Figure 3. Bathymetric map of the Shackleton Fracture Zone/Elephant Island intersection. The map's location correponds to the large box in figure 1. The thick, black line is the R/V Polar Duke cruise track and the gray-shaded segments are track lines along which the R/V Surveyor collected Seabeam data. Note that the thickness of the shaded tracks varies with depth to the ocean floor. The bathymetric interval is 200 meters. See text for an explanation of the tectonic features.
transpressional faults associated with the Shackleton Fracture Zone may have played a direct role in the uplift of the forearc terrane exposed on Elephant Island. The study of the intersection of a major active plate boundary with the Antarctic Peninsula is part of an ongoing effort to understand the tectonic structure of plate boundaries and the processes involved in the uplift of subduction complexes within the Scotia Arc. Work aboard RIV Surveyor was supported by Antarctic Living Marine Resources, the National Science Foundation's Division of Polar Programs, National Aeronautics and Space Administration Geodynamics Program, Texas Advanced Technology Research Program, and the University of Texas Department of Geological Sciences. The R'V Polar Duke cruise was supported by National Science Foundation grant DPP 86-15307. We thank the National Oceanic and Atmospheric Administration officers and staff, especially Captain Christian Andreasen (National Ocean Service, Charting and Geodetic Services, Rockville, Maryland) for their help with the Seabeam
Quaternary marine geology of the northwestern Ross Sea DAVID E. REID
Department of Geology and Geophysics Rice University Houston, Texas 77251
On the final scientific expedition of the USCGC Glacier during during Deep Freeze 87, 29 piston and gravity cores and 128
data, Rodger Hewitt and Rennie Holt for allowing us to participate in their program, and the captains and crews of RIV Surveyor and RIV Polar Duke for their invaluable assistance in the implementation of our surveys in often difficult conditions. Special thanks go to Craig Berg, Gary Nelson, and Dennis Dixon for their excellent work in Seabeam data aquisition and trackline preparations.
References British Antarctic Survey. 1985. Tectonic map of the Scotia Arc. 1:3000000. [BAS (Misc) 3.J Cambridge: British Antarctic Survey. Dalziel, I. 'N. D. 1984. Tectonic eon! ution of a forearc terrane, Southern Scotia Ridge, Antarctica. (Geological Society of American Special Publication
200.) Boulder, Colorado: Geological Society of America.
approximately 700 kilometers of sparker seismic data were collected from the outer continental shelf and slope of the north western Ross Sea (figure 1). The cores were subjected to a detailed sedimentological study resulting in the identification of basal tills, glacial-marine sediments, a wide range of sediment gravity flow units, and carbonate deposits (Reid 1989). A primary objective of the study was to resolve the stratigraphy and lithiofacies relationships generated by the advance and retreat of the late Wisconsin ice sheet. This is important because the identification of basal tills succeeded seaward by glacial-marine deposits would establish the northernmost limit of grounded ice in this region during the last glacial maximum. A second objective was to characterize carbonate deposition in a polar glacial-marine setting. In the past, the presence of ANTARCTIC JOURNAL
Figure 1. Bathymetry and Deep Freeze 87 data location map of the northwestern Ross Sea. Note the proposed late Wisconsin grounding limit. Basal tills were identified south of this limit while glacial-marine deposits were identified north of the proposed limit.
carbonates has been considered by some to directly exclude a glacial origin for any associated diamictites (Schermerhorn 1974). A better understanding of this polar carbonate-glacigenic depositional system may aid in recognizing analogous depositional systems in the ancient record. Myers (1982) identified basal tills and petrological provinces supporting deposition by grounded ice as far north as Coulman Island (figure 1). The cores examined in this study extend the transect of cores previously examined, northward to the shelf break. Basal tills were identified in Deep Freeze cores located about 40 kilometers north of Coulman Island extending the known limits of deposition by grounded ice to approximately 73°11'S latitude (figure 1). Petrologic, seismic, and bathymetric data suggest that these basal tills were deposited from grounded ice which had advanced eastward from northern Victoria Land. Cores recovered seaward of the basal tills were determined to contain glacial-marine sediments. The glacial-marine sediments were distinguished from basal tills on the basis of textural, petrological, and clast shape differences, as well as the presence of abundant, well preserved, calcareous microfossils including planktonic foraminifera. Both the basal tills and glacial-marine deposits have a gradational upper contact with coarsening upward glacial-marine sediments. These deposits are, in turn, sharply overlain by gravelly sands which are interpreted as Holocene sediments. 1989 REVIEW
Textural, petrological, and paleontological data suggest that the glacial-marine sediments were deposited by intense ice rafting in close proximity to a grounded ice sheet in an (at least seasonally) open marine environment. The coarsening upward glacial-marine sequence is interpreted as representing an increase in current intensity coincident with the retreat of the expanded ice sheet following the late Wisconsin glacial maximum. A radiocarbon date from this interval (DF 87 number 6 25-40 centimeters) yielded an uncorrected date of 18,590 ± 500 ago, supporting the interpretation that the coarsening upward sequence is post glacial maximum in age. The proposed grounding limit for the late Wisconsin ice sheet derived from the core data is supported by a change in seismic character from data collected south of the limit to data collected to the north. Seismic data from south of the proposed limit are characterized by relatively high relief on individual reflectors, cross-cutting relationships, and wedge-shaped zones of acoustically chaotic reflectors which are interpreted as till tongues (figure 2a). In contrast, seismic data collected north of the proposed grounding limit are characterized by generally flat, coherent "normal marine" type reflectors separated by acoustically transparent zones. Relief on individual reflectors is relatively broad and gentle (figure 2b). Piston cores containing carbonate deposits were recovered from the outer shelf and upper slope and from the seaward 129
flanks of Mawson and Pennell-Iselin Banks (figure 1). These sediments are composed primarily of varying amounts of bryozoans, barnacles, and foraminifera with echinoids, pelycephods, ostracods, gastropods, and corals as accessory components. Four distinct carbonate facies, which trend eastwest along the shelf margin of the northwestern Ross Sea were identified (Reid 1989). The faunal assemblages and textural data imply a general west-to-east decrease in current intensity during deposition of the carbonates. Radiocarbon dates indicate that widespread carbonate sedimentation was occurring on the outer shelf margin of the northwestern Ross Sea during much of the late Wisconsin glacial episode (uncorrected dates range from 22,730 ± 900 old to more than 35,750 ago) and are, therefore, contemporaneous with widespread glaciation and glacigenic sedimentation on the Ross Sea continental shelf. In fact, the widespread distribution and relatively diverse calcareous fossil content implies open marine (ice-free) conditions for the outer shelf of the northwestern Ross Sea during this time. Further support for this interpretation is provided by the presence of planktonic foraminifera in all the glacial-marine and carbonate sediments examined, as well as the widespread occurrence of barnacle detritus (since barnacles have a planktonic larval stage). Current research is focused on obtaining isotopic and detailed faunal assemblage data from these carbonate sediments to understand more fully the implications for the late Wisconsin paleogeology and paleoceanography for the northwestern Ross Sea. This research was supported by National Science Foundation grant DPP 85-16908 awarded to John Anderson and is part of a Master's thesis study conducted at Rice University by David Reid. Sigma Xi, the research society, provided some funds for radiocarbon dating and their contribution is gratefully acknowledged.
References Myers, N.C. 1982. Petrology of Ross Sea basal tills: Implications for antarctic glacial history. Antarctic Journal of the U.S., 17(5), 123-124. Reid, D.E. 1989. Late Cenozoic geology of the northwestern Ross Sea, Antarctica. (Masters thesis, Rice University, Houston, Texas.) Schermerhorn, L.J.G. 1974. Late Precambrian mixtites: Glacial and/or nonglacial? American Journal of Science, 274, 673-824.
Figure 2. A. Deep Freeze 87 sparker data collected between coring stations 31 and 32, located south of the proposed late Wisconsin grounding limit. Seismic characteristics of cross cutting relationships, high relief on individual reflectors, and wedge-shaped acoustically chaotic zones suggest glacial erosion and deposition. Numbers indicate two-way travel time in seconds. (Contrast with B.) B. Deep Freeze 87 sparker data collected between coring station 6 and 7, located north of the proposed late Wisconsin grounding limit. Reflectors in this area are relatively continuous and flat-lying suggesting marine deposition with only a limited glacial influence. Numbers indicate two-way travel time in seconds. (km denotes kilometer.)
130
ANTARCTIC JOURNAL