Wind projects do not emit CO2 or use much water, a big argument in favour of their deployment to combat man-made climate change. But what about the impact of climate change on global wind resources, or on wind turbines themselves through, say, high loads caused by extreme winds, or by rising sea levels and wave height?
A trend of long-term change in weather linked to carbon emissions is clear, but it is hard to tease out the natural variability of weather and climate, especially as wind patterns are harder to predict than temperature changes. The issue, however, has been pushed up the news agenda because of the slowest wind speeds recorded in 47 years in California and Texas in the first quarter of 2015, according to Finnish weather monitoring firm Vaisala. Figures for the second quarter showed the western US's wind-generated electricity production was still lower than normal. By contrast, winds speeds in central Canada and some US border states in the first quarter were higher, by up to 20%.
Preliminary evidence indicates that the impact of climate change on wind is likely to be mild over a project's 20-25 year lifetime. "Climate change doesn't factor into most wind investment or lending decisions at the moment," says Michael Brower, president and chief technical officer of renewable engineering consultancy AWS Truepower.
There are hints in the models of modest change over 40 to 50 years, he says. But researchers think that the likely impacts are small enough for turbine design and wind farm siting and layout to evolve and keep pace.
Pascal Storck, Vaisala's global manager of energy services, notes that there have been many climate-change studies looking at long-term wind-speed trends, both in past decades, based on measurements, and in the future, based on climate-model projections. "The take-away from these studies is that there are probably trends in wind speed connected to climate change, but they are small, vary regionally and are not consistently of the same (positive or negative) sign," he says.
However, larger short-term variability in wind can affect a project's ability to service its debt, and few, if any, studies have examined this, Storck adds. Of climate change and wind speed volatility, he says: "We are starting to see more extreme records broken, for example in terms of temperature, precipitation and snowpack. I have no expectation that wind will differ."
For the past three or four years AWS Truepower has added climate-change risk (or uncertainty) to its wind-resource assessments. It already factors in an 8% uncertainty ratio for ten-year projections, and 7.9% for 20-year projections. Adding climate-change uncertainty pushes these risk ratios up to 8.1% and 8.5% respectively.
For the financial community, AWS Truepower expresses the uncertainty as a P50:P90 ratio, the standard ratio used in wind power financing. P50 or P90 means there is 50% or 90% probability that the predicted energy will be met or exceeded. P50 is the mean, and therefore more likely to occur, while P90 is the high-confidence measure. Excluding climate-change risk, the company expresses the uncertainty as 90.7% for ten years and 90.8% for 20 years. When climate-change risk is factored in, the ratio falls slightly - to 90.6% for ten years and 90.2% for 20 years.
Regional climate modelling
Over time, temperate and sub-arctic zones are likely to see reduced wind because of the shift north of the polar jet stream and a weakening in the temperature gradient between the poles and the equator, Brower says. So far however, changes have been hard to detect. About three quarters of global wind-power capacity by the end of 2012 was installed between 30 and 60 degrees north in Europe, North America and China, notes a paper by Sara Pryor and Rebecca Barthelmie of Cornell University.
A research paper published in 2010 by atmospheric scientist Diandong Ren, then of the University of Texas in Austin, compared various climate projections and found that, on average, higher temperatures would reduce wind-power density at turbine hub height in China by about 14% by 2100, equivalent to a 5% reduction in wind speed. However, that is not enough of a reduction to affect long-range wind-development planning, says Brower.
A few years ago, Vaisala summarised the research as follows: on an annual basis, climate change is predicted to cause stronger surface-wind speed values across the boreal regions of the northern hemisphere, including much of Canada, Siberia and northern Europe, and in tropical and subtropical regions in Africa, Central and South America. However, Greenland, southern Europe, China, India, southern Australia, and much of the west coast of South America are expected to experience decreasing surface wind speeds.
It cautions that inter-model contrasts are largest in western North America, sub-Saharan Africa, broad swathes of Eurasia, Brazil and the Andean region of South America. In the US, most models concur that the mean annual wind speed values will increase across the Great Lakes region extending southward across the Midwest and into Texas. Decreased values are predicted across most of the western US. "However, these predicted changes have a strong seasonal dependence, with wind speed increases over most of the US during the winter and decreases during the summer," says Vaisala.
A study published in June in the Renewable Energy journal looked at the results of four regional climate models from the North American assessment programme. It found "strong agreement" that the wind-energy resource will increase from 2038-2070 in parts of Kansas, Oklahoma and northern Texas, a region that accounts for much of the US's wind development.
According to atmospheric scientist Sara Pryor, global climate change may alter either storm dynamics or storm tracks or both, particularly in the mid-latitudes, and may affect the wind resource in the context in which the industry operates at specific locations. But, she cautions, there has been very limited research so far.
Pryor says that simulations with regional climate models and using statistical downscaling techniques suggest no clear tendency in the next 50 years for the wind resource over the contiguous US, and no change or modest increases over northern Europe. Extreme winds are not projected to be likely to move beyond current safety margins or current climate variability.
Even fewer studies have been conducted for South America and Asia - including China and India - and there is a need for analyses in both regions, she says.
The uncertainties in climate science were shown by a study using PRECIS (proving regional climates for impacts studies), a regional climate modelling system developed by the UK's Met Office. It showed a startling decline of up to 60% in Brazil's wind resource by 2100, but the result may derive partly from the simplifying assumptions employed in the study, says Pryor.
As temperatures rise, icing on turbines could become less problematic, opening up more sites for wind turbine deployment. Similarly, more ice-free areas of sea would alter the accessibility of project sites, while drifting ice changes the loads on turbine foundations.
Less permafrost would also have a huge positive impact on road construction and repair at high elevations, opening up more new avenues for wind development. But, conversely, higher temperatures and higher altitudes reduce energy production.
A higher sea level may make foundations in coastal areas more prone to flooding, and more salinity can increase corrosion damage. And a change in waves - higher in the North Sea and lower in the Mediterranean - could increase or decrease the loading on offshore turbines, say the researchers. Component materials fare differently in different temperatures. Rubber seals can become more brittle in the cold, and different lubricants are needed if temperatures rise.
This year's unusually low winds in the western US were probably not attributable to climate change but to a high-pressure ridge forming over the northern Pacific, prompting above-average winter temperatures due to the polar jet stream being further north than usual. But the phenomenon has concentrated minds and raised some intriguing questions, not least regarding over-reaction to short-term weather changes. "We're worried a little that investors will panic," says Ed Zaelke, co-chair of the global project-finance practice at Akin Gump. Indeed, the stock market reacted sharply when yieldco NGR Yield said on 4 August that the second quarter's "historically low" winds had hit its earnings and revenues. Its share price dropped 11% that morning, a market response perhaps exaggerated because of longer-term fears about climate change.