Modeling the coupled ocean—Katabatic wind systems of the Antarctic

Modeling the coupled ocean—Katabatic wind systems of the Antarctic RICHARD T. MCNIDER, Department of Mathematical Sciences, University ofAlabama, Huntsville, Alabama 35899 Scorr L. G00DRIcK, Atmospheric Science Program, University ofAlabama, Huntsville, Alabama 35899

ents which develop along the sloping topography. This is hanhe wind stress distribution along the periphery of the dled in the current formulation by working with a perturbation T antarctic continent is like no other place on Earth (Gill pressure formulation and analytically balancing the base-state 1982). A large easterly component near the coast is evidently pressure. attributable to the geostrophic adjustment of the katabatic The ocean model boundary layer (and free ocean) winds driven by the cold antarctic plateau (Parish 1988). The employs either a local closure scheme based on the gradient large curl between these topographically driven easterlies and Richardson number (McNider and Pielke 1981) or a level-2.5 the strong Southern Hemispheric westerlies may have a sigclosure which includes a prognostic turbulent kinetic-energy nificant role in the upwelling and associated biological proequation (developed by Arritt and Physick 1989). Thus, both ductivity of the southern oceans. The direct downslope katashear-induced and convectively induced mixing can be parabatic flows and their alongshore adjustment may also have meterized. significance to polynya development, bottom-water formaWe report here on preliminary simulations for an ice-free tion, and the maintenance of the coastal currents including ocean including only one-way coupling. In the atmospheric the East Wind Drift (Bromwich and Kurtz 1984; Zwally, model katabatic flows developed rapidly in the first 12 hours Comisco, and Gordon 1985). of simulation. Gradually, due to Coriolis deflection, a crossThis paper reports on preliminary model studies of the slope or alongshore component developed. For the ice-free coupling between antarctic katabatic flows and the coastal ocean a substantial thermal gradient developed near the ocean. Although considerable past work in coupling largecoast. The katabatic flows in the coastal area were substanscale and low-order ocean/ atmosphere models has been tially enhanced leading to a secondary closed circulation. The done, relatively less work on coupling true mesoscale multididepth of the flow over the water was also substantially mensional models containing high-resolution boundary layincreased due to the thermal instability over the water. The ers has been undertaken. The following describes a coupled most interesting feature is that a secondary jet in the alongatmosphere/ ocean model which has been developed by coast direction developed offshore, and the combined kataMcNider from the frameworks of a mesoscale atmospheric batic/thermal contrast wind extended beyond 200 kilometers model (Pielke 1974) and a coastal ocean model (McNider and (km) offshore. Figure 1 shows velocity contours in the alongO'Brien 1973). The mesoscale atmospheric model is a hydrostatic mesoscale model based on 10 .0 an original formulation by Pielke (1974) with subsequent reformulations and improvements by several investigators 2.3 including Mahrer and Pielke (1977), 7.6 McNider and Pielke (1981), and Arritt and Physick (1989). The ocean model structure is the same as the atmospheric model, and in h fact, most of the model code was taken from the atmospheric mesoscale model. Therefore, coupling the two models is rather straightforward. The model employs a bottom topography-following 2. coordinate system with the upper (ocean surface) coordinate being the ocean free surface. This coordinate system has several numerical advantages because the 0 300 600 900 1200 1500 ocean surface is a coordinate surface, and KM yet topography can be dealt with in a simple manner. The major disadvantage of Figure 1. Alongshore wind velocity contours from the atmospheric model after 10 days (consuch a coordinate system is in dealing tour interval is 3 meters per second). Positive flow is from the east. (h denotes high. I denotes with the large hydrostatic pressure gradi- low. KM denotes kilometer.) I1IIIuI I IlITlIII!lII ilIll'] 111111 1111111' II

ø . -

01

ANTARCTIC JOURNAL - REVIEW 1994

291

1-,uuu,rI,uIIIuI

1.6

320.4

1.4

7.2 too

1.0 U) U) Cl)

•0

.8

60 120 180 240 300 KM Figure 3. Alongshore velocity contours in the ocean model after 10 days of forcing by the katabatic winds (contour interval is 2 centimeters per second). Positive flow is from the east. (M denotes meter. KM denotes kilometer.)

i. ' r.

A n 0 *-I I 0 100 300 500 700 900 1100 1300

Distance Offshore (Km)

and boundary effects led to downwelling in the ocean shoreward of approximately 100 km and upwelling from 100-300 km offshore. The sea surface slope actually sloped upward to the coast due to the shoreward Ekman transport. This led to an alongshore current which evidently corresponds to the East Wind Drift (see figure 3.) Later as the alongshore stress maximum propagated offshore, a more complicated flow pattern was produced with offshore flow returning near the coast and associated upweffing at the coast. This research was supported by the National Science Foundation grant ATM 91-20321 and by the National Institute for Global Environmental Change Southeast Center (Department of Energy Cooperative Agreement DE-FCO3-90ER61010). Financial support does not represent endorsement by the Department of Energy of the views expressed in this article.

1.3 1.2 1.1 1.0

U,

U)

.7

(0 C .6

References

d,-ot-.k L 6.. 0 100 300 500 700 900 1100 1300

Arritt, R.W., and W.L. Physick. 1989. Formulation of the thermal inertial boundary layer in a mesoscale model II. Simulations with a level-2.5 turbulence. Boundary Layer Meteorology, 49, 411-416. Bromwich, D.H., and D.D. Kurtz. 1984. Katabatic wind forcing of the Terra Nova Bay polynya. Journal of Geophysical Research, 89, 3561-3572. Gill, A.E. 1982. Atmospheric-ocean dynamics. Orlando, Florida: Academic Press. Mahrer, Y., and R.A. Pielke. 1977. A numerical study of the air flow over irregular terrain. Contributions in Atmospheric Physics, 50, 98-113. McNider, R.T., and J.J. O'Brien. 1973. A multi-layer transient model of a coastal upwelling. Journal of Physical Oceanography, 3,258-273. McNider, R.T., and R.A. Pielke. 1981. Diurnal boundary layer develop-

Distance Offshore (Km) Figure 2. Evolution of (top) meridional and (bottom) zonal wind stress components as a function of distance offshore. Curve A is the stress profile after 2 days with subsequent curves at 2-day intervals through 10 days (curve E). Values are in dynes per square centimeter. (Km denotes kilometer.) shore direction whereas figure 2 shows the evolution of the windstress distribution. Note the jet distribution and the substantial curl in the zonal stress distribution, both of which are important to the ocean dynamics. This alongshore atmospheric jet gradually propagated offshore. For the case of the ice-free ocean, the ocean model was run using the windstress produced by the atmospheric model. Although the early direct offshore component initially drove a weak near-surface flow offshore, the ocean responded primarily to the alongshore atmospheric component. This alongshore wind stress led to an onshore flow in the surface layer of the ocean. In the first few days of the simulation, because of the jet structure in the alongshore direction, wind stress curl

ment over sloping terrain. Journal

of Atmospheric Science, 10,

2198-2212. Parish, T.R. 1988. Surface winds over the antarctic continent: A review. Reviews of Geophysics, 26,169-180. Pielke, R.A. 1974. A three-dimensional numerical model of the sea breeze over south Florida. Monthly Weather Review, 102, 115-139. Zwally, H.J., J.C. Comisco, and A.L. Gordon. 1985. Antarctic offshore leads and polynyas and oceanographic effects. In S.S. Jacobs (Ed.), Oceanology of the antarctic Continental Shelf (Antarctic Research Series, Vol. 43). Washington, D.C.: American Geophysical Union.

ANTARCTIC JOURNAL 292

REVIEW 1994