Get access

Testing life history predictions in a long-lived seabird: a population matrix approach with improved parameter estimation


  • P. F. Doherty, Jr.,

  • E. A. Schreiber,

  • J. D. Nichols,

  • J. E. Hines,

  • W. A. Link,

  • G. A. Schenk,

  • R. W. Schreiber

P. F. Doherty, Jr., Dept of Fishery and Wildlife Biology, Colorado State Univ., Fort Collins, CO 80523-1474, USA ( – E. A. Schreiber, Natl Mus. of Nat. Hist. MRC 116, Washington D.C. 20560, USA. – J. D. Nichols, J. E. Hines and W. A. Link, USGS Patuxent Wildlife Research Center, Laurel, MD 20708 USA. – G. A. Schenk and R. W. Schreiber, G. A. Schenk, 4109 Komes Court, Alexandria, VA, 22306 USA.


Life history theory and associated empirical generalizations predict that population growth rate (λ) in long-lived animals should be most sensitive to adult survival; the rates to which λ is most sensitive should be those with the smallest temporal variances; and stochastic environmental events should most affect the rates to which λ is least sensitive. To date, most analyses attempting to examine these predictions have been inadequate, their validity being called into question by problems in estimating parameters, problems in estimating the variability of parameters, and problems in measuring population sensitivities to parameters. We use improved methodologies in these three areas and test these life-history predictions in a population of red-tailed tropicbirds (Phaethon rubricauda). We support our first prediction that λ is most sensitive to survival rates. However the support for the second prediction that these rates have the smallest temporal variance was equivocal. Previous support for the second prediction may be an artifact of a high survival estimate near the upper boundary of 1 and not a result of natural selection canalizing variances alone. We did not support our third prediction that effects of environmental stochasticity (El Niño) would most likely be detected in vital rates to which λ was least sensitive and which are thought to have high temporal variances. Comparative data-sets on other seabirds, within and among orders, and in other locations, are needed to understand these environmental effects.