Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 Antarctic Bottom Water Formation I. PRINCIPAL RECOMMENDATIONS 1. We recommend the initiation of planning for a program of direct observations of various processes for simultaneous test of several hypotheses on the formation of Antarctic Bottom Water in the Weddell Sea. The ele- ments of the initial program should include: (a) A study of the distribution and variation of the ice cover in the Weddell Sea; (b) A study of the hydrography of the Weddell Sea-we note that because of the technological problems imposed by the heavy ice cover, considerable planning and preliminary field work will be required for compre- hensive observations; (c) A study of the currents in the Weddell Sea in conjunction with the hydrographic work; (d) Studies of the bottom topography of the Weddell Sea, to be carried out in conjunction with the hydrographic workâall ships travers- ing the Weddell Sea area should be outfitted with accurate positioning and depth-sounding equipment; (e) An effort to obtain measurements from a nuclear sub- marine under the ice and the strong support of development of automatic data systems and mobile surface-effect-type laboratories for use in the Antarctic. 2. We recommend the continuation and completion, in order to iden- tify other bottom-water-formation areas, of the unfinished circumpolar hydrographic survey carried out by the USNS Eltanin through 1972. Em- phasis should be placed on coverage of areas not already studied (e.g., con- tinental margins), and atlases of the data should be compiled. Summer ice- breaker studies in any continental margin areas, e.g., Weddell Sea, Ross Sea, or Adelie coast, and efforts to gain any winter data anywhere on hydrog- raphy, chemistry, meteorology, and ice cover should be supported. 25
26 Southern Ocean Dynamics 3. We recommend the development of realistic theoretical and laboratory models of Antarctic Bottom Water formation. II. INTRODUCTION Water-mass analysis shows that the bottom water over a large portion of the world ocean has its origin in the waters surrounding Antarctica. The mechanism of formation of this Antarctic Bottom Water (AABW) has been discussed for many years and is one of the chief unsolved problems of antarctic oceanography. The principal questions are the following: What physical mechanisms produce Antarctic Bottom Water? Where are the poten- tial formation areas? How much is formed? When does the formation occur? What is the interaction of the bottom water with other phenomena in the Southern Ocean? III. BRIEF HISTORY The formation of cold, high-salinity water by the freezing of ice at the surface in the winter was first suggested by Brennecke [1921] and Mosby [1934]. They noted that the resultant cold surface water, whose salinity has been increased, could mix with the deeper, warmer, and saltier water at the edge of the Continental Shelf and sink to the-bottom. Later analysis of water masses in the Southern Ocean [Deacon, 1937; Wiist, 1938] suggested that the major portion of the AABW forms in the Weddell Sea. The absolute rate at which the bottom water is formed has not been determined directly, but estimates of the mean annual production range from 10 x 106 to 50 x 106 m3/s, with the lower values [Stommel and Arons, 1960] being more gen- erally accepted but not directly tested. More recent studies have supported the lower estimates [Gill, 1973; Mosby, in press; Warren, 1971; Warren and Voorhis, 1970; Wright, 1970]. The most recent evidence also shows that the Weddell Sea is not the sole source of AABW and that not all bottom water has the same characteristics as the Weddell Sea variety. For example, Jacobs et al. [1970] and Gordon [197la, 1971b] have discovered saltier bottom water that probably comes from the Ross Sea, and Gordon and Tchernia [1972] show a third source of AABW near the Adelie coast. The "Adelie" water has exactly the same characteristics as the Weddell Sea bottom water. Moreover, Gordon [in press] showed that shelf water at the freezing point with salinity sufficiently high to form AABW also occurs in the regions of the Shackleton and Amery ice shelves. However, the Weddell Sea appears to be the major source of AABW, and much of this water enters the rest of the World Ocean, where it has strong influence on the temperature and density of abyssal waters and on sedimentary processes.
Antarctic Bottom Water Formation 27 A number of physical processes that may play a role in the formation of bottom water have been proposed. Fofonoff [1956] suggested that the characteristics of the deep water may be determined primarily by the equation of state of seawater, just as the temperature distribution of deep water in lakes is determined primarily by the physical properties of water rather than the energy exchanges across the surface of the lake. He made use of the fact that when two water types of different temperatures and salinities are mixed together, the increase in density brought about by the small decrease in volume yields a mixture that is denser than either type. The mixture therefore sinks. He suggested that AABW is such a mixture of warm deep water and water from the Continental Shelf. He noted that since un- mixed shelf water has not been observed at great depths, bottom water is formed principally in shallow water. Formation of bottom water may also occur when the coastal-current water is cooled as it flows along and beneath the vast ice shelves in the southern Weddell Sea [Seabrooke et al., 1971]. This cooled coastal-current water would then mix with warm deep water to form AABW. However, this process appears to require the loss of more heat than can be conducted through the ice [Foster, 1972; Gill, 1973]. Gordon [197la] has suggested that the cold shelf ice may play a role in allowing the shelf water to attain sufficient density for bottom-water formation. In this way, the shelf ice influences the shelf water produced by sea ice. A double-diffusive mechanism (salt-fingering) has been proposed as still another physical process for formation of AABW [Gill and Turner, 1969]. This process can occur in a fluid whose density variations result from the distribution of two components with different molecular diffusivities. The motion is driven by drawing on the potential energy in the field of the component with a destabilizing gradient. The mechanism can be active when the surface water is considerably less dense than the deeper water, can be driven by melting of surface ice, and can operate in the summer. Gill and Turner suggest a current down the slope, carrying colder, fresher water to the bottom, under the warmer saltier water. The existence of such a bottom current in the summer is a testable prediction of this thermohaline mecha- nism. Although all of these mechanisms are plausible, none has been criti- cally tested, and the most generally accepted remains the freezing of the sea surface, originally suggested by Brennecke and Mosby. Moreover, we note that although primary production of the cold saline shelf water may occur in winter, it is possible that escape of this water to the deep ocean may not be confined to the winter. Gill [1973] has shown how there can be a net brine release over the Continental Shelf area because the pack ice is continually being blown offshore, and he suggests that bottom water is produced throughout the year, since there is always a supply of dense water from the Continental Shelf. He notes that the western part of the Weddell Sea is particularly important in the production process.
28 Southern Ocean Dynamics IV. RECOMMENDED PROGRAM A. GENERAL SURVEY DATA REQUIRED Using data collected by various national programs up to 1960 and by the USNS Eltanin from 1962 to 1972, oceanographers have begun to formu- late the questions previously mentioned. However, the survey data are insuf- ficient for resolution of any of these questions, and the nature of a broad survey precludes detailed studies of local areas. Moreover, we still have no winter data on water properties in potential sinking regions because ships cannot penetrate the ice. To extend the present survey data as efficiently as possible, we make the following three general recommendations for surveys of hydrography, bathymetry, chemistry, meteorology, and ice cover: 1. To identify other bottom-water-formation areas, the hydrographic survey carried out by the Eltanin through 1972 should be continued and completed, with emphasis on coverage of areas not already studied (e.g., continental margins), and atlases of the data should be compiled. 1. Summer icebreaker studies in any continental margin areas-e.g., Weddell Sea, Ross Sea, or Adelie coast-should be supported. 3. Efforts to gain any winter data anywhere on hydrography, chemis- try, meteorology, and ice cover should be supported. B. DATA REQUIRED FOR TEST OF HYPOTHESES At the conclusion of the general survey work, it will be necessary to turn to the collection of data for tests of the specific hypotheses of AABW formation. We divide the general problem into two parts here: the formation of cold, high-salinity water in general; its mixing with and escape from the surrounding water masses. A test of the idea that sea-ice formation leads to the formation of denser water will first require knowledge of the physical and chemical prop- erties of the ice, the rates of its formation, and its movement due to wind. The geographic distribution and thickness of the ice must be determined. Second, the hydrographic properties of the water under the ice must be known, and the rate of formation of cold, high-salinity water must be moni- tored by measuring both the vertical and horizontal circulation of local waters. The effect of freezing ice shelves on seawater density requires all these data, as well as the distribution of freezing and melting at the shelves.
Antarctic Bottom Water Formation 29 Cooling of the surface due to katabatic flow of air down the cold slopes of the Antarctic continent can also form high-density surface water. Such effects have been directly observed off the coast of France in the Mediterranean during the mistral, where deep mixing occurs in selected regions down to a depth of 2100 m [Medoc Group, 1970]. An experiment similar to the one in the Mediterranean would be valuable for the study of AABW formation, but it would be more complex, because of the effect of ice at the surface. The distribution of polynyas and leads and the hydrographic properties of the water under the ice, as well as the rates of formation of the dense water and its horizontal and vertical circulation, must be monitored in regions of preferred cold air flow. A number of direct measurements are required for tests of mixing and escape of AABW. To test the theories based on (a) mixing and sinking in proportion to production rate of dense shelf water, (b) mixing and sinking due to the nonlinear equation of state, and (c) mixing and sinking due to the salt-finger mechanism, we require detailed (microstructure) measurements of temperature, salinity, and chemistry and direct velocity (horizontal and verti- cal) measurements. Hypotheses that depend on change of outside constraints will, of course, require direct measurement of those constraints, e.g., possible interaction between the wind and the density field associated with deep geostrophic currents [Gordon and Tchernia, 1972; Killworth, 1973]. Finally, the question of the movement of the AABW away from the source regions must be considered. The question resolves into two parts: the movements and their dynamics; mixing with other water masses. We need to know the geographic distribution of currents and bottom-water charac- teristics (temperature, salinity, oxygen, silicate) and the transports and dynamics of these flows (geostrophic or nongeostrophic). The question of mixing will be attacked through physical and chemical measurements of microstructure and fluxes. C. SPECIFIC PROGRAM FOR THE WEDDELL SEA We recommend a plan for simultaneous test in the Weddell Sea of several hypotheses on the formation of bottom water. Until the International Weddell Sea Oceanographic Expedition (IWSOE) of 1968, the oceanographic data in the Weddell Sea were largely confined to the outer edges, mainly in the northern and eastern parts, with the exception of the few stations obtained during the drift of the German ship Deutschland across the central part of the Weddell Sea in 1912. In the southern summer of 1968, the USCGC Glacier was able to penetrate the central and southwestern parts of the Weddell Sea and to obtain a number of hydrographic stations in these regions. The only winter hydrographic stations
30 Southern Ocean Dynamics are still those obtained by the Deutschland in 1912, and, unfortunately, the winter data are probably the most important to the understanding of AABW formation. To understand the formation of AABW as it occurs here, we need to know the heat and salt budgets, the location and intensity of the sinking of surface waters and the subsequent subsurface currents, and the type of mixing processes involved. Understanding the budgets requires in turn a knowledge of the air-ice-sea interactions at the surface of the ocean, of the temperature and salinity structure in the entire Weddell Sea, and of the incoming and outgoing currents. An understanding of air-ice-sea interactions in the Weddell Sea is extremely important, since in the presence of open leads and polynyas, heat exchange is greatly increased. Consequently, the rate of freezing in these areas may be several orders of magnitude greater than in areas where the ice is a meter or more thick. Thus we must also know how the distribution of ice cover changes in time throughout the Weddell Sea. The hydrography of the Weddell Sea should be determined with a varying spatial resolution. Since the geostrophic constraint causes any subsur- face flow of AABW to move largely around the Weddell Sea parallel to the contours, hydrographic sections would be most strategically located normally to the general bottom contours. An idealized plan for the location of hydro- graphic stations is seen in Figure 4.1. The region in the vicinity of the shelf break, approximately the 600-m contour, is of special interest, as it is here that the warmer deeper water may mix with the shelf water. The summer measurements by the Glacier showed that there was a frontal structure here, but the stations were spaced too far apart to resolve the details. It is also important that measurements of the temperature and salinity microstructure of the top 200-300 m be made over the deeper parts of the Weddell Sea, as this is the region in which surface water mixes with the warmer deeper water. All these measurements are needed in both summer and winter and ideally would be carried out at about 3-month intervals. Existing winter data furnish only a rough indication of what may take place during the freezing season. Relative values of the transport of water in the Weddell Sea can be calculated from hydrographic data using the geostrophic equations, but absolute values can be obtained only from current measurements over a long enough period. A successful strategy might be to combine hydrographic stations with measurements of the bottom currents, especially those into the Weddell Sea in the east between about 25° W and 30° W, and the subsurface currents along the Continental Shelf. Bottom-current measurements and hydrographic data at selected stations in the hydrographic plan of Figure 4.1 would provide these data; such measurements ideally should be made over periods long enough to allow short-time effects to be filtered out. The bathymetry of the entire Weddell Sea area, especially the western
Antarctic Bottom Water Formation 31 FIGURE 4.1 Idealized plan for a hydrographic survey in the Weddell Sea. and southwestern shelves, needs to be known in much greater detail. The bottom topography of the Continental Shelf is of particular interest, since the flow of water along the shelf may be partially controlled by the configuration of the bottom contours. Of perhaps even greater importance is the bottom typography of the Continental Slope, since the flow of A AB w or its precursor could take place largely along the slope, and the presence of large canyons on the slope could lead to a concentration of flow in density currents down these canyons. Our recommendations for a field program in the Weddell Sea are the following: Ice Cover. We recommend that a program to study the distribution
32 Southern Ocean Dynamics and variation of the ice cover in the Weddell Sea be initiated as soon as possible. The first step would be the study of all the pertinent satellite photographs. Visual observations are not practical for complete year-round coverage, since the southern part of the Weddell Sea is dark in winter, and the entire region is commonly covered with clouds. We do not know whether the winter ice cover is broken by leads that are too small to be resolved by satellites; therefore, we suggest a program of remote sensing of the ice cover by aircraft. Both side-looking radar and passive microwave radiometers have the advantage of being all-weather instruments and thus may be ideal for antarctic work. We suggest regular monthly flights over the Weddell Sea. At present, it is possible to fly C-130-type aircraft from Ushuaia, Argentina, down the western part of the Weddell Sea across the southeastern part and back over the central part. If refueling were possible at a base on the Ant- arctic Peninsula, the coverage could be increased. The net transport of ice from the Weddell Sea shelf region should also be monitored to allow the calculation of the net amount of salt release during bottom-water formation. This monitoring could be done by satellite coverage. Hydrography. We recommend that a program to study the hydrog- raphy of the Weddell Sea be initiated as soon as possible, noting that because of the technological problems imposed by the heavy ice cover, considerable planning and preliminary field work will be required for comprehensive obser- vations. In addition, the results of the ice cover program may have a consider- able effect on the plans for the hydrographic work. At present, the most effective hydrographic program would be one that made use of a nuclear submarine. STD's are already available for use from submerged nuclear sub- marines, and it would be possible to obtain a rapid survey of the temperature and salinity structure of the entire Weddell Sea at any time of the year. A somewhat less desirable method, which would probably provide only part of the needed data, is the use of present-day icebreakers. Even in the summer, it is not possible for icebreakers to penetrate the central and western parts of the Weddell Sea, except during unusual ice conditions like those encountered during the summer of 1968 by the Glacier. If it appears that a large polynya in the southeastern Weddell Sea remains open during the winter, it might be desirable to have an icebreaker in this area to study the air-ice-sea interac- tions under winter conditions. Also technically feasible at present would be a manned station drifting across the Weddell Sea during the winter, as did the Deutschland and the Endurance. The least demanding drifting station would be an icebreaker with a skeleton crew. Such an icebreaker would have some mobility to search for the regions of bottom formation, which might be quite small. Unmanned automatic data buoys that are tracked and interrogated by satellites could provide winter hydrographic data. The successor to IRLS, the
Antarctic Bottom Water Formation 33 TWERLE system, will probably be operational in 1974, but the development of suitable salinity sensors has yet to be accomplished. Helicopters operating from icebreakers or shore stations also might provide platforms for obtaining hydrographic data in ice-covered regions, but operations under winter condi- tions remain untried. Unmanned submersibles are being tested in the Arctic and could provide a means to extend hydrographic data under ice cover; however, at present their range is still short, less than 50 km. Finally, the development of a mobile surface-effect-type laboratory that could operate on both ice and water could produce the ideal platform for year-round work in the Weddell Sea. Unfortunately, surface-effect-type vehicles for operation on rough ice surfaces are still in the early stages of development, and it may be many years before they can be used safely in Antarctica. We recommend that every effort be made to obtain measurements from a nuclear submarine under the ice and that the development of automatic data systems and mobile surface-effect-type laboratories for use in the Antarctic be strongly sup- ported. Currents. We recommend a program to study the currents in the Weddell Sea in conjunction with the hydrographic work. At present, it would be possible to make current measurements from an icebreaker during the summer in the eastern Weddell Sea in the relatively ice-free region at about 20° W using bottom-current meters and hydrographic stations. It might also be possible to retrieve current meters left over the winter in this region. In the remainder of the Weddell Sea, the year-round close ice cover will probably prevent retrieval of current meters using conventional techniques. The devel- opment of acoustically interrogated bottom-current meters might provide a means of obtaining long records of currents without the necessity of retrieval. This method would be most feasible in the northwestern parts of the Weddell Sea, where helicopters from land stations on the Antarctic Peninsula could emplace and interrogate the instruments through holes in the ice. Some of the testing of current-metering devices could be carried out in the more accessible Ross Sea region. Bathymetry. We recommend that studies of the bottom topography of the Weddell Sea also be carried out in conjunction with the hydrographic work and that all ships traversing the Weddell Sea area be outfitted with accurate positioning and depth-sounding equipment. Again, the most efficient means for determining bathymetry under ice cover would be a nuclear sub- marine. However, since the bathymetry does not change with the seasons, summertime penetrations by icebreakers could cover most of the Weddell Sea. The region just east of the Larsen Ice Shelf, which has so far been impenetrable, might require spot soundings through the ice by helicopters.
34 Southern Ocean Dynamics Related Studies. In addition to these four main programs concerned with studying AABW formation, some other related programs should be con- sidered. As mentioned earlier, the general hydrographic study of the seas around Antarctica carried out by the Eltanin from 1962 to 1972 should be continued to determine the extent of AABW formation in regions other than the Weddell Sea. In addition, the work on the abyssal circulation to deter- mine the transport of AABW northward away from Antarctica should be continued, with direct current measurements emphasized. Finally, if the studies of the oceanography under the Ross Ice Shelf that are planned in conjunction with the Ross Ice Shelf Drilling Project show that the ice shelf can modify the surrounding waters in a quantitatively important manner, then similar studies should be carried out under the Filchner Ice Shelf. It is clear from property distributions that some bottom water is formed in the Ross Sea, although at a much lower rate than in the Weddell Sea. Summer and winter hydrographic stations on the Continental Slope there would give a better idea of the source of this water and might suggest how to deploy current meters to measure the formation rate directly. The work in the Weddell Sea is given first priority here. D. DEVELOPMENT OF NEW TECHNOLOGY The Working Group notes that the existing point measurements of water velocity and temperature are probably not adequate for long-term monitoring of the variability and amounts of bottom-water formation. Some kinds of averaging techniques, e.g., electromagnetic or acoustic, will probably be required and are being developed as part of other oceanographic experi- ments such as NORPAX and MODE. We do not feel that special development will be required for the Southern Ocean processes but wish to encourage the use of the new averaging techniques there as they become reliable enough for that environment. E. THEORETICAL AND LABORATORY WORK We recommend that the development of realistic theoretical and laboratory models of AABW formation be encouraged. F. SUMMARY OF RECOMMENDED PROGRAM Tables 4.1 and 4.2 summarize the experiments and goals and show how the recommended program would fit into a ten-year time scale.
Antarctic Bottom Water Formation 35 TABLE 4.1 RECOMMENDED PROGRAM STRUCTURE Antarctic Bottom Water Formation Experiments and Goals Proposed Experiments and Theory Specific Goals Complete Weddell Sea summer hydrography. Collect winter data under ice pack. Mbnito_rlng_ _E_xper imen t s * Long-term variability of bottom velocity and temperature near continental margins. Collect data for model formu- lation and test of hypotheses. Determine "climate" of Antarctic Bottom Water through knowledge of * Properties of water masses * Rate of renewal of deep circulation Determine relation between water-mass formation and ocean-atmosphere exchanges for parameterization into global circulation models. * Formulate critical tests of Antarctic bottom water form- ation hypotheses. TABLE 4.2 RECOMMENDED PROGRAM STRUCTURE Antarctic Bottom Water Formation Time-Phased Diagram ACTIVITY TIME FRAME 1973 - 1976 1976 - 1979 1979 - 1WZ CONTINUAL PLANNING AND REVIEW 1 CaneraJ. Survey Oata Collection n Complete Clrcumpolar Survey Sun-irr Surveys Collect Winter Data Begin > | Specific Weddell Sea Program I Ice Cover and Transport ' *» ^Collaboration [ for FCCE [Bathymetry *,,J1\ i Iff Kerga into . . Hew Technology I Development of Averaging Techniques for Long-term I Monitoring Carried out under ! Vegm UM In other Programs ! Southern Ocean Reaaarch ,T,^°,r)r 1 Effort I Formation Hypotheses Activity Planning Continuing
36 Southern Ocean Dynamics REFERENCES Brennecke, W. 1921. Me ozeanographischen Arbeiten der Deutsch Antarktischen Ex- pedition 1911-1932. Arch. Deut. Seew. 39(1), 214. Deacon, G. E. R. 1937. The hydrology of the Southern Ocean. Discovery Rep. 15. Cambridge University Press, 124 pp. Fofonoff, N. P. 1956. Some properties of sea water influencing the formation of Ant- arctic Bottom Water. Deep-Sea Res. 4, 32-35. Foster, T. D. 1972. An analysis of the cabbeling instability in sea water. J. Phys. Oceanog, 2, 294-301. Gill, A. E. 1973. Circulation and bottom water production in the Weddell Sea. Deep-Sea Res. 20, 111-140. Gill, A. E., and J. S. Turner. 1969. Some new ideas about the formation of Antarctic Bottom Water. Nature 224, 1287-1288. Gordon, A. L. 1971a. Comment on the Weddell Sea produced Antarctic Bottom Water. /. Geophys. Res. 76, 5913-5914. Gordon, A. L. 1971b. Spreading of Antarctic Bottom Waters, II, pp. 1-17 in Studies in Physical Oceanography-a Tribute to George Wust on his 80th Birthday, A. L. Gordon, ed. Gordon & Breach, New York. Gordon, A. L. In press. A general ocean circulation. In Proceedings of the Symposium on Numerical Models of Ocean Circulation, October 1972, Durham, N.H. National Academy of Sciences, Washington, D.C. Gordon, A. L., and P. Tchernia. 1972. Waters of the continental margin off Adelie Land, Antarctica. Antarct. Res. Ser. 19, 59-69. American Geophysical Union, Wash- ington, D.C. Jacobs, S. S., A. F. Amos, and P. M. Bruchausen. 1970. Ross Sea oceanography and Antarctic Bottom Water formation. Deep-Sea Res. 17, 935-962. Killworth, P. D. 1973. A two-dimensional model for the formation of Antarctic Bottom Water. Deep-Sea Res. 20, 941-972. Medoc Group. 1970. Observations of formation of deep water in the Mediterranean Sea, 1969. Nature 227, 1037-1040. Mosby, H. 1934. The waters of the Atlantic Ocean. Sci. Res. Norwegian Ant. Exped. 1927-1928 Rep. 11, Norske Videnskaps-Akad. Mosby, H. South Atlantic Bottom Water. In Symposium on the Ocean World (Tokyo, 1970). Scientific Committee on Antarctic Research. (In press.) Seabrooke, J. M., G. L. Hufford, and R. B. Elder. 1971. Formation of Antarctic Bottom Water in the Weddell Sea. J. Geophys. Res. 76, 2164-2178. Stommel, H., and A. B. Arons. 1960. On the abyssal circulation of the world ocean-II. An idealised model of the circulation pattern and amplitude in oceanic basins. Deep-Sea Res. 6, 217-283. Warren, B. A. 1971. Evidence for a deep western boundary current in the South Indian Ocean. Nature 229, 18-19. Warren, B. A., and A. D. Voorhis. 1970. Velocity measurements in the deep western boundary current of the South Pacific. Nature 228, 849-850. Wright, W. R. 1970. Northward transport of Antarctic Bottom Water in the Western Atlantic Ocean. Deep-Sea Res. 17, 367-371. Wu'st, G. 1938. Bodentemperatur und Bodenstremin der Atlantischen, Indischen und Pazifischen Tiefsee. Beitr. Geophys. 54, 1-8.
