nitz and Barnes 1974; Josenhans, Zevenhuizen, and Klassen References 1986). The similarity of form and the availability of deep draft ice indicate to us that the antarctic seabed features were also formed Anderson, J. B., C. Brake, E. W. Domack, N. Myers, and R. Wright. 1983. Development of a polar glacial-marine sedimentation model from by iceberg keels. Antarctic Quaternary deposits and glaciological information. In B.F. The origin of the sub-circular, flat-floored depressions are Molnia (Ed.), Glacial-marine sedimentation. New York: Plenum Press. also postulated to be related to modern iceberg keels (Lien 1981; Barnes, P.W. 1986a. Distribution of water depths on the Antarctic continental Barnes 1986). Subglacial processes, permafrost collapse, slope shelf seaward of the continental land and ice mass. (U. S. Geological Survey failure, and mass movement are not favored as mechanisms to Open-File Report, 86-599.) Washington, D.C.: U.S. Government form these depressions. Their close association with ice goug- Printing Office. ing and wobbling also suggests iceberg keels are responsible for Barnes, P.W. 1986b. Morphologic studies of the Wilkes Land continental these features. Iceberg with drafts between 400 and 500 meters shelf glacial and iceberg effects. In S.L. Eittreim and M.A. Hampton are probably uncommon; however, recurrence intervals of tens (Eds.), The Antarctic continental margin: Geology and geophysics of offshore Wilkes Land. to a few hundred years may be sufficient to maintain the ob- (Circum-Pacific Council for Energy and Mineral Re- sources, Earth Science Series, Vol. 5A.) Houston, Texas. served morphologies. Josenhans, H. W., J. Zevenhuizen, and R. A. Klassen. 1986. The Quater- Vorren et al. (1983) introduced the term iceberg turbate for nary geology of the Labrador Shelf. Canadian Journal of Earth Sciences, seabed strata developed from the tilling action of iceberg keels 23, 1190-1213. on the seafloor. We suggest that ice keel turbates are widespread Keys, J.R. 1983. Iceberg quantities, shapes, and sizes in western Ross on the antarctic shelf. Ice reworking the surficial sediment may and DUrville Seas. Antarctic Journal of the U.S., 18(5), 125-127. be forming a surficial unstructured marine diamicite which may Lewis, J.C., and G. Bennett. 1984. Monte Carlo calculations of iceberg correspond to the glacial marine deposits reported by Ander- draft changes caused by roll. Cold Regions Science and Technology, 10, son et al. (1983). 1-10. Side-scan sonar data from the Weddell Sea and the Wilkes Lien, R. 1981. Seabed features in the Baaenga area, Weddell Sea, Ant- arctica. Port and ocean engineering under Arctic conditions 81, Proceedings. Land shelf indicate ice gouge incisions a few meters deep and Quebec, Canada, University of Laval. tens of meters in width, down to depths of over 500 meters. In Orheim, 0. 1980. Physical characteristics and life expectancy of tabular the Weddell Sea, a "washboard" pattern and hummocky bed Antarctic icebergs. Annals of Glaciology, 1, 11-18. features have been formed by the interaction between ice keels Orheim, 0., and A. Elverhoi. 1981. Model for submarine glacial deposi- and the seabed to depths exceeding 300 meters. The freshness tion. Annals of Glaciology, 2, 123-128. of seabed morphology, and Holocene sediment ponding at Reimnitz, E., and P.W. Barnes. 1974. Sea ice as a geologic agent. In J. depths greater than 500 meters, indicate that the seafloor is Reed and J. Sater (Eds.), The coast and shelf. Arlington, Va. presently being reworked by iceberg keels. We conclude that a Vanney, JR., and G.L. Johnson, 1985. GEBCO bathymetric sheet 5-18. modern ice keel turbate deposit may be widespread on the In S.S. Jacobs (Ed.), Oceanology of the Antarctic continental shelf. (Ant- portion (54 percent) of the Antarctic Shelf less than 500 meters arctic Research Series, Vol. 43.) Washington, D.C.: American Geophysical Union. deep, probably in the form of an unstratified to poorly stratified Vorren, TO., M. Hald, M. Edvardsen, and 0. Lind-Hansen. 1983. marine diamicton. Glaciogenic sediments and sedimentary environments on continent- The field work aboard the S.P. Lee was in part supported by al shelves: General principles with a case study from the Norwegian the National Science Foundations program at McMurdo Station Shelf. In J. Ehlers (Ed.), Glacial deposits of northwest Europe. Rotterdam: and Christchurch, New Zealand. Balkema.

Preliminary results of marine geological and geophysical investigations in the Bransfield Strait, Antarctic Peninsula

JOHN D. JEFFERS

Department of Geology and Geophysics Rice University Houston, Texas 77251

The 1986-1987 austral summer marked the first season iIv Polar Duke (figure 1) operated in antarctic waters as a fully equipped marine geophysical research vessel. Newly installed equipment included a single-channel seismic reflection data acquisition system, nuclear precession magnetometer, and 3.5- and 12-kilohertz echo sounders. Seismic reflection, magnetic, and bathymetric profiles were collected in the Bransfield Strait Figure 1. R/V Polar Duke at Palmer Station, Anvers Island.

1987 REVIEW 131 62W 62S 60W 58W 61S 56W

63S 54W

61S

64S

62W 62S

60W 58 W 56 W 63S 54W

Figure 2. Locations of seismic reflection profiles collected in Bransfield Strait. Broken lines show data collected by R/v Polar Duke during the 1986-1987 austral summer; dotted lines show data collected in previous seasons by USCGC Glacier. Bathymetry contoured at 500 meters.

region of the northern Antarctic Peninsula on two Polar Duke to the axis of the basin. The tectonic, volcanic, and sedimentary cruises (figures 2 and 3). evolution of each of these basin segments has been somewhat Bransfield Strait is a young basin (<4 million years; Barker different. Although the exact timing of the evolution of and Daiziel 1983) formed by rifting of the South Shetland Is- Bransfield Strait cannot be determined because the basement lands from the northern Antarctic Peninsula. Our objective was rock and overlying sediments have not been sampled, our seis- to use the high-resolution geophysical capabilities of Polar Duke mic data suggest that chronologies constructed by other work- to study the evolution of this geologically complex basin. Our ers (Birkenmajer 1982; Barker and Daiziel 1983) are essentially specific interests include the nature and timing of the ki- correct. The development of the Brans field Basin was intimately nematics which led to its formation, the important plate tectonic related to plate tectonics in the adjacent Drake Passage ocean controls on its development as a sedimentary basin, and the basin. The pre-middle Neogene magmatic arc of the South interplay between plate tectonics, volcanism, and sedimenta- Shetland Islands was deformed by strike-slip faulting during tion throughout its history. the Miocene and early Pliocene. Boyd Strait and Bransfield During October 1986, marine geophysical data were collected Strait opened in response to the cessation of subduction at the in the northeastern portion of Bransfield Strait. Approximately South Shetland trench during the Pliocene. From the Pliocene 750 kilometers of seismic reflection profiles were obtained; most until recent time, the Bransfield Basin has evolved from a sim- of our effort was directed toward acquiring data along the axial trend of the basin, from south of King George Island to south of Elephant Island. These data provide important insight into the SOUTH L_ 1L_LT TNORTH structure of the Bransfield Basin and allow correlation among the seismic reflection data collected in two previous seasons by t USCGC Glacier (figure 2). In addition, 19 geologic stations were occupied where bottom sediments were sampled by grab sam- pling, piston coring, and gravity coring. Heavy pack ice re- stricted our planned operations in the southeastern portion of Bransfield Strait. .1 Thirteen days of cruise IV-87 in April 1987 were devoted to marine geophysics. Single-channel seismic reflection profiles were collected in the southwestern portion of Bransfield Strait. The seismic and magnetic data collected in the vicinity of De- 2 : .- .. $ ception Island will be valuable in relating the recent volcanic activity there to the submarine volcanism farther northeast in Bransfield Strait. Data collected in Boyd Strait will be used to relate the opening of the Boyd Strait basin to the history of the 14. main part of the Bransfield basin. 3 Preliminary assessment of the data collected in the 1986-1987 season shows that Bransfield Strait is divided into three main sub-basins separated by the two bathymetric divides associated Figure 3. Seismic reflection profile PD86-1 1. Vertical scale in sec- with Bridgeman and Deception islands. Numerous smaller onds; horizontal extent approximately 45 kilometers. Vertical exag- basin segments were formed by faults trending perpendicular geration 14 x at seafloor.

132 ANTARCTIC JOURNAL pie rift basin to a marginal basin displaying the characteristics of daand P. Penhale for their assistance while they were aboard. the early stages of seafloor spreading (Jeffers, Anderson, and Research was supported by National Science Foundation grant Lawyer in press). DPP 85-16908. The scientific party of cruise VIII-86 included J.B. Anderson (chief scientist), L.R. Bartek, J.D. Jeffers, D.S. Kennedy, and References N.C. Myers, all of Rice University; L.A. Law yer of the Institute for Geophysics, University of Texas at Austin; and J. Breza of Barker, P.F., and I.W.D. Dalziel. 1983. Progress in geodynamics in the Florida State University. Participants in cruise IV-87 included Scotia Arc region. American Geophysical Union, Geodynamic Series, 9, J.A. Austin (chief scientist), J. Dunbar, J. Giltner, and D.S. 137-170. Sawyer, all of the Institute of Geophysics, University of Texas at Birkenmajer, K. 1982. Late Cenozoic phases of block-faulting on King Austin; and J.D. Jeffers and W. Smyth of Rice University. The George Island (South Shetland Islands, West Antarctica). Bulletin de support and enthusiasm of the captains and crews of LAcadémie Polonaise des Sciences, Serie des sciences de la terre, 30(1-2), Polar Duke 21-32. and Bruce Carter, Bob Crimmins, John Hutchison, and Vince Jeffers, J.D., J.B. Anderson, and L.A. Lawyer. In press. Evolution of the Kelly of ITT/ANS contributed immensely to the success of our Bransfield Basin, Antarctic Peninsula. Fifth International Symposium for scientific efforts. We thank G. Bloom, H.W. Detrich, T. Maran- Antarctic Earth Sciences, Cambridge, England.

high degree of meltwater activity, although there is little direct Late Quaternary sedimentation and observation of plumes to support this idea. It has been postu- glacial history in the Gerlache Strait lated that terrigenous material entering the water in meltwater region Graham Land, Antarctica plumes is doing so below sea level, from subglacial sources (Singer 1987). The surface sediments and seismic data indicate that there are major differences in sedimentation on opposite sides of the THOMAS W. GRIFFITH Gerlache Strait (Singer 1987). Along the Danco Coast, on the eastern side of the Gerlache Strait, surface sediments are di- atomaceous and ponded sediment is uncommon. However, on Department of Geology and Geophysics the western side of the strait, in the bays and fjords in the Rice University Palmer Archipelago, sediments have a much higher terrigenous Houston, Texas 77251 component and appear to be ponding in every available bathymetric depression. Indeed, in Lapeyrere Bay (northeast Anvers Island) data indicate that a confined delta is prograding During the austral summers 1982-1983, 1985-1986, and from the head of the bay. 1986-1987 scientists on board the USCGC Glacier undertook a Differences in the microclimate from side to side of the study of sedimentation within the Gerlache Strait region. The Gerlache Strait may account for the observed differences in data used in this study consist of 76 sediment samples (includ- sedimentation. A strong precipitation gradient exists in the ing piston cores and bottom grabs), 70 kilometers of high- region, with the Palmer Archipelago receiving over twice the resolution single-channel seismic data, and a videotape record precipitation as the Danco Coast. In addition, much more of this of approximately 90 percent of the coastline surrounding the precipitation is in the form of rain, which should be easily study area. These videotapes were used to outline the glacial advected to the base of the glaciers through the myriad cre- setting of the bays and fjords. While most of the smaller glaciers vasses that dissect the glacier surfaces. This may help explain are grounded within a few meters of sea level, major valley and the apparently abundant meltwater there. outlet glaciers may be grounded at depths as great as several Each bay posesses a highly characteristic sedimentary regi- hundred meters. A bathymetric chart was also made from avail- me, depending on factors such as microclimate, glacial setting, able data (figure). The major goals of the project were to charac- and subglacial geology. Although it will be possible to syn- terize sedimentation within the fjords and bays of the region, thesize an integrated model of fjord sedimentation for the entire and to shed some light on the areas Late Quaternary glacial region, it is important to recognize the individualistic character history. of each body of water sampled. Terrigenous material, in the form of ice-rafted debris, ava- There is copious geomorphic and bathymetric evidence for lanche debris, and fine-grained particles suspended in melt- the existence of a once-greatly expanded tidewater glacier oc- water plumes is entering the marine environment against a cupying the Gerlache Strait, but there is very little record of the background of diatomaceous sedimentation. Unlike other areas ices actual retreat to present-day conditions, either on the seis- of the antarctic continental shelf, water depth exerts only a mic data or in the sedimentary record. Using the criteria of subordinate control over sediment distribution. Most coarse Anderson et al. (1980, 1984), only one sediment sample has material is deposited within a few kilometers of the coast, in- been identified as basal till; the remaining samples represent dicating that ice rafting is not occurring today on a large scale. either compound glacial marine sediments or mass flow deposi- The volume of fine-grained terrigenous sediment indicates a ts of various types. Seasonal ice cover and the protection af- forded by land masses effectively limit the distribution of re- sidual glacial marine sediments in the region. The data imply a Present address: Union Pacific Resources Company, Houston, lixas 77002 rapid decoupling and retreat of the ice during Holocene warm-

1987 REVIEW 133