Data-driven solutions to predict pest population outbreaks are an increasingly important component of contemporary Integrated Pest Management (IPM). A prime opportunity to implement an automated solution exists with the cotton bollworm in North Carolina. Cotton bollworm has been the target of black light and pheromone trapping networks across the eastern US for decades. Information generated by these networks has been communicated to growers through traditional extension meetings and digital resources (e.g., blogs, twitter, and websites). Although the information indicates corn earworm activity, the lag time between observation and data availability prohibits accurate deployment of scouting and remedial measures. Through innovative sensor design targeting corn earworm, this project takes a first
step toward addressing the communication disconnect between growers and risk. We retrofitted insect pheromone traps to log moth catches and environmental conditions in realtime. Pheromone traps are made of two metal mesh units: a cone and a cylinder trap: the moths travel up the cone until they pass through the narrow tip, at which point they are caught in the cylindrical trap. Our prototype uses an InfraRed (IR) sensor system at the cone tip to count moths as they enter the trap. Through multiple iterations of lab and semi-field testing, we developed our first automated prototype and deployed it at the Central Crops Research Station in Clayton, NC. The trap automatically counted moths and was accurate within 2.5 moths of the true count. After improvements, 25 automated insect traps were built in the winter of 2019-2020. Traps were be deployed throughout eastern North Carolina during the summer of 2020 to monitor corn earworm populations in space and time. Access to real-time corn earworm data will improve the management of this pest with the goal of reducing pesticide use in multiple crops.