Science1 publisher3 min readPublished
Heavy wildfire smoke shows up as 3% fewer bird species the following breeding season
Sarah Meier and Eric Strobl matched wildfire particulate data to volunteer bird counts on about 2,800 US survey routes from 2008 to 2022, and the diversity dip held on routes where nothing had burned within five kilometres.
The Scientist · Science desk

What happened
- Sarah Meier of ETH Zurich and Eric Strobl of the University of Bern published their analysis of wildfire smoke and bird diversity in the Journal of Environmental Economics and Management.
- They paired high-resolution wildfire particulate data with counts from about 2,800 North American Breeding Bird Survey routes across the United States between 2008 and 2022.
- Each of those routes runs about 40 kilometres with some 50 stops where trained volunteers record every bird they see or hear, a protocol the survey has used since 1966.
- Going from an average smoke year to a significantly smokier one lined up with around 3% fewer species, roughly one to two species on a route that normally records 55.
- The dip remained after the analysis accounted for birds shifting into neighbouring areas to avoid smoke.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- constraint Damage assessment drawn at the fire perimeter cannot register this loss, because it turns up on ground where nothing burned within five kilometres.
- cost With no monetary figure attached, appraisals of fuel treatment and smoke management carry the biodiversity term at zero, and whoever depends on those birds absorbs the difference.
- exposure Growers and foresters depend on seed dispersal, pollination and pest control from birds, so a diversity decline downwind reaches operations far from any flame front.
- capability The panel approach shows smoke can be separated statistically from the heat and drought it arrives with, a test other smoke-exposure questions can now borrow.
Separating smoke from the weather that makes it is the hard part of this problem. Fires cluster in heat and drought, and heat and drought press on birds on their own. Meier and Strobl compared the same routes against themselves over many years, holding weather, land use and differences in how the counts were conducted [10]. Then they took on the two explanations that would have made the smoke result trivial. If burnt habitat were the cause, the effect should disappear where nothing burnt nearby; on routes with no fire inside five kilometres it persisted [11]. If birds were simply moving a valley over, netting out local shifts should erase it; it persisted there too [12].
Neither test identifies a pathway, and the phys.org account of the work describes a correlation evident across the United States as a whole [22]. The finding is also lagged: heavier smoke in one year, shorter species lists in the next breeding season [6]. That ordering rules out one nuisance. A volunteer straining to hear birds through haze is a detection problem, and a detection problem does not wait a year to appear.
The percentage is a route-level average [7]. On a route logging 55 species, 0.03 times 55 is 1.65 species [20]. At the most heavily smoke-affected monitoring sites, the count fell much more sharply than that [9]. Behind the estimate sit at most about 42,000 route-years, if all 2,800 routes were run in each of the 15 seasons [21]. Species number was one of three measures: the pair also examined how evenly birds were distributed across species, and how far apart those species sit in evolutionary terms [5].
"I wasn't sure whether we would see any effect at all across such a large and ecologically diverse area," Meier said [15]. She is plain about the size of what came back. "The effect isn't huge, but it is measurable. And we never know which species will disappear and which of these are particularly important for an ecosystem. Therefore, even a small decline can, in the worst-case scenario, cause an ecosystem to become unbalanced," she said [16].
Route counts cannot show what happened to individual birds, and the study does not establish what that is. Other research points to smoke weakening birds, possibly through weight loss and respiratory damage, and to altered behaviour during the breeding season [13]. The pair also left valuation out by design, choosing to demonstrate single impacts as reliably as possible first [14].
"In the case of burned-out buildings, we can relatively easily quantify the cost of the damage. With biodiversity, things are far more complex," Meier said [17]. Birds disperse seeds, pollinate plants and help control pests, and a decline in their diversity reaches other animals and plants and eventually sectors such as agriculture and forestry [19]. "There are other consequences of forest fires that we have barely taken into account to date, even though they have an impact on society," Meier said [18].
What to watch
- A distance gradient showing how far downwind the effect reaches, which would set the width of any accounting boundary.
- Replication with data on individual birds, such as mass or survival, to test the weakening pathway the route counts cannot see.
- Whether the lagged effect holds in seasons after 2022, as smoke-heavy years repeat over the same routes.