297 Statistical and Synoptic Analysis of Offshore Wind Variations

Monday, 23 January 2017
Talaat Bakri, University of Northern British Columbia, PRINCE GEORGE, BC, Canada; and P. L. Jackson

We present an analysis of offshore wind variations at different locations in the northern and central coasts of British Columbia. The study focuses on the storms activity at 10m height as well as on the wind variability at the wind turbine hub height. The variations of winds at both heights is then explained by a synoptic climatology analysis.

Wind speed observations from nine buoys measured at 3m height, spanning 22 years, is extrapolated to both 10m and 100m heights. Monin-Obukhov theory is used to predict the wind speed profile by estimating the Obukhov length using the bulk Richardson number method. The extrapolated wind profile shows a constant wind shear in unstable atmospheric conditions and a progressively increasing shear in neutral and stable conditions. The occurrence of persistent wind speed above a threshold of 11 m s-1 (14 m s-1) at 10m is 30% (8%) of time at durations up to 4 h, and above 10% (3%) of time at durations up to 12 h. At the 100m height, winds above cut-in wind speed occurs around 70% (60%) of time for durations up to 4 h (12 h). Less than 1% of time winds exceed the cut-out wind speeds for any given duration at the 100m height. Wind speeds at 10m height are analyzed to determine extreme values corresponding to a 5-yr and 50-yr return periods, which addresses the storm’s variability over long periods. Generalized Pareto distribution (GPD) using peaks-over-threshold method is applied in the extreme wind speed values analysis. The GPD goodness of fit was evaluated using several methods including Kolmogorov-Smirnov test, which indicates that the tested distribution has a good fit. A 5-yr return level exceeds 33 m s-1 at most locations, while a 50-yr return level ranges between 40 m s-1 to 46 m s-1 at some locations and exceeds 50 m s-1 at two locations.

The Spatial coherency of extreme wind speed events between all buoys is examined to investigate the contribution of the synoptic scale conditions in initiating such events. The spatial coherence is qualified by selecting three reference buoys and a wind speed threshold, and then comparing it with observations from other buoys. The results reveal high coherency for most buoys implying that the synoptic scale is the main deriver of high wind speed events. Therefore, synoptic scale circulations are investigated and correlated to different wind speed categories at both 10m and 100m heights. Synoptic circulation patterns are generated using the 6 hourly NCEP-NCAR reanalysis mean sea level pressure data, principle components, and k-means clustering. The correlations between the synoptic patterns and the wind categories reveals that high wind speed events at both heights are mainly derived by synoptic patterns which characterized by a low pressure center over the Gulf of Alaska, and a mean sea level pressure gradient over the coast of British Columbia. Such synoptic patterns are more frequent from October to March. The within-type variability of these patterns shows dominant south to southeast winds at three different locations. Synoptic patterns responsible for wind speeds less than 4 m s-1 at 100m height are characterized by a high pressure system (the Pacific High) that extends from the southwest of the domain toward the northern and central coasts of British Columbia. The extended Pacific High is a quasi-permanent summertime feature.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner