Atlantic Coastal Biome
259
of the equator by confluence with the weaker NEC (Boisvert, 1967). The NBC changes
character seasonally, forming a series of coastal eddies in boreal winter that transport
only 10 Sv, but the surge of the southeast trades into the western Atlantic during boreal
summer transforms it into a western jet that then transports as much as 35 Sv.
Although topographically locked to the continental edge, the NBC becomes unstable
between the mouths of the Amazon and Orinoco so that a small part of the flow which
overlies the continental slope is retroflected eastward into the open Atlantic in a series
of very large, persistent eddies: the Amazon eddy is formed at 4
N, and the Demerara
eddy at 8
N (Bruce et al., 1985). It is convenient to consider these eddies, which form
prominent features extending far seaward, as part of the western tropical Atlantic province
(see WARM); they are, as Fratantoni and Glickson (2002) remark, the largest oceanic
rings formed anywhere in the ocean, which is not surprising given the low value taken
by the Coriolis parameter at only 2-3
from the equator.
It is the effect of the discharges from the Amazon on the ecology of the Atlantic
shelf, and from the Orinoco on the Caribbean shelf, that will take most of our attention
in discussing this province. The Orinoco is not a negligible river, having the fourth
largest annual discharge of any, but the Amazon, of course, is a special case that is well
described by Geyer et al. (1996) both because it discharges more water than any other
river but also because it lies directly on the equator; moreover, unlike the Congo, it is not
associated with a deep cross-shelf canyon. The freshwater discharged by the Amazon has
a nitrate concentration of about 15 M, which is small compared with the discharge of
other great rivers: Yangtse, Yellow, and Mississippi rivers discharge freshwater containing
50–100 M nitrate. Perhaps when the Amazon rain forest is entirely cleared, and the
basin reduced largely to agriculture as in the other examples, its freshwater drainage will
come to have similarly high nitrate values. Nevertheless, the Amazon discharge does have
a very strong signature of CDOM, not distinguishable from the chlorophyll signature in
satellite imagery unless the data are processed with a specific algorithm (Hu et al., 2004).
At peak flow in May, the rate of discharge from the Amazon is about 0.2 Sv
(or ∼220000 m
−3 sec
−1 ), about twice the minimum flow in November. Such strong discharges prevent seawater from entering the river mouth, so that a strong salinity front
occurs about 100 km offshore, coincident with a turbidity front, and lying above a series
of transverse shoals formed by deposition of sediments. Even at this distance offshore,
water depth is only about 15–20 m above a submarine delta that is being built out across
the shelf. Although tidal streams are sufficiently strong that cross-shelf velocities reach
200 cm sec
−1 in the frontal zone, the suspended load in the bottom water is so great as
to dampen vertical mixing by tidally induced bottom stress.
The shallow, low-salinity plume of river water passes northwest along the coast, the
form taken by its bounding salinity front being wind-dependent. Southeast winds induce
a “fast” plume of brackish water to head along the coast, whereas northeast wind produces
a “slow” plume that balloons out across the shelf, with very little northward transport
from the river mouths. Here, another interesting example of the consequence of the
low value of the Coriolis acceleration at low latitudes (see Chapter 4) is observed: in
middle to high latitudes, as Nittrouer and DeMaster (1996) point out, the Coriolis force
would induce the plume to turn anticyclonically across the shelf (clockwise, that is, in
the northern hemisphere), but here the dynamics of the plume respond almost directly
to the sustained landward wind stress. This is sufficient to maintain a general flow along
the coast to the northwest, with anticyclonic retroflection occurring only in relation to
topographic features on the slope (Fig. 9.27).
Drogued, satellite-tracked drifters have been deployed in the NBC just offshore of
the Amazon shelf and near the river mouths. Most of these buoys subsequently track
northwest along the shelf, with those nearest the shelf edge describing a series of tight
anticyclonic eddies only as they approached and entered the Caribbean near Trinidad.
Several passed at a shallow angle toward the edge of the shelf and entered the retroflection
259
of the equator by confluence with the weaker NEC (Boisvert, 1967). The NBC changes
character seasonally, forming a series of coastal eddies in boreal winter that transport
only 10 Sv, but the surge of the southeast trades into the western Atlantic during boreal
summer transforms it into a western jet that then transports as much as 35 Sv.
Although topographically locked to the continental edge, the NBC becomes unstable
between the mouths of the Amazon and Orinoco so that a small part of the flow which
overlies the continental slope is retroflected eastward into the open Atlantic in a series
of very large, persistent eddies: the Amazon eddy is formed at 4
N, and the Demerara
eddy at 8
N (Bruce et al., 1985). It is convenient to consider these eddies, which form
prominent features extending far seaward, as part of the western tropical Atlantic province
(see WARM); they are, as Fratantoni and Glickson (2002) remark, the largest oceanic
rings formed anywhere in the ocean, which is not surprising given the low value taken
by the Coriolis parameter at only 2-3
from the equator.
It is the effect of the discharges from the Amazon on the ecology of the Atlantic
shelf, and from the Orinoco on the Caribbean shelf, that will take most of our attention
in discussing this province. The Orinoco is not a negligible river, having the fourth
largest annual discharge of any, but the Amazon, of course, is a special case that is well
described by Geyer et al. (1996) both because it discharges more water than any other
river but also because it lies directly on the equator; moreover, unlike the Congo, it is not
associated with a deep cross-shelf canyon. The freshwater discharged by the Amazon has
a nitrate concentration of about 15 M, which is small compared with the discharge of
other great rivers: Yangtse, Yellow, and Mississippi rivers discharge freshwater containing
50–100 M nitrate. Perhaps when the Amazon rain forest is entirely cleared, and the
basin reduced largely to agriculture as in the other examples, its freshwater drainage will
come to have similarly high nitrate values. Nevertheless, the Amazon discharge does have
a very strong signature of CDOM, not distinguishable from the chlorophyll signature in
satellite imagery unless the data are processed with a specific algorithm (Hu et al., 2004).
At peak flow in May, the rate of discharge from the Amazon is about 0.2 Sv
(or ∼220000 m
−3 sec
−1 ), about twice the minimum flow in November. Such strong discharges prevent seawater from entering the river mouth, so that a strong salinity front
occurs about 100 km offshore, coincident with a turbidity front, and lying above a series
of transverse shoals formed by deposition of sediments. Even at this distance offshore,
water depth is only about 15–20 m above a submarine delta that is being built out across
the shelf. Although tidal streams are sufficiently strong that cross-shelf velocities reach
200 cm sec
−1 in the frontal zone, the suspended load in the bottom water is so great as
to dampen vertical mixing by tidally induced bottom stress.
The shallow, low-salinity plume of river water passes northwest along the coast, the
form taken by its bounding salinity front being wind-dependent. Southeast winds induce
a “fast” plume of brackish water to head along the coast, whereas northeast wind produces
a “slow” plume that balloons out across the shelf, with very little northward transport
from the river mouths. Here, another interesting example of the consequence of the
low value of the Coriolis acceleration at low latitudes (see Chapter 4) is observed: in
middle to high latitudes, as Nittrouer and DeMaster (1996) point out, the Coriolis force
would induce the plume to turn anticyclonically across the shelf (clockwise, that is, in
the northern hemisphere), but here the dynamics of the plume respond almost directly
to the sustained landward wind stress. This is sufficient to maintain a general flow along
the coast to the northwest, with anticyclonic retroflection occurring only in relation to
topographic features on the slope (Fig. 9.27).
Drogued, satellite-tracked drifters have been deployed in the NBC just offshore of
the Amazon shelf and near the river mouths. Most of these buoys subsequently track
northwest along the shelf, with those nearest the shelf edge describing a series of tight
anticyclonic eddies only as they approached and entered the Caribbean near Trinidad.
Several passed at a shallow angle toward the edge of the shelf and entered the retroflection