Between April 2023 and October 2025, a research team led by Dr. Elena Markarian at the University of Massachusetts Amherst placed motion-activated cameras in 187 residential yards across three towns — Northampton, Easthampton, and Hadley — to record the actual rate of bird predation by free-roaming domestic cats. The study was funded by a small grant from the Massachusetts Audubon Society and published in the Journal of Urban Ecology in February 2026.
The cameras recorded 2,847 distinct predation events across the thirty months of fieldwork. Of those, 2,103 involved birds. The remaining incidents were small mammals, lizards, and one unfortunate juvenile rabbit.
Extrapolated to the cat population of the three towns — a known quantity from local licensing records and an estimated number of unowned cats — the data suggest that domestic cats killed somewhere between 38,000 and 62,000 birds in the study area across the thirty-month period. The total bird population estimate for the same area, derived from Breeding Bird Atlas counts, is roughly 240,000 individuals at the spring breeding peak.
These numbers will surprise no professional ornithologist. The 2013 paper by Scott Loss, Tom Will, and Peter Marra, published in Nature Communications, estimated that domestic cats kill between 1.3 and 4.0 billion birds annually in the contiguous United States. The Markarian study is a small, local replication of a much larger and well-established finding.
What the Markarian study added was the camera footage. Previous estimates depended on owner-reported predation, scat analysis, prey-return rates, and stomach contents from euthanized cats. The camera data showed, for the first time, what was happening in real time in the yards where the cats lived.
Three specific findings stood out.
First, the cats classified by their owners as 'not really hunters' killed birds at almost the same rate as the cats classified as 'good hunters.' Across 73 owner-surveyed households, the self-reported hunter cats averaged 18 predation events per camera-year. The self-reported non-hunters averaged 14. The owners simply were not seeing what their cats were doing.
Second, the bell-collar reduction effect, often cited as evidence that bells solve the problem, was smaller than usually claimed. Cats wearing bells killed approximately 31 percent fewer birds than unbelled cats — a real reduction, but not a solution. The remaining 69 percent of mortality was happening anyway.
Third, the colourful collar covers marketed under the brand name Birdsbesafe — which use bright fabric to alert birds visually — produced a 54 percent reduction in bird predation in the study sample. This is a larger effect than the bell, and consistent with previous research published in the United Kingdom and the United States.
Markarian is careful, in conversation, to distinguish between the population-level findings and the conclusions an individual cat owner can draw. The data do not mean that any specific cat is killing thirty birds a year. The distribution is heavily skewed. A few prolific hunters account for most of the kills. Most cats kill fewer birds, and a small minority kill none at all.
What the data do support, with high confidence, is that the average free-roaming domestic cat in a residential yard is a meaningful mortality source for songbirds, and that any cat owner who allows the cat outdoors unsupervised is participating in that mortality at some level.
The interventions, ranked by effectiveness, are a short list.
The most effective is the obvious one: keep the cat indoors. Indoor cats kill no songbirds. They also live, on average, about three times longer than outdoor cats — twelve to seventeen years rather than three to seven — because they are not hit by cars, mauled by other cats, or exposed to feline leukemia and other diseases. The veterinary case for indoor housing is at least as strong as the ecological case.
The second most effective intervention is the catio. A screened enclosure attached to the house, accessible through a window or pet door, gives the cat outdoor air and sunlight without giving the cat the yard. Designs range from a $200 prefabricated unit to a custom-built ten-by-twelve structure with shelves and plantings. The cost has fallen substantially in the past five years as the concept has become mainstream.
The third, for owners who cannot or will not keep the cat indoors, is the Birdsbesafe collar cover. The cost is about ten dollars. The fitment is non-trivial — the cat must accept a breakaway collar plus the fabric cover, and not all cats will — but the predation reduction is real and well-documented.
The bell, on its own, is the weakest of the practical interventions. It reduces but does not solve. It is better than nothing.
What the Markarian study did not address, and what the broader literature has not fully resolved, is the question of unowned cats. Feral and semi-feral cats — the colonies maintained by trap-neuter-return programs and the strays that move between households without ownership — represent a substantial fraction of the killing in most populated landscapes. The interventions available to individual cat owners do not reach those animals.
On that question, the conservation community remains divided. Some advocacy groups support continued TNR programs as a humane management strategy. Others, including the American Bird Conservancy, argue that the bird mortality from feral colonies is unacceptable and that the colonies should be progressively removed through adoption and lethal control. The debate has been ongoing for thirty years and is not close to resolution.
For the backyard birder with a single house cat, the question is more tractable. The cat can be kept inside. The cat can have a catio. The cat can wear a Birdsbesafe collar. None of these is a dramatic intervention. All of them, individually or together, materially reduce the contribution one household is making to a continent-wide mortality problem.
Markarian has now returned to a follow-up study, running cameras in fifty yards in which the cats have been moved to one of the three intervention categories above. The preliminary data, shared informally at a regional conference in March, suggest that the household-level interventions produce yard-level reductions in bird mortality of roughly the proportions found in the literature.
None of this is news. The science has been clear for at least fifteen years. What the cameras have done is make the abstract numbers visible, and harder to dismiss. The cat owner who watches the footage from her own yard, as several of Markarian's participating households did, is not usually the same cat owner she was the week before.
