Spectral analysis of long term measurements of wind and turbulence from tall masts –land and sea based.

Research output: Contribution to conferenceConference abstract for conferenceResearchpeer-review


In a continuing study we are analyzing wind speed data from tall masts with the purpose of providing answers to some of the outstanding questions concerning the flow in the atmospheric boundary layer. The right answers are
essential for one of the NEWA key activities: The building of the model chain and its use for computing the high resolution time series at every grid point which will be an important part of the final wind atlas database. The dataset
then allows us to investigate one of the most crucial issues in the concept of the model chain namely that of linking the two-dimensional mesoscale flow models to the three-dimensional microscale turbulence models.
The analysis is based on data from five sites: The land based data comes from the two Danish test stations for large wind turbines: Høvsøre and Østerild. The offshore data comes from the wind farms Horns Rev M2 and Horns Rev
M8 and from the German research platform FINO3. We are mainly performing spectral analysis on the extensive mean meteorological cup anemometer data and high frequency sonic anemometer data. In the first part of our study
we aimed at establishing the full scale spectrum of the boundary layer wind using the data from the 100m mast at Høvsøre and the 62m mast at Horns Rev M2. The results from the study were published in¹. One of the important
conclusions was: The spectral gap in the horizontal wind power spectrum exists and can be modeled. The linear composite of the wind variations from the mesoscale and microscale gives the observed power spectrum in the gap
range. Depending on the relative contribution to the variation from the microscale and mesoscale, the gap may be visible or invisible.
In the next phase of the current study, data from the Østerild mast is adding to our knowledge on the variation with height of the mesoscale and turbulence flows, as we have extended the analysis from 100 m to 241 m. Of a special
interest is the variation with height of the spectral gap, the turbulence quantities and the mesoscale spectra. The Østerild study also complements the Horns Rev (off-shore) and the Høvsøre (coastal) study in the sense that
Østerild shows clearly land based properties, such as a non-disappearing diurnal peak in the spectra for all the heights. Then a part of this study is devoted to investigate the variation with height of the diurnal cycle of the
horizontal wind speed; the difference between land and sea and the special variation over land with a minimum at around 80 – 100 meter and a continuing increase with height after the minimum. The study provides an explanation
of these variations. Clearly, it is relevant to examine if the NEWA modeled time series can reproduce the daily variations correctly. We are further analyzing the impact of the frequent occurrence of cellular structures over Northern Europe in the atmospheric boundary layer on the spectral properties, following often cold polar outbreaks. Open cells have a tendency to fill up the spectral gap and as a consequence, in such situations, make load calculations based on contemporary turbulence model schemes questionable. We also attempt to use the full lateral spectrum to estimate wake meandering for different atmospheric conditions, using reasonable estimates of β, the ratio between the Lagrangian – Eulerian timescales of correlation.
Original languageEnglish
Publication date2017
Publication statusPublished - 2017
EventWind Energy Science Conference 2017 - Kgs. Lyngby, Denmark
Duration: 26 Jun 201729 Jun 2017


ConferenceWind Energy Science Conference 2017
CityKgs. Lyngby
Internet address

Cite this