82
W. Alpers
4.3.1 Atmospheric Phenomena
Atmospheric phenomena in the marine boundary layer become visible on SAR images of the sea surface by the variations of the short-scale sea surface roughness.
The near-surface wind modulates the short-scale sea surface roughness and thus, via
Bragg scattering, also the backscattered radar power or the Normalized Radar Cross
Section (NRCS) (see e.g. Valenzuela 1978).
On SAR images, the modulation of the NRCS becomes visible as image intensity
or image brightness modulation. The high resolution SAR already gives valuable
information on the location and the spatial structure of atmospheric phenomena, but
more information is extracted from them when the NRCS values are converted into
near-surface wind vectors. Such conversion is achieved by applying Geophysical
Model Functions (GMFs) or wind scatterometer model functions used in scatterometry (see Sect. 4.2). However, quantitative information on the sea surface wind speed
can only be extracted from SAR images when the wind direction near the sea surface
is known.
The wind direction is sometimes inferred from the SAR images itself, e.g., from
the direction of wind streaks or from the direction of wind shadows behind coastal
mountains or mountainous islands (Monaldo et al. 2001, Monaldo and Karbaol
2003; Horstmann and Koch 2005). Another way to obtain wind direction is from
atmospheric models, like the one developed by the National Center for Environmental
Prediction (NCEP) in the US. Recently, a new wind retrieval has been developed in
which also Doppler shifts induced by motions of the sea surface is included (Mouche
et al. 2012). This new wind retrieval algorithm uses all three sources of information
on wind direction (model, linear features, and Doppler shift) and combines them
using the Bayesian method.
However, near-surface wind fields cannot be retrieved from SAR images at the
spatial resolution of the SAR because of the speckle inherent in all SAR images.
In order to reduce the speckle, one has to carry out averaging over pixels which
implies reduction of the spatial resolution. It requires a trade-off between accuracy
of the wind field and spatial resolution. Typically, wind fields are retrieved from SAR
images with a spatial resolution of 200–500 m.
4.3.1.1 Atmospheric Gravity Waves
Atmospheric Gravity Waves (AGWs), which sometimes are also called atmospheric
internal waves, are encountered in the atmosphere when it consists of layers of
different densities (potential temperature). They either occur as quasi-periodic waves
or as strongly nonlinear waves in the form of solitary waves or wave packets. AGWs
are often generated behind mountain ranges in which case they are called lee waves.
Usually lee waves are stationary with respect to the terrain feature, but they are
propagating relative to the mean air flow above the earth surface.
In Fig. 4.3 an ERS-1 SAR image is depicted which was acquired at 2239 UTC on
3 September 1993 during an ascending satellite pass. It shows in the lower section sea
W. Alpers
4.3.1 Atmospheric Phenomena
Atmospheric phenomena in the marine boundary layer become visible on SAR images of the sea surface by the variations of the short-scale sea surface roughness.
The near-surface wind modulates the short-scale sea surface roughness and thus, via
Bragg scattering, also the backscattered radar power or the Normalized Radar Cross
Section (NRCS) (see e.g. Valenzuela 1978).
On SAR images, the modulation of the NRCS becomes visible as image intensity
or image brightness modulation. The high resolution SAR already gives valuable
information on the location and the spatial structure of atmospheric phenomena, but
more information is extracted from them when the NRCS values are converted into
near-surface wind vectors. Such conversion is achieved by applying Geophysical
Model Functions (GMFs) or wind scatterometer model functions used in scatterometry (see Sect. 4.2). However, quantitative information on the sea surface wind speed
can only be extracted from SAR images when the wind direction near the sea surface
is known.
The wind direction is sometimes inferred from the SAR images itself, e.g., from
the direction of wind streaks or from the direction of wind shadows behind coastal
mountains or mountainous islands (Monaldo et al. 2001, Monaldo and Karbaol
2003; Horstmann and Koch 2005). Another way to obtain wind direction is from
atmospheric models, like the one developed by the National Center for Environmental
Prediction (NCEP) in the US. Recently, a new wind retrieval has been developed in
which also Doppler shifts induced by motions of the sea surface is included (Mouche
et al. 2012). This new wind retrieval algorithm uses all three sources of information
on wind direction (model, linear features, and Doppler shift) and combines them
using the Bayesian method.
However, near-surface wind fields cannot be retrieved from SAR images at the
spatial resolution of the SAR because of the speckle inherent in all SAR images.
In order to reduce the speckle, one has to carry out averaging over pixels which
implies reduction of the spatial resolution. It requires a trade-off between accuracy
of the wind field and spatial resolution. Typically, wind fields are retrieved from SAR
images with a spatial resolution of 200–500 m.
4.3.1.1 Atmospheric Gravity Waves
Atmospheric Gravity Waves (AGWs), which sometimes are also called atmospheric
internal waves, are encountered in the atmosphere when it consists of layers of
different densities (potential temperature). They either occur as quasi-periodic waves
or as strongly nonlinear waves in the form of solitary waves or wave packets. AGWs
are often generated behind mountain ranges in which case they are called lee waves.
Usually lee waves are stationary with respect to the terrain feature, but they are
propagating relative to the mean air flow above the earth surface.
In Fig. 4.3 an ERS-1 SAR image is depicted which was acquired at 2239 UTC on
3 September 1993 during an ascending satellite pass. It shows in the lower section sea