2012 Annual Report
1a.Objectives (from AD-416):
The objective of this cooperative research project is to develop an acoustical monitoring tool for detecting subterranean termites. Primary target pest is the Formosan subterranean termite (Coptotermes formosanus) in structures, trees, and soil.
1b.Approach (from AD-416):
Probably the most important factor in termite control is the early detection of new termite infestations. By nature, subterranean termite infestations are usually hidden behind walls and under floors such that they are hard to detect non-invasively. Acoustic detection offers the opportunity to detect infestations non-invasively by relying on the detection of termite activity such as eating, movement, nest construction or head banging by termite soldiers. Unfortunately, these acoustic and vibratory signals are weak and the attenuation of the signals through the wall or floor material is significant putting these signals at the limit of detectability. Previous work has focused on detectability in a qualitative fashion in terms of a yes/no detection of known infestations. Our work will continue to focus on improving the detection limits of acoustic sensors as well as identifying when detection is possible in terms of infestation density and location/distance of measurements to infestation. The use of alternate detection methods such as LDV’s will also continue to be investigated along with the use of infrared technology.
Our main mission is to develop instruments and devices that can help in detecting Formosan Subterranean termite (FST) infestations in a non-invasive and cost effective manner. One of the main ways to detect termites non-invasively is to monitor surface vibrations on walls infested with termite activity utilizing laser-based vibration measurements (Laser Doppler Vibrometry or LDV). Our objectives were to identify what types of acoustic signals could be detected from FST activity during a simulated infestation and quantify the detected acoustic signals during a month long infestation. Head banging (HB) was the most common signal encountered and would be the most promising for detection during termite home inspections. Since it occurred when the surface was disturbed and undisturbed, little agitation would be necessary to elicit head banging from soldiers in infested walls. Feeding behavior signals were only detected within the first 48 hours after infestation, but were barely audible and difficult to distinguish from the noise floor. Head banging (HB) levels were relatively low during first 2 days of infestation and showed the highest levels after 4 to 5 days of infestation. It is unclear why lower levels of head banging were seen after 3 to 4 weeks of infestation, but may be due to mortality levels in soldiers or a loss of aggression in colonies kept in laboratory conditions. Our next objective was to determine the sensitivity of the LDV to head banging signals at measureable distances on a mock wall (40.6 x 61 cm) infested with 5,000 termites. Head banging signals were detected at a maximum distance of 60 cm from source of infestation. HB signals at 10 cm consisted of both a low (1-2 kHz) and high (5-7 kHz) frequency components, whereas HB signals at 60 cm consisted of only the low (1-2 kHz) frequency component. The relative amplitude of the head banging signals at 60 cm was nearly half that of the HB signals at 10 cm for the low frequency component (1-2 kHz). LDV detection of HB signals should focus on the low frequency component to maximize the detectable distance. Home inspections using an LDV would require measurements to be taken from multiple points on the wall to ensure successful termite detection.