Abstract
The coastal Otago community of Waitati (45°45”S, 170°34”E, 19 km north of Dunedin) is considering renewable energy sources to help offset their electricity consumption. The Waitati Energy Project is focussed on the installation of a wind turbine for energy generation, and also for a visual testament to community achievement. The purpose of this research was to investigate the climatological viability of wind power in the Waitati area, with the assistance of boundary layer theory for understanding the interactions between synoptic scale air flow and land surface influences. This study utilised a combination of field work and numerical modelling to understand the local wind regime by characterising its spatial and temporal variability, determining the atmospheric and topographical influences on wind speed and direction, and predicting suitable locations for wind turbine sites. For the observations, twelve months of wind data from two automatic weather stations in Waitati (AWS1 and AWS2, located to the east and west of the township), as well as a 35 year wind record from Taiaroa Head were used. The numerical modelling was generated by the meteorological component of The Air Pollution Model (TAPM), to give three-dimensional numerical predictions of the Waitati wind regime at a 1 km resolution.
The average wind speed in Waitati over the study period (June 5 2009 to May 31 2010) was 4.34 m s-1 at AWS1 and 2.46 m s-1 at AWS2. This variation in average wind speed was attributed to a sheltering effect at AWS2. At the more exposed Taiaroa AWS, average wind speeds reached 6.71 m s-1. The 35 year record of average wind speed from Taiaroa Head allowed the data from the study period to be put into a longer temporal context. Seasonal analysis of the observational data revealed spring and autumn as the windiest periods, with west to southwest airflow the predominant flow direction. These data were supplemented by seasonal TAPM predictions across the Waitati area. Areas of higher elevation were more exposed to high wind speeds, and the eastern side of the Waitati valley was largely sheltered from gradient winds by high topography to the south and west. Week-long case study analyses of cyclonic and anticyclonic systems revealed the ability of TAPM to predict wind speeds in different conditions. During the cyclonic conditions wind speeds were well predicted for Waitati, and the model estimated topographical channelling of the southerly through the hills to the south of Waitati. During the anticyclonic conditions nocturnal drainage flows and daytime sea breezes were predicted to modify the weak synoptic northeasterly wind.
Of the four wind turbines featured in this research, the three larger, commercial ones were found to be climatologically unviable for this low wind speed environment. Their cut-in wind speeds of 4 to 5.5 m s-1 were so close to the (observed and modelled) average annual wind speed that it could be predicted that these turbines would only be operational for half of any given year. In contrast, the locally designed Thinair102 turbine, which is made for operation at the household level, would be more suitable with a cut-in wind speed of 3.5 m s-1.