GIS and RS for Modeling Landform Controls on the Distribution of Forest Canopy Vascular Epiphyte

In the tropical mountainous rainforest, the forest canopy structure is highly correlated with local topography. However, the relationships of forest canopy structure with canopy vascular epiphyte (one of its unique biodiversity richness) is poorly described and yet to be established. This study aims at modelling the distribution of canopy vascular epiphyte diversity in the Nyungwe National Park of Rwanda. Canopy vascular epiphyte distributions were mapped by applying a logistic regression equation relating the presence/absence data to the parameters of surface shape. The results indicated that there is a significant impact of terrain elevation and the geomorphic shape of the land surface (terrain shape index) on vascular epiphyte occurrence. Based on the most accurate logistic model assessed by the residual deviance information, the Akaike Information Criterion and the area under the Receiver Operating Characteristic Curve, the probability of vascular epiphytes to occur was predicted at 75 % level of accuracy. This confirms the role of elevation and landform to determine communities and site suitability in mountainous areas.