- York University
Summary
Although bat detectors are used with increasing frequency for the identification of bats in the field (Aldridge and Rautenbach, 1987; Crome and Richards, 1988), only one study comparing performance of different models has been published (Downes, 1982). Some bat detectors are tunable through different frequencies, but if the tuning is not accurate, results obtained with different instruments may not be comparable. The purpose of this study was to examine and to compare the performance of three models of bat detectors. We used two QMC Mini bat detectors, two QMC Model S200 bat detectors (QMC Instruments Ltd., 357 Mile End Road, London E1 4AA, England), and one Skye Instruments bat detector (Skye Instruments Inc., Llandrindod Well, Powys, Wales LD1 6DF) and considered accuracy of tuning, threshold at different frequencies, and detectability of echolocation calls of four species of bats.
Methodology
To determine accuracy of tuning, we compared responses of the detectors to mechanically generated sounds. Sine-wave sounds were produced by an Exact Model 126 (Exact Electronics Corp., Hillsboro, OR) sweep generator and a Technics speaker (Model 10TH400B). We monitored the frequency of signals with a Non-Linear Systems FM7 frequency counter (Non-Linear Systems, Delmar, CA), and their amplitude with a Textronix 212 oscilloscope (Tektronix Inc., Beaverton, OR). We tested responses of the detectors to signals of 20, 30, 40, and 60 kHz. We successively placed each detector 1.5 m directly in front of the speaker and tuned it until it emitted a clear, sharp signal. We recorded the frequency to which the detector was tuned and compared this value to the frequency of the emitted signal. This process was repeated twice for each detector at each test-signal frequency. All battery-operated instruments received new batteries at the beginning of the tests.
To determine threshold sensitivity of the bat detectors at each frequency, we used the oscilloscope to measure the weakest signal (in millivolts) required to elicit a response from the detector. During this procedure we did not change the gain settings on the detectors.
We compared the performance of the instruments in detecting echolocation calls of Myotis lucifugus (n = 7), M. septentrionalis (n = 3), M. leibii (n = 4), and Pipistrellus subflavus (n = 4) captured in a Tuttle trap (Tuttle, 1974) at a swarming site in eastern Ontario. We brought the bats to the Queen's University Biological Station (Chaffey's Locks, Ontario) and housed them in cylindrical cages with water ad lib. During daylight hours, bats were flown individually in an 8.9-m long by 5.0-m wide by 2.4-m high room for 8–10 min each or until we had tested each bat with each detector. For each detector we recorded the maximum distance at which the echolocation calls of each bat were detectable when each detector was tuned to the frequency of strongest output (from the detector). We analyzed the data by use of the Statistical Analysis System (SAS Institute, 1985).