Dahlheim et al.: Temporal changes in abundance of Phocoena phocoena inhabiting the inland waters of Southeast Alaska 
253 
is possible but unlikely, given the multiple years in- 
volved and the overall extent of the area coverage in 
our study, that a large grouping of animals was missed 
during a survey period. One explanation is that part 
of the population may have moved outside our study 
area (e.g., offshore waters) because of shifts in prey 
availability and abundance. Without knowing whether 
porpoise shift their distribution or what might drive it 
if they do (e.g., prey preferences or oceanographic con- 
ditions), we cannot determine whether porpoise move- 
ment patterns are a factor in the observed downward 
trend. In addition, a change in porpoise numbers may 
also be dependent upon year-to-year variations in habi- 
tat suitability, increased predation, increased mortality 
from bycatch, or a combination of all these factors. 
When examined on a regional scale, abundance 
was relatively consistent throughout the survey pe- 
riod in the northern region that included Glacier Bay. 
In contrast, a significant downward trend in abun- 
dance was estimated for the southern region that in- 
cluded the waters surrounding Wrangell and Zarembo 
Island between the early 1990s and the mid-2000s, 
and an increase was observed for that region as of 
2010. Of particular interest is that porpoise numbers 
declined only in regions where salmon (Oncorhyn- 
chus spp.) and Pacific herring net fisheries operate. 
Certainly, bycatch has been shown to be a significant 
source of porpoise mortality in other geographic areas 
(Gaskin, 1984; Read and Gaskin, 1988; Woodley and 
Read, 1991; Read et al., 1993) and may, in fact, be 
responsible for the downward trend observed in our 
data for the mid-2000s. We are unable, however, to 
attribute the decline in the mid-2000s to incidental 
takes in the net fisheries given that interaction data 
are not available. Regardless of the reasons for that 
decline, further studies are necessary to understand 
the possible causes of the variability observed in our 
study in the abundance of harbor porpoise in inland 
waters off Southeast Alaska. 
Insights into population structure 
As currently defined (see Introduction), the Southeast 
Alaska stock of harbor porpoise occurs from Dixon En- 
trance to Cape Suckling, including all inland and coast- 
al waters within this region (total area= 106,087 km 2 ). 
Studies that have addressed stock structure of harbor 
porpoise in Alaska are either not available or based on 
limited sampling from a particular area (Rosel et al., 
1995; Chivers et al., 2002). Outside Alaska, research 
has shown that harbor porpoise stock structure is of a 
finer scale than that of the stock structure reflected in 
the Alaska stock assessment report for 2011 (Allen and 
Angliss, 2012). Stock discreteness in the eastern North 
Pacific was, for example, analyzed by using mitochon- 
drial DNA from samples collected in California, Wash- 
ington, and British Columbia but from only one sample 
from Alaska (Rosel, 1992). Results of our initial investi- 
gation indicated little interbreeding of harbor porpoise 
along the western coast of North America. 
Further genetic testing by Rosel et al. (1995, 1999) 
showed that harbor porpoise in both the eastern North 
Pacific and North Atlantic were not panmictic and that 
movement was sufficiently restricted resulting in ge- 
netic differences. Furthermore, Chivers et al. (2002), 
using both mitochondrial and nuclear (microsatellite) 
DNA, examined the intraspecific structure of harbor 
porpoise that inhabited the eastern North Pacific and 
reported similar findings. These studies revealed sta- 
tistically significant genetic differentiation, indicating 
demographic independence of fairly small subunits 
within the population. 
Additional evidence that harbor porpoise restrict 
their movements has been obtained from both contami- 
nant research and satellite tagging studies. Investiga- 
tions on the pollutant loads of harbor porpoise from 
California to the Canadian border (Calambokidis and 
Barlow, 1991) also suggest restricted movement pat- 
terns of harbor porpoise. Pollutant studies produced 
similar results in the North Atlantic (Westgate and 
Tolley, 1999). Satellite-tagging operations conducted in 
Puget Sound, Washington, have shown that porpoise 
movements were fairly localized and did not occur be- 
tween Neah Bay (at the entrance of the Strait of Juan 
de Fuca) and the San Juan Islands; these 2 regions are 
separated by approximately 150 km of continuous open 
waterways (Hanson 8 ). 
During our study, harbor porpoise distribution was 
clumped in 2 major areas that were consistent through- 
out the 22-year period: the waters of Glacier Bay and 
Icy Strait and the waters surrounding Wrangell and 
Zarembo Island, including the adjacent waters of east- 
ern Sumner Strait (northern and southern regions, 
respectively; also see Dahlheim et al., 2009). These 
regions are separated by a distance of approximately 
400 km. If harbor porpoise within the inland waters of 
Southeast Alaska behave in the same manner as har- 
bor porpoise elsewhere have been reported to behave 
(i.e., movements are locally restricted), then a low level 
of mixing would be expected, indicating the potential 
for reproductive isolation between the 2 regions. The 
physical character of this region (e.g., hundreds of is- 
lands and a complexity of waterways) may also limit 
frequent movements of harbor porpoise between the 2 
regions. 
The difference in abundance trends between the 
northern and southern regions of the study area pro- 
vides additional evidence that these 2 regions, where 
consistent porpoise concentrations have occurred over 
2 decades, represent regions of population structuring 
for this species within the inland waters of Southeast 
Alaska. Of utmost concern is that incidental takes 
within a small region (e.g., Wrangell and Zarembo Is- 
land) could severely affect undefined localized stocks of 
harbor porpoise that could easily go undetected unless 
stock structure is identified. On a larger scale, given 
the wide distribution of harbor porpoise throughout 
8 Hanson, B. 2013. Personal commun. Northwest Fish. Sci. 
Cent., 2725 Montlake Blvd. East, Seattle, WA 98115-2097. 
