Department of the Environment

About us | Contact us | Publications

Header imagesHeader imagesHeader images

Publications

Disclaimer

Key departmental publications, e.g. annual reports, budget papers and program guidelines are available in our online archive.

Much of the material listed on these archived web pages has been superseded, or served a particular purpose at a particular time. It may contain references to activities or policies that have no current application. Many archived documents may link to web pages that have moved or no longer exist, or may refer to other documents that are no longer available.

State of the Marine Environment Report for Australia: The Marine Environment - Technical Annex: 1

Compiled by Leon P. Zann
Great Barrier Reef Marine Park Authority, Townsville Queensland

Ocean Rescue 2000 Program
Department of the Environment, Sport and Territories, Canberra, 1995

ISBN 0 642 17399 0

The oceanography of Australian seas

Jason H. Middleton

Centre for Marine Science
University of New South Wales
Sydney NSW 2033

Introduction

The study of physical oceanography encompasses the description and understanding of all physical processes occurring in the ocean. The magnitude of the processes ranges from those extending over many thousands of kilometres and several years for large scale ocean circulation processes, to those covering only centimetres and lasting for as short as seconds for small scale processes. In the space available here, all aspects of physical oceanography cannot be adequately covered. For example, discussions about possible sea level rise (Church et al. 1991) or climate variability studies (Ramage 1966, Meyers et al. 1991) are not included; and a description of the weather affecting the Australian region can be found instead in the 'Manual of meteorology' produced by the federal Bureau of Meteorology. Instead, this report concentrates on introducing the main areas of physical oceanography relevant to Australian waters and directing the reader to literature references for more detailed explanations.

Temperature and salinity in the upper ocean

The upper ocean consists of the top few hundred metres where changes occur on time scales of a few days to many months.

There are many factors controlling the physical properties of the upper ocean. Among the most important are incoming and outgoing radiation, precipitation and evaporation, sea-ice formation and melting, mechanical energy inputs and mixing processes. Some factors tend to make the water column more stable (more strongly stratified) by reducing the density of the upper layers. For example, incoming radiation tends to heat the surface layers through absorption and they are then less dense than the layers beneath. Precipitation, river runoff and sea-ice melting provide fresher water to the surface layers, also rendering them less dense as a result of the reduced salinity. Although vertical mixing is somewhat inhibited by a stably stratified water column, strong winds and wave action tend to mix the surface layers through mechanical action. This action creates a relatively homogeneous 'mixed layer' of less dense fluid in the upper 100 m of the ocean. The confluence of water masses at zones of convergence (ocean fronts) can also generate substantial vertical mixing.

In the tropics, water temperature decreases rapidly with depth. The region of maximum temperature change is called the thermocline. The main thermocline is centred at approximately 500 m depth in the tropics and mid-latitude oceans, while a seasonal (summer) thermocline often exists at approximately 100 m depth in mid-latitudes. While temperature and salinity may change with depth, the combined effect of the two usually is such that density increases with depth. The level of maximum salinity change with depth is called the halocline, while the level of maximum density change with depth is called the pycnocline.

Evaporation, outgoing radiation, cooling by contact with colder air or the formation of sea-ice tend to render surface layers denser than those below by virtue either of a lower temperature or an increased salt content (salt is rejected into the sea below during sea-ice formation). Denser fluid overlying less dense fluid is an unstable situation and it causes the water column to overturn, generating substantial 'convective' mixing. Since the density of the ocean usually increases with depth, the newly formed denser fluid at the surface usually is not dense enough to cause the whole water column to overturn. Instead, a convective mixed layer of some tens of metres depth is formed. In Antarctic waters, convective mixed layers of approximately 500 m depth form in winter.

The distribution of water properties in the world's oceans is outlined by Sverdrup, Johnson and Fleming (1942) and by Pickard and Emery (1982) and the annual and seasonal climatologies are now available in digital and microfiche form (Levitus 1982).

Figure 1

Figure 1: Major oceanic currents affecting the Australian coastline.

Large scale ocean circulation

Wind-driven circulation

The overall result of westerly winds in the mid-latitudes and easterly winds in the tropics is to drive the ocean currents in the major ocean basins in large closed circulation patterns called gyres. Gyres rotate anti-clockwise in the southern hemisphere (Pickard & Emery 1982). In the Southern Ocean, which extends all around the Antarctic continent, the prevailing westerly winds drive the Antarctic Circumpolar Current to the east although there is evidence of strong variability (Morrow et al. 1992).

Because the earth is spherical and rotating, ocean currents tend to intensify toward the western boundary of the ocean basins. These strong but highly variable currents are called western boundary currents, and their variability is often sufficiently strong to mask the general poleward direction of their flow. The East Australia Current (EAC) is an example of a western boundary current (Figure 1). In the western Pacific Ocean, the vast number of shallow areas containing comprised of coral reefs and islands have the effect of weakening the western boundary current effect - although at times the EAC flows very strongly off Australia's eastern coast. This current flows south from the Coral Sea (Church 1987), hugging the continental slope until the region of Forster (central New South Wales) where it tends to run further offshore. Once or twice a year the EAC strains itself into loops in the Tasman Sea off the New South Wales coast. These loops become detached from the current and form EAC eddies, warm disc-shaped water parcels having a diameter of 200-300 km, and extending 1500-2000 m into the ocean (Nilsson & Cresswell 1981, Bennett 1983). These eddies may be seen from satellites fitted with infrared detectors (Cresswell & Legeckis 1986).

Off the west coast of Australia the deep ocean currents are highly variable and do not necessarily flow to the north as might be expected from a traditional gyre model (Cresswell & Golding 1980, Bye 1983). The reasons for this are not clear but are generally thought to be associated with flow through the Indonesian Archipelago from the Pacific Ocean to the Indian Ocean (Godfrey & Weaver 1991), which tends to drive a fairly persistent westward flowing current off north-western Australia (Holloway & Nye 1985). Instead, the Leeuwin Current flows southward in the deeper waters off the Western Australian coast, and this is occasionally strengthened by winds. The Leeuwin Current is, however, primarily driven by thermosteric effects associated with the extremely warm waters which come from the region of the North West Shelf (Weaver & Middleton 1989). It is a seasonal current, reaching its maximum strength about mid-year (Cresswell 1991, Smith et al. 1991). It is also highly variable both in its overall structure and in the detail of its turbulent nature (Batteen & Rutherford 1990).

Winds blowing parallel to a coastline can produce upwelling or downwelling effects through the surface Ekman layer transport, which transports surface waters perpendicular to the wind direction. In particular, coastal upwelling is induced whenever a wind blowing parallel to the coast has the deeper ocean to the left of the direction of the wind vector. The earth's rotation causes a surface Ekman layer to transport surface waters offshore, and these must be replaced by the upwelling at the coast of colder, nutrient-rich waters from below. Wind-driven upwelling does not appear to be persistent along any particular section of the Australian coast, but happens from time to time at many locations (eg Rochford 1975, Schahinger 1987, Griffin & Middleton 1992).

Thermohaline circulation and water masses

In this section the principal characteristics of water mass types and the thermohaline circulation in the oceans around Australia are outlined. In the tropical Indian Ocean and western Pacific Ocean, surface temperatures often exceed 28C with reduced solar radiation causing a gradual decrease in temperature at higher latitudes. More rapid reductions in temperature occur at the Subtropical Convergence (40-45S) and again at the Antarctic Convergence (55-60S). Tropical sea surface temperatures in the deep ocean change little with season while a typical seasonal variation in most non-tropical oceanic Australian waters is approximately 5oC. On the continental shelf of north-western Australia the summer heating provides a substantial warm pool of surface waters, and this provides the source region for the Leeuwin Current. Surface temperature maps for February and August for the world's oceans are shown by Sverdrup, Johnson and Fleming (1942: Charts II and III) and by Pickard and Emery (1982: Figures 4.3 and 4.4), and here for Australian waters in Figures 2a (February) and 2b (August).

Surface salinities change little with the seasons being between 34 psu and 35 psu (Practical Salinity Units) in the tropics and at latitudes greater than 40S, and between 35 psu and 36 psu at mid-latitudes. Influences of Antarctic sea-ice melting on surface waters are not observed much further north than the Antarctic Convergence.

Below the surface waters to about 800 m depth lies South Pacific Central Water (to the east of Australia) and Indian Central Water (to the west of Australia) extending in latitude from the equatorial waters to the Subantarctic Convergence.

Low salinity, cold Antarctic surface waters meet subantarctic waters at the Antarctic Convergence. The resulting dense mixture sinks and flows slowly northward below the Central Water at an average depth of approximately 1000 m. This colder, lower salinity Antarctic Intermediate Water is seen in both the Indian and Pacific oceans as far north as the equator (England 1992). Below the Antarctic Intermediate Water, and flowing southward at depths between 1000 m and 3000 m in the Pacific, lies Pacific Deep Water. Pacific Deep Water is originally of Atlantic/Antarctic origin, and deep waters of the Indian Ocean have similar properties to those of the Pacific. The formation of Antarctic bottom water in the Weddell Sea and at other continental margins of Antarctica (Middleton & Humphries 1989) results in a dense, cold water mass which sinks to abyssal depths off Antarctica and flows slowly northward contributing to the properties of the deep water masses in the Atlantic, Pacific and Indian oceans.

Pickard and Emery (1982) presented temperature-salinity diagrams for the South Pacific and Indian oceans (Figure 7.13) and north-south vertical sections of water properties in the central Pacific (Figure 7.34). The general description of water masses and currents of the world's oceans given by Sverdrup, Johnson and Fleming (1942) is both lucid and generally valid.

Continental shelf circulation

Wind-driven circulation

Weather patterns change on both seasonal and synoptic (one to three weeks) time scales. In shallow continental shelf waters, where the water column has much less mass than in the deep ocean, wind stress can generate substantial currents in a day or so. These winds might change direction on seasonal and/or synoptic time scales, each time accelerating continental shelf waters to flow along the coast. The current pulses forced by the synoptic-scale wind reverse direction in response to the along-shore component of the wind stress, but the pulses themselves propagate in a wave-like manner along the continental shelf with the coast to the left of the direction of propagation. Such propagating current pulses are called continental shelf waves or coastal trapped waves.

Figure 2

Figure 2: Sea surface temperature maps of Australian waters in February (top) and August (bottom).

Continental shelf waves propagate at speeds of about 4-10 m a second, with faster speeds on wider continental shelves. The associated currents may be 0.1-0.2 m per sec, and over a two-week time period can transport particles several hundreds of km along-shore. Wind-driven continental shelf waves have now been identified as a primary source of current variability in practically all Australian continental shelf waters such as the Great Barrier Reef (Andrews 1983, Middleton & Cunningham 1984, Burrage, Church & Steinberg 1991, Cahill & Middleton 1993), the New South Wales shelf (Freeland et al. 1986), the Great Australian Bight (Provis & Lennon 1981), Bass Strait (Middleton 1991, Middleton & Viera 1991) and the North West Shelf (Webster 1985). It is likely that much of the variability observed by Godfrey, Vaudrey and Hahn (1986) off the south coast of Australia is also due to wind-driven continental shelf waves. On most Australian shelves there are mixed contributions of locally wind-forced shelf waves and freely propagating shelf waves (generated by wind-forcing farther upstream in the sense of continental shelf wave propagation), enhancing the difficulty of current prediction. A schematic diagram of surface and sub-surface circulation patterns for continental shelf waves in the southern Great Barrier Reef reproduced from Griffin and Middleton (1986) is shown in Figure 3. While this diagram is strictly valid only for the area for which it was drawn, the general features are typical and the complexity of the flow structure is well illustrated.

An upwelling of nutrient rich waters from the deeper waters of the continental slope is associated with that phase of the continental shelf wave which has current pulses flowing with the coast to the right of the current direction. At times this upwelling is strong enough to bring extra nutrients into the euphotic zone. Continental shelf waves may have a modal structure whereby waters both deeper and further offshore oscillate in opposite phase to the nearshore surface waters.

In some parts of the world, seasonal winds are sufficiently strong to produce consistent alongshore coastal currents and persistent upwelling. A few persistent upwelling zones appear to exist off the Australian coast (Andrews & Gentien 1982, Schahinger 1987) although more areas may be identified once a history of sea surface temperature pictures taken from satellites is acquired.

Figure 3

Figure 3: Schematic diagram of surface (a) and subsurface (b) circulation patterns for continental shelf waves in the southern Great Barrier Reef.

Between November and May cyclones are a common feature of Australian tropical waters. They can generate strong currents and high mean sea levels (Fandry & Steedman 1989, Hearn & Holloway 1990) as well as high seas.

In general, wind-driven continental shelf circulation patterns have along-shore currents which are approximately 10 times faster than across-shelf currents. This feature is essentially a result of the presence of the coast.

Deep ocean effects on circulation

In many cases the offshore deep ocean currents flow in the opposite direction to the local-wind-driven continental shelf currents, and in these cases there is often a (changing) zone of demarcation. Inshore of the zone the currents are primarily wind-driven while offshore from the zone the currents are primarily driven by deep ocean effects. In such circumstances, prediction and interpretation of current speed and direction can be extremely difficult.

The situation described above is often true off the New South Wales coast, and analyses of current data from the Australian Coastal Experiment (Freeland et al. 1986) was hampered by the existence of the strong southward-flowing East Australian Current (EAC). The EAC is a strong, highly versatile current which is fed by the South Equatorial Current in the Coral Sea (Burrage 1993) and which hugs the continental shelf as it flows poleward. It tends to separate from the coast near Cape Hawke (New South Wales) and often forms large eddies and meanders. The Leeuwin Current, flowing southward along the outer continental shelf of Western Australia, appears to be primarily driven by deep ocean effects (primarily the flow from Pacific to Indian oceans via the Indonesian Archipelago) but strongly affects shelf circulation along the entire west coast (Godfrey & Ridgway 1985, Weaver & Middleton 1989).

On occasions the different temperature characteristics or turbidity of shelf and ocean waters enable easy identification by airborne or satellite infrared scanners (Cresswell et al. 1983).

Tidal circulation

The apparent motion of the sun and the moon around the earth at regular intervals is a result of the forces of gravitational attraction. The balance between gravitational and centripetal forces for each celestial body is such as to result in a tidal response not only on the side of the earth nearest the celestial body but also on the farthest side. The response consists of both sea level variations of several metres (tides) and associated ocean currents (tidal streams). The apparent orbits of the sun and the moon around the earth are, however, neither circular, nor co-planar, nor aligned with the earth's equatorial plane. These irregular orbits cause a change in the tidal forcing and the ocean response with time. To allow for these variations it is usual to consider the tidal forcing as arising from many fictitious celestial bodies, each of which has a perfectly circular orbit, and each of which produces a perfectly regular response. The combined effect of all constituents gives the whole tidal response. Amplitudes and phases of the contributions to the tidal height field are given in the Australian national tide tables, while a practical guide to overall effects of tidal currents often appears on relevant navigational charts.

Because of the presence of the continents, the tides cannot follow the celestial bodies around the world as the earth rotates. Instead, tides propagate in the deep ocean around amphidromic points (places where the ocean surface does not vary at all). Tides tend to propagate clockwise around amphidromic points in the southern hemisphere, and anticlockwise in the northern hemisphere.

For narrow continental shelves such as that on the east coast of New South Wales the tides have ranges much the same as those occurring in the deep ocean (approximately 2 m). For very wide continental shelves such as the North West Shelf off Western Australia or the region offshore from Mackay in Queensland, the tidal oscillation is nearly resonant and the spring tidal range is approximately 10 m (Holloway 1983, 1984, Middleton, Buchwald & Huthnance 1984). Although sea level variations due to tides can dominate on wide shelves, the sea level fluctuations associated with wind-driven currents may contribute substantially to the sea level variability on narrow shelves, or during neap tides. In the complex coral reef systems offshore from Mackay, tidal currents in channels can reach 4 m per sec (8 knots).

To determine the tides for a particular region, it is usual to make measurements of tides at hourly intervals for periods of several months and at several locations. Amplitudes and phases of constituents, for both tidal currents and heights, are computed from the data and, since the tidal response for each constituent is quite regular, predictions for future times can then be made with a high degree of reliability. The Australian national tide tables, produced annually, contains predictions for a representative selection of coastal locations.

Some peculiar features of tides on the Australian coast are their large ranges (over 10 m) on the North West Shelf and in Broad Sound, the 'Dodge' tides in Spencers Gulf where the tides are absent for several days, and their ability in the Great Barrier Reef to draw nutrients into the shallower waters (Thompson & Guiding 1981).

Computer models are sometimes used to produce a synoptic view of tides in a region such as the Gulf of Carpentaria or Bass Strait. A good example of the presentation of tidal data, and of comparison with a numerical (computer produced) model of the tidal flow in a continental shelf region is given by Griffin, Middleton & Bode (1987, Figure 4), reproduced here as Figure 4.

Figure 4

Figure 4: Comarison of a numerical model's predicted M2 tidal height (a) and Greenwich phase lag (b) fields with the author's (Griffin, Middleton & Bode, 1987) own and previous observations.

Flow in embayments and semi-enclosed shelf seas

On outer continental shelf regions current flows tend to be directed primarily alongshore and are usually consistent, both in speed and direction, for distances of several hundred kilometres alongshore. However, coastlines are not straight: they are convoluted to varying degrees with promontories and embayments of various scales. These topographic features modify the regular outer-shelf flow, occasionally resulting in closed or semi-closed nearshore recirculation patterns. These 'eddies' may be several kilometres in diameter, and may have very strong currents flowing in various directions, depending on the exact location at which current measurements are made (Deleersnijder, Norro & Wolanski 1992, Middleton, Griffin & Moore 1993). They can serve to confuse interpretations of prevailing current directions when their existence is not appreciated. For example, measurements at a location a few kilometres directly offshore from a headland might show persistent offshore flow due to the redirection of a general alongshore flow by the headland, while a more general shoreward flow might occur some distance upstream or downstream of the headland. Such circulation features might be important to the local biology or to sediment transport. Robinson (1982) provides a useful review of the types of flow which might be expected. An example of a study of wind-driven flow in a semi-enclosed embayment is that undertaken by Steedman and Craig (1983) in Cockburn Sound, Western Australia.

With topographic scales of several hundred kilometres the flows are strongly controlled by the coastline and somewhat insulated from the open ocean. For example, wind-driven flows in the Gulf of Carpentaria (Forbes & Church 1983) and Bass Strait (Fandry 1983) are both extremely complex, and predictive capability for these regions has only been achieved through the use of numerical models.

The degree of isolation from open ocean waters afforded in some embayments and gulfs may allow water masses of distinctive properties to form. Evaporation during the dry South Australian summer renders the upper reaches of Spencer Gulf and Gulf St. Vincent extremely saline with salinities exceeding 48 psu (Practical Salinity Units) in Spencer Gulf and 42 psu in Gulf St. Vincent (Nunes & Lennon 1986). Although the temperatures are warmer than oceanic temperatures, the effect of the substantially increased salinity is to produce a water mass of density somewhat greater than that of the continental shelf waters outside the gulf. The dense, high salinity water flows seaward along the bottom of the gulfs as a gravity current, and cascades off the continental shelf into deeper oceanic waters (Nunes & Lennon 1987). During winter, surface cooling of Bass Strait waters causes them to be well mixed (Baines & Fandry 1983), but as a result of the limited depth of Bass Strait (approximately 100 m), waters cool substantially more than those in the adjacent deep ocean. The cool, dense Bass Strait waters then cascade down the continental slope into the Tasman Sea (Godfrey et al.1980, Villanoy & Tomczak 1991).

Conclusions

In writing this summary, many simplifications have been made in order to present the essence of the physical processes of importance to Australian seas. The simplifications have been made not only in the properties of each individual process, but also in the range of processes considered. In seeking additional knowledge of physical processes the reader is offered a selection of key references on each topic. These have been chosen because they are recent and relevant.

References

Andrews, J.C. 1983. Thermal waves on the Queensland shelf. Aust. J. Mar. Freshwat. Res., 34: 81-96.

Andrews, J.C. & Gentien, P. 1982. Upwelling as a source of nutrients for the Great Barrier Reef ecosystems: a solution to Darwin's question? Mar. Ecol. Prog. Ser., 8: 257-269.

Australian national tide tables 1993. Canberra: Australian Government Publishing Service. 260 pp.

Manual of meteorology, Part 1. General meteorology (1975). Canberra: Australian Government Publishing Service. 144 pp.

Baines, P.G. & Fandry, C.B. 1983. Annual cycle of the density field in Bass Strait. Aust. J. Mar. Freshwat. Res., 34: 143-153.

Batteen, M.L. & Rutherford, M.J. 1990. Modelling studies of eddies in the Leeuwin Current; the role of thermal forcing. J. Phys. Oceanogr., 20: 1484.

Bennett, A.F. 1983. The South Pacific including the East Australian Current. pp 219-242, in A.R. Robinson (ed.), Eddies in marine science. Berlin: Springer-Verlag.

Burrage, D.M. 1993. Coral Sea currents. Corella, 17: 135-145.

Burrage, D.M., Church, J.A. & Steinberg, J.A. 1991. Linear systems analysis of momentum on the continental shelf and slope of the central Great Barrier Reef. J. Geoph. Res., 96: 22169-22190.

Bye, J.A.T.</>A 1983. The general circulation in a dissipative ocean with longshore wind stresses. J. Phys. Oceanogr., 13: 1553-1563.

Cahill, M.L. & Middleton, J.H. 1993. Wind-forced motion on the Northern Great Barrier Reef. J. Phys. Oceanogr., 23: 1176-1191.

Church, J.A. 1987. The East Australia Current adjacent to the Great Barrier Reef. Aust. J. Mar. Freshwat. Res., 38: 671-683.

Church, J.A., Godfrey, J.S., Jackett, D.R. & McDougall, T.J. 1991. A model of sea-level rise caused by ocean thermal expansion. J. Climate, 4: 438-456.

Cresswell, G.R. 1991. The Leeuwin Current - observations and recent models. J. R Soc. W.A., 74: 1-14.

Cresswell, G.R., Ellyett, C., Legeckis, R. & Pearce, A.F. 198. Nearshore features of the East Australian Current system. Aust. J. Mar. Freshwat. Res., 34: 105-114.

Cresswell, G.R. & Golding, T.J. 1980. Observations of a south-flowing current in the south-eastern Indian Ocean. Deep-Sea Res., 27A: 449-466.

Cresswell, G.R. & Legeckis, R. 1986. Eddies off southeastern Australia. Deep-Sea Res., 33: 1527-1562.

Deleersnijder, E., Norro, A. & Wolanski, E. 1992. A three-dimensional model of the water circulation around an island in shallow water. Cont. Shelf Res., 12: 891-906.

England, M.H. 1992. On the formation of Antarctic Intermediate and Bottom Water in ocean general circulation models. J. Phys. Oceanogr., 22: 918-926.

Fandry, C.B. 1983. Model for the three dimensional structure of wind-driven and tidal circulation in Bass Strait. Aust. J. Mar. Freshwat. Res., 34: 121-141.

Fandry, C.B. & Steedman 1989. An investigation of tropical cyclone generated circulation on the Northwest shelf of Australia in a three-dimensional model, Deutsche Hydrograph. Zeitsch., 42: 307-341.

Forbes, A.M.G. & Church, J.A. 1983. Circulation in the Gulf of Carpentaria II. Residual currents and mean sea level. Aust. J. Mar. Freshwat. Res., 34: 11-22. Freeland, H.J., Boland, F.M., Church, J.A., Clarke, A.J., Forbes, A.J., Huyer, A., Smith, R.L., Thompson, R.O.R.Y. & White, N.J. 1986. The Australian Coastal Experiment: a search for co