PARENT SESSION

Natural History 1 -- Session Chair: Patrick Zollner-- Nelson Hall East, Goodwin Forum

ADVANCES IN THE ACOUSTIC MONITIORING OF BATS. Joseph M. Szewczak. Department of Biological Sciences, Humboldt State University, 1 Harpst Street, Arcata, CA.

ABSTRACT- Despite the ready abundance of acoustic information that bats provide, confident species recognition from this information has remained an elusive goal. Calls from bats that we think we know have been rendered uncertain when subsequent studies reveal a sympatric species capable of emitting similar call variants. Surmounting this challenge requires new analysis methodology and incorporating acoustic information beyond the basic time-frequency relationships on which many studies have concentrated. This limitation was necessitated from the practical need to reduce the burden of data inherent in digitizing ultrasound (approx. 300,000 data points per second) to a manageable scale. However, digital data storage and processors have advanced to enable practical analysis of bat vocalizations using full-spectrum acoustic information, as is common to other bioacoustic endeavors. Full-spectrum data also permits an evaluation of the acoustic environment for noise and other factors that can lead to spurious results if left unknown. The full–spectrum approach can glean new species–specific characteristics as we build larger datasets of acoustic information and call libraries. While many studies have analyzed call structure using static variables such as maximum and minimum frequency, call structure actually consists of a continuous multivariate effect of frequency, time, and power(amplitude). Using digital signal processing, related continuous acoustic parameters may be combined to produce transforms of time–varying interrelationships that can be represented in two–dimensional plots. These transform plots can render striking visual contrasts between species with inexplicit distinctions in standard time-frequency plots. Finally, full-spectrum acoustic processing holds the promise to enable automated species recognition using intelligent computing systems.

KEY WORDS: bats, acoustic information