| Publication Type: | Journal Article |
| Year of Publication: | 2003 |
| Authors: | Gorman, LR, Rosenberg, DK, Ronan, NA, Haley, KL, Gervais, JA, Franke, V |
| Journal: | The Journal of Wildlife Management |
| Volume: | 67 |
| Issue: | 3 |
| Date Published: | 2003 |
| ISBN Number: | 0022541X |
| Keywords: | Athene, Athene cunicularia, Strigidae |
| Abstract: | Obtaining reliable estimates of absolute and relative reproductive rates is challenging for avian species whose nests are difficult to observe, such as the burrowing owl (Athene cunicularia). We compared methods for estimating reproductive rates of burrowing owls, defined as the number of 21- to 28-day-old young per successful nest. We compared observations using (1) the mean and (2) the maximum number of young observed during 5 30-min observation periods, and (3) the maximum number of young videotaped during 2-hr video surveillance. We evaluated the reliability of these methods with the known number of young present in nest boxes. All 3 methods performed poorly as estimators of absolute reproductive rates (absolute bias >23%, root mean square error [RMSE] >42%). Video surveillance performed most poorly of the 3, with a high incidence of failing to detect any young at successful nests. The maximum number of young observed from direct nest observations was correlated with the known number of young (r = 0.82 ± 0.13, n = 21) and provided more reliable estimates of relative than absolute reproductive rates. The mean number of young observed from direct observations was correlated with the known number of young (r = 0.64 ± 0.18, n = 21), but had both higher bias and lower precision than the maximum number observed for estimation of relative reproductive rates. Our results suggest that using counts of young observed outside of the nest burrow may lead to incorrect conclusions on factors affecting reproductive rates. When counts are the basis of inference, the effort researchers use at each nest should be standardized and reported. Further work on field methods that allow estimation of detection probability, or ensuring that all young are observed, will be imperative in providing reliable estimates. |
| URL: | http://www.jstor.org/stable/3802707 |
| Short Title: | The Journal of Wildlife Management |
Taxonomic name: