47% of North American Bird Species Are Declining - and in Three Regions, the Rate Is Accelerating

The numbers describing North American bird decline have been known in broad outline for years. What has been harder to establish is whether those declines are speeding up, steadying, or beginning to slow - and which specific factors are pushing certain regions from ordinary decline into accelerating collapse.

A study published in Science by Francois Leroy and colleagues addresses both questions using 34 years of data from the North American Breeding Bird Survey, one of the most sustained monitoring programs in ecology. By analyzing 261 species across 1,033 survey routes from 1987 to 2021, the team mapped not just the direction but the dynamics of bird population change across the continent.

The Statistical Approach

Earlier analyses of bird population trends focused primarily on total counts: are there more or fewer birds than before? Leroy and colleagues applied advanced statistical modeling to estimate population dynamics over time - tracking not just abundance at a given point but the rate of change in abundance and whether that rate itself is increasing or decreasing.

This approach distinguishes between a species that is declining at a consistent rate and one whose decline is accelerating. The latter is more urgent: a population losing 2% per year is in a very different trajectory than one that lost 1% annually five years ago but is now losing 3%. Detecting acceleration requires a long enough time series and statistical methods capable of separating genuine trend changes from year-to-year variation.

Key Findings

Across all survey routes, the analysis detected an average 15% decline in bird abundance per route over the 35-year period. Among the 261 tracked species, 122 - comprising 47% of the total - showed statistically significant population declines. Within that declining group, 63 species showed evidence that their decline is accelerating: getting worse faster each year.

An additional 67 species show declines that outpace population recruitment - births and immigration are not replacing deaths at sufficient rates to stabilize numbers. This signals deeper demographic strain than simple abundance trends capture.

Geographically, the steepest losses concentrate in the southern and warmer regions of North America, but the sharpest acceleration in decline rates shows up in three distinct hotspots: the Mid-Atlantic, the Midwest, and California. These regions share intensive agricultural land use as a common feature - the alignment between accelerating bird decline and metrics of cropland extent, fertilizer application, and pesticide use was the strongest predictor the analysis identified.

What the Agricultural Connection Means

The study is correlational. It cannot establish whether cropland, fertilizers, pesticides, or the interaction among all three drives decline, nor can it rule out unmeasured co-varying factors. Agricultural intensity is the strongest predictor in the model, but the mechanism translating that intensity into bird population dynamics - whether through direct pesticide exposure, prey insect suppression, habitat loss within farmed landscapes, or some combination - requires additional research to establish.

The taxonomic breadth of the pattern is striking. Acceleration in decline was not limited to farmland specialists or species with known agricultural vulnerabilities. It appeared across families and functional groups with diverse habitat requirements and feeding ecologies, which is consistent with a systemic driver affecting entire ecosystems rather than specific species.

The findings extend and deepen prior work documenting aggregate bird losses across North America, adding the crucial dimension of acceleration - the knowledge that in some of the continent's most productive agricultural regions, the loss is not stabilizing but picking up speed.