Bird populations in North America: Decline accelerating—nearly every second species affected

Bird populations in North America have been declining for decades. But a new study now shows that not only are bird numbers falling, the speed of this decline is also increasing.

This is the conclusion of a study published in the journal Science by researchers from Ohio State University and the Czech University of Life Sciences Prague. The scientists analyzed over 35 years of population data for 261 bird species—with alarming results: In many regions, bird populations are not only shrinking, the decline is even accelerating, particularly in intensively farmed landscapes.

The long-term study by François Leroy, Marta A. Jarzyna, and Petr Keil thus provides a new perspective on an already well-documented problem. Until now, research has primarily focused on the question of whether bird populations are declining. Numerous studies show that this is indeed the case. A high-resolution analysis of North American bird populations found, for example, that approximately 75% of the species studied are in decline (Alexander Johnston et al. 2025). Another widely cited study estimated as early as 2019 that the total number of birds in North America has declined by approximately 29% since 1970—a loss of roughly three billion individuals (Kenneth V. Rosenberg et al. 2019).

The current study now poses a further question: Is this decline in bird populations accelerating? And if so, where are the geographic hotspots of this trend?

The data: 35 years, 1,033 routes, 261 species

The analysis was based on the North American Breeding Bird Survey (BBS)—one of the most comprehensive and longest-running bird population monitoring programs in the world. Since the 1960s, volunteers have been counting all birds they see or hear along designated routes each year. For the study, 1,033 of these survey routes from the period 1987 to 2021 were analyzed.

To derive the most realistic population estimates possible from the count data, the researchers combined modern statistical methods—so-called N-mixture models within a fully Bayesian analytical framework—with spatial modeling. This allowed them not only to determine whether bird populations were increasing or decreasing, but also to examine whether the rate of these changes shifted over time.

At the continental scale, the trend is clear: the average number of birds recorded per survey route fell from about 2,034 individuals in 1987 by roughly 304 birds by 2021. This represents a decline of about 15% over the 35-year study period — roughly one out of every seven birds counted at the start of the study.

The spatial analysis also shows a clear trend: 718 routes (70%) recorded a significant decline in bird numbers, while only 172 routes (17%) showed an increase.

Not just decline—but acceleration

What matters, however, is not just that bird numbers are falling. The study shows above all that this decline is increasingly accelerating in many regions.

To examine this dynamic, the researchers distinguish between two quantities. First, the population trend itself—that is, how much bird numbers change over time. Second, the change in this trend: Is the decline becoming stronger or weaker over the years? Mathematically, this corresponds to the first and second derivative of population change.

If a population’s annual growth rate—the difference between gains and losses—continues to decline over time, it means: Populations are not just declining, the decline is accelerating. This is exactly the pattern that emerges in the continental average of the study. The growth rate of the studied populations decreased significantly, indicating a large-scale acceleration of population losses.

This pattern is also evident at the species level: Of the 261 bird species studied, 122 species (47%) showed a significant population decline. 63 of these species—more than half of the declining species and roughly one quarter of all species studied—also experienced a significant acceleration of this decline.

A similar picture emerges at a higher taxonomic level. Of 21 bird families with significantly declining populations, the majority also showed an acceleration of losses. The trend is therefore not a problem limited to individual particularly sensitive species groups, but cuts across many different bird groups and habitats in North America.

Where it is worst—and why

The greatest average population losses were found primarily in the South and Southwest of the USA. Bird numbers declined particularly sharply in Florida, Texas, Louisiana, and Arizona.

The hotspots of acceleration, however, show a different geographic pattern. The decline accelerated particularly markedly in the Midwest—in states such as Indiana, Ohio, Kentucky, Illinois, Wisconsin, and Michigan—as well as in parts of the Mid-Atlantic region (Delaware, Maryland, New Jersey) and in California.

This spatial distribution is no coincidence. The researchers linked their population estimates with twelve environmental and land use variables, including temperature data, vegetation cover, the so-called human footprint, and several indicators of agricultural intensity—such as pesticide use, fertilizer use, and cropland area. Using machine learning analytical methods, they identified those factors that best explain the observed patterns.

Main causes: Intensive agriculture and climate change

The analysis reveals two central correlations. The decline in bird numbers itself is particularly strongly associated with high and rising temperatures. The acceleration of this decline, however, occurs primarily where agriculture is particularly intensive .

Regions with high pesticide use, heavy fertilizer use, and large cropland areas—three variables that the authors collectively interpret as an indicator of agricultural intensity—show the most pronounced accelerations of population losses.

Particularly problematic appears to be the combination of intensive agriculture and rising temperatures. Agricultural landscapes often warm more than natural ecosystems, for example due to lower vegetation cover or altered surface properties. This can further amplify climate-related stresses on bird populations.

The models also suggest that the strongest acceleration effect occurs in regions with mean annual temperatures of around 10 °C—precisely where bird populations are particularly dense and human land use is particularly intensive.

Which Habitats and Species Are Affected?

The study analyzed a total of ten habitat types. Only forest species showed a significantly positive population trend on average—albeit with a decreasing growth rate, suggesting that this increase is already slowing.

Four habitat types, however, recorded significant population declines: urban areas, grasslands, wetlands, and open woodlands. Particularly concerning is that losses in wetlands and open woodlands are additionally accelerating—populations there are not just declining, but the decline is growing stronger year after year.

At the species level, too, a broad spectrum of affected birds is evident. Notably, even common and widespread species are among the populations whose decline is accelerating. These include the American crow (Corvus brachyrhynchos), the common starling (Sturnus vulgaris), and the mourning dove (Zenaida macroura), which were long considered typical and stable components of North American bird communities.

Similar trends are shown by other well-known species. The red-winged blackbird (Agelaius phoeniceus), one of the most common birds in North America, is experiencing accelerated population losses just like the common grackle (Quiscalus quiscula), which is widespread in many agricultural landscapes. Also affected are the American goldfinch (Spinus tristis), a common garden bird, as well as the song sparrow (Melospiza melodia), one of the most widespread songbird species on the continent. The indigo bunting (Passerina cyanea), whose bright blue plumage makes it one of the most striking songbirds in North America, is also among the species with accelerated decline.

Other species also show declining populations, but their decline appears to be slowing. These include the house sparrow (Passer domesticus) and the chimney swift (Chaetura pelagica), a migratory bird that is heavily affected by the loss of suitable nesting sites in chimneys as well as by the decline of flying insects.

On the other hand, there are also some species with positive population trends. For example, populations of the Canada goose (Branta canadensis) continue to increase, while the Eurasian collared dove (Streptopelia decaocto)—a species that only arrived in North America in the 20th century—continues its expansion. However, such developments remain the exception: Overall, significantly more species show declining than increasing populations, and even among many growing populations, a slowdown in growth is already evident.

What Does This Mean for Conservation?

The researchers emphasize that not only the decline of bird populations itself is significant, but also its dynamics. Many populations are not just declining—the decline is even accelerating. The study also shows that regions with particularly strongly accelerated population losses frequently overlap with areas of intensive agriculture. However, a clear causal link cannot yet be derived from this. Robust causal conclusions would require targeted experiments or so-called quasi-causal study designs.

Despite this limitation, the methodological insight of the study is significant: Those who assess biodiversity changes solely on the basis of absolute population numbers may overlook important warning signs. The acceleration of population declines can be an early indicator of future extinction risks. In many monitoring programs and conservation strategies, this dynamic has so far been scarcely considered.

The authors therefore propose applying such acceleration measures to other biodiversity indicators in the future as well—such as species richness, changes in species composition, or species turnover rates. This would provide a more comprehensive picture of ecological changes.

If population declines are accelerating, it could mean that we have been overlooking important warning signs—and that the loss of biodiversity is progressing faster than many conservation programs currently account for.

---

Source

• François, L., Jarzyna, M. A., & Keil, P. (2026). Acceleration hotspots of North American birds’ decline are associated with agriculture. Science 391, 917–921. https://doi.org/10.1126/science.ads0871