Local ecological niche modelling to provide suitability maps for 27 forest tree species in edge conditions

Abstract

Ecological Niche Modelling (ENM) portrays the relationship between the actual geographical distribution of a species and the environmental factors that induced this distribution. Yet most models study species over the wider range of their distribution; thus, they are rarely appropriate for forest management and forest restoration on the local scale. This study aims to understand the major environmental factors affecting the distribution of 27 species, through limiting ENM at national level (Lebanon). MaxENT software was used for modelling. Area under the curve (AUC) values showed a very good robustness of the models. Minimal biogeographic and climatic parameters such as elevation, distance from the sea, annual mean precipitation, the average minimum temperature of the coldest month, the average maximum temperature of the warmest month, and Emberger Quotient were sufficient to obtain robust modelling results. Cloud coverage during summer was identified as a novelty factor explaining species distribution at the edge of their range. Composite soil and topography predictors such as Potential Direct Incident Radiation (PDIR) and the Integrated Moisture Index (IMI) were reduced to simple factors such as aspect, slope and available water content, whose contribution was conditioned to higher data resolution. The high number of presence points enabled us to study the range of species distribution gathering them according to their ecological characteristics. The generated reforestation suitability maps and the likelihood of occurrence of each species were achieved to define priority species for conservation and forest management. This information could be useful for decision-makers and foresters.