Estimation of habitat‐specific demography and population growth for peregrine falcons in california

American Peregrine Falcons in California (Falco peregrinus anatum) have been managed in one of the largest reintroduction programs in the history of endangered species management. However, as for most other widespread natural populations, the spatial and temporal dynamics of California Peregrines have remained poorly understood. Long‐term monitoring data on this highly managed species present a unique opportunity for retrospective analysis of the factors contributing to the successful recovery of Peregrine Falcons in California, as well as demographic differences between habitat types. We used a newly developed mark–recapture model (the Barker model), which allows the simultaneous use of recaptures, dead recoveries, and live resightings, to provide estimates of first‐year, second‐year, and adult survival of Peregrine Falcons in the mid‐ and south‐coast regions of California. Annual survival rates for second‐year and adult Peregrines were estimated at 86%. Through model fitting, we show positive effects of urban habitats on first‐year survival. In our best‐fit models, first‐year birds fledged in urban areas had a survival rate of 65%, whereas rural (non‐urban) first‐year survival was only 28%. The introduction method also influenced first‐year survival after dispersal; in rural areas, estimated survival of hacked young (after independence) was lower than estimated survival of wild‐reared young. We also show that birds in urban habitats have significantly higher fecundity rates than birds in rural habitats, even though the fecundity of rural breeders has increased significantly over the last two decades. We argue that the strong habitat differences in first‐year survival combined with lower fecundity rates in rural habitats (due to slower improvement in eggshell thinning rates) facilitates spatial structuring of the California Peregrine population. Matrix population models constructed for both rural and urban habitats support this assessment. The temporally averaged population growth rate in urban habitats was estimated as λ = 1.28, compared to λ = 0.99 in rural habitats. Yearly analytical λ values in rural habitat predict declining population growth (λ < 1) throughout the 1980s and increasing population growth (λ > 1) in the 1990s due to improved reproductive performance. These results indicate that the introduction effort was pivotal in recovering the rural population in this portion of its former habitat, because intrinsic growth rates alone would have been insufficient to yield the observed population recovery.