There is a global imperative to transform the energy system to reduce emissions of greenhouse gases and other pollutants and provide a sustainable source of power for the world. Wind energy is leading the charge.
Wind turbines are now providing a substantial fraction (approximately 7%) of global electricity generation and are operating in 90 different countries. The global installed capacity of wind-derived electricity over the last decade has grown at a rate of over 8% per year. Recent work (Pryor et al. 2020, Climate change impacts on wind power generation, Nature Reviews Earth and Environment) has suggested that continue expansion of global installed capacity at this rate would lead to cumulative avoided carbon dioxide emissions of upto 154 GtCO2 by 2050 and delay passing of the 2°C global warming threshold by 5 years.
Wind turbines are designed to withstand harsh conditions during the expected 20 to 30 years of continuous operation to avoid excess mechanical failures and provide clean, reliable and cost-efficient electricity. The work a wind turbine does is similar to you continuously driving your car around a football field at 80 miles per hour for 30 years all the while being buffeted by winds and rained/snowed and hailed on! You stop only once per year, or less, for maintenance.
But how can/should developers characterize those conditions and which wind turbines should they choose? A key aspect of selection of appropriate wind turbines for a given location is an assessment of the extreme aerodynamic and mechanical loads to which they will be subject. A key source of those loads is the extreme wind speed. That is the wind speed within the swept area of the wind turbine blades that is expected to be equaled or exceeded once during a 50 year period. We refer to that as the fifty-year return period wind speed.
To date the fifty-year return period wind speed has not been available to prospective wind farm developers so the convention has been to estimate it using the so-called reference wind speed. The reference wind speed is defined as five times the annual mean wind speed. The current design guidelines identify four key thresholds for this reference wind speed; 37.5 ms-1, 42.5 ms-1, 50 ms-1 and 57 ms-1. This last class of wind turbines are designed for regions that are subject to tropical cyclones and are thus likely to experience the highest wind loading.
This work generates estimates of the fifty-year return period wind speed for all regions of the globe at a 30 by 30 km grid resolution. The resulting digital atlas contains extreme wind speed estimates in 1 million grid cells. It is available for public download and also includes quantified uncertainties on the extreme wind speed estimates.
These estimates of extreme wind speeds at 100 m above the ground or sea level are derived using 40-years of hourly output from a global reanalysis developed by the European Center for Medium Range Weather Forecasts. We took hourly wind speeds in each grid cell and applied statistical approaches to quantify the wind speed expected to be equaled or exceeded once during a 50 year period. The methods use a branch of statistics called generalized extreme value distributions and permit us to provide a best-guess of the extreme wind speed along with quantified uncertainties. Beyond providing guidance to prospective developers regarding the extreme wind speed, we also show that the fifty-year return period wind speed is generally lower than the reference wind speed. This means that for many locations wind turbines are being over-engineered. That leads to excess costs because they are being strengthened more than is necessary. Think about it – it is possible that Florida will experience snow, but if you live there you don’t fit snow tires to your car because the chance of you needing them is small and the cost of buying the snow tires, fitting them and the associated loss of fuel efficiency isn’t worth it.
This digital extreme wind atlas is designed to facilitate the continued expansion of the wind energy industry and to aid deployment into areas with very limited observational data and to do so at the lowest levelized cost of energy.
Pryor S.C., Barthelmie R.J., Bukovsky M.S., Leung L.R. and Sakaguchi K. (2020): Climate change impacts on wind power generation. Nature Reviews: Earth and Environment 1 627-643 doi: 10.1038/s43017-020-0101-7.