Authors: Malcolm J. Bowman, David E. Dietrich, Avichal Mehra
Preliminary results from a Southwest Pacific Ocean regional model with 20 levels, 1/4 degree resolution, based on a new reduced dispersion, modified Arakawa "a" grid version of the dieCast model are described.
The model covers the Australian eastern seaboard from the southern tip of Tasmania up to Cairns, Queensland, the southern Coral Sea, Tasman Sea, a segment of the Southern Ocean and the southwestern Pacific Ocean surrounding New Zealand (Fig. 1a, 1b).
Boundaries are 146° E (Tasmania), 18° S (Fiji Islands), 160° W, and 65° S.
Topography is unfiltered etopo5 (Fig. 2, which has defects in nearshore waters that need correction), wind field is Hellerman annual mean (Fig. 3), temperature and salinity climatology is Levitus.
All four boundaries are open. Inflows are specified on the western and northern boundaries (Circumpolar Current and cross equatorial drift feeding the East Australian Current-(EAC). The eastern and southern boundaries are outflow boundaries in which the flow is adjusted according to the internal dynamics. The western boundary south of Tasmania admits a ~93 Sv Circumpolar Current.
The northern inflow off the east Australian coast is mainly a thermal wind associated with the Levitus annual mean climatology plus a weak barotropic component. The thermal wind is integrated from zero at level 14 (1450 m below msl) to the surface to give a maximum inflow of about 1.2 cm sec-1, which occurs about 28 points (7°, or about 700 km) east of the Australian coast. The total baroclinic inflow is about 7 Sv.
In addition to the thermal wind, a uniform barotropic northward drift of 0.002 cm sec-1 is specified at the northern boundary. This gives about 0.3 Sv outflow, so the net inflow is about 6.7 Sv.
The western inflow is mainly a thermal wind from Levitus annual mean climatology plus a weak barotropic component. The thermal wind is integrated throughout the water column to give a maximum inflow of about 12 cm sec-1, which occurs about 109 points north of the southern boundary (~ 51° S). The total baroclinic inflow is ~ 93 Sv.
In addition to ~ 93 Sv from the thermal wind, a uniform weak barotropic eastward drift of 0.1 cm sec-1 is specified at the western boundary. The total inflow from the west is ~ 101 Sv.
The eastern and southern boundaries are outflow-only boundaries. The outflow is adjusted by advecting the interior (upwind) velocity to the boundary. Longitudinal variations are damped near the southern and northern boundaries. Latitudinal variations are damped near the eastern boundary (this allows simple 1D upwind advection to update normal velocity at the outflow boundaries). Planned future nesting in a global model will decrease the guesswork and may eliminate the need for such sponge layers near outflow boundaries.
The temperature and salinity surface fields and the general circulation for day 720 is shown in Figs. 4-6. Details of surface features are discussed below.
The East Cape Current forms a western boundary current consisting of a train of counterclockwise rotating anticyclones off the east coast of the North Island of New Zealand (Figs. 7-9) with a particularly well developed persistent anticyclone located off the southern tip of the North Island, known as the Hikurangi Eddy (Bowman, 1985). This eddy can be seen in a color animation of model results from days 370-460.
This shows up clearly in the AVHRR thermal IR image of Fig. 10. The flow separates and flows zonally along the northern flank of the Chatham rise. The flow patterns should be compared with the dynamic topography calculations of Heath (1975; Fig. 11) which shows the presence of the Hikurangi Eddy and meandering flow off to the east.
A strong coastal current flows northward along the west coast of the South Island, splitting near Banks Peninsula with a minor component squeezing between the mainland and the Mernoo seamount, before flowing into the approaches to eastern Cook Strait (Fig. 9). Large scale meandering of the Circumpolar Current can be seen near the bottom of Fig. 9.
The Tasman Front (Denman and Crook, 1976; Stanton, 1976) has been observed to lie zonally across the Tasman Sea, stretching from the tip of the North Island (Fig. 12). The front shows up clearly in the surface salinity plots of Fig. 6 and 8, suggesting that the Tasman Front is the southern boundary of the southwest Pacific subtropical gyre. The Tasman front also is exhibited in the temperature field (Figs. 5, 7) and is centered at about 18° C in our calculations.
The East Australian Current pinches off vigorous warm core anticyclonic eddies south of Brisbane which have been observed to propagate slowly down the coast southward after pinching off (Figs. 13, 14).
Results (Fig. 15, 16) display an apparently realistic EAC as a series of pinched off anticyclonic eddies of ~ 250 km dia. and maximum swirl velocities ~ 1.5. m sec-1. They form south of Brisbane at about 30° S and propagate southwards down the coast past Sydney. Near Cape Howe (lat. 38° S) their southwards drift is arrested, and they bleed vorticity into the Tasman frontal zone and eventually dissipate, before being replaced by the next younger eddy in line. Eddy lifetime is ~ 2 year.
The along-shore propagation of the separated eddies is nonlinear with maximum propagation speeds of ~ 5 cm sec-1. Eddy intensities are quite sensitive to the strength of the undercurrent (see below).
The vortex shedding period of the model EAC is roughly 300 days (sheds eddies at days ~ 460, 860, 1095, ...). Eddy rotational periods are ~ 5 days, which compares favorably with observations (Cresswell, 1982; Bennett, 1983).
Younger eddies have been observed to overtake slower, older eddies and coalesce over a 20 day period (Cresswell, 1982; Cresswell and Legeckis, 1985). This has been observed in the model, centered around days 350 and 980. The sequence of events is captured in Fig. 16a,Fig. 16b,and Fig. 16cand a color animation.
Experiments show that even a weak EAC equatorward undercurrent significantly decreases the strength of the pinched off anticylonic EAC eddies. With no undercurrent (i.e., no barotropic addition to the baroclinic thermal wind as described above), we find that the modeled warm core eddies slowly spin up to unphysically large amplitudes as they continue to absorb vorticity from the north.
Thus, the undercurrent may play a significant role in the dynamics of the pinched off eddies even though it is weak and may be difficult to measure. We conclude that the undercurrent must be carefully specified to allow realistic behavior of pinched off EAC eddies.
As part of the EAC eddy shedding cycle, sometimes anticyclonic eddies form in the southern Coral Sea area and propagate to the EAC separation point near Brisbane, before flowing eastward across the Tasman Sea around the North Cape of NZ.
Bennett, A.F. 1983. "The south Pacific including the East Australian Current", Eddies in marine science, A.R. Robinson (ed). Springer-Verlag, NY, pp 609.
Bowman, M.J. 1985. "On the B-induced coastal trapping of a baroclinic eddy" J. Phys. Oceanogr. 15: 817-822.
Cresswell, G.R. 1982. "The coalescence of two East Australian Current warm-core eddies". Science 215: 161-164.
Cresswell G.R and R. Legeckis. 1986. "Eddies off southeastern Australia". Deep-Sea Res., 33: 1527-1562.
Denham, R.N. and F.G. Crook. 1976. "The Tasman Front". N.Z. J. Mar. Freshwater Res., 10:15-30.
Heath, R.A. 1975. "Oceanic circulation off the east coast of New Zealand". New Zealand Oceanogr. Inst. Memoir 55, Wellington, 80 pp.
Hamon, B.V. 1965. "The East Australian Current, 1960-1964". Deep-Sea Res., 12:899-921.
Nilsson, C.S. and G.R. Creswell. 1980. "The formation and evolution of East Australian Current warm core eddies". Prog. Oceanogr. 9:133-184.
Stanton, B.R. 1976. "An oceanic frontal jet near the Norfolk Ridge northwest of New Zealand". Deep-Sea Res., 23:1207-1219.
This research was supported by ONR(Office of Naval Research) and the School of Environmental and Marine Sciences at the University of Auckland. Computing facilities were provided by the Department of Engineering Science. Special thanks are due to Profs R. Meyer and M. O'Sullivan for assistance in supporting the research and to Steve Payne for assistance in setting up this web page.
Fig. 1a: Relative Sea Surface Height (SSH) at Day 370 from DieCAST Ocean Model.
Fig. 1bGeneral circulation and position of fronts in the south-west Pacific Ocean according to Heath (1985).
Fig. 2: Unfiltered water depths taken from the etopo5 data set. Defects near coastlines need to be repaired.
Fig. 3: Mean Hellerman winds for the model domain.
Fig. 4: Surface pressure contours and velocity vectors for the domain for day 720.
Fig. 5: Surface temperature field for the domain for day 720.
Fig. 6: Surface salinity field for the domain for day 720.
Fig. 7: Surface temperature for the seas surrounding New Zealand, day 151.
Fig. 8: Surface salinity for the seas surrounding New Zealand, day 151.
Fig. 9: Surface pressure contours and velocity vectors for the seas surrounding New Zealand, day 151.
Fig. 10: Thermal IR sea surface temperature image for 28 March 1990 (courtesy Andrew Shaw, Department of Marine Science, University of Otago).
Fig. 11: Composite dynamic topography contours relative to 1000 dbar for February/March 1963 and 1965 (from Heath, 1975).
Fig. 12: Sea surface temperature and geopotential anomaly for the Tasman Front (from Stanton, 1976).
Fig. 13: Surface dynamic topography, East Australian Current, Jan-Feb 1964 (from Bennett, 1983; Hamon, 1965).
Fig. 14: XBT track and surface dynamic topography, East Australian Current, Mar 1977 (from Bennett, 1983;Nilsson and Cresswell, 1980).
Fig. 15: Surface pressure contours and velocity vectors for East Australian Current, day 902.
Fig. 16a,Fig. 16b,Fig. 16c: Sequence of surface pressure and velocity vectors for days 902, 940, 950, 960, 970, 980, 990 and 1000, showing the coalescence of two anticyclonic eddies.
Authors: |
| Malcolm J. Bowman Marine Sciences Research Center State University of New York Stony Brook NY 11794-5000 Home Page |
| David E. Dietrich Center for Air Sea Technology Mississippi State University Stennis Space Center MS 39529 |
| Avichal Mehra Center for Air Sea Technology Mississippi State University Stennis Space Center MS 39529 |
| Back to the DieCAST Homepage |
 IDSL
Homepage |
