Sampling considerations for demographic and habitat studies of Northern Goshawks

Abstract

We used mark-recapture methods to monitor Northern Goshawks (Accipiter gentilis) and their nests over 12 yr in an increasing sample of breeding territories (37 in 1991 to 121 in 2002) in northern Arizona. As many as 8 yr of repeated nest searching were required to identify the population of breeders, as individuals skipped egg-laying on territories for up to 7 consecutive yr. Extensive temporal (within territory) and spatial (among territory) variation in reproduction and a high annual frequency of movements among dispersed alternate nests in territories made finding and monitoring goshawks problematic. Low detectability of nonbreeding goshawks (combined with uncertainties stemming from variations in breeding and use of alternate nests) made it difficult to categorize territories unequivocally as ‘‘unoccupied’’ by goshawks in non-egg-laying years. Temporal and spatial variation in reproduction required large numbers of territories to attain reliable estimates of reproduction of goshawks; such estimates were achieved only when samples approached or exceeded 60–100 territories. Our within-territory goshawk searching protocol, designed to increase the low and variable detectability of goshawks, required extensive sampling efforts to detect among-alternate nests movements. In lieu of such efforts, samples of territories occupied by goshawks may ‘‘decay’’ over time and lead to false inferences of population declines. Low detectability, variations in breeding, and large samples require that demographic and habitat studies of goshawks employ intensive and repeated searches for goshawks in large study areas over at least 8 yr.

Citation

Reynolds, Richard T.; Wiens, J. David; Joy, Suzanne M.; Salafsky, Susan R. 2005. Sampling considerations for demographic and habitat studies of Northern Goshawks. Journal of Raptor Research. 39(3): 274-285.