Using Mesoscale Meteorological Models to Assess Wind Energy Potential
Thesis DisciplineEnvironmental Sciences
Degree GrantorUniversity of Canterbury
Degree NameMaster of Science
As the demand for safe and clean electricity increases, the New Zealand wind energy industry seems poised to expand. Many generating companies have projects in the planning stage and there are likely to be many more potential sites yet to be identified. Reliable wind climate predictions over a wide area and for different heights above grounds are often vital to determine the viability of wind farm projects. This study investigates the use of meteorological mesoscale models to determine the wind and energy resource, particularly in areas of complex terrain. Complex terrain environments are likely to be typical of where New Zealand wind energy developments will take place. Using the prognostic mesoscale meteorological model TAPM (The Air Pollution Model), regions of relatively high mean wind speed were identified for a number of regions, including Banks Peninsula and parts of Canterbury and Otago. The simulations were conducted for a one-year period (2001) and at different heights above ground level. Depending on the resolution of the model calculations, speed-up effects from the forcing of some topographic features were accounted for by this model. Where the modelling was considered reliable, hourly wind data were obtained from grid points within the inner grid and used as input data for the industry-standard wind energy assessment model WAsP (The Wind Atlas Analysis and Application Program). As WAsP is able to account for detailed topography and surface roughness features, wind and energy predictions at a specific site or over a wider area surrounding the site were made. Limitations of both models in complex terrain were identified. These limitations were due to a number of factors, including the grid spacing used for mesoscale model calculations, the complexity of the terrain, and difficulties in modelling some regional scale airflow regimes. Being aware of when and where model limitations are likely to occur is important in being able to overcome and account for them.