Abstract

Introduction: Invasive species are considered the second cause of extinction of native species after habitat loss. The impacts of invasive species have serious economic implications since the presence of this type of species can result in a decrease in ecosystem services granted to humans. In marine systems, some human activities such as maritime transport and aquaculture have favored the dispersion of invasive species, especially those with commercial importance. This paper describes the potential distribution of the tiger shrimp, Penaeus monodon, an invasive species along the American Atlantic coast. Objective: To describe a potential distribution model of Penaeus monodon in the American Atlantic region and compare the environmental characteristics of this region with those of the Indo-Pacific original niche conditions. Methods: Using geographic and environmental data, we constructed and tested three models to determine the efficiency of MaxEnt v. 3.3 software in predicting new areas for the distribution of this invasive shrimp species. Geographic data were downloaded from such web sites as the Global Biodiversity Information Facility, the Ocean Biogeographic Information System, and the United States Geological Survey, as well from literature. Environmental data were downloaded from Bio-Oracle v2.0 data base. The three tested models were: 1) the first was created using only recordings of Penaeus monodon from the Indo-Pacific (its origin zone) and then projected to the Atlantic (native model); 2) the second was built using only recordings from the invaded area; the training and projection area of this model was the Atlantic (invasive model); 3) the third included recordings from both the Indo-Pacific and Atlantic regions, and the model was trained and projected jointly in both areas (complete model). We extracted the values of the three models for each tiger shrimp sightings in the invaded area; sightings with values ³ 0.5 were considered as valid prediction of occurrence of the species. Results: We found that the following variables explained 80 % of species distribution: phosphates from the ocean surface, coastal type, chlorophyll a, and maximum bottom temperature. In terms of the models’ ability to predict the occurrences reported in the Atlantic, results were as follows: Native model had a prediction index of 40 %; Invasive model was able to predict 81 % of recordings; and complete model predicted 92 % of total occurrences reported in the invaded area. Conclusions: Our findings suggest that based on the complete model, the countries where the tiger shrimp could establish itself are Mexico and Cuba. Continuous monitoring and conservation actions are relevant in the countries where this species is currently established, as well of those countries with potential for invasions.

Keywords: biological invasions, invasive prediction, realized niche, environmental factors model, spatial modeling