Abstract

The Pacific lionfish (Pterois spp.) invaded the Western Atlantic through an anthropogenic introduction off the coast of Florida in the mid 1980s. It is widely regarded as having the potential to become one of the most ecologically and economi-cally harmful marine invasions to date. Lionfish were first observed in Bermuda’s waters in 1999. Despite this, the true dis-tribution of the lionfish population across Bermuda’s reef systems is still unknown. This information is vital for effective management programs. This research attempts to address this lack of knowledge through the generation of a species distri-bution model (SDM) based on known lionfish presence locations and various environmental predictors. Approximately 330 presence-only lionfish observations were generated from a wider fish community census of Bermuda’s mesophotic coral ecosystems (MCEs) and the Bermuda Government’s lionfish catch data. Environmental predictor layers (depth, temperatu-re, benthic habitat and slope) were generated using ESRI ArcGIS 10.3.1. The lionfish SDM was generated using MaxEnt software 3.3.3. MaxEnt is a software package primarily used for species distribution and environmental niche modelling. The R Project dismo package 1.0 - 12 provided an interface between ArcGIS and MaxEnt. Evaluation of the model consis-tently returns an AUC (Area Under Curve) score above 0.75. This model is the first of its kind for Bermuda and not only has the capability to model lionfish but serves as a platform for future species distribution models. It is anticipated these preliminary efforts will become a vital tool for more effective marine spatial management and Pterois spp. eradication pro-grams.

PDF Preview