Geoscience Reference
In-Depth Information
production in the convective ABL (
∂/∂z
negative,
Chapter 11
) is from buoyancy; its
cross-isobaric angle thus tends to be smaller and its much more diffusive turbulence shifts
much of this adjustment to the interfacial layer.
• In the ABL the Coriolis terms in the turbulence second-moment budgets
(8.70)
,
(8.71)
tend to be small compared to the leading terms
(Problem 9.17)
; thus we'll neglect
them.
• The mean depth
h
of the ABL can range from tens of meters to a few kilometers. In
clear weather over land
h
typically has a diurnal cycle driven by the surface energy
budget. After sunrise
h
increases with time as surface heating drives buoyant convection;
h
typically reaches amaximum inmid to late afternoon. On a clear night amuch shallower,
stably stratified boundary layer develops at the surface in response to the surface cooling
through emitted radiation; the turbulence inwhat was the convective boundary aloft decays
(
Chapter 12
). The local, instantaneous, stable ABL top is the interface between turbulent
and nonturbulent air. This nocturnal ABL can be complicated by breaking gravity waves
that generate their own turbulence.
• Buoyancy effects make the convective and stable ABLs strikingly different. The neutral
(zero virtual temperature flux, mean virtual potential temperature constant with height)
ABL is rare because small virtual temperature differences in the ABL can cause large
buoyancy effects. Numerical simulations suggest that (given enough time, measured
in units of 1
/f
) an equilibrium neutral ABL can establish its own depth that scales
with
u
√
|
/ρ
. However, below this height in the atmo-
sphere there is often an inversion that establishes the ABL depth by extinguishing the
turbulence.
• In clear weather over land the mean wind speed in the surface layer can have a diurnal
cycle of substantial amplitude, with higher speeds in unstable, daytime conditions and
lower speeds in the stable conditions at night.
• The Reynolds number
R
t
of ABL turbulence is far larger than that of laboratory or com-
putational turbulence. By Reynolds-number similarity (
Chapter 2
) the energy-containing
structure of the ABL is believed not to be substantially different from that of a lower-
R
t
version, so laboratory simulation, also called
fluid modeling
, can be an effective
tool for studying the large-eddy structure of the ABL. But the difference in
R
t
causes
the fine structure of ABL turbulence to differ strongly from that of laboratory flows
(
Chapter 7
).
• The large time scales and high fluctuation levels of ABL turbulence can cause the
required averaging times (
Chapter 2
) for ABL statistics to exceed the times over which
the ABL can be considered quasi-steady. This tends to make the scatter in ABL turbulence
measurements larger than in engineering flows.
• The near-surface portion of the ABL, the
surface layer
, is the best understood, in part
because measurements are easiest to make there. But also its structure and dynamics
are dominated by local interactions with the surface, which minimizes the influence of
horizontal inhomogeneity and nonstationarity.
/f
, where
u
∗
=
surface stress
|
∗
We shall elaborate on these points in detail in this and the following chapters.
