Using MODIS ocean color satellite images
collected during March -- July 2004, an attempt was made
to observe a series of eddies in the central Sargasso Sea before
and during 2 first cruises of the EDDIES project. Five eddies
of interest were identified prior to the cruises using satellite
altimetry: 3 anticyclones and 2 cyclones. Using altimetry archive
each of the eddies was tracked to determine its trajectory
and amplitude. Then this data set together with cruise data was
used to identify exact position and strength of a particular eddies
in color images.
We have used SEADAS software to
display MODIS color images processed to Level 2 chlorophyll-a
maps using cylindrical map projection with the spatial resolution
~1 km. The chlorophyll levels in each of the eddies did not
remain constant over the time; so, the features are not clearly
evident in each image.
March-April 2004.
During this period, when the mixed
layer is deep or just starts to shoal, and the spring
phytoplankton bloom is triggered, satellite images show
increased concentrations of surface chlorophyll.
MODIS image acquired March 12,
2004 shows increased Chl-a concentrations tied with
a frontal zone located along 61W with a multiple small-scale
swirls. Anticyclone eddy marked as "A1", in its central part and
north of it is obscured by clouds. But a broad belt of increased
Chl-a concentration is clearly seen in the southern and eastern
parts of the eddy. It is not clear if this increase is associated
with the local upwelling inside the eddy, or is due to horizontal
advection of phytoplankton from the adjacent cold eddy located
NE of "A1". Another one Chl-a patch located at 31N, 60W may be
associated with western periphery of cold eddy "C1". Altimetry
maps show that center of this eddy is located at 31.5N, 59W.
Next image acquired 2 weeks later
March 28, shows Chl-a maximum located SE
and S of two patterns associated with 2 strong cold eddied (rings?).
Anticyclonic eddy center marked as "A1", is located approximately
50 km SE of that feature. At this time, increased Chl-a
concentration is evident in center of cold eddy marked "C1",
and in center of another one cold eddy marked "C2'. An interesting
feature is a tongue of low chl-a water intruding through the
south periphery of C1 from SW. This tongue has anticyclonic
rotation and may be associated with the anticyclonic eddy A2.
Color image obtained on April 1,
2004, shows Chl-a pattern consisting of two rings
and A1 eddy. But in this case the area of increased chl-a is
located right in the center of "A1" eddy. An interesting feature
is a long "sleeve" connecting eastern parts of A1 and cold core
eddy just to the north of A1. Taking into account direction of rotation
in both eddies we can conclude that A1 eddy advectively transports
high chl-a waters and accumulates them in its central part. But
these waters not originated from the adjacent cold eddy. Seem, both
of them transport high chl-a waters which are resulted from the upwelling
processes on the periphery of these eddies. In lower part of the
image a low chl-a tongue has moved further to the east and the remnant
of eddy "C1" is evident.
Image acquired April 4, 2004 shows
that chl-a concentration in area occupied by A1 eddy is
increased significantly. A tongue of low chl-a water which is
associated with A2 eddy, almost completed anticyclonic rotation.
Also, "C1" eddy is visible.
Two consequtive images obtained
on April 30 and May 1, do not show any features of interest.
May-June 2004.
Next image was obtained on May 8,
2004. At this time of the year the spring bloom rapidly
utilizes the nutrient in the surface waters, and the subsurface
deep chlorophyll maximum (DCM) begins to develop. Also, as water
column stratification continues to intensify below the shallow
mixed layer, the seasonal thermocline inhibits further nutrient
enrichment of the surface waters. Despite this reason color image
shows concentration in the area associated with the position of
A1 eddy is high enough to conclude that this increase is a result
of vertical processes inside the eddy but horizontal advection.
Position of eddy C2 is also indicated.
On images obtained on May 20, June
5, and June 6, 2004, eddy A1, migrating to the west, is
still may be tracked due to chl-a pattern showing anticyclonic
rotation. The higher chlorophyll values in A1 center appear
to be advected from the northern, cyclonic eddy. The drawing
of water from cyclonic eddy can explain why the anticyclonic eddy
has high chl-a concentration and the cyclonic eddies C1 and C2 have
low chl-a concentration. Examining these images showed that beginning
from early May anticyclonic eddy A1 draws chl-a rich chl-a
waters from the northern eddy. Cyclonic eddies moving through the
relatively clear water form DCM which may be not detected onboard the
satellite. Note a large area of specific (similar to triangle)
form with very low chl-a concentration centered on 30N, 62W. It is
interesting that this low chl-a pattern may be tracked down in all
images (until June 17) but it is unclear which mechanism forms this pattern.
A clear sky image acquired June
16 shows intense streamer drawing high chlorophyll waters
from north in the southward direction. Approximately near 33N,
this streamer changs its direction and high chl-a values appear
to be advected into the western, anticyclonic and cyclonic
eddies. Comparison to altimetry map shows that position
and direction of this streamer, originally formed
from Gulf stream, coinside with the eastern periphery of the long
anticyclonic feature extending from northern boundary up to A1
eddy.
Unfortunately, we don't have a possibility
to to identify this streamer in ADCP data: ship's track didn't
cross it. The only small part of ADCP data revealed a southwest
jet-like current near 32.5N, 62.5W. Position of C2 eddy is also
indicated.
MODIS image acquired next day, shows
a body of high chl-a water accumulated inside A1 eddy. A
small (southern) part of the streamer is still presented
on the image.
Althought the image obtained on
June 28, 2004 is 70% clear sky, it is difficult to identify
any chl-a patterns with A1 eddy. The higher chl-a values
near 30.75N, -60.75W appear to be cyclonic eddy C2. Location
of another one cyclonic eddy C2 is evident due to cyclonic-like
swirl at 30.25N, 65.0W.
On three consecutive images acquired
July 16, 17, and 18, the only location of the
cyclonic eddies C1 and C2 may be identified.
|