Predicting successful replacement of forest invaders by native species using species distribution models: The case of Pittosporum undulatum and Morella faya in the Azores

Invasive alien species have taken the place of native forest species worldwide, and particularly in island systems like the Azores. Species distribution models could be used to support the design of control strategies, including the definition of target areas. In this study we used species distribution modelling, performed with the ecological-niche factor analysis (ENFA) approach. Pittosporum undulatum is the most widespread woody invader in the Azores where it is considered as one of the priority species for control actions. On the other hand, Morella faya is a native woody plant facing several threats, including plant invasions. This paper aims to evaluate whether and where areas currently occupied by P. undulatum could also be favorable habitat for M. faya, thus providing support for future management actions. Occurrence data on P. undulatum and M. faya from a forest inventory were randomly split into training and testing datasets. Twenty six ecogeographical variables were used together with the training datasets in modelling. The modelling results were validated based on the continuous Boyce curve calculated with the testing datasets. We selected and compared the models of the highest predictive power and robustness for both species to define where M. faya could inhabit areas currently occupied by P. undulatum. The two species have quite similar environmental preferences, which correspond mainly to coastal lowlands and forested habitats characterized by relatively high temperature and wide ranges of low relative humidity values. The main difference between both species is that M. faya was considered more specialized according to current data. M. faya could potentially be used to replace P. undulatum in up to 24% of the Azorean area currently invaded, mostly in São Jorge and Pico Islands. According to our results, the ENFA modelling approach revealed to be a useful tool in the design of more efficient management strategies for woody plant invaders.

► P. undulatum and M. faya are woody invasive and native species in the Azores, respectively.
► Modelling was used to predict and compare both species habitat suitability.
► Both species prefer lowlands of high temperature and low relative humidity.
► M. faya could replace P. undulatum in up to 24% of the Azorean area currently invaded.