Line transecting with difficulties; lessons from surveying minke whales

Difficulties encountered in line transecting, like missed observations on t he track line, discrete cues, heterogeneity in cue strength or sighting eff iciency, and measurement errors, might cause estimates of abundance to be b iased. Methods for removing bias in abundance estimates based on difficult line transect survey data are considered, with an emphasis on hazard probab ility modelling and likelihood analysis with simulation as an integral part of the estimation. The shipborn survey of minke whales (Balaenoptera acutu rostrata) in the northeastern Atlantic in 1995 was intrinsically difficult. Experiences from this survey and its analysis are reviewed. If the survey had been designed and the data analyzed by traditional line transect method ology, results would be biased negatively by some 85%.