ICE SCOURING AS A GEOLOGIC AGENT: PLEISTOCENE EXAMPLES FROM SCARBOROUGH BLUFFS AND A NUMERICAL MODEL

David Jason Eden

A thesis submitted in conformity with the requirements for the degree of Master of Science Graduate Department of Geology

University of Toronto

O Copyright by David J. Eden (2000)

National Library Bibliothèque nationale du Canada

Acquisitions and Acquisitions et Bibliographie Services services bibliographiques

395 Weliingtm Street 395. NC WdlingtOri OnawaON K I A W Oriawa ON K1A ôN4 canada CaMda

The author bas granted a non- exclusive licence aiiowing the National Library of Canada to reproduce, loan, distribute or seil copies of this thesis in microfom, paper or electronic formats.

L'auteur a accordé une licence non exclusive permettant a la Bibliothèque nationale du Canada de reproduire, prêter, distribuer ou vendre des copies de cette thèse sous la forme de microfiche/nlm, de reproduction sur papier ou sur format électronique.

The author r e W o'vnership of the L'auteur consewe la propriété du copyright in this thesis. Neither the droit d'auteur qui protège cette thèse. thesis nor substantial extracts fiom it Ni la thèse ni des extraits substantiels may be printed or otherwise de celle-ci ne doivent être imprimés reproduced without the author's ou autrement reproduits sans son permission. autorisation.

Ice scouring as a geologic agent: Pleistocene examples from Scarborough Bluffs and a numerical model

David Jason Eden Degree of Master of Science, 2000

Graduate Department of Ceology, University of Toronto

In southem Ontario, extensive outcrops of glaciolacustrine muds and shoreface sands of Late

Pleistocene age are exposed near Toronto, and were deformed by floating ice masses in an

ancestral ice-damrned Lake Ontario. Outcrop mapping and a GPR survey identifi ice scour

structures up to 4 m, 10 m wide and at least several hundred metres long and underlain by sub-

scour sediment deformations up to 5 m below the scour. The scours also have ndged margins, as

observed in modem scours.

Paleoenvironrnental information can be derived fiom ancient ice-scoured strata using an FEM

numencal model of sub- ice scour deformation. For the largest ice scour exposed at Scarborough,

ice mass is 0.04 Mt, keei ciraft is 10 to 30 m and scouring force is 5 x 103 kN, consistent with a

small ice berg or pressure-ndged lake ice. Estimates of ice mass dimensions are consistent with

independent sedimentological reconstructions of paleobathymeûy.

I am greatly indebted to Nick Eytes for helpfiil discussion and supervision, and for introducing

me to the Cudia Park site and the world of glacial sedimentology. 1 thank Golder Associates for

the use of geotechnical modelling software, Kyle Hodder of the University of Toronto for

assistance with the ground penetrating radar swey , and Mike Doughty of the University of

Toronto for assistance with iliustrations. 1 am most gratefiil for ongoing encouragement fiorn

my wife and parents.

TABLE OF CONTENTS

ABSTRACT ... .........e.........m.u.m.m.mHI..- -..w.w.nw~.~.m.~..~n~.onn~o~w~~.o~.~~~~m~~~~~e.~n.m~.~n.~ ll

ACKNOWLEDGEMENTS ........................ ......... .- ......... - ..................... ...HH.....-w..n................ .. m

TABLE OF CONTENTS ... . .~ . .~ . . .~ .m.n~~.~~~Mt .~ . .~~~~.~~~~~. .~~ . .~ .~~~wH~~.~ .~~oH~. .~~ .~~w~.~~. .~~m~~~. .~ . . . . .... IV

................................................................................................................................. LIST OF FIGURES VI

.................................................................................. LIST OF TABLES .................... ........ ... .-.....- VI

INTRODUCTION ..... ........... ..... .. ....... ........ .... ...........*.......*..o............*...m"........................................... ............... I

Stratigraphy of study area ....................................................................................................... 4 2.1. I Scarborough-Sunnybrook- Thomd~fle stratipaphic succession ................................................................ 5

................................................................................................................... Ice scour structure 9 ................................................................................................... 3.1. 1 Character and ongin of infill sedimen~ I f

3.1.2 Formationoficescour ........................................................................................................................ I 2

PALEOENVIRONMENTAL ANALYSIS OF ICE SCOUR BY NUMERICAL MODELLING ................. 14

..................................................................................... Geotechnical properties of substrate 14 ............................................................................................................................. 4.1. f Slope stability analysis 16 . . ............................................................................... Fimte element model ......................... .. 17

......................................................................................................................... Model results 19 Size and draft of ice mass ..................................................................................................... 20 Ice scoured mud and ice-keel turbated facies ...................................................................... 22

RELATED ICESCOUR STRUCTURES AND FACIES OF THE SUNNY8ROOK .................................. 25

Ice-rafted debris within the Sunnybrook ............................................................................... 25 ............................................................................................................ Striated mud surfaces 27

Pebbly mud ice-keel turbated facies ..................................................................................... 29 ........................................................................... Shear zones at the base of the Sunnybrook 31

DISCUSSION ................... .. ................. ............ .......................................................................................... 33

LIST OF TABLES

Table 1: Pre.Pleistocene. Pleistocene and Modem ice scours ........................................................... 47

Table II: Pre.Pleistocene. Pleistocene and Modem ice-keel turbates ............................................. 50

Table III: Pre.Pleistocene. Pleistocene and Modem soft-sediment stnated surfaces . Note scarcity of Pleistocene examples ........................................................................................................ 51

Table N: Values of Cu and E for Sunnybrook and Scarborough sand used in parametric finite .................................................................................................................... element analysis 52

LIST OF FIGURES

Figure 1 : Location of Scarborough Bluffs and exposures of ice-scour structures and facies mentioned in text ................................................................................................................... 53

Figure 2: Stratigraphy of Scarborough Bluffs. ................................................................................. 53

Figure 3: A, B: Typical massive clast-poor mud facies of the Sunnybrmk exposed at Cudia Park. ........... C, D, E: Clusters of ice-rafted debns within Sunnybrook mud. F: Ice-rafted clasts. 54

Figure 4: Basal contacts of the Sunnybrook mud where it rests on underlying sands of the Scarborough Formation. ...................... ... ............ .., . 55

Figure 5: Prominent ice scour exposed on surface of Sunnybrook mud at Cudia Park on Scarborough Bluffs. ........................................................................................................ 56

Figure 6: A "Western" outcrop of ice scour stucture shown in Figure 5 exposed in Cudia Park ...... Ravine. B: Cross-section througb large scour exposed at location Rî on Rouge River. 57

Figure 7: A: Sketch of ice Sour depicted in Figure 5 showing ùifill strata. B: Nature of sub-scour deformation. .......................................................................................................................... 5 8

Figure 8: A: Profile and B: facies interpretation of ground penetrating radar survey conducted at Cudia Park adjacent to outcrop of ice-scour shown in Figures 5 and 6 ................................ 60

Figure 9: Ice-scoured surface on upper S m y b r w k exposed at Sylvan Park .................................. 61

Figure 10: Ice-scoured topography on uppennost Sunnybrook mu&. A, B, C: Sylvan Park. D, E, F: Rouge Valley. ................................................................................................................... 62

Figure 11 : A: Modern occurrences of ice-scoured seabed topography. Br Cross-section through modem ice scour. C: Floor of former Glaciat Lake Agassiz in Manitoba ............................ 63

Figure 12: A: Minimum slip surface used in numerical slope stability analysis. B: Finite element mesh and boundaxy conditions for numerical mode1 of ice scour. C: -Mode1 results showing contours of shear stress (see text for details). D: Muence of Sunnybrwk properties on calculated displacement. E: Influence of Scarborough sand properties on calculated displacement. ....................................................................................................... 64

Figure 1 3: Mode1 for ice-scoured topography seen on upper surface of Sunnybrook. .................... .67

Figure 14: Soft-sediment striations cut across mud at Highland Creek ............................................ 68

LIST OF APPENDICES

Appendix A: SIGMlVW Finite Element Numerical Model. .... .. .... . .... . . .. ...... .. ....-... .. . . . . . . . . . . . . 69

vii

1 INTRODUCTION

Large areas of high-latitude and polar oceans and seas, including the Beaufort Sea, the Grand

Banks of Newfoundland, and the Greenland and Norwegian shelves, are routinely scoured by

icebergs (e.g., King, 1976; Vorren et al., 1983; Barnes and Lien, 1988; Marienfeld, 1992; Shipp

et al., 1999). Scouts are typically 10 to 100 rn wide (exceptionally 1200 m), 1 to 2 m deep

(exccptionally 30 rn deep) and usually occur in water depths greater than 20 m, normally ranging

fkom 100 to 550 m (e-g., Luternauer and Murray, 1983; Bames et al., 1984; Lewis and

Woodworth-Lynas, 1990; Woodworth-Lynas, 1992; Woodworth-Lynas and Dowdeswell, 1994;

Davies et al., 1997 and refs therein; Table 1). In addition, the effects of scouring by seasonal

(and multi-yea.) pack ice and associated pressure ndges in inshore waters, lakes and estuaries are

also well documented (e-g., Tyrell, 1892; Koschenkin, 1958; Weber, 1958; Mollard, 1973;

Dionne, 1972, 1974, 1988, 1989; Dredge, 1982; Barnes et al., 1984; Grass, 1984, 1986;

Woodworth-Lynas et al., 1986). Hamelin (1959) introduced the term 'glaciel' to refer to al1

processes, sediments and features related to the action of floating ice (regardless of type and

environment; Dionne, 1989). Vorren et al. (1 983) introduced the term 'iceberg turbate' to refer to

deformation structures and sediments produced by ice scour; Woodworth-Lynas and Guigne

(1 990) used the non-specific term 'ice-keel turbate' to refer to ploughed sedirnents regardless of

ice type (Table II).

Reports of ice-infested polar oceans arising nom Arctic exploration and the discovery of

Antarctica in the early nineteenth cenhiry were a great stimulus to the new discipline of glacial

geology. Ice- keel turbation and ice-rafting was one of the earliest depositional mechanisms to be

invoked by nineteenth century geologists to explain poorly-sorted glacial deposits on land in the

mid-latitudes. Such strata were wrrespondingly calleci 'drift' deposits (Lyell, 183 1; Danvin,

1 839, 1 8 55; Geikie, 1 896). However, as argueci by Dionne (1 982) this model was controversial,

arising mostly from the uncntical application of the model to deposits of undoubted glacial

origin (tills), and interest in the geologic record of drifting ice consequently stagnated. In the

present day, despite the common occurrence of ice scouring processes in modem envimmnents,

few examples of ancient structures have been described fiom Pleistocene or pre-Pleistocene

outcrops (Table 0. Relict ice scours and ice-turbated sediments are reported from late

Pleistocene glaciolacusFnne sediments in Scotland (Thomas and Connell, 1985), in Ontario

(Eyles and Clark, 1988, see below), in Manitoba on the floor of the former Glacial Lake Agassiz

(Woodworth-Lynas and Guigné, 1990), from raised marine sediments of Arctic Canada

(Woodworth-Lynas et al., l986), Scotland (Geikie, 1896) and Norway (Longva and Bakkejord,

1990; Tables 1, II). Pre-Pleistocene outcrops of ice scours have been recognised h m Early

Permian glacially-intluenced marine strata of the southem Gondwana continents such as the

Dwyka of South Afnca (Savage, 1972), fkom the Itararé Group of Brazil (Rocha-Campos et al.,

1994) and from glacially-infhenced shallow marine strata of the southem Sydney Basin of

Australia (Eyles et al., 1997; 1998). Surprisingly, only a single large outcrop of pre-Pleistocene

ice-keel turbated sediment has been described to date (Eyles et al., 1998; Tables 1, II) some 170

years after the process was first recognised by geologists.

1.1 Purpose of this study and signifierince

One purpose of this study is to identiQ the known occurrences of ice scour structures and ice

turbated sediment in modem, Pleistocene and pre-Pleistocene sedimentary successions (Tables 1,

II). A theme that emerges is that despite ice scour being a comrnon modem &y process

affecting huge areas of high latitude seas, lakes and rivers, descriptions of ancient deformation

stnictures and associated sediments are surprisingly few in number. The scarcity of pre-

Pleistocene examples reported in the literature has been attributed fiindamentally to lack of

knowledge of ice scour processes and products, combined with a lack of published criteria for

recognition (Woodworth-Lynas and Dowdeswell, 1994).

Consequently, a second purpose of this study is to present outcrop data f+om a cold-climate late

Pleistocene (about 45 ka old) glaciolacustrine succession near Toronto, Canada, that was subject

to ice scour processes. A sirnplified finite element mode1 of sub-ice scour deformation was

developed to determine paleoenvironmental information, such as ice mass size and

paleobathymetry. Other smaller-scale structures attributed to floating ice, such as 'soft-seàiment

striations', 'dump deposits' and ice-keel turbated facies are also identified. It is hoped that the

data presented here will stimulate m e r identification of ice scouring in the geologic record.

2 BACKGROUND

In the Great Lakes region of mid-continent North Amenca, extensive ice-damrned lakes, much

Iarger than their modem descendants, were ponded against the advancing margin of the

Laurentide Ice Sheet during the 1 s t (Wisconsin) glaciation (Karrow and Callan, 1985). At

Toronto, Canada, a 60 m thick exposed succession of glaciolacustrine sediments was deposited

in a deepened ancestral Lake Ontario during the eariiest phase of the glaciation, sometime

around 45 ka (Berger and Eyles, 1994). These strata occur stratigraphically below younger late

Wisconsin tills that record maximum ice sheet expansion after 23 ka (Boyce et al., 1995; Boyce

and Eyles, 2000). Glaciolacustrine deposits are exposed along the rapidly-eroding Lake Ontario

shoreline at Scarborough Bluffs (Figs. 1,2). This study is concemed with the early Wisconsin

Scarborough Formation deltaic sands, an overlying glaciolacustrine pebbly mud ('Sunnybrook')

and shoreface sands (Thorncliffe Formation; Fig. 2).

2.1 Stratigraphy of study area

One problem in referring to the strata exposed at Scarborough Bluffs has been the past usage of

genetic terms to formally descnbe stratigraphic units without understanding of their

sedirnentology and origin, together with stratigraphic subdivision based on a few selected

outcrops. Thus the mud deposit with rare clasts that rests on the Scarborough Formation has been

variably described as 'Sunnybrook Till' (Terasmae, 1960; Karrow, 1967) in recognition of its

locally 'tiil-like' character and presumed subglacial ongin, and later 'Sunnybrook Drift' (Kar~ow,

1969) in recognition of the common presence of stratified and laminated silty clay facies. Eyles

and Eyles (1983) used the informa1 non-genetic term 'Sunnybrook diamict'. Most recently,

Hicock and Dreimanis (1989) sub-divided the Sunnybrook into a lowermost Sylvan Park

Member, Sunnypoint Member, and Bloor Member based on study of a few selected outcrops,

despite dernonstrateci sedimentological variability along outcrop (see Eyles et al., 1984; see

below) and recognised the Sunnypoint Member as a till. These varying interpretations are

discussed later; to postpone the issue of formal stratigraphie nomenclature until a later par? of

this thesis, the tem 'Sunnybrook' is used below.

2.1.1 Scarborou~h-Sunnvbrook-Thorncliffe stratinra~hic succession

The Scarborough Formation is a thick (50 m) coarsening-upward (mud to sand) deltaic

succession that can be traced in the subsurface over a large part of south-central Ontario (Eyles,

1987). This succession records southerly progradaticn of a large fluvially-dominated delta fed by

a sandy braided river draining the upper Great Lakes (Kelly and Martini, 1986; Martini and

Brookfield, 1995). Palynological investigations show that during Scarborough Formation time

the climate was arctic-like with a boreal forest (Terasmae, 1960). The lower part of the deltaic

succession consists of prodelta laminated silty clays that are overlain by channelised fluvial

sands of the delta plain showing large-scale trough and planar cross-stratification. Delta plain

facies occur about 40 m above the modem level of Lake Ontario, indicating a high water Ievel

resulting fkom the Laurentide Ice Sheet blocking the St. Lawrence Valley downstream (Karrow,

1967). A subsequent phase of relatively low water tevels is indicated by paleochannels cut into

the delta top such as the large channel at Bluffers Park some 50 m deep and 1500 m wide vig. 2)

but the ongin of such c h a ~ e l s is not well known. A subsequent m e r rise in lake level, when

the Laurentide Ice Sheet margin entered the basin, flooded the delta top and deposited the muds

(Sunnybrook) that blanket the underlying delta. The Sunnybrook is predominantly a massive and

poorly-stratified mud (typically 40% clay, 45% silt, 15% sand) containing few clasts (Fig. 3).

In a senes of papers, N. and C. H. Eyles and CO-workers described the sedimentoiogy,

paleontology, paleomagnetic and isotope stratigraphy of the Sunnybrook (e-g., Eyles et al., 1983,

19834 1984; Eyles et al., 1987, Westgate et al., 1987; Eyles and Westgate, 1987; Rutka and

Eyles, 1989; Schwartz and Eyles, 1991 ; Eyles and Williams, 1992) and presented a

glaciolacustrine origin where suspended glacial meltwater plumes and ice-rafting ('min-outg;

Eyles and Eyles, 1983) deposited mud in a deep icesontact glacial lake. Much of south-central

Ontario is floored by soft Ordovician shales and the Sunnybrook reflects glacial scour and the

trapping of large volumes of glacially-produced rnud in a deep ice-contact glacial lake. The

secondary effects of sediment gravity flows (debris flows, slumping) and traction current

activity were also identified. Facies are dominzted by massive and crudely-stratified mud that

locally contains suficient clasts to be termed 'diarnictg, together with rhythmically-laminated

'varve-like' silty-clays including silt-clast breccias and clots of ice-rafted debris. Massive diamict

and mud facies are thickest (> 1 5 m) within the large paleovalley on the underlying Scarborough

Fomiation delta at Bluffers Park (Fig. 2), where they p a s up transitionally into thick

rhythmically-laminated prodelta silty clays and storm-deposited sands of the Thorncliffe

Formation. Outside the channel, where the Sunnybrook rests on the low relief surface of the

underlying delta plain (Fig. 2). its thickness is reduced (< 10 m) and it has a simple blanket-like

planar tabular geometry that can be traced regionally in the subsurface on downhole geophysical

logs.

The Sunnybrook shows two types of relationship with underlying sand strata: conformable and

interbedded. The former occurs where massive mud rests confonnably on an angular

unconformity cut abruptly across underlying undisturbed sands Vig. 4A). Elsewhere, thin (< 3

rn) wave ripple-laminated and storm-deposited cross-strati fied sands thin upwards and are

interbedded with lowermost muds of the Sunnybrook giving rise to a transitional contact (Fig.

4B). Locally, both types of contact have been defomed by shearing and faulting (Figs. Figs. 4C,

D). Such zones occur sporadically at the base of the Sunnybrook and are thickest (up to 1 rn) on

the flanks of the large paleovalley at Bluffers Park (Randall, 1995). Defornation structures

exposed east of the paleovalley, dong the central part of the Bluffs, were the subject of a detailed

study conducted by Hicock and Dreimanis (1989; 1992). They confirmed that the upper part of

the Sunnybrook jwhich they formally recognised as the Bloor Member') was gla~iolac~trine

and that the Sunnybrook and Scarborough had a transitional, interbedded contact over much of

its outcrop recording deposition in a deepening glacial lake ('Sylvan Park member'). Io addition,

they argued that the abruptly codormable contact, together with the small-scale faulting and

shearing that sporadically affects the contact was the product of subglacial erosion and

deformation. They cited additional clast fabnc data to support a subglacial ongin below a west-

flowing ice sheet. Despite the lack of stratigraphic continuity and restncted number of outcrops

studied, they formally recognised a 'Sunny Point member' (Hicock and Dreimanis, 1992, p. 147)

at the base of the Sunnybrook and argued that it formed as a 'deformation till' when a 'partially-

grounded glacier' overrode and tectonized glaciolacustrine sediment. Below, structures and

deformed sediments from the upper and lower contacts of the Sunnybrook are describeci and

interpreted as the result of ice-scour. The discussion starts with a large structure exposed at

Cudia Park (Figs. 1,2, 5 , 7).

3 ICE SCOUR STRUCTURES AM) RELATED ICE-KEEL TURBATED FACIES

Eyles and Clark (1988) descnbed a large trough-like structure, attributed to ice-scour, nom

shoreface sands of the Thomcliffe Formation exposed in the sidewalls of Bellamy Ravine (Fig.

1) and suggested (p. 805) that other soft-sediment deformations could be the result of ice

turbation. Unfominately, these outcrops have since been covered by slope stabilisation measures.

A newly-discovered ice scour structure described below has been exposed by cliff falls at Cudia

Park between Meadowcliffe and Cudia Park ravines (Figs. 5,6,7, 8). Other related structures

have been recently identified at Sylvan Park, dong Highland Creek and in the Rouge Valley

(Figs. 1, 9, 10).

3.1 Ice scour structure

At Cudia Park, a trough-like shiicture, about 10 m wide and 4 m deep, is exposed in the cliff face

(Figs. 5 ,6 ). The v-shaped structure is cut into the upper surface of the Smybrook and filled

with sands of the Thomc5ffe Formation (Fig. 7). The Scarborough sands have been deformed

directly below the scour where the base of the Sunnybrook is displaced downwards. On the

southwest margin of the outcrop the Sunnybrook contains horizontal interbeds of poorly-

stratified diamict and rhythmicallplaminated 'varve-like' silty clay that are truncated by the

sidewall of the trough. Sidewalls of the tmugh are smooth and at a high angle (- 60"); uppermost

sidewalls &tend up to 75 cm above the stratigraphie top of the surrounding Sunnybrook forming

raised rims or 'berms' (Figs. 5,6, 7).

The interbedded contact of the Scarborough Formation and overlying Sumybrook has been

sheared and displaced downwards by about 1 m under the structure. Below this zone, about 2 m

of npple cross-larninated and horizontally-laminated sands and interbedded laminated mud

(facies Sr, Sh, FI respectively; Fig. 7) have been extensively deformed @art of the so-called

'Sunnypoint Member' of Hicock and Dreimanis, 1992; see above).

Excavation and repeated obsewation of the area after large cliff falls indicate that the structure is

the outcrop expression of a buried scour oriented approximately west-est. A ground penetrating

radar survey conducted along the cliff top (Fig. 8) confirmeci this orientation. The data shown in

Figure 8 are fiom a 20 m long traverse, parallel to and 7 m firom the cliff face.

The application of ground penetrating radar (GPR) to identiQ the geometry of shallow buried

structures in Pleistocene sediments and other rocks has been described in several recent papers

(e-g., Beres et al., 1999; Vandenberghe and van Overmeeren, 1999; DagaIIier et al., 2000). GPR

uses electromagnetic pulses at radio fiequencies that are reflected back off interfaces in the

substrate. A PulseEKKO IV unit was used, with 1000 V pulsers and 50 MHz antennae

permitting a vertical stratigraphie resolution of about 1.5 m. The antennae were used at a

separation of 2 m, with measurements taken every OSm, which is the manufacturer's

recommended configuration for 50 MHz (Sensors & Software, 1996).

Results of the s w e y indicate the subsurface continuation of the exposed trough, westward to a

poorly exposed outcrop of a sand-filled trough with lateral berms in the sidewalls, of the adjacent

Cudia Park Ravine (Figs. 6A, 8).

3.1.1 Character and oriein of iofill sdimen@

The trough infi 11 is dominated by storm-deposited fine sand showing well-developed hummocky

and swaley cross-stratification (facies Shc, Ssc; Fig. 7) of the Thomcliffe Formation; detailed

facies descriptions are given in Eyles and Clark (1988). A lower unit of and , about 2 m thick, is

deformed by numerous 'pillows' and narrow (< 10 cm) vertical 'pipes' (facies Sd; Fig. 7)

recording post-depositional liquefaction and upward escape of pore waters. In tum, defonned

fine sands are truncated by a thin (25 cm) bed of crudely-stratified gravel (Gs; Fig. 7) showing

poorly-developed megaripples with wave lengths up to 30 cm. Both the gravels and the lower

part of the overlying sand contain rounded balls, up to tennis bal1 size, composed of Sunnybrook

silty clay which are 'armoured' with a veneer of sand and small gravel clasts.

Hurnrnocky and swaley cross-stratified sands of the Thomcliffe Formation record the passage cf

storm waves over the lake floor and the reworking of fine sand falling through the water column

by strong oscillatory cunents (e.g., Walker and Plint, 1992). These facies identie a lower

shoreface setting sufficiently deep to prevent reworking by waves during fair weather. Direct

estimates of water depths and s tom wave heights during the filling of the scour can be made

fiom the form of wave-gerierated ripples (e.g., Clifton and Dingler, 1984; Diem, 1 985). Such

estimates provide usehl lirniting water depths during formation of the scour and thus, the size of

the ice mass involved (see below). Estimated water depths for the Thorncliffe Formation

shoreface are no more than 20 m deep, with wave heights of about 5 rn with periods of about 5

seconds (see Eyles and Clark 1986; 1988). These wave heights exceed the maximum height of

waves produced by extreme storms on present day Lake Ontario reflecting the much greater size

and fetch of the ice-dammed lake. S ymmetncal gravel ripples are the coarser-pineci facies

equivalent of hummocky and swaley cross-stratification (Leckie and Duke, 1984; Walker and

Plint, 1992). Eyles and Clark (1986) reported the presence of large ice-rafted boulders within

storm deposits of the Thorncliffe Formation; the gravel ripples at Cudia Park are probably lag

concentrations of clasts either reworked fiom the underlying Sunnybrook or introduced as ice-

rafted debris (RD). Associated armoured clasts fonn when a s t i ch cohesive clayey sediment is

rolled around by currents on a sandy substrate (Allen, 1984).

Uppermost Thorncliffe Formation sands at Cudia Park are truncated by a major unconforrnity at

the base of overlying Iroquois glaciolacustrine deposits (Fig. 7). Iroquois deposits accurnulated

in a rnuch younger late glaciai lake (Glacial Lake Iroquois; - 12,500 bp) with a surface elevation

about 60 m above the modem lake. An abandoned iroquois cliffiine lies immediately idand of

the Cudia Park site.

3.1.2 Formation of ice scour

The scour forms an elongate charnel-like feature, onentated west-east with a subsurface extent

of at least 100 m (Fig. 8). Diagnostic 'contextual' evidence of its origin cornes fiom associated

sediments, the presence of deformation structures directly below the feature and the occurrence

of raised berms either side of the depression. Given that the structure is cut into the upper surface

of deep water glaciolacustrine muds ('Bloor Member' of Hicock and Dreimanis, 1992; Schwartz

and Eyles, 1991) and is filled by lower shoreface sediments, a subaerial fluvial origin can be

mled out. Small-scale erosional scours can be produced subaqueously by storms (e.g., 'gutter

casts'; Aigner, 1985), but the large size and depth of the structure, cohesive geotechnical

properties of the Sunnybrook, simple v-shape and smooth side slopes, together wîth the presence

of raised laterd bems and defonned substrate below the structure are not consistent with

excavation by wave-induced currents. Similarly, the elongate nature of the trough, cohensive

substrate and the presence of sidewall bems rules out an ongin as a 'stmdel' crater excavated in

non-cohesive bed matenals in shallow water depths when surface meltwater drains rapidly

though seasonal ice covers (for example, as seen in the Alaskan Beaufort Sea; Reimnitz, 1997).

Downward displacement of the base of the Sunnybrook, together with the associated

deformation of underlying Scarborough Formation sands, indicate that lake floor sedirnents were

dispIaced immediately below the axis of the depression. Such deformation, combined with the

paleoenvironrnental context provided by shoreface infiil sediments and the presence of lateral

berms, are consistent with an ice scour origin produced by the dragging of an ice keel across the

lake floor (Woodworth-Lynas, 1996). The similarity in cross-sectional form and dimensions

between modem iceberg scours (e.g., King and Gillespie, 1986) and that at Cudia Park is striking

(Fig. 1 1) and in turn, is similar to the geometry of an ice scour previously reported along the

BIuffs (Eyles and Clark, 1988). The trend of the scour at Cudia Park as confirmed by radar

survey (west-east) is parallel to the inferred regional shoreline of the ice-darnmed lake, which is

consistent with modem Lake Erie (Grass, 1984) and other seasonally ice-stressed environments

where floating ice masses move into shallow water and are subsequently driven by wind, waves

or currents along the coast parallel to the bathymetric contours (e.g., Barnes et al., 1984, 1988).

4 PALEOENWRONMENTAL ANALYSIS OF ICE SCOUR BY NUMERICAL

MODELLING

Given the availability of good outcrop data at Scarborough with regard to the geometry of the ice

scour and the depth of sub-scour defonnation, a numencal mode1 was developed to detennine

paleoenvironmental idonnation such as the size and mass of the scouring ice. Numerical models

of modem iceberg scours have already been presented by several workers (e-g., Yang et al.,

1996; Clark and Landva, 1988; Lien, 1986; Chari, 1986) and include different styles of substrate

defomation such as plastic flow and displacement of rigid blocks along slip surfaces. Al1 these

models are constrained by a lack of field data regarding the subsurface geology of scours and

depth of sub-scour deformation; information that is readily available in this study. The first step

is to infer appropriate geotechnical properties for the Sunnybrook and Scarborough Sands at the

time of scouring. This can be done by using data from analoggus sediments whose properties are

well known and also by conducting a slope stability analysis of the scour side slopes.

4.1 Geotechnical properties of substrate

In the absence of test data fiom the Sunnybrook at the time it was scoured, appropriate

geotechnical properties used in this anaiysis are estimates based on studies of analogous

geological materiah. While there is disagreement as to the origin of the lower part of the

Sumybrook (see below) there is agreement that the upper Sunnybrook was deposited as a

norrnally-consolidated glaciolacustrine mud (Eyles and Eyles, 1983; Hicock and Dreimanis,

1992). Since deposition about 45 ka (Berger and Eyles, 1994), it has been heaviiy

overconsolidated by loading under yo-mger sediments and by as much as 1 km of ice during

expansion cf the Laurentide Ice Sheet over the Toronto region during the late Wisconsin,

sometime after 23 ka. A good analogue for the Sunnybrook shortly after deposition is provided

by the glaciolacustrine Tilbury Clay of southwestern Ontario (Soderman et al., 1968; T.C.

Kemey, pers. comm. 1999). This deposit is of much younger age (c. 13 ka) and unlike the

Sunnybrook has not been heavily overconsolidated under a sediment or ice load; in the Tilbury

clay overconsolidation is restricted to a stiff surface c ru t foxmed by postglacial lake drainage

and dessication.

Representative mean-values of undrained shear strength (Cu) and Young's modulus (E) for the

Tilbury Clay below the hardened surface crust are 50 kPa and 35 MPa respectively (Sodeman et

al., 1968). Values of Cu range fiom about 34 kPa to 57 kPa, and E ranges between 7 MPa and 55

Mpa. These values are incorporated in the parametric finite element analysis discussed be!ow.

A reasonable value of E for the underlying Scarborough Formation sand was estimated fiom a

published range of values for sand to be about 70 MPa (LeLievre and Matyas, 1991). E values

for sand Vary with confining pressure (Craig, 1992) and a range of values between 35 and 140

MPa was used to account for this variation. Values were systematically varied during mode1 nuis

(Table N).

Unit weights for both the Sunnybrook and Scarborough are assumed to be 19 kN/m, which is

typical for sand and clay. Poisson's ratio is theoretically 0.5 for saturated, undrained soils (Craig,

1992). For numerical convergence purposes, Poisson's ratio is assumeci to be 0.49 (e.g. Yang et

al ., 1 996; Geoslope International, 1997).

4.1.1 S l o ~ e stabiIitv analvsis

A stability analysis of the sidedopes of the scour provides additional information on the

geotechnical properties of the Sunnybrook at the time of scouring. The smooth side slopes of the

scour demonstrate that immediately after scouring, and before the open scour was filled with

sand, the side slopes were stable. A slope stability analysis can be used to determine the

minimum value of Cu necessary to allow a factor of safety of 1 .O (Le., a condition of stability).

This analysis was conducted using the SLOPEN limit equilibnum program (Geo-Slope

Intern&ional, 1997a) combined with Taylor's (1937) stability coefficients for cornparison (see

also Bromhead, 1992). The steepest scour slope at Cudia, with an angle (P) of 57", was analysed.

For the S L O P E N analysis, the Morgenstern-Pnce method (Bromhead, 1992) was used. Results

indicate that for the minimum factor of safety to be 1 .O, the value of Cu for uie Sunnybrook is

about 4 kPa (the minimum slip surface is shown on Figure 12 A). Using Taylor's slope stability

coefficients the following equation was used:

N d H

where F is the factor of safety of the slope against rotational slump failure, N, is the Taylor

stability coefficient, y is the unit weight of the soil, and H is the dope height. This method

assumes uniform soi1 properties in the slope, which applies in this case because the scour is cut

entirely within the Sunnybrook. Underlying deltaic sands are sufficiently far below the bottom

of the scour that the minimum slip sudace would not intersect the sand. From Taylor's design

chart (Taylor, 1937), and a slope angle (B) of 57O, the Taylor stability coefficient is Ns = 0.19.

Because the slope is submerged, the buoyant unit weight of the Sunnybrook was used (1 9 kN/m3

- 9.8 kN/m3 = 9.2 kN/rn3). The maximum slope height is about 3.75 m and assuming F = 1,

solving equation 1 for Cu yields a value of Cu of 7 kPa.

Based on the above results of Cu = 4 kPa (i3om SLOPE/W) and Cu = 7 kPa (fiom Taylor's

stability coefficients), a value of Cu = 5 kPa was estimated as the "minimum" undrained shear

strength for the Sunnybrook substrate at the time of ice scouring. A corresponding value of

Young's modulus was determined using the correlation between E and Cu reported for the

Tilbury Clay by Soderman et al. (1968) where E = 700 Cu. This correlation is consistent with

others cited in the literature, where E = 500 Cu for soft clays, and E = 1000 Cu for fim to stiff

clays (e.g., LeLievre and Matyas, 1991). Given this relationship, the "minimum strength" E for

the Sunnybrook at the time of scouring is estimated to be about 3.5 MPa,

4.2 Finite element mode1

Having established relevant geotechnical properties of the substrate at the time of scouring, ice-

scour forces were modelled using a commercially available 2-D finite element stress rnodelling

program SIGMNW (Geo-Slope International, 1997). The berms on the margins of the scour

(Figs. 5 , 6 , 7) are insignificantly small in terms of sub-scour stress distribution and are omitted

for simplicity. In addition, it is assumed that the maximum sub-scour deformation occurred when

the scouring ice-mass had penetrated to its maximum depth. As shown in Figure 12 B, a number

of additional conditions are assurneci:

O the bottom of the Sunnybrook was initially planar and horizontal;

a zero vertical displacement boundary condition is placed 20 m below the bottom of the

scour (i.e., well below the maximum observed depth of deformation);

zero horizontal displacement boundaries are placed 30 m to either side of the scour edges

and are assumeci to be beyond the influence of scow deformation;

a downward acting water pressure boundaq is placed on the top of the Sunnybrwk and

on the bottom margin of the scour, simulating a 20 m water depth at the tirne of scouring;

and

a downward force boundary is assumed to act on the scour and simulate the downward

force of the scouring ice mass.

A mode1 mesh was created, based on the geometry of the scour, and consists of 1840 elements

and 1938 nodes. SIGMAW pemits different matenal models to be used. For the sands of the

Scarborough Formation, a linear-elastic model was used, where the substrate deforms linearly in

proportion to applied stress. For muds of the Sunnybrook, the elastic-plastic model was used,

where the substrate behaves as a linear-elastic matenal up until the yield stress value is reached.

The yield stress depends in turn on the sediment's undrained shear strength (C,). The SIGMNW

calculation was set to iterate ten times or until the displacement convergence vector nom was

less than 1%. The displacement convergence vector nom is a measure of how much

displacement is calculated in a given iteration, relative to the cumulative displacement up to that

itcration. The solutions in this study converged to a vector n o m of a few percent or les, which

is acceptable practice (Geoslope International, 1997).

To investigate the e f fe t that self-weight has in causing downward displacement, the mode1 was

nin for an in-situ, pre-scow case with a flat lake bottom surface. Elastic settlements fkom the

self-weight of the soi1 were found to be of the order of a few millimetres and so can be neglected

in the interpretation of the overall settlements based on the downward scour force.

Figure 12 C shows contours of calculated shear stress below the ice scour during its fornation.

The mode1 also shows that changing the assumed values of Cu and E for the Sunnybrook has

little effect on calculated defornation at the Scarborough/Sunnybrook interface (Fig. 12 D). Cu

was varied fiom 5 kPa to 75 kPa, and E was varied Erom 17.5 MPa to 70 MPa, values that

bracket a credible range of Young's modulus and undrained shear strength (see above). In

contrast, the value of Young's modulus for the Scarborough sand had a strong influence on the

calculated deformation (Fig. 12 E). The assurned value for E of the Scarborough sand was 70

MPa and when doubled to 140 MPa, calculated downward displacements in the sand were

approximately halved. When an E value of 35 MPa was used, half the assumed value, the

calculated downward displacement in the sand was approximately tripled. Because Young's

modulus for a sand is stress-dependent and difficult to determine, this introduces uncertainty in

calculating expected deformation in the top of the sand that could lead to emors in calculated

deformation of up to a few hundred percent. Nonetheless, the mode1 is sufficiently robust to

provide a ~ ~ O ~ O U S test of whether an ice scour cut into the Sunnybrook could have caused the

observed sub-scour deformation in underlying sands. The maximum downward displacement of

the sand below the base of the Sunnybrook is about 1 m (Figure 7). Based on the modelling

results, this depth of deformation is created by a downward scouring force of about 5 x 10) kN

(Fig. 12 D). This is consistent with the findings of Palmer (1997) who reported that the scouring

force of modem icebergs is of the order of 1 x lo4 kN. The next step of the analysis is to identim

the size of ice mass that created the scour.

4.4 Size and draft of ice mass

The size of ice m a s that created the scour at Cudia Park c m be readily estimated using an

equation developed by Golder Associates (1982):

where

- p-x - maximum iceberg scour force in N

f - - compressive strength of ice or substrate sediment (whichever is less) in Pa

m - - effective ice m a s in kg (including hydrodynamic mas)

v - - ice velocity in m/s.

The term 'hydrodynarnic mass' refers to the additional mass of water dragged or pushed dong by

moving ice and which effectively ad& to the ice mass. Hydrodynamic mass depends on the

shape of the floaMg ice mass and ranges h m about 20 to 50 % of the ice mas . Average ice

velocities range fiom about 0.2 to 0.3 m/s but instantaneous speeds can be greater (Golder

Associates, 1982). The analysis assumes that vertical and horizontal components of the ice

scouring force are of the same order-of-magnitude. This asswnption is supported by the results

of Yang et al. (1996) who conducted two centrifuge models of ice scouring in the laboratory and

showed that vertical forces were 1.5 to 3 times greater than horizonta1 forces.

Given P, = 5 x 103 kN, and assuming that f = 50 kPa (assumed to be equal to the then

undrained shear strength of the Sunnybrwk), and v = 0.5 d s (assumed instantaneous speed),

then the effective mass of ice that produceci the scour at Cudia Park is calculated to be 0.04

million tonnes. An ice mass of this size is classifieci as a "Bergy Bit" according to the iceberg

size clzssification scheme of Gustajtis (1978). Given a range of assurnptions for height

(fieeboard above water) of 1.5 to 5 m for a "Bergy Bit" and typical draft to height ratio of 7: 1,

then the draft of the "Bergy Bit" is calcuiated to be between 10 and 30 m. This value agrees very

well with estirnated water depths of no more than 20 m for the storm-deposited shoreface sands

that fil1 and bury the scour (çee above). A draught of 20 m is well within the size range of keels

that develop under pressure ridges formed of seasonal or multi-year lake ice, but given that the

lake was ice-dammed, icebergs were almost certainly present.

The following section describes other structures on the upper surface of the Sunnybrwk that

result fiom ice-scour processes.

4.5 Ice scoured rnud and ice-keei turbated facies

Betsveen Cudia Park and Sylvan Avenue, wer an outcrop exposure of more than 2000 rn (Fig.

l), the strôtigraphic contact between the rnud of the Sunnybrook and overlyhg Thorncliffe

Formation sands has a highly inegular 'wavy' topography with a relief amplitude of up to 3 m

(Figs. 9, I OA, B, C). The same topography is also seen where the upper Sunnybrook crops out

inland dong the Rouge River near Metro Zoo (Fig. 1, 10D, E, F). The top of the Sunnybrook

shows sand-filled depressions up to 3 m deep bounded by upward-tapering pillars of rnud up to 1

m wide. Such structures are of considerable engineering interest as they control the movement of

groundwater across the surface of the Sunnybrook, which f o m s a major aquitard in the regional

hydrostratigraphy, and result in 'piping' and accelerated erosion dong the Bluffs ( e g , Eyles et

al., 1986).

Mapping of the outcrop at Sylvan Park in 1985 and 1995 afier the face had retreated in some

cases as much as 8 m, identifies the complex three-dimensional geometry of the mudfsand

contact (Fig. 9). The irregular surface is filled and blanketed with the same fine-grained storm-

deposited shoreface sands of the Thomcliffe Formation, as described above at Cudia Park.

These show erosive-based beds up to 1 m thick recording successive storms; lowemost sands

fi 11 ing the irregular topography are extensive1 y de formeci by so fi-sediment pillows and water-

escape structures (Fig. 9). Rafts and blocks of defomed sand 'float' in the rnud below (e.g., site

1 ; Fig. 9) and irregular 'clumps' of rnud also comrnonly occur within the overlying sand on the

flanks of rnud pillars. Dispened ice-rafted debris, including boulden up to 40 cm in diameter, is

also present (Fig. 9 ). Along the Rouge Valley at site R1 (Fig. 10Ey F), horizontally-laminateci

silty clays within the Sunnybrook have been depressed below one prominent basin and are

deformed to a vertical orientation below the adjacent rnud pillar.

Sand-filled basins that penetrate d o m into the Sunnybrook were intetpreted by Eyles and Clark

(1 986) as the product of sol?-sediment 'loading' following the rapid deposition of storm-

deposited sand on a soft mud. Paleomagnetic work at Sylvan Park showed that muds were very

sofi at the tirne of deformation (Eyles et al., 1987). However, lowemost sands cleady infill a

pre-existing 'scooped' topography on the underl-ying rnud (Figs. 9, 10). The extent of mud

deformation exhibited below scoops, where sand rafts are intermixeci with mud, also indicates

intense mechanical disturbance that cannot be reconciled with simple gravitational loading but is

consistent with ploughing by floating ice (Fischbein, 1987). Sand rafts within the underlying

muds attest to compIete disruption of storm-deposited beds by ploughing into underlying rnud

(see below). The irregular relief on the surface of the Sunnybrook likely results fiom multiple

episodes of ploughing of cohesive rnud and the intersection of numerous xour and trough

berms now preserved as mud pillars (Fig. 13). initial s tom deposits that fil1 the topography are

deformed by water escape structures resulting fiom rapid sedimentation on and around a highly

irregular rnud surface with scours and upstanding berms. Mud blocks within sands likely result

from the erosion and collapse of berms and storm-induced re-sedimentation of muddy debris into

scours (C. Woodworth-Lynas, Pers. Comm., 2000; Fig. 13). Cornparison of the geomorphology

of modem seafloors scoured by icebergs in Antarctica (e.g., Barnes, 1997) and that formed by

perennial ice in Alaska (Barnes et al., 1987) with that seen on the upper Sunnybrook shows

striking similarities (Fig. 11). Reirnnitz and Kempena (1982) described a similar environmental

setting on the upper shoreface of an Arctic barrier island where an 'ice wallow relief foms each

season by the stranding of ice floes.

Deformed sediment below each scour, where sand is mixed with rnud (Fig. 9) is interpreted as

having been ploughed ('turbated') by floating ice. Vorren et al (1983) introduced the term

'iceberg turbate' for substrate sedirnents ploughed by icebergs; Woodworth-Lynas and

Dowdesweil(1994) introduced the terms lake-ice turbate and sea-ice turbate to describe the

sediments produced by ice-scour either by seasonal or multi-year ice. Because of the lack of

data as to the type of floating ice present at Scarborough, the non-specific term ice-keel turbate is

used here for the upper surface of the Sunnybrook. Pleistocene and pre-Pleistocene ice-keel

turbated facies (tenned K T facies here) are only rarely reporteci in îhe Literature (Table III).

The Sunnybrook is easily mapped in the subsurface inland using downhole geophysical and

borehole logs and foms a tabiilar stratipphic 'market with a broadly consistent thickness (< 10

m) lying between about 110 and 125 m above sea level recording deposition on a flat or gently-

sloping delta top. The widespread presence of the same ice-scoured topography on the uppermost

Sumybrook surface where exposed inland along creeks (Fig. 1) indicates that the former lake

floor was criss-crossed by numerous scour structures perhaps comparable to that seen on the now

exposed fioor of former glacial Lake Agassiz (Clayton et al., 1965; Dredge, 1982; Fig. 1 1). In

future studies, high-resolution 3D seismic or ground penetrating radar investigations may reveal

a Sunnybrook surface cnss-crossed by ice scours.

5 RELATED ICE-SCOUR STRUCTURES GND FACIES OF THE SUNNYBROOK

So far in this thesis, the deformationai effects of scouring and ice-keel turbation have been

described only fiom the upper surface of the Sunnybrook. Such structures are easily recognised

because scours are infïlled with stormdeposited sand, and sand has been ploughed into mud. In

contrast, the record of ploughing is more difficult to recognise in deeper water offshore settings

where the substrate consists only of mud and pebbly mud (diamict) facies and turbation results in

non-diagnostic facies (e.g., Dowdeswell et al., 1993). The following section describes structures

and facies exposed at the base of, and within, the muddy Sunnybrook that may be the result of

deposition fiom, or defonnation by, floating ice. Shear zones, striated mud surfaces and stone

lines are descnbed that can be explained as the product of floating ice and it is speculated that

much of the glaciolacustrine Sunnybrook rnay be an ice-keel turbate deposit.

5.1 Ice-rafted debris within the Sunnybrook

Texturally, the Sunnybrook is extremely fine-grained, consisting of up to 80% mud, with a low

and highly variable component of sand and clasts. Massive mud grades laterally and vertically

into diarnict (where the volume of clasts is greater than 10% of sediment volume). Sporadic sub-

horizontal concentrations of clasts in otherwise stone-poor Sunn ybrook muds (e.g., Figs. 3C, D,

E, F) were reported by Eyles et al. (1983a) and referred to as 'stone lines' by Kelly and Martini

(1 986). They typically consist of a sma1l number (usually < 6) of dispersed and mostly rounded

or subrounded clasts ranging in diameter from a few centimetres to 30 cm. They extend along

any one stratigraphie level for no more than 1 m (Fiy. 3F). Such crudely-defïned 'beds' of clast-

rich mud show diffbse upper and lower boudaries (Fig. 3D). Their three-dimemional form is

that of small 'patches' of clasts no more than 2 or 3 m2 in extent.

'Stone lines' were re-named 'clast pavements' by Hicock and Dreimanis (1992) but the term is

incorrectly applied because clasts are not close-packed or in mutual coctact (see also Fip. 2b in

Hicock and Dremanis, 1989 and Figs. 4f and 5b in Hicock and Dreimanis, 1992). They also lack

the extensive lateral continuity and smoothed striated upper surface of tnie pavements (see Eyles,

1988; 1994; Visser, 1989; Benn and Evans, 1998, p. 392). Hicock and Dreimanis (1989) reported

clasts having consistently-oriented striations on their upper surfaces at four sites along the Bluffs

and also reported the presence of glacially-shaped 'flat-iron' clasts; exactly the same data set was

republished by Hicock and Dreimanis (1992). They suggested that clasts had been deposited as

part of a subglacial pavement striated by a westward-flowing ice sheet. In contrast, an extensive

stiidy by RandalI (1 995) along the entire outcrop length of the Sunnybrook including inland

valleys, was unable to replicate previously published clast orientation results. Clasts are few

within the Sunnybrook (Fig. 3A, B) and are dominantly composed of granite-gneisses fom

northem 'shield' source areas and show a sub-equant, blocky shape fiom which no usehl

directional data can be obtained. Relatively large numbers of steeply dipping, cornmonly

vertical, clasts were not reported by Hicock and Dreimanis (1992). Angular clasts of soft shale

are common. Striated clasts are extremely rare and where found consist of sub-rounded clasts

with multiple crossing sets on several surfaces and consequently were not used as directional

indicators in earlier studies (Eyles and Eyles, 1983; Kelly and Martini, 1986; Gravenor and

Wong, 1987; Sharpe, 1988). Glacially shaped 'flat-iron' clasts diagnostic of subglacially

transported debris (e.g., Boulton, 1978) are extremely rare.

Stone lines and clast-rich horizons ('zones') are interpreted here as concentrations of ice-rafted

debns characterised by non-oriented clasts (e.g., Domack and Lawson, 1985). The relative

abundance of sub-rounded clasts lacking any striations indicates that a glacial source is unlikely;

the very few sîriated clasts that are present identiQ a minimal glacial contribution. Dionne

(1 974, 1988, 1993) describes 'clurnps' of river ice-rafted debris, one or two clasts thick, on mud

flats along the St. Lawrence River similar to the 'stone lines' seen in the Sunnybrook (see also

Martini et al., 1993). So-called 'dump deposits' formed by this process are described fiom the

fiords and continental shelf of East Greenland (Dowdeswell and Dowdeswell, 1989;

Woodworth-Lynas and Dowdeswell, 1994, p. 253) and Alaska (the 'clast clustea' of Eyles et al.,

1 985, their Fig. 1 ). Von Engeln (1 9 1 8) used the tenns 'nests' and iceberg 'droppings'; large

'bergmounds' deposited by the stranding of icebergs, are described by Fecht and Tallman (1978).

The near absence of striated and glacially-shaped clasts fkom the Sunnybrook may indicate

rafting of poorly-sorted beach material perhaps incorporated into and released by seasonal or

perennial ice masses, rather than icebergs transporting glacial debris.

5.2 Striated mud surfaces

Pebbly mud Sunnybrook facies are well exposed inland along cut banks and valley sidewalls of

creeks that drain to Lake Ontario. An intrafonnational bedding plane exposure within massive

and laminated Stoney mud is exposed along Highland Creek about 1 km south of Old Kingston

Road along the east bank of the river (Fig. 1). This surface shows well-developed northwest-

trending 'soft-sediment sîriations' on rnud (Fig. 14A) consisting of small-scaie b w s with a

relief of about 1 cm, on which are superimposed milhetre-scaie grooves and ridga.

Additionally, srnall micro-ridges lie transverse to the trend of h o w s and are associated with

small offsets in the trend of striations (Fig. 148. C). Laminated and crudely stratified muds

below the pavement are contorted by small-scale shears and contain small, tightly-folded rafts of

laminated silty clay up to 20 cm in diameter. The Highland Creek exposure is the largest (3 m2)

seen to date; others are sporadically exposed within Sunnybrwk outcrops dong Scarborough

Bluffs.

The sûiated rnud surface at Highland Creek is identical to those described fiom modem

seasonally-cold tidal flat sediments scoured by ice floes along the St. Lawrence River in Quebec

(the 'dnft ice tracks' of Dionne, 1974, 1988). The transverse micro-ridges on the surface at

Highland Creek closely resemble the 'rippled mud tracks' descnbed by D ~ O M ~ (1 974) that are

formed where partly floating ice slides with an irregular 'stick-slip' motion across a soft substrate,

and are similar to the 'streaked mud surfaces' reported by Von Engeln (1 9 18) fiom Alaska

formed when icebergs glided over soft marine mud.

Pre-Pleistocene exarnples of striated rnud surfaces are listed on Table iII and have been

discussed by Woodworth-Lynas and Dowdeswell(1994). Such surfaces are interpreted almost

exclusively as subglacial (Table III) but these authors argued that an ongin below floating ice is

more plausible. They cited the lack of additional 'contextual' evidence for subglacial and ice-

proximal conditions and the special conditions required to fom and preserve such a surface

below a dynamic ice sheet. The author agrees with their argument and notes M e r that Visser

(1990) has descnbed a grooved and ridged surface, similar to that seen in Figure 14, near the

base of the Carboniferous-PemUan Dwyka Formation of southern Afkica The surface was

formed, so it was argued, when a glacier margin overrode soft sediment. It is significant that the

striated surface is cut across a thin (< 3 m) succession of interbedded nppled sandstones and

rnudstone containing ice-r&ed clasts and thin diamictite lenses (one no more than 30 mm thick

and al1 less than 1 m thick) that has been interpreted as till. The surface is in tum blanketed by

marine shale with dropstones. Sunilar 'fbrowed' surfaces of the same age and stratigraphy are

described by Savage (1972) and Von Brunn (1977) and are al1 overlain by transgressive

mudstones. Dionne (1974) briefly discussed the work of Savage (1972) and pointed to the lack of

glaciotectonic defornation and absence of thick tillites and suggested, in contrast, formation by

drifting ice masses. The same comment applies to the example of Visser (1990) who noted, very

significantly, possible iceberg scours on the top of the reported section (Visser, 1990, p. 238).

5.3 Pebbly mud ire-keel turbated facies

Eyles (1 982) and Eyles and Eyles (1983) identified two predominant 'Sunnybrook' mud facies

types: massive, with variable clasts, recording the dominance of tain-out' processes (Fig. 3A, B,

C, D, E, F), together with weakly-stratified facies resulting from the syn- and post-depositional

downslope flow of mud (Fig. 3G, H). These last facies oAen contain thin units of rhythrnically-

larninated silty clays recording density underfiows, and clast breccias indicating post-

depositional reworking of cohesive muds. Such facies are oAen gently folded by slumping (Eyles

et al., 1983a). Visser (1985, 1989) describes the same facies types and associations from Late

Paleozoic glaciomarine 'rain-out complexes' in the Karoo Basin of southem Afnca. A

Sunnybrook pebbly mud facies occurs where massive facies show isolated rounded or slab-like

'rafts' of fine sand often dipping at a hi& angle, wispy deformed silt laminae, defomed Stone

lines and contorted clusters of non-oriented clasts creating a massive to weakly-sbatified,

'disorganised' structure. The association of this facies with ice-keel turbated muds and sands

strongly suggests that the Sunnybrook has been twbated, at les t in part.

Ice-keel turbated deposits are conunon on modem glacially-influenced continental shelves where

'plume deposits' originate by rain-out of suspended mud and ice rafted debris but, in addition, are

turbated by the keels of icebergs (the iceberg turbates of Vorren et al., 1983; termed IBT facies

herein). Photographs and descriptions of modern IBT facies fiom the East Greenland fiords and

shelf (Dowdeswell et al., 1993; Woodworth-Lynas and Dowdeswell, 1994; Dowdeswell et al.,

1994; Smith and Andrews, 2000), fiom the Antarctic shelf (Shipp et al., 1999; ODP, 1999), the

Canadian continental shelf (Josenhans et ai., 1986; Davies et al., 1997) and fkom the Nonvegian

continental margin (Vorren et al., 1983) bear close comparison with the pebbly muds of the

Sunnybrook. Ostracode faunas present in the Sunnybrook, such as Candona Caudata, suggest

water depths greater than a few tens of metres (Westgate et al., 1987; Rutka and Eyles, 1989)

which is beyond the influence of keels formed below pressure-ridged multi-year pack ice. This

indicates the influence of icebergs and suggests that the Sunnybrook is partly cornposed of MT

facies. Detailed comparison between Sunnybrook facies and modem IBT facies is warranted,

perhaps concentrating on the fine structure of such deposits as revealed by X-rays and magnetic

anisotropy (e.g., Gravenor and Wong, 1987; Eyles et al., 1987), but is constrained by the scarcity

of descriptions of JBT facies in outcrop (Table II).

5.4 Shear zones at the base of the Sunnybrook

As related above, the contact between the Suanybrook and underlying sands can be divided into

a) confonnable (Fig. 4 A), b) interbedded Vig. 4B), and c) where either type has teen post-

depositionally deformed by shearing (e.g., Fig. 4C), such as found below the ice scour at Cudia

Park (Fig. 7). Similar zones of deformed sediment occur sporadically along the Bluffs (Randall,

1995) and were described as 'easily overlooked' by Hicock and Dreimanis, 1992, p. 150).

Randall(1995) systematically mapped the occurrence and thiclcness of defomed sedirnents

along the Bluffs and demonstrated that they are typically of limited extent along section (c 15 m)

and are separated by zones of undefonned sediment. Deformation commody takes the fomi of

parallel sub-horizontal shear surfaces forming the upper and lower bounding surfaces of a shear

zone of deformed and intemiixed mud and sand, up to 30 cm thick (Fig. 4D). In tum, massive

rnud of the Sunnybrook rests abmptly on the uppermost shear surface. Zones of deformed

sediment thin and pinch-out laterally along section. Deformation is most prevalent on the flanks

of the paleovalley at Bluffers Park (Randall, 1995).

Hicock and Dreimanis (1 992) lumped discontuiuous outcrops of defomed sediment into a

forma1 stratigraphie unit they named the 'Sunny Point Till membef. This unit was interpreted as

a 'deformation till' formed by a 'partly-grounded ice sheet' that was, however, sufficiently thick to

ovemn the Niagara Escarpment and extend 250 km southwestward to the Lake Erie basin.

Hicock and Dreimanis (1992) emphasised the presence of consistently-oriented and glacially-

shaped clasts found as 'pavements' within sheared sediment at the base of the Sunnybrook as a

key criterion for recognition of deformation till. As related above, such 'pavements' are readily

interpreted as the product of floating ice. Moreover, at Cudia Park deformation is unambiguously

the product of sub-scour deformation during localised scouring by a fkee floating ice mass.

Consequently, the localised and diseontinuous zones of deformed sediment at the base of the

Sunnybrook are more readily explained as sub-scour deformations rather than subglacial

glaciotectonic deformations It has been stressed many times, most recently by Eyles and

Williams ( 1 992), that the distinctive sedimentology, geometry and paleontology and isotope

stratigraphy of the Sunnybrook does not conform with the characteristics of tills deposited under

ice sheets. True tills are of complex geometry reflecting the development of landfoms at their

base and top, typically contain abundant striated clasts, and are associated with other ice-

proximal landforms and sediments occurring together in a 'subglacial depositional system' (see

Benn and Evans, 1996, 1998; Boulton et al., 1996; Boyce and Eyles, 2000).

It is significant that ice grounding structures occur at the base of the Sunnybrook, where it rests

on the upper surface of the Scarborough Formation delta. The change fiom sand to mud

deposition records an increase in water depths sufficient for mud to accumulate across the delta

top and to allow access to floating ice masses. As noted by Randall(1995), deformation is most

prevalent on the margins of the paleovalley at Bluffers Park, most likely reflecting floating ice

moving onshore along the deeper water of the valley and grounding against the slopes of the

valley.

6 DISCUSSION

Clearly, in the absence of outcrop descriptions of ancient ice-ploughed facies, a systematic

search of glaciolacustrine and glaciomaruie strata in Pleistocene and preZleistocene basins is

required. Data fiom Scarborough indicates that good candidates are the extensive Pleistocene

glaciolacustrine deposits of mid-continent North America.

Greatly-enlarged glacially-dammed water bodies formed repeatedly during Late Wisconsin

glaciation of the Great Lake basins and their deposits are widely exposed in coastal bluffs around

the modem lakes. Given the density of relict ice scours seen on the now-exposed floors of other

large glacial lakes of the time (e.g., Glacial Lake Agassiz; Clayton et al., 1965; Dredge, 1982;

Fig. 1 1 C), ice scour and ice-turbation of seàiments was a comrnon process. Eyles and Eyles

(1 992) suggested that muddy iceberg turbates have been described unwittingly in the literature as

tills or tillites; indeed, the Sunnybrook Till' of southern Ontario is the f h t such candidate ta be

identified. Karrow's (1969) redesignation of the Sunnybrook 'Till' as 'Sunnybrook Drift' in

recognition of the glaciolacustrine appearance of the deposit may be very appropriate in view of

Lyells' (1 83 1 ) intended use of the term 'drift' for deposits of icebergs.

An absence of scour structures reported from Pleistocene glaciomarine sediments in mid and

hi&-latitude coastal zones, now elevated above sea level by glacio-isotatic recovery, is also vety

surprising given the repeated release of vast armadas of icebergs f?om northem hemisphere ice

sheets (e.g., Bond et al., 1992; Hesse and Khodabakhsh, 1998). The same comment applies to the

sedimentary record of pre-Pleistocene glaciations, which is dominated by glacially-uitluenced

marine strata that accurnulated on continental shelves. Furthemore, recent models of a late

Proterozoic 'snowball Earth' where the oceans are postulateci to have been covered by perennial

ice up to 10 km thick (Hofian, 1998) suggests that massive 'canyon-scale' ice scows await

discovery in the rock record.

Given al1 the considerztions discussed above, ice-ploughed strata and structures should be

common, not rare as currently indicated by the literature (Tables 1, II, III). The author strongly

agrees with the conclusion of Woodsworth-Lynas and Dowdeswell(1994) that ancient ice-

scoured facies are in fact more common than the literature suggests but have been

misinterpreted. An important aid to recognition of ancient ploughed strata is stratigraphic

context. Such structures are much more likely to be formed and preserved either at the base of

deepening-upward stratigraphic successions, Le., on flwding surfaces (when floating ice masses

are able to uivade an area); or toward the top of shallowing-upward successions on regressive

,ause surfaces where the opportunity for ice grounding is maxunised. In addition, be-

transgression and regression results in vertical juxtaposition of very different sedimentary facies,

the deformational effects ofice keei turbation will be more evident. A good example is where

shoreface sands are ploughed into underlying muds (Fig. 9). In contrast, the sedimentary effects

of ice- scouring and deformation in offshore mud deposits of glacially-influenced shelves and

lakes (such as IBT and K T facies), may be more subtle and more difficult to recognise given

that ploughing produces non-descript massive pebbly mud facies. Clearly, there is an important

need for detailed outcrop descriptions of ploughed sediments paying particular regard to related

facies and their position within stratigraphic successions.

7 REFERENCES

Aigner, T., 1 985. Storm Depositional S ystems. Springer-Verlag, Berlin, 1 74 p.

Aitchison, J.C., Bradshaw, M.A. and Newman, J., 1988. Lithofacies and origin of the Buckeye Formation: Late Paleozoic glacial and glaciomarine sediments, Ohio Range, Transantarctic Mountains, Antarctica. Palaeogeography, Palaeoclimatology, Palaeoecology, 64,939 104.

Allen, J.R.L., 1984. Sedimentary Structures, Their Character and Physical Basis. Developments in Sedimentology 30, Elsevier, Amsterdam, 663 p.

Balderson, R.H. and Wilson, J.B., 1973. Iceberg plough marks in the vicinity of the Norwegian Trough. Norsk Geologisk Tidsskrift, 53,323-328.

Balderson, R.H., Kenyon, N.H. and Wilson, J.B., 1973. Iceberg plough marks in the northwest Atlantic. Paleogeography, Paleoclimatology, Paleoecology, 13,2 15-224.

Barnes, P.W., 1997. Iceberg gouges on the Antarctic SheK In: T.A. Davies and six others (Eds), Glaciated Continental Margins: An Atlas of Acoustic Images. Chapman and Hall, London, pp. 154-1 56

Bames, P. W., Rearic, D.M. and Reimnitz, E., 1984. Ice gouging characteristics and processes. In: Barnes, P. W., Schell, D.M. and Reimnitz, E., eds. The Alaskan Beaufort Sea: Ecosystems and Environments. Academic Press Inc., New York, p. 185-2 12.

Barnes, P. W., Asbury, J.L., Rearic, D.M. and Ross, C.R., 1987. Ice erosion of a sea floor knickpoint at the inner edge of the starnukhi zone, Beaufort Sea, Alaska. Marine Geology 76,207-222.

Barnes, P.W. and Lien, R., 1988. Icebergs rework shelf sediments off Antarctica. Geology, 16, 1130-1 133.

Barnes, P. W., Rawlinson, S.E. and Reimnitz, E., 1988. Coastal geomorphology of arctic Alaska. Arctic Coastal Processes and Slope Protection Design, TCCR Practice Report, ASCE, May 1988.

Benn, D.I. and Evans, D.J.A., 1996. The interpretation and classification of subglacially- deformed materials. Quaternary Science Reviews 15,23-52.

Benn, D.I., and Evans, D.J.A., 1998. Glaciers and Glaciation. Amold, 734 p.

Beres, M., Huggenberger, P., Green, A.G. and Horstmeyer, H., 1999. Using two- and three- dimensional methods to characterize glaciofluvial architecture. Sedimentary Geology 129, 1-24.

Berger, G. W. and Eyles, N., 1994. Themioluminescence chronology of Toronto-area Quatemary sediments and implication for the extmt of the midcontinent ice sheet(s). Geology, 22, p. 3 1-34.

Berkson, J.M. and Clay, C.S., 1973. Microphysiography and possible iceberg grooves on the floor of western Lake Superior. Geological Society of Amenca Bulletin, 84, 13 15-1328.

Beuf, S ., Biju-Duval, B., de Charpal, O., Rognon, P., Gariel, 0. and Bennacef, A., 197 1. Les grés du paleozoique inferieur au Sahara: sédimentation et discontinuités évolution structurale d'un craton. Paris, Editions Technip, 464 pp.

Blake, W., 1977. Iceberg concentrations as an indicator of submarine moraines, eastern Queen Elizabeth Islands, District of Franklin. Geological S w e y of Canada Paper 77-1B.

Bond, G.R and 1 3 others, 1992. Evidence for massive discharge of icebergs into the North Atlantic Ocean during the last glacial period. Nature, 360,245249.

Boulton, G.S., 1978. Boulder shapes and grain-size distributions of debris as indicators of transport paths through a glacier and till genesis. Sedimentology 25; 6: 773-798.

Boulton, G.S., Van der Meer, J.J.M., Hart, J., Beets, D., Ruegg, G.H.J, Van der Watern, F.M. and Jarvis, J., 1996. Moraine emplacement in a deforming bed surge - an example fkom a marine environment. Quaternary Science Reviews, 15, p. 96 1-987.

Boyce, J.I., Eyles, N. and Pugin, A., 1985. Seismic reflection, borehole and outcrop geometry of Late Wisconsin tills at a proposed landfill near Toronto, Ontario. Canadian Journal of Euth Sciences 32, p. 1331-1 349.

Boyce, J.I. and Eyles, N., 2000. Architectural element analysis applied to glacial deposits: interna1 geometry of a Late Pleistocene drumlinised till sheet, Ontario, Canada. Geological Society of America Bulletin, 1 12, 98-1 18.

Bromhead, E.N., 1 992. The Stability of Slopes. Blackie Academic & Professional, London. 41 1 p.

Chari, T.R., 1986. Iceberg-scour modelling at Memorial University of Newfoundland. In: Lewis, C.F.M., Parrott, D.R., Sirnpkin, P.G. and Buckley, I.T., eds. Ice Scour and Seabed Engineering, Environmental Studies Revolving Funds, Report No. 049, Ottawa. p. 109-1 17.

Clark, J.I. and Landva, J., 1988. Geotechnical aspects of seabed pits in the Grand Banks area. Canadian Geotechnical Journal, 25 ,44844 .

Clarke, J.M., 191 7. Strand and undertow markings of upper Devonian tirne as indicators of the prevailing climate. Bulletin of the New York State Museum, 196, 199-2 10.

Clayton, L., Laird, W.M., Klassen, R W. and Kupsch, W.O. 1965. htersecting minor lineations on Lake Agassiz Plain. Joumal of Geology, 73,652-656.

Clifion, H.E. and Dingler, J. R., 1984. Wave-formed structures and paleoenvironmental reconstruction. Marine Geology, 60, 1 65 - 1 98.

Craig, R.F., 1992. Soi1 Mzchanics, 5th Edition. Chapman & Hall, London. 427 p.

Dagallier, G., Laitinen, A.I., Malatre, F, Ignace, P.A.M., Van Campenhout, C. and Veeken, P.C.H. 2000. Ground penetrating radar application in a shallow marine Oxfordian limestone sequence located on the eastexn flank of the Paris Basin. NE France. Sedimentûry Geology 130, 149-165.

Darwin, C. R. 1 839. Journal of Researches into the Geofogy and Natural History of the various Countries visited during the Voyage of HMS Beagle round the World ( h t published 1 839 with many subsequent editions).

Darwin, C.R. 1855. On the power of icebergs to make rectilinear, unifonnly-directed grooves across a subrnarhe undulatory surface. London, Edinburgh and Dublin Philosophical Magazine and Journal of Science, 1 O, 96-98.

Davies, T. A. And six others. 1997 (Eds.). Glaciated Continental Margins: An Atlas of Acoustic Images. Chapman and Hail, London, 3 15pp.

Deynoux, M., 1980. Les formations glaciaires du Précambrian Terminal et de la fin de I'Orodovicien en Afiique de l'Ouest: deux examples de glaciation d'inlandis s u . un plateforme stable. Trauaux des Laboratoires dcs Sciences de la Terre, St.-Jérome, Marseille, \%), 17, 554 p.

Diem, B., 1985. Analytical method for estimating paleowave climate and water depth fiom wave ripple marks. Sedimentology, 32, p. 705-720.

Dionne, J.C., 1969. Tidal flat erosion by ice at La Pocatiere, St. Lawrence Estuary. Journal of Sedimentary Petrology, 39, 1 174- 1 18 1.

Dionne, J-C., 1972. Ribbed grooves and tracks in mud tidal flats of cold regions. Journal of Sedimentary Petrology, 41, 848-85 1.

Dionne, J.C., 1974. Polished and striatecl mud surfaces in the St. Lawrence tidal flats, Québec. Canadian Journal of Earth Sciences, 1 1,860-866.

Dionne, J.C., 1977. Relict iceberg furrows on the floor of Glacial Lake Ojibwa, Quebec and Ontario. Maritime Sediments 13, p. 79-8 1.

Dionne, J.C., 1982. Glaciel. In: M.L. Scwartz (Ed.,), The Encyclopedia of Beaches and Coastal Enviroments. Hutchinson Ross, Stroudsburg, Penn., p. 447.

Dionne, J-C., 1988. Characteristic features of modem tidal flats in cold regions. In: P.L.de Boer (Ed.,), Tide-idluencd Sedimentary Environments and Facies. D. Reidel Publishing, 301-332.

Dionne, J.C., 1989. Bibliographie du periglaciare du Quebec 1969- 1989, incluant le glaciel pour la period 1960- 1 989. Geographie physique et Quaternaire, 43,233-243.

Dionne, J.C., 1993 Sediment load of shore ice and ice rafüng potential, Upper St. Lawrence Estuary, Quebec, Canada. Jolrrnal of Coastal Research 9, 628-646.

Domack, E. W. and Lawson. D.E. 1985. Pebble fabric in an ice-rafted diamicton. Journal of Geology, 93,577-592.

Dowdeswell, J.A. and Dowdesweil, E.K., 1989. Debris in icebergs and rates of glacimarine sedimentation, obsewations and a simple model. Journal of Geology 97,22 1-23 1.

Dowdeswell, J.A., Villinger, H., Whittington, R.J. and Marienfeld, P. 1993. Iceberg scouring in Scoresby Sund and on the East Greenland continental shelf. Marine Geology 1 1 1 ,373 1.

Dowdeswell, J.A., Whittington, R.J. and Marienfeld, P. 1994. The origin of massive diamicton facies by iceberg rafting and scouring, Scoresby Sund, east Greenland. Sedimentology 41,21-35.

Dredge, 1982. Relict ice-scour marks and iate phases of Late Agassiz in northwestern Manitoba Cm. J.Earth Sci., 19, 1079-1087.

Eyles, CH. , 1982. The sedimentology of the Early and Middle Wisconsinan deposits at Scarborough Bluffs, Ontario. Unpublished M.Sc. thesis. University of Toronto.

Eyles, C.H., 1988. A model for striated boulder pavement formation on glaciated, shallow marine shelves: an exarnple firom the Yakataga Formation, Alaska. J. Sedimentary Petrology, 58,62-71.

Eyles, C.H., 1994. Intertidal boulder pavements in the northeastern Gulf of Alaska and their geological significance. Sedimenw Geology, 88, 16 1 - 173.

Eyles, C.H. and Eyles, N., 1983. Sedimentation in a large lake: A reinterpretation of the late Pleistocene stratigraphy at Scarborough Bluffs, Ontario, Canada. Geology, 11, p. 146- 152.

Eyles, C.H., Eyles, N. and Miall, A.D. 1985 Models of glaciomarine sedimentation and their application to the ancient record. Palaeogeog. Palaeoecol. Palaeoclimat., 5 1, 15-84

Eyles, C.H., Eyles, N. and Gostin, V.A., 1998. Facies and allostratigraphy of high-latitude, glacially-influenceci marine strata of the Early Pennian southem Sydney Basin, Australia. Sedimentology, 45, 121-163.

Eyles, N., 1987. Late Pleistocene depositional systems of Metroplitan Toronto and their engineering and glacial geologic significance. Can. J. Earth Sci., 24, 1009-1022.

Eyles, N., Eyles, C.H. and Miall, AD., 1983. Lithofacies types and vertical profile models; an alternative approach to the description and environmental interpretation of glacial diamict sequences. Sedimentology, 30, 393-41 0.

Eyles, N,. Eyles, C.H. and Day, T.E., 1983a Sedimentologic and paleomagnetic characteristics of glaciolacustrine diamict assemblages at Scarborough Bluffs, Ontario, Canada. In: E.B.Evenson et al., (Eds), Tills and Related Deposits, A.A. Bakema, Rotterdam, 23-45.

Eyles, N., Eyles, C.H. and Miall, A.D., 1984. Lithofacies types and vertical profile models; an altemate approach to the description and environmental interpretation of glacial diamict sequences: Reply to comments. Sedimentology, 3 1,891 -898.

Eyles, N. and Clark, B.M., 1986. The significance of hummocky and swaley cross-stratification in late PIeistocene lacustrine sediments of the Lake Ontario Basin, Canada. Geology, 14, 679-682.

Eyles, N. Buergin, R. And Hincenbergs, A., 1986. Sedimentological controls on piping structures and the development of scalloped slopes along an eroding shoreline, Scarborough Bluffs, Ontario. National Research Council of Canada, Symposium on Cohesive Shores, Ottawa, 69-86.

Eyles, N., Day. T.E and Gavican, A., 1987. Depositional controls on the magnetic characteristics of lodgernent tills and other glacial diamict facies. Canadian Journal of Earth Sciences, 24,2436-2458.

Eyles, N. and Westgate, J.A. 1987. Restricted regional extent of the Laurentide Ice Sheet in the Great Lakes basins during early Wisconsin glaciation. Geology, 15,537-540.

Eyles, N. and Clark, B.M., 1988. Storm-influencecl deltas and ice scouring in a late Pleistocene glacial lake. Geological Society of Arnerica Bulletin, 100, 793-809.

Eyles, N. and Eyles, C.H., 1992. Glacial Depositional Systems. in: Walker, R.G. and James, N.P. eds. Facies Models, Response to Sea Level Change. Geological Association of Canada, pp. 74-92.

Eyles, N. and Williams, N. 1992. The sedimentary and biological record of the 1 s t interglaciaVglacia1 transition at Toronto, Canada. Geological Society of Amenca Special Paper 270, 119-138.

Eyles, N. Eyles, C.H. and Gostin, V.A., 1997. Iceberg rafting and scouring in the Early Permian Shoalhaven Group of New South Wales, Australia: Evidence of Heinrich-like events? Palaeogeog., Palaeoecol and Palaeoch., 136, 1 - 17.

Fader, G.B., King, L.H. and Josenhans, H.W., 1982. Superficial geology of the Laurentian channel and the westem Grand Banks of Newfoundland. Geological Survey of Canada Paper 8 1-22.

Fecht, K.R. and Tallman, A.M., 1977. Bergmounds dong the westem margin of the Channeleci Scablands, southtentral Washington. Geological Society of Amenca, Abstracts with Programs 1 O, p. 400.

Fischbein, S.A., 1987. Analysis and interpretation of ice-deformed sediments fiom H d s o n Bay, Alaska. U.S. Geological Survey, Open File Report 87-262.

Geikie, J. 1896. The Great ice Age. Appleton, New York, Third edition. 850pp.

Geo-Slope International. 1997. SIGMACW. CAD software for finite element stress/deformation geotechnical modelling, Calgary. Alberta.

Geo-Slope International. 1997a. SL.OPE/W. CAD software for limit equilibrium geotechnical modelling, Calgary, Alberta.

Golder Associates Ltd.; in association with: Albery, Pullerits, Dickron & Associates, Consulthg Engineers; Beaver Dredging Company Ltd; and W.F. Baird & Associates, Coastal Engineen, 1982. Grand Banks Artificial Island Shidy. Report for Petro Canada Ltd.

Grass, J.D., 1984. Ice scour and ice ridging studies in Lake Ene. IAHR Ice Symposium 1984, Hamburg.

Grass, J-, 1986. Ice-scour and ice-ndging studies in Lake Erie. In: Lewis, C.F.M., Parrott, D.R., Simpkin, P.G. and Buckley, J.T., Eds. Ice Scour and Seabed Engineering, Environmental Studies Revolving Funds, Report No. 049, Ottawa. p. 29-30.

Gravenor, C.P. and Wong, T. 1987. Magnetic and pebble fabrics and origin of the Sunnybrook Till, Scarborough, Ontario, Canada. Canadian Journal of Earth Sciences, 24,2038-2046.

Gustaj tis, K.A. 1978. Iceberg Population Distribution Study in the Labrador Sea. Technical Report, C-Core, August, 1978.

Hamelin, L.E., 1959. Dictionnaire fiancais-anglais des glaces flottantes. Universite de Laval, Institute de Geographie, Unpublished manuscript. 83 pp.

Harris, LM. and Jollymore, P.G., 1974. Iceberg hirrow marks on the continental shelf northwest of Belle Isle, Newfoundland. Canadian Journal of Earth Sciences, 11, 43-52.

Hesse, R. and Khodabakhsh, S., 1998. Depositional facies of late Pleistocene Heinrich events in the Labrador Sea. Geology, 26, 103-106.

Hnatiuk, J. And Wright, B.D., 1983, Sea bottom s c o u ~ g of the Canadian Beaufort Sea. In: Proc., 1 5th Annual Offshore Technology Conf., Dallas, Texas, paper 4584,3,35-40.

Hicock, S.R. and Dreimanis, A. 1989. Sunnybrook Drift indicates a grounded eariy Wisconsin glacier in the Lake Ontario basin. Geology, 17, 169-1 72.

Hicock, S.R. and Dreùnanis, A., 1992. Sunnybrook Dnfi in the Toronto area, Canada: reinvestigation and reinterpretation. Geological Society of Amenca Special Publication 270, 139- 162.

Hoffhan, P.F., Kauhan, A.J., Halverson, G.P. and Schrag, D.P, 1998. A Neoproterozoic Snowball Earth. Science, Vol. 28 1, pp. 1342-1 346.

Josenhans, H.W., Zevenhuizen, J. and Klassen, R.A., f 986. The Quatemary geology of the Labrador Shelf. Canadian Journal of Earth Sciences, 23, 190-2 13.

Josenhans, H.W. and Woodworth-Lynas, C.M.T., 1988. Enigmatic linear fùrrows and pits on the upper continental slope, northwest Labrador Sea: are they sediment f u ~ ~ o w s or feeding traces? Maritime Sediments and Atlantic Geology 24, 149- 155.

Karrow, P.F. 1967. Pleistocene geology of the Scarborough area. Ontario Department of Mines Report 46,108 pp.

Karrow, RF. 1969, Stratigraphie studies in the Toronto Pleistocene. Geological Association of Canada Proceedings 20,4- i 6.

Karrow, P.F. and Calkin, P.E., 1985. Quatemary evolution of the Great Lakes. Geological Association of Canada Special Paper 30.

Kelly, R.I. and Martini, I.P., 1986. Pleistocene glaciolacustrine deltaic deposits of the Scarborough Formation, Ontario, Canada. Sedirnentary Geology, 47,2742-

King, E.L., 1976. Relict iceberg fùrrows on the Laurentian Channel and western Grand Banks. Can. J. Earth Sci., 13, 1082-1092.

King. E.L. and Gillespie, R.T., 1986.Regional iceberg scour distribution and variability on the eastem Canadian continental shelf. In: C.F. Lewis, D.R. Parrott. P.G. Simpkin and J.T. Buckely (Eds). Ice Scour and Seabed Engineering. Environmental Studies Revolving Funds Report No. 049, Ottawa, p. 172-1 8 1.

Koschenkin, B.I., 1958. Traces of ice floe action on the seabed in shallow waters of the northem Caspian Sea. Akademiia Nauk SSR Laboratoria Astrometodov, Tnidy, 6, p. 227-234.

Kovacs, A. and Mellor, A., 1974. Sea ice geomorphology and ice as a geologic agent in the northem Beaufort Sea. In: Reed, J.C. and Sater, LE., eds., The Coast and shelf of the Beaufort Sea Arctic Institute Publication, pp. 1 13-16 1.

Kfiner, A. and Rankama, K., 1972. Late Precambrian glaciogenic sedimentary rock in southem Afnca: a compilation with dehnitions and correlations. Precarnbrian Research Unit, University of Cape Town, Bulletin ïI.

Leckie, D.A. and Duke, B., 1984. %gin of hummocky cross-stratification; Part 1, Straight- cres ted s yrnmetrical grave1 dunes; the coarse-grained equivalent of HCS . Canadian Society of Peîroleurn Geologists, Symposium on the Sedimentology of Shelf Sands and Sandstones, Program with Abstracts, p.5 1.

LeLievre, B. and Matyas, EL- 1991. Notes on Foundation Engineering, Department of Civil Engineering, University of Waterloo.

Lewis, C.F.M., Josenhans, H.W., Simms, A., Sonnischen, G.V. and Woodworth-Lynas, C.M.T., 1989. The role of seabed disturbance by icebergs in mixing and dispersing sediment on the Labrador shelE a high latitude continental margin. In: Program with abstracts, Canadian Continental Shelf Symposium (C2S3), Bedford Institute of Oceanography, Dartmouth, October 2-7.

Lewis, C.F.M., and Woodworth-Lynas, C.M.T., 1990. Ice scour. In: Geology of the Continental Margin of Eastern Canada, Geology of Canada v.. 2. Geological Survey of Canada, 785- 792.

Lien, R.L., 1 98 1. Sea bed features in the Blaaenga area, Weddell Sea, Antarctica. In: Proceedings of 6th Internatiofial Coderence on Port and Ocean Engineering under Arctic Conditions, Quebec, July 23-3 1,2,706-716.

Lien, R., 1 986. Iceberg scouring and its influence on seabed conditions: investigations and proposed mode1 for the Norwegian shelf. In: Lewis, C.F.M., Pmon, D.R., Simpkin, P.G. and Buckley, J.T., eds. Ice Scour and Seabed Engineering, Environmental Studies Revolving Funds, Report No. 049, Ottawa. p. 87-96.

Longva, 0. and Bakkejord, K.J., 1990. Iceberg deformation and erosion in soft sediments, southeast Norway. Marine Geology, 92, 87- 104.

Longva, O. And Thoresen, 199 1. Iceberg scours, iceberg gravity craters and current erosion marks kom a gigantic Preboreal flood in southeastern Norway. Boreas, 20,47962.

Lutemauer, J.L. and Murray, J. W., 1 983. Late Quaternary morphologie development and sedimentation- central British Columbia continental shelf. Geological Survey of Canada, Paper 83-21: 1-37.

Lyell, C. 183 1. Principles of Geology. John Murray, London.

McLaren, P., 1975 Under ice diving observations in the coastal environments of southeastem Melville and western Byam Martin Islands. In: Current Research, part A, Geol. Surv. Can., paper 75-14 475-477.

Marïenfeld, P., 1992. Recent sedimentary processes in Scoresby Sund, East Greenland. Boreas 21, p. 168-1 86.

Martin, H., 1965. The Precambrian geology of South West Afnca and Namaqualand. Precambrian Research Unit, University of Cape Town, 1 59 pp.

Martini, LP., Kwong, J.K. and Sadura, S. 1993. Sediment ice rafting and cold climate fluvial deposits: Albany River, Ontario, Canada. In: M. Marzo and C. Puigdefabngas (Eds.,). Alluvial Sedimentation. h t . Ass. Sedimentologists, Spec. Pub., 17,63-76.

Martini, 1.P and Brookfield, M.E. 1995. Sequence analysis of upper Pleistocene (Wisconsinan) glaciolacustrine deposits of the north-shore bluffs of Lake Ontario, Canada. Journal o f Sedimentary Research, B65,388-400.

Miller, J.M.G., 1989. Glacial advance and retreat sequences in a Penno-Carboniferous section, central Transantarctic Mountains. Sedimentology, 36,4 19-430.

Mollard, J.D., 1973. Landforms and surface materials of Canada, a stereoscopic airphoto atlas and glossary. J.D. Mollard Ltd., Regina.

Montes, A.S.L., Gravenor, C.P. and Montes, M.L., 1985. Glacial sedimentation in the Late Precambrian Bebedouro Formation, Bahia, Brazil. Sedimentary Geology, 44, 349-3 5 8 -

Ocean Drilling Program, 1999, Initial Reports, Volume 178, Antarctic Glacial History and Sea- Level Change Sites 1095-1 103. National Science Foundation - Joint Oceanographic Institutions, Inc., ISSN 0884-5883.

Palmer, A C , 1997. Geotechnical evidence of ice scour as a guide to pipeline burial depth. Canadian Geotechnical Journal, 34: 1002-1003.

Pelletier, B.R. and Shearer, J.M., 1972. Sea bottom scouring in the Beaufort Sea of the Arctic Ocean. 24th International Geological Congress, Section 8, Montreal, Quebec., pp. 251- 261.

Pereira, C.P.G., Piper, D.J.W. and Shor, A.N., 1985. SeaMARC I rnidrange sidescan sonar survey of Flemish Pass, east of the Grand Banks of Newfoundland. Geol. Surv. Canada Open File Report No. 1161.

Powell, R.D. and Gostin, V.A., 1990. A glacially-influenceci, storm-dominated continental shelf sytem on the Permian Australian-Gondwana margin. Abstract of Papers, 13th International Sedimentological Congress, Nottingham, UK, August 26-3 1, pp. 435-436.

Randall, T.A., 1995. A sedimentological investigation of the Sunnybrook diamict, Scarborough Bluffs region, Ontario, Canada Unpublished M.Sc thesis, McMaster University, Ontario, Canada, 284 pp.

Rattigan, J.H., 1967. Depositional, soA sediment and post-consolidation structures in a Palaeozoic aqueoglacial sequence. Journal of the Geological Society of Australia, 14, 5- 18.

Reimnitz, E., 1997. Stnidel craters on shallow Arctic prodeltas. In: T.A. Davies and six others, (Eds.,) Glaciated Continental Margins: An Atlas of Acoustic images. Chapman and Hall, London, pp. 146- 147.

Reimnitz, E., Bames, P., Forgatsch, T. and Rodeick, C., 1972. infiuence of grounding ice on the Arctic shelfof Alaska. Marine Geology, 13,323-334

Reimnitz, E. and Kempema, E., 1982. Dynarnic ice-wallow relief of nort'iem Alaska's nearshore. Journal of Sedimentary Petrology, 52, 45 1-46 1.

Rocha-Campos, A.C., dos Santos, P.R. and Canuto, J.R., 1994. Ice scouring structures in Late Paleozoic rhythmites, Parana Basin, B e l . In: Deynoux, M., Miller, J.M.G., Domack, E.W., Eyles, N., Fairchild, I.J. and Young, G.M., eds., Earth's Glacial Record. Cambridge University Press, Cambridge, p. 234-240.

Rutka, M. and Eyles, N., 1989. Ostracod faunas in Late Pleistocene glaciolacustrine diamict facies, Southern Ontario, Canada, and their paleolimnological significance. Pdaeogeog. Palaeoclim.Palaeoeco1., 73. 6 1-76,

Sensors & Software inc., 1996. PulseEKKO N RUN User's Guide, Version 4.2. Sensors and Software Technical Manual 20.

Savagc, N.M., 1972. Soft-sediment glacial grooving of Dwyka age in South AfXca. Journal of Sedimentary Petrology, 42,307-308.

Sharpe, D. 1988. The intemal structure of glacial landforms: An example from the Halton Till plain, Scarborough Bluffs, Ontario, Canada. Boreas, 17, 15-20.

S c h w i ~ z , H. P. and Eyles, N. 1991. Laurentide Ice Sheet extent inferred fkom stable isotopic composition (0,C) of ostracodes at Toronto, Canada. Quatemary Research, 3 5, 3 05-320.

Shearcr. J.M. and Blasco, S.M., 1975. Further observations of the scouring phenornena in the Beaufort Sea. In: Cwent research, part A. Geological Survey of Canada, Paper 75-l A, pp. 483-493.

Shipp, S., Anderson, J.B and Domack, E.W. 1999. Late Pleistocene-Holocene retreat of the West Antarctic Ice Sheet system in the Ross Sea: Part 1- Geophysical results. Bulletin Geological Society of America. 1 1 1, 15 1 7- 1536.

Smith, L.M. and Andrews, J.T., 2000. Sediment c haracteristics in iceberg dominated fjords, Kangerlussuaq region, East Greenland. Sedimentary Geology, 1 30, 1 1 -26.

Sodernian, L.G., Kim, Y.D. and Milligan, V. 1968. Field and laboratory studies of modulus of elasticity of a clay till. In: Soi1 Properties From in-situ measurements, a symposium. Highway Research Record, National Academy of Sciences Publication 163 1.

Taylor, D.W., 1937. Stability of earth slopes. Journal of the Boston Society of Civil Engineers, 24, 197-246.

Terasmoe, 1. 1960. A palynological study of Pleistocene interglacial beds at Toronto, Ontario. Geological Survey of Canada Bulletin, 56,23-4 1.

Thomas, G.S.P. and Connell, R.J., 1985. Iceberg drop, dump and grounding structures fiom Neis tocene g lacio-lacustrine sediments, Scotland. Journal of Sedimentary Petrology 55, 243-249.

Tyrell, J.B., 1892. Report on northwestem Manitoba. h: Annual Report for 1890-9 1, part E., Geological Survey of Canada.

Vandenberghe, J. and van Overmeeren, R.A., 1999. Ground penetrating radar images of selected fluvial deposits in the Netherlands. Sedimentary Geology, 128,245-270.

van der t inden, W.I. and Fillon, R.H., 1976. Hamilton Bank, Labrador margin. Geological S urvey of Canada, Paper 75-40, 3 1 p.

Vaslcr, D., 1990. Upper Ordovician glacial deposits in Saudi Arabia. Episodes, 13, 147-1 61.

Visser. J.N.J., 1985. The Dwyka Formation along the north-western margin of the Kamo Basin i i i the Cape Province, South Anica. Transactions of the Geological Society of South ilfrica, 88,3748.

Visser, J.N.J., 1989. The Permo-Carboniferous Dwyka Formation of southern Afnca: deposition by a predorninantly subpolar marine ice sheet. Palaeogeography, Palaeoclimatology, Palaeoecology, 70, 377-39 1.

Visser, J.N.J. 1990. Glacial bedfonns at the base of the Permo-Carboniferous Dwyka Formation dong the western margin of the Karoo Basin, South Afkica. Sedimentology, 37,23 1-246.

Von B ~ u M , V., 1977. A furrowed intratillite pavement in the Dwyka Group of Northem Natal. Trans. Geol. Soc. South f i c a 80, 125-130.

Von Enseln, O.D., 1% 8. Transportation of debris by icebergs. Journal of Geology 26,74-81.

Vorren, T.O., Hald, M., Edvardsen, M. and Lind-Hacsen, O.W., 1983. Glacigenic sediments and sedimentary environments on continental shelves: general principles with a case study from the Norwegian shelf. In: Ehlets, J., Ed. Glacial Deposits in North-west Europe. Bakema, Rotterdam, p. 6 1-73.

Wahlgrcn, R.V., 1 979. Ice-scour tracks in eastem Mackenzie Bay and north of Pullen Island, Beaufort Sea. In: Current research, part B. Geological Survey of Canada, Paper 79-18, pp. 5 1-62.

Walker, R.G. and Plint, A.G., 1992. Wave- and stoxm-dominated shallow marine systems. In: Wallcer, R.G. and James, N.P. eds. Facies Models, Response to Sea Level Change. Geological Association of Canada Pp. 2 19-238.

Weber, J.N., 1958. Recent grooving in lake bottom sediments at Great Slave Lake, Northwest Temtories. Journal of Sedimentary Petrology, 28,333-34 1.

Westgatc, J.A. Chen, F.J. and Delorme, L.D. 1987. Lacustrine ostracodes in the late Pleistocene Sunnybrook diamicton of southem Ontario, Canada Canadian Journal of Earth Sciences, 21,2330-2335.

Woodworth-Lynas, C.M.T., 1992. The Geology of Ice Scour. Unpubiished Ph.D thesis, University of Wales, 269 pp.

Waod~wrth-Lynas, C.M.T., 1996. Ice scow as an indicator of glaciolacustrine environments. In: GlaciaI Environments: Sediments, Forms and Techniques. Butterworth-Heinemann Ltd., Oxford, p. 161-178.

Wcodu-orth-Lynas, C.M.T., Seidal, M, and Day, T., 1986. Relict ice scours on King William Island, N.W.T. In: Lewis, C.F.M., Parrott, D.R., Simpkin, P.G. and Buckley, J.T., eds. Ice Scour and Seabed Engineering, Environmental Studies Revolving Funds, Report No. 039, Ottawa. p. 64-72.

Woodwrth-Lynas, C.M.T. and Dowdeswell, I.A., 1994. Sofi sediment striated surfaces and i:iassive diamicton facies produced by floating ice. In: Deynoux, M., Miller, J.M.G., L)omack, E.W., Eyles, N., Fairchild, I.J. and Young, G.M., eds., Eartti's Glacial Record. Cambridge University Press, p. 24 1 -259.

Woodworth-Lynas, C.M.T. and Guigné, J.Y., 1990. Iceberg scours in the geological record: csamples fiom glacial Lake Agassiz. In: Dowdeswell, J.A. and Scourse, J.D., eds., G laciornarine Environments: Processes and Sediments, Geological Society of London, Special Publication 53, pp. 2 17-223.

Yang, Q.S., Poorooshab, H.B. and Lach, P.R., 1996. Centrifiige modeling and numerical simulation of ice scour. Soils and Foundations, 36, 85-96.

Table 1: Pre-Pleistocene, Pleistocene and Modern ice scours. ' where cross-section geometry is observed.

Refercncc Lot.tioa Ltngth Widtb k p t ô Water Mctbod Notes

, P r c c ~ ~ i a r ~ c ~ Manin (1 965); KrCIner Namibia 3040 10-20 - Late

1 and ~ & k - (1 972) cm m Neoproterozoic Beufet al. (1971) Algeria 1 rn - Ice k t t l scour

on Silun'an transgression surface, associateci with soft sediment de formation (see same refercnce in Table 6-2)

Powell and Gostin A u d i a 1-2 m - Permian (1 990)' g laciornarine la oc ha-campos et al. Brazil c80 m 0.24.5 ~0.2 - Outcrop in ( 1 994) m m w a w of

Permian (marine).

kyles et aI. (1997) Australia 50rn 4 m - Outcrop on Coast of Permian.

PIaslocene and eady H h e icr seours Berkson and Clay Lake Superior ( 1973) Dionne (1 977) glacial Lake

Barlow- Ojibway

Dredge (1982); Clayton norrhern Lake 300- 6-40 m (2 m 7-30 m Air photo, ( 1 965) Agassiz, 1800 m direct

Canada (most observation ~ 6 0 0 rn)

.Thomas and Conne11 Scotland >10m 3 m 2 m (20 m Direct Q u m (1 985) observation exposure;

ancient scour Woodworth-Lynas et al. King William 1.8 km >1 m 120 m Direct Scour f o d ( 1986) Island, arctic observation on ancient

Canada seafloor, present elev. 1 O- 1 5 m as1

.~yles and Clark (1988) Ontario, >ZOO m 9 m 2.5 m 20 m Direct Exposure of Canada observarion ancient scour in

natural outcrop Gilbert et al. (1992) Glacial Lake ~3.57 m < 174 m cl m 10 m Air photo,

Iroquois, Ontario,

direct observation

1 Reinterpreted as log impressions by Eyles et al. (1998).

Rcfcrencc Loc8tho Leagtb Width Dcptb Watcr M d o d Notes Dcpa

.Longva and Bakkejord Norway 100sof 15-70 (Zm (40 m Direct Scour f o d ( 1 990) m t o 2 m observation on ancimt

km seafloor during a flood following brcach of an icedam

Longva and Thoresen Norway (1991) '~oodworth-Lynas and southeni Lake 6-8.5 km 50 m < 3 rn (1 10 m Air photo, Guigne ( 1990) A g s k direct - -

& d a observation This studv Onm-O, >100m IOm 4 m -20m Direct

Canada observation M o d c i i i . u r u r ~ d ~ i œ r a r r n

~ e i r 6 n ~ et a. (1 972) Alaska 1-1.5m 1.5-2 Summer rn mundinp; ice

Kovacs and Mellor Beaufort Coast c0.5 -3 m ( 1 974) m Pelletier and Shearer Beaufort Sea 8 km 10-30 4 O m 10-50 Those at depth ( 1972); Hnatiuk and m m >75 m are rclict Wright (1 983) Reimnitz et al. (19721 Alaska (20m < l m % m Ice thickness 4

m; keels to 12 m Reimnitz et al. (1972) Alaska 3 m 70 cm SiIty to sandy

bottoms Kovxs and Mellor Beaufort 1.5 m 10-30 ( 1974) midshelf m McLaren (1 975) B yarn Channel 9-15 m 0.3- 30 m SCUBA

0.6 m Shearer and Blasco Beaufort 4 0 m 15,30 Width, depth (1 975) m increase as

water deepens Wahlgren (1 979) Beaufort Sea 0-1 km IO rn <lm 25-50

m Balderson and Wilson NE Atlantic (5.5 km 20 m 2 rn 14U- Submersible Scours trend (1 973), Baldenon et al. 150 m parallel to ( 1 973) troughs;

maximum berg - draft 400 m

Harris and Joll~rnore Ncwfoundland 3 km 30 m 4.1 m 155- Sonograph Rims -1 m (1 974) continental 300 m

shelf Kovacs and Mellor Beaufort Sea 300 m 10rn 0.6-1 c80m Keels to 5 rn ( 1 974) rn draupht Kovacs and Mcllor Beaufort outer 10 rn 30-80 ( 1 974) shel f m

4 5 m van der Linden and Hamilton Bank Many cens of 4 m e 5 0 m Sonograms Keet draught Fillon (1 976) kms rn 4 0 0 m Lien (1 98 1 ) Weddell Sea 4 5 0 m ( 2 5 m (320 m Sidescan

Referencc Location Lemgtb Widtb Dcptb Water Metbod Notes

losenhans and Eastern S e v d -0 rn <20 m CIO0 m Woodworth-Lynas Canadian kms (1988); Pereira et al. Seaboard (1 985); Fader et al. ( 1982) Marienfeld (1 992); East Green land -25 200- Acoustic Dowdeswell et al. rn (up 600 m profile ( 1 994) to 12

rn) Hequette et al. (1995) Inner shelf, mean mean (10 m Bathymecry

Beaufort Sea 1 I m, 0.3 rn, to -20 profiles, ~ 3 4 5 ~ 2 . 2 m sonographs m) rn

Modern kcuslrirreprck içr .cri rim icc sceurs TyrelI (1 892) Lake c33 Direct Boulder gouges - -

Winnipqzosis paces Observation Weber ( 1 958) Great Slave 4 7 6 û m c 3 3 m < l m Ridges 1.5 m

Lake Many h i 6 ~ 3 0 m

Dionne ( 1 969) St- Lawrence 1.5 - 2 0.3 - 0.2 - shdlow Direct W a r Y bn 0.8 m 0.35 (tidaI Observation

m flats) Grass (1 984); Grass Lake Erie <6 km 10-100 -=2 m 13-25 Geophysical - - ( 1986) m m Survey Dionne (1 998) SL Lawrence a few 0.2 - 1 c0.4 shallow Direct

esniary rnctrrs m m (tidal Observation flatsl

Table II: Pre-Pleistocene, Plehtoeene and Modern hc-keel turbates. Denota outcrop description.

Prc-PIcisrucene b ù e d tuthta ' ~ y l e s et al. (1 998) New South 4 m Pennian defonned bioturbated sheIf sandstones and

Wales, mudstones bclow large ice-scour. A d i a

PI&occue iceked mrkta anvi vin (1 839) Chile "Outcrops of ullstratifled formations of mud and sand,

containing roundcd and uigular fragments of al1 sizes which has originatcd in the repeated plouging of the sea bottorn by the stranding of icebergs." Chapter XI, Voyage of thc Beagle.

'~eikie (1 896) Scotland Marine clays dcfonmd by "the groundinp; of ice-raftsw. Woodworh-Lynas Lake Agassiz Ground observations, excavations, air-photo observations. ( 1 996) Ocean Drill ing Program Antamic Drill core, palaeontoIogical samples, seisrnic. ( 1 999) Peninsula

continental shclf

Modern ice-ked nirbrm Blake (1977) Eastern Q u m gbservations of icebergs stranded on submarine moraines.

Elizabeth Island, Canadian Arctk

Vorren et al. (1983) Nowcgian About 2 rn Gravity core samples. shelf

Fischbein (1987) Alaskan Gravity cotes, sonar, seisrnic. Beaufort Sea

Josenhans et al. ( 1986); Labrador shelf Sonar, seismic, piston cc're, grab samples. Lewis et al. ( i 989) Bames and Lien (1 988) Weddell Sea Sonar, scismic, sediment sarnples.

and off Wilkes Land, Antarct ica

L Marienfeld (1 992); East Greenland >5 m Gravity cores, box corcs, seismic, x-radiography of core. Dowdeswell et al. shelf ( 1 993; 1994); Smith and Andrews (2000) Maclean ( 1997) Baffin Island Seismic

Table III: Pre-Pleistocene, Pleistocene and Modem soft sediment striateci surfaces. Note scarcity of Pleistocene examples.

Refercnct L m t k Dimcasions Notes

Clarke (1 9 1 7) New York Up to 15 cm Upper Dcvonian Portage group sandstone. State long

Rattigan (1967) A u d i a Carboniferou Kutnuig Facies; cunent ripples, varying striation orientation, ovniain by dropstone-bcaring layers

Beuf et al. (1971) Algcria Uppcr Ordovician Tamadjert Formation Savage (1 972); Visser South Africa Cartjonifcrous Dwyka Formation; curmt ripples nomial to (1 985); Visser (1 990) striation Deynow (1980) Maucïtania Plccambrian; 'chatter-marks' roughly pcrpendicular Io - - -

striations sirnitar to those in Dionne (1 972) Montes et al. (1985) Brazil 1-5 cm rclicf Late Precambnan disconforrnity surface bctwttn

glaciogenic dropstone-bearing Bebedouro Formation and underGnp: del& deposits of MO~TCI do Chapéu Formation

Aitchison et al. (1988) Transantarctic Late Paleozoic Buckeye Formation; dropstone-bearing 1 Mountains diamictites, associated with humrnocky cross-stratifi&on

indicating shallow open-wata conditions Miller (1989) Transantarctic Pemio-Carbonifcrous Pagoda Formation

Mountains Vaslet (1 990) Saudi Arabia Upper Ordovician Sarah Formation glacial paleo-vailey

mis study Ontario, 1 cm relief 1 Canada

Modern so~-udùrent i r d sur/Ùa Dionne ( 1969; 1972; St Lawrence Up to 300 m Caused by river ice; ebb currents, no boulden, rirns on 1974) %ver, Quebec long tracks

Wodvonh-Lynas Cobequid Bay, 10s of cm to Caused by pan ice during spkg break-up on tidal flats. ( 1 992) Nova Scotia 2 m wide; up

to -25 cm L deep

Table IV: Values of Cm and E for Sunnybrook and Scarborough sand used in paramehic analysis.

! I Case # Sunnybrook diamict Scarborough sand 1 Cw (kPa) E (MPa) E (MPa) ( 1 I 50 35 ! 70

70 70 - 70

1 2 1 25 35

5 1 50 70 1 70

1 3 1 4

75 35 - 50 17.5

Figures 1 and 2. Figure 1 (top): Location of Scarborough Bluffs and exposures of ice-scow structures and facies mentioned in text. Figure 2 (bottom): stratigraphy of Scarborough Bluffs along section line shown in Figure 1 .

Figure 3. A, B: Typical massive clast-poor mud facies of the Sunnybrook exposed at Cudia Park (Fig. 1 C ) . C , D, E: Clusters of ice-rafted debns within Sumybrook mud. F: Ice-rafied clasts (outlined) described as a 'subglacial clast pavement' by Hicock and Dreimanis (1992).

Figure 4. Basal contacts of the Sumybrook mud where it rests on underlying sands of the Scarborough Formation (Fig. 2). A: Sharply conformable, non-erosive contact of mud on sand. B, D: Interbedded contact showing wave-rippled sand (Scarborough Formation) draped by rnud that thickens upwards into massive mud facies (Sunnybrook); Sylvan Park (Fig. 1C). C: Defomed where sand has been injected upward into mud. Outcrop is 1 m wide (Cudia Park, Fig. 1 C). E: Deformed sediment showing undisturbed confonnable contact of mud on sand at lefi with (centre) massive mud resting conformably on 30 cm thick shear zone with prominent augen and stringers of sheared sand. Shear zone thins to right and lefi. Location: Cudia Park (Fig. 1C).

Figure 5. Promirient ice scour exposed on surface of Sumybrook mud at Cudia Park on Scarborough Bluffs (Fig. 1C for location). Compare with Figure 11C. A: Surnrner view of scour showing infill strata (Fig. 7). B: Winter view with snow covered infill of storm-deposited sands (Thomcliffe Formation; Fig. 2) and lateral bems (arrowed). Sub-scour deformation extends down into underlying sands (Scarborough Formation; Fig. 7). Note planar conformable contact either side of sub-scour deformed zone.

Figure 6 . A "Western" outcrop of ice scour stucture s h o w in Figure 5 exposed in Cudia Ravine (Figs. lC, 8). Note lateral berms (arrowed). B: Cross-section through large scour at location R2 on Rouge River (Fig. 1B). Figure for scale. Scour is huncated by modem terrace gravels.

Park * exposed river

IROQUOIS SANDS

Drnm

SUNNYBROOK

Dmm

SCARBOROUGH FORMATION

Figure 7A. Sketch of ice scour depicted detail in Figure 7B.

'Approximate extentof deformed sediment

in Figure 5 showing infill strata. Boxed area show

Figure 78. Nature of sub-scour deformutioii (close-up of boxcd area shown in Figure 7A).

I Figure 7.B

Figure 9. Ice-scoured surface on upper Sunnybrook exposed at Sylvan Park as mapped in 1985; inset shows situation in 1995 aRer some 8 m of cliff retreat. Note prominent sand-filled ice scours ai 1,2 and 3 and presence of ice-keel turbated muds and sands below scours 1 and 2. Storm-deposited hummocky and swaley-cross stratified sands occur in erosively-based beds up to lm thick recording erosion and deposition during successive storms. Compare with Figures 1 I A, B.

Figure 10. Ice-scoured topography on uppermost Sunnybrook muds; outcrops likely provide a section through an extensive scoured lake floor (e.g., Fig. 1 ID). A, B, C: Sylvan Park (Fig. IC). D: Rouge Valley (site R3; Fig. 1B). E, F: Rouge Valley (site R1; Fig. 1B).

Disiance (kilomenes)

Figure 1 1 . A, B: Modem occurrences of ice-scoured seabed topography (from Atlas of Marine Acoustic Images). C: Cross-section through modem ice scour (from King and Gillespie, 1986). Compare with Figures 5,6,7. D: Floor of former Glacial Lake Agassiz in Manitoba, Canada, showing intersecting ice scours. Area shown is roughly 2 km across (Governmeni o f Canada Air Photo # A 15954-79, Central Manitoba).

.P- 2

Assumed water surface

Minimum slip s w h a . wheft factor of a f c l y = 1 .O

Figure 12 A. Minimum slip surface in SLOPEN analysis.

O o w w a d force

Figure 12 B. Finite element mesh and boundary conditions for numerical mode1 of ice scour. C. Mode1 results showing contours of shear stress.

Downward displacement (mm)

O 200 400 600 800 1000

Downward displacement (mm)

Figure 12 D: Muence of Sunnybrook properties on calculated displacement; E: Influence of Scarborough sand properties on calculated displacement. In D, each point is actually a cluster of points representing the range of assumed Swuiybrook properties (Table IV). The :ight clustering shows how variation in Sunnybrook properties has little effect on calculated displacement. In E, error bars for le2 and Se3 show the variation in calculated displacernent fiom the range of assumed Scarborough sand properties (Table IV).

Ice turbated mud and Sand

Bem debris reworked by

Hummocky (Shc) swaley cross- stratified sand (Ssc) rapidly- deposited on ice-scoured su bstrate

Figure 13. Model for ice-scoured topography seen on upper surface of Sunnybrook A. Formation of multiple scours on mud and generation o f ice-keel turbate by ploughing o f sand into mud. B. Blanketing of scoured topography by storm-deposited hummocky and swaley- cross-stratified sands. Lowennost sands are comrnonly deformed by rapid deposition over irregular substrate. Note ice-rafted debns in storm-deposited sands together with rnud debris reworked from adjacent high-standing berrns.

Figure 14. Soft-sediment striations cut across mud at Highland Creek (Fig. 1 A). A. Outcrop view. B/C. Oblique and vertical views o f striated surface on removed block showing smaii-scale micro-ridges oblique to trend o f striations.

APPENDIX A: SIGMA/W FINITE ELEMENT NUMERICAL MODEL

This appendix is a brief summary of theory behind the SIGMAlW program, drawn fiom program documentation (Geoslope International, 1997).

SIGMAN is formulated for either two-dimensional plane strain or axisymmetric problems using small displacement, small strain theory. Confomiing to conventional geotechnical engineering practice, the "compression is positive" sign convention is used. In the discussion below, the bracket sets < >, ( ), and [ ] are used to denote a row vector, a column vector and a matrix, respectively.

The finite element equation used in the SIGMAN formulation for a given time increment is

J [ B ] ' [ c ~ B ) ~ ((a) - bJc m'du + p&c Ab'dA + (F,)

where: [BI = strain-displacement matrix [Cl = constitutive matrix {a) = column vector of nodal incremental x- and y-displacements A = area along the boundary of an element v = volume of an elexnerit b = unit body force intensity cN> = row vertor of interpolating fiinctions P = incremental surface pressure {F,, ] = concentrated nodal incremental loads

Sumrnation of this equation over al1 elements is implied. SIGMA/W is formulated for incrementai analysis. For each time step, incremental displacements are calculated for the incremental applied load. These incremental values are then added to the values fiom the previous time step. The accumulated values are reported in the output files. Using this incremental approach, the unit body force (Le., the weight of an element) is only applied when an element is included for the f h t time during an analysis.

For a two-dimensional plane strain analysis, SIGMA/W considers al1 elements to be of unit thickness. For constant element thickness, t, Equation (A.1) can be written as:

in an abbreviated form, the finite elernent equation is,

where : [KI = element characteristic (or stifkess) matrix

which for plane strain is

{a) = nodal incremental displacements (F) = applied nodal incremental force which is made up of the following: (Fb) = incrementd body forces {Fs) = force due to surface boundary incremental pressures

which for plane strain is

(FnJ = concentrated nodal incremental forces

SIGMAiW solves this tùiite element equation for each time step to obtain incremental displacements and calculates the resultant Uicremental stresses and strains. It then sunis d l these increments since the first time step and reports the sunmed values in the output files. These output files were used to generate the contours of shear stress shown in Figure 12 C and the calculated downward displacement values shown in Figures 12 D and E.

The input parameters required for the SIGMAMT mode1 used in this study were undraineci shear strength Cu, Young's modulus E, unit weight, and Poisson's ratio. The values used are reported in Chapter 4.

SIGMAIW uses a displacement convergence vector n o m as a measure of how much dispiacement is calculated in a given iteration, relative to the cumulative displacement up to that iteration. In this study, the SIGMA/W calculation was set to iterate ten times or until the displacement convergence vector norm was less than 1%. The solutions converged to a vector norm of a few percent or less, which is judged to be acceptable based on guidelines given in (Geoslope International, 1997).