Burn and Doroff: Decline of Enhydra lutns along the Alaska Peninsula 



271 



1922, and by 1962 the population had grown to an es- 

 timated 1625 sea otters (Lensink, 1962; Kenyon, 1969). 

 Around the same time, the population in the Shumagin 

 Islands was estimated to be 2724 sea otters (Kenyon, 

 1969). 



The first systematic surveys of sea otter abundance 

 along the north side of the Alaska Peninsula were con- 

 ducted in the mid-1970s (Schneider, 1976), followed by 

 surveys in 1982 and 1983 by Cimberg et al. 2 Bruegge- 

 man et al. 1 conducted quarterly surveys of both the 

 northern and southern Peninsula in 1986 to assess 

 sea otter abundance and seasonal distribution. The 

 surveys conducted in 1986 provided seasonal estimates 

 of abundance during a single, ice-free year, and a clear 

 picture of habitat use in the mid-1980s along the Alaska 

 Peninsula (Brueggeman et al. 1 ). 



The sea otter surveys described above were concen- 

 trated along the western end of the Alaska Peninsula 

 where remnant populations existed and appeared to 

 have recovered. By the late 1980s, sea otters had also 

 returned to the nearshore waters of the entire penin- 

 sula as far east as Cape Douglas (DeGange et al. 3 ). 

 Prior to this survey in 1989, little was known about 

 sea otter distribution and abundance on the Alaska 

 Peninsula east of Kupreanof Point. 



The objectives of our study were 1) to assess current 

 sea otter distribution and abundance along the north- 

 ern and southern Alaska Peninsula, 2) to contrast our 

 results with prior surveys conducted in 1986 and 1989, 

 and 3) to relate these data to the observed sea otter 

 population declines observed elsewhere in southwest 

 Alaska. We repeated the aerial survey methods devel- 

 oped by Brueggeman et al. 1 for sea otter habitat along 

 the Alaska Peninsula which consisted of a combination 

 of strip transects in offshore habitat (to the 70-m iso- 

 bath) and coastline surveys (si km of shore) of island 

 groups within the study area. We also repeated the 

 coastline surveys of DeGange et al. 2 to determine the 

 eastward extent of the decline. 



Materials and methods 



Offshore survey areas 



The north Alaska Peninsula (NAP) study area ranged 

 from Cape Mordvinof on Unimak Island in the west to 

 Cape Seniavin in the east. This area was further subdi- 



2 Cimberg, R. L., D. P. Costa, and P. A. Fishman. 1984. Eco- 

 logical characterization of shallow subtidal habitats in the 

 north Aleutian Shelf. OCSEAP Final Rep. no. 4197, 99 p. 

 U.S. Dept. of Commerce, National Oceanographic and Atmo- 

 spheric Administration, Anchorage, Alaska 99501. 



3 DeGange, A. R., D. C. Douglas, D. H. Monson and C. M. 

 Robbins. 1994. Surveys of sea otters in the Gulf of Alaska 

 in response to the Exxon Valdez oil spill. Final report to 

 the Exxon Valdez Oil Spill Trustee Council, Marine Mammal 

 Study 6-7, 11 p. U.S. Fish and Wildlife Service, Anchorage, 

 Alaska 99503. 



vided into two subunits (NAPa and NAPb), and a line at 

 162°W longitude divided the two subunits (Brueggeman 

 et al. 1 ). The south Alaska Peninsula (SAP) study area 

 ranged from the Ikatan Peninsula in the west to the 

 Shumagin Islands in the east. The seaward extent of 

 both the NAP and SAP study areas was the approximate 

 70-m depth contour (Fig. 1A). 



The strip transect method developed by Brueggeman 

 et al. 1 consisted of a series of transects oriented north- 

 south which were spaced every three minutes of longi- 

 tude throughout the study area. In 1986, surveys were 

 flown in a DeHavilland Twin Otter aircraft equipped 

 with bubble windows at an altitude of 92 m and an 

 airspeed of 185 km/h. Two observers, one on each side 

 of the aircraft, relayed sea otter sighting information to 

 a data recorder seated in the aft section of the aircraft. 

 Sea otter sightings were grouped into three distance 

 intervals spaced at right angles to the transect line: 

 0.0-0.23 km, 0.23-0.46 km, and 0.46-0.93 km. These 

 distance zones were determined by using a clinometer 

 to place marks on the inside of the bubble windows. 

 Environmental information on sea state, visibility, and 

 glare was recorded throughout the survey. 



In May 2000 and April 2001, we repeated the survey 

 conducted by Brueggeman et al. 1 using similar meth- 

 ods, with the exception that our survey aircraft was 

 an Aero Commander equipped with bubble windows 

 and we grouped sea otter sightings into five distance 

 intervals: 0.0-0.115 km, 0.115-0.23 km, 0.23-0.345 km, 

 0.345-0.46 km, and 0.46-0.575 km. 



Coastline survey areas 



In 1986, Brueggeman et al. 1 also surveyed the coastlines 

 of 22 islands on the south side of the Alaska Peninsula 

 quarterly at a distance of 0.46 km from shore, using 

 the same aircraft, altitude, and airspeed as in the off- 

 shore area surveys (Fig. IB). In 1989, DeGange et al. 2 

 surveyed the coastlines of these same islands and the 

 Alaska Peninsula from False Pass to Cape Douglas 

 (Fig. 1C). The 1989 survey was conducted from Bell 

 206 and Hughes 500 helicopters at a distance of 0.2 

 km from shore at an altitude of 92 m and an airspeed 

 of 130 km/h. We used similar methods (0.23 km from 

 shore, altitude 92 m, airspeed 185 km/h) to survey the 

 coastlines of these 22 islands and the Alaska Peninsula 

 in April and May 2001. The area of the offshore surveys 

 was adjacent to, but did not overlap, the area of the 

 coastline surveys. Coastline surveys were not conducted 

 in the NAP study area. 



Offshore survey analyses 



Prior to the analysis of the 2000-01 offshore survey data, 

 we tested several assumptions made in the 1986 analysis 

 regarding the detectability of sea otters as a function 

 of 1) survey strip width, 2) survey conditions, and 3) 

 time of day. We examined the distribution of sea otter 

 sightings by distance zone using a chi-square analysis 

 to determine the appropriate survey strip width to use 



