Many bird species are unaffected by wind farms, concludes a study carried out by UK bird charities.
Scientists with the British Trust for Ornithology (BTO) and RSPB found that building the turbines was more disruptive than operating them.
Impacts varied between species, with red grouse numbers recovering after construction, curlews declining and not recovering, and skylarks increasing.
Their findings are published in the Journal of Applied Ecology .
This is the latest in a long line of studies on wind farms' interactions with birds, but differs from most in its scale.
Ten species of birds were included, and 18 wind farms in upland areas of the UK were studied - most were monitored before construction began, during construction, and again afterwards.
Defining the problem
"There's certainly no indication in the species we covered in this study that collision mortality is causing a big problem, but we need to bear in mind that it didn't cover the bigger raptors where we know collisions tend to happen," said Jeremy Wilson, head of science at RSPB Scotland.
"Where we are concerned is with two species, curlew and snipe, where we saw the density drop during construction and not recover afterwards.
"The overall picture is perhaps more positive than some of the wilder headlines have suggested; but that doesn't suggest there's no problem."
Curlews, which have seen declines of 30-90% in different parts of the UK in the last couple of decades, are known to be sensitive to human disturbance.
Here, they declined in density by about a third during wind farm construction, as did snipe, without recovering.
Red grouse showed a similar decline during construction, but numbers rebounded directly afterwards.
Skylark numbers soared as the tubines rose, and remained elevated after building finished - perhaps because turning over the vegetation helps them nest or find food.
Lapwings, golden plovers and dunlins showed little change.
The study is unlikely to change the charities' existing approach to onshore wind farms, which is generally to favour development because of the damage that climate change threatens to bring to wildlife worldwide, but to oppose it in areas where birds are likely to suffer.
"What this study does is refine the evidence available to use, to help us define better what is a risky development," said Dr Wilson.
"It's not a black and white picture; but this kind of finding is precisely why we get involved in this kind of research."
Follow Richard on Twitter