Frank T. Awbrey
Biol. Dept., San Diego State Univ., San Diego, CA 92182-0057
Sound-pressure levels of animal sounds have been measured in so many different ways that comparison between studies is very nearly impossible. The transient, repetitive nature of most vocalizations means that standard sound level meters may yield very different numbers, depending on meter time constant and frequency weighting. Integrating sound level meters, which average sound over various periods and measure sound exposure level directly, are an improvement, but are hard to use for measuring single, transient sounds such as a frog call or an echolocation pulse. A better way, with brief signals, is to record them through calibrated transducers, along with calibration tones of known amplitude, onto tape recorders with linear response to varying levels. These recordings can then be measured with digital spectrum analyzers in a way that is reliable, accurate, and repeatable. Possible measures include peak, octave, one-third octave, maximum, time-average, sound exposure, and spectrum levels. Comparability requires attention to and reporting of important details, such as averaging times, filter bandwidth and weighting functions, and adherence to standards for measuring and reporting sound.