Francois Leroy and colleagues have confirmed what field biologists have suspected for years: bird populations across North America are not just shrinking but shrinking faster, with the steepest acceleration concentrated in the continent’s most intensively farmed regions. Their study, published in the journal Science on February 26, 2026, analyzed 261 species across 1,033 survey routes spanning more than three decades and found that nearly half of those species are losing ground at an increasing rate. The findings sharpen the picture drawn by earlier research and point directly at industrial agriculture as a primary driver.
Three Decades of Data Reveal a 15 Percent Drop
The research team drew on the long-running Breeding Bird Survey, a federal monitoring program that has sent trained volunteers along fixed roadside routes since 1966. Leroy and colleagues used route-level point counts, observer data, and weather records from 1987 through 2021 to build a continent-wide picture of abundance change. On average, bird numbers along those routes fell by about 15 percent over the study period. Put another way, for every six birds counted on a route in 1987, only five remained roughly three decades later.
That average masks sharp geographic variation. While declines appeared across much of the continent, the researchers identified spatial hotspots where losses were not just large but accelerating, meaning each successive decade brought steeper drops than the one before. Of the 261 species examined, 122 species, or 47 percent, showed acceleration. The Midwest stood out as a consistent cluster of those worsening trends, with many routes in corn and soybean country showing particularly rapid downturns.
Farmland Is the Common Thread
The geographic overlap between acceleration hotspots and intensive cropland is the study’s most striking finding. Leroy and colleagues cross-referenced bird trends with federal land-cover data from the annual cropland layer, which classifies crop types and acreage across the lower 48 states in satellite-derived rasters. They also incorporated county-level pesticide-use estimates from the USGS pesticide synthesis, which models application rates from agricultural surveys and crop statistics. The result was a statistical association between areas of high agricultural intensity and the fastest bird declines.
This pattern holds even after accounting for other variables such as climate, habitat type, and baseline species richness. Routes running through corn and soybean belts, where monocultures dominate and chemical inputs are heaviest, consistently showed steeper downward trajectories than routes in less cultivated areas. The study does not claim a single-cause explanation. Factors like habitat loss, pesticide exposure, fertilizer runoff, and mechanized harvesting likely interact, but the spatial correlation is tight enough to identify agriculture as a leading suspect rather than a background factor.
Grassland Birds Bear the Heaviest Losses
The new findings build on a well-established baseline. In 2019, a team led by Peter Marra published research documenting a net loss of roughly 3 billion birds in North America since 1970. That earlier work, also published in Science, drew on multiple monitoring networks and flagged grassland species as suffering especially large proportional and absolute declines. Species such as meadowlarks, bobolinks, and many sparrows that depend on open grasslands for nesting and foraging have lost habitat at a pace that outstrips nearly every other guild.
What the 2026 study adds is evidence that those losses are not leveling off. The acceleration signal is strongest precisely in the regions where grassland conversion has been most aggressive, particularly in the Great Plains and upper Midwest. A separate USGS investigation found that expanding biofuel crops in North Dakota contributed directly to grassland bird losses by converting native prairie to row crops. That dynamic, replicated across much of the prairie pothole region, helps explain why grassland species keep falling even after decades of documented decline.
As more perennial grasslands give way to annual row crops, birds lose nesting cover, insect-rich foraging grounds, and the structural diversity they need to avoid predators. The timing of planting and harvest can destroy nests outright, while herbicides and insecticides reduce both plant diversity and invertebrate prey. The Leroy team’s acceleration metric suggests that these pressures are not simply continuing but intensifying in key agricultural zones.
Why Correlation Deserves Careful Reading
Most coverage of the Leroy study has treated the agriculture link as settled science, but the data warrant a more careful read. The Breeding Bird Survey methods rely on roadside counts conducted during a narrow breeding-season window, and the program’s own documentation notes limitations in translating route-level trends into absolute population estimates. Roads tend to run through agricultural land, which could amplify the apparent connection between farming and bird loss simply because survey coverage is densest there.
The pesticide data, too, carry caveats. USGS county-level estimates are modeled from survey responses and crop acreage, not from direct field measurements, and the agency has flagged discontinuities in how seed treatments and certain chemical classes are counted over time. None of this invalidates the study’s central finding, but it does mean the strength of the causal link between specific farming practices and bird declines remains an active area of investigation rather than a closed question.
Statistical models that link bird trends to land use are also sensitive to how spatial scales are chosen. A county-level pesticide estimate may not capture conditions on an individual farm or field, and some species range widely beyond the narrow strip of habitat visible from a roadside route. Leroy and colleagues acknowledge these constraints, framing their work as a high-level diagnosis of where and when declines are accelerating, not a definitive indictment of any one chemical or crop.
What Accelerating Losses Mean for Farmland Ecosystems
Birds are not just scenic additions to rural life. They consume vast quantities of crop-damaging insects, disperse seeds, and serve as indicators of broader ecosystem health. When bird populations drop, pest pressure can rise, potentially increasing the very pesticide use that contributed to the decline in the first place. That feedback loop is difficult to quantify with precision, but the directional risk is clear: fewer birds in farm country could mean higher costs for the farmers whose practices are driving the losses.
The Leroy analysis underscores that risk by showing how quickly trends can worsen once they begin to accelerate. A modest decline that compounds over multiple decades can translate into local extirpations, especially for species already squeezed into remnant habitat patches along fencerows, ditches, and shelterbelts. As those birds disappear, farms may lose a free line of defense against outbreaks of grasshoppers, caterpillars, and other herbivores.
There are potential bright spots. Practices such as maintaining field margins, restoring prairie strips, reducing pesticide use, and diversifying crop rotations have all been associated in past studies with higher bird abundance. The new acceleration maps could help target such interventions to the places where they are likely to yield the greatest ecological return. Still, the scale of the documented losses suggests that incremental tweaks will not be enough without broader shifts in how much land is devoted to intensive monoculture.
A Warning Signal, Not the Final Word
Leroy and colleagues have provided a continent-wide early warning system for avian collapse in farm country. By focusing on acceleration rather than just net change, their work highlights where conservation efforts are most urgently needed and where agricultural policy may be out of step with biodiversity goals. For grassland birds in particular, the message is stark: the bottom is falling out faster than many models anticipated.
At the same time, the study is a reminder of the limits of large-scale monitoring. Roadside surveys, modeled pesticide data, and coarse land-cover maps are blunt tools for dissecting the fine-grained realities of modern agriculture. Translating their patterns into concrete policy (whether that means reforming crop insurance, incentivizing habitat set-asides, or tightening pesticide regulations) will require more detailed, local research that can pinpoint which practices are most harmful and which offer the best hope for recovery.
For now, the clearest conclusion is that in North America’s most heavily farmed landscapes, birds are disappearing faster than before, and that acceleration is closely tied to how the land is worked. Whether those trends can be slowed, or reversed, will depend on decisions made not just by scientists and conservationists, but by the farmers and policymakers who shape the continent’s fields each season.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
