VERTICAL WIND COMPONENT
129
(it) there is no separation of the points according to height or site;
(h) there is a clear relation between c~,/u+, and z/L,; and
(c) there is an approach to a one-third power law at large values of
Those points are very important as they confirm the use of ziL, as a stability
parameter and point to the relevance of local parameterisation even when
z * 1 km. They also tend to confirm the use of z as a scale length up to this
altitude. However it is important to bear in mind that none of the measurements used in this paper were made in or above an inversion.
The two lines previously discussed are also included in Fig. 2 using the
same notation as in Fig. 1. It can be seen that though they represent the
general trend of the data when 1 -z/L, I > 20. one tends to lie above
the points while the other tends to lie below them. This was confirmed by
regressing the two variables for various minimum values of I -z/L, I and it
was found that the relation
I - 4 L , I.
represented the data quite well when 1 -z/L, I 2 20; this line has also been
included in Fig. 2. Extending the range of acceptable stabilities down to
I -z/Ll I 2 I or 10 reduces the power law to - 0.29 and increases the
constant of proportionality. Thus it appears that the convective limit will
not be attained unless I -z/LI 1 2 20; for smaller values the slope will
decrease and the constant of proportionality increase, as is apparent in
Fig. 1 of Wyngaard er al. (1971).
The line corresponding to Eq. ( 5 ) has also been included in Fig. 1. and it
can be seen that the more unstable data scatter quite well about it.
3. CONCLUSIONS
The results presented in this paper have clearly established the dependence of 0, on the local state of the atmosphere and have indicated the
existence of a convective limit when I -z/L, I 2 20. These conclusions
appear valid up to heights - 1 km provided an inversion is not present. The
value of the constant of proportionality in the convective limit AS''3 appears
to lie in the range 1.10-0.96. If these values are combined with the neutral
limit of rr,/uII ( = A / I c " ~ ) , the value of (5 may be derived. Pasquill (1972)
recently tabulated the various estimates of this neutral limit and it appears
that it value of 1.3 is reasonable. With k = 0.4. this gives
If0 I 6 5 1.5
which in thc lower limit may imply that convection and wind shear are
L
129
(it) there is no separation of the points according to height or site;
(h) there is a clear relation between c~,/u+, and z/L,; and
(c) there is an approach to a one-third power law at large values of
Those points are very important as they confirm the use of ziL, as a stability
parameter and point to the relevance of local parameterisation even when
z * 1 km. They also tend to confirm the use of z as a scale length up to this
altitude. However it is important to bear in mind that none of the measurements used in this paper were made in or above an inversion.
The two lines previously discussed are also included in Fig. 2 using the
same notation as in Fig. 1. It can be seen that though they represent the
general trend of the data when 1 -z/L, I > 20. one tends to lie above
the points while the other tends to lie below them. This was confirmed by
regressing the two variables for various minimum values of I -z/L, I and it
was found that the relation
I - 4 L , I.
represented the data quite well when 1 -z/L, I 2 20; this line has also been
included in Fig. 2. Extending the range of acceptable stabilities down to
I -z/Ll I 2 I or 10 reduces the power law to - 0.29 and increases the
constant of proportionality. Thus it appears that the convective limit will
not be attained unless I -z/LI 1 2 20; for smaller values the slope will
decrease and the constant of proportionality increase, as is apparent in
Fig. 1 of Wyngaard er al. (1971).
The line corresponding to Eq. ( 5 ) has also been included in Fig. 1. and it
can be seen that the more unstable data scatter quite well about it.
3. CONCLUSIONS
The results presented in this paper have clearly established the dependence of 0, on the local state of the atmosphere and have indicated the
existence of a convective limit when I -z/L, I 2 20. These conclusions
appear valid up to heights - 1 km provided an inversion is not present. The
value of the constant of proportionality in the convective limit AS''3 appears
to lie in the range 1.10-0.96. If these values are combined with the neutral
limit of rr,/uII ( = A / I c " ~ ) , the value of (5 may be derived. Pasquill (1972)
recently tabulated the various estimates of this neutral limit and it appears
that it value of 1.3 is reasonable. With k = 0.4. this gives
If0 I 6 5 1.5
which in thc lower limit may imply that convection and wind shear are
L