A critical appraisal of population viability analysis

Abstract

Population viability analysis (PVA) is useful in management of imperiled species. Applications range from research design, threat assessment, and development of management frameworks. Given the importance of PVAs, it is essential they be rigorous and adhere to widely accepted guidelines; however, quality of published PVAs is rarely assessed. We evaluated the quality of 160 PVAs of 144 species of birds and mammals published in peer‐reviewed journals from 1990 to 2017. We hypothesized that PVA quality would be lower with generic programs than with custom‐built programs; be higher for those developed for imperiled species; change over time; and be higher for those published in journals with high impact factors (IF). Each included study was evaluated based on answers to an evaluation framework containing 32 questions reflecting whether and to what extent the PVA study adhered to published PVA guidelines or contained important PVA components. All measures of PVA quality were generally lower for studies based on generic programs. Conservation status of the species did not affect any measure of PVA quality, but PVAs published in high IF journals were of higher quality. Quality generally declined over time, suggesting the quantitative literacy of PVA practitioners has not increased over time or that PVAs developed by unskilled users are being published in peer‐reviewed journals. Only 18.1% of studies were of high quality (score >75%), which is troubling because poor‐quality PVAs could misinform conservation decisions. We call for increased scrutiny of PVAs by journal editors and reviewers. Our evaluation framework can be used for this purpose. Because poor‐quality PVAs continue to be published, we recommend caution while using PVA results in conservation decision making without thoroughly assessing the PVA quality. Article Impact Statement: Quality of PVAs is low because published guidelines are not followed and because of unskilled use of generic PVA programs. This article is protected by copyright. All rights reserved