ABANDONED PENGUIN COLONIES
AND
CLIMATE CHANGE IN THE
ANTARCTIC PENINSULA
Steven D. Emslie, Department of
Biological Sciences,
University of North Carolina, Wilmington, NC
28403
INDEX
Abstract
Introduction
Methods
Identification and Analysis
of Organic Remains
Radiocarbon Analysis
Discussion
Paleodiet
Conclusion
References
Six abandoned colonies of Adelie penguin (Pygoscelis adeliae) were excavated near Palmer Station, Anvers Island, Antarctic Peninsula, to investigate the occupation history of this species. Sediments excavated from each site were sifted through fine-mesh screens and produced abundant fish bones and otoliths and squid beaks that represent prey remains deposited by penguins during the nesting period. Radiocarbon analyses indicate that colony occupation began prior to the Little Ice Age (LIA; A. D. 1500-1850), with the oldest site dating to 644 years before present (B. P.; average reservoir-corrected date with 1s range, 603-679 B.P.). Food remains indicate that the non-euphausid prey of penguins consisted primarily of a mesopelagic squid (Psychroteuthis glacialis) and two species of fish (Pleuragramma antarcticum and Electrona antarctica). The relative abundance of the first two prey taxa varied significantly among six sites (X2 > 34.6; df = 10; p < 0.001) with colonies dating prior to the LIA having greater representation of squid, and less of silverfish, than those occupied during the LIA. Data from control excavations at three modern colonies indicate a diet similar to that of the pre-LIA sites. These results suggest that Adelie penguins may have shifted their diet in response to warming and cooling cycles in the past. In addition, only Adelie penguins are known to have nested in the Palmer Station region prior to the 1950’s; Gentoo (Pygoscelis papua) and Chinstrap (P. antarctica) penguins now breeding in this region have expanded their ranges southward in the Peninsula within the past 50 years in correlation with pronounced regional warming.
Abandoned penguin colonies are becoming increasingly important for paleoenvironmental research in Antarctica. The excellent preservation of organic remains in ornithogenic soils (formed from bird guano) in these sites can provide significant amounts of information on the paleoecology and paleoclimatology of this ecosystem. Bones and other remains of penguins are well preserved at these sites and can provide further information on the age and paleoecology of the species that once occupied them. Although studies of abandoned colonies in east Antarctica have provided data on past fluctuations in penguin populations with climate change (Baroni & Orombelli 1991, 1994), detailed investigations in the Peninsula region are lacking. Studies at Palmer Station, Anvers Island (64° 04' W; 64° 46' S), Western Antarctic Peninsula, are indicating that abandoned colonies contain a wealth of information not only on the occupation history of penguins, but also on their past diet and foraging strategies. Moreover, Fraser et al. (1992) have proposed that populations of Adelie and Chinstrap penguins have fluctuated over the past 50 years in response to changing sea-ice conditions concurrent with a regional warming of 4-5 ° C. Data from abandoned rookeries may reflect similar responses by these species with climate change in the past.
Three species of penguins currently breed near Palmer Station. The most abundant species is the Adelie penguin (Pygoscelis adeliae) which has been studied in this region since 1973 (Parmelee 1992). In 1957, Chinstrap penguins (P. antarctica) were recorded breeding in small colonies on the Joubin Islands (35 nests) and at Dream Island (2 nests); no Gentoo penguins (P. papua) were known in the vicinity of Palmer Station until 1975 (Parmelee 1992). Both Gentoo and Chinstrap penguins have been increasing in this area since these first sightings, while Adelie penguins have declined (Fraser & Patterson 1997). The former two species may be expanding their breeding populations southward in the Antarctic Peninsula in response to climate change and a warming trend over the past 50 years that has increased the average mid-winter surface air temperature in this region by up to 4-5 ° C (Fraser et al. 1992; Smith et al. in press). If so, evidence for similar population responses may be found in the occupation record at abandoned penguin colonies. To address this question, six abandoned and three modern colonies were excavated in the Palmer Station region in austral summer 1997. Here, results of these excavations are presented with interpretations on the late Holocene environmental history of this region. These data also provide information on past diets and ecological responses by penguins to climatic change.
Nine abandoned penguin colonies were located in the Palmer Station region
by ground surveys (see map). Areas rich in vegetation and/or where
concentrations of pebbles were apparent were considered sites most likely
to be former breeding colonies. Vegetation (moss, lichens, and grass)
often grows more densely in former colonies, due to soil nutrient
enrichment from penguin guano, than in areas that were not used by
penguins. Pebble (used in nests) and bone concentrations on the surface
also characterize these sites. Areas meeting these criteria were tested by
probing beneath the surface to locate additional evidence for a former
colony (concentrated organic remains including bones, feathers, and
eggshell fragments). Once an abandoned colony was identified, a 1x1 m test
pit was established within the site and excavated in 5-10 cm arbitrary
levels, or by natural stratigraphic layers when apparent, until underlying
bedrock was reached. Volu metric measurement of all excavated sediments
from each site was accomplished using 20-liter buckets; these sediments
were then washed through three nested screens with mesh sizes of 0.64,
0.32, and 0.025 cm. Organic material trapped in the top two screens was
sorted and collected in the field; matrix in the 0.025-cm mesh screen was
dried, weighed to the nearest 0.1 g, and sorted under a low-power (5-10 X)
binocular scope. In this manner, most organic remains preserved in the
sediments, including penguin bones, feathers, eggshell fragments, fish
bones, scales, and otoliths, and squid beaks, were recovered. All test
pits were backfilled, and any surface vegetation replaced, at the
conclusion of the excavations.
Three modern colonies also were
excavated using similar techniques to act as controls against the number
and kinds of organic remains recovered from the abandoned sites. These
modern colonies were excavated late in the breeding season, after nesting
birds had departed, so that no disturbances to active nests were incurred.
In addition, four sites in marine terraces away from any abandoned or
modern colonies were excavated to act as additional controls on the amount
and kinds of organic remains that accumulate in sediments outside of
penguin colonies. Small portions (< 2 g) of bone were removed from
identified elements from each site for accelerator mass spectrometry (AMS)
radiocarbon analysis.
IDENTIFICATION AND
ANALYSIS
OF ORGANIC REMAINS
All penguin bones recovered from the abandoned colonies were identified as Pygoscelis sp. or P. adeliae; no bones of P. antarctica or P. papua were recognized. The majority of bones from each site represented juvenile individuals, recognized by their incomplete ossification and porous surface, as would be expected at a breeding site (Emslie 1995). Taxa of cephalopods and fish identified from abandoned colonies indicate that the most abundant taxa identified from all sites are a single species of squid (Psychroteuthis glacialis), Antarctic silverfish (Pleuragramma antarcticum), and lantern fish (Electrona antarctica). The number of identifiable specimens and MNI’s for the first two of these taxa vary significantly between the abandoned colonies (X2 > 34.6; df = 10; p < 0.001), but lantern fish do not (X2 < 16.1; df = 10; p < 0.1). In contrast, the modern Adelie penguin colonies at Litchfield and Torgerson Islands show no difference in numbers or MNI’s between sites for squid and lantern fish (X2 < 4.4; df = 2; p < 0.05).
The proportion of represented prey taxa based on the amount of sediments excavated may explain the significant variations in number of specimens and MNI’s from each site. These data indicate much higher proportions of Psychroteuthis glacialis, and lower of Pleuragramma antarcticum, from Cormorant Island and Sites 1 and 5 on Biscoe Point compared to Humble Island and Sites 1-2 on Litchfield Island. The two modern Adelie penguin colonies on Torgerson and Litchfield Island show few differences in the proportions of squid and fish represented at each.
Except for Beach Test 1, control samples from marine terraces show much lower proportions of squid and fish remains (total number of specimens and MNI’s per taxon) than in the modern and abandoned colonies. BT 1 had the highest proportion of Psychroteuthis in the sediments, similar to data from the abandoned colony on Humble Island, and higher than the modern Chinstrap colony on Dream Island. However, the total number of specimens of squid and silverfish remained significantly higher at Humble Island compared to BT 1 (X2 > 4.9; df = 2; p < 0.05); though the MNI of squid did not (X2 = 3.6; df = 1; p < 0.1).
Species richness of cephalopods was lowest from the abandoned colony at Humble Island (three taxa) and highest at Site 2 on Litchfield Island (six taxa); fish species richness also was highest at this latter site (nine taxa). In the modern colonies, species richness of cephalopods (five taxa) and fish (seven taxa) was highest at Litchfield Island, and similar between Torgerson and Dream Islands.
Photos showing an abandoned colony (top) recognized by the concentration of nest pebbles and rich vegetative growth on the site.
Excavations of abandoned colonies reveal layers of nest pebbles (middle) with glacial or beach sediments, consisting of angular gravel and/or sand below the colony sediments (bottom).
Radiocarbon dates on penguin bones and squid beaks provide an occupation history of Adelie penguins in the Palmer region for at least the past 644 years (based on average reservoir-corrected dates). The oldest dates are from the lower strata at Site 1 on Biscoe Point and Cormorant Island. Multiple dates from two sites (Site 1 on Litchfield and Cormorant Islands) suggest a relatively brief occupation period. Site 1 at Humble Island produced two dates that also suggest a relatively brief period of occupation at 254-421 B. P. (before present), and two modern dates. These last two dates may represent another brief occupation period in the more recent past, or are intrusive specimens that were mixed within the shallow sediments at this site. Site 5 at Biscoe Point is similar in having an older occupation period in the lower strata, with dates ranging from 450-598 B. P., as well as a younger occupation from 329 B. P. to the present. Site 2 at Litchfield Island exhibits the greatest range in dates and penguin occupation from 507 B. P. to the present, though it is unknown if these dates represent a single or multiple occupation period(s). Radiocarbon dates on squid beaks were similar to those on penguin bones from the same stratigraphic level at Biscoe Point Site 5 and the modern colony at Torgerson Island indicating correspondence in age of penguin bones and prey remains in the colony sediments.
Squid beaks from an
abandoned penguin colony.
The presence of bones of adult and juvenile Adelie penguins at all sites except on Cormorant Island, and the absence of bones that could be verified as other pygoscelid species, suggests that these colonies were occupied in the past only by the former species. The total radiocarbon chronology by site, and by stratigraphic levels within each site, indicates that all abandoned penguin colonies in the Palmer region were occupied either before (> 450 years old) or during (100 - 450 years ago) the Little Ice Age (LIA; A. D. 1500-1850; Grove 1988). Osteological identifications further indicate that all abandoned sites, except for Site 1 on Cormorant Island, were occupied by Adelie penguins.
These results allow comparison of ecological responses by Adelie penguins to past climatic fluctuations as evinced by the ice-core and marine sediment records. In the Antarctic Peninsula these records indicate that, prior to the LIA, it was relatively warmer than today (Smith et al. in press). Cool and warm periods occurred throughout the LIA (with more of the latter than the former) with temperatures steadily rising again over the past 50 years. Moreover, the warming over the past three decades is the most pronounced that has occurred in over four centuries as shown by a 480-year ice-core record from the Dyer Plateau (Thompson et al. 1994; Smith et al. in press).
The radiocarbon chronology of penguin bones so far indicates that only Adelie penguins have bred in the Palmer region beginning up to 603-679 years ago and that only recently have Gentoo and Chinstrap Penguins expanded their breeding distribution to this region. It is possible that the relatively rapid regional warming now occurring in the Antarctic Peninsula has favored the expansion of these species southward, while the Adelie penguin has declined (Fraser et al. 1992; Smith et al. in press). The absence of Gentoo and Chinstrap penguins in the Palmer region prior to 1957 supports this hypothesis.
In addition to this occupation history, the significant variations in prey species richness and relative abundance represented in the colony sediments suggest that Adelie penguins may have shifted their diet during different occupation intervals, or that diet varies by colony. Dietary remains recovered from the lower levels (Levels 2 and 3) at two sites excavated at Biscoe Point (Site 1 and 5) indicate a greater relative abundance of squid (Psychroteuthis glacialis) represented at Site 1, but few differences otherwise. Interestingly, one other site of similar age at Cormorant Island also produced a relatively high abundance and species richness of squid remains in the sediments. These results suggest that Adelie penguins consumed a greater proportion of squid from 679 to 450 B.P., or prior to the LIA, perhaps due to greater squid availability (especially P. glacialis) over other prey taxa at that time.
Dietary remains from colonies that were occupied during the LIA on Litchfield and Humble Islands indicate a lower proportion of squid was consumed, especially at Site 1 on Humble Island, but that fish species richness and abundance was significantly higher in these sites compared to pre-LIA sites on Cormorant Island and Biscoe Point. These data further suggest that Adelie penguins may have shifted their diet from a relatively greater proportion of squid prior to the LIA to a greater proportion of fish during the LIA, perhaps in response to warming and cooling cycles that may have affected prey abundance.
The high abundance of squid beaks in the sediments of the modern Adelie penguin colonies also is of interest as squid are rarely encountered in stomach samples from living Adelie penguins (Williams 1995). A low frequency of Pyschroteuthis glacialis has been reported in Adelie penguin diet at Adelie Land, but only from samples gathered late in the breeding season (Offredo et al. 1985). These authors speculate that this squid species may move to shallower depths in February and therefore become accessible to predation by penguins only late in the breeding season. If so, a similar response may be occurring in the Palmer region. Diet samples collected from Adelie penguins in January over the past 10 years have failed to produce any substantial remains of squid, yet squid beaks are common in the colony sediments. The samples from the three screen traps placed on the modern colonies verify that these penguins presently are feeding on squid at some point during occupation or attendance of their breeding sites. These results suggest that the penguins must be taking squid early or late in the season, while they are still occupying their breeding sites, but feed primarily on krill (and some fish) during the chick-rearing season as shown by the stomach samples. Additional research is needed to determine the importance of squid in Adelie penguin diet and the environmental circumstances in which this prey species is taken.
A penguin feeding its chick by direct regurgitation into the chick's mouth.
Investigations of abandoned colonies have provided an occupation history of Adelie penguins in the Palmer Station region over the past 644 years. In addition, this research raises new questions on the ecological response by penguins, especially dietary shifts, to changing environmental conditions, e.g., do the proportions of various prey species change with warming and cooling trends in a predictable pattern in Adelie penguin diet? Investigation of abandoned colonies also may provide information on sea-level and ice-margin changes in the past (Goodwin 1993). Adelie penguins require ice-free terrain for their nesting sites. Excavations indicated that sediments below ornithogenic soils were deposited either from glacial melt water or on marine terraces at or near sea-level in the past. Thus, radiocarbon dates on organic debris in the ornithogenic soils immediately above the glacial or marine deposits can provide a minimum age for the retreat of the ice margin or lowering of sea-level.
A Chinstrap penguin inspects
a backfilled excavation pit
in an abandoned colony, King George Island.
Baroni, C. & Orombelli, G. 1991. Holocene raised beaches at Terra Nova Bay, Victoria Land, Antarctica. Quaternary Research, 36, 157-177.
Baroni, C. & Orombelli, G. 1994. Abandoned penguin colonies as Holocene paleoclimatic indicators in Antarctica. Geology, 22, 23-26.
Emslie, S. D. 1995. Age and taphonomy of abandoned penguin colonies in the Antarctic Peninsula region. Polar Record, 31, 409-418.
Fraser, W. R., Trivelpiece, W., Ainley, D. G. & Trivelpiece, S. G. 1992. Increases in Antarctic penguin populations: reduced competition with whales or a loss of sea ice due to environmental warming? Polar Biology, 11, 525-531.
Grove, J. M. 1988. The Little Ice Age. London: Methuen, 498 pp.
Parmelee, D. F. 1992. Antarctic birds. Minneapolis: University of Minnesota Press, 202 pp.
Smith, R. C., Domack, E., Emslie, S. D., Fraser, Ainley, D. G., Baker, K., Kennett, J., Leventer, A., Stammerjohn, S., Mosley-Thompson, E. & Vernet, M. In Press. Climate change and ecological responses of the Antarctic marine ecosystem. Bioscience.
Thompson, L. G., Peel, D. A., Mosley-Thompson, E., Mulvaney, R., Dai, J., Lin, P. N., Davis, M. E. & Raymond, C. F. 1994. Climate since AD 1510 on Dyer Plateau, Antarctic Peninsula: evidence for recent climate change. Annals of Glaciology, 20, 420-426.
Williams, Tony D. 1995. The penguins. Oxford: Oxford University Press, 295 pp.
A complete published version of this paper is available at:
Emslie, S. D., W. Fraser, R. C. Smith, and W. Walker. 1998. Abandoned penguin colonies and environmental change in the Palmer Station area, Anvers Island, Antarctic Peninsula. Antarctic Science 10 (3): 257-268.