Reassessing economic threshold levels for rice pest management: Transitioning from manual field observations to AI-driven precision systems

Background : The economic threshold level (ETL) and economic injury level (EIL) have served as the fundamental decision making indices for integrated pest management (IPM) since the mid-20th century. However, the rapid intensification of rice cultivation, characterised by high-tillering cultivars and heavy synthetic inputs, has fundamentally altered agricultural microclimates and pest dynamics. This review reassesses the validity of traditional ETL models in the context of contemporary technological advancements and shifting ecological realities. Methods : This critical review synthesises literature published between 1951 and 2026 across multidisciplinary databases, including Web of Science, Scopus, ScienceDirect, and Google Scholar. The methodology employs a transparent, PRISMA-guided framework to evaluate the physiological impacts of climate change on host-pest dynamics, analyse the modulating influence of the rhizosphere microbiome, and contrast traditional manual scouting against emerging artificial intelligence (AI)-driven monitoring technologies. Results : Findings indicate that elevated carbon dioxide (CO 2 ) and drought stress recalibrate the metabolic relationship between host plants and herbivores, often accelerating pest population buildup. Furthermore, the study identifies the “rhizosphere zoo” as a critical but overlooked modulator of plant immune pathways, specifically jasmonic and salicylic acid, which significantly enhances host tolerance to biotic stress. Traditional manual sampling methods were found to be increasingly inadequate for modern food security demands due to labour intensity and lack of real-time predictive capacity. Conclusions : The article proposes a transition toward “Precision ETL,” a framework underpinned by the integration of the Internet of Things (IoT), unmanned aerial vehicles (UAVs), and artificial intelligence (AI). This paradigm shift moves rice pest management from reactive scouting to proactive, predictive modelling, facilitating targeted interventions that reduce pesticide reliance while preserving beneficial entomofauna.