Geoscience Reference
In-Depth Information
with a height-dependent
a
.Emeis(
2004
) puts
a
1.6 in a lower layer that includes
the surface layer (he chooses 55-200 m for this layer),
a
=
=
2.0 for a middle layer
(200-600 m), and a
2.5 for an upper layer (600-1000 m). The approximation
(
4.22
) to turbulent viscosity (
4.16
) has been made because (
4.22
) can be com-
puted from SODAR data without the need of additional information. This offers
the opportunity to determine the diurnal course and the vertical structure of turbu-
lent viscosity and to estimate the range of values this quantity can take from longer
time series of ground-base
d me
asurements with a SODAR. Because a quadratic
expression (
=
w
) instead of
u
w
is used in (
4.22
), the information about the sign
of the turbulent momentum flux is lost in (
4.22
). In order to assure the positive-
ness of the turbulent viscosity, only the absolute value of the shear must be used
in (
4.22
).
Figure
4.30
shows a time series of the t
ur
bulent viscosity for two different heights
above ground over 12 days together with
∂
u
∂
2
σ
σ
w
, which have been used to derive
this variable. The most prominent feature in Fig.
4.30
is the daily cycle of the magni-
tude of the turbulent viscosity with a maximum around noon and a minimum around
midnight. This daily cycle is not only due to such a cycle in
z
and
σ
w
but especially in
Fig. 4.30
Time series of turbulent viscosity in m
2
s
−
1
(top), standard deviation of the vertical
wind component in m s
−
1
(middle) and vertical shear of the horizontal wind in 1 s
−
1
(below) for
two height ranges (55
200 m above ground:
full lines
), 200-600 m above ground:
dashed lines
)
from SODAR measurements using eq. (
4.22
).
−
Search WWH ::
Custom Search