From Pest Surveillance to Precision Management: Emerging Technologies for Monitoring, Forecasting and Sustainable Management of Rice Stem Borers

Abstract

Rice stem borers continue to be major pests in rice cultivation because their larval forms exhibit cryptic behaviour, asynchronous development and dependence on crop phenology, which undermine conventional scouting approaches. The yellow stem borer, Scirpophaga incertulas (Walker), and the striped stem borer, Chilo suppressalis (Walker), are among the major rice pests that have been widely studied. Studies have demonstrated that a combination of pheromone and light traps, weather, phenology, remote sensing, geographic information systems, computer vision and machine learning can contribute to improved integrated pest management. Weather-based, spatio-temporal and phenology-driven models can be used to predict the temporal abundance of crop pests at specific developmental stages, while satellite and unmanned aerial system (UAS), remote sensing allows pest infestation to be assessed at large scales. Automated trapping systems, hyperspectral imaging and convolutional neural networks can improve detection speed and accuracy. Nevertheless, adoption of these technologies is limited by data shifts, difficulties in species- and life-stage-specific identification limited field validation of modern technologies”high sensor costs, and the absence of links between predicted risks and management actions. Biological control agents, mating disruption, host plant resistance, RNA interference technology and resistance-informed insecticide applications can help bridge existing gaps in integrated pest management. This review focuses on recent advances in the surveillance, prediction, assessment and management of rice stem borers published after 2015, while also providing an overview of their integration into precision pest management.