Numerical Lagrangian drifters
from the computer SALMON model. The tails are ten days
long and the small wiggles are due to the semidurnal
tide. The colors show the depth, with dark red at the
surface and the blues below 100 meters. The black contours
are isobaths.
|
The ACC, which extends for more than 1500 km along the coast of
Alaska, is one of the strongest and most persistent coastal currents
in the world. Driven by both wind and buoyancy, this current directly
influences the addition and distribution of freshwater, biota and
pollutants around the Gulf of Alaska by confining the freshwater
nearshore and modifying coastal sea levels. In comparison, the transport
of water through Shelikoff Strait is equivalent to the flow from
about six Mississippi Rivers.
Run at a resolution of three kilometers, Hedstrom’s model
is aligned and nested within two larger models: the North Pacific
ocean grid model running at a resolution of 20 to 40 kilometers,
and the Northeast Pacific (NEP) model running at 10 kilometers, both
of which were developed by Hedstrom (ARSC), and Al Hermann and Liz
Dobbins of the Pacific Marine Environmental Laboratory (PMEL).
Data relating to atmospheric weather conditions that affect ocean
circulation is from the Comprehensive Ocean-Atmosphere Data Set (COADS).
COADS monthly mean fields were used in early test runs for the surface
forcing—the net sea surface heat flux and wind stress response
to atmospheric conditions. Hedstrom predicts that future runs will
be forced with daily winds from the National Centers for Environmental
Prediction (NCEP), or possibly the 5th-generation Penn State/NCAR
mesoscale model (MM5). Eventually, runs may combine ROMS with the
Weather Research and Forecast (WRF) model, a fully compressible,
three-dimensional, non-hydrostatic, cloud-resolving mesoscale model
designed to simulate phenomena on scales of 1 km to 10 km.
Mike Foreman of the Institute of Ocean Sciences in British Columbia,
Canada supplied the tidal-forcing data. Because tides cause the movement
of sediment along the littoral zone of the coast, they strongly influence
the form and structure of coastal deltas, tidal flats, barrier islands
and wave energy along the beach profile, as well as impact the mixing
of salt and fresh water in estuaries.
The freshwater boundary conditions for Hedstrom’s model are
taken from Tom Royer’s work on deep ocean and coastal hydrography
and currents. Currently with Old Dominion University in Virginia,
and formerly with UAF for many years, Royer’s work led to the
identification of the impact that freshwater discharge has in driving
the Alaska Coastal Current (ACC). It’s something like squirting
a fast stream of water from a hose into another, slower-moving body
of water. As the hose water mingles with the slower-moving water,
instabilities occur on the outer edges of the stream, causing eddies.
One way to record the impact freshwater discharge has on the ACC
is through the use of mechanical drifter buoys designed to follow
currents and collect physical data about the ocean. These Lagrangian
drifter buoys are based on a system in which two large bodies, Sun-Earth
or Earth-Moon, are the points to which a small third body, the buoy,
will maintain a fixed position relative to the other two. Named for
French astronomer Louis Lagrange, this type of buoy provides scientists
with direct observations of ocean surface currents. A unique feature
of Hedstrom’s SALMON Project model is that it can track, from
one processor to another, computer simulations of numerical Lagrangian
drifters along the coast of Alaska. The model uses 440 numerical
drifters broken into groups of twenty and set at depths of 5 and
fifty meters.
The latest model run has simulated three climatological years of
ocean circulation in the northern Gulf of Alaska. The numerical drifters
indicated highly variable flow and small-scale eddies between the
nearshore ACC and the farther offshore Alaska Stream. These eddies
appear to be seasonal, increasing during the summer when the fresh-water
influx is strongest. Observational data collected along one transect
line has previously shown that the ACC and the parallel Slope Current,
an area where salinity is increased just inside the Alaska Stream,
generally flow continuously in one direction, while another strong
current consistently flows between them in the opposite direction.
However, because the physical data were observed from one transect
line, and only sampled every other month, the eddy flow was not apparent.
To test the previously collected eddy flow data against the results
of Hedstrom’s model run, the SALMON Project, working in conjunction
with the Northeast Pacific GLOBEC (Global Ocean Ecosystem Dynamics)
program, undertook one of the most comprehensive surveys ever conducted
in Alaska waters. In May and July-August 2003, the GLOBEC Mesoscale
Survey was launched to map physical and biological oceanographic
characteristics associated with the ACC, focusing particular interest
on the continental shelf region south of Seward, Alaska.
During Part One of the survey conducted in May aboard the Alpha
Helix, an oceanographic ship operated by UAF’s Institute of
Marine Science, researchers towed a vehicle platform, called a SeaSoar™,
which was used to deploy a wide range of oceanographic monitoring
equipment. Because the SeaSoar™ is capable of undulating from
the surface to depths of 500 meters at tow speeds of up to 12 knots,
the SALMON team was able to obtain data much faster and with more
efficiency than in the past. They were also able to get five times
the resolution of the front they were surveying than is normally
achieved through the use of traditional hydro-cast methods, which
take longer and do not supply data while the ship is moving from
one station to the next.
“ We had thought the return flow was permanent, but Hedstrom’s
model showed that it changes over time,” said Dave Musgrave,
physical oceanographer with the Institute of Marine Science (IMS)
at UAF. “In the May survey, with the help of SeaSoar™,
we found that when we came back to the same location for another
cruise track seven days after the first track, the previous flow
had completely disappeared. Our new observations proved that Kate’s
model was acting correctly.” |
