Basic Introduction to Species Distribution Modelling

Species distribution models (SDMs) have become the most widely used method for wildlife management and have been applied in the fields of ecology, biogeography, and conservation. Species distribution modelling commonly requires two categories of data: (1) species data and (2) environmental data. Species data can be nominal (presence/absence records), ordinal (ranked abundances), or ratio (abundance and richness). Environmental data refers to both biotic and abiotic conditions. The most common types of environmental data in species distribution modelling are climatic and topographical variables since these two sets of variables represent, respectively, the large-scale conditions relevant to species' physiology and small-scale conditions which affect solar energy input and availability of moisture. There are various techniques for species distribution modelling. The choice of modelling technique is affected by the availability of data and in turn affects modelling outcomes. The accuracy of SDMs can be measured with respect to two characteristics: discrimination capacity and reliability; generally, discrimination capacity has been seen as a more crucial metric of model performance. Accuracy is an important challenge faced by SDMs. Several factors affect the accuracy of SDMs such as environmental data, species data, the ecology of the species, available computational resources, the model being utilized, and spatial resolution.