Univ. of Florida College of Veterinary Medicine, Gainesville, Florida. Class of 2008.
Mentor: Richard Mankin
Acoustic and Visual Monitoring of Female Mediterranean Fruit Fly (Ceratitis capitata) Capture Events in a Bioassay Tunnel
Abstract: The medfly is a worldwide pest to citrus and some deciduous fruits. Quarantine officials have successfully kept medfly infestation under control in the mainland U.S. by deploying early warning traps to target accidental introductions. However, the traps that are currently used in medfly surveillance programs require considerable time and labor to monitor and maintain, and medfly populations can build up rapidly between scheduled inspections. Researchers are attempting to develop automated trapping systems that would employ microphones to identify the medfly by its wingbeat frequency. Such devices would decrease labor costs and provide immediate notification of medfly detection. While these devices work well in a laboratory setting where interfering background noise can be eliminated, it is uncertain how these detection devices will perform under “real world” environmental conditions. In this project, female medflies were observed and recorded while being captured at a male-baited trap in a bioassay tunnel. The acoustic environment of the tunnel was similar to the outside, in that the microphone was exposed to street, wind, and thunderstorm noise. It is expected that the results of analysis of these recordings will help improve the capability of an automated system to detect medfly wingbeats under non-ideal conditions.
Amanda Tovey is recording a video of female medflies that have flown in a wind tunnel to a lure baited with male medflies and are being captured on a yellow sticky ball nearby. Wingbeats of flying females and females attempting to takeoff after capture are being recorded simultaneously for subsequent signal analysis of flight sound characteristics.
Amanda Tovey is conducting an analysis of female medfly wingbeat signals that will be used to help identify features that distinguish among different insect species. Ultimately, the goal is to automate the process of identifying insects that are captured by a trap.