that these surface ripples are in equilibrium with
the local wind stress. The backscatter depends not
only on the magnitude of the wind stress, but also
the wind direction relative to the direction of the
radar beam (the azimuth angle) (e.g. Jones et al.,
1978). Space-based scatterometers, operating on
polar-orbiting satellites, provide both wind speed
and direction through multiple looks at any one
pixel in the swathe. Two wavelength bands, at
frequencies of approximately 5 and 14 GHz, have
been used with satellite scatterometers. The higher
frequencies of the Ku-band allow greater sensitivity at low wind speeds to wind direction, but also
exhibit stronger influence from atmospheric precipitation, as compared with the lower frequencies
of the C-band. The geophysical model function,
from which ocean surface wind vectors are
retrieved from the radar cross-section, are largely
based on empirical fits of data (e.g. Jones et al.,
1978; Freilich and Dunbar, 1993; Thiria et al.,
1993; Stoffelen and Anderson, 1997; Wentz and
Smith, 1999).
The European Remote Sensing (ERS) satellites
ERS-1 and 2 (Attema, 1991), launched in 1991
and 1996, respectively, carry the Advanced
Microwave Instrument (AMI). This operates at the
C-band (5.3 GHz) frequency and provides wind
vectors at 50 km resolution over a 500 km swathe
to the right of the satellite subtrack (Fig. 3.4.1b,
see Plate 3.4.1, p. 172). This narrow swathe limits
the daily coverage to 40% of the global ocean, and
it requires 3 days to provide almost full coverage,
but the ERS-2 continues to provide data in 2000.
Bentamy et al. (1998) found good agreement
between the interpolated wind fields derived from
ERS-1 data and weekly and monthly averaged
NWP wind fields. Liu et al. (1995) simulated
Kelvin waves and anomalous ocean warming in
the equatorial Pacific by forcing an ocean general
circulation model with ERS-1 scatterometer winds.
Weekly wind fields derived from ERS scatterometers were shown by Grima et al. (1999) to be
superior to wind fields from a climate model in
simulating tropical ocean circulation in that the
observed upper ocean structure agreed better with
buoy-measured ocean structure and currents.
The National Aeronautics and Space Administration (NASA) launched the very first scatterometer
on the SEASAT satellite in June 1978, which lived
for only 3 months. It operated at the Ku-band
(14.6 GHz). Four fan-beam dual-polarized antennas
illuminated two 500 km swathes, one on each side
of the spacecraft, providing wind vectors at 50 km
resolution. However, only one side was in operation most of the time. The US developments since
have built on that experiment. In 1996, NASA Scatterometer (NSCAT) was launched on the Japanese
spacecraft, Midori (ADEOS-1). Its six fan-beam
antennas provided 600-km wide swathes on both
sides of the spacecraft. It measured at the Ku band
(14 GHz) and provided ocean surface winds at
25 km resolution, and covering 77% of the global
ocean every day (Fig. 3.4.1a, see Plate 3.4.1,
p. 172), until its early demise in June 1997, owing
to loss of the solar panel on the spacecraft. The
quality of data was shown to be above expectation
through comparison with in-situ measurements (e.g.
Bourassa et al., 1997; Freilich and Dunbar, 1999),
through comparison with winds from the analysis of
NWP models (e.g. Liu et al., 1998; Atlas et al.,
1999; Ebuchi, 1999), and through forcing of ocean
circulation models (e.g. Chen et al., 1999; Chu et al.,
1999; Milliff et al., 1999; Verschell et al., 1999).
A new scatterometer, SeaWinds, was launched
on the NASA mission QuikSCAT in June 1999.
SeaWinds uses pencil-beam antennas in a conical
scan. The antennas radiate Ku-band microwaves
at 46° and 54° incident angles and measure the
backscattered power across a continuous 1800 km
swathe. SeaWinds is capable of providing wind
speed and direction at 25 km resolution over 93%
of the earth’s ice-free oceans every day, under both
clear and cloudy conditions (Graf et al., 1998).
The power of synoptic global coverage and high
spatial resolution by a space-based scatterometer is
clearly illustrated in Fig. 3.4.2 (see Plate 3.4.2,
p. 172). The impact of scatterometer winds in the
analysis of Hurricane Floyd, shown in the insert of
Fig. 3.4.2, is discussed by Liu et al. (2000). The
SeaWinds data used in the insert was specially produced to have a spatial resolution of 12.5 km and
for the strong wind and high precipitation conditions in tropical cyclones. Practical utility for
obtaining the gale force wind radius in tropical
cyclones is reported by Uhlhorn et al. (2000). SeaWinds was also found to detect the circulation of
tropical depressions early in their development
(Katsaros et al., 2000). Both the scatterometers on
SEASAT and NSCAT have a data gap at nadir
between the two swathes. The continuous 1800km wide swathe of SeaWinds is a tremendous
technical advance.
3.4 Air–Sea Fluxes from Satellite Data
175
Liu and Katsaros
the local wind stress. The backscatter depends not
only on the magnitude of the wind stress, but also
the wind direction relative to the direction of the
radar beam (the azimuth angle) (e.g. Jones et al.,
1978). Space-based scatterometers, operating on
polar-orbiting satellites, provide both wind speed
and direction through multiple looks at any one
pixel in the swathe. Two wavelength bands, at
frequencies of approximately 5 and 14 GHz, have
been used with satellite scatterometers. The higher
frequencies of the Ku-band allow greater sensitivity at low wind speeds to wind direction, but also
exhibit stronger influence from atmospheric precipitation, as compared with the lower frequencies
of the C-band. The geophysical model function,
from which ocean surface wind vectors are
retrieved from the radar cross-section, are largely
based on empirical fits of data (e.g. Jones et al.,
1978; Freilich and Dunbar, 1993; Thiria et al.,
1993; Stoffelen and Anderson, 1997; Wentz and
Smith, 1999).
The European Remote Sensing (ERS) satellites
ERS-1 and 2 (Attema, 1991), launched in 1991
and 1996, respectively, carry the Advanced
Microwave Instrument (AMI). This operates at the
C-band (5.3 GHz) frequency and provides wind
vectors at 50 km resolution over a 500 km swathe
to the right of the satellite subtrack (Fig. 3.4.1b,
see Plate 3.4.1, p. 172). This narrow swathe limits
the daily coverage to 40% of the global ocean, and
it requires 3 days to provide almost full coverage,
but the ERS-2 continues to provide data in 2000.
Bentamy et al. (1998) found good agreement
between the interpolated wind fields derived from
ERS-1 data and weekly and monthly averaged
NWP wind fields. Liu et al. (1995) simulated
Kelvin waves and anomalous ocean warming in
the equatorial Pacific by forcing an ocean general
circulation model with ERS-1 scatterometer winds.
Weekly wind fields derived from ERS scatterometers were shown by Grima et al. (1999) to be
superior to wind fields from a climate model in
simulating tropical ocean circulation in that the
observed upper ocean structure agreed better with
buoy-measured ocean structure and currents.
The National Aeronautics and Space Administration (NASA) launched the very first scatterometer
on the SEASAT satellite in June 1978, which lived
for only 3 months. It operated at the Ku-band
(14.6 GHz). Four fan-beam dual-polarized antennas
illuminated two 500 km swathes, one on each side
of the spacecraft, providing wind vectors at 50 km
resolution. However, only one side was in operation most of the time. The US developments since
have built on that experiment. In 1996, NASA Scatterometer (NSCAT) was launched on the Japanese
spacecraft, Midori (ADEOS-1). Its six fan-beam
antennas provided 600-km wide swathes on both
sides of the spacecraft. It measured at the Ku band
(14 GHz) and provided ocean surface winds at
25 km resolution, and covering 77% of the global
ocean every day (Fig. 3.4.1a, see Plate 3.4.1,
p. 172), until its early demise in June 1997, owing
to loss of the solar panel on the spacecraft. The
quality of data was shown to be above expectation
through comparison with in-situ measurements (e.g.
Bourassa et al., 1997; Freilich and Dunbar, 1999),
through comparison with winds from the analysis of
NWP models (e.g. Liu et al., 1998; Atlas et al.,
1999; Ebuchi, 1999), and through forcing of ocean
circulation models (e.g. Chen et al., 1999; Chu et al.,
1999; Milliff et al., 1999; Verschell et al., 1999).
A new scatterometer, SeaWinds, was launched
on the NASA mission QuikSCAT in June 1999.
SeaWinds uses pencil-beam antennas in a conical
scan. The antennas radiate Ku-band microwaves
at 46° and 54° incident angles and measure the
backscattered power across a continuous 1800 km
swathe. SeaWinds is capable of providing wind
speed and direction at 25 km resolution over 93%
of the earth’s ice-free oceans every day, under both
clear and cloudy conditions (Graf et al., 1998).
The power of synoptic global coverage and high
spatial resolution by a space-based scatterometer is
clearly illustrated in Fig. 3.4.2 (see Plate 3.4.2,
p. 172). The impact of scatterometer winds in the
analysis of Hurricane Floyd, shown in the insert of
Fig. 3.4.2, is discussed by Liu et al. (2000). The
SeaWinds data used in the insert was specially produced to have a spatial resolution of 12.5 km and
for the strong wind and high precipitation conditions in tropical cyclones. Practical utility for
obtaining the gale force wind radius in tropical
cyclones is reported by Uhlhorn et al. (2000). SeaWinds was also found to detect the circulation of
tropical depressions early in their development
(Katsaros et al., 2000). Both the scatterometers on
SEASAT and NSCAT have a data gap at nadir
between the two swathes. The continuous 1800km wide swathe of SeaWinds is a tremendous
technical advance.
3.4 Air–Sea Fluxes from Satellite Data
175
Liu and Katsaros