Workshop on measurement, techniques and questions for observations of echolocating animals 

September 13, 2009, Kambaikan Doshisha University, Kyoto

James A. Simmons, Shizuko Hiryu, Annemarie Surlykke, John Hallam & Herbert Peremans

Theme 1

Workshop on Measurement, Analysis and Simulation of Directional Acoustics for Biosonar

ChiRoPing. John Hallam, Annemarie Surlykke, Herbert Peremans

Recent years have seen increasing awareness of the importance of shape-dependent directionality in bat biosonar. Techniques are becoming available for simulating the acoustic beam-patterns generated by facial anatomy and use of array-measurement systems, able to capture remote recordings of bat vocalisations across a large area, is growing in field and laboratory studies. Similar approaches are also being pursued in recent underwater bioacoustic work.

Complementing this, work in computational modelling of animals' sensorimotor systems has for some years been emphasizing the importance of embodiment, in particular the exploitation of physical shape and sensor deployment to simplify the computational burden placed on the nervous system.

This workshop provides the opportunity for researchers from both computational and from biological backgrounds to come together to discuss their respective techniques and explore their potential for generating compelling explanations of biosonar behaviour and strategies.

Theme 2:

Workshop on equipment, techniques and questions for field observations of echolocating bats 

Jim Simmons, Shizuko Hiryu

For most of the past 70 years, essentially since the discovery of echolocation, our understanding of the behavior of echolocating bats in natural conditions has depended largely on listening to bats with bat detectors, which translate the bat’s ultrasonic sonar sounds into human-audible displays.  The most widely-available equipment for acoustic pick-up has consisted of either heterodyning or zero-crossing bat detectors, which began as cumbersome laboratory devices but have progressed to hand-held field-portable tools for convenient observation of bat behavior.  Once miniaturization of commercial devices took effect, various types of bat detectors have become ubiquitous.  A web search for “bat detectors” today yields a plethora of devices at costs ranging from about $50 US to $5000 US depending on their capabilities.  These include commercial instruments from Pettersson Elektronik, Titley Electronics, Ultra-Sound Advice, Magenta Electronics, and Alana Ecology, plus a number of “build-it-yourself” designs that are available as kits.   For several decades, the audio-frequency signals from bat detectors have been recorded on audio cassette tape recorders to better serve the needs of documentation.  The most recent bat-detector designs offer digital recording (see below). 

From the beginning of work on echolocation, recordings of the ultrasonic signals themselves were made on analog tape recorders with higher tape speeds than typical audio recorders.  The earliest field work was done with bulky Ampex recorders that were only notionally “field portable,” but the advent in the 1960s and 1970s of two lighter-weight instruments—the Pemtek tape recorder and the Racal recorder—have given us clear records for most of the different types of echolocation signals used by bats.  The advent of digital recording methods has led to several instruments being available in the last 10 years or so that offer the ability to record the ultrasonic waveforms on laptop or notebook computers, or on digital storage cards originally developed for use in cameras.  These include the Pettersson and Avisoft devices.  Ultrasonic microphones often now are the weak link in the field-equipment chain, because condenser microphones can be too sensitive to damage from humidity and dust.  The Titley bat detector is equipped with a particularly rugged ultrasonic microphone (an encapsulated Polaroid transducer) that is easily incorporated into other systems.  For ultrasonic signal recording, the practical constraint that ultrasonic frequencies higher than 80-90 kHz do not ordinarily survive the journey through the air from the bat to the microphone can be exploited with the best current audio technology to simplify field studies.  Several sturdy, very small commercial digital audio recorders now offer sampling rates up to 192 kHz on 2 to 4 channels, including units from Audio Devices, Tascam, and Foster.  The recording medium is a small high-capacity flash card.  These are effective solutions for nearly all field recording problems that require one or more channels of ultrasonic signals, and they can be used for recordings with microphone arrays.  At present, the top-of-the line purpose-built instrument is the Sony SIR-1000W recorder, which has up to 4 channels of 384-kHz digital sampling, or 8 channels of 192-kHz sampling.  (Tascam makes a similar high-end multi-channel device.)  With its plug-in video card, the Sony recorder gives video capability at the cost of halving its analog signal sampling capabilities. 

In the early studies of aerial interceptions of insects or insect evasive action, strobe-flash photography, sometimes with synchronized acoustic recording, was indispensable for direct documentation of the behavior, which, however, was not often done in the field for some years thereafter.  When night-vision optical devices became available, several studies capitalized on this technology, but the results usually were described as notes of observations because video recording equipment still was inconvenient to use and bulky.  In the past 15 years, a wide range of sophisticated acoustic, photographic, and video equipment has been brought to bear on studying echolocation in the field.  These methods provide the means to document the bat’s behavior along with the sounds.  First, the strobe-flash photographic method has been revived using infrared flashes and two cameras to build multiple-exposure stereo records of flight and interception.  Both the bat’s flight-path and the sound sequence can be reconstructed in time and three dimensions.  For video, night-vision lenses, infrared illumination, and thermal-imaging video cameras are available.  They offer different capabilities and limitations.  Some infrared cameras with night-vision lenses work by themselves as camcorders, but the thermal-imaging cameras usually require a separate video recording device.  Almost any camcorder with external video input, or a Sony Video Walkman, will do.  Combined with recording the output of a bat detector on the video recorder’s soundtrack, infrared or thermal-imaging video is especially useful for understanding the kinds of behavior that occur with particular patterns of echolocation sounds. 

One special category of recording technique is the use of a microphone array to track the bat from its sounds.  Some array methods using 4 to 8 microphones are being used for full-scale three-dimensional tracking of bats over a wide area while they forage.  Other array methods concentrate on a smaller area and try to track the bat with fewer microphones.  Some array techniques have also been used to record sounds from flying bats in the lab, to estimate broadcast directionality or to ensure that no sounds are missed during turns in flight.  One useful method is to combine a 2- or 3-microphone array with video recording to locate the sources of sounds within video images of multiple bats.

Another new technique is the use of a radio microphone (Telemike) to record the flying bat's sounds with an on-board microphone and miniature radio transmitter.  The current device weighs only 0.6 grams and can be carried by a pipistrelle!  Several new findings have emerged from this procedure, and one intention is to discuss these at the Kyoto meeting.

The above is a summary of methods that I know about, and the idea is that, at the meeting, or by email prior to the meeting, we fill out this summary with whatever additional information you want to include.  I know that many of you are using novel methods involving acoustic and video recordings that everyone else will want to know about, too.   The final document will be made available on the meeting website (http://cse.fra.affrc.go.jp/akamatsu/AnimalSonar.html ).