THE CLIMATE ADJUSTMENT PROCESS - Wind Resource Assessment: A Practical Guide to Developing a Wind Project - page 159

related to the timescale at which wind fluctuations may be experienced simultaneously
by the reference and target sites. If the interval is too short, then a large proportion
of the speed fluctuations may contain no useful information about the relationship
between the two stations; they are just noise. If the interval is too long, then important
information about the relationship may be lost.
The optimal time interval is, in turn, related to the size of typical weather distur-
bances and their rates of motion. As a rule of thumb, the duration of a wind “event,”
whether it is a gust occurring in a matter of seconds or a sustained period of high
winds lasting several days, approximately equals the size of the associated weather
disturbance (which may range from a small turbulent eddy to a large storm system or
front) divided by its speed relative to the observer. A wind fluctuation cannot occur
simultaneously at two points unless both are within the realm of influence of the same
disturbance. Thus, the shortest timescale over which correlated fluctuations can occur,

t (s), is, very approximately, the distance between the target and reference stations,
D (m), divided by the typical or average background wind speed, v (m/s):

D
v

t ≈ (12.2)

Suppose the typical mean wind speed is 7 m/s. Then the shortest reasonable timescale
to correlate stations that are, say, 100 km apart is about 14,285 s, or 4 h.
As a general guideline, when the reference is an ordinary surface weather station
located some distance away from the target tower, daily averaging serves well. This
has the advantage that it is simple to apply, and it reduces the influence of differences
in diurnal wind speed patterns related to tower height and station location (which
can also introduce noise in the correlation). The only time a shorter interval such
as 1 h or 10 min might reasonably be used is when the reference station is within a
few kilometers of the target site. This occurs most often when secondary masts are
correlated with a primary mast within the project area.

12.2.2 Homogeneous Wind Speed Observations
The requirement for a long, homogeneous reference data record can also be difficult to
meet. One problem is that measurement standards change from time to time as national
weather agencies seek to improve their measurement technology and data products.
Unfortunately, this runs counter to the wind industry's interest in having consistent,
long-term wind data sets. In the United States, for example, almost all leading weather
stations were converted to the ASOS standard in the middle to late 1990s and early
2000s. In the process, tower heights changed from (typically) 6.1 to 10 m, many
towers were moved, and the previous manual recording technology was replaced by
automated digital equipment. The result was a substantial discontinuity in the recorded
wind speeds, which rendered data collected before ASOS effectively useless for MCP.
More recently, cup anemometers, which have been the standard for many decades at
US weather stations, were replaced in the United States by ultrasonic “ice-free wind”,
or IFW, anemometers, resulting in another disruption in the continuity of wind speeds.