Wind energy is a rapidly growing alternative to hydrocarbons. Global wind capacity reached 597 GW at the end of 2018, meeting nearly 6% of electricity demand. This production, which is proportional to the speed of wind cubed, has nonetheless been adversely affected since the 1980s by a reduction in the average surface wind speed, a phenomenon known as "global terrestrial stilling".

Drawing on data from wind power plants around the world, the researchers have shown that this trend reversed in 2010. Wind speeds have returned to 1980 levels in just eight years, with a rate of growth three times higher than the slowdown observed before 2010. This recovery has been the most pronounced in North America, Europe and Asia.

After eliminating factors related to vegetation and urbanization, the researchers can now attribute these variations to climatic mechanisms and, more specifically, to the driving forces associated with decadal variability in large-scale oceanic and atmospheric circulations. According to their work, the "Pacific Decadal Oscillation", the "North Atlantic Oscillation" (NAO) and the "Tropical North Atlantic Index" explain to a large extent both the global terrestrial stilling phenomenon and the recent trend reversal.

Changes in wind speed have had and therefore will have an enormous impact on the wind industry.

In effect, the upswing between 2010 and 2017 has led to a 17% increase in wind power potential worldwide, and 2.5% in the United States. This demonstrates that wind speed plays at least as important a role as technological innovations, which were long considered to be the only factor contributing to the increase in wind generation.

If the current trend persists in the medium term, wind generation will increase systematically by 3% per decade.

This new approach paves the way for long-term planning within the global wind industry.

As part of this collaboration led by Princeton University (USA) and featuring many Chinese and European partners, researchers from the LSCE (CNRS/CEA/UVSQ) participated in interpreting data from the different global wind datasets.