Chapter 7 - WORKING GROUP REPORTS

7.1 Zooplankton/Krill Working Group Report

Chairman: Mark Huntley

Rapporteur: Steve Nicol

Members: Charles Greene
Suzanne Razouls
Tsutomu Ikeda
Sigrid Schiel
Victor Marin
Jon Watkins
Langdon Quetin
Stanislaw Rakusa-Suszczewski
Robin Ross

7.1.1 Site selection

The group recommends the consideration of two sites for field studies: (1) the Bellingshausen Sea, and (2) an area directly west of the Ross Sea, bounded by the continent to the south, 65 S, and 140 to 160 E. Most of the discussion which follows refers to the Bellingshausen Sea site in particular, but applies in general to both sites.

7.1.2 Climate change context

Both sites provide an opportunity to assess the role of sea ice in the life cycle and habitat of Antarctic krill, Euphausia superba. Climate change might be expected to alter: (1) the areal extent of seasonal ice cover, (2) the thickness of ice, (3) the rate of formation and retreat, (4) and the percentage of open water within the pack. Furthermore, both sites afford the opportunity to observe the effects of glacial meltwater on physical stability, circulation patterns, and their effects on ecosystem structure and dynamics. Both areas have been subject to significant fisheries for krill, so there exists the possibility for examining fisheries/climate change interactions.

7.1.3 Target species

Krill Euphausia superba is designated as a target species due to its clear importance from an ecological and economic viewpoint. Furthermore, there exist abundant historical data on the species. The following species were also indicated as important species:
** Salpa thompsoni
* Euphausia crystallorophias
* Calanoides acutus
* Calanus propinquus
Themisto gaudichaudi
Metridia gerlachei
Rhincalanus gigas
Thysanoessa macrura
Sagitta gazellae
Salpa thompsoni is designated as especially worthy of consideration due to its propensity to dominate in years and locations where E. superba are scarce. Euphausia crystallorophias is a high Antarctic, neritic euphausiid, particularly associated with the pack ice. Calanoides acutus and Calanus propinquus represent important copepod species with contrasting life cycles: C. acutus exhibits seasonal, ontogenetic vertical migration, whereas C. propinquus does not.

7.1.4 Definable populations

The gyral circulation of the Bellingshausen Sea is thought to contain a functional population of E. superba and, by inference, of other significant holozooplankton species. Similarly, the 150—E study site is known to have supported a consistent krill fishery in a restricted area along the continental shelf break. However, the details of the regional circulation in these areas are not well known, and are expected to benefit from a combination of WOCE and GLOBEC field studies.

7.1.5 Population dynamics

Understanding the population dynamics of the key species in these two areas will require two separate types of cruises: (1) quasi-synoptic demographic surveys and (2) processoriented cruises. Particular attention will be paid to the analysis of demographic parameters during the winter season and their influence on the size of populations during the productive summer season. Recommendations regarding the measurement of population dynamics parameters are made below (Recommended Research Strategy).

7.1.6 Focus on processes and mechanisms

The working group identified key gaps in our knowledge regarding the life cycle of Antarctic krill and other holozooplankton. Particular amongst these problems is the question of overwintering strategies, encompassing under-ice behavior, potential benthic interactions, and physiological adaptations to the winter environment. The group characterized certain critical processes ith respect to their temporal/spatial scales (Table 1 ).
TABLE 1

Reproduction       days - months       1 - 10 km (H)
                                       1000 m (100 m resolution)

Migration (H)      days - weeks        10 - 100 km

Migration (V)      hours - seasonal    300 m
Swarming           hours - weeks       10 m - 10 km

Mortality     
 a) "natural"      weeks               microscale - gyre scale
 b) predation      seconds - months

Growth             days - weeks

Life cycle         2 - 3 years         gyre scale
                   (hatch - Hatch)

*Note: Sample to 1000 m for copepods

7.1.7 Historical database

The group acknowledged that, while meager, data do exist for both study regions, particularly in regard to krill in the Bellingshausen Sea study area. Relevant data have been reported from the work of the BIOMASS program, V. Siegel, the Discovery expeditions, the Eltanin expeditions, and a variety of Polish and Russian expeditions.

7.1.8 Modeling

The group recommends the development of coupled biological-physical models for holozooplankton populations in the study areas. These models should include the parameters of sea ice formation and retreat and its effect on stability of the water column.

7.1.9 New technology

New developments in remote sensing technology were encouraged. These might include moored sensing systems which would provide long-term Eulerian data, complementing shorter-term Lagrangian information available from research cruises. Satellite and aircraftbased observing technologies were also endorsed.

7.1.10 Relation to other programs

It was noted that other programs and scientific activities in the Bellingshausen Sea in the near future would include BOFS (JGOFS), WOCE, the US LTER (a ten-year ecosystem study in Palmer Basin), RACER, AMLR, and a Polarstern cruise to the Bellingshausen-Amundsen Seas in 1993-94. In addition, shore-based laboratories in the Antarctic Peninsula region, as well as Rothera (Adelaide Is.) would provide facilities for laboratory-based process studies and long-term continuous observations.

7.1.11 Recommended research strategy

It is recommended that the field research program include three principal activities: (1) Quasi-synoptic survey cruises, (2) Process-oriented cruises, and (3) Remote sensing.

The quasi-synoptic survey cruises should take place at approximately monthly intervals for a minimum period of three years, in order to resolve the full life cycle of E. superba and other key species. These cruises should encompass shelf, slope and oceanic environments (Fig. 1), with the aim of resolving mesoscale features in both the biotic and dynamic physical environments.

Process cruises should focus on the processes and mechanisms identified above, and should be directed on a phenomenological basis from information arising from results of the survey cruises.

Remote sensing aspects of the program should incorporate new and existing technologies capable of yielding synoptic data over long time scales.

In general, the working group recommends that this project could feasibly begin by 1996-97. It would involve multiple ships from different nations, and would require that a common quasi-synoptic survey grid be occupied at different times by all participants.

It was noted that sampling strategies may differ depending upon the species being investigated. For example, while the 200 m depth horizon may be adequate for a study of juvenile and adult krill, the 1000 m depth horizon may be more appropriate for copepods and early larval stages of krill.

7.2 Benthic Working Group Report

Chairman: Donal Manahan

Rapporteur: Jackie Grebmeier

Members: Wulf Arntz
Rennie Holt
Jim Barry
Ken Smith
Ulrich Bathmann
Bill Stockton
Paul Dayton
Martin White
In the context of GLOBEC, the benthic working group discussed process-oriented objectives that would be investigated at various shallow and deep water sites surrounding Antarctica to be occupied by various international scientific programs. The objective at these sites would be to investigate effects of climate change on the benthos by studies of community composition, population dynamics and energy flow.

7.2.1 Site selection criteria

Selection of research sites was based on historical long-term data records, logistical constraints (ship, field station availability), and the location of high and low Antarctic sites and species for latitudinal comparisons. The group selected the following 5 sites surrounding Antarctica:
  1. Ross Sea/McMurdo Sound: high Antarctic (USA)
  2. South Orkney/South Shetland Islands: low Antarctic (eg. UK, POL, D)
  3. SE Weddell Sea: high Antarctic (D)
  4. Davis Sea: high Antarctic (AUS)
  5. Antarctic Peninsula: low Antarctic (various countries)

7.2.2 Climate change context

Benthic structures in the Antarctic are notable for their persistence, which is a valuable characteristic to investigate the effects of climate change. The benthos holds a long-term record of ecological processes both within the fauna and sediments. Within the global climate change context, variation is likely to be measurable as latitudinal changes in community structure and in benthie dynamics, both surrounding islands as well as the main land mass. In particular, Antarctic benthic fauna appear to have higher temperature sensitivity than temperate species, often with Q10's > 10, indicating they may be critical indicators of small temperature changes due to a climatic shift. A second major variable influenced by climatic change is likely to be a shift in organic carbon supply and food chain disruption, resulting in changes in the benthie structure. A. Criteria for Selection Five characteristics were identified as criteria for selection of target species, including:
  1. Measurable growth parameters
  2. Abundant
  3. Wide or restricted distribution (either group will react to climate change differently)
  4. Known life history
  5. Amenable for reproductive studies
B. Species (pelagic and nonpelagic larvae) Three groups of benthie fauna were selected as potential target species. Examples are presented for each, inclusive of both wide ranging species with pelagic (p) larvae and restricted fauna, often characterized by brooders (b).
  1. Bivalves, e.g. Adamussium (p), Laternula (p), Mysella (b), Gamardia (b)
  2. Echinoderms, e.g. Odontaster (p), Sterechinus (p), Ophionotus (p), Diplasteria (b)
  3. Crustaceans, e.g. Notocrangon (p), Chorismus (p), Glyptonotus (p)

7.2.4 Definable populations

The group agreed that selected benthic populations were very tractable using genetic techniques in addition to current data bases on populations. Populations with well-known interactions with other species within the community and known fluctuations, would be desirable.

7.2.5 Population dynamics and physical processes

Coincident measurements of benthic population dynamics (e.g. recruitment, life history strategies, production) with physical processes (e.g. ice cover, temperature, salinity, currents) are essential for determination of key variables influenced by climatic change. Currently standard population dynamic studies of various target species are underway in McMurdo Sound (USA) and areas of the Weddell and Scotia Seas (Germany, U.K.).

7.2.6 A. Processes and mechanisms inducing change

Six major processes were identified that could presently be investigated as mechanisms indicative of global change. These include:
  1. carbon flux
  2. ice conditions
  3. current flow
  4. temperature and salinity
  5. light regimes
  6. redox profiles in sediments.

B. Effects resulting from climate change

Based on input information from physical processes and the mechanisms that could induce changes in the benthos, the group discussed tractable measures in the benthos that could result from climatic perturbations. Four major areas of studies were determined, which include:
  1. energy flow
  2. physiology
  3. population dynamics
  4. community studies.
The energy flow studies are the major area for coordination with JGOFS, which proposes to undertake studies of carbon flux, carbon mineralization in the sediments, and bioturbation. The remaining three areas of study are specific to benthic faunal structure specifically in line with GLOBEC directives. Physiological studies of target species would provide essential information on rates and processes, e.g. the effects of temperature and carbon supply fluctuations on both larval and adult fauna. Population dynamics would include studies of recruitment, production and reproduction. Community studies would include measurements of species composition, abundance and biomass.

7.2.7 Historical database

The group identified the importance of background studies utilizing the large database available from the support and use of Antarctic field stations as well as past offshore benthic studies.

7.2.8 Modeling input

A modeling effort of the population dynamics for target species would be undertaken as the data sets become available. Input from the carbon/energy flow studies would be valuable.

7.2.9 New technology

(to be discussed in another workshop group)

7.2.10 Relation to other programs

GLOBEC will benefit from close cooperation between JGOFS (energy flow), CCAMLR (resource interactions), WOCE (circulation), and FRAM (modeling).

7.2.11 Research strategies

The benthic working group agreed that the selection of permanent stations along designated transects that crossed from the shallow shelf to the deep offshore areas were essential to adequately investigate possible climatic changes to benthic structure. Sampling at these stations would include recruitment and reproductive output measurements, along with visual observations of the sediments by both still camera/video and by SCUBA. Measurements would include physical, biological and sediment coring. Interaction experiments (short-term) would be undertaken both at permanent stations and land-based field stations. Long-term measurements at permanent stations would also occur on the designated transects on a ship-ofopportunity basis during maintenance of field stations. Studies of the historical record in the sediments (e.g. radioisotope methods of dating, stable isotope measurements of bivalves, foraminifera) and sea level change information recorded in raised beaches would be valuable.

7.3 Top Predators Working Group Report

Chairman: Inigo Everson

Rapporteur: Valerie Loeb

Members: William Fraser
Adolf Kellermann
Tony Koslow
Richard Veit

7.3.1 Approach

This section, originally intended to concern only fishes, has been expanded to include penguins, other seabirds and seals. This modification was made due to the important coupling between these predators and nekton in the southern ocean food web (Croxall et al., 1988). These birds and seals, like fishes, are directly or indirectly dependent on krill as a food source (Laws, 1985) and are likely to be sensitive indicators of environmental change.

The reproductive success of Southern Ocean birds and seals has been shown to depend on interannual variations in prey abundance (Croxall et al., 1988); their long term fluctuations in abundance have been related to changing sea ice conditions (Fraser et al., in press). Therefore these higher predators are especially valuable to a program designed to detect the biological effects of global warming.

There are other practical reasons for their inclusion. Birds and seals can be more easily and inexpensively surveyed than other pelagic animals. While foraging, they perform spatially and temporally integrated "sampling regime" over a substantial area. Lengthy time series already exist on the reproduction and abundance of several of the numerically dominant species (e.g. Adélie and chinstrap penguins, fur and crabeater seals) in various locales (Palmer Station, King George Island, Signy Island and South Georgia). Furthermore, the recently established CCAMLR Ecosystem Monitoring Program (CEMP) which involves studies on selected bird and seal species. The results of this program have direct relevance to GLOBEC. Studies focused on fishes and higher predators will provide information on the relationships between predator-prey and their environment which is critical to understanding variability in the Southern Ocean ecosystem. Such studies may also provide the bases for monitoring the effects of man-induced perturbations.

The group agreed that in order to detect changes in fish and higher predators induced by climatic change it is essential to establish long term base line monitoring. These studies are dependent on having good information from field studies and modeling exercises identifying critical processes.

7.3.2 Site selection

We suggest that the Atlantic sector, including the Antarctic Peninsula, South Georgia and the South Orkney Islands, is the most appropriate study region because of its known sensitivity to variation in the Antarctic Circumpolar Current and its existing historical data bases. Within this area, the Weddell and Bellinghausen Seas are considered important and contrasting sea ice zones worthy of study; South Georgia is an open water area with considerable commercial fishing activity (krill and frofish), a historical data base and ongoing monitoring programs; the South Orkneys are within the Weddell-Scotia Confluence and experience commercial fishery activity (especially summertime krill harvesting). The wide latitudinal range was felt to be important for assessing larger scale ecological changes which would be associated with climate change.

7.3.3 Target species

We have selected a relatively long list of target species with the justification that ecological changes among groups of species across the broad study area are more likely to reveal compelling evidence of climatically related change. The target fish species all have broad distributional ranges and represent commercially harvested forms, abundant non-harvested holopelagic forms and accessible shallow water species.
Commercially harvested species:
Champsocephalus gunnari
Notothenia larseni (by-catch)
Electrona carlsbergi

Non-harvested holopelagic species:
Pleuraggramma antarctica
Electrona antarctica

Non-harvested nearshore species:
Notothenia neglecta
Trematomus hansoni
Harpagifer sp.
The commercial species are fished primarily in the northern areas of the Atlantic sector and are included in CCAMLR monitoring operations. The icefish Champsocephalus gunnari is an important fisheries resource and has a long term CCAMLR data base; Notothenia larseni is an abundant by-catch in fisheries operations; Electrona carlsbergi, a myctophid, is the basis of a developing open ocean commercial fishery. The non-harvested holopelagic species Pleuragramma antarctica and Electrona antarctica are abundant and important in food webs in high Atlantic waters and represent contrasting ecological patterns. Notothenia neglecta, Trematomus hansoni and Harpagifer may be conveniently collected at the shore stations.

The target penguin and seabird species are primarily krill predators considered important by CEMP: Adélie, chinstrap, macaroni and gentoo penguins; cape and Antarctic petrels; Black browed albatross. Because of their different feeding activities we also feel it would be useful to include the grey headed albatross (fish and squid prey) and South Polar skua (which feeds on Pleuragramma antarctica ) as target species. The mammalian target species are crabeater and Antarctic fur seals. Both are dependent on krill, but occupy different habitats analogous to those of Adélie and chinstrap penguins.

7.3.4 Definable populations

We are uncertain whether various populations can be distinguished at the present time. The CCAMLR subareas under consideration are felt to be reasonable management units for the commercial fish species. Some bird and seal species show distributional differences which may represent distinct populations. For these species populations could probably be distinguished using mitochondrial DNA or other molecular techniques.

7.3.5 Population dynamics

The commercial fish stocks are monitored and analyzed annually by CCAMLR and cohort analyses have been performed on the South Georgia stocks. Through traditional fishcries techniques, spawning stock biomass for these and the other finfishes can be established through ongoing base line studies of growth and reproduction. Continuing national and CEMP bird and seal programs are monitoring growth rates, breeding success and cohort survival.

7.3.6 Focus on processes and mechanisms

Included are studies providing data important for understanding population dynamics relative to direct and indirect effects of environmental change. Direct effects include:
  1. The effect of temperature on growth and development rates of different ontogenetic stages of fishes:
  2. Overwintering studies of higher predators to gain knowledge of a critical mortality period;
  3. The possible effects of increased UV radiation on near surface fish eggs and larvae. Indirect effects include:
  4. Hydrographic conditions and dispersal of pelagic early life stages of fishes:
  5. The importance of food (e.g. krill) availability on fish condition, reproductive behavior (e.g. gonadal development, timing of spawning);
  6. Foraging. dynamics of higher predators in relation to prey abundance and aggregation behavior.

7.3.7 Historical database

Historical data bases on the commercial fish species have been established by CCAMLR and are being augmented by national programs. CCAMLR has catch statistics and has undertaken cohort analyses of the major commercial species. The BIOMASS program also established a data base on both commercial and nonharvested fish species as well as bird and mammal species. National programs in the U.S., U.K., Germany, France, Australia, New Zealand and South Africa have also provided data on a variety of fish and bird species. Long term data bases have been established for Adélie and chinstrap penguins, crabeater and Antarctic fur seals, various other seal species, and ca. 15 sea bird species nesting on South Georgia and the South Orkneys. Predator populations have been monitored at Palmer station since 1977, King George Island since 1976, the South Orkneys since 1953 and South Georgia since 1962.

7.3.8 Modeling

Specific suggested models based on data resulting from the focused studies and historic bases include:
  1. Physical oceanography (e.g. temperature, water column stability, dynamics) and resulting effects on food supply, growth and development rates, survivorship, and dispersal during early life stages of fish;
  2. A standard population dynamics model for seabirds, integrating physiological data and environmental variables;
  3. Movement and dispersal of foraging predators (based on behavioral data) relative to acoustically detected prey target distributions to determine how seabirds and seals locate food patches.
  4. Fisheries vs. climate related effects on harvested species.
  5. Trophodynamic models of multispecies interactions between fish and higher predators and their prey.

7.3.9 Technology

Technological improvements or developments which would be useful in the suggested studies of fish, sea birds and seals include the following:
  1. Improved acoustics hardware and software for the location, identification and quantification of fish;
  2. Underwater visual systems for assessing prey (krill, pelagic fish) distribution;
  3. Improved satellite tracking and time depth recording devices for predators;
  4. Improved finer-scale resolution in remote sensing of sea ice coverage with differentiation of sea ice condition and other hydrographic conditions;
  5. Improved finer-scale resolution of sea ice conditions through aircraft observations;
  6. Biochemical methods for evaluating fish condition factors;
  7. Genetic markers for determining stock identity;
  8. Increased usage of Lagrangian drifters for assessment of current transport and advection.

7.3.10 Relation to other programs

The goals of this research are related to those of CCAMLR with respect to the effects of commercial fisheries operations and also to CEMP. Information derived from the JGOFS, WOCE, and FRAM programs are directly applicable to our studies. The research of relevant groups within SCAR is also relevant. The focused interest on fisheries dynamics and potential impact of climatic change in the Southern Ocean is shared with CSIRO which is developing a project to examine the trophodynamics of fish stocks along the continental slope of southern Australia.

7.3.11 References

Croxall, J. P., T. S. McCann, P. A. Prince and P. Rothery. 1988. Reproductive performance of seabirds and seals at South Georgia and Signy Island, South Orkney Islands, 1976-1987: Implications for Southern Ocean monitoring studies. In: D. Sahrhage (ed.), Antarctic Ocean and Resources Variability, Springer-Verlag, Berlin, 261-285.

Fraser, W. R., W. Z. Trivelpiece, D. G. Ainley and S. G. Trivelpiece. Increases in Antarctic penguin populations: Reduced competition with whales or a loss of sea ice due to environmental warming. Polar Biology, in press.

Laws, R.M. 1985. The ecology of the Southern Ocean. Am. Sci. 73, 26-40.

7.4 Physics/Climate Working Group Report

Chairman: Eileen Hofmann

Rapporteur: David Webb

Members: Joey Comiso
Jian-Hwa Hu
John Klinck
Peter Niiler

7.4.1 Overview

The Working Group's discussions focused on three broad categories of physical environments in the Southern Ocean that potentially could be affected by climate change which in turn could have effects on associated ecosystems. The first of these is the large scale circulation of the Southern Ocean. This was thought to be least known from a biological perspective and the one that may show the least effect of climate change. However, the large-scale system is one that is well represented in models and therefore climate change effects due to changes in circulation patterns could be investigated with modeling studies. The second, the sea-ice region, was considered to be an important environment and one in which climate change may have a noticeable effect. Observations of changes in sea ice can be made routinely with satellites and enough of an historical data base now exists to begin analysis and correlative studies of interannual variations in sea-ice cover. The final environment is that of the coastal ocean. This was considered to be the least studied of the three environments and the one that potentially may show the most effect of " climate change.

7.4.2 Climate change

Possible climate change scenarios have recently been reviewed by the International Panel on Climate Change. Work by a number of researchers indicates mean temperature increases, at the Earth's surface, of 4 degrees C are to be expected by 2150. Flux changes at the sea-surface will be of the order of 2 W m-2 (compared to a total flux of a few hundred W m-2). The percent of interannual variability in the heat flux is much larger than the expected climate change.

The accuracy of present climate predictions is limited by the relatively coarse (300 km grid) atmospheric and oceanic models used for climate research. However, the models indicate that in sea-ice regions the increase in summer temperatures will be substantially smaller than the global mean. Global warming will be delayed over the oceans with the greatest delay occurring in the Antarctic Ocean, just south of the belt of minimum westerlies.

The present models predict a reduced temperature contrast in the atmosphere between the equator and the poles. This will result in a reduced strength of the westerlies and a corresponding change in the strength of the Antarctic Circumpolar Current.

Finally, the models predict increases in the cloudiness of the atmosphere (resulting from increased evaporation). As the Southern Ocean is cloud covered, typically 80% of the time, even small increases in cloud cover could result in large changes in the biological productivity of the region.

7.4.3 Large-scale system

The large scale structure of the circulation in the Southern Ocean is controlled largely by the surface wind stress and the shape of the ocean bottom. The wind stress, in general terms, controls the strength (total transport) of the Antarctic Circumpolar Current. The bathymetry, on the other hand, controls the location of the current. In particular, the ACC is constrained to flow through Drake Passage, north of the Kerguelen Plateau, south of the Campbell Plateau and through the Eltanin Fracture in the East Pacific Rise. These gateways for the ACC determine its path through the Southern Ocean. The location of the atmospheric Westedies would have to shift by ten or more degrees of latitude in order to have any major effect on the structure of the circulation in the Southern Ocean.

The polar gyres near the Antarctic Continent may be differrent from the large-scale structure. The location of the Weddell gyre is strongly influenced by the Antarctic Peninsula. The other, suspected, gyres (e.g. in the Ross Sea) may become much more evident if the Easterlies along the continent became stronger. These gyres might also extend farther into the Southern Ocean if the winds were to change.

Within the large scale, Antarctic Circumpolar Current, there are narrow (about 50 km in width) high speed current jets that are associated with density fronts. These jets are separated by relatively low speed zones of about 100km width. A study of surface drifters shows that these fronts are associated with a secondary circulation that leads to flow convergence at the surface (surface drifters tend to collect over the fronts). The strength and importance of this secondary circulation has not been investigated nor has its effect on biological processes.

The high speed, narrow currents in the ACC are subject to flow instability which leads to mesoscale eddies. This eddy variability is evident in satellite altimetric observations, specifically in the measures of the time variation of the height of the sea surface. In fact, the ACC stands out in the Southern Ocean as a band of large flow variation. There is also measureable flow variation near the Antarctic Continent but it is not clear how much of this is due to the presence of ice.

Within the band of high variability associated with the ACC, areas of even higher variations exist. The largest magnitude of the eddy kinetic energy occurs in the Agulhas Retroflection and near the collision of the Brazil Current and the Falkland Current. Lesser hot spots are over the Kerguelen Plateau, the Macquarie Ridge, the Campbell Plateau, the East Pacific Rise and the Scotia Arc. The implication of this observation is that mesoscale variability is driven to some extent by the interaction of flow in the Southern Ocean (which penetrates to the bottom with only slightly diminished speed) with relatively shallow (less than 1 km) parts of the Southern Ocean.

A comparison of phytoplankton maps from CZCS and bathymetry reveals a striking necessary condition: high phytoplankton occurs in regions of large bottom slope. However, not every region of strong bottom slope is associated with high phytoplankton concentrations. Some of the regions of high phytoplankton are also areas of high flow variability, but not all. The relationship among flow variability, bathymetry and high phytoplankton concentration is not clear at this time.

7.4.4 Sea-ice region

Sea ice in Antarctica is one of the most seasonal parameters on the surface of the earth. In winter, it is a very extensive habitat, covering an area about 20 x 106 km2 and a large percentage of the Southern Ocean south of 50 S. In summer, only 20% of the winter ice cover remains. The immediate effect of the large seasonality is to cause seasonal modifications in the vertical structure of the underlying ocean. During growth, in fall and winter (about 9 months), the formation of ice causes the ejection of salt thereby decreasing stability of the mixed layer. During spring and summer, the retreat of the ice causes the introduction of large amounts of low salinity melt water to the surface providing vertical stability in the water column. It has been postulated that the presence of melt water is a key factor leading to phytoplankton blooms near ice edges. Melt water provides vertical stability in the water column and allows phytoplankton to grow in high-light high-nutrient environments.

During winter, the presence of leads and polynyas are also significant factors affecting the environment. Their presence is known to cause a considerable change in heat fluxes between the ocean and the atmosphere and salinity fluxes between the ice and ocean. Leads are linear and random features of open water (or new ice) in the ice pack and are known to constitute less than 10% of the ice cover. Polynyas are more rounded features and have been classified as either sensible heat polynyas or latent heat polynyas. The sensible heat polynyas which are usually in the deep ocean and can cover large areas are believed to be caused primarily by upwelling over topographical features (e.g. the Maud Rise). Latent heat polynyas are usually located along the coast and are formed by katabatic (or synoptic) winds. Biological populations have been observed to be considerably enhanced in lead and polynya regions. A careful monitoring of these features is therefore important.

Consistent records of ice extent from satellite observations have indicated no significant change in ice cover during the past seventeen years. However, there have been large regional variations. Large polynyas in 1974 through 1976 were observed in the Weddell Sea, but not in other regions. In 1980, the ice cover in the Weddell Sea was 15% larger than normal. This was compensated by smaller than average sea ice extents in other regions such as the Ross Sea and the Indian Ocean during this period. Long term effects of global warming would reduce the seasonality of sea ice and perhaps result in the eventual absence of summer ice. However, on the short term, the effect is not too obvious because of the complex feed backs that exist between ice, ocean, and the atmosphere.

7.4.5 Coastal circulation

Much of the physical oceanography research that had been done in the Antarctic has focused on the processes associated with the large-scale flow of the Antarctic Circumpolar Current or on processes that contribute to bottom water formation. With few exceptions, the regional and coastal circulation of the Antarctic has been ignored.

The historical hydrographic and current measurements that exist for the Antarctic are primarily concentrated in the Bransfield Strait-South Shetland Island region. These data reveal that the coastal flow in this region consists of complex circulation patterns that exhibit seasonal variability in strength and direction, in response to changes in wind stress and ice cover. The coastal currents are relatively narrow, being on the order of a few kilometers in width, but having large horizontal extent. For example the narrow westward flowing current on the north of the South Shetland Islands is thought to be circumpolar in nature. The coastal currents are influenced by bottom topography and coastal geometry, which can result in small scale variability.

Coastal regions such as the Bransfield Strait are areas where different water masses meet. This results in the formation of small scale frontal regions that can and do exhibit considerable variability in space and time. It is also likely that coastal flows are influenced by the amount of melt water from ice shelves and glaciers that is introduced each year.

Climate change effects could potentially affect the coastal circulation in the Antarctic through such processes as reduced inputs of melt water and/or changes in solar radiation. Either of these processes could alter water column stability, which would affect the intensity of veritcal mixing in coastal regions. Also, changes in the wind stress field would alter the intensity of the seasonal surface circulation.

7.4.6 Recommendations

The Working Group recommended that: There is a need for assembly and analysis of historical information. In particular the observations from shore-based stations in the Antarctic should be put into a standard format and made available. Such data sets would help in filling in the lack of long term observations of environmental parameters in the Antarctic.

Understanding the processes associated with sea-ice extent and variability are an important part of determining what (if any) effect climate change will have on the

Understanding of coastal circulation is a necessary component of addressing questions that relate to marine population fluctuations. Many species, such as krill, spawn on or near the continental shelf where their larval forms are dispersed by the coastal currents. Thus, understanding the factors that result in the successful recruitment of these species requires first a knowledge of the coastal current systems. The existence of shore-based laboratories makes coastal programs logistically feasible for the Antarctic.

There is a need for consistent and synoptic observations of sea ice and currents in the Antarctic. Attention should focus on designing measurement programs that use satellites, moored instrumentation and drifters.

7.5 Modeling Working Group Report

Chairman: Eileen Hofmann

Rapporteur: Victor Marin

Members: Joey Comiso
Jian-Hwa Hu
Tony Koslow
Dick Veit
David Webb

7.5.1 Overview

The Working Group's initial discussions focused on several broad issues that dealt with general aspects of modeling marine systems. Many of these general issues are already treated in the GLOBEC document on theory and modeling (GLOBEC, 1990). The Working Group suggests that interested individuals refer to this document for a discussion of the general modeling philosophy and issues that are relevant to the GLOBEC program. Issues that pertain to the development of models specifically for animal populations in the Southern Ocean were discussed by the Working Group. One area that needs development is that of sea ice models. Many of the marine populations in the Antarctic depend on sea ice during some or all of their life history. Hence, correct representation of interannUal variability in the extent of sea ice cover and/or its effect on these populations in models is important. It was also recognized that the results of large scale circulation models, such as FRAM, are a valuable resource. The Working Group discussed how the output from this type of model can be combined with finer scale regional models. The need for development of models that simulate the aggregation behavior of animals such as that observed for krill and its predators was noted. Much of the mortality of krill populations is due to predators such as penguins and seals. The inclusion of higher predators, that are decoupled from flow fields, in planktonic models was discussed by the Working Group. Futhermore, animals such as krill also decouple from the circulation field in the latter part of their life cycle. The Working Group discussed the approaches that could be taken to address this type of model. Expanded discussions of these points is given in the following sections.

7.5.2 Modeling issues

7.5.2.1 Modeling with uncertainties

For many of the zooplankton species of interest in the Southern Ocean there is incomplete knowledge of their life cycle. This presents problems in designing a model to investigate the biology/ecology of the species. Consequently, an approach would be to focus on those species for which most complete information exists (e.g. Euphausia superba, Calanoides acutus). A second approach is to focus on more conceptual models of life strategies, for example seasonal migrating and non-rmgrating species. These models should incorporate active migration as well as passive dispersal.

7.5.2.2 Matching fine and coarse scale resolution models

Any program undertaken in the Antarctic as part of the GLOBEC initiative most likely will have a regional focus (i.e., Bellingshausen Sea). However, even for a regional focus circulation models will need to include larger scale circulation effects. One way to incorporate these effects is to imbed a high resolution regional circulation model in a coarser resolution large scale model. A second way is to use a coarse large scale circulation model to provide the boundary forcing for the regional model. The techniques for matching solutions between different scale models are not well developed and this is an area of research that needs attention.

7.5.2.3 Compatible space and time scales between physical and biological processes

Biological processes encompass a large range of spatial and temporal scales. Often the scales of importance are not resolved adequately in circulation models. In particular, the vertical resolution of circulation models is often inadequate for considering biological processes. On the other hand, high vertical resolution is necessary to correctly represent air-sea exchange processes that are expected to occur due to climate change.

7.5.2.4 Sea-ice models

Thermodynamic models of sea ice are reasonably well developed. These models describe the growth and melting of a uniform ice field over the course of a year. Schemes for incorporating thermodynamic sea ice models into general circulation ocean models have also been developed. The simplest is essentially to advect the ice with the local current field and more complicated models treat ice as a plastic medium.

The present models do not realistically predict the ridging of sea ice or the formation of caverns by the rating of sea ice. Such models need development. The present models also do not attempt to describe the details of the flow field below the sea ice. However, this may be possible using a general circulation model with high resolution in the top 200 m.

7.5.2.5 Models for recruitment

There is a critical need to use Lagrangian calculations to look at the dispersal of holoplanktonic species and the planktonic stages of benthic, micronektonic, and nektonic species. This requires proper circulation models and also measurements of growth and the migration pattern. These calculations are relatively inexpensive and yield considerable insight on the dispersal and distribution of the species.

7.5.3 Recommendations

  1. Preliminary modeling efforts should be undertaken before field studies to ensure collection of optimal data sets. This part of the modeling would be based upon existing data and scientific intuition and where available incorporate historical time series data.
  2. Models of biological processes should be based upon physiological principles and basic biology. This is a basic premise of the GLOBEC modeling philosophy.
  3. The group strongly recommends the use of existing models especially mixed layer models, general circulation models, regional circulation models and sea ice models. The results of these models should be interfaced with biological models to investigate the role of physical and biological processes in determining biological distributions.
  4. There is a critical need to invest resources in developing sea ice models to give a realization of the circulation associated with the sea ice field.
  5. The group detected a need to transfer aggregation theories developed in terrestrial ecology to marine populations. This is of particular importance for models developed to study the swarming behavior of krill and models that treat the aggregation of predators in response to the swarms/patches of krill.
  6. Physiological and basic principles models require measurements on rates and processes. This in turn requires close cooperation between models and experimentalists.
  7. Biological models should explicitly incorporate large-scale variability in the physical environment.
  8. A comprehensive model should be constructed in the end to integrate the aforementioned results.

7.5.4 References

Theory and Modeling in GLOBEC: A First Step, Report to the GLOBEC Steering Committee from the Working Group on Theory and Modeling, February 1990, 9 pp.

7.6 Physiological Rates Working Group Report

Chairman: Sigrid Schiel

Rapporteur: Langdon Quetin

Members: Tsutomu Ikeda
Donal Manahan
Stanislaw Rakusa-Suszczewski
Robin Ross
Martin White

7.6.1 Needed physiological investigations

I. Determinations of physiological state

A. Considered GLOBEC Report Number 3 "Biotechnology Applications to Field Studies of Zooplankton" very important in regard to the need to investigate the following:

  1. Metabolism and locomotory ability
  2. Morbidity
  3. Diapause
  4. Egg production rates
  5. Growth rates
  6. Developmental stages
  7. Age
  8. Feeding rates and diet
[The group felt that GLOBEC Report Number 3 covered many important issues well. We sought not to duplicate the effort of that workshop, but raise some new issues and/or those specific to antarctic work.]

B. Determinations need to reflect different time intervals in the physiology of the animal, i.e., long term versus short term changes in the state of the animal.

C. Need to consider all of the approaches in Report Number 3 and work toward

  1. Reducing the number of measurements required
  2. Simplifying measurement techniques
D. Special importance should be given to resolving the debate about temperature compensation in antarctic species. Need to emphasize techniques to help address this issue. Need to contrast different life stages and organisms from different biotopes (new for Antarctic GLOBEC).

II. Seasonal studies need to be emphasized regarding

A. Environmental triggers of behavioral and metabolic events - in polar environments very small changes (i.e. 0.5 degrees C) may trigger change.

B. Comparisons of metabolic responses between extremes in the environment, e.g. summer versus winter. Most organisms investigated show seasonal cycles in metabolic activity.

C. May need new resources for station/ocean work throughout the year.

III. Physiological "plasticity" needs to be considered, especially to understand the capacity of antarctic animals to respond to environmental change (new to antarctic work)

A. Need detailed laboratory experiments to interpret simpler shipboard measurements in the context of an animal's metabolic history and consequences to its future.

B. Need to measure physiological responses by the target organism to conditions outside the normal environmental range normally encountered.

  1. Consideration given to likely environmental changes based on model predictions for effects of global change
  2. Identify possible changes in temperature, salinity, food, and ultraviolet radiation
  3. Vary environmental parameters in small increments because of high Q10's found to date in invertebrates
C. Need to understand physiological ontogeny of the organism, or how different developmental stages respond to variables in the environment
  1. Priority should be given to the early stages of development since their survival may be particularly sensitive to environmental conditions, concept of critical period during early life history
D. Need more effort toward understanding physiological state of the target organism in reference to environmental processes

E. Important to explore the idea of the ability of Antarctic species to procrastinate a physiological decision, especially whether it is a generality for Antarctic species

IV. Additional specific questions

A. Krill and salps do not generally occur together. Is this separation in part due to physiological differences between the two species, or primarily due to physical conditions in the environment?

B. What physiological parameters should be measured during the life cycle or at specific stages of the target organism that would most likely be an index of processes affecting population dynamics?

V. Target species

A. Attention should be given to not only studying a few species in detail, but also to studying a greater number of species in less detail.

  1. Detailed studies on a few species are important to evaluate which measurements are the most appropriate to emphasize.
  2. Studies on a broad range of species should illustrate the generality of results from detailed measurements on a particular species, i.e., using the chosen parameters
B. Logistical selection criteria for detailed studies
  1. Is a particular species suitable for experimentation?
  2. Is it possible to obtain all life stages?
  3. Is there historical data?
C. Target species for detailed study
  1. Plankton
    1. Euphausiids, recommend Euphausia superba
    2. Copepods, recommend Calanoides acutus
    3. Salpa thompsoni, because it generally appears to be found in the absence of krill
  2. Benthos
    1. Echinoderms, Sterechinus sp.
    2. Crustaceans, Notocrangon antarcticus
  3. Fish
    1. Pelagic
      1. Pleuragramma antarcticum (high latitude) or
      2. Nototheniops larseni (lower latitude)
    2. Demersal (near shore)
      1. Harpagifer or (larvae spend less time in water column, restricted distribution close to shore)
      2. Trematomus hansoni (near McMurdo Station, pelagic larvae, widely distributed over shelf)
  4. Seabirds
    1. Adélie penguin.
    2. chinstrap penguin
The above outline was agreed upon by all of the participants in the workshop, or at least it was clearly written on a chalk board and no strong objections were voiced. The following is the rapporteur's abbreviated account of notes taken during discussion. These notes may not be the consensus of the group, but are added at this point for further information.

The complexity of the benthos is greater than the planktonic community. For the benthos site selection should consider historical long term records, logistical constraints, and high and low latitude sites. The Ross Sea, McMurdo Sound and the South Orkney Islands should be considered possible sites. In the context of climate change the persistence of the benthic structure, the long term faunal and sediment record, and the ecological community structure should be considered. Some of the criteria for selection of target species shotrid be their abundance, whether they have a wide or restricted distribution, and whether measurable growth parameters exist.

The group discussed the merits of looking at a few species in detail or many species in less detail. The detailed approach was favored, but with a note that a broader range of species needs investigation at some level. For the detailed approach we need to consider how an organism responds to different carbon inputs, determine the physiological state of the organism and have the laboratory data to understand the implications. Ground truthing of physiological measurements and how different environmental variables affect the physiology of an organism is essential if we are to use these measures as assessments of the physiological state of an organism in the field (Fig. 1). Possible measurements included metabolic rate, growth, and (particularly for larvae) enzyme activity, amount of total protein and the pattern of synthesis of specific proteins. Concern was expressed about whether techniques were too sophisticated for field use and that they would not be something that "everyone" could do. Citrate synthase activity was suggested as an example of a useful and appropriate assay since the assay is simple, material needs to be frozen only at -80 degrees C, and citrase synthase is thought to be a good index of metabolism.

It may also be useful to better understand an organism's maximum potential versus what we actually see in the field. Growth rates would be a good example. Another question mentioned was how do we relate physiology to birth and death rates? or How do we relate physiological status to light, temperature, salinity and other abiotic factors? What followed was a general discussion of what to measure. Krill in Prydz Bay experience a constant low temperature compared to those west of the Antarctic Peninsula that encounter a 4-5—C range in temperature. The same species from different areas may show different physiological responses that may make "physiological state" difficult to interpret. It was mentioned again that growth was a good integrator of recent past environmental events. A combination of the biochemical and physiological approaches may be most suitable for GLOBEC. However, we need to be aware of potential problems. One example mentioned was the contribution of enzymatic activity from bacterial enzymes in krill stomachs to any analysis of krill.

Physiological rates should be evaluated in terms of their relevance to population dynamics. In addition, the relative importance of stages to measure since the work load may need to be prioritized. It was also emphasized that there should be a thorough review about what we know of particular species important to the study, and a critical evaluation of past research.

7.7 Population Dynamics Working Group Report

Chairman: Jarl-Ove Stromberg

Rapporteur: Adolf Kellermann

Members: Wulf Arntz
Tony Koslow
William Fraser
Suzanne Razouls
Jacqueline Grebmeier
William Stockton
Mark Huntley
The group felt that much of the parameters and processes relevant to study the dynamics of populations have already been tackled in the format I working groups. The apparent paucity of information needed to define populations of benthic, planktonic and warm-blooded organisms was acknowledged. As a pragmatic approach, it was agreed that spawning units may be identified in certain species which for a first attempt may be regarded as populations although exchange between them is documented but yet unresolved. The primary task for the group was then agreed as to identify the key gaps in the array of demographic parameters for the various taxonomic groups, and what variables influence population dynamics of these groups that may be sensitive to climate change. It was pointed out that even small water temperature changes may have significant impacts on growth and developmental rates, and hence on parameters relevant to population dynamics.

The potential study areas were the Bellingshausen Sea and adjacent waters to the east, but it was felt that because of logistic constraints of national ongoing and planned projects, other areas such as the Weddell Sea or Prydz Bay should be taken into consideration. These may also be utilized to look at a given target species under different latitudinal regimes, or the shipborne work in the primary study areas may be complemented by shore-based studies in different regions on e.g. rates and processes. In general, the group acknowledged that the most striking andimportant gap is the lack of data from the winter months for all taxonomic and ecological groups.

7.7.1 Benthos

Shipborne sampling periods are limited to the austral summer months. This may be improved by the use of ice-strengthened research vessels, and by future and present establishment of shore-based research. In analyzing length frequency data, the apparent longevity of many benthic organisms may obscure patterns that are useful for age and growth estimated. An apparent feature with respect to early life history seems to be the decrease of species having meroplanktic larvae 1) with latitude, and 2) with bottom depth. Another feature is the long developmental times of embryos which may contribute to circumantarctic distribution patterns, but may also be interpreted as waiting stage for favorable environmental conditions during larval drift. Field studies indicate that recruitment may be sporadic and irregular. Colonization should be studied in areas which are exposed after major ice shelf calving. Similarly, re-colonization and the succession of species may be studied in areas of high iceberg grounding frequency.

7.7.2 Fish

Although the shortcomings of traditional fishing methods were recognized, it was understood that there are no new techniques readily available. In recent years, population dynamics of the commercially harvested species has been studied in detail. It was agreed that stock assessment should not be the main objective in the study areas, although there is at present no commercial fisheries going on. Instead, the existing gaps are the proper assessment of larval and juvenile growth and developmental rates as related to biotic and physical environments. Key events in the life history such as hatching, settlement and first maturity have to be determined.

7.7.3 Zooplankton

The group reiterated the gaps that were identified by the Zooplankton and Krill WG format I, i.e., the need for both quasi-synoptic demographic surveys and processoriented cruises. Among particular processes, reference was made to the processes identified by that group.

7.7.4 Higher level predators

Populations in sea birds can be clearly identified and followed. Marking and tracking is feasible in sea birds. The bottleneck is apparently the winter months, especially with respect to foraging dynamics, i.e., food consumption, distribution relative to prey. Since more than 90% of the bird biomass consists of penguins, study efforts should focus on these. Adélie and chinstrap populations in the Antarctic Peninsula area have shown a decrease and increase in population size, respectively, over the past 40 years, which may well be related to changing degrees of pack-ice cover. Environmental conditions for these species seem to vary more in the Weddell Sea than in the Bellingshausen Sea. Some seal species may be regarded as ecological equivalents of these penguin species, e.g. the crabeater and fur seals.

7.7.5 Techniques

Standardized techniques have to be agreed upon in order to make comparisons in space and time possible.

7.7.5.1 Benthos

Both semiquantitative and quantitative sampling gears should be used. The first includes Agassiz-trawl (4 mm mesh) and video-systems, while quantitative gear comprise box cores, multiple corers and meiofauna corers. For the megafauna, video-systems and cameras can be considered as being quantitative. The minimum mesh size for sieving is 0.5 mm. Aging methods need to be developed, such as the use of hard parts in sea urchins or appendices in crustaceans, and these estimates need to be validated. Rearing experiments have not shown to hold great potential for this due to the lack of growth in some species in captivity.

7.7.5.2 Zooplankton

A continuous recording device is the Optical Plankton Counter (OPC) which is towed at 8-10 knots with a depth range of 300-0 m, and which is now commercially available. Net sampling has to be vertically stratified down to 2000 m with desirable free scale sampling within strata of 100 m. Multiple opening-closing nets should be used with mesh sizes around 250 um. These may be complemented by acoustic doppler systems and moorings. Under ice studies may be performed by SCUBA diving or hand operated horizontal tows through holes in ice floes or fast ice.

7.7.5.3 Fish

Early life stages should be obtained with the RMT 1+8, but also the international young cod net was recommended. For adult fish, standard 100-300 feet bottom trawls, and benthopelagic, high fishing nets should be used. Rearing experiments provide insights into the capacity of otoliths as recorders of past growth and environmental histories of fish.

7.7.5.4 Higher level predators

The techniques used are internationally standardized (CCAMLR). All data are considered quantitative. Tracking devices should be developed and utilized. Studies of the microstructure of seal teeth have revealed important insights into foraging patterns and success of fur seals, providing indicators of unfavorable conditions and environmental disturbances. Similar techniques should be tested for possible application in other seal species.

7.8 New Technology Working Group Report

Chairman: Jon Watkins

Rapporteur: Ken Smith

Members: Ulrich Bathmann
John Klinck
Inigo Everson
Valerie Loeb
Charles Greene
Stanislaw Rakusa-Suszczewski
In view of the large amount of discussion devoted to new technology in previous GLOBEC meetings (see Initial Science Plan, 1991; GLOBEC; North Atlantic Program 1991, Workshop on Biotechnology applications to field studies of zooplankton, 1991) the working group concentrated on developments necessary due to the unique physical environment of the Southern Ocean or specific requirements of target species that may not have been addressed elsewhere. Key areas identified were the effect of working in and under the ice and the problems associated with studying krill and salps. A summary of existing, developing and desirable new technology will be found in Table 1.

7.8.1 Physical oceanography and meteorology

a) Ice position, thickness and quality: there are many developments under way for obtaining this information through remote sensing by satellites and instrumentation on aircraft (see for instance Comiso, this report). The group recognized that there was a need for real time data on small spatial scales e.g. to investigate the local environment surrounding a ship. Potentially such information (e.g. at a scale of 1 - 100 km) could be collected from aircraft, balloons, drones and drogues. There was no expertise within the present group to assess the recent developments in remote measurements of ice quality. This was considered to be a very important pararneter and therefore should be addressed. Recommendation: aircraft logistic support for operations coupled with real time satellite data would be necessary. Satellite receiving stations capable of collecting such data either on ship or on adjacent bases would be necessary. b) Local weather and sea surface conditions: data such as wind speed/direction are available from WOCE meteorological buoys in addition to local and large scale surface circulation. Recommendation: there should be close coordination between GLOBEC and WOCE concerning meteorological and physical oceanographic data for study areas. Table 1: Summary of technologies now available (*), under development (**), or desirable but requiring development (***), for Southern Ocean GLOBEC investigations.

SENSOR        INSTRUMENTATION                    DEPLOYMENT PLATFORMS 

Mobile survey cruises

 Acoustics   Low-frequency acoustic array,       towed
             for school detection or tracking**

             Multi-frequency surface acoustics   towed body, hull mounted
             (existing), dual-, split-beam* (**)  

             Multi-frequency remote acoustics    multiple nets, towed bodies
             (prototype), dual-, split-beam**    and vechicles

             Acoustic Doppler Current Profiler   hull mounted
             (ADCP)*

 Optics      Optical particle counter**          nets and towed bodies
						            
             Video camera systems* (**)          towed bodies, vehicles and
                                                 benthic sledges

 Sampling    Automated sample processing         multiple and high-speed trawls
						             

Process oriented cruises

 Acoustics   Acoustic volume imaging systems**    ROV's, profilers,submersibles

             Side-scan sonar (others from above)*

 Optics      TV and still cameras, still* and     Profilers, ROV's, divers
 


Fixed location experiments

 Acoustics   Low-frequency acoustic array**
                        
	     Acoustic volume imaging**

             Acoustic transponder and receiving
             arrays for predator-prey studies**

 Optics      Micro video cameras for predator-
             prey studies (High definition)*(**)

Moorings
    
 Acoustics   Multi-frequency acoustics*(**)       vertical profiling arrays
	     ADCP's*(**)			     

 Optics      Longterm cameras* and videos         cameras such as Bathysnap
						             
 Sampling    Bottom landers for growth and
             physiology***
c) Structure of water column: in general it was thought that systems used or being developed by oceanographers for use in other areas were likely to be suitable for studies in a Southern Ocean GLOBEC (although see constraints under 2.a). The Group stressed that it was most important that oceanographic and biological measurements were coordinated and were measured over the same scales. Frequently the oceanography was determined at larger scales than those applicable to biological processes, especially those implicated in the swarming of krill. Recommendations: physical oceanography for small scale phenomenon - on the scale of meters to centimeters - would need to be accorded high priority. Relevant temporal and spatial scales of study for oceanography, phytoplankton, krill and predator dynamics are discussed in detail in Murray et al. (1988; see especially Fig. 8). A general treatment of scale-related issues for zooplankton is discussed in Marine Zooplankton Colloquium 1 (1988). d) Bathymetry: An understanding of this is vital because of the effect on currents. Recommendations: bathymetry of the study region should be well defined with multi-beam echosounders (such as SeaBeam) and side scan sonar.

7.8.2 Ice biology

a) Distribution and abundance of organisms: the presence of ice presents extra sampling problems in comparison to other areas. Once in the ice, ships are effectively stationary or cause much disturbance if steaming is attempted. Therefore remote sensing techniques must be developed further. Development of remotely operated or autonomous vehicles would allow under-ice surveys. In addition ice islands and ice-anchored drifting buoys could be used to provide extra information. Under-ice profiling could be carded out from moored arrays which could contain instrumentation such as transmissometers, fluorometers, ADCP, sediment traps, multifrequency acoustic profiling instruments. It is stressed that deployment of such equipment under the ice is not a simple case of using techniques and equipment developed elsewhere due to the remote location of the study sites and the inaccessbility of the equipment for much of the year. The development of equipment to make in situ observations on animals living within the ice was also thought to be necessary. Recommendation: non-invasive techniques to observe krill and zooplankton distribution, abundance and behavior in both ice-free and ice-covered areas should be accorded high priority (e.g. use of optical holography, multifrequency acoustics, etc.). Recommnendation: close coordination with the Sea Ice Working Group of SCAR and with SO-JGOFS should be established concerning the biological data for sea ice, under ice and water biota of the Southern Ocean. b) Physiology: It was felt that equipment capable of making in situ measurements of respiration, growth, etc. would be beneficial for benthic animals and the under-ice environment. It was pointed out that use of captive populations in large tanks (see Price eta/. 1987 for use of such a tank in Canada) or in enclosures in sheltered bays would provide valuable information on behavior and physiology (see Foote et al. 1989 for use of rafts and cages at South Georgia). Recommendation: large enclosures need to be developed further to simulate natural conditions for physiological studies of krill and other organisms. For example, enclosures in Admiralty Bay could be more cost effective than building large scale laboratory facilities ashore.

7.8.3 Species specific problems

a) Antarctic krill Euphausia superba: krill are frequently found at or near the sea surface (0-10 m). This depth range is particularly poorly sampled by nets and acoustics (see for example Everson and Bone, 1986a, on results from an upwardlooking echo-sounder). Moored upward-looking acoustic arrays could be capable of distinguishing water movement and acoustic backscatter from targets in the upper 5-10m in both ice-free and ice-covered areas. The effect of high sea states on the distribution of krill and other zooplankton/micronekton in surface waters was discussed. While this creates problems for all observation techniques it may not be severe because of the downward migration of animals under these conditions. Recommendation: instrumentation be improved or developed to examine the upper 10 rn of the water column and the undersurface of the ice. Many animals and krill in particular have been shown to avoid nets (see for instance Everson and Bone, 1986b). It was agreed that for krill, stealth nets capable of sampling with minimum avoidance at relatively high speed were desirable. The Group recognized the need for the development of non-invasive sampling techniques but stressed that these should be validated at the earliest opportunity. There was good evidence that krill were able to avoid divers, submerged cameras and other "non-invasive" systems. Recommendation: evaluate avoidance/attraction effects of measuring devices and deployment platforms on krill behavior. b) Salps: in addition to krill, salps must be adequately studied. Because of the delicate nature of salp aggregates, it is important to use a combination of nets and video systems to quantify them and distinguish aggregation sizes. Recommendation: improve large volume sampling techniques to determine abundance, biomass and distribution of salps with minimal disturbance to aggregates. It was suggested that a large volume water sampler monitored by video camera for triggering at appropriate times might be developed. It is important that sampling devices for both krill and salps are routinely available to take advantage of the alternate occurrence of these two species. c) Copepods: the Group did not discuss instrumentation specifically for copepod studies. It is likely that techniques mentioned in the North Atlantic proposal and under-ice biology would form the core of new developments.

7.8.4 Data management

The Group recognized that this was not the best forum to discuss data management but that a number of points should be highlighted at this time. Timely interchange of data, data access, data entry protocols and validation were all areas that could cause problems. A number of other international programmes have experience in setting up and administering databases (e.g. BIOMASS, WOCE). Recommendation: data management must be considered early in the development of GLOBEC in concert with other existing international programs.

7.8.5 References

Everson, I. and D. G. Bone. 1986a. Detection of krill (Euphausia superba) near the sea surface: preliminary results using a towed upward-looking echo-sounder. British Antarctic Survey Bulletin 72: 61-70.

Everson, I. and D. G. Bone. 1986b. Effectiveness of the RMT-8 system for sampiing krill (Euphausia superba) swarms. Polar Biology 6: 83-90.

Foote, K. G., I. Everson, J. L. Watkins and D. G. Bone. 1990. Target strengths of Antarctic krill (Euphausia superba) 38 and 120 kHz. Journal of Acoustic Society of America 87: 16-24.

Marine Zooplankton Colloquium 1, 1989. Future marine zooplankton research - a perspective. Marine Ecology Progress Series 55: 197-206.

Murphy, E. J., D. J. Morris, J. L. Watkins and J. Priddle. 1988. Scales of interaction between Antarctic krill and the environment. In: D. Sahrhage (ed.), Antarctic Ocean and Resources Variability, pp. 120-130, Springer, Berlin.

Price, H. J. 1989. Swimming behavior of krill in response to algal patches: a mesocosm study. Limnology and Oceanography 34: 649-659.