Citizen Science For The Birds
If you were to venture a guess, how many birds do you think there are in the world? Are there more birds than humans, or fewer? How do you think this number has changed over time?
These imponderable questions turn out to be very important to scientists and conservationists. They need good estimates of the numbers and geographic distributions of animals in order to prioritize conservation efforts, study species, and track the wellbeing of species over time. As you might imagine, though, reliable estimates are extremely difficult to generate. Animals move and migrate. They can be difficult to spot and count in the wild. And there are not enough trained scientists to conduct detailed censuses in every remote location.
Fortunately, scientists are not alone. In parks, wilderness areas, and neighborhoods around the world, amateur bird watchers are providing the critical missing piece to the puzzle. By recording the precise times, locations, numbers, and species of birds spotted and submitting the information to eBird, they become the eyes and ears of researchers, extending their reach beyond what would be possible for the limited number of scientists studying bird populations.
The data collected by citizen scientists and governments is indispensable, but it is also non-standardized, and can contain inaccuracies due to lack of professional training. In the past, this has made it difficult for researchers to glean useful statistical information from the wealth of bird watcher data available.
Inspiringly, a group of researchers has refused to be deterred by what may seem like insurmountable obstacles. For a study from the National Academy of Sciences, researchers combined the best available scientific data with the huge amounts of data collected by hobby birdwatchers to generate what they hope will be the best estimate to date of global bird abundance. With the help of these citizen scientists, they’ve developed a process that can provide a snapshot of current bird abundance, track changes in bird populations over time, and even (with a few tweaks) be applied to other animal groups.
The researchers generated their Global Species Abundance Distribution estimates using a five-step process:
Step 1 — The researchers built a model to estimate the relationship between relative abundance (the number of individual birds spotted by citizen scientists) and the actual density of a species in a region. They used a few hundred bird species for whom they had excellent scientific estimates of density in at least some regions (training species), and modeled the relationship between these reliable estimates and the data gleaned from amateur birdwatchers in those same regions.
Step 2 — The researchers divided the globe into 5° squares. In each square, they used birdwatcher data from eBird and the model from step 1 to estimate each training species’ density in that square.
Step 3 — The researchers refined the model by incorporating life history traits of different bird species. These are physical or behavioral traits like body size, color, and flock size that might make some species easier for birders to spot than other species. By incorporating these factors in the model, researchers could more accurately relate birdwatcher sightings with actual species density.
Step 4 — After completing steps 1-3 to train a model to generate population density estimates for the training species, researchers used the model to generate population density estimates for non-training species, for which good data is not currently available.
Step 5 — After estimating the population density for a given species in each square where it is found, researchers calculated the global abundance estimate for each species by multiplying the density estimate by the area in each square and adding up the totals of all squares where a species is found.
The resulting estimates tell us that there are about 50 billion birds alive in the world today, or about 6 birds for every human. Many species of birds have small populations, with about 12% of bird species having populations of less than 5,000. Most individual birds live in the palearctic (Europe, Asia, & North African) and nearctic (North America) realms, while the Madagascar and Antarctic realms are home to the fewest individual birds.
Aside from providing fun facts for your gee whizz collection, this new technique for modeling bird populations will enable scientists to track population changes over time, giving them a better idea of how events, human-caused and otherwise, impact animal populations and which populations need the most urgent attention from advocates, conservationists, lobbyists and legislators.
https://www.pnas.org/content/118/21/e2023170118