Many Arctic birds are renowned for their magnificent migratory achievements not only in the sense that they carry out the longest journeys among birds, but also because of their efficient travel and orientation capacity under complex polar conditions. We used a tracking radar on board the Canadian Coast Guard icebreaker Louis S. St-Laurent to record bird migration at the various sites visited on the Tundra Northwest 1999 expedition, in order to address the following issues about evolution and mechanisms in polar bird migration:

1. Routes

The shortest route between two Points on the Earth’s surface is along the great circle (orthodrome) and not along the path of constant compass course (loxodrome; loxodromes and orthodromes coincide only for the special cases of routes with geographic courses due North or South along the longitudes). The greatest differences between orthodromes and loxodromes occur at polar latitudes. On the basis of the radar studies we hope to answer the question of whether the routes of the migrating birds show adaptations to the spherical geometry of the Earth.

2. Orientation

We have suggested, on the basis of our earlier tracking radar studies in Siberia, that many tundra birds use a time compensated sun compass to fly along orthodrome-like routes. This seems to be a more likely orientation principle than alternative ways of orienting by, for example, a magnetic compass system (Alerstam & Gudmundsson 1999a). We aim to evaluate this hypothesis in Arctic Canada, a region with uniquely distinctive geomagnetic field features (widely variable magnetic declination and very steep inclination angles) close to the magnetic North Pole, which is situated within the study area and which was visited during the expedition. Predicted flight routes assuming sun or magnetic compass orientation in this region show considerable differences, permitting a critical testing of different orientation principles.

3. Process

The migrants’ responses to weather and wind, as well as their flight altitudes and speeds will be investigated in order to evaluate flight strategies and long distance migratory performance of Arctic birds.

4. Patterns

Mapping the geographical patterns of bird migration in the study area, where there have been no previous radar studies (except in the Beaufort Sea region; cf. Johnson & Herter 1990), may reveal migratory divides and contribute to clarifying the differentiation between populations from the Pacific and Atlantic migration systems. This will be important for understanding the evolutionary process behind the migration patterns of Arctic birds.

5. Comparison

We wish to compare the results from Arctic Canada with the results obtained during extensive radar studies of bird migration along the North-east Passage and the central Arctic Ocean on earlier expeditions (Alerstam & Gudmundsson 1999b, Gudmundsson & Alerstam 1998). Hence, we will attempt to put our findings in a circumpolar perspective.

Methods

We placed a tracking radar above the helicopter deck (altitude of radar antenna about 21m above sea level) near the stern of the expedition ship. The radar was used to track birds (flocks or single individuals), usually during a time interval of between 1 and 6 minutes at ranges between 2 and 20 km. The radar recorded distance, azimuth-and elevation angles to the targets every 2 seconds, from which altitude, track direction and ground speed are calculated for each target. Measurements of wind directions and speeds at different altitudes were obtained by radar tracking of helium-filled balloons with attached aluminium foil reflectors. Heading, speed and levelling of the ship, as well as its geographical position, were recorded simultaneously with the radar tracking, and the radar data were corrected with respect to these factors. The technical data for the radar have been presented in Gudmundsson & Alerstam (1998) and analysis procedures are essentially the same as described by Alerstam & Gudmundsson (1999b).

Some results

Radar operation took place at 24 sites, A–Z, as shown in Figure 1, giving as a result a total of 768 tracks of migrating birds. The ship was usually situated 2–8 km from the nearest shore, except at site M where this distance was 27 km (due to very shallow waters closer to the shore) and at site N that was 100 km north of the shore in the southern Beaufort Sea. Bird movements during the study period in July and August consisted mainly of post-breeding migration of waders (several species), skuas (especially long-tailed skua Stercorarius longicaudus) and Arctic terns Sterna paradisaea.

 Intensities

At some sites the number of bird echoes were few (or none) due to inclement weather, such as at sites B (low pressure situation with extensive rain) and P (gale winds), or because of unsuitable conditions for radar operation such as at site V (steep mountains at close range). However, at most sites the number of radar echoes from birds tracked per hour of radar operation may be used as a very rough measure of bird movement intensity. There was important geographic variation in intensity with the highest traffic rates occurring in the westernmost parts of the study area, in the South-east Beaufort Sea. In comparison, bird movements over the barren tundra regions at or north of 75°N were, on many occasions, exceedingly sparse in spite of suitable weather and radar conditions. Migration in the Baffin region never reached intensities comparable to the mass movements recorded in the Beaufort Sea region (and to an even greater degree along the Siberian Arctic coast).

Altitudes

The overall preliminary height distribution shows that the birds travelled at widely different altitudes up to a maximum of 4900m. While this altitudinal range is the same as recorded for the migration of tundra birds in Siberia, the altitudes from Canada were on average slightly lower than those from Siberia. Hence, the median altitude in Arctic Canada was about 700 m versus 1000m in Siberia and Russia, with 3% of all migrants recorded above 3km altitude in Canada versus 9% above this level in Siberia (cf. Alerstam & Gudmundsson 1999b, Alerstam & Jönsson 1999).

Directions

Migration patterns differed in the most interesting and important ways between different regions of the study area. At site H the distribution was bimodal with one peak of directions towards south-east (probably mainly waders) and another peak towards WSW-NW (possibly including, among others, long-tailed skuas). At the most westerly sites M-P mass migration was recorded with the vast majority of the birds travelling in the ENE-SE sector, even 100km off the coast at high altitudes where some of the migrants may have originated from Siberia (cf. Alerstam & Gudmundsson 1999a). Waders were dominant, including species such as the American golden plover Pluvialis dominica, the semipalmated sandpiper Calidris pusilla, the white-rumped sandpiper Calidris fuscicollis and the pectoral sandpiper Calidris melanotos. Future analyses will investigate the possible link between these easterly mass movements and south-eastward departures of waders over the Atlantic Ocean from eastern Canada (e. g. Richardson 1979), and consider different possibilities for continental flight routes and orientation principles. The wide scatter of migratory directions, with very few tracks having southerly courses but many moving towards the north, on Banks and Melville Islands (sites R and S) is surprising. Possible participants in these flights are waders, especially grey phalaropes Phalaropus fulicarius, and perhaps also long-tailed skuas and Arctic terns, exploiting post breeding resources in newly ice-free waters in the eastern and north-eastern Beaufort Sea region. Finally, a well-defined peak is shown in the sector between south-east and SSW for migration in south-eastern Baffin Island (sites Y and Z), involving waders as well as passerines.

Prospects

The radar results, together with the supplementary weather data and field observations that we have now obtained on the successful expedition to Arctic Canada, open up highly promising possibilities for answering many questions about flight routes, orientation, process and pattern in Arctic bird migration.