bats

Dolphins possess a highly sophisticated auditory system and a keen capability for echolocation. Signals are emitted in the form of high intensity, short duration, broadband exponentially decaying pulses. The frequency spectra of echolocation signals used by many dolphins are dependent on the output intensity of the signals and not on any fine tuning by the animals. When the output intensity is low, the center frequency of the click tends to be low. As the output intensity increases, the center frequency also tends to increase. The pulses propagate from the dolphin's melon in a relatively narrow beam, and echoes are received via the lower jaw, with a slightly wider beam. Echolocating dolphins can detect targets at ranges of approximately 100 plus meters, depending on the size of the targets. Target discrimination experiments have shown that dolphins can discriminate the shape, size, material composition and internal structure of targets from the echoes. The broadband, short duration properties of the signal allow the echoes to have high temporal resolution, so that within the structure of the echoes a considerable amount of information on the properties of the target can be conveyed. A brief comparison between the bat and dolphin sonar system will also be made. Bats typically emit much longer signals and a wider variety of different types of signals than dolphins. Signals used by some bats are suited to detecting Doppler shift, whereas the dolphin signal is designed to be tolerant of Doppler effects.

1. A method for the identification of bat species from time-expanded broad-band recordings of their echolocation calls is presented. The method may be used for the assessment of habitat use by bats.
2. Recordings were made of echolocation calls produced by 536 bats of known species identity, belonging to 15 species found in Great Britain. One call was analysed per individual, and sonograms and descriptive statistics of six time and frequency variables of calls are presented. British bats can be placed in three groups according to the structure of their calls: high duty cycle FM/CF/FM bats (Rhinolophus spp.), low duty cycle FM bats (Myotis spp. and Plecotus spp.) and intermediate duty cycle FM/CF bats (Pipistrellus and Nyctalus spp. and Eptesicus serotinus).
3. FWCF/FM bats could be identified from the peak frequency of their calls. Two separate quadratic multivariate discriminant analyses were carried out on the time and frequency parameters of calls produced by FM bats and FM/CF bats. For FM bats 67%, and for FM/CF bats 89%, of unknown calls were classified to species.

The use of bat detectors to monitor bat activity is common. Although several papers have compared the performance of different brands, none have dealt with the effect of different habitats nor have they compared narrow- and broad-band detectors. In this study the performance of four brands of ultrasonic bat detector, including three narrow- band and one broad-band model, were compared for their ability to detect a 40 kHz continuous sound of variable amplitude along 100 metre transects. Transects were laid out in two contrasting bat habitat types: grassland and forest. Results showed that the different brands of detector differed in their ability to detect the source in terms of maximum and minimum detectable distance of the source. The rate of sound degradation with distance as measured by each brand was also different. Significant differences were also found in the performance of different brands in open grassland versus deep forest. No significant differences were found within any brand of detector. Though not as sensitive as narrow-band detectors, broad-band models hold an advantage in their ability to identify species where several species are found symmetrically.

Bat detectors are commonly used to monitor bat behaviour. Earlier research has suggested that there may be systematic differences in the response of different detectors to bat calls. Such differences would have important implications for the comparability of quantitative surveys conducted with bat detectors. The present study examines variability within and between brands of bat detector in accuracy of tuning, directionality and sensitivity to different types of bat echolocation call in bat detectors from three manufacturers. The consistency of results from a field survey incorporating the three brands in a standardised methodology are also examined. Significant differences were found within and between brands in directionality and sensitivity which would lead to bias in bat surveys. The implications of these findings for bat surveys are discussed, as are the design features of importance for species identification.

We recorded echolocation and ultrasonic social signals of the bat Myotis septentrionalis. The bats foraged for insects resting on or fluttering about an outdoor screen to which they were attracted by a 'backlight'. The bats used nearly linearly modulated echolocation signals of high frequency (117 to 49 kHz) with a weak second harmonic. The orientational signals from patrolling bats were about 2.4 ms in duration and occurred at a repetition rate of about 18 Hz. The signals used by bats as they approached the screen were of shorter duration (0.72 ms) and occurred at higher rates (33.8 Hz). We registered one feeding 'buzz'. We recorded social signals when two bats patrolled the hunting area. The social signals were characterized by their longer durations (6 ms), lower frequencies (70 to 30 kHz), and curvilinear sweeps. We calculated the source levels of orientational and social signals using the differences in arrival times at three microphones in a linear array. The source levels were on average 102dB peSPL at 10 cm. We could not calculate source levels of the signals used by bats as they approached the screen at close range, but these signals were much weaker (about 65d8 peSPL at the microphone).

Bats of the genus Myotis cannot be identified reliably using conventional acoustic analyses. Here we use morphology of echolocation calls to discriminate between Myotis spp. This method may be used to identify unknown roosts to species level in the field. Echolocation calls of M. daubentonii, M. mystacinus and M. nattereri, were recorded in emergence flights from roosts. Images of echolocation calls were extracted for morphological analysis performed in SHAPE, a program that transforms twodimensional outline data into Elliptic Fourier Descriptors. Species typical call shapes were described with Mahalanobis models. Discriminant Function Analyses (DFA) were applied with Mahalanobis scores of typical shape alone and with a spectral call parameter, maximum frequency. DFA achieved an overall correct classification rate of 88.9% using typical outline shapes alone. Correct classification of 100% of both M. daubentonii and M. mystacinus was achieved by both typical call outlines. For M. nattereri, 79.6% of calls were correctly classified by call morphology, but the addition of maximum frequency improved this to 96.3%. Shape analyses provide a quick and easy method of distinguishing Myotis species under field conditions and could be extended to include other species of bats that share conventional acoustic parameters.