1. Canadian Wildlife Service, Edmonton, Canada
King Eider winter movements in the Bering Sea tracked by satellite telemetry
Steffen Oppel
Department of Biology and Wildlife, University of Alaska, Fairbanks, AK, USA
Lynne Dickson
Canadian Wildlife Service, Edmonton, Canada
Abby Powell
U.S. Geological Survey, Alaska Cooperative Fish and Wildlife Research Unit,
Fairbanks, AK, USA

2. King Eiders spend ca. 10 months per year at sea
Background
King Eiders spend ca. 10 months per year at sea
forage on benthic invertebrates by diving to sea floor
King Eiders are large sea-ducks that breed in arctic tundra ecosystems around the world, and in Alaska they breed on the North Slope. These ducks are called sea-ducks because they spend most of their life at sea. At sea, King Eiders forage on benthic organisms, that is mussels, clams, shells, crustaceans, urchins and other invertebrates that they obtain by diving to the sea floor. They can dive 50 m, but they do need open water to have access to their food source.King Eiders are migratory, but remain in subarctic or north temperate waters even throughout winter, and the birds breeding in western Canada and Alaska all fly to the Bering Sea in winter, and this map shows winter locations of 95 birds tracked from 2002 – 2006 just to give you an impression of the winter distribution.
As we have heard this morning, the environmental conditions in the Bering Sea are changing rapidly as a result of global warming, and this may have an effect on sea ducks by changing sea ice conditions and availability of food resources.
However, we know very little about the general ecology of sea ducks in winter, which limits our ability to predict how they may be affected by climatic changes in the future. King Eider populations have declined in the past 30 years, and these ducks are an important food source for many native Alaskan subsistence hunters in coastal areas, so their population status is of conservation concern.
In this project we therefore tried to provide some baseline data on the winter ecology of King Eiders, and we were most interested in the following questions:

3. birds from western North America winter in Bering Sea
Background
birds from western North America winter in Bering Sea
winter period from August through May
Alaska
Russia
Bering
Sea
King Eiders are large sea-ducks that breed in arctic tundra ecosystems around the world, and in Alaska they breed on the North Slope. These ducks are called sea-ducks because they spend most of their life at sea. At sea, King Eiders forage on benthic organisms, that is mussels, clams, shells, crustaceans, urchins and other invertebrates that they obtain by diving to the sea floor. They can dive 50 m, but they do need open water to have access to their food source.King Eiders are migratory, but remain in subarctic or north temperate waters even throughout winter, and the birds breeding in western Canada and Alaska all fly to the Bering Sea in winter, and this map shows winter locations of 95 birds tracked from 2002 – 2006 just to give you an impression of the winter distribution.
As we have heard this morning, the environmental conditions in the Bering Sea are changing rapidly as a result of global warming, and this may have an effect on sea ducks by changing sea ice conditions and availability of food resources.
However, we know very little about the general ecology of sea ducks in winter, which limits our ability to predict how they may be affected by climatic changes in the future. King Eider populations have declined in the past 30 years, and these ducks are an important food source for many native Alaskan subsistence hunters in coastal areas, so their population status is of conservation concern.
In this project we therefore tried to provide some baseline data on the winter ecology of King Eiders, and we were most interested in the following questions:

4. What movements occur during winter?
Questions
What movements occur during winter?
What factors are correlated with movements?
First, when do King Eiders arrive on their wintering areas, and how long do they stay there? This is of interest in order to understand whether a shifting climate could affect the birds’ migration schedule.
Second, we were interested in whether birds stayed in one wintering site throughout the entire winter, or whether they moved between different sites.
And lastly, and I guess I’m giving away the punchline of question 2 here, we tried to explore whether sea ice could be a motivating factor for the winter movements we did observe.

5. 94 birds fitted with satellite transmitter in June 2002-2005
Methods
94 birds fitted with satellite transmitter in June
winter movement defined as >50 km
calculated sea ice concentration for all movements (
We addressed all these question by marking a total of 94 birds with implantable satellite transmitters between 2002 and 2005, and we captured 80 birds on their breeding grounds in Alaska and 14 in Canada.
(NOTE: 111 birds were marked, but only 94 were used in analysis here).
We programmed the transmitters to receive location information every 4-7 days and this resulted in transmitters lasting approximately months, so that for most birds we were able to receive information on a full annual migration cycle. We filtered all the location data to remove inaccurate locations and calculated the travel distances between subsequent locations using Dave Douglas ARGOS filter program.
We defined the beginning and end of every season for each individual separately based on its movement rate and trajectory. This means that the wintering period does not start for all individuals at some given calender date, but rather at the precise time when that individual completes its migration to the wintering grounds. Likewise, winter ends for an individual when it starts for spring migration, so the winter period I will be talking about today is the time period between two migration events for each individual.
During that time we defined a wintering site as all the locations that were within 25km of each other, and a winter movement was defined as a movement between two different wintering sites, so generally more than 50 km in length.
And lastly we used online sea ice data from the National Ice Center to calculate the sea ice concentration for the wintering sites from which King Eiders departed in winter, and we calculated the difference in sea ice concentration at those locations over the time period during which a movement occurred.

6. Results: winter movements
59.1% of tracked birds used >1 wintering site
some birds use up to 4 wintering sites
During this time we were surprised to see very low within-year site fidelity, and almost 60% of the birds we tracked used more than 1 wintering site.
On average birds used at least two wintering sites, and some birds used up to 4 different wintering sites, and the number of movements was on average greater than the number of wintering sites, with up to nine winter movements for some individuals.

7. Results: winter movements
Mean winter movement range 12,000 km2 (±18,000 km2)
no movement between three wintering regions
Alaska
Russia
In order to visualize these movements we calculated movement ranges as 95% minimum convex polygons of all winter locations of each individual, and this map shows the movement ranges of 94 individual King Eiders.
These movement ranges are analogous to home ranges, but it is important to understand that these only show the extent of movements, and do NOT imply that a bird uses the entire area that is covered by the polygon.
The movement ranges had enormous variation, ranging from just under 50 km2 like the one bird sitting around St Matthew island to almost 80,000km2 (which is the size of the entire Bristol Bay area), with a mean of about 12,000 km2
Despite some extensive movements, there were 3 discrete wintering regions in the Bering Sea between which birds did not move from late December through early April. These regions are shown here as red ovals.
Bering Sea

8. Results: correlates of winter movements
only 16 % of movements synchronized between individuals
sea ice concentration constant for most movements
number of movements
difference in sea ice concentration (%)
So our last question was to find some explanation for these movements. If deteriorating weather or sea ice conditions would force birds to move away from a site then we would expect that all birds at a given site would leave roughly around the same time – but this was not the case. In only 16% of the movements for which we had information from more than one bird at the same site did birds leave a site simultaneously, in the other 84% at least some birds stayed when others left.
Another hypothesis was that sea ice could form at certain wintering sites, preventing birds from accessing their food source when there is no open water anymore. So we expected birds to move away from an area during a time when sea ice concentration at that site increased. But this was not the case either. This histogram shows the number of winter movements by King Eiders over the range of changes in sea ice concentration at the departure site, and you can see that for the vast majority of movements the sea ice conditions at the departure site did not change at all during the time the movement occurred. The positive numbers indicate that sea ice concentration increased, and negative numbers indicate a decrease. There are slightly more movements with increasing sea ice cover, but this does not appear to be a biologically significant difference.

9. sea ice does not appear to cause movements
Discussion
wintering more dynamic than in other sea ducks (Petersen et al. 2002, Merkel and Mosbech 2007, Iverson and Esler 2006)
sea ice does not appear to cause movements
movements may be exploratory
So, in summary, wintering in King Eiders appears to be way more dynamic than was previously known. The movements of King Eiders are generally an order of magnitude larger than what has been documented for other sea-ducks like Harlequin Ducks by Iverson and Esler or even closely related Eider species both in Greenland (Merkel and Mosbech) or in the Bering Sea (Spectacled Eider: Lovvorn, Petersen).
As of yet we have not found a convincing explanation for the variability in winter movements, and sea ice alone does not seem to be the most important causal agent triggering those movements. The movements could be exploratory in nature, which means that King Eiders check out a number of different sites so they know where to go when at some point in the future the site where they are suddenly becomes unsuitable. Such exploratory movements are known for other bird species that live in unpredictable and fluctuating environments, and are a distinct possibility for King Eiders.

10. Management implications
potential to adapt to changing environment
numbers using an area may be substantially higher than number counted at any given time
Regardless of whether King Eiders actually use these movements to increase their spatial knowledge or not, the high variability among individuals alone is good when we consider the challenges sea ducks may face with a rapidly changing environment in sub-arctic seas. King Eiders do not seem to be fixed to winter in a certain area, they have the flexibility to move around, so if certain areas become inhospitable due to climatic changes and changes in sea ice distribution, it should be possible that the population can respond to these changes by utilizing different areas.
The high flexibility however creates a problem when assessing the importance of any given area for King Eiders. This is particularly important for areas like Bristol Bay, which may see some industrial development in the future, and where impact assessments are generally based on aerial surveys or other censuses at discrete points in time. With those techniques we will only count a number present at a given time, and from what I’ve shown you today we have very good reason to believe that the ducks you count on one day may have been replaced by different individuals on another day, so that the total number of King Eiders using an area throughout the winter may be a lot higher than the number you see in a given area at any given time. So by simply using census data we will most certainly underestimate the importance of areas and this is something that should be kept in mind for impact assessments.

11. Minerals Management Service Coastal Marine Institute
Acknowledgements
Minerals Management Service
Coastal Marine Institute
Canadian Wildlife Service
US Fish and Wildlife Service
US Geological Survey
Sea Duck Joint Venture
North Slope Borough
Conoco Phillips, AK
USGS
ABR, Inc.
Service Argos, Inc.
Microwave Telemetry, Inc.
German Academic Exchange Service (DAAD)
Troy Ecological Research Associates, Inc.
Alaska Cooperative Fish and Wildlife Research Unit
Robert Suydam
Dave Douglas
Dave Verbyla
Ed Murphy
Rebecca Bentzen
Andrea Hoover
Cheryl Scott
…and a large number of
field assistants…
With this we would like to thank the whole world, especially …(pick 2-3 from the audience)…