The research, conducted by scientists at the Cornell Lab of Ornithology and United States Geological Survey (USGS), compared population trend estimates from the North American Breeding Bird Survey (BBS), an annual structured survey led by USGS and the Canadian Wildlife Service, with those from eBird, a global participatory science project led by the Cornell Lab of Ornithology. The study analyzed data from 372 bird species across 32 Bird Conservation Regions<\/a> between 2012 and 2022.<\/p>\n\n\n The research found that when both surveys detected significant population trends, they agreed on the direction of change 92% of the time. Only 1.3% of all species and region combinations showed significant trends moving in opposite directions between the two datasets.<\/p>\n\n\n\n Some of the differences in population trend estimates, according to the study, are to be expected because the methodologies differ. The BBS is conducted by trained observers along roadsides at specific locations during daylight hours while eBird data is collected by participatory scientists at numerous locations any time of the day or night. These approaches complement each other, allowing for combined datasets that sample a wide range of habitats, locations, and temporal periods. <\/p>\n\n\n\n \u201cWe\u2019ve done a handful of studies now looking at how eBird products align with other surveys and what we see is that estimates are often species or location dependent,\u201d said Orin Robinson, lead author and senior research associate at the Cornell Lab of Ornithology. \u201cWe looked through the data sets for rules of thumb, but they really don’t exist\u2014a lot of the estimates are going to be species and location dependent.\u201d <\/p>\n\n\n\n The team found no systematic differences between the surveys that would suggest one dataset always estimates a higher trend value than the other. Instead the team found that population increases or decreases were challenging to determine in more than 80% of the trends estimated by BBS and 33% of trends estimated by eBird because the range of estimated trends contained zero. <\/p>\n\n\n\n \u201cThere are reasons why uncertainty differences exist between data sets, but just because one seems to be more uncertain doesn’t necessarily mean it is an inferior estimate\u2014you’d rather be more accurate with a little more uncertainty on either side than be precisely incorrect,\u201d said Robinson. <\/p>\n\n\n\n Robinson adds an important reminder that both surveys provide estimates of population trends while the exact reflection of nature is unknown. \u201cIt comes down to how much uncertainty researchers and managers are willing to live with,\u201d said Robinson. For Robinson, it\u2019s not an either or situation when it comes to using the data. \u201cBoth surveys are valuable and will continue to provide valuable information,\u201d said Robinson. <\/p>\n\n\n\n Robinson recommends that scientists focus on the credibility metrics provided by both surveys when assessing data sets. \u201cIf credibility metrics are low, there\u2019s going to be a lot more uncertainty and researchers should pay attention to that,\u201d said Robinson. He also recommends that scientists looking to integrate both data sources pay attention to species. For example, integrating both data sets for species such as owls may not produce good results but for species like woodpeckers that are adequately sampled by both BBS and eBird, integrating the data sets will likely reduce uncertainty and provide better estimates of the population trend than either alone. <\/p>\n\n\n\n The team hopes the results of this study will help increase confidence in results and help future researchers integrate both eBird and BBS data where appropriate to improve our understanding of bird population trends. <\/p>\n\n\n\n ####<\/p>\n\n\n\n Robinson, O. J., A. J. Johnston, W. M. Hochachka, J. A. Hostetler, J. R. Sauer, T. Auer, M. E. Strimas-Mackey, S. Ligocki, N. A. Faraco-Hadlock, V. Ruiz-Gutierrez, A. D. Rodewald, and D. Fink (2025). Same view through a different lens: comparing population trends for North American birds using eBird and the Breeding Bird Survey. Ornithological Applications 00, 1\u201314: https:\/\/doi.org\/10.1093\/ornithapp\/duaf077<\/a><\/p>\n\n\n\n About the Cornell Lab of Ornithology<\/strong>More Cornell Lab News<\/h2>
![\"\"](\"https:\/\/www.birds.cornell.edu\/home\/wp-content\/uploads\/2025\/02\/YellowBelliedSapsucker_StevenHunter.jpeg\")
<\/figure><\/div>
The Cornell Lab of Ornithology is a nonprofit, member-supported organization dedicated to the understanding and protection of birds, wildlife, and our shared planet through research, education, participatory science, and conservation. The mission of the Cornell Lab of Ornithology is to interpret and conserve the earth\u2019s biological diversity through research, education, and citizen science focused on birds. birds.cornell.edu<\/a><\/p>\n\n\n\n