Survival and philopatry of female redheads breeding in southwestern manitoba

Understanding the relationships between survival and philopatry of adult and juvenile waterfowl is fundamental information for developing effective management programs. We used combined recovery-resighting models to estimate band-reporting (r), fidelity (F), survival (S), and resighting (p) probabilities of female redheads (Aythya americana) breeding in southwestern Manitoba, Canada, during 1983-1993. We banded and nasal-marked 194 hatch-year (HY) females during 1983-1988 and 149 after-hatch-year (AHY) females during 1984-1990 and subsequently obtained 298 resightings and 26 band recoveries during 1984-1993. Recovery data were sparse, and band-reporting rates were best modeled as a constant (r= 0.082, 95% CI: 0.055 to 0.121). Breeding-site fidelity (F= 1 - permanent emigration) averaged 0.899 (95% CI: 0.703 to 0.971) across age classes, or 0.744 (95% CI: 0.287 to 0.955) and 0.923 (95% CI: 0.665 to 0.986) for models that treated HY and AHY females as different. Annual survival rates varied by age, averaging 0.417 (95% CI: 0.311 to 0.532) for HY females and 0.722 (95% CI: 0.610 to 0.812) for AHY females in a model with constant F, or 0.487 (95% CI: 0.273 to 0.705) and 0.708 (95% CI: 0.597 to 0.799), respectively, in models with age-specific estimates of F. Resighting estimates varied by age and year; averaging 0.513 (95% CI: 0.263 to 0.764) for HY females returning as yearlings and 0.724 (95% CI: 0.604 to 0.843) for AHY females. We suspect that lower resighting rates by HY versus AHY females were the result of greater temporary, emigration by HY females (i.e., lower breeding propensity), and that some among-year variation in resighting rates was also caused by temporary emigration (i.e., drought avoidance). Our data demonstrate that temporary and permanent emigration are important processes affecting local redhead populations, and that failure to model these processes can lead to biased estimates of survival, especially among HY females.