Geoscience Reference
In-Depth Information
(
1987
). Some of the linear approaches are quite old and date back to work of, e.g.,
Jackson and Hunt (
1975
). These analytical approaches have always been accom-
panied by numerical efforts, see e.g., the work of Taylor (
1977
). Also the well-
known WAsP model is based on such linear analytical approaches (Troen and
Petersen
1989
).
4.2.1 Potential Flow
The simplest case for a description of flow over a hill is frictionless potential flow.
This implies a laminar flow of a non-viscous fluid with no surface friction. It is
presented here in order to present an analytical model that shows first-order effects
of flow over hills. The main feature is the speed-up of the wind speed over the hill,
a slight wind speed reduction upstream of the hill and a considerable reduction of
the wind speed over the downwind slope of the hill.
For a flow perpendicular to a two-dimensional ridge (i.e. a ridge which is
infinitely long in the direction perpendicular to the flow), the speed-up of the
potential flow over the hill can be described using the thin airfoil theory (Hoff
1987
):
Du
pot
ð
x
;
z
Þ¼
u
1
ð
L
Þ
H
L
;
z
x
L
r
ð
4
:
1
Þ
L
where x is the direction perpendicular to the ridge, z is the vertical coordinate, H is
the height of the ridge, L is the half-width of the ridge (the distance from the crest to
the place where the height is H/2), u
?
(L) is the scaling wind speed in the undis-
turbed flow at height L. Therefore, all heights in this simple model scale with L. r is
the form function of the ridge cross-section. H/L is the aspect ratio of the ridge and
describes the magnitude of the slope. Adding (
4.1
) to the undisturbed flow,
u
?
(z) yields for the wind profile in the potential flow over the ridge:
u
pot
ð
x
;
z
Þ¼
u
1
ð
z
Þþ
u
1
ð
L
Þ
H
x
L
;
z
L
r
ð
4
:
2
Þ
L
In contrast to all wind profile relations given in
Chap. 3
, the wind profile
relation (
4.2
) does not only depend on the vertical coordinate but also contains a
horizontal coordinate. The form function r can be given analytically as long as the
ridge cross-section h(x) can be described by the inverse polynom (see Fig.
4.4
):
¼
x
L
1
1
þ
L
2
h
ð
4
:
3
Þ
The associated form function r for this ridge cross-section (
4.3
) reads:
Search WWH ::
Custom Search