Factors affecting perpendicular sighting distances on shipboard line-transect surveys for cetaceans

Factors that affect cetacean perpendicular sighting distances are investigated using a Generalised Additive Modelling (GAM) framework to analyse 8,203 sightings of 34 species seen on 200,000km shipboard line-transect surveys in the eastern Pacific 1986-96. Perpendicular distance is modelled as non-linear function following predictor variables: species; an priori grouping; logarithm group size; Beaufort sea state; presence rain or fog; cue; sun glare; geographic stratum; observer; ship; year; cruise; and, 1991-96, visibility and swell height. Based Akaike Information Criteria (AIC), best model for 1986-96 included all variables except rain/fog code. For 1991-96 data, height anomaly was also important replaced ship year best-fit model. independent subsets GAM coefficients were highly correlated, indicating many same factors acting different areas at times. Species methods (25x binoculars vs unaided eye) had largest effects distances. The groups captured much, but not among-species differences. Two other species-related (group size cue) models. related search conditions (Beaufort state anomaly) searchers themselves (individual observer) important. We anticipate this information relative magnitudes affecting can be used improve both design analysis data.