Wind has become an important element of climate for consideration due to the its growing presence as a renewable source of energy. Variations in wind create uncertainty which adversely impacts investment and planning decisions and can lead to structural damage to equipment when extreme events occur which are not able to be adequately planned for. Long-term wind variations are related to climate variability over periods ranging from months to years, with the so-called teleconnections possibly driving significant changes in power output. The goal of this study was to assess the potential impacts of modes of variability within the climate system on wind energy output in the Upper Midwest (UMW: 40-52°N, 87-105°W), a North American region rich in wind resources and experiencing rapid turbine deployment. First, to facilitate this goal, the representation of wind resources by reanalysis models was tested, as were methods of extrapolating 10-meter wind speeds to heights more common of wind turbines (hub-height, often around 80-100 meters). Reanalyzed wind fields were found to capture many of the mean, variational and distributional characteristics of wind speeds at 10-meters as measured by weather stations, though declining trends in the observations were not found to be accurately replicated in the reanalysis models. Next, four methods of wind speed extrapolation commonly used in the literature were tested for their capacity to capture the mean and variations in wind speeds from tall towers measuring at heights ranging from 39-100 meters above ground level. Each method was applied to four reanalyses and results compared against tall tower data. All of the method-reanalysis combinations produced wind speeds which were too slow than observed and less variable than those measured at the tall towers, though the variable exponent power rule applied to MERRA was able to achieve relatively close results with small mean biases. 80-meter wind fields were generated from MERRA using the variable exponent power rule for application in the final section of this study. The 80-meter wind fields were utilized to derive wind energy output. This power output data was then used in a multiple linear regression model to assess the influence of several teleconnections important to the UMW, as well as potential effects of solar forcing variations. This model was applied to each grid cell and season, allowing for spatial and temporal variations in the relationships between the modes of variability and power production to manifest. The magnitude and significance of the teleconnections and solar forcing vary throughout the year and across the region. These influences are shown to fit with expectations of flow set by sea level pressure anomaly patterns. Extreme monthly wind energy anomalies are explored, with the strongest extremes affecting most of the region simultaneously and negative power anomalies found to persist for periods of several months to a year. Negative power output episodes are shown to follow from a combination of synoptic and teleconnection-driven factors while strong, positive output episodes are mostly short-lived and the result of synoptic factors (favorable positions of high and low pressure and strong pressure gradients). These findings have important implications for long-term energy planning and have the potential to improve seasonal and interannual predictions for the industry.
University of Minnesota Ph.D. dissertation. August 2020. Major: Geography. Advisor: Katherine Klink. 1 computer file (PDF); xi, 151 pages.
Wind in the Upper Midwest: Assessing Wind Resource Variability and Representation in Reanalyses.
Retrieved from the University of Minnesota Digital Conservancy,
Content distributed via the University of Minnesota's Digital Conservancy may be subject to additional license and use restrictions applied by the depositor.