As far as many environmentalists are concerned, wind farms should be called “wind harms.”
Ornithologists in particular have been raising their voices over avian deaths since the dawn of large-scale turbine installations.
This doesn’t mean that all wind energy companies are in conflict with environmental preservation. One Swiss outfit conducted its own collaborative monitoring, and the results are not as expected.
How much faith do you have in the recorded and reported stats that spell wind turbine disaster for birds?
Leading by example: How Vattenfall is proving wind energy and wildlife can coexist
Leading European energy company Vattenfall has a mission of “fossil freedom.”
This means enabling fossil-free living within one generation—a noble endeavor.
The wholly Swedish state-owned outfit is setting a good example with its handling of the environments around its installations.
Recently, it set up a collaborative study with Spoor, a biodiversity technology company. The target was avian life around one of its turbines at the Aberdeen Offshore Wind Farm, Scotland.
And the results were more positive than anyone—not just environmentalists—would have expected.
This is a surprise, considering that the common negative narrative around wind turbines revolves around how many avian deaths they cause through birdstrikes.
But it turns out that the glowing reports may not be down to changes in bird behavior or changes made to turbines.
Next-level standards of environmental monitoring: AI is coming to the party
There are new standards for environmental monitoring, and artificial intelligence is at the helm.
Vattenfall and Spoor collected over 19 months of continuous, high-res AI monitoring info around the coastal Aberdeen turbine. The focus was on seabirds’ behavior and the detrimental impact of wind installations.
An incredible 2,007 flight paths were covered, with 95% of daylight hours scrutinized for behavior patterns.
The research strategy is a prime example of a methodological shift from theoretical modelling to real-world, continuous data collection.
A particularly beneficial aspect of this strategy is how the “AI + Expert Review” system works to eliminate false positives and ensure data integrity.
After artificial intelligence offered so much input, the results were striking.
Predictions were made in 2018 about the danger the turbines would pose to avian life. Around eight birdstrikes were predicted per year, per turbine.
But the results revealed in a 2026 news release from Vattenfall turn those stats on their heads.
Bridging the gap: How real-world evidence replaces the “worst-case scenario” assumptions
The data show that original models significantly overestimated the danger to avian life.
It was observed that the birds started displaying natural avoidance behavior, proactively steering 350 to 450 feet around turbine blades.
The actual recorded collisions were zero, with a statistical estimate of only 0.002.
Not eight, as predicted eight years ago.
Bird species were actually recorded going about their lives as normal. Not one disruption to behavior patterns or injury was noted.
These species exhibited “micro-avoidance,” making split-second adjustments that rendered the 2018 mortality predictions obsolete.
The AI lens: How high-tech tracking debunked the birdstrike myth
The changes were not so much in bird behavior, but in the methodology used to track it. A leap in observation technology was all that was needed.
Traditional impact assessments relied on human observers or static math. Researchers were effectively “blinded ” to the birds’ natural spatial awareness.
By utilizing Spoor’s AI-driven bio-tracking, Vattenfall captured what humans couldn’t.
Birds are treating turbines as predictable landmarks rather than chaotic threats.
The utilization of AI changed the narrative from “unavoidable impact” to “coexistence.”
And it’s now proven that modern energy infrastructure can demonstrably operate with a negligible footprint on biodiversity.
How many other green solutions are we stalling based on outdated fears rather than real-world evidence?