FISHERY BULLETIN: VOL 77. NO .3 



cuts and channels run through the area. Five 

 channels, two man-made, open to the Gulf of 

 Mexico. 



Operating from Aransas County Airport, a 

 high-wing, four-seat airplane was flown along 21 

 transect lines spaced at approximately 4.63 km 

 (2.5 n.mi.) intervels across the study area (Figures 

 1-3 ). With some exceptions the transect lines were 

 oriented due east to west. To provide a reference 

 point with a previous population study (Shane 

 1977) the first two lines were bent to conform to 

 the narrow Corpus Christi and Aransas ship 

 channels (Figure 1). Line 8 was jogged slightly to 

 the north over the Lamar Peninsula so that its 

 western extension would cross Mission Bay (Fig- 

 ure 1). Lines 14 and 15 were altered to overfly the 

 Pass Cavallo and ship channel entrances into 

 Matagorda Bay in the region of Port O'Connor, the 

 location of a proposed T. truncatus study. In 12 

 cases the transects were interrupted by land that 

 divided them into two or more parts, so that in all, 

 42 overwater crossings were flown. Eight of these 

 crossings were 2 km or less in length while the 

 longest was 42 km. Their average length was 10.2 

 km. Time of these crossings ranged from < 1 to 

 about 18 min. 



Most transects were flown at 167 km/h and an 

 altitude of approximately 152 m (500 ft). The first 

 part of transect 1 was flown at 213 m (700 ft) to 

 safely maneuver around large cranes and other 

 structures. When not fully occupied with flying 

 the plane, the pilot searched for bottlenose dol- 

 phins. An observer sat in the right front seat next 

 to the pilot. This observer also functioned as the 

 "navigator," talking the pilot onto transect land- 

 marks, calling out the start and stop times for each 

 transect, and charting the dolphin sightings. Two 

 observers sat in the rear of the plane. The observer 

 in the right seat mainly functioned as a recorder 

 who kept a transect log noting the time of starting 

 and ending of each transect and comments on visi- 

 bility, weather, and other observations of interest. 

 A sighting form was also kept in which was noted: 

 the observer making the sighting, the nature of 

 the observation which first alerted us to the pres- 

 ence of a dolphin herd, the sighting cue; the esti- 

 mated numbers of adult animals and calves and 

 their assumed behavior; and the estimated right 

 angle, or perpendicular, distance of the sighted 

 dolphin herds from the plane's track. While a strip 

 transect design had been planned, the perpendicu- 

 lar distance estimations were essential for alter- 

 native dolphin density calculations utilizing line 



transect theory (Saber 1973). If time allowed, the 

 herd configuration relative to the environment 

 was also sketched. 



Because of the low flying speed, the airplane was 

 relatively quiet and voice communication between 

 party members was feasible. The shortness of the 

 transects and rest intervals between transect lines 

 alleviated observer fatigue. 



Observers searched outward to about 400 m (we 

 estimated distances in yards). This distance was 

 estimated with the aid of tape markings on the 

 wing struts that had been calibrated against 

 range marks on the landing strip. When a dolphin 

 sighting was made, the pilot deviated from the 

 transect line and usually orbited the herd twice 

 while all observers counted the animals and noted 

 the presence of calves. A consensus opinion was 

 scored for these counts. Rarely only one circle was 

 necessary, and on occasion three or more circuits 

 were flown before the observers felt confident with 

 the count. On occasion, individual animals or 

 small herds could not be relocated and limited 

 data based on the original sighting were logged. 



Two observers worked all the flights, whereas 

 one person was relieved as recorder-observer for 

 the last three flights. The same pilot flew the plane 

 on surveys 1-4. A different pilot took over on the 

 last survey. 



RESULTS 



Operations 



The survey design called for six replicate tran- 

 sect runs on successive days. The period of the 

 operation (26 March-1 April 1978), however, was 

 plagued by strong winds ( 33-46 km/h) that caused 

 a 1-day postponement of survey 4, cancellation of 

 survey 6, and affected the results of survey 5 to the 

 extent that those data are of limited value (the 

 specific effects of weather on the survey will be 

 discussed later). Weather conditions were good to 

 excellent on two runs, surveys 2 and 4, and mar- 

 ginal to fair on surveys 1 and 3. A malfunctioning 

 airplane engine caused curtailment of the last 

 three transects on survey 2. These were made up 

 at the end of survey 4 under similar environmen- 

 tal conditions. A total of 436 track kilometers ( 235 

 n.mi.) was flown on each survey. Assuming a 

 402.25 m scan on each side of the aircraft, an area 

 of 351 km^ (102 n.mi.^) was searched. With the 

 4.63 km transect line spacing, this would repre- 



588 



