A Spectrographic Analysis of Burrowing Owl Vocalizations

ON. of the first studies of the Burrowing Owl (Speotyto cunicularia) appeared almost 100 years ago (Coues 1874). Most of the many brief accounts published since deal with its food habits or are casual observations of behavior. Recently Thomsen (1971) and Coulombe (1970, 1971) published major studies of this species, Thomsen on its food habits, behavior, and population dynamics, Coulombe on its physiological attributes, food habits, behavior, and seasonal movements. Neither author spectrographically analyzed any of the vocalizations they described. The main purpose of this study is to analyze Burrowing Owl vocalizations, the first such analysis of any North American member of the Strigiformes. Also Burrowing Owl vocalizations, primarily song, are compared to information gathered from passerines to illustrate that other orders of birds may follow the general rules by which passerine song appears to be structured. METtIODS AND MATERIALS I studied a population of 15 breeding pairs of Burrowing Owls and their offspring from 5 May 1970 to 20 May 1971 near Albuquerque, Berna]il]o County, New Mexico. Observation time totaled over 400 hours. Most observations were made from a car or on foot and were aided with 7 X 50 binoculars and a 30 X spotting scope. A portable blind was used for close observations and for recording vocalizations. As lights disturbed the owls, none were used. The study area contained two concentrations of nesting holes 3 rni]es south of Albuquerque at an elevation of 5,300 feet. One was a 1.8-rni]e-]ong section at Tijeras Arroyo, the second a 1/2-rnile-long section where a railway cut through a hillside. The two sites were 1/2-rnile apart and the intervening land served as mutual foraging ground. Voca]izations were recorded with a Uher 4000 Report-L Tape Recorder and Uher M-514 microphone at 7-2 ips. A 24-inch parabolic reflector sometimes was used. Spectrograms were produced with a Kay Electric Company Sonagraph (6061-B) equipped with an Arnplitude Display Unit (6076-C). Data for time analysis came from frequency-time spectrograrns, using wide band-pass setting. Those analyzed for arnp]itude were produced at a narrow band-pass setting with the Amplitude Display Unit. Frequency analysis came from frequency/time spectrograrns at a narrow bandpass setting. A spectrogram using a Sona-Marker Time Marker and frequency marker units was used as a template for measuring various parameters of the voca]izations accurately. From this a transparency with frequency and time increments was produced and used as an overlay to measure spectrogram prints. Of 15 breeding pairs, 9 fernales and 9 males were color-banded for individual recognitions. At least one adult of each pair was banded at a]] but two burrows. One