heat (LH) to the mean parameters measured in the
atmospheric surface layer.
:C d (u9u s )
(3.4.1)
SH:C h (T9T s )
(3.4.2)
LH:LE:C e u(q9q s )
(3.4.3)
where is the surface air density, L is the latent
heat of vaporization, and E is evaporation. u, T
and q are the equivalent neutral wind speed,
temperature and specific humidity at a reference
height (usually taken to be 10 m). The subscript ‘s’
indicates the values at the ocean surface. The
transfer coefficients of momentum, heat and moisture under neutral stability, C d , C h and C e , have
been determined through fitting field measurements or solving flux–profile relations (e.g. Liu
et al., 1979; Large and Pond, 1982; Bradley et al.,
1991; Smith et al., 1992; DeCosmo et al., 1996;
Fairall et al., 1996). Charnock (1955) postulated that C d is a function of wind stress in a
well-developed sea and Donelan et al. (1997) and
others have found that C d depends on sea state.
Wind, temperature and specific humidity at any
height in the atmospheric surface layer under any
stratification can be converted to u, T and q, using
the flux–profile relations (e.g. Liu and Tang,
1996). In general, u s is assumed to be zero and q s
is the saturation humidity at T s multiplied by 0.98,
to account for the reduced-saturation specific
humidity over salt water.
Atmospheric stratification is difficult to estimate from space, but the enterprise is saved by the
roughness of the sea being a function of the wind
stress. As long as the original satellite calibration
for wind speed is done for neutral atmospheric
stratification, our stress estimates from satellite
measurements (equivalent neutral wind) will be
correct using the bulk formulae. This fortuitous
advantage is to some extent carried over to the
sensible and latent heat flux estimates. In any case,
over most of the ocean, the atmospheric temperature is closely adjusted to the sea surface temperature, which implies near-neutral stratification. In
addition, whenever the wind speed is greater than
10 m s
91
, the atmosphere approaches neutrality
since the mechanical turbulence dominates over
the buoyancy forced turbulence (e.g. Kraus and
Businger, 1994).
3.4.3 Wind forcing
There are several sets of space-based global wind
fields available now and for the future. The Special
Sensor Microwave/Imager (SSM/I) operates at 19,
22, 37 and 85 GHz in dual polarization (except for
22 GHz, which has only vertical polarization), on
the polar-orbiting operational spacecraft of the
Defense Meteorological Space Program (DMSP).
It has provided continuous wind speed measurements over the global ocean since July 1987.
Recently, several DMSP satellites have been in
orbit at the same time, providing complete daily
coverage with ever better algorithms based on the
long data record (e.g. Wentz, 1996; Krasnopolsky
et al., 1999). SSM/I has a wide scan and, therefore,
good coverage (1400 km) (Fig. 3.4.1, see Plate
3.4.1, p. 172), but does not provide information
on the surface wind direction. The wind speed
from SSM/I has been combined with surface wind
data from a numerical model through a variational
method to produce wind vector fields at 6-hour
intervals and at 2° by 2.5° resolution (Atlas et al.,
1996). These large-scale wind fields over the tropical Pacific were evaluated by Busalacchi et al.
(1993) by comparison with operational Numerical
Weather Prediction (NWP) products and interpolated wind fields from ship measurements and
cloud motions. They found large-scale similarity
between SSM/I and other wind fields, but suggested that the dense space–time coverage of
SSM/I offers a distinct advantage in forcing ocean
circulation models. The SSM/I wind fields were
found by Liu et al. (1996) to generate more realistic anomalous ocean cooling in an ocean general
circulation when compared with the results simulated by the same model but forced by NWP winds.
The SSM/I wind fields have more structure and
energy than NWP winds but have the same directional characteristics as the NWP winds. SSM/I is an
operational sensor that will be operated well into the
twenty-first century. The production of wind vector
data from these assimilated microwave radiometer
measurements, however, is a research effort.
Scatterometers send microwave pulses to the
earth’s surface and measure the backscattered
power (so-called radar cross-section) from the
surface roughness. Over the ocean, the backscatter
depends on ocean surface roughness due to small
(centimetre scale) waves. The idea of remote sensing of ocean surface winds was based on the belief
SECTION 3 NEW WAYS OF OBSERVING THE OCEAN
174
atmospheric surface layer.
:C d (u9u s )
(3.4.1)
SH:C h (T9T s )
(3.4.2)
LH:LE:C e u(q9q s )
(3.4.3)
where is the surface air density, L is the latent
heat of vaporization, and E is evaporation. u, T
and q are the equivalent neutral wind speed,
temperature and specific humidity at a reference
height (usually taken to be 10 m). The subscript ‘s’
indicates the values at the ocean surface. The
transfer coefficients of momentum, heat and moisture under neutral stability, C d , C h and C e , have
been determined through fitting field measurements or solving flux–profile relations (e.g. Liu
et al., 1979; Large and Pond, 1982; Bradley et al.,
1991; Smith et al., 1992; DeCosmo et al., 1996;
Fairall et al., 1996). Charnock (1955) postulated that C d is a function of wind stress in a
well-developed sea and Donelan et al. (1997) and
others have found that C d depends on sea state.
Wind, temperature and specific humidity at any
height in the atmospheric surface layer under any
stratification can be converted to u, T and q, using
the flux–profile relations (e.g. Liu and Tang,
1996). In general, u s is assumed to be zero and q s
is the saturation humidity at T s multiplied by 0.98,
to account for the reduced-saturation specific
humidity over salt water.
Atmospheric stratification is difficult to estimate from space, but the enterprise is saved by the
roughness of the sea being a function of the wind
stress. As long as the original satellite calibration
for wind speed is done for neutral atmospheric
stratification, our stress estimates from satellite
measurements (equivalent neutral wind) will be
correct using the bulk formulae. This fortuitous
advantage is to some extent carried over to the
sensible and latent heat flux estimates. In any case,
over most of the ocean, the atmospheric temperature is closely adjusted to the sea surface temperature, which implies near-neutral stratification. In
addition, whenever the wind speed is greater than
10 m s
91
, the atmosphere approaches neutrality
since the mechanical turbulence dominates over
the buoyancy forced turbulence (e.g. Kraus and
Businger, 1994).
3.4.3 Wind forcing
There are several sets of space-based global wind
fields available now and for the future. The Special
Sensor Microwave/Imager (SSM/I) operates at 19,
22, 37 and 85 GHz in dual polarization (except for
22 GHz, which has only vertical polarization), on
the polar-orbiting operational spacecraft of the
Defense Meteorological Space Program (DMSP).
It has provided continuous wind speed measurements over the global ocean since July 1987.
Recently, several DMSP satellites have been in
orbit at the same time, providing complete daily
coverage with ever better algorithms based on the
long data record (e.g. Wentz, 1996; Krasnopolsky
et al., 1999). SSM/I has a wide scan and, therefore,
good coverage (1400 km) (Fig. 3.4.1, see Plate
3.4.1, p. 172), but does not provide information
on the surface wind direction. The wind speed
from SSM/I has been combined with surface wind
data from a numerical model through a variational
method to produce wind vector fields at 6-hour
intervals and at 2° by 2.5° resolution (Atlas et al.,
1996). These large-scale wind fields over the tropical Pacific were evaluated by Busalacchi et al.
(1993) by comparison with operational Numerical
Weather Prediction (NWP) products and interpolated wind fields from ship measurements and
cloud motions. They found large-scale similarity
between SSM/I and other wind fields, but suggested that the dense space–time coverage of
SSM/I offers a distinct advantage in forcing ocean
circulation models. The SSM/I wind fields were
found by Liu et al. (1996) to generate more realistic anomalous ocean cooling in an ocean general
circulation when compared with the results simulated by the same model but forced by NWP winds.
The SSM/I wind fields have more structure and
energy than NWP winds but have the same directional characteristics as the NWP winds. SSM/I is an
operational sensor that will be operated well into the
twenty-first century. The production of wind vector
data from these assimilated microwave radiometer
measurements, however, is a research effort.
Scatterometers send microwave pulses to the
earth’s surface and measure the backscattered
power (so-called radar cross-section) from the
surface roughness. Over the ocean, the backscatter
depends on ocean surface roughness due to small
(centimetre scale) waves. The idea of remote sensing of ocean surface winds was based on the belief
SECTION 3 NEW WAYS OF OBSERVING THE OCEAN
174