Nearly 3 billion birds gone

The finding

In September 2019, the journal Science published what became one of the most widely reported ecological findings in recent memory. A team of scientists led by Ken Rosenberg at the Cornell Lab of Ornithology had spent years pulling together population monitoring data for 529 bird species across the continental United States and Canada. When they combined those estimates and ran the numbers, they landed on a figure that stopped them cold: 2.9 billion fewer birds than there had been in 1970.¹

That is not 2.9 billion species. It is 2.9 billion individual birds -- a net loss of roughly 29% of the entire North American breeding bird population within half a century. If you were alive in 1970, more than one in four of the birds alive at that time no longer exist.

What makes this study different

Conservation science typically focuses on individual threatened species or specific habitats. What made the Rosenberg study unusual -- and why it generated front-page coverage across major publications -- was its scope. This was not a report on one declining species or one disappearing ecosystem. It was an aggregate accounting of all birds, common and rare, across all habitats, across an entire continent.

The researchers drew on multiple independent datasets to build their picture: decades of standardised breeding bird surveys run by the US Geological Survey and Canadian Wildlife Service, Christmas Bird Count data going back to 1900, and -- most strikingly -- weather radar data. Radar stations across North America detect migrating birds in the same way they detect rain, and analysis of radar records since 2007 showed that the total biomass of birds passing through the night sky had decreased by approximately 14% in just over a decade.¹ The convergence of those independent lines of evidence made the finding hard to dismiss.

Which birds have declined most

The losses are not spread evenly. Some groups have been hit far harder than others, and the pattern of who has declined reveals a great deal about what is driving the losses.

The grassland figure is particularly stark. A 53% decline means that of every two grassland birds alive in 1970, only roughly one remains. Species like the Eastern Meadowlark, Bobolink, and various sparrow species -- birds that were genuinely common within living memory -- have become a fraction of what they were. The cause is largely the transformation of varied grassland into monoculture farmland, which removes the structural diversity these birds need for nesting and feeding.²

The losses among common backyard species are arguably the most significant from a public awareness perspective. When threatened species decline, it often goes unnoticed by anyone who isn't actively looking. When sparrows, robins, and swallows thin out, it is something anyone who spends time outdoors can feel, even if they can't put a number on it.

What is driving the losses

The study identified no single cause -- the declines are driven by multiple interacting pressures that have intensified over the same half-century period.

Habitat loss and agricultural intensification

The conversion of grassland, wetland, and woodland to intensive agriculture is the largest single driver. Modern farming practices have removed hedgerows, drained wetlands, eliminated field margins, and replaced diverse plant communities with monocultures that support very few insect species. Since birds depend on insects for food -- particularly during breeding season when chicks need protein -- the collapse of insect populations in agricultural landscapes has cascaded up to birds.³

Cats and glass

Free-roaming domestic and feral cats kill an estimated 1.3 to 4 billion birds annually in the United States alone -- making them one of the largest direct sources of bird mortality.⁴ Window collisions kill a further 600 million birds per year in the US. Both are human-caused mortality sources that operate at enormous scale and receive relatively little policy attention.

Pesticides

Neonicotinoid insecticides -- now the most widely used class of pesticide in the world -- are acutely toxic to birds and persistently toxic to the insects they eat. Studies have linked neonicotinoid exposure to weight loss, navigational impairment, and reduced survival in migratory birds. The near-elimination of insects from intensively farmed landscapes removes the food base that many bird species depend on entirely during the breeding season.⁵

Light pollution and building collisions

Most songbirds migrate at night, navigating by stars and the earth's magnetic field. Artificial light disrupts both systems, pulling birds off course, concentrating them around illuminated buildings in cities, and increasing the rate of collision with glass. Radar analysis in the study showed the steepest declines in bird migration density across the eastern United States -- the most densely lit half of the continent.¹

Climate change

Climate change compounds all of the above. It shifts the timing of insect emergence relative to bird breeding seasons, pushes habitat ranges northward faster than birds can follow, intensifies drought and wildfire, and alters the phenology of the plants that insects and birds both depend on. It is not the primary driver of current losses -- habitat destruction holds that title -- but it is increasingly a multiplier that amplifies every other pressure.

What the radar data showed

One of the most compelling elements of the study was its use of weather radar. Scientists had known for decades that radar picks up migrating birds, but it had never been systematically analysed at continental scale to track population trends. When the team did this, they found that the total volume of birds moving through the night sky over North America had dropped by roughly 14% between 2007 and 2017 alone -- a decline visible in the data even in the absence of any species-level survey.¹

This matters because it is an entirely independent line of evidence. It cannot be explained by survey methodology, observer bias, or changes in monitoring effort. The skies are simply moving less biological material than they were a decade ago.

Are any birds recovering?

The picture is not uniformly bleak. Waterfowl -- ducks, geese, and swans -- have recovered substantially since the 1970s, largely because of sustained investment in wetland conservation driven by hunting interests. Raptors including eagles and ospreys bounced back after the banning of DDT in 1972. These recoveries demonstrate that targeted, sustained intervention works. They also demonstrate that the broader crisis is a choice, not an inevitability.¹

What it means for rewilding

Birds are not just a barometer of ecological health -- they are active participants in ecosystem function. They pollinate plants, disperse seeds, control insect populations, scavenge carcasses, and cycle nutrients through ecosystems. Their decline is not a parallel story to the biodiversity crisis; it is the same story, told through one of its most visible chapters.

Rewilding approaches that restore habitat connectivity, remove pesticide pressure, reintroduce apex predators that shape vegetation structure, and allow natural succession offer some of the most promising pathways back. The wood duck recovery -- driven by nest box programs and wetland protection -- and raptor recoveries after DDT show that birds respond quickly when conditions improve. The challenge is creating the conditions at sufficient scale.¹

Frequently asked questions