During the winters of 1991-92 and 1992-93, I conducted a radio-telemetry study of gyrfalcons (Falco rusticolus) wintering in central South Dakota. Five immature and three adult gyrfalcons were trapped, all of which were females. Of these gyrfalcons, adult #304 was captured and used as a study subject during both field seasons. Three of the study subjects were found dead during the study. One gyrfalcon had been shot, and a cause of death could not be determined in the other two cases.
Age structure of the population was analyzed using all recorded sightings of gyrfalcons for both winters. Of 38 sightings, 15 (39%) were adult gyrfalcons, 17 (45%) were immature gyrfalcons, and 6 (16%) were of unknown age. First observation dates of gyrfalcons were very similar in both winters, as were numbers of gyrfalcons sighted. The results of this study, combined with past banding data, indicated that the study area was regularly used by migrant gyrfalcons during winter.
Immature gyrfalcons did not establish winter home ranges in the study area. Adult gyrfalcons, however, remained within the study area during the entire field season and established well-defined home ranges. Three extracts subsets of locations for each adult were analyzed using incremental polygon clusters (Kenward 1987) and harmonic mean measures of activity areas (Dixon and Chapman 1980). Estimated home range sizes varied greatly among data sets and estimation techniques. Maximum areas were taken as 100% polygon clusters [equivalent to the 100% minimum convex polygon (Mohr 1947)], and high use areas were defined with the 85% harmonic mean isoline. Area estimates using comparable analyses were similar to reported winter home ranges for gyrfalcons in Washington (Dobler 1989). Minimum convex polygon home ranges of contemporary adults overlapped less than 10%
A pair-wise of data collected from random locations was compared to gyrfalcon locations for distance to reservoir, proximity to waterfowl, and proximity to human activity. Distance to reservoir was found to be significantly less for gyrfalcon locations, indicating a marked degree of association with the reservoir. Gyrfalcon locations did not differ significantly from random for either waterfowl or human activity. Gyrfalcons, however, were observed to associate with large feeding flocks of waterfowl when hunting, and were intolerant of close approach by humans.
Perch use was recorded and analyzed using a contingency table G-test. Gyrfalcons, as a group, selected perches different from rant availability. Significant variability, however, existed among individuals.