Minimizing the cost of translocation failure with decision-tree models that predict species’ behavioral response in translocation sites
MetadataShow full item record
The high number of failures is one reason why translocation is often not recommended. Considering how behaviour changes during translocations may improve translocation success. We used data from five simulated translocations, considering the short term responses of an endangered Australian skink when released under alternative conditions, to derive decision tree models. We used four different decision tree algorithms (decision tree, decision tree parallel, decision stump and random forest) with four different criteria (gain ratio, information gain, gini index and accuracy) to investigate how environmental and behavioural parameters might affect the success of a translocation. We assumed that behavioural changes that increased dispersal away from a release site would reduce translocation success. The trees became more complex when we included all behavioural parameters as attributes, but these trees gave us more detailed understanding about why and how dispersal occurred. Decision trees based on parameters related to release conditions were easier to follow and might be used by conservation managers to make translocation decisions in different circumstances.
This is the peer reviewed version of the following article: [Ebrahimi M, Ebrahimie E, Bull CM (2015) Minimising the cost of translocation failure by using decision tree models to predict species behavioural response in translocation sites. Conservation Biology, vol. 29, no. 4, pp. 1208-1216.], which has been published in final form at DOI:10.1111/cobi.12479. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Self-Archiving (http://olabout.wiley.com/WileyCDA/Section/id-820227.html#terms)