We have drawn on new developments in statistical modelling based on machine learning to develop environment-based models describing the distributions of thirty native fish species.
These models were developed using data from the New Zealand Freshwater Fish Database, along with environmental data from the River Environments Classification. The models not only indicate the relative importance of different habitat factors in determining the distributions of each species, but can also be used to predict the probability of capture, species by species, for all of New Zealand’s river and stream segments (more than half a million), including those for which no inventory data are available.