Submitted to: Journal of Economic Entomology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/22/2010
Publication Date: 12/1/2010
Citation: Mankin, R.W., Hodges, R.D., Nagle, H.T., Schal, C., Pereira, R.M., Koehler, P.G. 2010. Acoustic indicators for targeted detection of stored product and urban insect pests by inexpensive infrared, acoustic, and virbrational detection of movement. Journal of Economic Entomology. 103:1636-1646. Interpretive Summary: Pests that live hidden in grain and wood are difficult to detect. Scientists at the Center for Medical, Agricultural, and Veterinary Entomology Gainesville, Florida, North Carolina State University, and University of Florida developed an inexpensive insect detection instrument using infrared, acoustic, and vibration sensors and then collected and analyzed signals produced by three species of stored product insect pests and two common household pests. The project had two goals—one to determine if such a device could be constructed inexpensively, and the second to determine if the signals could be used to distinguish which of several species had been detected. The device worked successfully, and signal processing methods were devised to distinguish the insects from each other.
Technical Abstract: Crawling or running, scraping or shuffling, and wriggling activity of three stored-product pests, Sitophilus oryzae (L.) (Coleoptera: Curculionidae), Tribolium castaneum (Herbst) (Coleoptera: Tenebrionidae), and Stegobium paniceum (L.) (Coleoptera: Anobiidae), and two urban pests, Blattella germanica (L.) (Blattodea: Blattellidae) and Cimex lectularius L. (Hemiptera: Cimicidae) were monitored individually by infrared sensors, microphones, and a piezoelectric sensor in a small arena to evaluate effects of insect locomotory behavior and size on the ability of an inexpensively constructed instrument to detect insects and distinguish among different species. Adults of all species could be detected when crawling, scraping, and wriggling. The smallest insects in the study, 1st-4th-instar C. lectularius nymphs, could not be detected easily when crawling, but could be detected when scraping or wriggling. Sound and vibration sensors detected brief, 3-10-ms impulses from all tested species, often grouped in distinctive trains (bursts), typical of impulses in previous acoustic detection experiments. There were statistically significant differences among species in distributions of infrared signal durations, sound impulse-burst durations, and sound pressure levels (energy) of impulses that best matched an averaged spectrum (profile) of scraping behavior. Thus, there is potential that signals collected by an inexpensive, polymodal-sensor instrument could be used in automated trapping systems to identify targeted species, 0.1-mg or larger, in environments where servicing of traps is difficult or when timeliness of trapping information is important.