¿Dónde buscar troglobiontes? Ensayo de una cartografía predictiva con MaxEnt en Gran Canaria (islas Canarias)

Abstract

83 Arxius de Miscel·lània Zoològica, 12(2014): 83–92 Naranjo et al. ISSN: 1698–0476 El © [2013] del artículo pertenece al autor o autores; estos autorizan a la revista Arxius de Miscel·lània Zoològica la publicación del artículo bajo la licencia de Creative Commons Reconocimiento 3.0 que permite un uso no restringido, la distribución y la reproducción en cualquier medio, siempre que se citen los autores y la revista. ¿Dónde buscar troglobiontes? Ensayo de una cartografía predictiva con MaxEnt en Gran Canaria (islas Canarias) M. Naranjo, Á. C. Moreno & S. Martín Naranjo, M., Moreno, Á. C. & Martín, S., 2014. ¿Dónde buscar troglobiontes? Ensayo de una cartografía predictiva con MaxEnt en Gran Canaria (islas Canarias). Arxius de Miscel·lània Zoològica, 12: 83–92. Abstract Where should we search for troglobionts? A study of predictive cartography using MaxEnt in Gran Canaria (Canary Islands, Spain).— A total of, 160 terrestrial troglobiont species are known to date in the Canary Islands. These species are mainly located on the youngest islands that have abundant volcanic tubes (Tenerife, La Palma and El Hierro). On Gran Canaria, an older island with few volcanic caves, the hypogean fauna was considered poor until recent explorations in the mesocavernous shallow substratum and water mines were carried out, with remarkable results. The island covers an area of 1,560 km² and has a maximum height of 1,949 m. As ecological diversity is high, it is fundamental to identify the best areas to conduct effective sampling. Software for species habitat modelling such as MaxEnt, can be useful to select such areas, creating species distribution maps from environmental data. We obtained a potential distribution map using MaxEnt and including all the known troglobiont species of the island. This predictive map is statistically significant and accurately classifies a high percentage of the observed data. Lithology and average rainfall are the two variables that best predict the presence of these species. Basaltic materials —preferably modern— and gravitational landslides are good places for finding subterranean fauna. Locations with high average rainfall appear to be the most appropriate for this purpose.