Recipient Organization
BEE ALERT TECHNOLOGY, INC.
1620 RODGER ST., SUITE #1
MISSOULA,MT 59802
Performing Department
(N/A)
Non Technical Summary
This project is aimed at helping plant production and protection by providing a means of rapid, autonomous reporting of pesticide incidents impacting honey bee colonies. We propose research leading to the design of an electronic device to monitor bee colonies for exposure to pesticides, specifically neonicotinoids in Phase I, and to remotely notify appropriate parties (e.g., beekeeper, grower, researcher) of the event. Project objectives include research to: 1) subject honey bees to controlled pesticide exposures and record the acoustic responses, 2) construct an Artificial Neural Network to discriminate changes in bee sounds for different chemical categories of pesticides, and 3) design an autonomous system to remotely alert the beekeeper or relevant agency.
Animal Health Component
30%
Research Effort Categories
Basic
(N/A)
Applied
30%
Developmental
70%
Goals / Objectives
Subjecting honey bees to controlled chronic and acute pesticide exposures and recording audio.Construction of Artificial Neural Networks (ANNs) specifically tuned for pesticide exposure events.Design the autonomous monitoring device to include the necessary communication and monitoring hardware.Testing the operation of the hive-mounted monitoring device
Project Methods
Testing will be performed on both chronic and acute exposures. Separate testing using an isolation chamber will determine the dose amounts to deliver chronic and acute exposure events.We will use two different methods of introducing the pesticide to the colony; ingestion and contact. For ingestion, the pesticide will be dissolved in sucrose water for direct uptake by the bees. For contact, a nebulizer will be used to introduce a vapor to the bees. Each test will be conducted on 6 different bee colonies, with 6 control colonies included for each dosing level.Audio collection will consist of a digital Marantz PMD670 recorder (saves all data to solid-state memory flash cards - no tape), a microphone amplifier (SME 2100) from Saul Mineroff Electronics and two 1/8 inch diameter, 12 inch long, probe microphones. We will also be using our ultrasonic recording system developed.Audio data will be preparded for analysis using the Stuttgart Neural Network Simulator, a powerful tool for designing and testing Artificial Neural Networks. For the Phase I project, verification of the ANN accuracy will be limited to 1) the ANN training results (how well the ANN has learned the pattern), and 2) percentage of correct identification of test signals.These devloped algorithms will be incorporated into an embedded system of our design with a satellite communication option.