• breeding success;
  • reproductive cycle;
  • seabird conservation;
  • simulation

Estimating total breeding populations (I) for species that exhibit biennial breeding is generally done from counts of individuals that breed in each year (N), but can be complicated by the fact that the proportion of individuals breeding varies from year to year. Partly, this reflects the proportion of individuals that re-breed in successive years (re-breeding rate, p), which is largely, although not exclusively, governed by reproductive failure. Here we show that variation in counts of breeding individuals not only reflects changes in total breeding population but can be sensitive to and powerfully driven by variation in p. A simulation of annual field counts of a bird exhibiting biennial breeding was constructed to explore the effect of re-breeding attempts on estimations of the total breeding population. The model was used to simulate the consequences of adult mortality and different annual patterns of nesting failures on total breeding population estimates, and to explore the consequences of variation in p on N, when total breeding population remains constant. N is shown to be very sensitive to variations in p, so that even short-term fluctuations in p can cause changes in N that oscillate for many years ahead. We compare our modelled results with real data for Grey-headed Albatrosses Thalassarche chrysostoma and demonstrate that, when I is held constant in the model, actual counts may be simulated by variations in p only. Normally, I is unknown and is extrapolated from N on the assumption that N mirrors changes in the size of the total population. Consequently, applying average values of p can result in misleading estimates of total breeding population. We recommend that annual counts of breeding individuals are supplemented with annual estimates of p. Field protocols that aim to estimate annual breeding population size from counts of breeding individuals should be complemented by independent measures of rates of re-breeding and nest failure.