274 
Fishery Bulletin 106(3) 
Table t 
Total acceptable survey effort (i.e., where visibility was 
at least 3.7 km, Beaufort sea state was 3 or lower, air- 
craft altitude was below 325 m, and observers were on 
watch) for right whales (Eubalaena glacialis) in the Cape 
Cod Bay, Massachusetts, study area, summarized by two- 
week periods across the entire study, 1998-2002. 
Period 
Effort (km) 
1-14 December 
900 
15-28 December 
550 
29 December-11 January 
3203 
12-25 January 
5476 
26 January-8 February 
8242 
9-22 February 
7953 
23 February-8 March 
7535 
9-22 March 
4983 
23 March-5 April 
7965 
6-19 April 
9006 
20 April-3 May 
7863 
4-17 May 
2790 
pooled across periods before classification; therefore the 
maps showing different time periods were comparable. 
As a test of the definition for critical habitat boundary 
(i.e., a definition that did not include a potential effort 
bias), SPUE values were compared between the 3-min- 
ute cells inside the critical habitat, outside and west of 
it (western Cape Cod Bay), and outside and east of it 
(outside of Cape Cod). 
All computations and statistical analyses were per- 
formed with Statistical Analysis System procedures 
(SAS for Windows, vers. 6.13, SAS Inst. Inc., Cary, NC). 
Calculations of effort and SPUE were performed with 
custom SAS programs developed at the University of 
Rhode Island. Whenever possible, we used statistical 
tests of hypotheses with nonparametric methods (in 
SAS PROC NPAR1WAY) to avoid the assumptions of 
normal distributions and homogeneous variances as- 
sociated with parametric tests. The test most often 
used was the Kruskal-Wallis nonparametric ANOVA. 
When the comparison involved only two distributions, 
the Wilcoxon rank sum test was used (equivalent to 
the Mann-Whitney U test or a parametric T-test). The 
only parametric test that we used was a Duncan’s mul- 
tiple range test (in SAS PROC GLM), which allowed us 
to perform multiple pairwise T-tests while correcting 
the critical significance probabilities for the number of 
comparisons. There is no nonparametric test equivalent 
available in SAS. 
Results 
During the 1998-2002 seasons, 167 aerial surveys 
were conducted, yielding 66,466 km of acceptable 
Table 2 
Mean (±standard error [SE] ) and maximum number of 
sightings of North Atlantic right whales (Eubalaena gla- 
cialis) per unit of effort (no. of whales/1000 km of survey 
effort) in Cape Cod Bay, Massachusetts, summarized by 
year. 
Year 
Mean 
SE 
Maximum 
n 
1998 
14.58 
2.45 
789.4 
819 
1999 
11.48 
1.49 
417.3 
874 
2000 
18.44 
2.49 
625.7 
728 
2001 
12.74 
1.70 
575.7 
802 
2002 
2.11 
0.61 
210.7 
739 
All 
11.91 
0.84 
789.4 
3962 
survey effort. Of the 103 3-minute cells available to be 
sampled in the study area, 101 (98.1%) were sampled 
at least once, and 88 (85.4%) were sampled in every 
year. Because the basic sampling unit represented 
aggregated data within a cell across a 2-week period, 
the maximum possible sample size was 6180 (103 
cells x 5 years x 12 periods), and the actual number of 
samples (effort >2 km) was 3962 (64.1% of the maxi- 
mum possible). Overall mean effort was 658 km of 
survey track per 3-minute cell (range: 2.5 to 1189.5 
km). Spatial distribution of effort was fairly uniform 
throughout the study area; lowest effort occurred in 
the peripheral cells, some of which were partially over 
land. Temporal distribution of effort was relatively 
consistent during the study period, with the excep- 
tion of December, when surveys were conducted on an 
ad hoc basis (Table 1). Effort was lower in May than 
from January through April because surveys in May 
were suspended after two or three consecutive surveys 
when right whales were not sighted. The remaining 
temporal variability in survey effort was primarily 
due to weather conditions or sea ice that prohibited 
or shortened surveys. 
During the aerial surveys, 1553 right whales were 
sighted, the majority of which were sighted during 
February-April. Cumulative sighting totals included 
multiple sightings of photographically identified indi- 
vidual whales within and between seasons. The earli- 
est sighting within a season occurred on 13 December 
1998, and the latest sighting occurred on 2 May 1999. 
The highest number of right whales sighted on a single 
day was 47 on both 4 March 1998 and 27 March 2000. 
Sightings of 203 whales occurred during unacceptable 
survey conditions (mostly due to high sea state), and 
eight whales were documented for cells with acceptable 
conditions but during which there was less than 2 km 
of effort during a particular 2-week period; therefore 
sightings of 1342 whales were included in the SPUE 
analysis. Right whales were sighted during acceptable 
conditions in 69 of the 101 sampled cells (68.3%) over 
the five years of the study. Sightings were concen- 
