Underlying the surface water all locations was the Weddellwinter water. This layer reflected cooling and freezing whichoccurred during the preceding winter and had temperaturesbetween about -1.5° and -1.7°C and a salinity of about 34.5 partsper thousand. The winter water extended from just beneath thesurface water to about 100 meters and was separated from theupper waters by a region of very strong vertical gradients intemperature, salinity, and density. A weak (0.1°-0.2°C) tem-perature maximum layer 10-20 meters thick was often presentbetween the surface and winter water layers. This thermal fea-ture was a remnant due to absorption of incoming solar radia-tion during the preceding summer.

The deepest water mass observed was the Weddell warmwater, which extended from the bottom of the winter water atabout 100 meters to more than the 1,500 meters maximum castdepths. Temperatures in this layer varied between 0°C and0.5°C, and salinities were 34.65 parts per thousand at the upperboundary of the layer increasing to 34.7 parts per thousand atthe greatest observed depths. The greatest observed tem-peratures in this layer occurred at about 500 meters depth. Theregion between this temperature maximum and the top of thelayer at about 100 meters was characterized, over much of thestudy region, by large (more than 100 meters in vertical extent)temperature and salinity steps. These features have also beennoted by Middleton and Foster (1980). There is as yet no satisfac-tory explanation for their presence, though they may play asignificant role in vertical heat fluxes.

Dynamic heights relative to the 1,500-decibar level were com-puted to estimate whether appreciable baroclinic circulationwas present over the study region. The total dynamic relief wasof the order of only a few dynamic centimeters over the entirearea, indicating that the baroclinic circulation was negligible.

During the field program, two separate "rapid transects"were occupied by both vessels. There were attempts to samplealong transects over time intervals which were short relative tothe periods over which changes might be expected to occur inthe water column. No significant variations were detected be-

tween occupations of these transects, suggesting that the phys-ical system was reasonably steady-state during the 30-day dura-tion of the cruise.

Finally, no strong horizontal gradients, or frontal systems,were observed. This was in contrast to the AMERIEZ I program,which sampled in November 1983 along the strong frontalsystem separating the outflowing Weddell Sea water from thewaters of the Eastern Scotia Sea. The region sampled during theMarch 1986 program was characterized only by very gradualhorizontal variation—primarily in the upper layers, in movingfrom west to east away from the sea-ice edge.

The northwestern Weddell Sea MIZ region was intensivelysampled, revealing three water masses: the Weddell surfacewater, the Weddell winter water, and the Weddell warm water.The dominant upper layer feature was the widespread surfacewater layer, which was underlain by very strong vertical gra-dients. Such gradients imply that little vertical mixing is occur-ring through the bottom of the upper layer. The dominant deepfeature was the presence of vertical temperature steps whichmay be related to region-wide vertical heat transfer.

This work was supported in part by National Science Founda-tion grants DPP 84-20646 (to Muensch) and DPP 85-13098 (toHusby), by ONR contract N00014-82-C-0064 with Science Ap-plications International Corporation (to Muensch) and by theNational Marine Fisheries Service/National Oceanic and At-mospheric Administration (to Husby).

References

Carmack, E.C. 1977. Water characteristics of the Southern Ocean southof the Polar Front. In M. Angel (Ed.), A voyage of discovery. New York:Pergamon Press.

Middleton, J.H., and T.D. Foster. 1980. Fine structure measurements ina temperature-compensated halocline. Journal of Geophysical Research,85(C2), 1107-1122.

Summer-winter comparison of WeddellSea surface water and its productivity

C.-T.A. CHEN*

Institute of Marine GeologyNational Sun Yat-Sen University

Kaohsiung, TaiwanRepublic of China

The U.S.-U.S.S.R. Weddell Polynya Expedition (WEPOLEX)

(Chen 1982; Gordon 1982; Jennings, Nelson, and Gordon 1982)generated chemical data near the outflow region of the WeddellSea in the late winter/early spring. These data permitted inves-

* On leave from College of Oceanography, Oregon State University,Corvallis, Oregon 97331.

tigation for the first time of the cumulative winter effects on thedistribution of chemical properties in the Weddell Sea surfacewater.

Temperature, salinity, oxygen, and nutrient data all indicatethat the temperature-minimum layer found in summer is theremnant winter surface water. In summer, solar heating andmelting of sea ice transform the top layer of the winter surfacewater into a warmer, less saline water than the winter surfacewater which in summer is partially preserved beneath the toplayer. Subsequently, biological activity reduces the nutrient con-centration in the summer surface layer. The oxygen con-centration in the surface layer, however, increases in summerbecause the oxygen-depleted winter surface water picks upoxygen from the atmosphere after sea ice is melted. In sum-mary, the winter Weddell Sea surface water is colder, saltier,contains more nutrients but less oxygen than the summer sur-face water (Chen 1984; Gordon, Chen, and Metcalf 1984; Jen-nings, Gordon, and Nelson 1984).

Comparison of WEPOLEX and GEOSECS (Geochemical OceanSection Studies) normalized total carbon dioxide (NTCO 2 =TCO2 x 35/salinity, TCO 2 is the total amount of carbonate spe-

128 ANTARCTIC JOURNAL

20WEPOLEX GEOSECS

Summer Surface

L)°.. 0CD

-10

Winter Surface Remnant Winter

Surface Water

20 1/22002250 2300 2200 2250 2300

NTCO2 , )JmoI kg-'

Potential temperature (0) plotted vs. normalized total carbon diox-ide (NTCO2 ; "ii. mol/kg" denotes "micromoles per kilogram") for theWEPOLEX data (all stations) and the GEOSECS data (stations 79, 82,85,and 89).

cies dissolved in sea water) data (Takahashi etal. 1980; Huber etal. 1983; Chen 1984) also indicate that the winter NTCO7 valuesin the surface layer agree with the GEOSECS summer values in thetemperature-minimum layer (figure). The summer surfaceNTCO2 and normalized alkalinity values are lower than thewinter values by about 50 micromoles per kilogram and 10microequivalents per kilogram, respectively. These data indi-cate a production of 5 micromoles per kilogram of calciumcarbonate and 45 micromoles per kilogram of soft tissue, a ratiosomewhat lower than an average ratio of 1-to-4 for the worldoceans (Broecker and Peng 1982). The lower ratio is expectedbecause siliceous rather than calcareous organisms dominate inthe cold waters of the southern oceans.

The average amount of soft tissue production in the surfacelayer above the temperature-minimum layer is about 22.5 mi-cromoles per kilogram. This value translates to a productivity of300-450 milligrams per square meter per day between latewinter WEPOLEX and summer (cEosEcs) if we assume averagesurface layer to be 100 meters thick. The average calcium carbo-nate productivity is 33-50 milligrams per square meter per day.Although the effect of meltwater in summer would decrease theestimated productivity by about 2 percent, the effect of air-to-sea transport of carbon dioxide would probably offset the melt-water effect.

Primary productivity in the southern oceans shows muchspatial and temporal variability (Balech et al. 1968; Koblentz-Mishke et al. 1970; El-Sayed 1970; Lisitzin 1970; Goodell 1973).For instance, Lisitzin (1970) reports the diatom concentration inseawater to be between 0.5 x 10 6 and 1.0 x 10 per cubic meter.Such a large variability may explain partially why some directproductivity measurements have yielded low values when thegeneral assumption is that the productivity is high in the south-ern oceans (El-Sayed and Turner 1977; Jennings et al. 1984).

Our estimates were obtained following the method of Jen-nings et al. (1984) based on nutrient data. This method gives atemporally and spatially averaged result. Our estimated pro-ductivity of 300-450 milligrams per square meter per day agreesvery well with the results of Jennings et al. (220-420 milligramsper square meter per day) and supports the notion that the

productivity is indeed high in the southern oceans. Even if weassume that the average yearly productivity is reduced by halfas a result of the winter ice cover, we still have a high organiccarbon productivity of 55-82 grams per square meter per yearand calcium carbonate productivity of 6-9 grams per squaremeter per year. These values agree very well with the result, 54and 7.6 grams per square meter per year, respectively, obtainedfrom the recent sediment trap experiments in the southernoceans (Noriki, Harada, and Tsunogai 1985).

I acknowledge support from the U.S. Department of Energy(subcontract 19X-89608 C under Martin Marietta Energy Sys-tems, Inc. contract DE-ACO5-84 OR 21400 with Department ofEnergy), and the National Science Council of the Republic ofChina (NSC 76-0209-M110-03).

References

Balech, E., S.Z. El-Sayed, C. Hasle, M. Neushul, and J.S. Zaneveld.1968. Primary productivity and benthic marine algae of the Antarctic andSubantarctic. (Antarctic Map Folio Series, 10.) New York: AmericanGeophysical Society.

Broecker, W.S., and T.H. Peng. 1982. Tracers in the sea. Palisades, N.Y.:Eldigio Press.

Chen, C.T. 1982. Carbonate chemistry during WEPOLEX-81, AntarcticJournal of the U.S., 17(5), 102-103.

Chen, C.T. 1984. Carbonate chemistry of the Weddell Sea. (U.S. Departmentof Energy Technical Report, DOE/EV/10611-4.) Washington, D.C.:U.S. Government Printing Office.

El-Sayed, S.Z. 1970. On the productivity of the Southern Ocean. InM.W. Holdgate (Ed.), Antarctic ecology. New York: Academic Press.

El-Sayed, S.Z., and J.T. Turner. 1977. Productivity of the Antarctic andtropical/subtropical regions: A comparative study. In M.J. Dunbar(Ed.), Polar oceans. Calgary, Alberta: Arctic Institute of NorthAmerica.

Goodell, H.G. 1973. Marine sediments of the Southern oceans, the sediments.(Antarctic Map Folio Series, 17.) New York: American GeophysicalSociety.

Gordon, A.L. 1982. The U.S.-U.S.S.R. Weddell Polynya Expedition,Antarctic Journal of the U.S., 17(5), 96-98.

Gordon, A.L., C.T.A. Chen, and W.G. Metcalf. 1984. Winter mixedlayer entrainment of Weddell Deep Water. Journal of Geophysical Re-search, 89, 637-640.

Huber, BA., J. Jennings, C.T. Chen, J. Marra, S. Rennie, P. Mele, and A.Gordon. 1983. Reports of the U.S.-U.S.S.R. Weddell Polynya Expedition.(Hydrographical data, LDGO 83-1.) Palisades, N.Y.: Columbia Uni-versity Press.

Jennings, J.C., L. Gordon, and D.M. Nelson. 1984. Nutrient depletionindicates high primary productivity in the Weddell Sea. Nature, 308,51-54.

Jennings, J . , D. Nelson, and L.I. Gordon. 1982. Nutrient chemistryprogram during the U.S.-U.S.S.R. Weddell Polynya Expedition, Ant-arctic Journal of the U.S., 17(5), 101.

Koblentz-Mishke, O.J., V.V. Volkovinsky, and J.G. Kabanova. 1970.Plankton primary production of the world ocean. In W.S. Wooster(Ed.) Scientific Exploration of the South Pacific. Washington, D.C.: Na-tional Academy of Sciences.

Lisitzin, A.P. 1970. Sedimentation and geochemical considerations. InW.S. Wooster (Ed.), Scientific Exploration of the South Pacific. Wash-ington, D.C.: National Academy of Sciences.

Noriki, S., K. 1-larada, and S. Tsunogai. 1985. Sediment trap experi-ments in the Antarctic Ocean. In A.C. Sigleo and A. Hattori (Eds.),Marine and Estuarine Geochemistry. Chelsea, Mich.: Lewis Publishers,Inc.

Takahashi, T., W.S. Broecker, A.E. Bainbridge, and R.F. Weiss. 1980.Carbonate chemistry of the Atlantic, Pacific and Indian Oceans: The results ofthe GEOSECS expeditions, 1972-1978. (Lamont-Doherty Geological Ob-servatory Technical Report No. 1, CV1-80.) Palisades, N.Y.: ColumbiaUniversity Press.

1986 REVIEW 129