Mapping Evidence of Cumulative Harm to Birds from Offshore Wind Farms

Offshore wind power has become one of the world’s fastest-growing sources of renewable electricity, with governments racing to decarbonize their grids and replace fossil fuel generation. Yet a new systematic review reveals a striking imbalance in the scientific evidence underpinning one of the technology’s most contested environmental questions: how the combined, cumulative effects of hundreds of wind installations will affect the world’s birds. The study, published in the journal Environmental Management by researchers at Rutgers University and Map of Life Solutions, mapped nearly three decades of research and found that while scientists know a great deal about which bird species are vulnerable to offshore wind, they know almost nothing about how impacts from multiple facilities add up across populations.

The research team, led by Leon A. Green-Tkacenko with Michael C. Allen and Julie L. Lockwood, conducted an evidence-mapping exercise using Tethys, a comprehensive repository of marine energy research maintained by the Pacific Northwest National Laboratory. Of the approximately 9,920 publications in the database, the team filtered, screened, and synthesized 354 empirical studies on offshore wind effects on birds, spanning publication dates from 1997 to the present. Each study was then mapped against a widely cited framework for understanding cumulative adverse effects, developed by Goodale and Milman in 2016, which identifies the categories of information required to assess how wind development affects wildlife at the population level rather than the level of a single project.

The framework distinguishes three broad categories of evidence. The first is hazards, meaning the physical changes wrought by human activities that could cause harm, including other wind farms and unrelated anthropogenic stressors. The second is vulnerable receptors, which encompasses identifying which species are most at risk, establishing clear baseline population data before construction, and stating thresholds of mortality that populations can tolerate. The third is exposure, covering the spatial and temporal boundaries over which birds encounter wind facilities, including the lifespan of installations, future development pressures, the life histories and phenology of the birds themselves, and the spatial dynamics of populations and development areas. The researchers subdivided these into eleven measurable subcategories to create a fine-grained picture of what the literature does and does not contain.

The results expose a field that has mastered its fundamentals while leaving its most consequential questions almost untouched. Nearly half of all studies, 46 percent, focused on refining which bird species are vulnerable to offshore wind effects, and 40 percent provided pre-construction baseline data. These are the building blocks of any cumulative assessment, the “known knowns” of the evidence base. But the picture darkens dramatically at the analytical frontier. Only four publications, a mere 1 percent of the dataset, examined the nature of effect pathways, that is, whether impacts from multiple wind farms are additive, synergistic, or countervailing. Just 10 studies, roughly 3 percent, attempted to define acceptable mortality thresholds beyond which bird populations would decline.

The distinction among effect pathways is far from academic. If the mortality caused by individual wind farms simply sums up as more facilities are built, an additive pathway, then current projections may be roughly sound. But if effects are synergistic, meaning they compound non-linearly so that combined impacts exceed the sum of individual facilities, population-level harm could be far worse than anticipated. Conversely, countervailing pathways could partially offset one another; for example, if birds displaced from one wind farm area experience reduced collision risk elsewhere, the combined effect would be less than the sum of its parts. The review found that additive mortality is the default assumption in most analyses, frequently without evidence to justify that choice, and at least one cited study openly acknowledged the lack of support for the assumption. If reality follows a multiplicative or countervailing pathway, cumulative effects are being systematically under- or over-estimated.

The threshold question is equally urgent. Setting a ceiling on tolerable mortality, the point at which additional deaths from wind development would push a population into decline, requires detailed knowledge of species’ life histories and demographic rates. The authors point to potential biological removal, a method originally developed for marine mammal management and later adapted for seabirds, which estimates a maximum sustainable yield of inadvertent mortality in a manner analogous to fisheries models. The technique has been applied only sparingly to offshore wind contexts, but the authors argue it could be deployed far more widely, particularly since the baseline population data it depends on are among the better-studied elements of the evidence base. Individual-based models and population viability analyses offer complementary routes, though uncertainty will always demand a precautionary approach.

Geography skews the evidence base as much as topic does. Of the 305 country-specific publications, 250, or 71 percent, came from Europe, followed by 92 from North America and just four from Australasia. The United Kingdom led with 104 publications, the United States followed with 89, and Belgium, Denmark, Germany, and the Netherlands each contributed ten or more. When the researchers compared publication counts against installed turbine numbers, the United States and United Kingdom emerged as disproportionately well represented, while China and Vietnam, the former hosting the largest number of offshore wind turbines in the world, were conspicuously absent. No publication in the dataset provided evidence on cumulative effects on birds in Chinese offshore waters, and none came from Vietnam at all. Less than 10 percent of studies considered waters beyond the Atlantic Ocean and its adjacent seas.

This spatial bias carries a hidden taxonomic cost. Petrels, shearwaters, and albatrosses, the order Procellariiformes, are far more species-rich in the Pacific than the Atlantic, and the East Asian-Australasian Flyway is the most diverse and threatened migratory bird corridor on Earth, home to critically endangered species such as the Spoon-billed Sandpiper. The review also found that the single-order studies it examined clustered heavily on a few groups: Charadriiformes, the gulls, terns, and shorebirds, with 34 publications; Anseriformes, the waterfowl, with 26; and Suliformes, the cormorants and gannets, with 13. Procellariiformes, despite their reliance on offshore habitats, were the subject of just one study.

The divide between marine and non-marine birds proved equally telling. Of the 354 publications, 189 focused exclusively on marine birds, only 20 on non-marine birds, and 145 on both. For landbirds and other non-marine species that migrate over the ocean, evidence for four entire categories was completely absent: effect pathways, mortality thresholds, facility lifespan effects, and development-area siting. The authors attribute this gap in part to a research bias toward breeding-season studies and to the technical limits of tracking small passerines, which are often too small for high-precision GPS tags. Understanding their over-ocean migrations instead relies on light-level geolocators, radar, and Motus tags, none of which deliver comparable spatial or taxonomic resolution.

The review also highlighted an underappreciated role for so-called gray literature. Government and industry reports, which made up 208 of the 354 included documents, were far more likely than peer-reviewed articles to contain baseline population data, appearing in 50 percent of reports versus 25 percent of journal articles, and discussions of facility-lifespan timescales were nine times more common in reports. Peer-reviewed papers, by contrast, more often addressed spatial patterns at broader scales. The authors argue that both streams of evidence are indispensable and that incorporating underrepresented countries’ gray literature into accessible databases such as Tethys would deliver some of the greatest gains. With 68 gigawatts of offshore wind already installed and a pipeline of 454 gigawatts, and with nearly half of all bird species already in decline, the team concludes that building a more balanced and complete evidence base is essential to decarbonizing the grid without deepening the toll on biodiversity.