Habitat evaluation of the Honduran emerald hummingbird Amazilia luciae (Trochilidae) using species distribution models

Abstract

Species distribution models are fundamental for planning and conservation, due to their predictive capacity to fill knowledge deficient. Amazilia luciae is the only endemic species of hummingbird in Honduras and according to the IUCN it is in danger of extinction, so it is important to describe its habitat and potential distribution for its conservation. The objective of this analysis was to conduct a habitat assessment of the Honduran emerald hummingbird A. luciae using species distribution models. Nineteen bioclimatic variables were used in combination with the elevation variable and the ecosystem variable of Honduras. Species occurrence data were obtained from the Global Biodiversity Information Facility (GBIF). Five models were generated, and the selection of the best model was determined by the Area Under the Curve (AUC) and True Skill Statistic (TSS) metrics. The raster, rdismo, SDMtune, and java packages were used using the statistical language RStudio 4.0.2 and ArcGis 10.5. The results indicate that Random Forest is the best model, finding that Lempira, Cortés, Santa Bárbara, Olancho, and Yoro are the predicted areas with very high suitability where the species has already been recorded. Comayagua indicated with very high suitability and Copán and Intibucá with high suitability are areas where A. luciae has not been recorded. The most important variable was precipitation in the warmest quarter.