Application of generalized additive models to butterfly transect count data

We investigate the use of generalized additive models for describing patterns in butterfly transect counts during the flight period. Models were applied to sets of simulated data and to transect counts from the British Butterfly Monitoring Scheme (BMS) recorded at a large number of sites in the UK. The models successfully described patterns in counts in a range of species with different life cycles and the approach can be used to estimate an index of butterfly abundance allowing for missing counts. The method could be extended to include other factors such as temperature, sunshine, windspeed and time of day, and to examine potential biases arising from variation in these factors.