What citizen scientist birdwatchers can tell us about hummingbird migration

This is a public research summary of: Supp, Sarah R., Frank A. La Sorte, Tina A. Cormier, Marisa C. W. Lim, Don R. Powers, Susan M. Wethington, and Catherine H. Graham. 2015. Citizen-science data provides new insight into annual and seasonal variation in migration patterns. Ecosphere 6: art15.  http://dx.doi.org/10.1890/ES14-00290.1 [Open Access]

Our recent paper in Ecosphere uses citizen science data from eBird to look at the movement patterns for five migratory  species of North American hummingbirds. Migratory animals need to use resources in different locations throughout the year, and depend on being able to find food, mates, or the correct weather conditions (temperature, rainfall) when they arrive at those locations. Because human-caused environmental change, including land conversion, pollution, and climate change, are happening at different rates across migratory paths, migrating animals are often considered to be at high risk for failing to adapt to these changes.

One way to begin to predict how well species might be able to adapt to environmental change is to understand how flexible their current migration routes already are. For example, if individuals are able to postpone or advance migration in different years to match weather patterns, we could potentially use this variation as a baseline for what response the species is capable of, given expected future shifts in the environment. So for our paper, instead of averaging migration timing and routes across different years, we wanted to know what the variation in migration routes might be across years.

This research was part of a large collaborative effort (including several related projects) to assess hummingbird response to environmental change. The co-authors in this study all brought their own expertise and perspectives, which made our collaboration a lot of fun! We focus on hummingbirds, a group often studied in the Graham lab to assess eco-physiological  and diversity response to change, because they are so cute! No really, they are! And people love them! But really, hummingbirds are a fascinating group to study and a great model organism when considering sensitivity to environmental change. Because they are so small, and have very high metabolic rates, they are particularly sensitive to extreme temperature (something we are concerned about in future climate change scenarios) and fluctuating flower nectar resources (to stay alive, they must eat often). Many hummingbird species do not migrate long distances, but there are five occuring here in North America that do (most of the >350 hummingbird species are in the tropics). Some of these species migrate from Mexico in the winter all the way to Canada for breeding – Amazing!

Bands to mark hummingbirds are extremely tiny.

Bands to mark hummingbirds are extremely tiny.

It would take a single scientist many years to collect data on hummingbird migration across such a large extent. Even a network of scientists would be limited in the number of birds they could actually capture and observe across sites spanning an entire migration pathway. New technology is now allowing scientists to recover extremely detailed information on the migration behavior for many species, but at only 2-5 grams, North American hummingbird species are too small to use satellite tracking technology. Although hummingbirds can be banded, it is an extremely delicate process, it requires much more training than to band other bird species, and recapture rates are very low. However, hummingbirds are commonly observed by a very large network of dedicated specialists – birdwatchers! eBird has an excellent platform for birdwatchers to publicly record and share their observations, and anyone, including scientists, can explore or download this treasure trove of data for FREE! We call this dedicated network of birdwatchers citizen scientists – because their skilled observations provide critical data needed for this kind of broad-scale study.


The maps on the left shows where each species was seen in by eBird birdwatchers (red, more observations; yellow, fewer observations). The maps on the right show the summarized migration routes for each species. There are six lines per map, one for each year 2008-2013. Blue colors indicate spring, orangey colors indicate autumn (modified from Fig 1. in the published paper).

So sometime last year, we downloaded all the eBird data for 2003-2013 on ruby-throated hummingbird, black-chinned hummingbird, calliope hummingbird, broad-tailed hummingbird, and rufous hummingbird. eBird observations really picked up around 2008, so we focused on results for the last six years in our time-series, 2008-2013. We were able to use the data from citizen birdwatchers to estimate when each species begins to move northward in the spring, when they have all arrived at their breeding grounds, when they have finished migrating southward in the fall, and whether these dates were different in different years. We also mapped out the average migration route for each species, where we detected more east-west variation than north-south variation – potentially indicating that individuals exhibit strong north-south movement (it’s important to get to the breeding ground at the optimal time), while being able to adjust the east-west component of their location to account for resources or weather patterns.  We did not assess the impact of environment on migration patterns for this paper, but that is something we are working on right now using remotely sensed data from satellites! One interesting observation was that long distance migrants (rufous hummingbird) had less variation in their migration route or timing than short distance migrants (broad-tailed hummingbird).

Using the eBird data, and data from citizen bird banders, we also noticed that many rufous and black-chinned hummingbirds are being observed during the winter in the southeastern United States. This is a pattern that many birdwatchers and scientists are very curious about, but there is no definite explanation right now. It is possible that the increased sightings of these birds outside of their typical migration pathways and wintering grounds could simply be because more birders are watching for them, reporting them, and keeping their feeders up year-round. It is also possible that land use change, such as increased urbanization has actually increased year-round blooming of nectar flowers and artificial nectar feeders for hummingbirds, thus encouraging them to stay. Finally, warmer winter temperatures could already be shifting the winter range northward for hummingbird species, to include southeastern United States.
This paper is only a first step in describing hummingbird migration in North America and we have ongoing work to understand the physiological and environmental drivers for variation in hummingbird migration, funded through Catherine Graham’s NASA grant, with co-PIs Scott Goetz, Don Powers, and Susan Wethington. I became involved with this research during my first postdoc in Catherine’s lab. Stay tuned to see what we’re working on next! If you’re curious to see what others have done using eBird data, check out their publication list here.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s