SATELLITE MEASUREMENTS
175
complementary orbits it would be difficult to provide a reliable operational
wave monitoring service based on observations alone. However, when the
satellite data are assimilated into ocean wave forecasting models, their
global coverage, their accuracy and their detailed resolution along the track
ensure that they make a measurable improvement to the skill and reliability
of wave forecasts.
The major shortcoming of this type of wave
measurement is that it contains no information about the wave direction.
3.4.2
Directional wave spectra from synthetic aperture radars
When a serious discrepancy is found between a wave model forecast and
the measured wave height, it is desirable to know the directional properties
of the waves so that incorrect wave energy sources in the model can be
located and corrected. Directional wave spectra can be estimated using
synthetic aperture radar (SAR). SARs view the sea surface obliquely and
produce image maps of the backscattered microwave energy, at a spatial
resolution of about 25m. The signal processing required to achieve this
resolution is computationally intensive and can generate interesting artefacts,
especially when the scattering surface is in motion (see Chapters 9 and 10 of
Robinson (2004) for an introduction to SAR ocean imaging). The radar
backscatter signal itself needs to be interpreted carefully. It represents a
measure of the roughness of the sea surface at length scales similar to the
radar wavelength, since a type of Bragg scattering mechanism is responsible
for the radar echo from obliquely incident microwave pulses.
Swell waves appear on SAR images as approximately linear patches of
bright and dark corresponding to regions where the backscatter is greater or
less than the mean. There are three different mechanisms by which the swell
waves modulate the short waves which determine the backscatter. A twodimensional spectrum of the image amplitude is related to, but not the same
as, the directional wave spectrum. The modulation transfer function which
relates the ocean wave spectrum to the image spectrum depends on the swell
wave frequency, their amplitude and their direction relative to the radar
azimuth. The problem of accurately and reliably inverting image spectra to
retrieve wave spectra has been a challenging field of research for two
decades, although recent techniques which use the complex form of the SAR
image (including the phase as well as the amplitude of the backscatter) look
most promising (see Chapron et al. (2001) for a review of the subject). It is
most important that users of SAR-derived wave spectra should understand
that SAR can provide very little information about the shorter waves with
period less than abut 10 s, and this frequency cut-off is worse for waves
propagating in a direction parallel to the satellite track.
The SARs of the European Space Agency can operate in a wave-mode in
which small (5 × 10 km) imagettes are acquired every 200 km along track.
Thus wave spectra can be sampled from across the global ocean every day.
175
complementary orbits it would be difficult to provide a reliable operational
wave monitoring service based on observations alone. However, when the
satellite data are assimilated into ocean wave forecasting models, their
global coverage, their accuracy and their detailed resolution along the track
ensure that they make a measurable improvement to the skill and reliability
of wave forecasts.
The major shortcoming of this type of wave
measurement is that it contains no information about the wave direction.
3.4.2
Directional wave spectra from synthetic aperture radars
When a serious discrepancy is found between a wave model forecast and
the measured wave height, it is desirable to know the directional properties
of the waves so that incorrect wave energy sources in the model can be
located and corrected. Directional wave spectra can be estimated using
synthetic aperture radar (SAR). SARs view the sea surface obliquely and
produce image maps of the backscattered microwave energy, at a spatial
resolution of about 25m. The signal processing required to achieve this
resolution is computationally intensive and can generate interesting artefacts,
especially when the scattering surface is in motion (see Chapters 9 and 10 of
Robinson (2004) for an introduction to SAR ocean imaging). The radar
backscatter signal itself needs to be interpreted carefully. It represents a
measure of the roughness of the sea surface at length scales similar to the
radar wavelength, since a type of Bragg scattering mechanism is responsible
for the radar echo from obliquely incident microwave pulses.
Swell waves appear on SAR images as approximately linear patches of
bright and dark corresponding to regions where the backscatter is greater or
less than the mean. There are three different mechanisms by which the swell
waves modulate the short waves which determine the backscatter. A twodimensional spectrum of the image amplitude is related to, but not the same
as, the directional wave spectrum. The modulation transfer function which
relates the ocean wave spectrum to the image spectrum depends on the swell
wave frequency, their amplitude and their direction relative to the radar
azimuth. The problem of accurately and reliably inverting image spectra to
retrieve wave spectra has been a challenging field of research for two
decades, although recent techniques which use the complex form of the SAR
image (including the phase as well as the amplitude of the backscatter) look
most promising (see Chapron et al. (2001) for a review of the subject). It is
most important that users of SAR-derived wave spectra should understand
that SAR can provide very little information about the shorter waves with
period less than abut 10 s, and this frequency cut-off is worse for waves
propagating in a direction parallel to the satellite track.
The SARs of the European Space Agency can operate in a wave-mode in
which small (5 × 10 km) imagettes are acquired every 200 km along track.
Thus wave spectra can be sampled from across the global ocean every day.