Predicting Patterns in Spatial Ecology Using Neural Networks: Modelling Colonisation of New Zealand Fur Seals

Conventional mathematical models for ecological processes are often complex and restricted in their predictive capability through the non-linear and non-gaussian properties of the input data. In this paper we discuss the capability of an artificial neural network (ANN) model to predict the colonisation potential of New Zealand fur seals (Arctocephalus forsteri) around South Island, New Zealand. We used the distribution of food sources, sea configuration and coastline terrain to predict the potential condition of pups for coastline segments around South Island. We suggest that ANNs can be used effectively in combination with geographic information systems for ecological modelling.