FISHERY BULLETIN: VOL. 83, NO. 2 



the David Starr Jordan; a summary of the obser- 

 vations is listed in Table 1. Seven of the tracks were 

 terminated prematurely, and of the remaining 19, 

 6 schools passed undetected by shipboard observ- 

 ers. These 6 schools did not appear to adjust their 

 direction of movement in reaction to the survey 

 vessel. 



Of the 13 schools sighted by ship, 1 school altered 

 its direction of movement in reaction to the ap- 

 proaching ship, prior to the detection of a sighting 

 cue by the shipboard observers, and 12 schools did 

 not appear to react before detection by the ship. 

 One of the 12 schools was composed of rough 

 toothed dolphins, Steno bredanensis, which are 

 not a target of abundance surveys. Thus, from the 

 results of this experiment, it is expected that 8% 

 ( V12) of the target schools encountered on a survey 

 will have moved (in reaction to the observer) prior 

 to detection. This does not imply a corresponding 



degree of survey bias. Nonrandom movement, 

 prior to detection, will alter the distribution of 

 sighting distances and the detection function fit 

 to the distribution; the survey will be biased to 

 the extent that the functional form is sensitive to 

 the data (see Discussion). Survey bias may also 

 exist as a result of schools that react to the ship 

 and are subsequently never seen by shipboard 

 observers; if these schools would have been ob- 

 served (the expectation is certain if they are on the 

 transect line, less certain if they are off the line), 

 then the bias is proportional to the fraction of 

 schools that escaped detection. As stated above, no 

 schools were observed to react to the ship and 

 avoid detection. 



The data suggest that dolphin schools may alter 

 their direction of movement in reaction to the ap- 

 proach of a survey vessel. Thirty-eight percent 

 (V13) of the schools which were tracked by helicop- 



TabLE L — Summary of dolphin school tracking data. 



'School did not appear to react to the approach of the survey vessel. 

 'School passed undetected by shipboard observers 

 ^Track prematurely tenninated. 

 "Cue observed on the horizon. 



190 



