Subglacial discharge velocity and iceberg hampering: the role of bedrock topography; Axelsberg, Nyn�shamn, south central Sweden


The results of a case study of full-pipe conduit discharge in a proglacial lake of the Pleistocene provide improved knowledge for recognizing the genetic morphology of isolated glaciofluvial deposits. Most published studies investigate the elongated deposits of esker systems in either stagnant or retreating ice sheet environments only. Lack of genetic description for flow through full-pipe conduits facing localized cross-section variation of the conduit has restricted recognition of changes in the dynamics of the flow in such situations. Axelsberg gravel pit, located about 40 km south of Stockholm, shows a succession of glaciolacustrine deposits, both on the lee-side and on top of a Precambrian gneiss body that served as a local hill in the Baltic Ice Lake during deposition of the sediments under study. The succession is subdivided into five lithofacies, two of which consist of coarse-grained gravel. These two facies are separated from one another by an even coarser-grained gravel facies rich in blocks of up to 3 m. This coarsest facies occurs mainly on top of the gneiss hill. The lower gravel facies is interpreted as a remnant of the deposits formed during a phase when the turbulent current of water discharged into the lake through a subglacial conduit started to become stronger. The block-rich gravel facies formed at the phase of maximum velocity of the turbulent current. The concentration of blocks in this facies is explained by extremely strong turbulent flow conditions during a sudden, local discharge through the subglacial conduit. The presence of the gneiss hill just in front of the conduit is thought to have created the exceptionally high-velocity turbulent current capable of transporting metres-large boulders. The upper, somewhat less coarse-grained but still boulder-rich, facies was formed when the turbulent conditions had become less extreme. Boulders up to 1 m occur in this facies.
The depositional features indicate that the succession in the Axelsberg gravel pit was formed within the frontal zone of the ice sheet while it was experiencing a stagnant position. Lack of similar deposits further to the north along the deglaciation path suggests a local, short-term event. Clay-varve choronology suggests an stagnant ice sheet at the time this sudden discharge occurred. This study suggests clastic sedimentary sections are the only reliable record of short-lived events in the past.


INTRODUCTION

There has long been a need to investigate the mechanisms of short-lived subglacial conduit discharge and deposition on the proglacial areas, where the ice sheet is in an stagnant situation (Gilbert and Shaw, 1994; Shreve, 1985, R�thlisberger and Lang, 1987). Although many workers have described the genetic depositional environment of eskers, elongated glaciofluvial systems (cf. Warren and Ashley, 1994), much less attention has been paid to the isolated deposits. Reliable knowldege of sudden subglacial discharge during stagnation requires case studies considering: (1) initiation processes, (2) discharge mechanism, and (3) appropriate hydrodynamics of conduits during the discharge event. This paper implements a detailed case study to address on aspect of the stagnent ice sheet discharge event-routing of �thightened� conduit channel due to bedrock morphology. The approach used is to examine turbulent, closed conduit depositional environment, with a particular focus on the influence channel cross-section geometry and the sudden deviation of the steady, non-turbulent hydological system into turbulent flow system. The results of this study are suitable for genetic interpretation of irregularly distributed glaciofluvial deposits in the glaciated areas of the Pleistocene. Although the mechanics of turbulent flows are not the subject of this paper, it should be emphesized that the results presented here are applicable only to the full-pipe conduit environment (cf. Saunderson, 1977)
Concentration of large blocks in glaciofluvial deposits of the area was first described by Persson (1977). In the course of regional mapping of the Nyn�shamn area, southern Stockholm, 26 localities with concentration of large boulders were identified (Fig. 2x, Persson, 1997). In most localities (22) the large boulders are concentrated in moraine deposits. A few single, large blocks are reported to occur with similar mineralogy to the surrounding bedrock (cf. Persson, 1977, page 33).
Axelsberg is the only site where the fresh outcrop made it possible to study the sedimentological relationship of blocks of various mineralogy in the area. The distribution of boulders in Pleistocene deposits are ususally interpreted as result of iceberg sedimentation mechanism. In the past few years, however, increasingly more attention was paid to the erosional effects of large-scale, powerful subglacial meltwater floods beneath continental ice sheets; these provide also a possible transport mechanism for such boulders (Russel 1994, Shaw & Sharpe 1987, Brennand, 1994). Such a mechanism affects the possible genetic interpretation and the depositional models for the large, elongated, deposits of coarse glacial sediments that are generally known as eskers. Table 1 outlines some important flow properties and sedimentary phenomena associated with floods (Baker 1984).
During an investigation of glacigenic deposits in the Stockholm region in the summer of 1996, a 190 m long section of glaciofluvial deposits was studied in an abandoned, elongated, gravel pit at Axelsberg (Fig. 1). The outcrop is divided into two stratigraphic levels; the contact between them is obscured by talus scree. The section at Axelsberg is a remnant of the western wall of an exploited gravel pit (Figs 2-3). The site is situated approx. 40 km south of Stockholm at the north-eastern border of the Younger Dryas ice-marginal zone in Sweden (cf. Brunnberg 1995). This zone is bound by two ice-recession lines, represented by the Sk�vde and Billingen moraines in the western part of the country, and by scattered ice marginal deposits in the eastern part (Lundqvist 1995). The Sk�vde and Billingen moraines are interpreted as the boundaries between which the ice front was halting or had local oscillations during the Younger Dryas cold event, which is dated as 12,600-11,400 clay-varve years BP (Bj�rck et al. 1996). The highest coastline of the area is about 150 m above sea level (Svensson 1991), indicating a water depth of approx. 120 m when the site was deglaciated.

Based on the geographical position of the Axelsberg site in the clay-varve chronology (Brunnberg 1995), the deposits formed in a proglacial lacustrine environment (the Baltic Ice Lake) about 11,400 clay-varve years BP. In the lowest part of the site, the present elevation is 25 m above sea level. Persson (1983) proposed four zones of ice-marginal sediments in south-east central Sweden. The Axelsberg locality is located in his zone IV. Glacial striae in the area are directed towards the southeast and indicate the youngest ice movement (Persson 1977).
The objective of the present study is a reconstruction of the impact of a suddenly strongly increased discharge in the area at the end of the Younger Dryas. Regional studies suggest the final drainage of the Baltic Ice Lake as the specific event for the termination of the Younger Dryas (Bj�rck et al. 1996). Based on detailed mapping of raised beaches in southern Sweden, the approximate position of the ice margin during the final drainage phase of the Baltic Ice Lake was about 7 km north of the Axelsberg site (Svensson 1991: his Fig. 118). This position of the ice margin is confirmed by the suggested date for the final drainage of the ice lake inferred from clay-varve studies (cf. Brunnberg 1995). The outcrop at Axelsberg is the only available section for study of the depositional environment of glaciofluvial deposits in the Stockholm region that is dated before the formation of the Uppsala and Stockholm eskers. Detailed study of this site thus provides a unique possibility to improve our knowledge of the imprint of sudden events in the subglacial environment of the Scandinavian Ice Sheet at the end of the Younger Dryas.
In order to establish lateral and vertical variations, both within individual beds and throughout the deposit, fabric analyses were undertaken. Sediments were characterised (Figs. 2 -3) using descriptive facies types. The term `block` is used here for boulders larger than 2000 mm size.

SEDIMENT DESCRIPTION AND INTERPRETATION
The sediments are grouped into five lithofacies (Table 2). Facies 1, 2 and 5 are exposed in the lower level of the Axelsberg outcrop (see Figs. 2-3), while facies 3 and 4 are exposed in the upper level. The area is covered by about 20 cm of postglacial beach sand and gravel, which is not included in the present study. Fabric studies (Fig. 4) indicate a weak preferred orientation in the block-rich lithofacies suggesting transport towards the southeast, similar to the recognised paleocurrent of the massive gravel facies of the lower level.

Facies 1: massive gravel
This is stratigraphically the lowermost facies in the lower topographic level (Figs. 2-3, section A). The lower boundary of the facies is hidden by talus scree. The facies consists of at least 2 m of clast-supported
gravel of pebble and cobble size clasts; it is massive to poorly bedded, with a sandy matrix and with locally imbricated clasts pointing at a transport direction towards the south. The main section of facies 1 (section A) gradually fines upward. In the northern part of the lower level, two large boulders of approx. 1.5 m in diameter occur (Figs. 3 and 2b, section B, marked with arrow). The facies passes laterally towards the south into facies 5 and 2 (Fig. 3, sections A and F).

Interpretation
The coarse character of the heterogeneous, crudely stratified gravel suggests powerful currents. Cheel and Rust (1982) considered a large clast size, similar to that of facies 1, as evidence for deposition from a subglacial meltwater conduit in which the current velocities were high. The abundance of cobbles and coarse pebbles in the basal strata is common in an ice-proximal depositional environment (Eyles et al. 1987). The presence of two large boulders in a limited section of this facies (Fig. 3, section B) reflects a vertical facies change with a coarsening upward pattern towards the northern part of the outcrop. The inversely graded gravel beds are deposits of a clast-rich debris flow (cf. Whipple, 1997; Miall 1996). The lateral grading into facies 2 and 5 is characteristic of the proximal part of a subaquatic fan. Both lateral changes in the particle size and facies changes are common in such deposits. The occurrence of a gradational lateral change into the fine-grained facies 5 and the absence of finer-grained particles in facies 1 suggest transport and deposition by a debris flow (Broster & Hicock 1985).

Facies 2: glaciofluvial sand and gravel
Facies 2 is exposed in the lower level (Figs. 2-3, section B) and dips towards the southwest. It is up to 8 m thick and has a conformable contact with facies 1 at its base (Fig. 3, section F). The lower part consists of poorly developed large-scale trough cross-bedded pebbly sand and gravelly planar sand sets. The thickness of the individual beds varies between 0.1 and 0.5 m. Each sand and gravel bed is normally graded, showing laminated fines and fine sand in the top part.

Interpretation
On the basis of the sedimentary structures, facies 2 represents typically the foresets of a subaquatic fan (cf. Lunkka & Gibbard 1996). This indicates that a subaquatic fan can be formed close to the glacial margin (cf. Smith 1985).
The lateral facies change into facies 1 suggests that facies 2 was formed at a position where the energy level of the discharging current was lower than was present directly in front of the subglacial conduit.

Facies 3: 'blocky' massive gravel
Facies 3 occupies the lower part of the outcrop in the upper level (Fig. 2a). The base of this facies is visible only locally (Fig. 3) because it is obscured by talus material in most parts.
The particles range from silty sand to large blocks. A striking feature is the high concentration of blocks (Figs. 5). Sand and gravel are commonly present between the blocks and some of the cobbles are angular. The layers are sorted with a dominance of silt, clay and fine-grained sand. The thickness of the layers varies from a few mm to about 2 m. Silt and fine sand are distinctly bedded and minute climbing ripples are present. Facies 3 is overlain by the boulder-rich gravel of facies 4. The contact between these facies is gradational in some parts and erosive in others (cf. Fig. 3, sections C-D). Fabric analyses of the long axes of clasts in facies 3 show a preferred orientation due southeast (cf. Fig. 4. I). The particle size of the matrix and the presence of planar sand and pebble beds amid the blocks make this facies very much the same as the material deposited between the large boulders in facies 1 in the lower level of the outcrop.

Interpretation
The gradational character of the vertical change between facies 1 and 3 is best shown by the similarity of the particle size of the matrix in both facies, by the sedimentary structures of the material between the blocks in facies 3, and by the characteristic, relatively fine-grained matrix amid the large boulders of facies 1 (Fig. 2a, section B, marked with arrow).
The abundance of blocks in facies 3 on top of the gneiss hill is striking. Three features of the blocks (i.e. petrology, distribution, and roundness) suggest that they were not derived from a local source:
(1) the local bedrock consists of Precambrian gneiss and rocks with a different mineralogical composition are not found throughout a vast area. The nearest known dolerite outcrop is located about 300 km north of the site, at least one of the blocks of facies 3 consists of dolerite (Fig. 3, section C) which implies that this block cannot have a local origin. Transport by ice forms the only reasonable explanation for its presence at Axelsberg;
(2) the blocks are scarce in other parts of the gravel pit but are frequent in facies 3,
(3) the blocks are sub-rounded, indicating a rather long transportation path (cf. Tucker 1988).
Considering these factors, it seems reasonable to conclude that the boulders have been transported over long distances to their present site by ice. Transportation of the blocks by an iceberg is considered a reasonable explanation for the blocks visible in Fig. 5. For other blocks of facies 3, however, this transportation medium is ruled out due to the lack of any deformation in the adjacent more fine-grained material, which should be present if large blocks are dropped from a melting iceberg and fall into a relatively fine-grained, water-saturated lake bottom.
Research on subglacial erosion has led to the conclusion that depositional fields are the result of turbulent subglacial meltwater discharge over a wide area. A velocity of 2 m/s is suggested as a minimum to remove large blocks in a subglacial drainage system (Elfstr�m 1987). A turbulent subglacial flow with a velocity of 1 m/s is estimated necessary for removing boulders with a size of 1 m (cf. Williams 1983).
Facies 3 is interpreted as overlaying facies 1 on the lee-side of the southwards dipping gneiss substratum (Fig. 2a). The appearance of large boulders in the lower level (Fig. 2a, marked with arrow) is interpreted as evidence of a sudden increase in the flow velocity on the bedrock hill and its lee-side in the south. The presence of deformations (impact depressions) in the silty sand beds below two of the blocks in the upper level suggests an ice-rafted origin for these blocks. The ice-rafted material is found in the very top part of facies 3. Consequently, these blocks are, in spite of their close spatial occurrence with the blocks of facies 3, of different origin and they are -in contrast to the other units of facies 3- not covered by facies 4.

Facies 4: boulder-rich gravel
Facies 4, which is exposed in the upper level of the outcrop, is a heterogeneous, and matrix-supported facies although an isolated outcrop of clast-supported facies 4 (Fig. 6) is documented too. The gravelly facies 4 is boulder-rich with a maximum thickness of about 4 m. The lower boundary (with facies 3) is gradational in some parts and erosive in other parts (Fig. 3, section E ). The facies is erosively overlain by postglacial beach gravel (Fig. 2a). The matrix consists of sand and silt. Beds of sand and gravel with syndepositional deformation structures are present. Boulders up to 1.5 m form the largest clasts. Small packs of deformed sub-horizontal laminae of silty sand and fine to medium gravel are visible in this unit (Fig. 7). Fabric analysis revealed a weakly developed preferred orientation of pebbles suggestive of transport towards the southeast (Fig. 4, II).

Interpretation
The heterogeneous character and variable clast fabric, together with the folded character of several sand and gravel beds suggest deposition from a gravel flow. The sorted beds were deposited by sediment-laden meltwater sheet and rills (cf. Lawson 1989). Sand grains supplied by the meltwater filled up depressions, among others those between the blocks, producing massive, fine-grained lenses of silty sand. Many of the laminae in facies 4 were incorporated and deformed during plastic flowage of the previously deposited coarser sediments.
This facies possibly consists entirely of flow tills (cf. Dreimanis 1989). Most likely this facies represents the vertical `tail` of facies 3. This interpretation implies a loss in stream velocity within the glaciofluvial system from a very high velocity (facies 3) to a lower velocity (facies 4). The partly erosive lower boundary may reflect a local mobilisation of material in the subaquatic environment.

Facies 5: diamicton
A succession of about 1.5 m massive pebble and cobble forms this diamicton facies. Facies 5 occur in the southernmost part of the lower level of the outcrop (Fig. 3, section A, and Fig. 8). The boundary with the underlying facies 1 is partly gradual and partly erosive. This facies is covered directly by postglacial beach gravel.
Laterally facies 5 passes into facies 1. Several small folds of sand and fines are present, mostly around the cobbles and the scarce boulders resemble `dropped particles`. Small lenses of clast-supported cobbles are present. The clast fabric was investigated (lower section, Fig. 2a); it shows no preferred orientations (Fig. 4, III).

Interpretation
The concave structures of small folds around larger stones indicate lateral dragging during deposition or shortly thereafter. The presence of dropped boulders and their related impacts on the finer material indicates a subaquatic environment with moderate depositional energy. Facies 5 might therefore, at first sight, be interpreted as ice-rafted till from an iceberg dumped into a lake with almost stagnant water. The lateral change into facies 1, however, makes this interpretation less likely. Particularly, the lack of graded beds makes it difficult to explain traction carpet as an origin for facies 5. Alternatively, this facies may have resulted from grain flow (Bagnold 1956). Bedload moved by a fluvial system is actually supported by dispersive pressure forming a traction carpet of grains driven by the fluid shear of the moving water above. The grain shear acts in a way to move grains by the action of fluid drag forces. Submarine grain flows are common phenomena (Middleton and Wilcock 1994). Highly concentrated, non-turbulent traction carpet are already known for a long time (Dzulynski et al. 1959). The carpet is driven by the shear stress exerted by the weight of the carpet and, additionally, by the shear stress exerted by the superjacent, faster-moving, turbulent flow (Todd 1989).



DISCUSSIONS
It is clear that the succession at the Axelsberg site has several characteristic features:
(1) the limited lateral occurrence of massive cobble-sized gravel with a sandy and pebbly matrix (facies 1);
(2) the occurrence of the fine-grained facies 5 at the lee-side of the gneiss hill, which suggests a lateral facies change of a subaquatic discharge lobe indicating lateral variation in sediment concentration;
(3) a mosaic-like vertical distribution of coarse-grained facies (northern outcrop of facies 1 and facies 3
(4) a centrally positioned blocky gravel facies 3, reflecting a maximum-energy turbulent current;
(5) a gravity flow of facies 4 formed at the top of the succession;
(6) the location of a deposit mainly on top and at the lee-side of the gneiss hill at least 12 m above the base of the sedimentary surface existing simultaneously at the position of the lower level in the Axelsberg outcrop;
(7) the location of blocks directly related to the top of the gneiss hill;
(8) the boulder-rich facies 4 situated at a higher topographic level than similar deposits in the area;
(9) the mostly gradual boundaries between the glacial deposits above the glaciofluvial deposits (Table 2), suggesting a single event to be responsible for deposition of the coarsening�upward part of the succession and ruling out a temporary advance of the ice as an alternative for deposition of the boulder-rich and blocky facies 4 and 3, respectively.
The above features are believed to be diagnostic of a series of subaquatic processes that took place during a single event of a sudden, high-energy meltwater discharge in the frontal zone of a temperate glacier.
The concentration of clast-supported cobbles in facies 1 (lower level) suggests deposition in a conduit core. Hence, the appearance of large boulders in facies 1 (Fig. 2a, shown with arrow in section B) suggests a change in the mechanical properties of the current. The presence of the gneiss hill must have changed the mechanical aspects of the current in the subglacial conduit locally. This change was obviously intensified in an upward direction as documented in the upper level, where blocks of up to 3 m - transported in subglacial streams over a long distance - are present. In general, the change in sediment characteristics from facies 1 to facies 3 reflects a coarsening-upward deposition known to be formed under non-Newtonian flow conditions (cf. Fisher & Mattinson 1968). Later, a loss of stream velocity resulted in deposition of the material constituting facies 4.
Examination of the facies configuration in the lower topographic level (facies 1 and its lateral equivalents facies 2 and 5) indicates that a single mechanism of flow and sorting is inadequate to explain these variations. Lateral mass emplacement is suggested to explain the presence farther downslope of finer proximal fan material deposited after the formation of the fan deposits in front of the conduit mouth. This is consistent with Broster and Hicock (1985), who showed that transformation between current types and sediment support mechanisms of subaquatic fans can occur both simultaneously and serially.
The long-distance subaquatic transport of both fine and coarse material over a basin floor may involve two principal processes: turbidity currents of various concentration and debris flow (cf. Ghibaudo 1992). Debris-flow deposits are recognised through their poorly sorted, reverse grading, matrix-supported and roughly aligned elongate fragments (Fisher 1971).
The condition for such a flow mechanism is that the transported particles exceed 60% of the flow water volume (K. Whipple, pers. comm.). As a result, due to particle interactions, the flow develops non-Newtonian characteristics. The presence of a wide range of particle sizes in facies 3 suggests that debris flows predominated. Gravity-flow deposits in subglacial environments are known mostly as intercalations of gravity flow tills and glaciofluvial deposits of the glacial frontal zone (Menzies & Shilts 1995, Dreimannis 1989, Brodzikowski & Van Loon 1991). In the absence of data on the size and duration of the drainage in the area, it is, however, not possible to determine with any reasonable accuracy the total volume of water discharged in the drainage system.
Although some blocks can for good reasons be interpreted as dropstones, the concentration of several other large boulders in the area requires other explanations.
The presence of these blocks might be considered as either due to detachment of parts of the gneiss substratum as the result of regional uplift after rapid deglaciation of the area (cf. Persson 1977) or as a consequence of a subglacial earthquake (cf. M�rner 1993). The latter interpretation would be acceptable only if all angular blocks had a mineralogy similar to the regional bedrock, which is, however, not the case at the Axelsberg site. The mechanism by which many large blocks were carried subglacially by the meltwater is therefore not yet completely clear.
A catastrophic outburst of subglacial meltwater need not have taken place, there is no regional evidence for a true catastrophe. The blocky facies (facies 3) is of local origin only, and limited to the part of the section on top of a bedrock hill. The bedrock hill in the central part of the outcrop may have partly blocked the subglacial drainage conduit. An unbalanced fluid pressure, according to the Bernoulli equation (Middleton & Wilcock 1994), would then have increased the meltwater velocity alongside of the bedrock obstacle (Fig. 9). The coarsening-upward part of the succession (facies 1 through 3) would be the result of such a mechanism. Shortly after deposition of the blocky facies 3, and due to the quick loss of extreme velocity and the onset of a less-turbulent flow, the boulder-rich facies 4 was formed (Fig. 10). During the final stage of the development, grain flows deposited finer-grained material between the boulders (cf. Blatt et al. 1980).

CONCLUSIONS

It is likely that the bedrock topography and the variation in the discharge velocity through a subglacial conduit were the most important controlling factors with respect to the configuration of the various sedimentary units. The succession at Axelsberg contains clear evidence of three successive events in a subaquatic glacial frontal zone. It is likely that the lowermost part of the succession was formed in a subglacial conduit environment in the frontal zone of the Scandinavian Ice Sheet. The middle part, i.e. facies 3, probably formed at the phase of maximum velocity of the turbulent flow conditions when the subglacial drainage system encountered an obstruction (Fig. 11). The transition between facies 3 and 4 and the formation of facies 5 probably reflects the rapid loss of velocity and the resulting change from turbulent to Newtonian flow conditions. The succession was later covered by beach deposits of the Baltic Ice Lake.

ACKNOWLEDGEMENTS
I thank Kelin Whipple and Nick Eyles for a critical review of the manuscript. I am grateful to Lloyd Burckle, Jan Lundqvist, Bertil Ringberg and Lars Brunnberg for constructive review of an earlier version of the manuscript. Laszlo Madarasz drafted the figures. Thanks to Per-Einar Tr�ften for introducing the site. A field grant from the Earthwatch Institute (Boston) and enthusiastic field assistance of the Earthcorp volunteers of the project �The glaciers of Sweden� are also much appreciated.

REFERENCES
Bagnold, R. A. 1956: The flow of cohesionless grains in fluids. Philosophical Transactions of the Royal Society of London 249 A, 235-297.
Baker, V. R. 1984: Flood sedimentation in bedrock fluvial systems. In Koster, E. H. & Steel, R. J. (eds.): Sedimentology of gravels and conglomerates. Memoir of the Canadian Society of Petroleum Geology 10. 87-98.
Bj�rck, S., Kromer, B., Johnsen, S., Bennike, O., Hammarlund, D., Lemdahl, G., Possnert, G., Lander Rasmussen, T., Wohlfarth, B., Hammer, C.L, & Spurk, M. 1996: Synchronized terrestrial atmospheric deglacial records around the North Atlantic. Science 274, 1155-1160.
Blatt, H., Middleton, G. & Murray, R. 1980: Origin of sedimentary rocks. Prentice-Hall, Inc. 782 pp.
Brennand, T. A. 1994: Macroforms, large bedforms and rhythmic sedimentary sequences in subglacial eskers, south-central Ontario: implications for esker genesis and meltwater regime. Sedimentary Geology 91, 9-55.
Brodzikowski, K. & Van Loon, A. J. 1991: Glacigenic sediments. Elsevier, Amsterdam. 674 pp.
Broster, B. E. & Hicock, S. 1985: Multiple flow and support mechanisms and the development of inverse grading in a subaquatic glacigenic debris flow. Sedimentology 32, 645-657.
Brunnberg, L. 1995: Clay-varve chronology and deglaciation during the Younger Dryas and Preboreal in the easternmost part of the Middle Swedish Ice Marginal Zone. Quaternaria, A:2, 94 pp. Dept. of Quaternary Research, Stockholm University.
Cheel, R. J. & Rust, B. R. 1982: Coarse-grained facies of glaciomarine deposits near Ottawa, Canada. In Davidson-Arnott, R. N., Nickling, W. and Fahey, B. D. (eds.): Research in glacial, glaciofluvial and glaciolacustrine systems. Geo Books, Norwich. 279-295.
Dreimanis, A. 1989: Tills: their genetic terminology and classification. In Goldthwait, R. P. and Matsch, C. L. (eds.): Genetic classification of glacigenic deposits. Balkema, 17-83.
Dzulyn`ski, S., Ksiazkiewicz, M., & Kuenen, Ph. H. 1959: Turbidites in flysch of the Polish Carpathian Mountains. Geological Society of America Bulletin 70, p. 1089- 1118.
Elfstr�m, �. 1987: Large boulder deposits and catastrophic floods. Geografiska Annaler 69, 101-121.
Eriksson, K.-G. 1960: Studier �ver Stockholms�sen vid Halmsj�n. Geologiska F�reningens i Stockholm F�rhandlingar 82, 43-125.
Eyles, N., Clark, B.M., & Clague , J. J. 1987: Coarse-grained sediment gravity flow facies in a large supraglacial lake. Sedimentology 34, 193-216.
Fisher, R. V. 1971: Features of coarse-grained, high-concentration fluids and their deposits. Journal of Sedimentary Petrology 41, 916-927.
Fisher, R. V. & Mattinson, M. 1968: Wheeler gorge turbidite-conglomerate series, California; Inverse grading. Journal of Sedimentary Petrology 38, 1013-1023.
Gilbert R., & Shaw J. (1994): Inferred subglacial meltwater origin of lakes on the southern border of the Canadian Shield. Canadian Journal of Earth Sciences 31, 1630-1637.
Ghibaudo, G. 1992: Subaqueous sediment gravity flow deposits: Practical criteria for their field description and classification. Sedimentology 39, 423-454.
Lawson, D. E. 1989: Glacigenic resedimentation: classification concepts and application to mass-movement processes and deposits. In Goldthwait, R. P. & Matsch, C. L. (eds.): Genetic classification of glacigenic deposits, Balkema, 147-169.
Lundqvist, J. 1995: The Younger Dryas Ice Marginal Zone in Sweden. Quaternary International 27, 171-176.
Lunkka, J. P. & Gibbard, Ph. 1996: Ice-marginal sedimentation and its implications for ice-lobe deglaciation patterns in the Baltic region: Pohjankangas, western Finland. Journal of Quaternary Science 11, 377-388.
Menzies, J. & Shilts, W. W. 1995: Subglacial environments. In Menzies, J. (ed.): Past glacial environments, sediments, forms and techniques. Butterworth and Heinemann, 15-136.
Miall, A. D. 1996: The geology of fluvial deposits: sedimentary facies, basin analysis, and petroleum geology. Springer-Verlag, Berlin. 582 pp.
Middleton, G. V. Wilcock, P. 1994: Mechanics in the earth and environmental sciences. Cambridge University Press. 459 pp.
M�ller, H. & St�lh�s, G. 1969: Beskrivning till geologiska kartbladet Stockholm SO. Sveriges Geologiska Unders�kning Ae 2. 101 pp.
M�rner, N.-A. 1993: Boulder trail from a subglacial earthquake, �sp�, Sweden. In Stewart, I., Vita-Finzi, C. & Owen, L. (eds.): Neotectonics and active faulting. Z. Geomorph. N.F. Supplementband 94, 159-166.
Persson, Ch. 1977: Beskrivning till jordartskartorna Nyn�shamn NV/SV och Nyn�shamn NO/SO, Sveriges Geologiska Unders�kning, Serie Ae. 31-32, 83 pp.
Persson, Ch. 1983: Glacial deposits and the Central Swedish end moraine zone in eastern Sweden. In Ehlers, J. (ed.): Glacial deposits in north-west Europe. Balkema, 131-139.
R�thlisberger, H. & Lang, H. 1987: Glacial hydrology. In Gurnell, A. M & Clark, M. J.(eds.) Glacio-fluvial sediment transfer, an Alpine perspective. 207-284. John Wiley, Chichester.
Russell, A. J. 1994: Subglacial j�kulhlaup deposition, Jotunheimen, Norway. Sedimentary Geology 91, 131-144.
Saunderson, H. C. 1977: The sliding bed facies in esker sands and gravels: a criterion for full-pipe (tunnel) flow? Sedimentology 24, 623-638.
Shaw, J., & Sharpe, D. R. 1987: Drumlin formation by subglacial meltwater erosion. Canadian Journal of Earth Sciences 24, 2316-2322.
Shreve, R. L. 1985: Esker characteristics in terms of glacier physics, Katahdin esker system, Maine. Geological Society of America Bulletin 96, 639-646.
Smith, N. D. 1985: Proglacial fluvial environment. In Ashley, G. M., Shaw, J. & Smith, N. D. (eds.): Glacial sedimentary environments. SEPM short course No. 16, 85- 134.
Svensson, N-O. 1991: Late Weichselian and Early Holocene shore displacement in the central Baltic Sea. Quaternary International 9, 7-26.
Todd, S. P. 1989: Stream-driven, high-density gravelly traction carpets: possible deposits in the Trabeg Conglomerate Formation, SW Ireland and some theoretical considerations of their origin. Sedimentology 36, 513-530.
Tucker, M. 1988: Techniques in sedimentology. Blackwell Scientific Publications, Oxford. 394 pp.
Warren, W. P. & Ashley, G. M. 1994: Origins of the ice-contact stratified ridges (eskers) of Ireland. Journal of Sedimentary Research A64, 433-449.
Whipple, K. X. (1997) Open-channel flow of Bingham fluids: applications in debris-flow Research. The Journal of Geology. 105, 243-262.
Williams, G. P. 1983: Paleohydrological methods and some examples from Swedish fluvial environments. 1. Cobble and boulder deposits. Geografiska Annaler 65, 227-243.