Using software tools developed by the marketing group Near Zero, which has developed open-source software tools to examine where experts agree and disagree and why, a research group hosted by the Carnegie Institution for Science's Department of Global Ecology has completed the largest expert survey yet of wind energy. 

Using that survey platform, the authors gathered responses from 163 of the world's foremost experts on wind energy to forecast future costs for this energy source - as you might expect, they gushed about its future. On average, the wind advocates surveyed predicted wind power costs would continue falling for the next several decades, for three major classes of wind turbines, both onshore and offshore, with prices falling by 24-30% by 2030, and 35-41% by 2050.

Why would that happen, when subsidies have never been more than a crutch in any other area of the market? The survey did not ask that, though every non-partisan energy expert would.

To gauge how much wind power production might scale up if the optimistic projections were real, scholars often use energy system models, such as the International Energy Agency's World Energy Outlook model, or the integrated assessment models used for generating scenarios assessed by the Intergovernmental Panel on Climate Change (IPCC). The models' assumptions about future costs of wind energy have an important influence on modeling results for the future potential for wind energy, Wiser and colleagues argue. 

Future costs of wind energy are often forecasted by looking at how past costs have fallen, following a "learning curve," and then extrapolating that curve into the future. Another method for estimating future costs is through bottom-up engineering assessments, looking at the costs of various parts of wind turbines.

Surveys of experts--known as expert elicitations--are one method for forecasting the future, which allows for more scope for experts to draw on technical knowledge as well as their informed opinion about future developments.  But like push-polling in politics, you already know the answer you want to get. It's like asking how great something is on a scale of 9 to 10.

The new elicitation asked wind energy evangelists to estimate future wind power costs under a range of scenarios--including a median scenario as well as "high cost" and "low cost" cases. This allowed the authors to assign probabilities to the range of possible scenarios.

The paper, whose lead authors was Ryan Wiser, Group Leader in the Electricity Markets and Policy Group at Lawrence Berkeley National Laboratory, was published in Nature Energy.