As interest in biodiversity within working landscapes has increased, and greater focus has been given to conservation outside of protected areas, there is a corresponding need to evaluate the capacity of different land uses to meet conservation needs. Coffee agriculture has garnered interest as a land use with high conservation potential. Studies investigating avian and arthropod diversity have found that coffee farms harbor similar numbers of species as the surrounding natural landscapes, however the degree of alteration and intensity of management greatly affect this potential. Herptiles have received less attention in this area of investigation, but the few previous studies on herpetofaunal diversity in agroecosystems have found similar results. Coffee is a dominant agricultural product in Puerto Rico, with a long history of production in the central mountains of the island. This region is also home to eleven of the sixteen species of native Eleutherodactylus (coquí) frogs. To compare coquí diversity and abundance between forests and farms, bioacoustic recorders were placed in 7 secondary forest sites and 19 coffee farms. Farm vegetation surveys were conducted to allow for comparison of management intensity between farms. Landcover was digitized in a half kilometer buffer around each recorder to examine landscape level effects.
Results/Conclusions
Four species of coquí were detected in the bioacoustic data. Results showed no significant difference in mean number of species, abundance, or Shannon-Weaver diversity index between forest and farm sites. This suggests that coffee farms and forests in Puerto Rico are effectively similar in habitat provisioning. There was significant difference within farm sites and within forest sites though. Species diversity ranged between two to three species on farms, and one to three species in forests. Importantly, while the number of species between site types were statistically similar, the Endangered Species E. wightmanae, a forest specialist, was not detected in any of the 19 farms. Of the three forest sites in which E. wightmanae was detected, two were outside of its mapped range, suggesting that it may occur in a wider area across the island than previously recorded. Landcover effects on abundance and diversity were found to be significant, with observed effects following the habitat preferences of the species. Increases in open habitat types corresponded to increasing grass/shrubland species abundance, while increases in disturbance accompanied decreases in E. wightmanae abundance. No farm management variables were found to have an effect except a weak effect of total coffee planted on E. coqui.