Journal of Coastal Research Charlottesville, Virginia

"Submerged Shorelines" on Glaciated Continental Shelves: Solving the Puzzle?

M. G. Grosswald

Institute of Geography Academy of Sciences of the USSR 29 Staromonetny Street Moscow 109017, USSR ABSTRACT _

GROSSWALD, M.G., 1988. "Submerged shorelines" on glaciated continental shelves: solving ~,""' , the puzzle? Journal of Coastal Research, 5(1), 113-121. Charlottesville (Virginia). ISSN 0749­ .. •.. 0208. - ~s pr~posed A hypothesis which has been based upon an analysis of physical processes operating ~ ~ ",,0 at the grounding li ne of the Ross Ice Shelf, Antarctica. It suggests that the submarine terrace­ m:i shaped benches of glaciated shelves were formed at the grounding lines of the Pleistocene marine ice sheets, well below the former sea levels. Consequently, present-day depths of the -+ S-- benches cannot indicate the sea levels of the past.

AD,?ITIO~ALIN~EX WORDS: Submarine terraces, ice shelves, grounding line, tidal pump, glacial erosion, marine ice sheet, glacioeustasy, sea level changes.

INTRODUCTION ues to be a puzzling problem, In this context, the problem of origin of the former and latter Submarine terrace-shaped benches that benches must be treated separately. In fact, this broadly occur on the continental shelves and separation should be the starting point of the continental slopes of the world keep attracting present analysis, interest of geologists and oceanographers who There is little doubt that all submarine believe that the depths and ages of the land­ benches of "normal" shelves were worked out forms may give insight into the history of sea by waves at the ancient shores during gla­ level changes and vertical movements of the cioeustatic lowerings of sea level, but we must earth's crust, Such benches are characteristic of leave this out and focus on similar benches of all shelves and slopes, they are known on both glaciated shelves which are our subject. Prac­ "normal" and glaciated continental shelves tically all high-latitude shelves of both North­ being everywhere considered as features of the ern and Southern Hemispheres fall into this same origin, The latter, however, has never latter group, As it was pointed out, the benches been proved and seems extremely doubtful. of glaciated shelves were also regarded as for­ While the submarine benches of "normal" mer shorelines that became submerged by the shelves, e.g. those ofCaribbean islands, are real postglacial glacioeustatic transgression of the wave-cut platforms formed during sea level ocean. On the other hand, nearly all the shelves lowerings caused by former glaciations (which in question were definitely inaccessible for was extensively confirmed by uniformity of wave erosion and other subaerial geomorphol­ their depths as well as by numerous finds of ogical processes during cold epochs of the Pleis­ submerged subaerial materials on the plat­ tocene, in particular during Weichselian, or forms and by their age determinations), the ori­ Wisconsinan time, since they were covered by gin of similar features occurring on glaciated extensive ice sheets and downwarped by ice shelves and adjacent continental slopes contin- loads. Hence, it is hard to believe that the gla­ ciated shelves could bear any traces of ice-age 87050 received 12 December 1986; accepted in revision 20 September or preglacial sea levels, so that for the terrace­ 1988. like benches of the shelves one must look for Editorial note: This is an updated translation of a paper originally published in: Sea Level Changes During the Last 15,000 Years, P.A. some more appropriate, and non-controversial, Kaplin, (Ed.), 1982. Moscow: Nauka, pp. 125-133. (in Russian). explanation, -----"-- ---I

114 Grosswald

PRESENT STATE OF THE PROBLEM, marine benches exist in the vicinity of Bjor­ HISTORY OF RESEARCH noya; their soundings yielded depths of 70-80 m, 160-200 m and 250-350 m. The benches, 5 to Judging by publications, the submarine ter­ 50 km wide, are separated from one another by races and similar forms of the sea bottom have several (at least two) parallel submarine been most extensively studied in the North moraine belts that extend continuously Atlantic region and adjacent parts of the Arctic between the shores of northern Norway and Ocean. The discovery of the terraces is tradi­ Sorkapp, Svalbard, along the western margin of tionally connected to works by NANSEN (1904, the Barents shelf (Figure 1). Individual 1924), although, to the best of my knowledge, morainic ridges of the belts are up to 40 or 50 the earlier paper on the subject (describing sub­ km long, their sides slope very gently, and merged benches of the submarine outskirts of widths (2 to 8 km) are strikingly large relative the British Isles) dated back to the last decade to heights (10 to 15, sometimes 30 to 50 ml, The of XIX century (HULL, 1898). ridges are built up chiefly by aquatic till. The As for Soviet students of the problem, the ear­ outermost morainic belt appears to be a direct liest and most significant contribution was extension of the Egga moraine of the Norwe­ made by KLIONOVA (1938, 1948, 1960). Var­ gian shelf and, like the Egga, likely was formed ious aspects of the submarine-bench problem during the late Weichselian glaciation. As for were examined and discussed by SAKS (1953), the outermost benches, they have been worked LISITSYN and USINTSEV (1953), VINOGRA­ out in the continental slope making its profile DOVA et ai. (1959), KULIKOV and MARTY­ stepped. Such shape of profile is characteristic NOV (1961), GRABOVSKY (1968), RVACHEV of the Norwegian continental slope too: accord­ (1968), STRELKOV (1968), SHCHUKIN ing to KOTENIOV et al., the slope profile has (1969), LINDBERG (1972), MATISHOV (1972), been complicated by benches, having average MYSLIVETS (1973), KOTENIOV et al, (1976). widths of 5 km, at the depths of 350-300 m, 600­ Also, a number of foreign publications on the 800 m, 1150-1300 m, and 1650-1800 m. problem should be considered, and among them Geographical position ofthe most pronounced the papers by RIGG (1960), STANLEY et al . submarine terrace-shaped benches in the Eur­ (1968) and FILLON et ai. (1978). asian Arctic was shown on the "Map of Quater­ The above authors presented data on areal nary Deposits of the Arctic and Sub-Arctic" extent and depths of terrace-shaped benches (1965). There, and in the majority of other occurring in the North, Norwegian, Greenland, places, "submarine terraces" were identified on Barents, Kara and Labrador Seas. Judging the basis of purely morphological evidence, from available information, the landforms are though in a few sites their occurrence was also broadly developed within a considerable range confirmed by specific data on the lithology of of depths-from moderate (less than 200 m) and bottom sediments associated with the benches. significant (400 to 600 m) down to quite great Several hypotheses were advanced in order to (1500 m and deeper). In particular, according to account for the origin of the submarine forms: RVACHEV (1968), seven submarine benches the eustatic, neotectonic, and "astronomic", as occur on the western edge of the Barents con­ well as certain explanations based on their dif­ tinental shelf in the vicinity of Bjornoya, hav­ ferent combinations. Klionova who paid partic­ ing depths ranging from 50-70 m to 320-330 and ular attention to submarine benches of the Bar­ 370-380 m, while on the bottoms of Bjornoy­ ents continental shelf adhered to the first renna and Sorkapprenna their depths reach hypothesis, being sure that the benches had 450-510 m. For the island shelf of Newfound­ been formed during the planetary sea-level low­ land Rvachev pointed out submarine benches erings of the Pleistocene ice ages: "The occur­ with depths of 260-280 m, 290-330 m and 350­ rence of two main submarine terraces on the 380 m. According to FILLON et al, (1978), such floor of the Barents Sea should be considered as landforms off the shores of the northern Labra­ a reflection of two latest glaciations of the Pleis­ dor Peninsula occur at the depths of 130 to 400 tocene whose traces are so clearly imprinted m, some of them being tilted due north. upon landscapes of the Russian Plain", she Detailed surveys by KOTENIOV et al. (1976) wrote in 1938 (p. 78). Klionova believed that have confirmed that a set of pronounced sub- the upper terrace having the depth of 60-70 m

Journal of Coastal Research, VoL 5. No, 1, 1989 "Submerged Shorelines" on Glaciated Shelves 115

------/ - -- - _.------, --- - ( 1------15 ----=- ((( ------i : -~-( ( ( -- (( -- ( ( ( ( E2]7 ( ( ( ( ( l l ~l~V~I§~ l l ~ l I... l \..>.. ~10 --=-~-=-~ ~ ~ ------__-'7

t~jf2

Figure 1. Submarine geomorphic features, western margin of the Barents continental shelf and adjacent continental slope. Sim­ plified after Koteniov et 01.,1976. Legend: (1) land; (2) submarine terrace-shaped benches; (3) seabed zones ofridge-and-hiIl topog­ raphy; (4) individual morainic ridges; (5) surface of marine and glaciomarine deposition within submarine troughs; (6) outline of trough bottom; (7) smaller troughs; (8) boundaries oflargest riegels; (9) continental slope; (101 continental rise, upper part; (11) submarine outwash fans; and (12) continental rise, lower part.

was connected with the Valdai, or Weichselian, the last (the Valdai) glaciation which he called glaciation, while the deeper one having the the Sartan and Zyrianka glacial stades. This depth of about 200 m dated back to the epoch of (glacioeustatic) concept was further elaborated Dnieper, or Saale, glaciation (KLIONOVA, by STRELKOV (1968) who examined all the 1948, 1960). available results of hydrographical surveys in Klionova's views were shared by SAKS (1953) the Barents and Kara Seas and concluded that who, however, deemed more appropriate to the topography of their floors suggests not two, associate the two aforementioned benches not but three submarine terraces with the depths of with the Valdai and Dnieper glaciations but 70-100 rn, 200-300 m and 400-500 m which he rather with two-the late and early-stages of thought proper to associate with the late and

Journal of Coastal Research, Vol. 5, No.1, 1989 116 Grosswald 1 early stades ofthe Valdai and the main stage of ered the data on their depths to have been an the Dnieper glaciation, respectively. adequate basis for their long-distance, even The glacioeustatic hypothesis remains popu­ trans-oceanic, correlations. lar for the North Atlantic shelves till now, and The tectonic hypothesis that explains the all the evidence attesting to their Weichselian submarine forms in question by regional and ice-sheet glaciation are practically ignored. For local crustal subsidence within continental instance, MYSLIVETS et at. (1976) hold that margins has been mostly applied to account for two sets of the "submerged shorelines" occur­ the benches which either lie at exceptionally ring in the North Atlantic within the depth great depths or have irregular-sloping, undu­ range of 100-150 m must be connected with the lating or "broken"-longitudinal profiles. This late and early stages of the last glaciation, hypothesis was particularly often resorted to by while the deeper shorelines recorded at the NANSEN (1904), VINOGRADOVA et al. depths of 200-300 m belong to older ice ages, (1959), RIGG (1960) and others. Global neo-tec­ hence are also of glacioeustatic nature and good tonism was presumed to be a cause of the "sub­ for calculations of volumetric changes of the merged shorelines" by LINDBERG (1972), too, world's ocean. Two authors, LASTOCHKIN who hypothesized about huge worldwide Pleis­ AND FIODOROV (1978), believe that the Bar­ tocene changes in sea level that had resulted ents and Kara shelves bear several "surfaces of from the "geo-hydrocratic processes", i.e. marine accumulation and abrasion" and judge changes in volume of the oceanic basins. them to be vast submarine terraces which have Some researchers argue that the submarine recorded different stages of former, relatively benches can be adequately explained only on stable, levels of the Arctic Ocean; they even the basis of an "astronomic" theory, the theory attempted to date the surfaces by comparing connecting the origin of these forms with their depth with the curves of eustatic changes changes in the motion parameters of the Earth in sea level. Meanwhile, according to STAN­ as a space body. Thus, according to certain cal­ LEY et at. (1968), submarine benches of the culations by G. D. KHIZANASHVILI (1960) Nova Scotian shelf occurring at the depths of and G. G. KHIZANASHVILI (1963), main lev­ 145-155 m were worked out during the Illinoian els of marine terraces, both submerged and (Saalean) low stand of sea level, whereas simi­ emerged, have been formed as a result of "skew­ lar forms ofthe same shelfthat are at the depth ing" the ocean surface caused by abrupt of about 120 m, during another low stand which increases in the tilt of the Earth's axis. was associated with Wisconsinan (Weichselian) It is necessary to emphasize once more that cold time. all the above-mentioned hypotheses, despite The deepest submarine benches are com­ their differences, have one feature in common: monly thought to be of pre-Quaternary age. For instance, FILLON et at. (1978) held that the they all stem from the assumption that the sub­ terrace-shaped benches of the North-Labrador marine terrace-shaped benches of the discussed continental slope lying as deep as about 400 m North-Atlantic shelves were formed by seawave had been formed approximately two million activity operating at the former sea levels, and years ago. Similarly, RVACHEV (1968) the present-day position ofthe benches relative believed that all submarine benches of the to sea level is a result of subsequent submer­ North Atlantic with the depths in excess of 200 gence. However, this assumption looks most m must be considered as Pliocene formations; in implausible now, since the recent research has particular, the benches of such far-away demonstrated the following (HUGHES et al., shelves as the Barents and Newfoundland 1977; GROSSWALD, 1980, 1983; DENTON and shelves with the same depths of 300-330 m had HUGHES, 1981; GROSSWALD and VOZOVIK, been formed synchronously, during a hypothet­ 1984): ical pre-Quaternary regression of the World's (a) The North-Atlantic (as well as North­ Ocean. In other words, Rvaches, like so many Pacific) continental shelves were repeatedly other geologists, adhered to the concept that the glaciated by extensive Pleistocene ice sheets, benches of glaciated shelves were real coastal and it seems probable that the ice-sheet mar­ terraces which originated at former sea levels gins extended far into adjacent deep basins pro­ and subsequently submerged; hence, he consid- ducing ice shelves in the Arctic Ocean, the

Journal of Coastal Research. Vol. 5, No.1, 1989 "Submerged Shorelines" on Glaciated Shelves 117

Greenland, Norwegian, Labrador and even Ber­ tradict to the concept of the continental shelves ing Seas; glaciations. Contrary to all other explanation, (b) The latest ice-sheet glaciation of the con­ it suggests direct cause-and-effect relations of tinental shelves occurred during Weichselian the benches to the geomorphological processes (Wisconsinan) time, reaching its maximum operating at the grounding lines of former some 17-21 thousand years ago; marine ice sheets. (c) The weight of the marine ice sheets was Grounding lines are the boundaries delimit­ sufficient to depress underlying continental ing the internal i.i,e, grounded upon bedrock) shelves by hundreds meters, i.e. by vertical dis­ from the external (i.e. floating) parts of marine tance that considerably surpassed the range of ice sheets. The latter are the ice sheets that glacioeustatic lowerings in sea level. ground on bedrock lying well below sea level, With all this taken into account, it is but log­ their ice being in energy- and mass-exchange ical to conclude that during the eustatic regres­ with sea-water masses. Actual positions of the sions of ice ages the shelves in question were grounding lines result from interactions of sev­ beyond reach of the coastal processes, so that no eral factors, such as sea level, bedrock depth, real marine terraces could form on them. mass balance and velocity of ice sheet-and Besides, the shelves underwent glacial erosion, they change (advance or retreat) in response to while the adjacent continental slopes were variations of the factors. It is noteworthy that repeatedly buried by the masses of glaciomar­ even the tidal fluctuations of sea level cause ine sediments. For this reason, it would be illog­ quite perceptible diurnal displacements of ical to assume that there are real marine ter­ grounding lines, so that attempting to deter­ races of different ice ages or of pre-glacial age mine their positions one should speak either preserved on glaciated shelves or nearby con­ about zones of their daily migrations or about tinental slopes. Also the great amplitude of gla­ certain averaged situations. Judging by data of cioisostatic movements of the North Atlantic theoretical research, a grounding line of a continental margins, ensuing from calculations marine ice sheet can be stable only on a bed, i.e. and evidence, invites failure to all the attempts on a shelfor continental slope, which is inclined of trans-oceanic correlations of submarine "ter­ in a distal (offshore) direction; and the varia­ races" based on their depths. bility of its position is strongly dependent on Thus, we have to state that the eustatic the angle of bed inclination, decreasing on steep hypothesis is not applicable for an explanation bed and increasing on gentle. of the submarine terrace-shaped benches of the During maximum stages of each of the North North Atlantic. The same is true of the tectonic and "astronomical" hypotheses: once the gla­ Atlantic and North Pacific glaciations, includ­ ciated shelves can possibly bear only the ter­ ing the last of them, grounding lines of marine races formed after the last deglaciation, all the ice sheets were positioned on the edges of con­ dramatic events suggested by the hypotheses tinental shelves or on upper parts of adjacent must date from the Holocene, i.e. from last 9 to slopes. Waning of the ice sheets and eustatic 13 thousand years. But this appears quite rise in sea level during late glacial periods improbable since the crustal movements meant a retreat of the grounding lines-the related to internal processes of the Earth take retreat that always proceeded in a step-by-step much longer time, and no significant wandering way, so that the intervals of abrupt changes of the poles has been tracked down for the last alternated with intervals of temporary stabili­ 90 million years (KERR, 1987). All in all, it zations. We may infer therefore that there were seems clear that the whole set of above-men­ several, sometimes many, stages in stabiliza­ tioned explanations of the North Atlantic sub­ tion of grounding lines, corresponding to a sin­ marine terrace-shaped benches is incompatible gle glaciation, the last glaciation included; that with the concept of ice-sheet glaciations of the those different positions of grounding lines may polar continental shelves. range within a rather wide spectrum of depths; and that the difference of ages between extreme ATTEMPT TO SOLVE THE PROBLEM members of such spectra may be rather short­ not in excess of a few thousand years. The new hypothesis explaining the benches We propose that the terrace-shaped benches has been presented below which does not con- of the discussed glaciated shelves and slopes

Journal of Coastal Research, Vol. 5, No.1, 1989 r 118 Grosswald I were formed not by surficial waves at former water mass (CLOUGH and HANSEN, 1979; sea levels, but in submarine-subglacial envi­ ZOTIKOV et al., 1980), whilst observations on ronments at the temporarily stabilized ground­ the periodic level fluctuations of that ice shelf ing lines of the marine ice sheets that covered proved the operation oftidal pumping late Pleistocene lodgement bed erosion exerted by the grounded ice sheet till of the Ross Sea bed being, in turn, overlain gives way to glaciomarine deposition; (2) that by marine silts of the Holocene, suggests rather the water mass underlying the ice shelf in the high efficacy of the submarine reworking of the vicinity of the grounding line consists of two till surface. distinct layers-the lower one which is rela­ Thus, the proposed hypothesis not only per­ tively saline and warm and the upper one which mits to reconcile the fact of occurrence of sub­ is relatively fresh and cold; thus, being strati­ marine terrace-shaped forms on the polar con­ fied, that water mass should be susceptible to tinental shelves with the evidence suggesting formation of internal gravity waves; (3) that their ice-sheet glaciations, but also points to a the tidal pumping produced by the ice shelfmay possibility of another explanation of the be considered an adequate generator of the forms-of considering them as glacier-ero­ internal waves, and (4) that shoaling of the sional benches which were also subjected to waves can result in turbulence which, in turn, impact of internal gravity waves. Judging by may lead to bottom-sediment transport and the depths of the grounding line of the present­ sorting. day West Antarctic Ice Sheet, such benches Presently, these suggestions were confirmed could be formed some 500-900 meters below sea by the data of physical modeling, laboratory level, and in certain cases, e.g. in the mouths of experiments and field observations. In partic­ subglacial trough-valleys, even much deeper­ ular, an extensive program of drilling and geo­ down to 1200-1800 m. As for the submarine physical research in the Ross Sea, Antarctica, zones of irregular ridge-and-hill topography demonstrated that during the Pleistocene adjoining to outward margins of the benches, expansions of ice the floor of the sea was repeat­ they seem to be the products of glaciomarine edly subjected to glacial erosion while after deposition which operated on the distal (exter­ deglaciations it turned into an area of contin­ nal) side of the grounding line at the same time uous glaciomarine deposition, and that the as the abutting benches were being formed on boundary between the two seabed environ­ the proximal (internal) side of the line (Figure ments coincided and moved back and forth with 2). Hence, the typical occurrence of "terrace­ the grounding line of the Ross Ice Shelf (DEN­ staircases" on the polar shelves with the steps TON and HUGHES, 1981; KELLOGG et «i., lying at wide ranges of depths may suggest not 1979). Also, the thermophysical investigations repeated lowerings of sea level during different and radio echosounding in Antarctica suggest Pleistocene and pre-Pleistocene glaciations, as that basal melting takes place there over con­ it is broadly believed, but rather a step-by-step siderable areas of the ice sheet, which produces retreat of grounding lines of former marine ice annually, according to ZOTIKOV (1977), rather sheets in the course of a single (most probably a great masses of meltwater of which some 30 the latest) deglaciation hemicycle. Likewise, to 50 krn" comes out beneath the Antarctic ice the other features of submarine benches in shelves. In addition, direct measurements of question and zones of irregular (morainic) temperature and salinity distribution in water topography, such as their reciprocal parallel­ under the Ross Ice Shelf near the J -9 borehole ism clearly visible in Figure 1, an unusual for confirmed the double-layered structure of the land moraines-broad and flat-profiles of sub-

Journal of Coastal Research, Vol. 5, No.1. 1989 "Submerged Shorelines" on Glaciated Shelves 119

..

I <>: i....~:~:)+~ ...<:'.::::://.:\<.<:". ". ....:,::.', ,',. ," :. . ' ...... '.. ..-.', " .' ".:.: ::: ::'.. :.': :....:.:: ....: ...... " ". ',:'., ' ..',..' .. .. ', ,

... ' . ", ... '.

c..-'--'-~...L.-'-~ ) J'~-.'" t: '.' ~.-:?~-.. ~. ~···7/ II /0 ~~.-:--O;O-.."T.••. ~~ y~ xgr /'/ ly>12

Figure 2. Formation of ice-erosional benches and abutting zones of ridge-and-hill moraines associated with rise in sea level and step-like retreat of an ice shelfgrounding line. zb and zH, consecutive positions of sea level; x], and X~lr, zones of diurnal migrations of grounding lines; I and II. consecutive stages of ice-sheet retreat. Legend: (1) glacier ice; (2) water circulation beneath the ice shelf (relatively warm and saline water); (3) the same (relatively cold and fresh water); (4) submarine accumulations of glaciom­ arine sediments; and (5) bedrock.

marine ridges and hills, and geochemical affin­ ments at the grounding lines of the past ice ity of the sediments the ridges and hills are sheets. made up of to the Holocene marine deposits, This attempt at solving the puzzle of subma­ reported by KOTENIOV et al., (1976), are all in rine terrace-shaped benches occurring on polar good agreement with the hypothesis suggesting shelves was first made by the author several formation of both the beaches and ridge-and­ years ago (GROSSWALD, 1980), and it is hill zones in submarine and subglacier environ- encouraging for the hypothesis that in follow-

Journal of Coastal Research. Vol. 5. No. I, 1989 I 120 Grosswald I

ing years some marine geologists arrived inde­ KELLOGG T.B.; TRUESDALE R.S., and OSTERMAN pendently at the same conclusion (FILLON and L.E., 1979. Late Quaternary extent of the West Ant­ arctic Ice Sheet: new evidence from Ross Sea cores. HARMES, 1982). Geology, 7(5), 249-253. Hence, the depths of the terrace-shaped KERR R.A., 1987. Tracking the wandering poles of benches of the polar glaciated shelves provide ancient Earth. Science, 236 (4798), 147-148. evidence neither on former sea levels nor on the KHIZANASHVILI G.D., 1960. Dynamics of the Earth's Axis and Ocean Levels. Tbilisi: Uchpedgiz. range of the vertical movements of the Earth's (In Russian). crust. Instead, the forms, being the traces of KHIZANASHVILI G.G., 1963. Origin of the sub­ temporally stable positions of the marine ice merged marine terraces in the light of a hypothesis sheet grounding lines, shed light on the former of the Earth's axis dynamics. Okeanologiya, 3(5), geography of the boundary between grounded 930-935. (In Russian). KLIONOVA M.V., 1938. Submerged shorelines in the and floating portions of the ice sheets at differ­ Barents Sea. Proceedings of the Soviet Section, ent stages of their history. More generally, this International Association Studies Quaternary all suggest that a specific, the deepest on the Period, 4: 3-5. (In Russian). Earth, "denudation level" appears to be asso­ KLIONOVA M.V., 1948. Marine Geology. Moscow: Uchpedigiz, 495p. ciated with grounding lines of the marine ice KLIONOVA M.V., 1960. Geology of the Barents Sea. sheets, both former and existing, at which a Moscow: Publishing House Academy of Sciences, specific complex of geomorphological processes 367p. (In Russian). operated or operate. KOTENIOV B.N.; MATISHOV G.G.; BELYAYEV A.V., and MYSLIVETS V.I., 1976. Geomorphology of the shelf and continental slope between Spitsber­ LITERATURE CITED gen and Northern Norway. Priroda i Khozyaistvo Severa, 4, 30-38. (In Russian). CACCIONE D.A. and SOUTHARD J.B., 1974. Incip­ KULIKOV N.N. and MARTYNOV V.T., 1961. ient sediment movement by shoaling internal grav­ Ancient shorelines on the floor of the Kara Sea. In: ity waves. Journal ofGeophysical Research, 79(15), Morskiye Berega, Tallinn (Proceedings Institute 2237-2242. Geology, Academy Sciences Estonian SSR), pp. 147­ CLOUGH J.W. and HANSEN B.L., 1979. The Ross Ice 154. (In Russian). Shelf Project. Science, 203(4379), 433-434. LASTOCHKIN A.N. and FIODOROV B.G., 1978. DENTON G.H. and HUGHES T.J., (Eds.), 1981. The Relief and recent history of the North- Eurasian Last Great Ice Sheets. New York: Wiley-Intersci­ shelf. Geomorfologiya, 3, 19-27. (In Russian). ence, 477p. LINDBERG G. U., 1972. Major Changes of the Ocean FILLON R.H.; FOLINSBEE R.A., and PALMER R., Level During the Quaternary. Leningrad: Nauka, 1978. Deep shelf and slope terraces off northern 548p. (In Russian). Labrador. Nature (London), 273(5665), 743-746. LISITSYN A.P. and UDINTSEV G.B., 1953. About FILLON R.H. and HARMES R.A., 1982. Northern ancient shorelines of the sea floor. Izvestiya AN Labrador Shelf glacial chronology and depositional SSSR. Ser. geogr., 1,23-31. (In Russian). environments. Canadian Journal ofEarth Sciences, MATISHOV G.G., 1972. Geomorphology of the sub­ 19(1), 162-192. marine margin off Western Greenland. In: Promys­ GRABOVSKY N.A., 1968. Seabed geomorphology in lovyye ryby severo-zapadnoi Atlantiki i usloviya the vicinity of the Great Newfoundland Bank. ikh obrazovaniya, Murmansk (Proceedings of Okeanologiya, 8(4), 92-96. (In Russian). PINRO, No. 28). (In Russian). GROSSWALD M.G., 1980. Late Weichselian ice sheet MIROPOL'SKY YU.Z., 1975. On the effect of current of Northern Eurasia. Quaternary Research, 13(1), 1­ 32. with a velocity shift on the generation of short­ GROSSWALD M.G., 1983. Ice Sheets of the Continen­ period internal waves in the ocean. Izvestiya AN tal Shelves. Moscow: Nauka, 216p. (In Russian). SSR .FizikaAtmosfery i Okeana, vol. 11(9),933-941. GROSSWALD M.G. and VOZOVIK Y.I., 1984. A (In Russian). marine ice cap in South Beringia (a working MIROPOL'SKY YU.Z., 1977. Theory ofInternal Grav­ hypothesis). Polar Geography and Geology, 8(2), ity Waves in the Ocean. Moscow: VINITI (Results of 128-146. Science and Technology). (In Russian). HUGHES T.J., 1977. West Antarctic ice streams. MYSLIVETS V.I., 1973. Terraces on the external Reviews of Geophysics and Space Physics, 15(1), 1­ shelf of Eastern Canada. In: Metody prikladnoi i 45. regional'noi fizicheskoi geografii. Moscow: MGU HUGHES T.; DENTON G.H., and GROSSWALD Publications, pp. 83-87. (In Russian). M.G., 1977. Was there a late-Wiirm Arctic ice sheet? MYSLIVETS V.I.; KALININA L.I., and SOLOV'IOVA Nature (London), 266(5603), 596-602. G.D., 1976. Attempt to calculate the volumetric HULL E., 1898. Further investigations regarding the changes of the World's Ocean during the Pleisto­ submerged terraces and river valleys bordering the cene. In: Problemy Paleogidrologii. Moscow: Nauka. British Isles. Transactions Victoria Institute (Lon­ (In Russian). don), Volume 30. NANSEN F., 1904. The bathimetrical features of the

Journal of Coastal Research, Vol. 5, No.1, 1989 "Submerged Shorelines" on Glaciated Shelves 121

North Polar seas, with a discussion of the continen­ Submerged terraces on the Nova Scotian Shelf. Zeit­ tal shelves and previous oscillations of shore-line. schrift Geomorphol. N.F., Supp!. Bd. 7, 85-94. In: The Norwegian North Polar Expedition 1893­ STRELKOV S.A., 1968. Centers of glaciation in the 1896: Scientific Results, Volume 4. Christiania. north of Siberia, conditions and stages in their NANSEN F., 1924. Spitzbergen. Berlin. development. In: Neogenovyye i chetvertichnyye RVACHEV V.D., 1968. On some geomorphological otlozheniya Zapadnoi Sibiri. Moscow: Nauka, pp. features of the Newfoundland and Labrador shelves 88-98. (In Russian). and continental slopes. Okeanologiya, 8(4), 659-665. SHCHUKIN 1.8., 1969. On certain cosmic and plane­ (In Russian). tary factors in formation of the Earth's topography. RIGG J.B., 1960. On the possibility of existence of a In: Problemy planetarnoi geografii. Moscow: MGU submerged shore line at a depth of 300 metres in the Publications. (In Russian). north-eastern Atlantic. Weather, 15(7), 226-231. ZOTIKOV LA., 1977. Thermal Regimen of the Ant ROBINSON E.S., NEUBURG H.A.C. and WILLIAMS arctic Ice Sheet. Leningrad: Gidrometeoizdat, 168p. R., 1974. Ocean tides beneath the Ross Ice Shelf. (In Russian). Antarctic Journal of the U.S., 9(4), 162-164. ZOTIKOV LA.; ZAGORODNOV V.S., and RAIKOV­ SAKS V.N., 1953. The Quaternary ofthe Soviet Arctic. SKI YU.V., 1980. Core drilling through the Ross Ice Leningrad-Moscow: Minmorrechflot (Proceedings of Shelf, Antarctica, confirmed basal freezing. Science, NIIGA, Volume 77), 627p. (In Russian). 207(4438), 1463-1465. STANLEY D.J.; DRAPEAU G., and COK A.E., 1968.

o RESUMEN 0 Se propane una hip6tesis basada en el analisis de los procesos fisicos que tienen lugar en la linea de base de la plataforma de hielos Ross en la Antartida, A partir de ella se sugiere que las terrazas submarinas de las plataformas glaciares se formaron sobre la linea de base de las placas de hielo marino del Pleistoceno bastante por debajo del nivel del mar primario. Consecuentemente, la prfundidad actual de las terrazas no puede indicar el nivel del mar pasado.~DepartmentofWater Sciences, University ofCan tab ria, Santander, Spain.

o RESUME 0 On propose une hypothese basee sur I'analyse des processus physiques qui agissent la au 1a banquise de Ross, dans I'Antarctique, touche le fond de la mer. Elle suggere que les terrasses sous-marines des platesformes continentales qui on ete englacees se sont formees 130 ou les glaces marines du Pleistocene touchaicnt Ie fond, bien en dessous des anciens niveaux de la mer. Aussi, la profondeur actuelle de ces terrasses n'est-elle pas unn indicateur des niveaux de la mer dans le passe.-Roland Paskoff, Uniuersite Lumiere de Lyon, France.

o ZUSAMMENFASSUNG 0 Es wird eine Hypothese zur Losung des Problems der "untergetauchten Kustenliriien" des glazial ubcrformten kontinentalen Schelfes vorgeschlagen. Diese Hypothese beruht auf der Analyse physikalischer Prozesse, die an der Basis des Ross-Ice-Schelfre­ gion der Antarktis wirksam sind. Diese Analysen lassen den Schlup zu, dajs die submarinen terrassenformigen Stufen des glazial uberforrnten Schelfes an der Basis des ple istozanen marinen Eisschildes geformt wurden, und zwar deutlich unterhalb des fru­ heren Meeresspiegels. Konsequenterweise kann die heutige Tiefenlage der Stufen kein Indikator fur die Lage des Palaomeeres­ spiegel sein.-Ulrich Radtke, Geographisches Institut, Universitiit Dusseldorf, F.R.G.

Journal of Coastal Research. Vol. 5, No.1, 1989