The use of generalized additive models (GAMs) for the modelling of wildlife population abundance is becoming increasingly common. The population trajectory is modelled as a smooth function of a number of explanatory variables, or covariates. GAMs offer greater flexibility than traditional log-linear Poisson regression methods, and are likely to give more realistic impressions of true trend. A 1996 contract to RUWPA funded by MAFF demonstrated that the GAM approach can be successfully applied to data sets such as the Common Birds Census coordinated by the British Trust for Ornithology (BTO).

Publication

Here are the SPLUS functions used in the paper for applying GAMs to wildlife surveys.
Code is also provided here for the free statistical package R.

Contract report

• Back to Rachel Fewster's home page