196 
Fishery Bulletin 1 14(2) 
Probability of seals entering the water 
Odds ratio 
Figure 5 
Odds ratios of a harbor seal {Phoca vitulina ) entering the water during random- 
ized, focal seal observations (2001-2006). Bars indicate 95% confidence intervals 
(CIs). Solid black symbols indicate factors that were statistically significant (i.e., 
where the Cl does not cross 1). The odds ratios are relative to what we used as the 
comparison category for each variable (in parentheses). Position=seal’s position in 
relation to the overall patch of ice (e.g., interior versus near an edge next to a sec- 
tion with open water or less ice). Vessel type and vessel distance results are based 
on 2001 data only. Vessel distance=distance from the focal seal or group. 
conditions than nonpups, but they did show a midday 
peak in counts (Hoover-Miller et ah, 2011). Because 
pups suckle when they are hauled out, the haul-out 
cycles of lactating females may be influenced by the 
needs of their pups. We found that, when >1 mother- 
pup pair was present, seals were less likely to enter 
the water during a random 10-min observation than 
were seals in groups without a pup present (Fig. 5), 
which also suggests longer haul-out bouts by mother- 
pup pairs. In Johns Hopkins Inlet, GBNP, in June, 
the TOD also was not associated with counts of either 
nonpups or pups or with the proportion of hauled-out 
seals that were pups (Mathews and Pendleton, 2006). 
Our analyses of our randomized observations did not 
reveal any relationship between seals entering the 
water and local ice dispersion (i.e., scattered or dense) 
(Fig. 5). In contrast, Young et al. (2014) observed a 
pattern in which seals were more likely to enter the 
water when an ice cover index (a combination of cover 
and density) was high. 
