As more birders join eBird and learn to enter their checklists from specific locations and in the best possible way, the database becomes more and more useful to science. Every eBird record is made freely available to scientific, conservation, and other non-profit uses via our 'Explore Data' page and the 'Data Download' tool. Each checklist you submit to eBird provides scientists with an increasingly valuable resource for answering questions about the distribution and abundance of birds.
Researchers in the fields of ecology and conservation frequently conduct studies aimed at answering two questions: Where does a given species live? and How abundant is it? Knowing where species live, what habitats they use, and how abundant they are is the most basic information needed to protect a species. Knowing whether these patterns are changing with time is perhaps even more critical, since changes in bird occurrence can often be one of the first signal of more widespread environmental changes.
eBird data is contributed by a community of tens of thousands of users and our philosophy is that the data should be freely available. All eBird data are available via our Data Download tool on the 'Explore Data' pages. This is the eBird Basic Dataset (EBD) and it is updated quarterly. In addition, yearly versions of the eBird dataset are generated and associated with a suite of climatic and landscape variables and provided free to scientists. This is the eBird Reference Dataset and it is accessible via the Avian Knowledge Network page.
Fig. 1. Orchard Oriole migration as modeled by the Spatio-Temporal Exploratory Model. The model predicts frequency of occurrence for each day and the animation shows the flow of migrant orioles in and out of the country. Larger map.
Documenting Bird Distribution
Your checklists document a species' presence (and absence, when you report all species) at a specific time and geographic location. Plotting all of the observations of that species over days, weeks, months, or years allows us to determine the species' range, including movement patterns and changes in distribution.
Each species' seasonal movements become evident when comparing its distribution from various times throughout the year. We can discover the precise migratory paths followed by Neotropical migrants, we can look for patterns in the post-breeding wanderings of Wood Storks, or we can watch the movements of nomadic species during the winter months. By comparing a bird's range among different years, scientists can keep track of changes in distribution. Where will Eurasian Collared-Doves colonize next? Where do Lawrence's Goldfinches winter each year? Are breeding and wintering ranges of different species expanding, shrinking, or staying the same? Do migratory paths change between years? Answering these questions will help us understand birds more completely and aid in their conservation.
From a scientist's perspective, knowing if a bird species is absent is equally important as knowing if it is present. For this reason, it is important that you report all of the birds you saw or heard whenever possible, so researchers analyzing your observations will know not only which species you saw, but also which species you didn't see. If birders only report the birds they went out looking for, the resulting distribution map ends up only showing where birders like to go birding and not where the birds are distributed. Accurate distribution maps need to reflect where birds are and where they are not.
Modeling Bird Distribution
eBird data is not uniform in distribution. There is much more eBird data from New York City than there is for rural North Dakota or from Somalia. This lack of uniformity means that in order to understand bird distribution it must be considered in a way that accounts for these biases in coverage. The Spatio-Temporal Exploratory Model (STEM) was developed to standardize time of day and effort and to account for the uneven distribution of the data. The result is that STEM results for several hundred species now provide modeled distributions for across the Lower 48 states. Although the landscape information used to date has only been available for the Lower 48 states, we hope to expand this modeling across the hemisphere.
You can read more about this modeling process and see several dozen examples on our occurrence map pages (see Fig. 1). It is important to understand that these models do not simply produce predictions of species occurrence at hundreds of thousands of points on the landscape, but are also able to predict the relative importance of a suite of landscape characters in defining the species' distribution. The results of these models are forming the basis for a number of additional research projects to investigate and understand bird distribution in new ways.
The full discussion of this methodology can be found here:
Fink, D., Hochachka, W. M., B. Zuckerberg, D. W. Winkler, B. Shaby, M.A. Munson, G.J. Hooker, M. Riedewald, D. Sheldon, and S. Kelling. 2010. Spatiotemporal Exploratory models for Broad-scale Survey Data. Ecological Applications 20(8): 2131-2147.
Completing the cycle - using the data for conservation
To reach its maximal level of utility, a bird observation will be submitted on a complete checklist that reports all species, will be shown on eBird data exploration tools, will be downloaded and used by science, and then that scientific analysis will feed back in to conservation efforts to protect the species. We refer to this as the full data cycle of an eBird record, and we are now seeing the cycle completed.
In 2011, eBird data were used for the State of the Birds Report. That report used STEM results of eBird data to provide models of summer and winter bird distribution. Those models were overlaid with maps of protected lands (e.g., Bureau of Land Management, U.S. Forest Service, National Park Service) in order to understand how well certain guilds of species are being protected by existing government land. This report has been able to guide additional research by the Bureau of Land Management, the U.S. Forest Service and others in order to inform better management and protection of birds on those lands.
eBird data is being used at much more local scales as well. In the Chicago area, the Bird Conservation Network eBird manages grasslands and uses eBird to monitor the success of management actions. Using eBird data submitted by volunteers, it is possible to see if land managers are able to create habitat that draws Henslow's Sparrows, Sedge Wrens, and other grassland birds to areas that did not previously have them.
In other areas, National Wildlife Refuge managers refer to eBird to guide their decisions on managing impoundment water levels. eBrid records submitted by birders help define the migration periods for shorebirds and ducks and, when submitted at the fine scale, reveal how successful refuges have been at creating habitat for target species in certain areas.
There are many other such examples, but the above should give an indication of what is possible.
Research at the Cornell Lab of Ornithology
There are major gaps in our understanding of birds’ migration strategies because it is often very difficult to follow individual birds as they travel within and between continents on their annual journeys. For larger species, it is possible to follow their travels in almost real time, by attaching tracking devices that communicate with researchers via satellite links. Recent technological developments now allow less precise tracking of birds as small as warblers using devices that collect information on the time of sunset and sunrise to estimate locations… as long as the recording devices can be recovered after the migratory circuit is complete. None of these technologies, however, allow for the tracking of a large number of individuals of any species, let alone following the migrations of a large and diverse set of species. This presents a significant limitation to the analysis of populations of migrating species, which is necessary for their broad-scale conservation and management.
A new study takes a novel approach to studying avian migration by using crowdsourced eBird data to quantify the day-to-day population-level movement of bird species during migration. The unique perspective provided by eBird has allowed researchers at the Cornell Lab of Ornithology to document for the first time population-level variation in migration speeds and routes:
La Sorte, F.A., Fink, D., Hochachka, W.M., DeLong, J.P., and Kelling, S. 2013. Population-level scaling of avian migration speed with body size and migration distance for powered fliers. Ecology, 94: 1839-1847. http://www.esajournals.org/doi/abs/10.1890/12-1768.1
Using eBird data, the authors confirm that established theories regarding the optimal speed of migration — theories that have only been tested with information from handfuls of individual birds and only from large species — are able to describe differences in migration speed among populations of broadly distributed species based on their body size. Additionally, the study’s findings suggests that the total distance birds must traverse to and from breeding and non-breeding grounds influences the speed at which they travel. More generally, this work has demonstrated the viability of using eBird to provide knowledge about the ecology and evolution of avian migration strategies, which will play an important role in the management and conservation of migratory species worldwide.
The above examples are just one way that eBird data are being used for science and conservation, but they are not the only way. An ever-growing list of publications use or discuss eBird data. For example, Hurlbert and Liang (Hurlbert, A.H., Liang, Z. 2012. Spatiotemporal Variation in Avian Migration Phenology: Citizen Science Reveals Effects of Climate Change. PLoS ONE 7(2): e31662. doi:10.1371/journal.pone.0031662) used 10 years of eBird data to demonstrate and explore shifts in migration dates due to temperature fluctuations. Read more here.
See our eBird publications list for many others.
Whether you submit actual counts for all species your counts are especially useful. Comparing how birds' numbers change over time is especially important in conservation biology and your counts will be instrumental in helping scientists determine whether populations are increasing, decreasing, or remaining the same.
We are currently pursuing grants to develop modeling techniques, based upon the STEM foundation, that will model abundance of birds. As those results are looked at from year to year, we believe it will be possible to better understand large-scale population change. Other long-running datasets, like the Christmas Bird Count and Breeding Bird Survey, have long provided insight on population trends, but some species are missed by those counts. Your eBird observations will help to fill those gaps.
Some fluctuations in populations are very drastic and obvious. Below is an example from the Great Backyard Bird Count, but the same pattern can easily be found by exploring the eBird Range maps.
Fig. 2. Redpoll counts in two years (2001 upper, 2002 lower). Note the increase in the number of redpolls reported along the Canadian - U.S. border in 2002 when compared with one year earlier. Using data from Project FeederWatch and the Great Backyard Bird Count, scientists have found a two-year cycle in Common Redpoll irruptions. Scientists will be able to use your information in the eBird database to uncover movement patterns in all species, and from time periods throughout the year.