Polar Biol (2008) 31:1397–1407
DOI 10.1007/s00300-008-0479-zORIGINAL PAPER
Population trends of Adélie penguin (Pygoscelis adeliae) breeding
colonies: a spatial analysis of the eVects of snow accumulation
and human activities
P. K. Bricher · A. Lucieer · Eric J. Woehler
Received: 11 February 2008 / Revised: 20 May 2008 / Accepted: 26 May 2008 / Published online: 2 July 2008
 Springer-Verlag 2008
Abstract Many of the Adélie penguin colonies used for
long-term demographic studies are located near research
stations, and there is a need to disentangle the eVects of
human activities and environmental variability on Adélie
penguin population trends. This study used Geographic
Information Systems and decision trees to examine whether
potential changes in snow cover and/or proximity to human
activities were able to explain the varying population trends
of colonies at two breeding localities near Casey, East Ant-
arctica. At the less visited site, Whitney Pt, wind exposure
and snow accumulation correctly predicted the broad trends
of colonies in most (84%) instances, whereas at Shirley I
their predictive accuracy fell to 58%. At Shirley I, proxim-
ity to human activity correctly predicted the broad trends
for 84% of colonies. While snow accumulation patterns are
a primary driver of variation in population trends among
colonies, the eVect of snow accumulation is outweighed by
proximity to human activities near Casey.
Keywords Adélie · Penguin · Human · Snow · Spatial
Introduction
Adélie penguins (Pygoscelis adeliae) have been widely
used as indicators of changes in Antarctic Ocean ecosys-
tems (Micol and Jouventin 2001; Croxall et al. 2002; Kato
et al. 2002, 2004). There are a number of reasons for this,
including their perceived primary role in these ecosystems
and the relative ease with which their populations can be
monitored (Micol and Jouventin 2001; Kato et al. 2002).
However, it has also been acknowledged that the use of
birds as bioindicators of climate change may be problem-
atic due to the complex and numerous interactions in Ant-
arctic ecosystems (Croxall et al. 2002), and the potential
confounding eVects of human impacts at local scales. Adé-
lie penguin colonies are known to be abandoned and recol-
onised as the climate changes (Ainley 2002; Emslie and
Woehler 2005).
Many recent investigations of Adélie penguin population
trends have focused on the eVects of climate variability,
mostly at meso- and macro-scales (Trivelpiece and Fraser
1996; Croxall et al. 2002; Kato et al. 2002, 2004; Forcada
et al. 2006). Others have attempted to separate the eVects of
climate change and human activities (Fraser and Patterson
1997; Micol and Jouventin 2001, Patterson et al. 2003).
Long-term studies of Adélie penguin populations have gen-
erally been conducted at sites that are easily accessed from
research stations, where human activities are focused (e.g.
Woehler et al. 1994; Fraser and Patterson 1997; Micol and
Jouventin 2001; Woehler et al. 2001). Variability in the
sizes of stations, the activities associated with them histori-
cally and currently, and their distances from penguin colo-
nies makes it diYcult to establish a broad understanding
of the eVects of human activities on Adélie penguins
(Thomson 1977; Ainley et al. 1983; Woehler et al. 1994;
Fraser and Patterson 1997; Micol and Jouventin 2001;
P. K. Bricher (&) · A. Lucieer
School of Geography and Environmental Studies,
University of Tasmania, Private Bag 76,
Hobart 7001, Australia
e-mail: pbricher@utas.edu.au
E. J. Woehler
Australian Antarctic Division, Channel Highway,
Kingston, TAS 7050, Australia
Present Address:
E. J. Woehler
School of Zoology, University of Tasmania,
Sandy Bay, TAS 7005, Australia123

1398 Polar Biol (2008) 31:1397–1407Patterson et al. 2003). This makes it diYcult to disentangle
the relative eVects of climatic variability and human
activities on trends in breeding Adélie penguins (Fraser and
Patterson 1997; Micol and Jouventin 2001; Patterson et al.
2003). As the number of people visiting Antarctica increases,
so does concern about the potential impacts of human dis-
turbance on Antarctic wildlife (PfeiVer and Peter 2004). In
2005/06, more than 26,000 people visited Antarctica on
tourism vessels (IAATO 2006). In addition, almost 4,000
people live in research stations located throughout Antarc-
tica during the summer (COMNAP 2006).
Very few studies have attempted to investigate the spa-
tial variability within and among Adélie penguin colonies
within a breeding locality. Only Fraser and Patterson
(1997) and Patterson et al. (2003) simulated snow accumu-
lation patterns based on a digital elevation model (DEM)
and hillshade analysis. They used human visitation records
to investigate the relative eVects of snow accumulation pat-
terns and human impacts on breeding success of resident
penguins.
This study aimed to determine the contribution of snow
accumulation patterns, wind exposure and proximity to
human activities to the observed long-term population
trends of Adélie penguin colonies at two breeding localities
in the Windmill Is, Antarctica. In addition, this study
attempted to diVerentiate between environmental and
human-induced eVects on penguin population trends at a
local scale. It improved on previous investigations of the
eVects of human activities and snow accumulation patterns
on population trends in Adélie penguin colonies, by using a
more accurate physically-based snow accumulation model
and a high-resolution DEM to model Wner scale patterns.
Methods
Study areas
The Windmill Is are the islands and coastline covering an
area of about 80 km
2
around Casey (66° 17
S, 110° 32
E)
in Wilkes Land, East Antarctica (Fig. 1). They comprise
four large peninsulas and more than 30 islands (Murray and
Luders 1990; Kirkup et al. 2002). During summer, the
Windmill Is contain the only extensive areas of snow-free
land in approximately 800 km of coast (Murray and Luders
1990; Kent et al. 1998). The Windmill Is contain extant
Adélie penguin colonies on 14 islands and peninsulas, with
a total population estimated at 93,000 § 9,300 pairs in
1990 (Woehler et al. 1991). The Windmill Is weather is
frigid-Antarctic (Melick et al. 1994). Weather observations
between 1989 and 2004 showed that in the warmest month,
January, mean daily temperatures ranged from ¡2.6° to
2.1°C. October was the coldest month of the breeding
season, with a mean daily temperature range of ¡15.3° to
¡8.3°C.
Census data
A breeding colony is here deWned as an area of contiguous
nest territories sensu Woehler et al. (1991, 1994). A breed-
ing locality is a geographical feature, either an island or a
discrete area of mainland, on which breeding colonies are
found. This contrasts with the use of the term “colony” to
describe a breeding locality sensu Ainley (2002).
Whitney Pt (66° 15
S, 110° 32
E) is one of two main-
land Adélie penguin breeding localities in the Windmill Is
and is located on the Clark Peninsula, which was desig-
nated a site of special scientiWc interest (SSSI) in 1985. In
1996, the Clark Peninsula SSSI was redesignated Antarctic
Specially Protected Area 136 (AAD 2006). Shirley I (66°
17
S, 110° 29
E) lies approximately 750 m west of Casey,
across a 100 m-wide channel that is blocked with sea-ice
for part of the year. Adélie penguin census data were avail-
able for the colonies at Whitney Pt for 22 breeding seasons
between 1959/60 and 2005/06, and for the Shirley I colonies
Fig. 1 Map of the Windmill Is, East Antarctica. Localities mentioned
in the text are shown123