Environmental Engineering Reference
In-Depth Information
Figure 7: Mean energy in wind for consecutive 5-year periods based on a time
series from Hesselø, Denmark, 1873-1982 [17].
variability in Northern Europe shows that variations in wind energy of up to 30%
can be expected from one decade to another (see Fig. 7). In another study [15] it
was found from an analysis of the expected power output for a 45 m high wind
turbine over a 22-year period that the inter-annual variation in power corresponds
to a mean relative standard deviation of approximately 13%. For the proper assess-
ment of the economics of wind power utilisation, such variability must obviously
be borne in mind [16].
Any short-term measurement should therefore be correlated with long-term
data covering ideally 30 years from a representative station. Unfortunately it is
very diffi cult in most cases to obtain reliable, consistent long-term data since
weather stations are frequently subject to changes in environment by, for exam-
ple, growing vegetation or building activity. Therefore - and particularly in
very remote areas - the use of re-analysis data might be a feasible alternative.
Most commonly used is NCEP/NCAR data from NCEP (National Centre for
Environmental Prediction) and NCAR (National Center for Atmospheric
Research) in the US. These institutes continuously perform a global re-analysis
of weather data. The objective is to produce homogeneous data sets covering a
decade or more of weather analysis with the same data assimilation systems.
This means that data from synoptic weather stations, radio sondes, pilot bal-
loons, aircraft, ships, buoys and satellites are collected, controlled, gridded, and
prepared for initialisation of a global numerical weather prediction model. By
running past-periods on a global scale, the data is re-analysed in a globally
consistent manner which enables it to be used in some situations for long-term
correlation purposes.
The advantages are that the period of available re-analysed data can be much
longer than the period of available on-site data, the data is neither subject to icing
Search WWH ::




Custom Search