Project

EUROS

Excellence in Uncertainty Reduction of Offshore wind Systems

EUROS is a STW-funded project aimed at finding ways to reduce the costs of design, construction and logistics of installation and maintenance of off-shore wind energy installations. The project is divided into three sub-projects. KNMI is involved in the one focusing on external conditions, especially the weather. For instance, better (i.e., more reliable) weather forecasts can improve the determination of maintenance windows, and a better understanding of long-term variability can improve yield estimates.

Together with WUR (Wageningen University & Research) KNMI is investigating the long- and short-term variability of the wind and wave fields on the North Sea and their predictability. Research questions are:

  1. How predictable are wind and wave variations in the North Sea?
  2. What are observed wind and wave variations across time scales in the North Sea?
  3. How does it vary across data sets?
  4. What are the prediction horizons?
  5. How is local variability related to large scale (more predictable) patterns of variability in the atmosphere?

During the first year of the project the focus was on points 2, 3 and 5. Using data from 20CR (Twentieth Century Reanalysis, Compo et al. 2011) and ERA-20C (Poli et al. 2016) the long-term variability of the wind field on the North Sea and its link with large-scale variability patterns was investigated. The results where compared to those from modern-time reanalyses (MERRA, Rienecker et al. 2011, and ERA-Interim, Dee et al. 2011).

Figure 1 shows that the averaged wind on the North Sea exhibits large interannual variations around a level of about 7.6 m/s, except for a period in the 1990s where the level is remarkably higher. Interannual variability is consistent between different reanalyses, but absolute values are different. Furthermore, ERA-20C shows a clear increase in wind speed during the 20th century, while 20CR shows no trend. The trends in ERA-20C are increasing with wind speed. They are higher for higher percentiles (Figure 2) and higher during winter than during summer (not shown). The reasons for the discrepancies between the reanalyses are currently under investigation.

Next steps in the research are:

  • Assess predictability of wind and waves (probabilistically) for lead times up to 9 months
  • Forecast calibration by combing forecasts and observations a posteriori
  • Compare high resolution reanalyses products (HARMONIE, WRF) with other products

 

Figure 1: Comparison of 10m wind speed over the North Sea from two reanalyses spanning the whole 20th century, but assimilating only a limited set of observations, and two modern-time reanalyses (MERRA and ERA-Interim) that use all available observations.
Figure 1: Comparison of 10m wind speed over the North Sea from two reanalyses spanning the whole 20th century, but assimilating only a limited set of observations, and two modern-time reanalyses (MERRA and ERA-Interim) that use all available observations.
Figure 2: Time series of different percentiles of North Sea averaged wind speed.
Figure 2: Time series of different percentiles of North Sea averaged wind speed.