Purple Martin animated map and graph

The animated maps and graphs on this page are based upon observational data of two species of swallows (Purple Martin and Tree Swallow) collected via eBird. The two maps provide a fascinating comparison to one another, in that they reveal quite different phenologies of migration between the two species, but also highlight the similarities in their migration strategies. 

 For both species, the maps on the left indicate average counts location by location. The white dots indicate low counts with progressively higher counts being represented by the warmer colors, red dots representing the very highest counts. Both species of swallows move north in small groups during March and April and spend May and June on their breeding grounds. In spring, notice how the dots tend to be white and yellow, indicating rather low counts. After breeding, both species aggregate in large roosts prior to their southward migration. Both species tend to aggregate in these large staging areas to take advantage of late summer insect hatches in coastal saltmarshes, inland lakes, and other wetlands. (To see staging Purple Martins, visit Presqu’isle State Park, PA, or the Nanticoke River marshes of Maryland in July or August; to see Tree Swallow swarms try Plum Island N.W.R., MA, in August and September, or Cape May in October. Tree Swallow counts can easily top 50,000 at either location!) On the map, these staging areas appear as red dots on the maps, since the red dots indicate higher counts from a single location. Notice how the Purple Martins stage up to a month earlier than the Tree Swallows and are largely gone from the United States by early October. Tree Swallows, by contrast, stage from August to October and large numbers winter in the Southeastern United States.

Mass of staging Tree Swallows at Sachuest Pt., RI, in September 2007

The animated graph at the right shows the same characteristics visible on the map, but in a different visualization. Two different metrics are represented on the graph: number of locations reporting the species and average count per location. Notice how the number of locations reporting both species is high in spring and summer as the swallows are spread over the landscape as northbound migrants and as widely dispersed breeders. As the young fledge and the adults no longer need to feed them, both adults and immature begin to gather with others of their species at traditional staging sites. As they do this they disappear from their breeding grounds and thus the number of locations begin to decline while the average count begins to increase. By late summer, almost all swallows are concentrated in these staging grounds, so average counts reach their peak while average number of locations reach their nadir. Notice how the Purple Martins withdraw from the country a month before Tree Swallow numbers peak. But also notice how early some Purple Martins may return—as early as late January or early February in Florida and Texas. Sometimes overshoots may occur in Indiana or Maryland in February!

The excess of food in the staging areas allow these birds to put on fat for their migration. Purple Martins eat insects exclusively, so they depart well before the cold weather, and make a long migration (some crossing the Gulf of Mexico) to Brazil. Tree Swallows can also survive on berries (especially Bayberries (Myrica sp.)), and have a shorter, less risky migration. Many winter in Florida, and some ambitious birds may spend the winter as far north as New Jersey. The different biological requirements of these birds thus create these different migration phenologies.

Staging Purple Martins at Presqu'isle S.P., PA, in August 2006

One other interesting thing to watch on this graph, is the pattern of Purple Martins west of the Great Plains. Western Purple Martins comprise two different subspecies, Progne subis deserticola and Progne subis arboricola, neither of which is as well-represented in eBird data as is their eastern cousin, Progne subis subis. The wintering grounds of these subspecies are unknown and could be totally separate from nominate subis. Both of these subspecies do not use artificial cavities such as martin houses, but rather nest in cavities in large cacti (deserticola) or old growth trees (arboricola). Both have declined in recent years. We do not yet know much about where these birds stage for their migration, though anecdotal evidence suggests that the Baja California Peninsula (and possibly other areas in West Mexico) may be important staging grounds for these birds.

This is just one example of how AKN data can be very powerful in showing visualizations of bird movements and also how it can be used to highlight gaps in our knowledge. With more reporting from America’s west and from Mexico in late summer and early fall, perhaps in a few years we will have some answers to the questions above. With more reporting from Central and South America, perhaps we can shed some light on the wintering grounds of the Purple Martins that breed in the western United States.