Dahlheim et al.: Temporal changes in abundance of Phocoena phocoena inhabiting the inland waters of Southeast Alaska 
243 
perceived areas of low porpoise density, but to date 
there has been no analysis of genetic or individual 
movement to assess the validity of these designations. 
The preference of harbor porpoise for shallower wa- 
ters makes them highly vulnerable to incidental cap- 
ture during net-fishing operations (Jefferson and Curry, 
1994; Read, 1994; Barlow et a!., 1995). The nature and 
magnitude of incidental takes are currently unknown 
but could be significant in some gill-net and purse- 
seine fisheries targeting Alaska salmon ( Oncoi'hynchus 
spp.) and Pacific herring ( Clupea pallasi). 
Obtaining abundance estimates for harbor porpoise, 
a small, inconspicuous cetacean species, is challenging. 
For example, the ability to detect harbor porpoise is 
highly sensitive to environmental conditions; surveys 
should be limited to relatively calm sea states and 
good lighting conditions. Despite such challenges, es- 
timates of both density and abundance for this species 
do exist for Alaska waters. Taylor and Dawson (1984) 
reported on a shore-based study that yielded density 
estimates for Glacier Bay National Park and Preserve. 
In 1991-1993, and again in 1997-1999, aerial surveys 
of coastal waters in Alaska, ranging from the south- 
eastern Bering Sea to Dixon Entrance, yielded more 
recent abundance estimates (Dahlheim et ah, 2000; 
Hobbs and Waite, 2010). 
In this study, we report the results from dedicated 
line-transect surveys conducted to determine the den- 
sity and abundance of harbor porpoise in Southeast 
Alaska over a 22-year period from 1991 through 2012. 
The objectives of these surveys were 1) to obtain rela- 
tive abundance estimates of harbor porpoise within 
the inland waterways of Southeast Alaska, 2) to in- 
vestigate porpoise density and abundance by different 
strata (i.e., smaller regions), 3) to establish a baseline 
for detecting changes in harbor porpoise abundance 
through time, and 4) to report on significant insights 
on this species as a result of these investigations. 
Materials and methods 
Study area 
The study area included the inland waters of Southeast 
Alaska (Fig. 1). Surveys covered all major channels or 
bays from Juneau to Ketchikan: Lynn Canal, Icy Strait, 
Glacier Bay, Cross Sound, Chatham Strait, Stephens 
Passage, Frederick Sound, Sumner Strait, and Clarence 
Strait. When time permitted or weather precluded the 
surveying of major channels, many adjacent smaller 
bodies of water (bays, inlets, and passages) were sur- 
veyed. The coverage of these smaller areas varied con- 
siderably among surveys and across the years. 
Field methods (1991-1993) 
During 1991-1993, surveys were carried out aboard 
the NOAA Ship John N. Cobb , a 28.4 m long research 
vessel with a combined bridge and average observer 
height of 5.9 m. A line-transect method was employed 
to survey predetermined tracklines. At the start of this 
study, distribution, habitat preferences, and seasonal 
occurrence of harbor porpoise within the study area 
were unknown. Tracklines were designed throughout 
the study area with either a zig-zag or straight-line 
path, depending upon the size of the different areas. 
The survey was designed to include all major water- 
ways and a selection of smaller bays and inlets to ex- 
amine both deepwater and nearshore habitats through- 
out the entire study area. The same trackline design 
was employed for all surveys completed between 1991 
and 1993, although alterations were made during some 
surveys depending on weather and other unforeseen 
circumstances (e.g., mechanical breakdowns or engage- 
ment in rescue operations). 
During line-transect surveys, sighting data were col- 
lected by a team of 3 observers, with 1 observer at each 
of 3 stations: starboard, port, and recorder station. In 
the early 1990s, the total number of biologists partici- 
pating in the survey was 6; therefore, a full observer 
rotation took 2 h, with each observer spending 40 min 
at each station or watch, followed by a 2-h rest period 
for each observer after each full rotation of watches. 
Schedules for observer rotations were selected random- 
ly each day. 
Port and starboard observers used 7x50 Fujinon 5 
binoculars (model 56A2, Fujifilm Holdings Corp., To- 
kyo) to search from 0° (at the ship’s bow) to 90°. Scan- 
ning techniques were standardized with nearly 32 min 
(or 80%) of the 40-min watch spent scanning with the 
binoculars and about 8 min spent scanning with the 
naked eye. To reduce observer fatigue, binoculars were 
supported by adjustable metal poles that were either 
handheld or rested on the observer’s hips. When not 
entering data, the recorder searched for porpoise by 
scanning both sides of the ship from the bridge with 
the naked eye. Binoculars were only used by the re- 
corder to confirm sighting identifications and numbers. 
Sightings made by the officers, crew, and off-watch ob- 
servers were recorded as “off effort” and were not used 
in calculations of density estimates. 
A GPS unit was connected directly to a portable 
computer on the bridge. The date, time, and position 
of the ship were automatically entered into a data file 
every 10 min and whenever data were entered by the 
recorder. Search effort was recorded on the computer 
by marking the beginning and end of each transect. 
Beaufort sea state, a weather description (rain and 
fog), a visibility index, and observer positions (port, re- 
corder, and starboard) were also entered. A new entry 
was made whenever a change in course, weather, or 
personnel occurred. 
When a sighting was made, the recorder entered 
the following data: all sightings made by the port 
and starboard observers, the sighting angle, number 
5 Mention of trade names or commercial companies is for iden- 
tification purposes only and does not imply endorsement by 
the National Marine Fisheries Service, NOAA. 
