Characteristics of the Wind - Wind Turbine Technology: Fundamental Concepts of Wind Turbine Engineering - page 478

many parts of the world are either not available or not nearly as comprehensive as those
for the U.S. A preliminary world-wide wind resource assessment based on statistical and
subjective analyses is given by Elliott et al. [1981]. Global pressure and wind patterns, up-
per-air wind data, boundary layer meteorology, and assumed annual frequency distributions
were used to obtain consistent estimates of the wind energy resource on a planetary scale.
By necessity, such assessments are limited in detail. However, this survey does give a rough
indication of the world's mean annual wind energy resource. The reference elevation used
for this assessment was 50 m, which is high enough above the ground for the wind to be rela-
tively independent of minor surface features and directly useable for medium- and large-scale
turbines. At the same time, it is close enough to the ground to allow meaningful interpolations
for small-scale machines.
The results of this preliminary world-wide wind energy resource assessment have been
summarized on a Molleweide map projection that shows the estimated distribution of the
seven wind energy classes defined in Table 8-1. As expected, the map shows greater
complexity over land than over sea and generally reflects only large-scale patterns. Much
greater spatial variability exists than is depicted, and the same is true for variations in time,
since the map only gives annual mean estimates. Because of these limitations, this map
should be used only for its intended purpose: to provide wind power project planners with
rough estimates of the mean annual wind energy resource in countries other than the U.S.
Analysis and Assessment Methodologies
Three basic methods have been used in wind energy resource assessments: (1) statistical
and subjective analysis of existing wind measurements, other meteorological data, and
topographical information [Elliott et al. 1986]; (2) qualitative indicators of long-term wind
speed levels [ e.g. Putnam 1948]; and (3) application of boundary layer similarity theory and
the use of surface pressure observations [Petersen et al. 1981].
Statistical and Subjective Analysis Methodology
This method depends on the availability of wind measurements representing a wide
variety of geographical, topographical, and climatological conditions. As an example, the
surface wind data which are the basis of the Wind Energy Resource Atlas of the United
States [Elliott et al. 1986] were obtained from a wide variety of sources including the
National Climatic Data Center (NCDC), the U.S. Forest Service, university research
projects, and power plant sites. Data sets from almost 7,000 recording stations were
screened, with almost half retained for further analysis. In unpopulated areas, adequate
summarized data are often not available, and an effort must be made to identify other
sources, such as private organizations and other government agencies. These data
frequently exist as unreduced strip-chart records or as partial compilations of hourly data
records collected for very specialized purposes. Formats are not always suitable for a wind
energy assessment, and the adequacy of the data set should be reflected in its certainty
rating.
In general, wind data in summarized or digitized formats are preferred. For stations
where both are available, the digitized data can be used to improve the existing summaries.
Analyses of wind speed records for periods with constant anemometer elevation, locations,
and observation frequency are more useful than routine summaries. For stations having
several different types of summarized wind data covering various time periods, one or two
of the better summaries for those stations should be selected considering