity. Findings indicate that El Niño events often reduce rainfall, increasing drought stress, while La Niña leads to excessive precipitation and flooding-both of which adversely affect rice productivity. AI-based studies have shown that models such as Random Forest (RF), Long Short-Term Memory (LSTM), and Convolutional Neural Networks (CNNs) demonstrate strong potential, although limitations remain in terms of scalability and local adaptation. • Hybrid modeling approaches that integrate physical and statistical methods are essential. • Future research must enhance data quality and integrate adaptive technologies to support climate-resilient agriculture.