Use of topographic predictors for macrobenthic community mapping in the Marine Reserve of La Palma (Canary Islands, Spain)

Evaluations of the coastal and marine conservation require detailed maps of types of communities that occur in the zone. This paper describes how distribution models are used to develop benthic distribution maps with biological data collected from surveys, and environmental variables derived from a Digital Elevation Model (DEM), including different indices of terrain complexity. We compared the success of two algorithms, Maxent and ENFA, commonly used in the marine environment to identify best suited methods to modelling the distributions of six benthic communities identified in the marine protected area of La Palma (Canary Islands, Spain). The environmental variables depth, slope, type of substrate, Bathymetric Position Index (BPI) and Vector Ruggedness Measure (VRM) were the variables with higher influence on the distribution of communities. The distribution models of both techniques were coincident and congruent, although Maxent produced more constrained predictions than ENFA, highlighting the significantly better performance of the Maxent models for communities with fewer presences, in this study, black coral and brown garden eels. The resulting distribution maps were evaluated and reclassified and they were represented in a unique map that summarises all of the individual maps. Given that the distribution models were made on the same study area and based on presences data collected at the same time, it was possible to make a preliminary analyse of the interactions between the studied communities. In conclusion, distribution models of benthic communities are suitable tools to design reliable and full coverage distribution maps of benthic communities and they provide new information about the behaviour of communities on the range of environmental conditions studied and useful information for management of marine and coastal areas.