Abstract
Wire cages (exclosures) that exclude nest predators are regularly used for management of federally threatened and endangered populations of piping plovers (Charadrius melodus) and similar species. However, recent evidence that apparent nest abandonment related to exclosure use is indicative of adult mortality has led to concern about the technique. We developed a decision support tool named PiperEx that uses site-specific nest-fate data to inform a stochastic population projection model to predict population growth rate at the site level with and without exclosure use. Underlying PiperEx is a Bayesian logistic-exposure multinomial nest fate model, with informative priors based on analysis of 1,312 piping plover nests monitored throughout the U.S. Atlantic Coast in 2015 and 2016. We used simulated nest fate data sets and real nest data from 3 management areas from 2013 to 2018 to test tool performance. Based on simulations, the probability of making the correct decision increased with sample size, exceeding 80% with 20 nests. Decisions recommended by the tool were robust to variation in demographic parameters. Using real data, the decision for a given year only matched the decision for the immediately previous year 16% to 67% of the time, whereas data pooled across years predicted the best decision for a particular year up to 100% of the time for a given area. We used a case study of Edwin B. Forsythe National Wildlife Refuge to demonstrate that the optimal decision is sensitive to how a site was defined. We recommend that data for PiperEx be collected on an annual basis, and data for the previous 5 or 6 years be pooled for making a decision at the start of the season. Further development of the tool should focus on its use to make regional recommendations, and use of multiple years of data to improve the capability of the tool to make robust site-specific recommendations. (c) 2020 The Wildlife Society.