250 
THE WILSON JOURNAL OF ORNITHOLOGY • Vol. 124. No. 2. June 2012 
TABLE 3. Estimated Leach’s Storm-Petrel breeding population on Grand Colombier Island in 2008. Parameters are 
presented as mean ± SE. The 95% confidence interval is in brackets. 
Sector 
Parameters 
Plaleau 
Southern side 
Northern side 
Steep ana 
Number of plots 
54 
67 
36 
5 
Sector area (nr) 
105.542 
187.785 
160.794 
32.662 
Sector slope ( ) 
16.3 ± 1.2 
31.6 ± 0.9 
36.0 ± 1.1 
40.0 + 2.2 
Adjusted sector area (m’) 
109.962 ± 31 
220.476 ± 51 
198.752 + 67 
42,637 + 71 
Burrow density 
Detection probability 
Occupancy rate 
Breeding success (10 Jul) 
0.32 ± 0.06 
0.89 ± 0.01 
0.55 ± 0.03 
0.93 ± 0.03 
1.15 ± 0.09 
1.22 ± 0.11 
0.45 ± 0.13 
Number of breeding pairs 
Total number of breeding pairs 
23,426 ± 4.063 
363,787 ± 19.991 
167,557 ± 12,722 
[295,502-432,0721 
160.267 ± 14.427 
12.536 ± 3.626 
hatching (mostly due to infertile eggs). Leach's 
Storm-Petrel surveys should be conducted no later 
than the first half of the incubation period to 
minimize underestimation and if breeding failure 
cannot be estimated precisely to correct survey 
estimates. 
Population Estimate on Grand Colombier 
Island. The world Leach’s Storm-Petrels popula¬ 
tion is estimated to be >8 million breeding pairs 
(Huntington et al. 1996). The species’ breeding 
range is centered in the northwestern Atlantic 
Ocean in eastern Newfoundland, Canada, where 
over hall of the world's breeding population (up to 
5 million pairs) occurs (Huntington et al 1996) 
Most colonies are in the Newfoundland region with 
the world’s largest colony reaching 3,360 000 
breeding pairs at Baccalieu Island (Sklepkovych 
and Montevecchi 1989). Thus, the Grand Colomb¬ 
ier colony contributes -6% of the North Atlantic 
breeding population. 
The Grand Colombier Leach’s Stonn-Petr 
colony ranked second with -363.000 breedir 
pairs among Leach’s Storm-Petrel colonies in tf 
northwestern Atlantic (i.e., Newfoundland Lai 
rad° r , Canada; and St Pierre and Miquelo, 
rranee). Previous surveys conducted during if 
l e nd 0s and : n 2004 0ra " d C “ 
— 178 OOfW n 11 y e y ' estimatcd population size ; 
- 43 On K a mSSeS 3nd Ktehebe "y 1989) an 
143.000 breeding pairs (Robertson et al 2006 
Our eat, mat e is twice those in previous years. Thi 
Leach\"I, “ “S kC ,' y 10 e "' irel >' from 
at ^ ^ St °rm-Petrel population increase. A mea 
S Z di M 8 0f 5 ^ (Huntington et al 
). «nd overall mean annual survival of 0 7 < 
(Huntington et al. 1996) s U a a&Kt * 
annual growth rate n \ n • maxima 
max), following Niel ant 
Lebreton (2005), is 1.1 (i.e., a maximal 10‘S 
annual increase). Large Leach’s Storm-Petrel 
colonies appeared stable in the northwestern 
Atlantic from the 1970s to early 2000s. and only 
small colonies showed declines (Robertson et al. 
2006). The difference between the 2004 and 2008 
surveys probably partly results from a sampling 
artefact as sampling effort was highly variable 
between surveys. The sampled area consisted of 8 
X 25-nr plots in the late 1980s (200 nr). 90 * 
16-trr plots in 2004 (1.440 nr), and-162 X 28.3- 
nr plots in 2008 (4,585 nr). Burrow density was 
30% higher in 2008 than in 2004. whereas 
occupancy rate was only 11% lower in 2008. 
I he 2008 survey was the first to include burrow 
detection probability, which indicated that burrow 
density could be underestimated bv !!%• We 
showed that breeding failure from mid-June to the 
start of the survey could account for a 7% loss in 
breeding pairs. 
Identification of potential threats to this popu¬ 
lation, because of the significant size of this 
colony, should be encouraged for effective 
conservation. We observed Leach's Storm-Petrel 
remains in regurgitation pellets at Herring Gull 
(Units argentatus) nests. Predation of Leach’s 
Storm-Petrels by Herring Gulls has been repotted 
(Stcnhouse et al. 2000) with up to 9% of a colony 
ol 269.765 breeding pairs being killed by 2.I+* 
gull pairs in one breeding season. Predation was 
mostly by specialized individuals or pairs (116% 
ot the gull breeding population: Stenhouse el al. 
2000). These authors considered that, despite 
large losses, the Leach’s Storm-Petrel breeding 
population did not appear to substantially decline, 
probably because recruitment could maintain the 
population. We estimated that 60-100 gull pairs 
