Introduction
Drift-ice abrasion marks are common along rocky shores in cold regions. Although they have been reported several times from many regions during the last century (Reference DionneDionne, 1970), their importance has been pointed out only recently, at least in English and French (Reference McLellanMcLellan, 1971; Reference DionneDionne, 1973; Reference Laverdière, Lavardière, Guimont and DionneLaverdière and others, 1981; Reference Dionne, Dionne, Lavardière and GuimontDionne and others, 1983). In the past, drift-ice abrasion marks were generally considered of little interest. In text-books, they are usually classified with other erratic abrasion marks on rock surfaces made by solifluction, landslides, mud flows, and avalanches (Reference Embleton and KingEmbleton and King, 1975, p. 184).
In coastal areas glaciated during the Quaternary, such abrasion marks are sometimes confused with glacial abrasion features. As a result, some authors reported multi-directional striations, sometimes up to seven or eight directions in the same locality, and they attributed the various sets to changes in glacier-flow direction.
Recent field work along the east coast of Hudson Bay in sub-Arctic Quebec shows that several sets of striations observed along present-day rocky shorelines are drift-ice abrasion marks (Reference Dionne, Dionne, Lavardière and GuimontDionne and others, 1983). The purpose of this paper is to report and characterize abrasion marks made by drift ice, to discuss briefly their significance, and to focus the attention of Quaternary geologists on a neglected topic.
Drift-Ice Abrasion Marks
The various drift-ice abrasion marks include polished surfaces, scratches, striations, small grooves, and even minor friction cracks (Reference Laverdière, Lavardière, Guimont and DionneLaverdière and others, 1981). These marks usually occur on relatively soft rocks such as shale, sandstone, limestone, dolomite, and basalt, and occasionally on harder rocks such as granite and gneiss.
Polished surfaces
Polished surfaces resulting from abrasion by drift ice are generally easily distinguished from glacially polished surfaces by the following characteristics. First, drift-ice polished surfaces occur only on relatively soft rocks. The surface is very small compared to a glacially polished surface: a few square meters compared to many hundreds of square meters for glacially polished surfaces. Secondly, the surface, even when wet, has a dull aspect, strongly contrasting with the shining aspect of the glacially polished surface which, in addition, is commonly intensely striated. Another subtle difference is that on a glacially polished surface there is often a thin varnish-like crust, about 1 mm in thickness, which spalls with weathering. This feature is absent on drift-ice polished surfaces, which closely resemble mechanical abrasion surfaces produced by sand and gravel washed by waves and currents or by a cavitation process in a fluvial environment.
Thus, because of their entirely different appearance, an experienced geologist can easily distinguish between drift-ice and glacier-ice polished surfaces. However, difficulties may occur if the exposed bedrock surface is small. In such a case, a careful and cautious examination of the surface is necessary.
Despite the relatively common occurrences of drift-ice or iceberg-polished surfaces, there are only a few mentions of this phenomenon in the literature, and many of these are suspicious (Reference ForchhammerForchhammer, 1847, p. 1177; Reference HayesHayes, 1868, p. 415; Reference HindHind, 1877, p. 230; Reference LoeweLoewe, 1935, p. 267; Reference BirdBird, 1967, p. 221; Reference DionneDionne, 1970, p. 176–77). Considering their use in reconstructing the geologic events in an area, drift-ice polished surfaces are still of little interest.
Scratches
Scratches are surficial abrasion features scattered over rock surfaces (Reference Laverdière, Laverdière, Guimont and PharandLaverdière and others, 1979). Although varied in form and size, they are usually 2 mm or less in depth and occupy only very small and scattered areas usually less than a few square decimeters. They reflect the erratic action of drift-ice over a rock surface. These features, often reported as striations, are quite frequent along present rocky shorelines (Figs 1 and 2). However, due to the mode of formation, they are rare on glacially polished surfaces which are more commonly dominated by striations, grooves, and chattermarks. As scratches differ in many aspects from striations, it is useful to classify these abrasion marks in a different category (Reference Laverdière, Laverdière, Guimont and PharandLaverdière and others, 1979).
Striations and small grooves
Striations are the commonest drift-ice abrasion feature reported in the literature (Reference HayesHayes 1868, p. 415; Reference Feilden and De RanceFeilden and de Rance, 1878, p. 486, 556; Reference GarwoodGarwood, 1899, p. 215; Reference LeiviskäLeiviskä, 1905, p. 216; Reference LeffingwellLeffingwell, 1919, p. 174; Reference GouldGould, 1928, p. 37; Reference MathiassenMathiassen, 1933, p. 32, 69, 94; Reference FrӧdinFrӧdin, 1956; Reference Strӧmberg and SundiusStrӧmberg and Sundius, 1956; Reference IgnatiusIgnatius, 1958, p. 11–12; Reference BirdBird, 1967, p. 221; Reference StrӧmbergStrӧmberg, 1971). Though frequently confused by some authors with glacial striations, they can be distinguished by the following characteristics. First, they usually cover small surfaces of a large exposed bedrock surface. Secondly, in a given locality, they are commonly multi-directional and show a criss-cross pattern (Figs 3–5), contrasting with glacial striations which are usually mostly parallel over long distances. Thirdly, they are usually short (a few centimeters to a few decimeters compared to 1 m or more), shallow, isolated, intermittent, divergent, discontinuous, and irregular in width, length, and depth. In addition, they are often sinuous, curved, and crooked. Because drift-ice striations are restricted to modern or former coastal areas, their geographical distribution is another distinguishing criterion, since glacial striations can occur widely throughout a glaciated area.
Reference RussellRussell (1890, p. 117) indicated that on river beds and shores “the scratches are rudely parallel to the direction of the river current, but are not deeply engraved. The scratches are made by sand and pebbles frozen in the ice”. A few authors have also provided some characteristics of drift-ice striations (Reference NansenNansen, 1897, p. 488; Reference PrestPrest, 1901, p. 337–38; Reference WashburnWashburn, 1947, p. 47–48; Reference FrӧdinFrӧdin, 1956, p. 26–29; Reference Strӧmberg and SundiusStrӧmberg and Sundius, 1956, p. 140–41).
In deeper water, drift-ice striations are also made by icebergs (Reference GeikieGeikie, 1882; Reference ChalmersChalmers, 1886, Reference Chalmers1895; Reference LoeweLoewe, 1935; Reference FrӧdinFrӧdin, 1956). Long ago, Reference ChamberlinChamberlin (1888, p. 228) mentioned that striations made by icebergs “are curved, crooked, and intermittent, of irregular width and depth”. He also indicated that “curvature of a higher degree than those likely to be produced beneath a glacier would undoubtedly be formed by grounding bergs and floes”. However, the distinction between drift-ice and iceberg striations is a difficult task even for an experienced field geologist, particularly in coastal uplifted areas.
In addition to striations, small grooves and even minor friction cracks made by drift ice have been reported recently (Reference Laverdière, Lavardière, Guimont and DionneLaverdière and others, 1981). Thus the criterion of the absence of friction cracks in small grooves discussed by Reference FrӧdinFrӧdin (1956) to distinguish between glacial and drift-ice abrasion marks is not conclusive.
Drift-ice striations are also found on rock fragments: boulders, cobbles, and pebbles (Reference LyellLyell, 1846; Reference DawsonDawson, 1886; Reference JennessJenness, 1952; Reference NicholsNichols, 1961; Reference TremblayTremblay, 1967). Reference RussellRussell (1890, p. 120) mentioned that “the scratches on the smooth face are less regular and less firmly drawn than grooves and striations on typical glaciated pebbles”. The characteristics of drift-ice striations on pebbles, cobbles, and boulders have been provided by Reference WentworthWentworth (1928) and Reference QuéréQuéré (1966). Despite some differences, it is usually difficult to distinguish between abrasion marks on erratics made by floating ice and glacier ice, particularly when the rock fragments have been transported and redeposited. Along modern shorelines, such as the St. Lawrence Estuary and eastern Hudson Bay, many boulders in the tidal zone are abraded on one face only, i.e. the exposed surface. In this case, it is relatively easy to determine the origin of striations and scratches. As underlined by Reference FlintFlint (1971, p. 95), “a great deal of judgment is needed in the identification of drift-ice abrasion marks on rock surfaces and particularly on erratics”.
Drift-ice abrasion marks are usually made by boulders, pebbles, and any other rock fragment frozen at the base of an ice cover or an ice floe, and then pushed along the rock surface by wind, tide, waves, or currents. They are also made by isolated blocks or boulders in a rock platform pushed by shore-ice ride-up and pile-up in a marine and lacustrine environment. Abrasion marks are made either by a movement of boulders in front of drift ice and below it. Depending upon the availability of debris, the characteristics of the rocky shoreline, the nature and the varying direction of the pressure, abrasion marks occur as a scattered feature throughout the surface. They are thus oriented in various directions and have different ages.
Discussion
Drift-ice abrasion marks are common along the present rocky shorelines of the St. Lawrence Estuary and particularly Hudson Bay. In sub-Arctic Quebec, they occur everywhere along the eastern shore of Hudson Bay, particu-larly on the gently sloping, volcanic rock (basalt) platforms at Long Island, Manitounuk Islands and Manitounuk Peninsula, Nastapoka Islands, and all along the mainland rocky shoreline (Reference Allard and TremblayAllard and Tremblay, 1983). They are usually superimposed on the former glacially polished and striated surface. In places, they are so abundant that the former glacially polished and striated surface is difficult to identify (Fig. 3). In many localities, up to eight or nine directions of striations and scratches have been measured. Of course, no more than one or two can be attributed to glacier flow which has a constant direction shifting from 250° to 300° in this area (Reference Hillaire-MarcelHillaire-Marcel, 1976). The other directions are thus questionable. Figure 4 and 5 illustrate drift-ice striations superimposed on a glacially polished and striated surface (the stylo points to the glacial flow direction).
It is difficult to understand how a glacier can striate successively a bedrock surface in several directions, preserving at the same time most of the abrasion marks made previously. If a glacier moving over a bedrock surface is able to polish, scratch, striate, or groove it once in a well-defined direction, how can the same glacier, in changing its flow direction several times, subsequently striate again the same bedrock surface such that the first, second, third, and even fourth sets of striations are partially preserved and not erased? There is a fundamental problem here that has not yet been resolved, possibly because the capacity of a glacier to striate and scratch bedrock surfaces repetitively without erasing the previous marks has not been questioned seriously until recently.
In coastal areas, ice flow of a relatively thin glacier stranded in shallow water is apt to change direction several times according to the calving bergs process. In addition, surficial abrasion marks related to the thinning out of the ice, in combination with the buoyancy effect, can be produced without destroying completely a former striated bed-rock surface. As a result, it is likely that two or three sets of well-defined striations are true glacial abrasion marks. However, in this particular environment, multi-directional sets of striations should be regarded as problematic abrasion marks, and the drift-ice hypothesis should, at least, be considered seriously (Reference FrӧdinFrӧdin, 1956, p. 26–29; Reference Strӧmberg and SundiusStrӧmberg and Sundius, 1956, p. 140–41).
Because drift-ice processes are erratic and irregular from year to year, the resulting abrasion marks have various orientations related to the direction of ice floes pushed onshore by winds or by pressures from pack ice. As mentioned above, the nature and size of abrasion marks vary considerably from place to place and only small surfaces are scratched or striated at the same time. The relatively small vertical pressures of ice floes compared to those of a glacier permit the formation of new sets of striations and scratches without entirely destroying the older ones. As a result, several sets of striations of different ages can occur at the same site. This is the case along the rocky shorelines of Hudson Bay.
Because drift-ice abrasion marks have not always been considered seriously in the past, at least in North America, some misinterpretations about bedrock striations do exist. For example, up to five sets of striations have been reported by Reference LeggetLegget (1980) at Grand Manan Island, Bay of Fundy, and up to eight directions have been recorded by Reference Andrews and FalconerAndrews and Falconer (1969) at Ottawa Islands, Hudson Bay. The multi-directional striations at these two localities were considered exclusively as glacial abrasion marks related to very “excited” glaciers moving successively in various directions during the deglaciation of these areas. Since the drift-ice hypothesis was not discussed or even suggested, it is questionable whether glaciers produce all the striation sets reported. It is worthwhile mentioning that these islands have been largely dominated by drift-ice processes since deglaciation.
In other areas, the situation is more complex and misinterpretations may have resulted. This is the case for Hopen Island, Svalbard (Reference Hoppe, Hoppe, Schytt, Häggblom and ӦsterholmHoppe and others, 1969), where scattered striations occurring on rocky shorelines were used first to indicate Pleistocene glaciation. In a recent paper, Reference HoppeHoppe (1981, p. 68) recognized that “the striae most probably are an effect of movement of drift or pack ice on the shore, a common feature in Swedish coastal areas”. It is relevant to mention here that in Scandinavia many glacial geologists (Reference LeiviskäLeiviskä, 1905; Reference FrӧdinFrӧdin, 1956; Reference Strӧmberg and SundiusStrӧmberg and Sundius, 1956; Reference StrӧmbergStrӧmberg, 1971), having worked in coastal areas where well-exposed bedrock surfaces occur, have recognized and discussed the drift-ice hypothesis in the case of problematical abrasion marks.
The criss-cross and multi-directional striations and scratches which occur on a limestone outcrop in New York State, as reported by Reference RitterRitter (1978, p. 379), are so similar to the drift-ice striations found along the rocky shorelines of Hudson Bay (Figs 4 and 5) that it is difficult to attribute all of them to variations in the glacial flow direction during the late Pleistocene. In the same way, the drag-mark casts on Precambrian sandstones in Australia as reported by Reference MiltonMilton (1966, p. 293), could also be attributed to drift-ice rather than to drifting algae, because no kelp is known to have existed at that period, while glaciations have been reported during Precambrian time in Australia (Reference Harland, Herod, Wright and MoseleyHarland and Herod, 1975; Reference Young and JohnYoung, 1979).
Finally, it is useful to point out that variations in the direction of ice flow may reflect purely local bedrock influence, i.e. bedrock irregularities which deflect ice flow. As a result, it is likely that in a given locality two or three sets of striations effected by the same glacier could be explained without invoking the drift-ice hypothesis. However, these striations are rarely superimposed and seldom intersect at right-angles.
Conclusions
Drift-ice abrasion marks along modern shorelines are quite common. The purpose of this paper is to focus the attention of Quaternary geologists, not yet aware of the problem, on the occurrence of such features not only along present-day shorelines but also on any former marine and lacustrine shorelines. With this in mind, it is suggested that many ambiguous or problematical so-called glacial striations (Reference McLellanMcLellan, 1971) reported in the past could perhaps be satisfactorily explained without the necessity of involving several changes in the direction of the glacier ice flow in a given locality during the late Pleistocene. Any set of striations restricted to a small surface and differing in direction from other glacially oriented features, such as streamlined outcrops, stoss-and-lee features, crag-and-tails, and chattermarks, and not related to local bedrock irregularities, should be questioned seriously in areas which have been submerged during late- and post-glacial time. Consequently, a re-examination of the data presented in some reports is necessary for a more adequate glacier ice-flow reconstruction in many coastal areas.
Acknowledgements
The author gratefully acknowledges the financial assistance of the National Research Council of Canada (NSERC Programs) and the Centre d’Études Nordiques for field work along Hudson Bay and the St. Lawrence Estuary shorelines. P.B. Clibbon (Université Laval) has reviewed a first draft of this paper, while C. Laverdière and P. Guimont (Université de Montréal), and D.R. Grant (Geological Survey of Canada) provided useful information and discussion. M.K. Seguin (Department of Geology, Université Laval) helped in translating some Swedish papers. The author appreciates very much the useful comments and remarks made by three unidentified referees.