16 



Fishery Bulletin 93(1). 1995 



encompasses California waters out to a distance of 

 185-278 km (100-150 nmi) from the coast and 

 roughly a depth of 3,000-4,000 m (Fig. 1). It was 

 defined on the basis of the distribution of fisheries 

 that are known to take marine mammals and does 

 not reflect a distributional boundary for any marine 

 mammal population. Surveys were conducted along 

 transect lines forming two nearly uniform, overlap- 

 ping grids (Fig. 1). The resulting overall grid lines 

 were spaced 41^46 km (22-25 nmi) apart. The loca- 

 tion of the transect grid was chosen without refer- 

 ence to specific areas or topographical features. To 

 avoid potential differences in regional coverage, an 

 attempt was made in each year to complete all 

 transects of the first grid, providing coarse coverage 

 of the entire study area, before beginning the second 

 grid. However, in both years, poor weather conditions 

 prevented the completion of both survey grids. In 

 1991, 85% (5,326 km) of transect grid 1 and 27% 

 (1,739 km) of grid 2 were completed, and in 1992, 



81% (5,065 km) of transect grid 1 and 14% (890 km) 

 of grid 2 were completed. The relative proportions of 

 survey effort in different sea state and cloud cover 

 conditions were similar for the two years (Table 1). 

 The survey platform was a twin-engine turbo-prop 

 DeHavilland Twin Otter, flown approximately at an 

 altitude of 213 m (700 ft) and an airspeed of 165-185 

 km/h (90-100 knots). All cetacean and sea turtle 

 sightings were recorded, but because of the high den- 

 sities of pinnipeds near rookeries, these species were 

 recorded only when seen farther than 10 km from 

 land. Two "primary" observers searched through 

 bubble windows on the left and right sides of the air- 

 craft. These windows allowed observers to view to 

 the side and directly beneath the aircraft with at least 

 10° of overlap between sides. To achieve higher sight- 

 ing efficiency near the transect line, observers 

 searched for cetaceans only out to a declination angle 

 of 12° (1,004 m perpendicular distance). An addi- 

 tional "secondary" observer monitored the trackline 

 area out to 55° declination angles (on 

 both sides) through a round 45-cm ( 18- 

 in) viewing hole in the belly of the air- 

 craft and reported sightings missed by 

 the primary team. A fourth person re- 

 corded all sighting, effort, and environ- 

 mental data. To minimize observer fa- 

 tigue, all observers rotated between 

 these four active positions and one 

 resting position roughly every 30 min- 

 utes. All observers had previous experi- 

 ence in identifying cetacean species from 

 aerial or shipboard platforms, or both. 

 All survey data were recorded on a 

 laptop computer connected to a LORAN 

 or GPS (Global Positioning System) 

 navigational receiver, providing a con- 

 tinuous record of position (updated 

 every few seconds), altitude, air speed, 

 and survey conditions. Environmental 

 conditions, such as Beaufort sea state, 

 percent cloud cover, and glare, were 

 updated whenever changes occurred. 

 Conversation in the aircraft was re- 

 corded on a central cassette recorder 

 as a backup to the computer record. 

 Observers also recorded individual 

 sighting information into personal 

 notebooks. Surveys were conducted only 

 in Beaufort sea states 0-4. 



Following the methods described in 

 Forney and Barlow ( 1993) and Carretta 

 and Forney (1993), the aircraft circled 

 for each sighting to obtain species iden- 

 tifications and school size estimates 



