**6. Chapter summary**

Several factors are influencing the accurate assessment and prediction of wind energy production. A primary issue is adequate understanding of the effects of wind variability, atmospheric stability, turbulence and air density variability on the wind turbine energy production. Non-negligible and quite often significant error is incurred when the effects of shear, TI, and atmospheric stability on the wind turbine power performance are ignored, as in the IEC standard, 61400-12-1 (2005). The standard procedures are valid only for ideal neutral conditions and a small wind turbine. Besides the dominant cubic dependence of the wind speed on the wind power density, there are smaller but still important corrections to the air density that are important to harvesting wind energy at high-elevation sites. Corrections that account for these factors must be included in the power output estimates, and more accurate predictions will help alleviate production-consumption imbalances. These imbalances can also be ameliorated through the use of storage devices. The field of wind resource assessment is evolving rapidly, responding to the increasingly stringent requirements of large-scale wind farm projects often involving investments of several hundred million dollars. Traditional cup anemometry is being complemented with ultrasonic sensors providing information on all three components of the wind velocity vector and enabling a better assessment of turbulence. Remote sensing devices like sodar and lidar are becoming more popular as turbine hub heights and rotor diameters increase, often placing the upper edge of the swept rotor area at heights of 130 m or more. While the traditional, conventional approaches of measuring the wind speed and direction at a few heights below hub height and extrapolating based on a logarithmic profile is still very common, the use of both vertical profiling devices and more accurate modeling tools considering the full terrain complexity and atmospheric stability is quickly moving into the mainstream. Measure-correlate-predict (MCP) methods are used to estimate wind speeds and directions at a target site where wind power is assessed for development. These methods use two sets of in-data. To begin with a series of measured speeds and directions from the target site during a period of time (usually one year) is needed. In addition to this, a reference series from a much longer period needs to be obtained. On the other hand, the advanced hybrid MCP method uses the recorded data of multiple reference stations to estimate the long-term wind condition at a target plant site. Because each reference station has the flexibility to use any of the available MCP techniques, the multiple reference weather stations were combined into the hybrid MCP strategy with the best suitable MCP algorithm for each reference station. Climate projections of wind resources in changing climate are a topic of a debate in the literature, requiring a thorough investigation of uncertainties and understanding the complex interaction of atmospheric dynamics. This will contribute to understanding the extent to which some of the predicted trends are the result of the weather and climate variability or the result of inadequate physical parameterizations in global and regional climate models.

*Entropy and Exergy in Renewable Energy*
