ABSTRACT
There has been a great increase in the development and use of species distribution models also referred to as bioclimate envelope models habitat suitability models or ecological niche models. These models generally aim at explaining and predicting variation in the occurrences or abundances of species as a function of environmental and spatial variables. Common purposes of applying species distribution models include those related to inference, prediction, and projection. Questions related to inference (or explanation) ask which environmental variables (including e.g. climatic, geographic, anthropogenic and spatial variables) control the distributions of species. Within the generalized linear modelling framework, a response variable (y) is assumed to be generated by a particular distribution, such as normal, Bernoullli or Poisson. Accounting for co-occurrences in the statistical model makes it possible to utilize information about other species when constructing model predictions. This chapter exemplifies both species-level and community-level approaches to species distribution modeling.
