Estuaries and coastal oceans are of immediate concern to us but are also
most challenging places to investigate and make predictions. They are constrained
by irregular coastlines and variable bathymetry, and forced by a complex array
of tidal, wind and buoyancy forces on a broad range of space and time scales.
Inhabited in these dynamic ocean regions are complex and diverse marine
ecosystems that are being threatened by human activities. Given the complex
nature of the coastal ocean, we are developing a hierarchy of numerical models,
ranging from simple box models to full-blown three-dimensional hydrodynamic
models.
Chesapeake
Bay
Chesapeake Bay is a partially-mixed estuary,
featuring a two-layer circulation with net seaward motion in a surface layer
and net landward flow in a bottom layer. Typical vertical salinity differences
of 2-8 stratify the water column, while salinity differences in the
longitudinal direction lie between 20 and 30, depending on the fresh water
input from rivers. The tidally-averaged
residual flows are of order of 0.1 ms-1. Moreover, the salinity
distribution shows large cross-channel variations, especially in the broader
reaches in the south end of the Bay. As fresh water plume moves seaward,
tending to the western shore, high-salinity ocean
water moves landward primarily through deep paleochannels.
Both tidal and wind mixing play an important role in determining the salinity
distribution and residual circulation. Compared with other estuaries, tidal
forcing in the Bay is relatively modest with tidal range rarely exceeding 1 m.
Winds are episodic with dominant periods of 2-7 days. Northwesterly winds
dominate in winter months (November-February) whereas southerly winds of
several days each are more frequent in the summer. During the fall transition
period, strong wind storms can occasionally destratify
the entire water column.
We have applied Regional Ocean Modeling System
(ROMS) to the Chesapeake Bay and used it to
investigate a number of dynamical processes in this estuary. The
following provides some examples.
a)
Tidal dynamics
We
have used a three-dimensional baroclinic model to
investigate tidal energy flux and dissipation in the Chesapeake
Bay. The model currents are validated by a comparison with
observed tidal current ellipses collected during previous field surveys. The
model elevations are validated against sea-level records collected at tidal
gauges. The total amount of tidal energy flux entering the Bay mouth is found
to be 177 W, 90% of which is associated with the semidiurnal lunar (M2)
component. Dissipation of tidal energy is highly non-uniform in the Chesapeake Bay. Most energy dissipation occurs in four
topographic hotspots: the Bay mouth region around the headland of Delmarva
Peninsula, the region south of the Rappahannock sill, the constriction near the
Bay Bridge
and the constriction north of Baltimore.
Figure 1.
Depth-integrated barotropic energy flux vectors for
the M2 constituent and distribution of energy dissipation rate per unit area
(b).
b) Hurricane-induced storm surges,
currents and destratification
Semi-enclosed
bays and estuaries are usually protected from hurricane-generated storm surges.
When a hurricane travels on the land side, however, it may induce high storm
surges, strong currents and destratification in the
water column. Real-time observations and numerical model prediction both show a
slab-like sloshing in Chesapeake Bay when it
was hit by Hurricane Isabel in September 2003. Strong southeasterly winds in
the right front quadrant of the storm forced water in Chesapeake Bay to move
northward as a single layer, producing high sea levels and flooding in the
northern Bay region including Baltimore and Annapolis. Furthermore,
the strong landward winds erased water-column stratification and caused a
strong intrusion of high-salinity shelf water into the Bay. After Isabel’s
passage, the longitudinal salinity gradient produces restratification
and two-layer circulation in the Bay.
Figure 2. Storm surges in Chesapeake Bay and its tributaries. ROMS-predicted
sea-level distribution: (a) a 3D view showing high storm surges in the Potomac
River and northern Bay at 0400 LST 19 September; (b) comparison of time series
of the predicted (black) and observed (red) sea levels at a few selected tidal
gauge stations.
Figure 3.
Water-column destratification due to wind mixing.
Predicted along-channel salinity distributions (a) before (0000 LST 16
September); (b) during (0000 LST 19 September); (c) after (0000 LST 22
September) Isabel’s passage, and (d) predicted non-tidal velocity after
Isabel’s passage (0000 LST 22 September).
c) Dynamics of estuarine circulation
As a
partially-mixed estuary, the Chesapeake Bay
features a two-layer circulation. The ROMS model reproduces this circulation
pattern. Figure 4 shows distributions of salinity and residual velocity along
the axis of Chesapeake Bay. Both salinity and
longitudinal velocity are averaged over the month of April to filter out the
tidal effects. Isohalines in the upper reaches of the estuary appear to be
vertical, reflecting strong vertical mixing in this shallow region. South of 39.3 latitude, isohalines slope upward toward the sea, as
is shown in Figure 4a. Due to strong tidal mixing in the bottom boundary layer,
isohalines are nearly vertical near the bottom. Top-bottom salinity differences
in the middle and lower reaches of the Bay vary between 5 and 8, while the
head-mouth salinity difference lies between 20 and 25. The longitudinal
salinity difference in the deep channel (between 37.7 and 39 latitudes) is
around 10. Figure 4b shows the axial component of the residual velocity. Water
moves seaward in the surface layers and landward in the bottom layers. Maximum
residual velocity is about 0.2 ms-1. The level of no motion
separating the two counter-flowing layers lies somewhere between 5 and 8
meters. The landward flow in the deep channel reaches the maximum speed at a
depth of 15 meters, well above the bottom. Figure 5 shows the tidally-averaged
salinity and residual current at the surface of Chesapeake
Bay. Due to the effect of Coriolis
force, the fresh water plume hugs the western shore as it moves seaward and
exits as a buoyancy-driven boundary current along the adjacent shelf. This
fresh water plume provides an important mechanism for transporting land-derived
inorganic and organic materials from watersheds to the ocean.
Figure 4. Distributions of
monthly-averaged (April, 1996) salinity (a) and longitudinal velocity (b) in an
along-channel vertical section.
Figure
5. Tidally-averaged salinity (a) and residual flow (b) at the sea surface of Chesapeake Bay.
Strait of Georgia and Juan
de Fuca Strait
The Strait
of Georgia and Juan de Fuca Strait are two coastal
ocean basins situated between Vancouver Island and the mainland coasts of British Columbia and Washington State.
The two straits are connected through Haro Strait-- a narrow channel with sills at its southern
and northern ends. A primary forcing of this system is the fresh water runoff
from the Fraser River, which peaks during the summer
freshet. A plume of brackish, silt-laden water is often seen to spread over the
oceanic water in the southern part of the Strait of
Georgia. The estuarine flow is characterized by a seaward outflow
in the upper portion of the water column and a landward inflow below.
Figure 6. A map of
Georgia-Fuca estuary on the west coast of North America.
We have developed a box model for the
Georgia-Fuca estuarine system. We divide this system
into three basins: Georgia
basin, Haro
Strait and Fuca basin, each of which consisting of an upper box and a
lower box. The main exchange between the estuary and the Pacific Ocean is at the western end of the Juan de Fuca Strait. The
following figure gives an example of multi-year simulations of the box model.
Both the river off (top panel) and the salinity in the Pacific
Ocean (second panel) exhibit year-to-year fluctuations. These
fluctuations transmit to the Georgia-Fuca estuary, as
shown in the time series of box salinities (third panel, green lines for Georgia boxes,
red lines for Haro boxes and blue lines for Fuca boxes). The bottom panel simply shows that the
salinity budget in the estuary remains balanced every year. Our model reveals a
rapid response of the estuarine circulation to interannual
variability in the fresh water forcing and in the properties of the continental
shelf water. Hence the Georgia-Fuca estuary will
respond passively to the large-scale climate variability in the North
Pacific.
Figure
7. Time series of river runff (a), offshore salinity
(b), box salinities (c) and volume fluxes (d) predicted from the box model of
Georgia-Fuca estuary.
South-China Sea
The South China
Sea (SCS) is a semi-enclosed marginal sea in the western North
Pacific Ocean. The circulation of the northern SCS is strongly
influenced by the Kuroshio, the western boundary
current of the North Pacific, which frequently intrudes into the northern SCS
through the Luzon
Strait, the only deep
passage connecting SCS to the open Pacific. The intruded Kuroshio
moves westward and collides with the Dongsha Islands
off the China
coast. The collision forces most of the flow to return as a strong return flow
to the north and a splinter current to the southwest along the upper
continental slope. The return loop occasionally spawns anti-cyclonic rings
which are instrumental in transporting tropical heat and salt into the northern
SCS. The splinter current called South China Sea Branch of the Kuroshio (SCSBK), on the other hand, feeds a northeastward
shelf-break current that arises due to the alongshore pressure variation
imposed offshore by the collision action, at the Dongsha
Islands, of the intruded Kuroshio.
Figure
8. A map of South China Sea.
The GFDL ocean model is applied to study the
circulation in the northern South China Sea
and it reproduces the main features. Model output shows the existence of
step-like pressure distribution along the shelf break. In a reduced-gravity inviscid model ocean, the step-like pressure distribution
results from the deflection of a constant-potential vorticity
current at a step-shelf fronted coast. This step-like pressure distribution
accounts for the year-round existence of the northeastward shelf-break current
(the so-called South China Sea Warm Current). A SCRUM model demonstrates that SCSBK
feeds the northeastward shelf-break current all along the shelf-break, making
the latter a warm current.
Figure
9. Simulated sea level and currents in South China Sea.
Publications:
Li, M., L. Zhong, and W. C. Boicourt. 2005. Simulations of Chesapeake
Bay estuary: Sensitivity to turbulence mixing parameterizations
and comparison with observations, J. Geophys. Res., 110, C12004, doi:10.1029/2004JC002585.
Li, M., L. Zhong, W. C. Boicourt, S. Zhang and D.-L. Zhang. 2005. Hurricane-induced
storm surges, currents and destratification in a
semi-enclosed bay. Geophys. Res. Lett.,
In press.
Zhong, L. and M. Li. 2005.
Tidal energy fluxes and dissipation in the Chesapeake Bay. Cont. Shelf Res., In review.
Hsueh, Y., and L. Zhong. 2004. A pressure-driven South China Sea
Warm Current. J. Geophys.
Res., 109, C09014, doi:10.1029/2004JC002374.
Hsueh, Y., and L. Zhong.
2003. A note on the deflection of a baroclinic current by a continental shelf. Geophys. Astrophy. Fluid Dyn., 97(5), 393-415.
Li, M., A. Gargett and K.L. Denman. 1999. Seasonal and interannual variability of estuarine circulation in a box
model of the Strait of Georgia and Juan de Fuca Strait. Atmos-Ocean, 37,
1-19.
