362
S. S. Karimova and M. Gade
(Ufermann and Romeiser 1999). As a criteria for the discrimination between “white”
eddies and atmospheric phenomena we used the presence of thin shear lines typical
for “white” oceanic eddies. Such shear lines can be found, e.g., in Fig. 18.3a, b inside
the manifestations of “white” eddies marked with the ellipse and with a rectangle.
All SAR images used for the present investigation were visually inspected and for
each detected eddy, its sign of rotation (cyclonic or anti-cyclonic), center coordinates
and diameter were recorded using the BEAM Visat software. The rotational sign
was defined based on the spiral structure of the eddies: if the spiral was swirling
counter-clockwise (clockwise) the eddy was regarded as cyclonic (anti-cyclonic).
The results of our statistical analyses of sub-mesoscale eddies (with diameters less
than approximately 30 km) are provided in the following section, while those of
meso- and basin-scale eddies (with larger diameters) are discussed in Sect. 4 and 5,
respectively.
18.3 Sub-mesoscale Eddies
In total, 582 “black” and 440 “white” sub-mesoscale eddies were detected in the
492 SAR images analyzed. It is noteworthy that the number of eddies per SAR
image is much smaller than that discovered for the Baltic, Black, and Caspian seas,
where about 14,000 eddies were detected in approximately 2,000 images (Karimova
2012). Presumably this is due to the greater depth of the upper mixed layer, which
is typically about 100 m for the Red Sea, compared to 10–40 m (depending on the
season) for the above mentioned basins. Following Tragou and Garrett (1997) the
baroclinic Rossby radius in the Red Sea is approximately 30 km; therefore, we chose
this length to discriminate between sub-mesoscale (smaller) and meso- to basin-scale
(larger) eddies. The vast majority (> 90 %) of eddies found in the ASAR imagery are
cyclonical, which is in line with theoretical considerations and numerical modeling
(Kleppin 2012).
We have to note here that the very mechanisms of sub-mesoscale eddy generation,
especially in case of spatial densely-packed groups of such eddies, are still under
investigation. In some particular cases, the mechanisms of eddy generation are quite
obvious and coincide with those of bigger eddies, such as lateral friction of the
near-coastal flow, influence of a coastline inhomogeneity, baroclinic instability at
the hydrological fronts, and barotropic (shear) instability in the shear current zones.
For a general case, when none of the hypotheses mentioned can be applied, eddy
origination was explained by current’s baroclinic instability (Eldevik and Dysthe
2002; Akitomo 2010; Kleppin 2012), quasi-periodic impulsive impact on the water
surface (Voropayev and Afanasyev 1992), and convection in the near-surface layer
(Boubnov and Golitsyn 1995). The results on sub-mesoscale eddy spatio-temporal
distribution provided here and in (Karimova 2012) seem to support the influence of
water baroclinicity on sub-mesoscale eddy activity.
In the following subsections we will analyze spatial scale and spatial distribution
of sub-mesoscale eddies detected.
S. S. Karimova and M. Gade
(Ufermann and Romeiser 1999). As a criteria for the discrimination between “white”
eddies and atmospheric phenomena we used the presence of thin shear lines typical
for “white” oceanic eddies. Such shear lines can be found, e.g., in Fig. 18.3a, b inside
the manifestations of “white” eddies marked with the ellipse and with a rectangle.
All SAR images used for the present investigation were visually inspected and for
each detected eddy, its sign of rotation (cyclonic or anti-cyclonic), center coordinates
and diameter were recorded using the BEAM Visat software. The rotational sign
was defined based on the spiral structure of the eddies: if the spiral was swirling
counter-clockwise (clockwise) the eddy was regarded as cyclonic (anti-cyclonic).
The results of our statistical analyses of sub-mesoscale eddies (with diameters less
than approximately 30 km) are provided in the following section, while those of
meso- and basin-scale eddies (with larger diameters) are discussed in Sect. 4 and 5,
respectively.
18.3 Sub-mesoscale Eddies
In total, 582 “black” and 440 “white” sub-mesoscale eddies were detected in the
492 SAR images analyzed. It is noteworthy that the number of eddies per SAR
image is much smaller than that discovered for the Baltic, Black, and Caspian seas,
where about 14,000 eddies were detected in approximately 2,000 images (Karimova
2012). Presumably this is due to the greater depth of the upper mixed layer, which
is typically about 100 m for the Red Sea, compared to 10–40 m (depending on the
season) for the above mentioned basins. Following Tragou and Garrett (1997) the
baroclinic Rossby radius in the Red Sea is approximately 30 km; therefore, we chose
this length to discriminate between sub-mesoscale (smaller) and meso- to basin-scale
(larger) eddies. The vast majority (> 90 %) of eddies found in the ASAR imagery are
cyclonical, which is in line with theoretical considerations and numerical modeling
(Kleppin 2012).
We have to note here that the very mechanisms of sub-mesoscale eddy generation,
especially in case of spatial densely-packed groups of such eddies, are still under
investigation. In some particular cases, the mechanisms of eddy generation are quite
obvious and coincide with those of bigger eddies, such as lateral friction of the
near-coastal flow, influence of a coastline inhomogeneity, baroclinic instability at
the hydrological fronts, and barotropic (shear) instability in the shear current zones.
For a general case, when none of the hypotheses mentioned can be applied, eddy
origination was explained by current’s baroclinic instability (Eldevik and Dysthe
2002; Akitomo 2010; Kleppin 2012), quasi-periodic impulsive impact on the water
surface (Voropayev and Afanasyev 1992), and convection in the near-surface layer
(Boubnov and Golitsyn 1995). The results on sub-mesoscale eddy spatio-temporal
distribution provided here and in (Karimova 2012) seem to support the influence of
water baroclinicity on sub-mesoscale eddy activity.
In the following subsections we will analyze spatial scale and spatial distribution
of sub-mesoscale eddies detected.