60
J.D. Albertson, G. Kiely and M.B. Parlange
profiles of mean meteorological quantities using Monin-Obukhov similarity theory, (ii) remote
sensing of ABL turbulence with lidar or scintillometry, or (iii) in situ point measurements
of turbulent . fluxes using eddy correlation or dissipation techniques in the ABL. For these
techniques to provide measurements that represent integrated fluxes over large spatial scales,
the measurements must be made above the blending height; here we rely on the thorough
mixing provided by the extremely high Reynolds number turbulent flow of the ABL. For a
complete discussion of the idea of a blending height and the relationship between measurement
height and spatial integration of surface fluxes see the recent review paper by Parlange et al.
(1995b).
Although the eddy correlation (EC) method provides a direct measurement of the vertical fluxes, there are applications for which it is inappropriate, particularly over water and for remote
field experiments requiring long term unattended operation. Fluxes from the mean profile method or the bulk aerodynamic method have limitations as well. The flux-dissipation method
is used widely over water surfaces, typically using power spectra to determine the dissipation
rates, followed by an iterative procedure to determine fluxes. However, this approach suffers
from uncertainties due to empirical inertial subrange constants, the jumpy (or noisy) nature of
power spectra, and the reliance on an iterative solution. In this paper we introduce a one-step
(non iterative) dissipation method for determining fluxes of momentum, sensible heat and water
vapor. This new method which uses third order structure functions to determine dissipation
rates, avoids many of the pitfalls previously assumed unavoidable with the dissipation method.
The results presented here may also improve the reliability of flux estimates from the ABL remote sensing instruments, such as lidar and scintillometers, since these techniques typically rely
on an empirical relationship between scalar fluxes and the dissipation rate of scalar variances
(see Hill et aI., 1992 and Eichinger et aI., 1993).
3.1.1 The Atmospheric Boundary Layer
The ABL is that layer of air, directly above the Earth's surface in which the effects of the
surface (friction, evaporation, heating and cooling) are felt directly on time scales less than a
day. The ABL is broadly made up of two layers, the atmospheric surface layer (ASL) and the
mixed layer (ML). The ASL is affected primarily by surface fluxes while the ML is affected
by both surface fluxes and boundary layer entrainment of the free atmosphere air from above.
Figure 3.1 is a schematic of the composition of the ABL. The ASL occupies about the lowest
10% of the fully developed daytime ABL, or approximately the first 100 m above the Earth's
surface (Parlange et aI., 1995b).
In the ASL, the turbulent flow is often assumed to be statistically stationary, when considering
periods of 10 minutes to 1 hour. The principle mechanism for the mechanical production of
turbulence in this time period is the vertical gradient of the mean wind. Figure 3.2 shows
schematics of the vertical profiles of the means of velocity U, potential temperature e, and
humidity q for the daytime convective boundary layer (CBL). It is seen from Figure 3.2, that
the steepest gradients occur in the lower 10% of the boundary layer (i.e. the ASL), while in the
upper 90% of the CBL strong convective mixing and entrainment of the free atmosphere air at
the top of the ABL smooth out almost all vertical variations in the mean profiles.
J.D. Albertson, G. Kiely and M.B. Parlange
profiles of mean meteorological quantities using Monin-Obukhov similarity theory, (ii) remote
sensing of ABL turbulence with lidar or scintillometry, or (iii) in situ point measurements
of turbulent . fluxes using eddy correlation or dissipation techniques in the ABL. For these
techniques to provide measurements that represent integrated fluxes over large spatial scales,
the measurements must be made above the blending height; here we rely on the thorough
mixing provided by the extremely high Reynolds number turbulent flow of the ABL. For a
complete discussion of the idea of a blending height and the relationship between measurement
height and spatial integration of surface fluxes see the recent review paper by Parlange et al.
(1995b).
Although the eddy correlation (EC) method provides a direct measurement of the vertical fluxes, there are applications for which it is inappropriate, particularly over water and for remote
field experiments requiring long term unattended operation. Fluxes from the mean profile method or the bulk aerodynamic method have limitations as well. The flux-dissipation method
is used widely over water surfaces, typically using power spectra to determine the dissipation
rates, followed by an iterative procedure to determine fluxes. However, this approach suffers
from uncertainties due to empirical inertial subrange constants, the jumpy (or noisy) nature of
power spectra, and the reliance on an iterative solution. In this paper we introduce a one-step
(non iterative) dissipation method for determining fluxes of momentum, sensible heat and water
vapor. This new method which uses third order structure functions to determine dissipation
rates, avoids many of the pitfalls previously assumed unavoidable with the dissipation method.
The results presented here may also improve the reliability of flux estimates from the ABL remote sensing instruments, such as lidar and scintillometers, since these techniques typically rely
on an empirical relationship between scalar fluxes and the dissipation rate of scalar variances
(see Hill et aI., 1992 and Eichinger et aI., 1993).
3.1.1 The Atmospheric Boundary Layer
The ABL is that layer of air, directly above the Earth's surface in which the effects of the
surface (friction, evaporation, heating and cooling) are felt directly on time scales less than a
day. The ABL is broadly made up of two layers, the atmospheric surface layer (ASL) and the
mixed layer (ML). The ASL is affected primarily by surface fluxes while the ML is affected
by both surface fluxes and boundary layer entrainment of the free atmosphere air from above.
Figure 3.1 is a schematic of the composition of the ABL. The ASL occupies about the lowest
10% of the fully developed daytime ABL, or approximately the first 100 m above the Earth's
surface (Parlange et aI., 1995b).
In the ASL, the turbulent flow is often assumed to be statistically stationary, when considering
periods of 10 minutes to 1 hour. The principle mechanism for the mechanical production of
turbulence in this time period is the vertical gradient of the mean wind. Figure 3.2 shows
schematics of the vertical profiles of the means of velocity U, potential temperature e, and
humidity q for the daytime convective boundary layer (CBL). It is seen from Figure 3.2, that
the steepest gradients occur in the lower 10% of the boundary layer (i.e. the ASL), while in the
upper 90% of the CBL strong convective mixing and entrainment of the free atmosphere air at
the top of the ABL smooth out almost all vertical variations in the mean profiles.