Studying many aspects of the population ecology of animals requires the ability to identify individuals. In birds, this is usually accomplished by retrapping marked individuals. However, in some cases, such as when trapping is difficult or likely to cause disturbance, alternative methods are desirable. I investigated the possibility of using the defensive “cack” call of the peregrine falcon for individual identification and then applied the technique to estimate turnover in a falcon population.
I made sonagrams of calls recorded in 1989-1991, and used three frequency variables, three temporal variables, and two derived slope variables to allocate calls among birds with discriminant function analysis (DFA). The average classification success of calls within a year using DFA was 83.5% (89% in 1990, 78% in 1991). Calls were classified to the correct sex 91 % of the time.
A nested analysis of variance was used on a sample of four birds that were recorded in two years and on two days within each year to assess stability of the calls over time. The results showed that, for the frequency variables, there were significant differences in pitch among birds and days, but not between years. Estimation of variance components, using the maximum likelihood method, was used to determine which factors contributed most to the observed variability. More than 60% of the variation in each of the frequency variables resulted from differences among birds. Variation attributable to recording birds on different days was less than 12%, and variation attributable to recording birds in different years was less than 0.0001 %. The results suggest that while an individual falcon’s voice varies within a day or among days within a year, there is much less variation in calls within a bird than there is among birds. Furthermore, it appears that an individual falcon’s calls are relatively stable among years, and could therefore be used as a method to identify individual falcons over time.
I used sonagraphic analysis, band information, and facial markings to estimate turnover and dispersal in the study population. Natal dispersal distances of four female Peregrine Falcons ranged from 63-781 km. Turnover as determined through band identification alone occurred in at least 2/11 females and 2/4 males. The use of sonagraphic analysis increased sample size slightly, and helped to categorize falcons that were suspected to be the same in successive years based on inconclusive or no band data. Estimates of gross annual turnover using band data and sonagraphic analysis were 16.7% to 25% in females and 50% (3/6) in males.