Limited contributions of released animals from zoos to North American conservation translocations/Methods

by Typhenn A. Brichieri-Colombi, Natasha A. Lloyd, Jana M. McPherson, Axel Moehrenschlager

Methods

We used the North American Conservation Translocation (NACT) data set compiled by Brichieri-Colombi and Moehrenschlager (2016). This data set comprises publications on North American conservation translocations involving terrestrial, marine and freshwater animals published between 1974 and December 2013. Publications were compiled using the ISI Web of Science and Academic Search Complete search engines, which primarily identified journal publications but also some grey literature, such as agency reports and newspaper and magazine articles. We searched all publications in the data set for the words “zoo∗,” “aquarium,” “safari,” and “society.” We selected these terms by running a word frequency query on the names of all AZA accredited institutions to ensure that our search captured the majority of zoos; 85% of AZA-accredited institutions include one or more of these terms in their name. We selected all articles that contained the terms in the author affiliation, abstract, main text, or acknowledgements sections and then recorded the species involved, type of conservation translocation, zoo name or names, type of zoo involvement (e.g., captive breeding, authorship, funding, veterinary care, etc.), source population (captive or wild), and release location. We included articles that mentioned zoo involvement but did not mention the name of the zoo, but we eliminated articles that only included a general statement about zoos (e.g., “Zoo and botanic garden managers are rethinking their organizational missions”). We used R statistical software (R Core Team 2016) to derive descriptive statistics and run Pearson’s correlations on trends over time. Although we recognize that year of publication is an imperfect indicator of the timing of conservation translocations, we do not believe the variable time lag between implementation and publication introduces systematic biases in our analyses.