A deeper look into a groundbreaking study and how its most significant finding might be its least reliable.
When Brian McGill, a macroecologist at the University of Maine, read a groundbreaking study in Science about the deaths of 3 billion North American birds, a small piece of text caught his eye (Schulson, 2019). Amidst the trappings of a conventional scientific article, the authors had included a hashtag: #BringBirdsBack. The Bird Conservancy, of which one of the authors is president, quickly registered the domain 3billionbirds.org, with the hashtag used as the website’s title. On the same day as the study’s release, the study was accompanied by an op-ed in the New York Times, written by one of the study’s authors, that proclaimed “The Crisis for Birds Is a Crisis for Us All” (Fitzpatrick & Marra, 2019). In the days following the article’s publishing, countless news stories ran headlines declaring “Global warming could make 389 species of North American bird EXTINCT” (Pero, 2019) and “An avian apocalypse has arrived in North America” (Woodward, 2020).
Lead author Ken Rosenberg, from the Cornell Lab of Ornithology, admitted that he and his team had “professional communicators right there with the professional scientists, and… a team of people who understand media and communication” (Schulson, 2019). Combined with the media team at the journal Science, Cornell released a media package, a flashy downloadable document, and Instagram-ready infographics. It was clear that the authors were trying to draw attention to their findings, and it was working.
The authors, led by Rosenberg, estimated that nearly 1 in 3 North American birds had disappeared since 1970. With the exception of wetlands, every type of avian habitat experienced bird declines exceeding -15%. Speaking to the magazine Undark about their findings, Rosenberg said “It was kind of a shocking result” (Schulson, 2019). But these dramatic numbers, coupled with the encouraged media attention, gave Brian McGill some pause. Writing in the academic blog Dynamic Ecology, McGill raised some questions over whether there was actually an “avian apocalypse” across North America (McGill, 2019). For example, of the 3 billion birds estimated to have died since 1970, 15% of them were from 2 invasive species: the European Starling and the House Sparrow (Schulson, 2019). More surprisingly, these birds have been subject to extermination efforts by both local and federal governments across the continent (NYIS Information, 2019). This means that nearly 450 million of the 3 billion bird deaths are made up of just 2 species that have been intentionally killed. When combined with other invasive bird declines, that number is likely to be much higher.
McGill further explored the authors’ original data and found that a handful of species declined drastically, a handful increased significantly, and most only changed a little (McGill, 2019). He specifically looked at each of the top 10 “biggest losers,” or the top 10 species that have declined the most. Of the top 10, 8 are labeled as “Least Concern” (farthest from extinction) and 2 are labeled as “Non-Threatened” (second farthest from extinction) on IUCN’s endangered species list. Coincidentally, these 10 “biggest losers” are some of the most abundant bird species across North America. To drive this point home, McGill notes that of the 40 most abundant bird species, 31 are showing declines and make up 2.1 billion of the original 3 billion figure. In other words, and perhaps counterintuitively, bird species with the most individuals are experiencing the worst declines. Rare birds, on the other hand, are generally experiencing modest declines, staying the same, or drastically increasing.
McGill further looked into the actual methodology for determining the 3 billion dead birds figure, and highlighted the uncertainty in the authors’ calculations (McGill, 2019). The value itself combines many estimates of avian abundance across North America, with much of the data coming solely from the United States. While McGill admits that “[he doesn’t] have a better method…this is clearly a step that introduces a lot of uncertainty” (Schulson, 2019). These estimates paint a broad picture of bird population trends in North America, but they inevitably hide a more nuanced story beneath. The nuances of the research were not ignored or purposefully hidden by the original study’s authors. The authors themselves point out that “[while] not optimized for species-level analysis, our model indicates that 19 widespread and abundant landbirds (including two introduced species) each experienced population reductions of >50 million birds” (Rosenberg et al., 2019). So, generally speaking, bird populations are decreasing across North America, but that hides a significant amount of variation at the species level.
To McGill’s and Rosenberg’s credit, both admit that they don’t have a way of understanding the full picture, either from the study itself or from the critiques of the study. The original authors were likely constrained by the publishing format of a top journal like Science. Rosenberg, in his Undark interview, admitted that they “had to cut out 90 percent of what [they] wanted to say” (Schulson, 2019). With limited space in a competitive journal, the authors had to highlight their most dramatic findings which both Science and the media would choose to emphasize.
Overall, Rosenberg’s study does point to some serious declines in many bird species, but to call it an “apocalypse” might be overblown. Many of the species with millions of individuals once had far fewer birds, but agricultural development led to an increase in favorable habitats like prairies and grasslands. While only the roughest of estimates exist for pre-agricultural bird populations, scientists are pretty certain that the most numerous species are returning to a somewhat more stable level (McGill, 2019).
This reality should not detract from native, endangered species who are significantly decreasing, but the study’s central “3 billion dead birds” figure and its encouraged media attention gloss over a much more nuanced story. At the same time, a more nuanced presentation of their findings would not have sounded as drastic nor would they have landed in Science or the public’s eye. Rosenberg and his team were even conflicted about which words to use, whether that be “crisis,” “collapse,” or something more drastic (they ultimately call it a crisis and refer to a “potential avian collapse”) (Rosenberg et al., 2019).
This case raises questions about how scientists should communicate complicated topics that do require attention, but may not be as simple as a single number. It is apparent that researchers need to weigh being perfectly nuanced in their findings versus calling attention to a subject with oversimplified conclusions. If McGill’s concerns – or even this article – were to garner more public attention, how would that hurt the credibility of the original authors? Or hurt the odds of actually helping the birds that need it the most? How much responsibility rests on scientists to promote their findings? Should they just complete their research and leave it at that?
There’s no easy answer to any of these questions. For Rosenberg’s original study, this much is true: 3 billion birds have died since 1970, but 2.1 billion of those deaths come from bird species that have millions of individuals. These declines are undoubtedly human-caused, both directly and indirectly. When we reduce avian habitats, increase the world’s temperatures through climate change, expose birds to pesticides, introduce invasive species, and undertake countless other detrimental activities, birds are going to decline. Policymakers and conservationists need to take a closer look at the birds that are rare and declining, even if they are not declining as drastically as more numerous species. It is these rare and declining species that should receive most of our attention when it comes to conservation and management strategies. While North America might not be losing all of its birds, we must realize that such a story cannot be limited to a single number nor be presented as a broad generalization. Only with a more nuanced understanding of the situation can we start to protect the birds that need our help the most.
References
European Starling. (2019, May 31). New York Invasive Species Information.
http://nyis.info/invasive_species/european-starling
Fitzpatrick, J.W., & Marra, P. P. (2019, September 19). The Crisis for Birds Is a Crisis for Us All.
The New York Times. https://www.nytimes.com/2019/09/19/opinion/crisis-birds-north-
america.html
McGill, B. (2019, September 20). Did North America really lose 3 billion birds? What does it
mean? Dynamic Ecology. https://dynamicecology.wordpress.com/2019/09/20/did-north-
america-really-lose-3-billion-birds-what-does-it-mean/
Pero, J. (2019, October 10). Global warming could make 389 species of North American bird
EXTINCT including robins and grouse if temperature rises 3 degrees C, new study finds.
DailyMail. https://www.dailymail.co.uk/sciencetech/article-7559433/Study-Rising-
global-temperatures-puts-HUNDREDS-North-American-bird-species-extinction-
risk.html
Rosenberg, K., Dokter, A., Blancher, P., Sauer, J., Smith, A., Stanton, J., Panjabi, A., Helft, L.,
Parr, M., & Marra, P. (2019). Decline of the North American avifauna. Science, 366(6461), 120-124. https://www.science.org/doi/10.1126/science.aaw1313
Schulson, D. (2019, September 25). In New Study on Bird Loss, Some Critics Say Subtlety Is
Lost, Too. Undark. https://undark.org/2019/09/25/in-a-new-study-on-bird-loss-
some-scientists-say-subtlety-is-lost-too/
Woodward, A. (2020, January 16). An avian apocalypse has arrived in North America. Birdsong
could become a rare sound. Insider. https://www.businessinsider.com/birds-disappearing-
In-north-america-2020-1
