AUTOMATED ACOUSTICAL MONITORING OF TRIBOLIUM-CASTANEUM (COLEOPTERA, TENEBRIONIDAE) POPULATIONS IN STORED WHEAT

An automated acoustical detection system for monitoring Tribolium castaneum (Herbst) populations in stored wheat was evaluated using 16 microphones per 5 bu (176.2 liters) wheat. A regression equation explained 93.9% of the variation in the number of insect sounds over a range of 5-640 adult insects per 5 bu. For one microphone during a 10-s interval, the probability of detection increased rapidly from almost-equal-to 0.12 with 10 insects per 5 bu to 0.44 with 80 insects per 5 bu, and then more slowly to 0.66 with 640 insects per 5 bu. The probability of detection was almost-equal-to 0.90 with 40 insects per 5 bu and 3 microphones, 20 insects per 5 bu and 4 microphones, 10 insects per 5 bu and 5 microphones, or 5 insects per 5 bu and 12 microphones. More frequently monitoring a single microphone improved the probability of detection 60-80% as much as adding the same number of microphones. The number of insect sounds was unaffected by the sex or mating status of the insects and decreased logarithmically with increasing distance between insect and microphone. Adults produced 80 times more sounds than larvae. Automation of insect monitoring should increase reliability and reduce labor costs.