Assessing predictions of population viability analysis: Peregrine Falcon populations in California
Population viability analysis (PVA) has been an important tool for evaluating species extinction risk and alternative management strategies, but there is little information on how well PVA predicts population trajectories following changes in management actions. We tested previously published predic...
| Main Authors: | , |
|---|---|
| Format: | Article in Journal/Newspaper |
| Language: | unknown |
| Published: |
Figshare
2016
|
| Subjects: | |
| Online Access: | https://dx.doi.org/10.6084/m9.figshare.c.3296333 https://figshare.com/collections/Assessing_predictions_of_population_viability_analysis_Peregrine_Falcon_populations_in_California/3296333 |
| Summary: | Population viability analysis (PVA) has been an important tool for evaluating species extinction risk and alternative management strategies, but there is little information on how well PVA predicts population trajectories following changes in management actions. We tested previously published predictions from a stage-structured PVA of Peregrine Falcons ( Falco peregrinus ) in California, USA (Wootton and Bell 1992), against population trajectories following the 1992 termination of statewide, active management (population supplementation of captive-reared young). In the absence of extensive post-management monitoring, we developed surrogate estimates of breeding population size by calibrating several citizen science data sets (Christmas Bird Count, CBC; and North American Breeding Bird Survey, BBS) to intensive population surveys taken primarily during the active management period. CBC abundance data standardized by observer effort exhibited a strong relationship to intensive survey data ( r 2 = 0.971), indicated significantly reduced annual population growth rates after management was terminated (λ = 0.023 ± 0.013 SE) than when supplementation occurred (λ = 0.089 ± 0.023 SE), and demonstrated an increasing population as predicted by the PVA. The population trajectory fell within the 95% CI of stochastic simulations of the model either with or without density dependence and assuming either measurement error or process error, but models with process error were most strongly supported by the data. These results indicate that PVA can quantitatively anticipate population trajectories following changes in management, highlight the importance of post-management monitoring of species of concern, and illustrate the benefits of using management changes as large-scale experiments to more rigorously test PVA. |
|---|