Abstract
One of the crucial choices when modelling species distributions using pseudo-absences approaches is the delineation of the background area to fit the model. We hypothesise that there is a minimum background area around the centre of the species distribution that characterizes well enough the range of environmental conditions needed by the species to survive. Thus, fitting the model within this area should be the optimal solution in terms of both quality of the model and execution time.
MinBAR is an R package that calculates the optimal background area. The version 1.0.0 is implemented for MaxEnt and uses Boyce Index as a metric to assess models performance.
Two case studies are presented to assess the hypothesis and to illustrate the package.
Partial models trained with part of the species distribution often perform better than those fitted on the entire extension. MinBAR is a versatile tool that helps modellers to objectively define the optimal solution.
