Tracking Foraging Trips of Penguins at Ross Island, Antarctica

Jean Pennycook ©

A Research Paper Presented to Paul Veisze

In Partial Fulfillment of the Requirements for

Geography 350, Data Acquisition in GIS

By

Jim Myers

May 2009

Abstract:

The Adelie penguins are one of the indicator species of Antarctic sea ice conditions, and in a larger sense, an indicator of global climate change.

Optimal foraging theory assumes that the fitness of an animal is a function of the efficiency of chick provisioning and self-feeding. Time and energy constraints weigh heavily on the allocation between these two demands. Balancing optimal foraging trip lengths, available food density, and resource allocation to self-maintenance or chick provisioning weigh heavily on penguin foraging patterns and habits. The ecological model is currently complicated by a need to better understand the spatial relationship between sea ice cover and not only the Adélie penguins, but also for the subsequent location/production/and possibly survival of the penguins’ food source.

This study tracks telemetry data collected from seven Adélie penguins (Pygoscelis adeliae) in the Ross Island area of Antarctica. These movement patterns are influenced by the distribution of ice and the presence of their prey (food source). The data for this study was collected by Dr. David Ainley and his research team during austral summers of 2007 and 2008. Dr Ainley and his research teams have studied Adelie penguins for over 25 years. The results were found to be within the ranges and characteristics observed by other researchers. Understanding the life of penguins is an interesting study in its own right, but many interconnected influences are at work in the association between penguins, near-shore and sea ice, their local environments, foraging needs and patterns, and global environmental change. Many researchers now study the effects of global climate change on polar and coastal ecosystems. This is a part of an indicator species research effort.

Introduction:

Purpose

Antarctica is magical and intriguing part of the world that few human beings have ever set foot upon. Due to recognition of the special nature of this area, and the unique creatures that inhabit it, many areas have already been designated with protected limited access. Numerous countries have research facilities in Antarctica. McMurdo Station is the main US research facility.

In the austral summer, from about mid-November until early February, Adélie penguins on Ross Island are busy with the process of courtship, nest building, egg laying, incubation, and rearing chicks to the point where the offspring can fend for themselves. During this time, foraging efforts of the parents must be adequate to meet the nutritional needs of both the chicks and the adults. Typically both parents lose body mass during this chick-rearing period. This has also been postulated as a factor for other birds, such as the Black-footed Albatross.

Foraging trip lengths and durations are a few of the factors of interest to scientists studying these birds. They are also interested in where the foraging areas are, as well as the commute routes to the feeding grounds.

Scope:

The scope of this project is to acquire and analyze telemetry data from seven Adélie penguins, tracked during a period of December/January of 2008 and 2009. The purpose of this paper is historical documentation of foraging trip tracking, rather than making strict scientifically-supported statements, interpretations, or observations about the birds. It is a focus on the basic biological question of how far and where do Adélie penguins currently go to get food for their chicks. This study is not a rigorous biological study. Additional information needed for such a research design is not currently available for consideration in this paper. Such information is discretionary proprietary property, and not available for general release. Discussions and any conclusions reached in this report are intended for historical information and background only, as influenced by the literature review.

Image Courtesy of Google™

Exhibit ‘1’

Procedure:

Raw tracking data will be parsed and imported into ArcGIS for geo-referencing, geospatial review. Routes taken by the telemetry-tracked birds will be mapped. Each bird’s route will be plotted for visual analysis, with a tabular summary of each bird’s foraging trip characteristics.

Acknowledgements:

This geospatial analyst is deeply appreciative to those who have made this data available. The telemetry data was collected as a part of the work of Dr. David G. Ainley and his research team, who are studying Adélie penguins at Ross Island, Antarctica. Dr. Ainley has been involved in penguin research for over twenty-five years, and is heavily published and respected expert on Adélie penguins. I am also indebted to my friend Jean Pennycook, the educational liaison for Dr. Ainley’s research. She has spent the past seven seasons in Antarctica, involved in various aspects of Antarctic research. She is also the source of much of the photography in this work.

Jean Pennycook©

Background (The Antarctic Research Setting):

Ross Island, Antarctica is located approximately 800 miles from the geographic South Pole. New Zealand is the closest northerly access route to this region, and Christchurch is the departure point for the trip to McMurdo Station. The flight from Christchurch to McMurdo Station is approximately 2200 miles. Exhibit 2A and 2B (below) illustrate the geographic site location. Exhibit 3 shows a more detailed map of Cape Royds on Ross Island.

Exhibit 2A Exhibit 2B

Courtesy USGS

Exhibit 3

A complete large-size map of Ross Island can be viewed and zoomed by using the following link: http://usarc.usgs.gov/drgs/dir1/c77190s1.jpg

Air transportation for U.S. research efforts in Antarctica is provided by the 109^thAirlift Wing at Stratton ANGB (New York Air National Guard) base in Scotia, New York. This support is fully funded by the NSF.

Ski-equipped LC-130 Hercules Transport Aircraft

U.S. Coast Guard icebreakers open a route for marine access.

USCG ice breaker Polar Star

The ice breaker Polar Star is able to ram her way though ice up to 6.4 meters (21 feet) thick and steam continuously through 1.8 meters (6 feet) of ice at 3 knots (5.6 km/hour). There are three pairs of connected tanks located on opposite sides of the ship to create a movement necessary to efficiently work through the ice. Pumps transfer a tank's contents (133 kiloliters, 35,000 gallons) to an opposing tank in 50 seconds and generate 64,800 kilowatt-seconds (24,000 foot-tons) of torque on the ship. Operations in the remote, hazardous, and unforgiving Polar Regions make it necessary for the crew to be highly self-sufficient. Crew members have extensive training in a variety of special skills.

McMurdo Station is the main U.S. research station area, with a peak austral summer researcher/support crew population of approximately 1100. Raytheon Polar Services currently holds the contract to provide logistical and facilities support for the U.S. Antarctic scientific research effort. All Raytheon personnel supporting the Antarctic logistics are deployed from Raytheon’s Centennial, Colorado location. There is a small contingent of core of personnel who remain during the severe Antarctic winters, but their numbers are very limited. Most on-site penguin research occurs from approximately mid November to end of January. Temperatures at McMurdo during this time can range from -20°C to as high as +10°C. Temperatures at the geographic South Pole during the austral summer can range between -30°C to -20°C.

Background (Telemetry and Satellite Imagery for Penguin Research):

Much research has been conducted on penguin diets, foraging habits, reproductive success, sea ice, and the interconnected biological relationships between the birds and their environment. Other species such as the Emperor, Magellanic, and Chinstrap penguin colonies occur in areas of Antarctica, and are studied by different research groups. GIS and satellite technology is currently being used for scientific research purposes. The following link, http://link.brightcove.com/services/player/bcpid1256280131?bctid=1258472794

is a four-minute video describing how Dr. Jerry Kooyman, Scripps School of Oceanography is using imagery from Digital Globe satellites to track and count Emperor penguin populations. He and his research team are still trying to figure out how to best use the satellite imagery tools to conduct their studies. Because most of the area where the penguins live is inaccessible to researchers, satellite imagery and telemetry tracking are powerful tools to obtain information not available by traditional methods. Another link to a long video presentation at UCSD relating to Dr. Kooyman’s work, with additional input from Dr. Ainley on Adélie penguins, is in the “Links” appendix at the end of this paper.

Satellite telemetry is expensive, both in terms of the cost of transmitters ($2000-$3,000 each), and the costs of accessing data from the satellite system (Bradshaw, et al.). Telemetry transmitters must not impede the penguin’s swimming or normal habits. Attachment difficulties can result in interrupted tracking durations and data losses, when the units become detached. Some of these units also contain transducers for recording diving depths of the penguins. All of this is spatial data, which can readily be analyzed and displayed in a GIS. Despite the transmitter attachment and cost challenges, GIS tools and methods are expanding.

During the austral summer of 2008, the Antarctic Geospatial Information Center (AGIC) created a geospatial analyst contact position at McMurdo Station. AGIC is an NSF-funded program that has been established to serve the geospatial (GIS) needs of USAP operations, research, and educational communities. The current contact is Michelle LaRue, who coordinates Antarctic geospatial research support through the University of Minnesota.

Currently there is a research proposal waiting for funding between NASA and Dr. Ainley’s group. This study would use remote sensing to further analyze the effect of sea ice dynamics and penguin demographics.

Background (The Research Efforts):

Exhibit ‘1’ (page 3) shows the location of Adélie penguin colonies in the Ross Island area. The birds have colonial nesting habits, and typically return to their native colony. A small percentage of birds have been found to move between colonies. Ice and food availability appears to influence the movement decisions. The Cape Royds penguin colony is geographically at 77.5440° S, and 166.0869° E. See the map on Exhibit ‘3’ (page 5) to identify the location. The sea is the sustaining base for life in Antarctica. Both predator and prey depend upon it.

Optimal foraging theory assumes that the fitness of an animal is a function of the efficiency of provisioning and self-feeding. Penguins must forage for food to sustain their metabolic needs, as well as sustaining the metabolic needs of their chicks, until the chicks are large enough to fend for themselves. Time and energy constraints weigh heavily on the allocation between these two demands. Low reproductive rates of long-lived sea birds have been attributed to the limitations on the acquisition of energy during the breeding season (Lack 1968). Fluctuations in the availability of nearby food sources and the distances involved for transporting food have a direct impact on the reproductive success of the birds.

Typical lifetimes are ~15, up to 20 years (Ainley), and it appears that there is increased reproductive success with age and experience. Chicks usually do not return to their native colonies for about three to five years. The extent of the sea ice affects the duration of foraging trips, as the penguins must walk over the ice to reach open water in their search for food. Foraging trips may be long or short, depending upon the stage of chick rearing, and the nutritional condition of the adult. Short trip durations may be around 20 kilometers, while the longer trips are in the 80-120 kilometer range. There are differences between species, and some penguins go much further in their long foraging trips. The penguin diet consists of fish, squid, and krill. An amazing aspect of this foraging activity is that apparently the penguins can stop or delay digestion, so their stomachs can act as a storage repository until mealtime for the chicks (Boresma).

After the eggs are laid, the male incubates the clutch for about two weeks, while the female goes to sea to feed. In the last week of incubation (about 34 day incubation), the pair relieve one another every other 1-2 days. The guard stage lasts about three weeks after the chicks hatch. Feeding requirements are high, and chicks must always be protected against the predatory Skua. At post-guard or crèche stage, both parents must forage for food and leave the fledglings un-guarded. Until the end of the guard stage (about the beginning of January) the small, weak, and uncoordinated chicks can be attacked directly by the skuas.

Skuas attacking an unguarded chick

Changes in body mass and foraging trip lengths have been found to be different for male Adelie penguins versus the females. The males have statistically higher body weight than the females at the beginning of the nesting period. In one study, after the eggs hatch, both sexes lost significant body mass, but the males typically lost about twice as much. Males typically undertook shorter foraging trips than the females. This study also indicated that the females made longer foraging trips and consumed more and different food, as determined by an automated weighing device (PIT microchip monitoring) and guano analysis. Birds are marked with a ‘flipper band’ number on their left flipper (for easy band location and visual identification), and some birds have implanted PIT (passively-interrogated transponder) tags which can identify individual penguins remotely as they walk over a computerized weighbridge (wb). Much collaborating data can be collected and associated with these trips.

Short trips occur most frequently during guard stage, when the chicks are small, and must be tended constantly. Departure weights prior to long trips were significantly lighter than those for short trips. The decision to forage far from the colony probably occurs when body condition reached a threshold level.

Background (Sea Ice Influences):

The ice influences populations through a number of processes operating at different temporal and spatial scales. Study results at Bechervaise Island, Antarctica show that the presence and timing of different types of ice have an impact on reproduction of Adélie penguins (Emmerson). The implications of near-shore and off-shore ice and Adélie penguin populations are complicated by timing as well as presence of the ice, which has an annual dynamic. To illustrate these differing conditions, compare Exhibit 4 and Exhibit 5, below. Exhibit 4 shows the area west of Ross Island and Cape Royds as being open water.

Year 2000, prior to ice berg B-15 and B-16 Exhibit 4

However, in Exhibit 5 ice exists almost all the way north to Cape Bird.

Year 1995 Exhibit 5

(Ross Island is in the center of Exhibit 5). Refer to earlier exhibits to orient features on Ross Island. In 2008, the ice had not moved out in usual fashion, and the birds had to walk 50 miles to reach the water’s edge until the ice finally did open up. See Exhibit 6, below. The red arrow in exhibit 6 indicates the approximate location of Cape Royds. The sea- ice edge was approximately 50

miles away, which was a two-day walk, one way, and prior to any feeding. This extensive ice resulted in birds abandoning their eggs and nest because of long-overdue returns of the mate. Once the nest is abandoned, the eggs are quickly scooped up by skuas.

There is concern about the impact environmental change will have on the presence of sea ice and the resulting impact on penguin populations. The type, extent, and location of the ice are thought to have varying impacts on penguin populations. Near-shore ice has a strong negative impact on penguin productive performance. At some sites, this has been associated with the lengthened duration of foraging trips created by the extensive near shore ice. At the same time, offshore ice (usually pack ice, located further out) appears beneficial when it has greater presence than the near shore ice during the guard stage (from the time the chicks are hatched plus 3 to 4 weeks). This appears to be due to the influence of prey (food) availability and the proximity of the food to the nesting areas. During the guard stage, short foraging trips are needed for frequent feedings critical to chick survival.

In the Bechervaise Island study, a number of models were tested for significance over a 17 year period, ending in 2006/2007. Some models were poorly represented during the years studied. There was an advantage to considering temporal influences of ‘winter’ vs. ‘summer’ ice rather than an average annual value. Similarly, considering guard stage ice in both near-shore and off-shore regions demonstrated better explanations of the data. The strongest model indicated that breeding success is a factor of sea ice cover, but it also suggests that only certain combinations of spatial and/or temporal measures of sea ice are important. The combination of reduced near-shore ice and extensive pack ice further off-shore during the guard stage produce optimal conditions for good access and suitable foraging habitat.

The understanding at this juncture is complicated further by a need to better understand the spatial relationship between sea ice cover and the subsequent location/production/and possibly survival of the prey (food source) of the Adélie penguins. Because of the complexity in the food web, and interaction between species in relation to the physical environment, it may be difficult to describe a clear connection (Ainley & Nichol et. al. 2007). It is clear that sea ice influence penguin populations, but more research is need to clarify the mechanisms at work for different populations experiencing very different ice environments. For example, Ross Island is further south than Bechervaise Island, and in a scenario of reduced sea ice due to global climate change, Ross Island populations may benefit. Because of the strong link between reproductive performance and the ice environment, Adélie penguin populations are likely to be sensitive to future changes in the ice environment.

Sometimes things happen to an adult while foraging for food.

Leopard seal and penguin

Methods:

The raw data were prepared in Microsoft Excel, and put in the proper formats for input into ArcGIS software. The raw data for each bird was imported into a separate GIS file. The data were displayed, and track-line was digitized to connect the date-referenced data points, creating a scaled plot of the data. Total travel distances for each bird, as well as measurements of trip statistics revealed by the data plots, were extracted. A table was developed with the collected and averaged statistics. A scatter plot was created to show all data points for all of the birds (See figures1 and 1A).

Results:

Shown below are seven telemetry tracking plots (Figures 2 through 8) of the birds in the study sample. Raw date, latitude and longitude data are also displayed below the plots. Figure 1 is a scatter plot, showing all data points for all of the tracked subjects. Figure 1A is a foraging trip summary table of notable statistics developed for each bird. The model bird is 70278, who easily has the most locations close to the nest. The short star-pattern foraging trips appear to demonstrate systematic searching for the best foraging grounds, with one long trip during the middle of the tracking dates.

In the displayed data files, southern hemisphere latitude is always negative, east longitudes are positive, and west longitudes are negative. With the exception of penguin 79610, all of the observations were east longitudes.

Penguin Foraging Trip Summary Table

Figure 1A

Jean Pennycook©

Analysis:

The plots represent the tracking paths in a spatially-correct layout, but do not show detail of ice conditions at the time. No-cost, high-resolution imagery of the area during the two austral summer years covered did not appear to be available. The technical and financial challenges to be to be addressed were greater than the benefit to be gained for the purpose of this study. With the limited raw tracking data available for this study, greater detail would probably not add significantly to the interpretive value. The coordinate system used for the plots was a south pole stereographic, with a stereographic projection.

The available latitude/longitude coordinate data were to the nearest hundredth of a degree (0.01). By definition, one minute of arc measurement in latitude is the same distance everywhere on earth (~6080 feet or one nautical mile). A hundredth of a degree is roughly 0.6 nautical miles in latitude, or 3,648 feet. Accuracy of the latitude position plot using this measurement scale would be approximately half of that amount, or 1824 feet (0.34 statute miles). A 0.0001 degree difference corresponds to 36.5 feet, which is common for GPS position reporting. Latitude positioning in this case is very coarse.

The distance difference in longitude values is different at every latitude because longitude lines converge at the poles. At latitude -76°S, a 0.10 degree difference in longitude arc distance would be significantly less distance than a 0.10 degree difference in latitude arc distance. However, latitude positioning accuracy (1824’) still compromises the latitude position plot, without any calculation of the longitude component. Considering current technology, the original raw data must have been much more accurate (at least in latitude measure).

Not having access to additional information about the individual birds (weights, gender, age, etc.) was unfortunate, and therefore left outside of the developed scope for this paper. Food loads returned increase with trip length (Ainley), but data was not available to explore this. The tracking observations would be more helpful if they were reported/available more often (maybe every six hours). Penguins act like optimal foragers as long as the feeding trips are less than or equal to two days (Ainley). With one observation (assuming equally-spaced temporal observations) data point per day, durations of nest-sitting are difficult to ascertain. However, this is probably a moot point, since the data location plots are so coarse. A bird could actually be a quarter-mile north of the nest and still have reported coordinates that are the same as the nest coordinates. It is quite possible that weighbridge data was used to monitor nest time.

Jean Pennycook©

Conclusions:

The tracking of these seven birds demonstrated the diversity of trip lengths and different geographic patterns covered. Due to wide variations in the habits of the subjects, and a small sample size, it is difficult to say much in a definitive sense. The ‘usual’ short-trip distance was around 40 KM or less, with an approximate maximum long-trip distance of 120 KM to 170 KM. The literature indicates that the usual foraging trip length requirements of the Cape Royds penguins might be shorter than in some other areas. Winter migration distances are another study.

Tracking studies from Bechervaise Island in 1997 indicated local foraging trips taken particularly by males during the guard stage would be about 20 KM. This appears consistent with some of the trip lengths in this sample. Offshore trips at Bechervaise Island were carried out by males or females, 80 KM to 120 KM offshore (Clarke et al. 1998). This also appears to be consistent with the observations of this tracking study.

From literature reviews it is clear that studies are being done which have a lot more data to work with in conjunction with the tracking data. More frequent tracking intervals (maybe six-hour rather than daily) would allow better analysis of time spent on the nest. Knowing the gender and weights as monitored by the weigh-bridges, and retrieving guano samples after the trip might provide more spatial information for analysis about the type of food collected in a particular region of the ocean. All of this and more may have been done, but the information at hand does not tell that story.

Telemetry and GIS analysis can provide powerful analytical capability, as well as interactive data access/display. This is much more than just static mapping. GIS can also use of current imagery for identification of sea ice limits (and possibly depths from remote sensing methods), since these ice factors significantly influence foraging routes and distances. Models of sea ice influence on penguin population would benefit from in-depth study of spatial and temporal scales of the ice, and its influence on demography. GIS technology is the tool for this analysis.

Currently, Adélie penguins are not endangered. Although their populations are declining in some colonies, they are stable in others. These species are important to watch, as they are bio-indicators of environmental change.

Jean Pennycook©

Field Accommodations

References:

1. Ainley, D.G., E.F. O’Connor, “The Marine Ecology of Birds in the Ross Sea, Antarctica”, Am Orthinol Union Monogr, (1984), 32:1-97.

2. Ainley, D. G. Ballard, L. Balance, K. Dugger, N. Nur, G. Rau, C. Rubic, “Understanding Penguin Response To Climate and Ecosystem Change”, pp 1-13, Retrieved April 23, 2009, from

http://penguinscience.com/current_sum.php

3. Australian Antarctic Division, “Adélie Penguin Satellite Tracking”, Retrieved March 5, 2009, from http://www.aad.gov.au/default.asp?casid=2945.

4. Bardsley, R.J, “DigitalGlobe Satellite Imagery Plays Key Rolein Emperor Penguin Protection and Research”, Retrieved March 5, 2009 from http://www.redorbit.com/modules/news/tools.php?tool=print&id=1123042.

5. Boersma, P.D., “Penguins as Marine Sentinels”, Bio Science, July July/August 2008 / Vol. 58 No. 7 doi: 10.1641/B580707, Retrieved March 5, 2009, from

http://www.aibs.org/bioscience-press-releases/resources/Boersma.pdf

6. Clarke, J.R., “Partitioning of foraging effort in Adélie penguins provisioning chicks at Bechervaise Island, Antarctica”, Polar Biology, (Spring 2001), 24:16-20.

7. Davis, L.S., Harcourt, R.G., C. J. Bradshaw, “The Winter Migration of Adélie Penguins Breeding in the Ross Sea Sector of Antarctica”, Polar Biology, (2001), 24:593-597.

8. Emerson, L., Southwell, C., “Sea Ice Cover and Its Influence on Adélie Penguin Reproductive Performance”, Ecology, (2008), 89(8): 2096–2102

9. Finn, L., Leah, “UW Researchers to Increase Penguin Tracking”, the Daily of the University of Washington, August 15, 2007.

10. Pennycook, J., Educational Liaison for D, G. Ainley, Personal Interview, February, 2009.

11. Shannon, S., “Researchers Track South American Penguins, Well Dressed and Well Traveled”, Retrieved April 22, 2009 from http://faculty.washington.edu/boersma/updates/penquinsinthelamp.pdf.

12. University of Washington (2007, August 10), “Satellite Tracking Will Help Answer Questions About Penguin Travels”, ScienceDaily, Retrieved March 5, 2009, from http://www.ScienceDaily.com/releases/2007/08/070806123601.htm

13. Yoda, K, and Y. Ropert-Coudert, “Temporal Changes in Activity Budgets of Chick Rearing Adélie Penguins”, retrieved April 22, 2009 from http://www.springerlink.com/content/41747120123h1847/.

Exhibit Sources and Photo Credits:

http://astro.uchicago.edu/cara/vtour/mcmurdo/ for un-referenced imagery

http://penguinscience.com/ Viola Toniolo, via e-mail use request

Jean Pennycook, personal use authorization

Links:

Education perspectives on ocean science videos

http://www.ucsd.tv/oceanscience/

National Science Foundation - Antarctica logistics and research station videos

http://www.nsf.gov/news/overviews/arcticantarctic/interactive.jsp

Penguin Survival in a Changing World (Dr. Kooyman, with Dr. Ainley - 55 minutes)

http://www.ucsd.tv/search-details.asp?showID=13814

Research site for Dr. David Ainley.

http://penguinscience.com/

Just For fun: An Australian video on physical facts of the universe.

http://dingo.care-mail.com/cards/flash/5409/galaxy.swf