Hydraulic-Jump and Hyperconcentrated-Flow Deposits of a

JOURNAL OF SEDIMENTARY RESEARCH, VOL. 73, NO. 6, NOVEMBER, 2003, P. 887–905Copyright q 2003, SEPM (Society for Sedimentary Geology) 1527-1404/03/073-887/$03.00

HYDRAULIC-JUMP AND HYPERCONCENTRATED-FLOW DEPOSITS OF A GLACIGENIC SUBAQUEOUSFAN: OAK RIDGES MORAINE, SOUTHERN ONTARIO, CANADA

H.A.J. RUSSELL* AND R.W.C. ARNOTTDepartment of Earth Sciences and Ottawa-Carleton Geoscience Centre, University of Ottawa, Ottawa, Ontario, K1N 6N5, Canada

e-mail: [email protected]

ABSTRACT: The Oak Ridges Moraine in southern Ontario is a poly-genetic moraine constructed of a number of coalesced deposits of gla-cifluvial and glacilacustrine origin. A detailed study of the facies ar-chitecture has been completed on a series of pit sections extending ;300 m subparallel to the paleoflow direction. Eight major lithofaciesand five facies associations have been described. These data have beeninterpreted to be upper-flow-regime hyperconcentrated-flood-flow de-posits emplaced under a regime of rapid flow expansion and loss oftransport capacity within a plane-wall jet with an associated hydraulicjump. Deposition from the plane-wall jet with jump occurred in threezones of flow transformation: zone of flow establishment, transitionzone, and zone of established flow. Massive gravels with unconsolidatedsand intraclasts and open-work gravel / gravel–sand couplets were de-posited in the zone of flow establishment by hyperconcentrated andsupercritical flows, respectively. Immediately downflow low-anglecross-stratified sand incised by steep-walled scours infilled by diffuselygraded sand define the transition zone, the zone of maximum vortexerosion, and the distal limit of deposits emplaced under upper-flow-regime conditions. These strata record rapid bed aggradation fromsediment-laden supercritical flows that episodically were scoured bylarge vortices generated within migrating hydraulic jumps. Strati-graphically upward and downflow strata consist only of lower-flow-regime sedimentary structures. Medium-scale, planar cross-strata andsmall-scale cross-lamination related to migrating 2-D dunes and cur-rent ripples, respectively, characterize the zone of established flow. Thefacies and sediment architecture suggest that this fan was depositedduring a relatively short period of time (days, weeks) by energetic sed-iment-laden floods.

INTRODUCTION

Sedimentary deposits formed in areas of flow expansion are of significantacademic and economic interest. The most extensively studied types ofthese deposits are alluvial fans, crevasse splays, deltas, tidal-channel deltas,and submarine fans. All of these deposits form important hydrocarbon res-ervoirs. Glacigenic, ice-contact subaqueous fans (grounding-line fans), onthe other hand, have received much less attention, partly because they havelimited hydrocarbon reservoir potential. Rather, their principal economicvalue is as groundwater aquifers or for aggregate mining. Nevertheless,many similarities in sedimentary processes and subsequent deposition existamongst all four depositional environments despite major differences ingeographic setting, sediment source, and mode of sediment delivery (e.g.,Reading 1996).

Each of the deposits outlined above commonly have sedimentary struc-tures and facies created by high-energy flows and hyperconcentrated flows.Accommodation space permitting (subaqueous case), the flow expands anddecelerates initiating deposition. In the case of sediment-charged flows,common to glacigenic settings, rapid bed aggradation occurs. If the dis-charge flow is supercritical, a hydraulic jump occurs as the flow transformsfrom a supercritical to a critical state. Hydraulic jumps have been used toexplain flow transitions, bedding styles, and scour geometries by research-

* Present address: Geological Survey of Canada, 601 Booth Street, Ottawa, On-tario, K1A 0E8, Canada.

ers in a range of sedimentary environments (Komar 1971; Carling 1995;Weirich 1989), flume experiments (Garcia 1993; Kubo and Yokokawa2001; Mulder and Alexander 2001; Hand 1974), and sedimentary deposits(Skipper and Bhattacharjee 1978; Gorrell and Shaw 1991; Brennand 1994;Brennand and Shaw 1996). Deposits commonly reported to be associatedwith hydraulic jumps are antidunes (e.g., Hand 1974; Skipper and Bhat-tacharjee 1978), climbing ripples and dunes (e.g., Daub 1996), and steep-walled scours and diffusely graded sand (e.g., Gorrell and Shaw 1991).Nevertheless, the sedimentological signature of hydraulic jumps remainspoorly understood, particularly for jumps at high Froude numbers (. 2.5).

Objectives

This paper describes a small glacilacustrine subaqueous fan deposit thatforms part of the Oak Ridges Moraine in southern Ontario. The site is thelocation of an active quarrying operation that has generated new sectionsthrough the deposit, providing a unique opportunity to identify the three-dimensional relationships of the sediment facies. The sediment facies areinterpreted to be deposits of high-energy flows, including supercriticalflows, hydraulic jumps, and hyperconcentrated dispersions. These interpre-tations are well constrained by downflow and vertical sediment facies tran-sitions and the facies architecture. The depositional processes are inter-preted within the context of a model of a plane-wall jet with jump efflux.The dynamics of plane-wall jet with jump systems are relatively well un-derstood from experimental studies; however, only a few sedimentary stud-ies have explicitly applied the plane-wall jet with jump model (e.g., Gorrelland Shaw 1991; Powell 1990). For this reason we review the jet-effluxmodel at the beginning of the paper. The assemblage of facies are inter-preted to have been deposited within a single meltwater season from asingle outbreak flood or jokulhlaup.

Glacigenic Subaqueous Fans

Glacigenic subaqueous fans were first described in the Ottawa area fromdeposits of the Pleistocene Champlain sea (Rust and Romanelli 1975; Rust1977) and subsequently have been described from numerous moraine (e.g.,Cheel and Rust 1982; Postma et al. 1983; Cheel and Rust 1986; Burbidgeand Rust 1988; Rust 1988; Sharpe 1988; Lonne 1995; Hunter et al. 1996;Paterson and Cheel 1997; Plink-Bjorklund and Ronnert 1999; Lonne 2001),esker (e.g., Banerjee and McDonald 1975; Diemer 1988; Henderson 1988;Gorrell and Shaw 1991; Brennand 1994; Spooner and Dalrymple 1994;Lajeunesse and Michel 2002), and modern settings (e.g., Powell 1990). Forthe most part these studies have focused on proximal deposits developedwithin the zone of inertial jet efflux into the basin or in buoyant plumeupwelling glacimarine settings. These studies have provided a wealth ofinformation on the spectrum of lithofacies, downflow lithofacies transitions,and stratigraphic architecture of glacigenic subaqueous fan deposits (Fig.1). The deposits have been interpreted in terms of rapid flow expansionand loss of transport competence at the ice margin as the flow evolvedfrom confined conduit flow to either an open or ice-covered aqueous basin.Poorly understood elements of many of these fan deposits are steep-sidedscours filled with diffusely graded sand or massive sand. Earlier workersinterpreted these deposits to be gravity-flow deposits (Postma et al. 1983;Burbidge and Rust 1988; Rust 1988) or the product of scour erosion andrapid sedimentation beneath a hydraulic jump (Gorrell and Shaw 1991).

888 H.A.J. RUSSELL AND R.W.C. ARNOTT

FIG. 1.—A) Composite vertical stratigraphicsections developed from literature cited in thetext. B) Probable association of sections andfacies with different jet-efflux models, listedfrom proximal to distal.

The correct interpretation of the scour and infill sand facies association hassignificant implications for the jet-efflux model (e.g., plane-wall jet, plane-wall jet with jump) used to explain local depositional conditions and theintegration into more regional-scale paleoenvironmental reconstructions.

Jet-Efflux Model

Glacigenic subaqueous fan deposits are well explained by the jet-effluxmodel, which, depending upon the position of efflux into a water column,can be an axisymmetric jet, a plane jet, or a plane-wall jet (Figs. 2, 3; Bates1953; Powell 1990). The plane-jet model was first applied to deltas (Bates,1953). In plane jets the rate of lateral spreading decreases downstream suchthat the expanding efflux develops a parabolic shape with width increasingat about three times the square root of the distance measured downstreamfrom the river mouth (Bates 1953). For subaqueous fans with flow exitingalong a basin floor the jet is more appropriately modeled as a plane-walljet (Launder and Rodi 1983; Powell 1990). The plane-wall jet has a muchgreater ratio of lateral spreading to longitudinal distance than the plane jetand will form a more semicircular or broad triangular shaped deposit (Fig.4; Wright 1977). Additionally, the plane-wall jet has a lateral to verticalspreading ratio of between 5 and 9, and as a consequence produces a broadlow-profile efflux and a streamwise velocity that decays as the inversesquare root of the downflow distance (Launder and Rodi 1983; Wu andRajaratnam 1995). An additional type of jet can be recognized when theefflux flow is supercritical and a hydraulic jump forms basinward of theefflux point (e.g., Rajaratnam and Subramanyan 1986). Hydraulic jumpsrecord the conversion of kinetic energy of the supercritical flow (Fro . 1)

to potential energy of the subcritical flow (Fro , 1) as the flow slows andthickens (Fig. 3). This transition is commonly defined by the densimetricFroude number,

UFr 5 (1)o

(d 2 d )o ag y! da

where U is the mean velocity, do the jet density, da the basin water density,y the flow depth, and g the acceleration of gravity (9.81 m s21). The stream-wise velocity decay for submerged jumps is similar to free jumps or plane-wall jet depending upon the parameter given by S*:

S* 5 12Fo21.3 (2)

(Wu and Rajaratnam 1995). When S . S*, where

S 5 (yt 2 y2)/y2 (3)

and yt the basin depth and y2 the subcritical sequent depth, the velocitydecay is similar to a plane-wall jet and when S , S* it is similar to a freejump. The rate of velocity decay in a free jump is substantially faster thanfor a plane-wall jet. Literature on the lateral spreading ratio of the plane-wall jet with jump, however, is sparse; but it is likely similar to or greaterthan the plane-wall jet given the similar or increased streamwise velocitydecay (Fig.4).

For the plane jet, the plane-wall jet, and the plane-wall jet with jumpthe intrusion and subsequent assimilation of the efflux jet into the basin

889HIGH-ENERGY HYPERCONCENTRATED GLACIGENIC SUBAQUEOUS FAN SEDIMENTATION

FIG. 2.—Illustration of four styles of jet efflux,A) plane jet, B) axisymetric jet, C) plane-walljet, D) plane-wall jet with jump. In part D thearrow indicates position of hydraulic jump.Drawing by John Glew.

FIG. 3.—Streamwise cross section showing stages of flow evolution from conduitto basin for A) plane-wall jet and B) plane-wall jet with jump. Variables: x is thestreamwise distance, yo is conduit diameter, y2 subcritical sequent depth, yt basinwater depth, Uo is maximum conduit velocity, and Um is the maximum efflux ve-locity.

evolves through three stages (Bates 1953; Long et al. 1990; Powell 1990).For each of the three kinds of jets these three stages are characterized bydifferent processes and therefore are commonly described by different ter-minology. For the plane jet and the plane-wall jet the three zones of flowevolution are commonly termed: (i) zone of flow establishment, (ii) tran-sition zone, and (iii) zone of established flow (Figs. 3, 4). These zones arebest defined for the plane jet but can be applied loosely to the other jettypes. For plane-wall jets the flow is fully established downflow of theefflux point at:

x/yo $ 15 (4)

where x is the streamwise distance and yo is the conduit diameter (Figs. 3,4). A major difference in the plane-wall jet with jump is the transition zone(developed zone), which is equal to ; 85% of the roller length for aturbulent hydraulic jump (Long et al. 1990), and it is in this reach that theflow thickens substantially (Rajaratnam and Subramanyan 1986) and de-creases in velocity. Within each of these three zones, which are commonlysimplified to a proximal-to-distal continuum, distinct lithofacies are depos-ited (Fig. 1). These lithofacies record sedimentation from the evolving jetas it expands and entrains ambient basin water.

STUDY AREA AND GEOLOGICAL SETTING

The study focused on a site on the north side of the western Oak RidgesMoraine located 15 km east of the Niagara Escarpment and 60 km north-west of Toronto, Ontario, Canada (Fig. 5). The Oak Ridges Moraine is aregional physiographic feature that extends for 160 km from the NiagaraEscarpment eastward to the vicinity of Trenton (Chapman and Putnam1984) and is up to 20 km wide and up to 160 m thick (Barnett et al. 1998).The western Oak Ridges Moraine consists predominantly of gravel, sand,and silt with less than 2% interbedded diamicton and clay (Russell 2001).

The ORM has been interpreted to consist of glacifluvial and glacilacus-trine sediment deposited in four stages: (i) subglacial, (ii) subaqueous fan,(iii) subaqueous fan/delta, and (iv) ice-marginal (for details see Barnett etal. 1998). These four stages are interpreted to record extreme changes inthe magnitude of meltwater discharge from regional subglacial reservoiroutbreak floods (Barnett et al. 1998; Brennand and Shaw 1994) to seasonal,climate-modulated melt and/or precipitation events (Gilbert 1997). Onlythe third stage, represented by the upper 20–60 m of the moraine, was


FIG. 4.—A) Schematic plan view of development of inertia-dominated plane-walljet into basin with velocity profile in flow shown by arrows (modified from Wright1977). B) Velocity profiles in plan and vertical for a three-dimensional plane-walljet (simplified from Launder and Rodi 1983). Equation defines the streamwise ve-locity decay (Rajaratnam 1976). Variables: x is streamwise distance, yo conduitdiameter, Uo is maximum conduit velocity, and Um is maximum efflux velocity.

deposited within a glacilacustrine setting dominated by subaqueous fansedimentation (Barnett et al. 1998). The ice-supported glacilacustrine basin(water surface ; 420 m asl, depth up to 200 m) was bounded to the westby the Niagara Escarpment (Chapman and Putnam 1984). This ice-sup-ported basin was , 40 km wide at the escarpment and narrowed eastward.This study investigated sediment deposited during the final stage of mo-raine formation.

Vertical sections 6–10 m high and ; 250–300 m long, oriented sub-parallel to the mean paleoflow direction (east–west), were measured in anactive aggregate pit in the period 1993–1998. In addition, a number ofshorter sections oblique and transverse to flow were measured (Fig. 5).Field observations of texture, bedding contacts, sedimentary structures, etc.,have been grouped into facies (e.g., Miall 1996). The study of the depositarchitecture is based on photo-mosaics (e.g., Miall 1996).

SEDIMENTOLOGY

Measured sections consist mostly of sand and gravel overlain by 1–3 mof laminated-massive silt that had been stripped back from the pit face bythe pit operators (Fig. 6). From east to west, grain size changes very rapidlyfrom gravel to sand. The sections generally fine upward. Only minoramounts of diamicton ware observed at the study site. From these exposedsections, eight principal facies have been identified (Table 1). Facies aredescribed from highest energy to lowest energy, which for the most partcorresponds to the proximal to distal or stratigraphically upward change offacies.

Poorly Sorted Gravel With Intraclasts (Gd)

The poorly sorted gravel facies consists of matrix- to clast-supported,poorly sorted gravel (Figs. 7, 8) and diffusely graded sandy gravel. Com-mon to these deposits are silty sand intraclasts up to 3 m long (Fig. 9A,B). The facies is up to 3 m thick, and sharply overlies the sand and cross-bedded gravel facies. In a streamwise direction the gravel typically gradesinto plane-bedded and cross-bedded sandy gravel and coarse sand (Fig. 7).Strata are generally poorly bedded or massive, with minor interbeds ofplanar-stratified gravel facies, particularly toward the top of the fan deposit(Fig. 9C). The dominant clast size is pebbles with minor cobbles. Diffuselygraded sandy gravel was observed at one location with a slump-coveredlower contact. The bed fined upward to coarse sand and the matrix wasreversely and normally graded. Local grain-size clusters are matrix sup-ported (Fig. 9B). Boulder-size intraclasts of fine and medium sand have asubparallel-to-bedding fabric and are low in the section (Fig. 9A, B). Thelarger intraclasts are mostly enveloped in a sand-rich matrix that coarsensaway from the intraclast (Fig. 9B). These finer-grained envelopes weredisplaced toward the west (downflow) along the upper margin of the coreintraclast. Upward in the facies the size and abundance of intraclasts gen-erally decreased. Clast fabric of these small intraclasts is random near largerintraclasts but with distance becomes more organized and planar.

Interpretation.—The poorly sorted gravel and diffusely graded sandygravel are interpreted to be hyperconcentrated-flow deposits. Hyperconcen-trated flows or traction carpets are commonly stratified and have been in-terpreted to consist of a subordinate basal hyperconcentrated frictional layerand an overriding dilute fully turbulent collisional layer. The thickness ofeach zone depends on variations in applied shear stress, grain size, anddownward sediment flux (Sohn 1997). Flows with a thick collisional layerform in coarser-grained flows and deposit well-sorted, distinctly gradedstrata. In contrast, a thicker frictional layer produces more massive deposits.Sediment is interpreted to have been supported by a number of mecha-nisms, including shear supplied by the overlying flow, dispersive pressure,buoyancy, hindered settling, and turbulence (Costa 1988; Smith and Lowe1991). In gravel, dispersive pressure and hindered settling were the primarysupport mechanisms. In contrast, in the sand-rich gravel, dispersive pres-sure was less significant because of the smaller mass of individual grains(Lowe 1976) and buoyancy and hindered settling were likely more impor-tant support mechanisms.

Hyperconcentrated traction-carpet deposits aggrade grain by grain (Sohn1997) and, depending on flow dynamics, matrix-supported or framework-supported gravel was deposited. Massive or normally graded, poorly bed-ded, matrix-supported gravel has been interpreted previously to be deposits


FIG. 5.—A) Location of study area in southern Ontario and areal extent of exposed western Oak Ridges Moraine sediment. B) Location of sections plotted on an aerialphotograph of the pit in 1991 (photo from Ontario Hydro). Paleoflow data presented as frequency percent: outer circle is 10 percent. Only measurements from cross-bedsare plotted.

of hyperconcentrated flows (Lowe 1982; Smith 1986). Less commonly,clast-supported gravel has been interpreted to be hyperconcentrated-flowdeposits (Dorsey and Falk 1998; Russell and Knudsen 1999). Variations inclast support are most likely related to differences in traction carpet dy-namics due to changes in sediment flux, suspended sediment concentration,support mechanisms, etc. A number of characteristics differentiate thesandy gravel deposits from the coarser poorly sorted gravel. In the sandygravel deposits, intraclasts are enveloped by finer sediment, there is abroader range of intraclast sizes, matrix grading is better developed, andpebble intraclast fabric is more disorganized (Fig. 9D). Deposition fromnon turbulent, high-concentration flows is indicated by the preservation ofpoorly consolidated intraclasts and grading in intraclast matrix haloes. Thematrix grading and downflow-displaced matrix haloes of the intraclastssuggests that deposition was not as a rigid plug but instead incremental.Increased frictional strength and the consequent increase in resistance toflow was possibly associated with disaggregation of sand intraclasts and acommensurate increase in local sediment concentration (e.g., Postma et al.1988). Similar facies from other subaqueous fan deposits have been inter-preted to be gravity flows (Rust 1988) or hyperconcentrated flows (Bren-nand and Shaw 1996).

Erosion of large unconsolidated friable sand intraclasts was probablyrelated to scouring in a hydraulic jump or scouring at the base of the flowand then followed by rapid deposition. The rapid dissipation of flow energyand generation of high volumes of entrained sediment some distance down-flow of a hydraulic jump would have developed stratified flows, causingrapid sedimentation. Sand intraclasts are commonly observed in esker (e.g.,Brennand 1994), subaqueous fan (e.g., Gorrell and Shaw 1991) and joku-lhlaup outwash deposits (Russell and Knudsen 1999) and have been attri-buted to scour within a hydraulic jump. Less commonly, such intraclastsoccur in fluvial settings, where they may originate by slumping along chan-

nel margins (e.g., Martin and Turner 1998) or by erosion from the channelbottom (Wan and Wang 1994).

Planar-Stratified Gravel (Gh)

The planar-stratified gravel facies occurs as laterally extensive tabularsheets 0.5–2 m thick overlying, and less commonly interbedded with, thecross-bedded gravel facies (Figs. 8, 10). It is most extensive in the vicinityshown in Figure 8, where it was traced for 8–10 m parallel to flow. Itgrades upflow into the poorly sorted gravel facies (see below) and down-flow is interbedded with or grades into low-angle cross-stratified or troughcross-bedded sand. Basal bed contacts are planar to undulatory with scours, 10 cm deep. Scours are generally infilled by coarse bimodal gravel. Thefacies varies from open-framework unimodal pebble beds to sandy pebblebeds, commonly forming upward-fining couplets (Fig. 10). Open-frame-work gravel beds are generally 1–5 cm thick. Sandy pebble beds are 2–10cm thick, more poorly sorted, and more laterally continuous than open-framework beds. The maximum observed clast size is ; 8 cm. Clast clus-ters occur locally and commonly have a cobble-size clast at the core withpebble-rich and sand-rich stoss and lee deposits, respectively. Stoss-sidedeposits are characterized by imbricate pebbles with a medium sand matrix,whereas lee-side deposits are finer grained and have a more disorganizedfabric. Clast imbrication is predominantly a(t), b(i) to flow and indicatespaleoflow toward the northwest.

Interpretation.—The upward-fining couplets, open-framework beds,clast clusters, transverse clast fabric, and an absence of cross-bedding areall considered to be evidence for deposition from a turbulent high-energyflow (e.g., Collinson and Thompson 1989). These characteristics have beenattributed to antidune flow conditions under both supercritical (Blair andMcPherson 1994; Iseya and Ikeda 1987; McDonald and Day 1978) and


FIG. 6.—Fence diagram of principal sections and representative stratigraphic logs (see Fig. 5 for section location). Fence diagram is coded by facies associations.Horizontal and vertical scale are approximate. Vertical scale may change by ; 3 m over the length of a single fence panel.

TABLE 1.—Summary of subaqueous-fan sediment facies described in this study.

Facies Facies CodeBest Exposes

in Figure CharacterThickness

(m) Interpretation

poorly sorted gravel Gd 7 normally and reversely graded, clast clusters, sand intraclasts 2–3 hyperconcentrated dispersions within the ZFEand ZFT, from axial flow and beneath a hy-draulic jump

planar stratified gravel Gh 8 sharp-based graded beds, heterogeneous to well-sorted, open frame-work to closed framework

2 from bed-load sheets flows within the ZFE

cross-bedded gravel Gt, Gp 7, 8 sharp-based, trough cross-bedded and planar cross-bedded, sandypebble to pebble gravel, rare faults

3 2-D and 3-D dunes and scour fill within the ZFE

low-angle cross-stratified sand Sh 12, 13 sharp-based, minor undulation, horizontal to low-angle dip 3–5 beneath in-phase waves of an undular hydraulicjump and upper-flow-regime conditions, withinZFT

diffusively graded/massive sand Sd 12, 14 sharp-based irregular scours, massive to bedded, amalgamated beds,sand intraclasts, water-escape structures

3 from hyperconcentrated dispersions beneath a hy-draulic jump, within ZFT,

cross-bedded sand St, Sp 15 sharp-based, planar-tabular (Sp) and trough (St) cross-stratified, me-dium to coarse sand, minor pebbly sand, rare convolute bedding

5–7 by 2-D and 3-D dunes, within the ZFE and ZFTand axial ZEF,

small-scale cross-laminated fine sand Sr 16 sharp-based, stoss-erosional to stoss-depositional, minor micro-faults 5–8 under depletive flow and combined suspensionbed-load transport, within the ZEF

silt-clay F basal contact sharp, fining upward, micro-laminate to beds ,5 cmthick,

1–3 basinal underflow and suspension sedimentation


FIG. 7.—A) Photo mosaic and B) line drawing of section. Section is subparallel and ; 10 m south of section in Figure 8. Note large intraclast in the poorly sortedgravel and rapid downflow transition to interbedded planar-bedded and cross-bedded medium sand; flow was from left to right. Note rising and thickening beds downflowdue to rapid bed aggradation (thick arrows). Shovel is 1 m long. Facies codes are defined in Table 1.

critical flow conditions (Koster 1978). Modern alluvial-fan gravel coupletswith basal open-framework beds have been shown to represent depositionunder high-energy antidune conditions of ; 1.5 , Fro , 1.7 (Blair 1999).To account for the absence of cross-strata and the abrupt upward grading,Blair (1999) invoked an autocyclic mechanism involving oversteepeningof the surface wave, rapid breaking of the curling wave crest, followed byrapid downslope shooting (washout) of water. During washout, large vol-umes of sediment are momentarily suspended and the antidune structure ispartially or completely eroded. Subsequent rapid sedimentation forms thefiner-grained upper portion of each gravel couplet. Clast clusters are pos-sibly transverse ribs that form under upstream-migrating hydraulic jumps(McDonald and Day 1978) or are relics of antidunes formed under in-phasewaves (Koster 1978). The transverse ribs are marked by the largest clastsand consequently are interpreted to have formed beneath breaking standingwaves when finer sediment was transported downflow. The succession ofgravel couplets is interpreted to have been deposited by a single flow event

and is consistent with modern alluvial-fan deposits. For example, on analluvial fan in Colorado a gravel facies 5 m thick, and consisting of 15couplets, was deposited from a single flash flood that lasted only 5 hours(Blair 1987).

Cross-Bedded Gravel (Gt, Gp)

The cross-bedded gravel facies forms tabular deposits 1–3 m thick (Figs.7, 8) that overlie a sharp, planar to scoured lower contact truncating low-angle cross-stratified or dune-scale cross-bedded sand. The facies consistsof normally graded, medium-scale cross-bedded, bimodal pebble to sandypebble gravel (Fig. 11). Trough cross-bedded gravel is more poorly sortedand is generally thicker, forming sets 0.5–1 m thick (Fig. 11A). The max-imum clast size is ; 10 cm and averages ; 1–3 cm. Planar cross-beddedbimodal gravel forms tabular sets 4–10 cm thick and cosets 25–30 cm thick.


FIG. 8.—A) Photo mosaic and B) line drawing of section. Lower, poorly exposed beds consist of cross-stratified gravel and minor planar-stratified gravel. The upper leftside consists of poorly bedded gravel overlain by cross-stratified sand (scale is 1.5 m at arrows). The right side of the face consists of cross-stratified sand and diffuselygraded and massive sand infilling a scour into small-scale cross-laminated fine sand. Flow was from left to right and is approximately parallel to section face. Facies codesare defined in Table 1.

Cross-beds dip at 20–258, are well sorted, and are normally graded (Fig.11B). Single sets are typically mantled by a fine silty sand.

Interpretation.—This facies was deposited by migrating two- and three-dimensional subaqueous dunes (e.g., Harms et al. 1975) or bars (e.g.,Boothroyd and Ashley 1975). Normally graded cross-stratification reflectsgravity sorting during slipface avalanching.

Low-Angle Cross-Stratified Sand (Sh)

Cropping out exclusively at the locality shown in Figure 12, the low-angle cross-stratified sand facies is up to 3.5 m thick (Fig. 6). The faciesis interbedded with diffusely graded or cross-bedded sand, and it gradesupflow into sandy gravel. It consists predominantly of well-sorted, mediumsand with minor coarse sand (Fig. 13) and minor moderately sorted pebblysand. Strata are , 1 cm thick, are graded or massive, form cosets , 40cm thick but locally up to 1.5 m thick, and overlie parallel or low-angle(, 158) erosion surfaces (Fig. 13). Low-angle dipping strata onlap ero-sional surfaces toward the east. Within a coset, laminae dip at a low angleand are parallel or diverge slightly. Laterally, strata are continuous for upto 8–10 m and pinch and swell along strike. Massive sand interbeds 2–3cm thick occur locally. Coarse sand occurs as discontinuous laminae 30–40 cm long within medium sand or fill shallow scours that are , 2 cmdeep and 5 cm long. Larger scours are 20–30 cm deep, 1–2 m long, and

filled with massive, medium sand. Locally, low-angle downflow- and up-flow-dipping laminae are preserved.

Interpretation.—On the basis of stratal geometry, onlap and dip direc-tion, normal grading and changes in bed-contact conformity, two deposi-tional processes are interpreted. Where strata are planar, normally graded,and conformable to the underlying bedding surface, deposition under up-per-flow-regime, plane-bed conditions is interpreted (e.g., Cheel et al.1990). Conversely, where strata are bounded by low-angle erosion surfaces,and laminae dip at shallow angles toward the inferred paleoflow direction,strata are interpreted to indicate antidune flow conditions. Both the plane-bed and low-angle strata were deposited from high-energy fluidal flows.Only the low-angle strata are discussed further.

The combination of low-angle cross-stratification dipping , 108 up-paleoflow, a decrease in dip angle that becomes asymptotic to the set base,and a paucity of high-angle cross-stratification suggest that deposition oc-curred under supercritical flow conditions (Fro .1) and migrating in-phasewave conditions (Alexander et al 2001; Cheel et al. 1990; Middleton 1965).In-phase waves probably developed within an undular (1 , Fro , 1.7) ora weak (1.7 , Fro , 2.5) hydraulic jump when a supercritical flow un-derwent rapid flow expansion (e.g., Chow 1959). Unsteady flow conditionsmay have fluctuated between the undular and weak hydraulic jump and asa result changed the local position of the hydraulic jump and, consequently,


FIG. 9.—Poorly sorted gravel facies. A) Large unconsolidated sand intraclast inmassive gravel. Shovel handle is 1 m long. B) Unconsolidated sand intraclast insandy gravel with fine matrix haloes, subvertical pebble-size intraclasts. Knife is 23cm long. C) Poorly bedded gravel showing local clast clusters and upward changein bedding style. Increments on scale are 10 cm. D) Close-up of clast cluster outlinedin part C. Flow was from left to right in all photos. Coin is ; 2 cm diameter.

FIG. 9.—Continued.

the location of erosion and deposition. Alternatively, stationary or migrat-ing in-phase waves may have evolved, in response to bedform aggradation,into breaking waves (Kubo and Yokokawa 2001). Massive beds were de-posited rapidly from hyperconcentrated flows. In fluvial environmentsbreaking in-phase waves have been reported to produce near-bed suspendedsand concentration of up to 80% (Simons et al. 1965). Such elevated sed-iment concentrations may have momentarily suppressed local fluid turbu-lence and deposited massive beds.

Diffusely Graded and Massive Sand (Sd)

Diffusely graded sand is exposed predominantly at locations shown inFigures 8 and 12 and abruptly overlies and is interbedded with low-anglecross-stratified sand. Locally it overlies small-scale cross-laminated sand.It fills irregular scours that are up to 3 m deep, have margins dipping upto 608, and locally overhang. Strata occur also in sharp-based tabular beds5–40 cm thick. Sediment ranges from medium sand to granular coarse sand.Granules and isolated clay and silt intraclasts occur only in scour fills (Fig.14A). Strata are massive or diffusely graded where sand is faintly planarstratified. Within the scour fill, beds are , 10 cm thick and average 5 cmthick. Stratification is most clearly defined toward the scour-fill margin andupward in the fill (Fig. 14). Locally, subhorizontal undulatory stratificationis traceable for tens to hundreds of centimeters until it grades into faintlystratified or massive strata. Thicker fills consist of multiple erosional sur-faces that are curvilinear and have apparent dips of up to 708.

Interpretation.—The steep-walled scours with locally overhanging mar-

gins and the fill of diffusely graded or massive sand of this facies areinterpreted to represent rapid deposition from sediment-charged flows. Al-though both scour-fill and tabular facies of diffusely graded sand are in-terpreted to indicate rapid deposition, there are important differences.Scour-fill units are interpreted to have been deposited immediately belowor downflow of a hydraulic jump. Conversely, tabular units are interpretedto have been deposited under high-energy, possibly upper-plane-bed flowconditions. Scours in unconsolidated medium sand with margins dippingat up to 608 and with local overhangs must have been eroded and filledvery rapidly. In the absence of evidence suggesting liquefaction, slumpscars, or slump deposits, or faults in underlying strata, erosion is interpreted


FIG. 10.—Planar-stratified gravel facies showing upward-fining couplets, clastclusters, and open-work pebble beds. Flow was from left to right.

FIG. 11.—Cross-stratified gravel. A) Poorly sorted trough-cross stratified gravel;scale is 8 cm long. Flow is from left to right. B) Normally graded, medium-scalecross-bedded pebbly gravel. Flow was from right to left. Knife is 23 cm long.

to have been by vortex impingement beneath a hydraulic jump. Hydraulicjumps form at the transition from supercritical flows to subcritical flowsand the conversion of kinetic energy to potential energy through an increasein flow thickness and decreased velocity. Additionally, there is always aloss in mechanical energy that is brought about by the generation of tur-bulence that is eventually dissipated into heat (Komar 1971). Intense tur-bulence generated within the hydraulic jump, particularly in high-Froude-number oscillatory jumps, erodes deeply and entrains large quantities ofsediment. The thicker, lower-energy flow downflow of the jump has sig-nificantly lower transport capacity that results in rapid sedimentation. Bedsof diffusely graded or massive sand have previously been attributed totraction carpets (Sohn 1997), sandy debris flows (Postma et al. 1983; Shan-mugam 1997), hyperconcentrated flows (Gorrell and Shaw 1991; Knellerand Branney 1995; Maizels 1989; Smith 1986), laminar sheared layers(Vrolijk and Southard 1997), or grain flow (Cheel and Rust 1982; Lowe1982). Collectively, these processes involve highly concentrated sedimentdispersions in which fluid turbulence is effectively damped. Sediment sup-port mechanisms for each of the processes may include one or a combi-nation of the following: turbulence, dispersive pressure, hindered settling,and buoyancy. Deposition of the diffusely graded sand in steep-sided scoursis interpreted to have occurred almost immediately after erosion becauseof rapid loss of transport capacity by the flow. The downflow transfor-mation of the flow probably involved a number of depositional processes.Where suspension sedimentation was highest, and the depositing sedimentflux was predominantly normal to the bed, deposition was possibly from asurge and a simple laminar sheared layer (e.g., Vrolijk and Southard 1997).With decreasing rates of suspension sedimentation, flow stratification andthe transfer of shear from the overlying dilute fluidal flow, deposition wasfrom sustained traction carpets (Sohn 1997) (see poorly sorted gravel withintraclasts).

Where tabular, diffusely stratified sand grades laterally into or is overlainby low-angle cross-stratified sand, the lack of stratification in the diffuselystratified part was probably related to higher rates of suspension sedimen-tation and more rapid upbuilding. In a flume study, Arnott and Hand (1989)showed that under upper-plane-bed flow conditions stratification becameprogressively more diffuse with increasing rates of bed aggradation. Thereason for this change was attributed to reduced lateral segregation ofgrains, and a steeper angle of climb of persistent low-amplitude bed forms.Consequently, interbedded diffusely stratified and low-angle cross-stratifiedsand is likely related to spatial and temporal variations in sediment con-centration and local differences in rates of sedimentation and flow speed.

Cross-Bedded Sand (St, Sp)

The cross-bedded sand facies crops out extensively at the locationsshown in Figure 15. It is sharp-based, predominantly medium-scale, planar-tabular and trough cross-bedded medium to pebbly coarse sand. Planarcross-sets are , 50 cm thick and form cosets 2–3 m thick (Figs. 15, 16A).The thickest sets occur in the lower part of cosets and thin upward. InFigure 15 strata dip at a low angle (, 108) to the west. Trough cross-setsare generally coarser than planar cross-sets, are coarse-tail graded, and formcosets 30–50 cm thick. At one location, climbing cross-sets 10–20 cm thickform a coset 1–1.5 m thick. Deformation is rare but where present consistspredominantly of small-scale normal faults with 2–5 cm offset, and minorconvolute bedding. Paleoflow measurements of predominantly troughcross-beds are to the southwest, whereas planar cross-beds indicate flowvarying from the southwest to northwest (Fig. 5). The combined measure-ments have a bimodal distribution with a primary southeast mode and asecondary northwest mode.

Interpretation.—The cross-bedded sand facies was deposited by tractionsediment transport and migrating dune bedforms (e.g., Church and Gilbert1975; McDonald and Vincent 1972). Tabular cross-stratification is the prod-uct of 2-D dunes that had limited erosion in their troughs (Saunderson andLockett 1983). Trough cross-stratification was formed by 3-D dunes withintense downflow scouring (Harms et al. 1975). Climbing, medium-scalecross-stratification indicates high rates of sedimentation under waning flowconditions (e.g., Ashley et al. 1982; Gorrell and Shaw 1991). It is useful tonote that climbing medium-scale cross-bedding is not common in the geo-logical record, most probably because of the high bed-load transport rates incombination with downflow scouring associated with dunes.

Small-Scale Cross-Laminated Fine Sand (Sr)

Small-scale, cross-laminated fine sand is the most common facies, form-ing units 4–5 m thick; however, it also crops out locally elsewhere (Fig.


FIG. 12.—A) Photo mosaic and B) line drawing interpretation of section. Strata are dominated by low-angle cross-stratified (Sh), diffusely graded and massive (Sd),trough cross-stratified (St), and small-scale cross-laminated sand (Sr). Note large scour and lateral transition of sand to sandy gravel from left to right. Figure numbers online drawings refer to figures showing details. Flow was obliquely out of the page toward the left. Facies codes are defined in Table 1.

6). It abruptly overlies the cross-bedded sand facies or, less commonly, thegravel facies. Strata consist of small-scale cross-laminated fine sand com-posed predominantly of stoss-erosional or stoss-depositional small-scalecross-laminae (Fig. 16B; ripple-drift cross-lamination of Jopling and Walk-er 1968). Heavy-mineral laminae are common on the foresets of the cross-laminae. Less common are sinusoidal laminae with steep angles of climb(. 608) and no heavy-mineral laminae. Cross-laminated silty fine sand tofine sand forms sharp-based sets , 4 cm thick and cosets up to 3 m thick.Where silty the coset thickness is generally thinner. Scours and intrafor-mational clay clasts are rare.

Interpretation.—Small-scale, cross-laminated silty fine sand was de-posited by low-energy current ripples. Climbing cross-lamination is theresult of combined traction and suspension sedimentation and rapid bedaggradation as the flow loses transport capacity (Ashley et al. 1982; Joplingand Walker 1968). Rapid bed aggradation and climbing cross-stratificationcommonly occur under expanding-flow conditions of density underflows,turbidity currents (Walker 1992), and overbank sedimentation (Harms et

al. 1975). Thick sequences (hundreds of centimeters) of small-scale cross-laminated sand are common in glacilacustrine deposits (e.g., Gorrell andShaw 1991; Rust and Romanelli 1975; Jopling and Walker 1968), but arecomparatively thinner, generally less than tens of centimeters thick, andless common in deep submarine fans (e.g., Pickering et al. 1989; Walker1992). This reflects the difference between quasi-continuous flows in gla-cial settings compared with surge-type turbidity currents in deep-sea sub-marine fans.

Silt–Clay (F)

Silt–clay up to 2–3 m thick is stratigraphically highest and in most placesgradationally overlies the small-scale cross-laminated facies. Fresh sectionsare mostly massive; however, sharp-based, fine sand or silt beds 1–3 cmthick are generally discernible upon drying. Rare, small-scale cross-lami-nated fine sand beds , 1 cm thick occur at the bases of individual couplets.Couplets fine upward and form rhythmic successions that are up to 20–30cm thick. A series of graded sand–silt couplets are commonly overlain by


FIG. 13.—A) Photo mosaic and B) line drawing of low-angle cross-stratified medium sand facies, showing bedding relationships with scour surfaces. Note theonlapping bed relationships with erosional surfaces and down-flow transition to conformable succession of beds, change in bed dip and thickness vertically, and smallscours infilled with massive sand (arrow). C) Close-up of bedding relationships in photo A. Steep-walled scour truncates beds to right. Succession is interpreted asantidune stratification formed beneath in-phase waves of a hydraulic jump recording rapid bed aggradation. Flow was obliquely out of the photo from right to left.Meter stick for scale; increments are 10 cm.

a clay stratum 0.5–2 cm thick that is faintly normally graded from under-lying silt and becomes massive.

Interpretation.—This facies is interpreted to have been deposited pre-dominantly by low-density turbidity currents and suspension sedimentationwithin a glacilacustrine basin (e.g., Gilbert 1997). Rhythmic, normallygraded sand–silt–clay strata were deposited by underflows and suspensionsedimentation. Clay strata were deposited from suspension. Suspension sed-imentation of clay requires low energy in the water column and is inter-preted to have occurred under winter conditions with negligible or no melt-water discharge and surface ice cover of the basin (Banerjee 1973). Con-sequently, thicker clay strata are interpreted to mark the end of an annualsedimentary cycle and the top of a varve.

FACIES ASSOCIATIONS

The facies identified in this study (Table 1) have been organized intofive facies associations, which are discussed below and are summarized onthe fence diagram in Figure 6 and schematically in Figure 17.

Gravel Association

The gravel association consists of poorly sorted gravel and planar-strat-ified gravel with minor cross-bedded and low-angle cross-stratified sand. Itis tabular and erosionally overlies cross-bedded sand and gravel. In outcropit is ; 10–20 m wide, 7–8 m thick, and has a length parallel to flow ofmore than 40 m (Figs. 6, 7, 8). Upward the strata change from medium-scale cross-bedded gravel to poorly sorted gravel with sand intraclasts andplanar-stratified gravel facies. Locally, there is well-developed lateral gra-dation from poorly sorted gravel with sand intraclasts to interbedded planar-stratified gravel, and farther downflow, to cross-bedded gravel and cross-bedded sand. The upward change in the succession is interpreted to reflectdeposition from progressively higher-energy flows and flows of increasingsediment concentration, whereas the downflow transitions indicate rapidflow expansion and a commensurate loss of flow energy and transport ca-pacity. Rapid downflow facies changes and erosional contacts characterizethis association and reflect the unsteady temporal and spatial nature of thejet efflux.


FIG. 14.—Diffusely graded and massive medium sand infilling a steep-walledscour. A) Note details of scour margin and infill: i) local steep angle of scour margin,ii) pebbles near base of fill, iii) internal scour margin, iv) variable continuity ofgraded bedding, v) abruptly overlying trough cross-bedded sand of similar grain sizebut with surface veneer of silty sand. B) Continuation of scour to left of photo A,i) locally overhanging scour margin, ii) massive sand lateral to diffusely gradedsand, iii) localized disruption of strata beneath scour margin, suggesting local slump-ing. The scale is 1 m.

Inclined-Sand Association

The inclined-sand association occurs adjacent to the gravel associationand consists of cross-bedded sand, low-angle cross-stratified sand, and mi-nor diffusely graded and massive sand (Figs. 6, 12). It ranges from 2 to 8m thick depending on the depth of incision by the steep-walled-scour as-sociation (see below). Beds oriented transverse to flow locally offlap thegravel association with a dip of 5–108 or elsewhere fine laterally from thegravel association. Offlapping beds form a succession of low-angle cross-stratified and diffusely graded sand with multiple low-angle truncations andsmall, shallow scours (Fig. 13). In contrast, strata that grade laterally aregenerally tabular and consist of horizontal gravelly sand beds that finegradationally away from the gravel association (Fig. 12).

Steep-Sided-Scour Association

The steep-sided-scour association consists of steep-sided, irregular scourseroded into the inclined sand association (Figs. 6, 12) and infilled withdiffusely graded sand facies. The largest scour is 10 m wide and 3 m deepand has margins dipping 10–608 (Figs. 12A, 14A). Locally the scour mar-gin consist of overhangs 20–30 cm long (Fig. 14B). In some cases thesubstrate of the scour margin is offset, suggesting minor slumping. The dipof the scour surface generally decreases upward and laterally, eventuallybecoming subhorizontal and overlain by cross-bedded sand. The coarserscour-fill sediment is poorly exposed but consists of sandy gravel with silty

sand intraclasts that fine upward to diffusely graded sand. The predominantinfill sediment is diffusely graded medium sand. Preservation of the steep-walled scour geometry, especially the overhangs, in a cohesionless sandysubstrate indicates that erosion was followed immediately by deposition ofthe diffusely graded and massive sand.

Large, Gently-Inclined-Bedset Association

The gently-inclined-bedset association consists of a thick succession ofgently-dipping surfaces that are overlain by planar cross-bedded sand orsmall-scale cross-laminated fine sand facies (Figs. 6, 15). The associationis 6–8 m thick and extends downflow subparallel to the paleoflow directionfor . 150 m. Individual bedsets are up to 60 cm thick and generally thinupward. Bed surfaces dip at 5–108 toward the west-northwest and the dipangle increases gradually basinward (westward). Small-scale cross-lami-nated sand is more abundant basinward. Local scours 1–2 m wide and 1m deep are infilled with cross-bedded medium sand.

Shallow-Channel-Fill Association

The shallow-channel-fill association consists of stacked and nested chan-nel fills of cross-bedded, small-scale cross-laminated sand and minor dif-fusely graded or massive sand (Figs. 6, 12, 15). It erosionally overlies theother facies associations and is thickest (3 m) where it overlies sand-dom-inated facies associations (Fig. 6). Channels are at least 1–2 m deep and5–20 m wide. Channels are deepest and widest near the base of the asso-ciation and become progressively shallower upward and downflow. Thebases of the channels have rare overdeepened troughs, and coarse lags arerare. This facies association has a consistent vertical and downflow faciestransition to progressively lower-energy, finer-grained silt and clay strata.

DEPOSITIONAL MODEL

Regional Setting

The facies and facies associations discussed above are interpreted to havebeen deposited in the proximal region of a glacilacustrine subaqueous fan(Fig. 17), similar to those that occur at the downflow terminus of eskers(e.g., Diemer 1988; Warren and Ashley 1994), onlap or are lateral to eskers(e.g., Banerjee and McDonald 1975; Brennand and Shaw 1994; Gorrell andShaw 1991), or form parts of large moraines (e.g., Fyfe 1990; Sharpe andCowan 1990). Glacilacustrine subaqueous fans have been interpreted toform in response to seasonal steady-state (e.g., Powell 1990), or episodicglacial hydraulic events (e.g., Gorrell and Shaw 1991). In some cases mo-raine deposits represent the coalescence of multiple adjacent fans (e.g., Rustand Romanelli 1975). Although no esker deposit was observed to mergeinto the fan, the study area is in the upland region of the Oak RidgesMoraine, which is a large polygenetic glacial landform (Barnett et al. 1998).To the east, esker ridges north of the moraine appear to connect with sub-aqueous fan deposits (Paterson and Cheel 1997). Because the moraine is apositive topographic landform, deposition in a glacilacustrine basin re-quired ice support and a glacial meltwater system for sediment delivery.This glacilacustrine basin was bounded on most sides by ice and to thewest by the Niagara Escarpment. Basin geometry and depth were controlledby the location where ice impinged along the escarpment and by the ele-vation of outlet channels that breached the escarpment. On the basis ofvalley morphology, sedimentary deposits, and elevation of moraine depos-its, Chapman (1985) identified four key outlet channel elevations at 420,380, 367, and 290 m asl. For the study area, with a surface elevation of290 m, these outflow levels probably drained a lake 70–120 m deep. Theapproximate surface extent of the lake can be estimated from the locationsof the drainage channels, the fan location, and the surface extent of HaltonTill. On this basis a rough estimate of the north–south extent of the lakeis ; 30–40 km. The east–west extent of the lake basin between the north-


FIG. 15.—A) Photo mosaic and B) line drawing interpretation of section. Note predominance of planar cross-bedded sand (Sp) in the lower part of the section andoverlying trough cross-bedded (St) and small scale cross-laminated fine sand (Sr) in the upper part. Lower section consists of the gently inclined bed set association,whereas, the upper part of the face (arrow) forms part of the sandy shallow-channel association. Flow is from left to right. Scale is 1 m long.

ern Simcoe and southern Ontario ice lobes is more difficult to estimate butwas possibly as short as 30 km but no more than 100 km.

The supply of meltwater-transported sediment to the subaqueous fan wasmost probably from unsteady episodic flood discharge rather than fromsteady seasonal flow. A flood interpretation is supported by the lack offine-grained horizons indicating stratigraphic evidence for annual sedimen-tation (varves), the high-energy facies, and the stratal architecture. Theabsence of clay interlaminae, either in situ or as intraformational clastswithin the coarse-grained fan succession, suggests that the fan was con-structed during a single meltwater season. The high-energy assemblagesconsist of poorly sorted gravel and diffusely graded sand deposited fromhyperconcentrated flows that are most commonly generated during floodevents (e.g., Costa 1988; Maizels 1993). Furthermore, steep-walled scoursare commonly formed by intense erosion beneath hydraulic jumps whichare a characteristic of flood discharges (e.g., Gorrell and Shaw 1991).

The proposed flood that constructed the fan possibly originated from oneof several individual reservoirs of either subglacial or supraglacial position,or possibly a combination of sources including precipitation. Studies ofspring events in alpine glaciers indicate a tripartite contribution to peakdischarges including: (i) subglacial water pockets, (ii) existing supraglacialstorage, and (iii) precipitation input (Warburton and Fenn 1994). Routingand fluctuations of discharge are influenced by changes in the geometryand connectivity of the subglacial cavity, changes in the contribution fromindividual meltwater sources, and development and collapse of conduits.These controls produce flow pulsations that influence the discharge, and

sediment load. Discharge at the terminus, regardless of the exact source(s)of the meltwater, was subglacial, or if englacial was delivered at the sameelevation as the fan surface.

Depositional Setting

Cross-bedded sand and gravel in the Oak Ridges Moraine have beeninterpreted as either braided fluvial, deltaic, or subaqueous fan deposits(Barnett et al. 1998; Duckworth 1979; Paterson and Cheel 1997). Addi-tionally, incision of underlying subaqueous fan deposits and infill has beenused to support an interpretation of falling lacustrine base levels, subaerialexposure, and subsequent rising base levels (Paterson and Cheel 1997).The similarity of sedimentary structures and facies produced by unidirec-tional currents is a common interpretive problem, whether in submarinefan (Hein 1984), glacifluvial esker, or subaqueous fan deposits (Rust andRomanelli 1975). In this study a subaqueous-fan setting is favored for thefollowing reasons: (i) rapid deposition and aggradation of coarse-grainedsediment; (ii) the succession is overlain by basinal silt–clay facies; and (iii)similarity with other subaqueous fan deposits. The inferred rapid aggra-dation of these deposits indicates deposition in a relatively deep subaqueousenvironment unconstrained by base level or accommodation space. Thereis also extensive evidence of rapid flow expansion and rapid suspended-sediment deposition from both climbing medium-scale (dune) and small-scale current-ripple stratification. Rapid bed aggradation and flow expan-sion are conditions more characteristic of subaqueous fans rather than sub-


FIG. 16.—A) Planar cross-bedded medium sand; flow was from left to right; meterstick is graduated in 10 cm increments. B) Climbing small-scale cross-laminatedfine sand (Sr), commonly referred to as ripple-drift cross-lamination. Paleoflow wasfrom left to right. Note false bedding produced by silt-rich layers suggestive of largecross-strata. Pencil is 14 cm long.

aerial, braided-fluvial deposits. Additionally, although the shallow-channel-fill association has similarities with braided fluvial systems, similar faciesassociations have been observed in glacigenic subaqueous fans (Burbidgeand Rust 1988), deep-water submarine fans (Hein and Walker 1982), andsubmarine-canyon fills (Arnott and Hein 1986). Finally, the entire fan isconformably overlain by a 2–3 m succession of silt and clay interpreted tobe glacilacustrine deposits. Thus, in the absence of evidence for water-levelfluctuations the entire succession is interpreted to have been deposited su-baqueously in a glacilacustrine environment.

Stage 1: Model of Plane-Wall Jet With Jump

A submerged plane-wall jet with jump forms where a supercritical flowdebouches from a conduit along the basin floor and undergoes rapid de-celeration (Fig.3). Such flows are characterized by the formation of a hy-draulic jump and the generation of intense turbulence within the jet efflux(Rajaratnam and Subramanyan 1986). A plane-wall jet with jump evolvesthrough three zones of flow development as it penetrates the ambient basinfluid (Long et al. 1991). The first zone, and standardizing the terminologywith that used for plane jets, is the zone of flow establishment (ZFE). Thiszone of jet flow is similar to that of a plane-wall jet with flow entrainmentof ambient fluid (Long et al. 1991). Downflow is the zone of flow transition,which corresponds with ; 85% of the roller length of the jump, and furtherbasinward is the zone of established flow (ZEF). The zone of establishedflow is marked by subcritical flow and continued fluid entrainment. The

plane-wall jet with jump is interpreted to have deposited a wide range ofsediment textures from high-energy, sediment-charged flows. This depo-sitional model interprets the sediment facies and facies associations withinthe context of the model of flow development of a plane-wall jet with jumpoutlined above.

Zone of Flow Establishment.—Within the ZFE, only poorly sorted andplanar-stratified gravel was deposited (Fig. 17). These strata were depositedfrom high-energy flows that partially or completely eroded underlying low-er-energy deposits of cross-bedded sand and gravel. The poorly sorted grav-el with intraclasts is interpreted to have been deposited from hyperconcen-trated flows that were the basal part of thicker stratified flows. The planar-stratified gravel was in turn deposited farther downflow from more dilutecritical flows, perhaps concentrated flows (e.g., Mulder and Alexander2001). Deposition in this zone was controlled both by allocyclic mecha-nisms of the glacier hydraulic system and sediment transport mechanismof bedload sediment. Consequently, deposition in this zone does not recordthe effects of flow expansion into the basin but rather conduit flow con-ditions.

Zone of Flow Transition.—Downflow of the ZFE, the zone of flowtransition (ZFT) developed beneath the hydraulic-jump roller. Here the flowexpanded rapidly at the supercritical-to-subcritical transition. The magni-tude of flow thickening depends on the Richardson number (Rajaratnamand Subramanyan 1986), which is equal to the inverse of the square of thedensimetric Froude number. For example, using a suspended-sediment con-centration of 26 kg m23, an inflow depth of 10 m, and a flow velocity of5 m s21, Gorrell and Shaw (1991) estimated the flow would thicken by 4to 12 times. For hydraulic jumps of low Fro (, 2.5), bed erosion may beminor and the most significant effect is the loss of transport capacity andconsequent rapid sedimentation. Increased sediment concentration in thebasal part of the flow could have formed a basal hyperconcentrated zoneand accordingly a stratified flow. Deposition from the basal hyperconcen-trated zone could then have caused rapid bed aggradation with the em-placement of the 2–3 m thick, massive or poorly organized gravel deposits.Rapid downflow evolution from the hyperconcentrated dispersion to morefluidal flow conditions is suggested by the lateral transition to plane-bedand cross-bedded deposits over a distance of ; 5 m (Fig. 7). These depositsare strikingly similar to coarse sand and gravel rhythmites described fromrecent jokulhlaup events in Iceland (Russell and Knudsen 1999). The 8–10 m thick Icelandic succession was deposited over a maximum of 17 hoursand consists of sharp-based, graded couplets of sandy gravel deposited byunsteady flow pulses that lasted from seconds to minutes (Russell andKnudsen 1999).

Flows with larger Fro, for example in an oscillatory hydraulic jump (2.5, Fro , 4.5), would have had the turbulent energy to erode rapidly anddeeply into underlying sediment and subsequently resediment this materialrapidly as flow energy was dissipated. Additionally, vertical oscillation ofthe jet may have locally increased scour depth and size. Alternatively, scoursize and depth may have been linked to the growth in vortex size by stream-wise vortex pairing until a single large vortex occupied the complete flowdepth (Long et al. 1991). Development of such large vortices would havedistributed entrained sediment throughout the flow and established the con-ditions for rapid loss of transport competence and capacity downflow ofthe jump. With the rapid decrease in turbulence downflow of the jump avoluminous downward flux of suspended sediment would have occurred,stratifying the flow and forming a basal hyperconcentrated layer. Accord-ingly, sediment was deposited by a variety of mechanisms that evolvedrapidly downflow from en masse suspension sedimentation, to depositionfrom laminar-shearing surge flows and traction carpets to eventual depo-sition from fluidal flows. Locally, in areas of highest suspended-sedimentconcentration, sediment was deposited as poorly sorted sandy gravel withintraclasts (Figs. 7, 9, 17). Elsewhere, diffusely graded sand beds 5–10 cmthick infilled scours during multiple stages of scour and deposition (Figs.12A, 14). Consequently, the steep-sided-scour association can be explained


FIG. 17.—Schematic block diagram of proximal subaqueous fan model showing lithofacies and zones of jet-efflux development. Drawing by John Glew.

by a continuum of events within a dynamic environment of vortex growth,vertical jet oscillation, and rapid streamwise loss of transport capacity.

Zone of Established Flow.—The more dilute overlying flow continuedbasinward as subcritical underflows within the zone of established flow(ZEF). Immediately downflow of the transition zone, migrating dunes con-structed a gently dipping accretionary wedge 4–6 m thick and 250 m longin the proximal ZEF (Figs. 15, 16). Within the first 100 m of the ZEF, 2-D dune-scale cross-bedding sets dominate (Figs. 6, 15). The cross-beddedsand grades rapidly downflow to mostly ripple-scale cross-lamination withonly minor dune-scale cross-bedding recording the basinward accretion ofthe fan (Fig. 17).

Stage II: Model of Plane-Wall Jet

Lateral capture of the main conduit flow, or cessation of the jokulhlaupdischarge and a return to seasonal diurnal meltwater flow, is defined by theabrupt superposition of the shallow-channel facies association (Figs. 12,15). The uniform streamwise character of the shallow-channel-fill associ-ation and downflow transition to planar-tabular cross-bedding and small-scale cross-lamination is used to infer deposition from a subcritical flowof a plane-wall jet. The stacked succession of broad shallow channels re-cord fan aggradation under a semicontinuous jet efflux. The length of ex-posure and the relatively uniform sandy facies suggest that deposition waswithin the zone of established flow rather than more proximal to the fanapex.

The complete fan succession is overlain by the silt–clay facies, 2–4 mthick. Silt rhythmites in this facies probably record the distal sedimentation

signal of efflux discharge elsewhere in the basin and the deposition ofBouma Tbc turbidites. Where only graded Td silt–clay couplets occur, de-position was predominantly by suspension settling. The rhythmites thusrecord the end of fan development.

DISCUSSION

Most subaqueous fans are deposited from either a plane jet, a plane-walljet, or a plane-wall jet with jump (Fig. 2). These three efflux types havedistinctly different efflux geometries, an elongate parabola for the planejet, and a stubby fan shape for the plane-wall jet and the plane-wall jetwith jump. The geometry of the jet efflux, and in turn the fan deposits, arerelated to the relative magnitude of the efflux-jet diameter, the inertial en-ergy and the interfacial shear. The downflow trend of a plane jet is welldefined (Bates 1953). No similar treatment exists, however, for the plane-wall jet or the plane-wall jet with jump. It is clear that in the case of highinterfacial shear, as in the case of a plane-wall jet, the fan is shorter inaxial length and more triangular in shape (Powell 1981; Wright 1977). Theplane-wall jet has a lateral to vertical spreading ratio of between 5 and 9(Launder and Rodi 1983). It has been noted that initial flow behavior ofthe plane-wall jet with jump is similar to that of the plane-wall jet (Longet al 1990). However, depending upon the Richardson number (flow den-sity) there may be a significant departure in the ratio of lateral spreadingto vertical thickness across the hydraulic jump, where, for example, theflow may thicken considerably (Rajaratnam and Subramanyan 1986). Con-sequently, the distribution of facies, particularly downflow, may be ex-pected to differ considerably between a plane-wall jet and a plane-wall jet


with jump. It might be anticipated that the distribution of coarse facies willbe considerably compressed in a plane-wall jet with jump compared to theplane-wall jet. Additionally the range of sediment facies will differ in asmuch as deposits from a plane-wall jet consist predominantly of subcriticalbedform stratification, whereas, deposits from a plane-wall jet with jumpwill have a proportion of sediment facies with bedform structures of criticaland supercritical flow in addition to deposits from hyperconcentrated andconcentrated flows.

The deposits described in this study have a number of facies and faciesassociations that are probably diagnostic of a plane-wall jet with jump. Themost distinctive is the steep-sided-scour facies association, which gradesrapidly downflow into cross-bedded sand deposited by subcritical flow. Thesteep-sided-scour association is interpreted to have been formed beneath ahydraulic jump. The jump, in turn, forms an important energy fence thatis indicated by an abrupt change in grain size and sedimentary structuresbetween the proximal-fan ZFE and ZFT deposits and those in the moredistal ZEF zone (Fig. 17). A similar interpretation has been proposed pre-viously by Gorrell and Shaw (1991) for steep-walled scours in a subaque-ous fan deposit near Ottawa, Ontario. The lateral and vertical facies asso-ciations described in this study provide additional evidence for the for-mation of such scours by vortex impingement associated with turbulenthydraulic jumps. The well constrained vertical and lateral facies associa-tions appear to preclude the possibility that the steep-sided-scour associa-tion was produced by slumping as invoked in numerous previous studiesof subaqueous fans (e.g., Rust and Romanelli 1975; Rust 1988; Cheel andRust 1982; Postma et al. 1983)

Evidence of supercritical flow, however, is not necessarily confirmationof deposition from a plane-wall jet with jump. Flow conditions within aplane-wall jet may locally attain supercritical flow conditions as the jetencounters basin-floor (fan) topography. Increased velocity due to flowthinning across topographic obstacles is a common phenomenon. The resultof such changes in the flow could be deposition of antidune cross-stratifi-cation and downflow development of hydraulic jumps. In general, faciesand facies associations related to such local flow conditions would be lo-cally developed and therefore not common in the stratigraphic record. Fur-thermore such deposits might occur downflow of a succession consistingof subcritical-bedform stratification. In this study, however, there is a con-sistent streamwise mechanistic and depositional continuum according tograin size, inferred flow concentrations and conditions, and stratification ofsupercritical and subcritical bedforms.

This study examined sediment cropping out along a 250 m section sub-parallel to the paleoflow direction. An interesting exercise would be toestimate the proportion of the exposure compared to the streamwise lengthof the entire fan complex. Because of the large number of variables in-volved in estimating boundary conditions for a plane-wall jet with jump,most of which are unknown in this example, the following estimate is basedon a model of the simpler plane-jet. The first basic parameter is an estimateof the conduit diameter. Using a conservative 10 m diameter (D) conduit,a scale common to flow estimates made by previous workers (Gorrell andShaw 1991), the ZFE should have an approximate length of four times thediameter (4D), or 40 m long. This is approximately twice the length of theobserved gravel exposure, suggesting either upflow continuation of the fancore association or discharge from a somewhat smaller conduit. The ZFThas a similar scaling relationship (4–8 D) and therefore should be about40280 m long. Using the extent over which diffusely graded sand wasobserved as a proxy for the length of the ZFT provides a distance of ;30–40 m. Using an ideal model of an inertia-dominated plane jet the 300m long section observed in this study corresponds to only 1.5% of the totaldistance of jet intrusion of 2000 times the conduit diameter (2000D), or; 20,000 m. As noted previously, the intrusion distance for friction-dom-inated plane-wall jet or plane-wall jet with jump is significantly shorter.Assuming an arbitrary decrease in jet penetration of 50–75%, or 10–15 km,the outcrop still represents only ; 3–6% of the total potential length of

intrusion of the jet efflux into the basin. This suggests that the fan succes-sion farther downflow would be dominated by sediment of a caliber finerthan lower medium sand and most probably dominated by stratification ofripple-scale cross-lamination or layers deposited from suspension.

Discriminating between subaqueous fan strata deposited by episodicfloods versus a seasonal steady-state meltwater regime is crucial to theunderstanding of late Laurentide deglaciation. From Icelandic sandurs, Mai-zels (1993) recognized facies and facies associations that can be used todifferentiate seasonal diurnal flood deposits from outbreak jokulhlaup de-posits of both nonvolcanic and volcanic origin. Key elements include: suc-cession thickness, recognition of deposits of traction versus hyperconcen-trated-flow origin, and cyclicity. On the basis of these and other criteriarelated to supercritical flow conditions, the subaqueous fan of this studyhas been interpreted to have accumulated during a single flood. In thiscase, however, the flood was not necessarily the consequence of an out-break release, but may also have been triggered by a large rainstorm. Un-fortunately, there are no known sedimentological characteristics that candifferentiate these mechanisms.

Resolving meltwater sources for such events has important implicationsfor understanding the genesis of moraines, which have generally been in-terpreted to be ice-margin deposits recording long-term occupation andseasonal sedimentation (e.g., Hillaire-Marcel et al. 1981). For example,recent work has suggested that the Oak Ridges Moraine consists of low-energy seasonal basinal sedimentation (Gilbert 1997), seasonal proximalfan deposition (Paterson and Cheel 1997), and mixed seasonal and higher-energy episodic flood deposition (Barnett et al. 1998). The current studyand the work of other authors (e.g., Barnett et al. 1998; Brennand andShaw 1996; Sharpe and Cowan 1990) suggests that some large morainesare built, at least in part, by short-lived, high-energy voluminous episodesof sediment flux that cause extremely rapid upbuilding of the moraine.

CONCLUSIONS

Eight facies and five facies associations were described from a 250–300m exposure oriented subparallel to the mean paleoflow direction in the OakRidges Moraine. The facies and facies associations can be best explainedin terms of a jet-efflux subaqueous fan model. Deposition is interpreted tohave occurred within an evolving single event involving semicontinuousjet efflux. The downflow evolution can be subdivided into three distinctstages of flow evolution. From proximal to distal these zones are: i) zoneof flow establishment (ZFE), ii) zone of flow transition (ZFT), iii) zone ofestablished flow (ZEF). The gravel association of planar-stratified and mas-sive gravel marks the ZFE. It is flanked laterally and downflow by low-angle cross-stratified and diffusely graded sand facies of the ZFT. Sedi-mentological interpretations suggest that the transition from hyperconcen-trated-flow deposition to fluidal-flow deposition occurred over a distanceof only 2–5 m and was the result of rapid flow expansion and attendantloss of transport capacity. Within the steep-walled scour association of theZFT, the scours were formed and diffusely graded and massive infill wasdeposited immediately downflow of hydraulic jumps that formed as a resultof rapid flow expansion and a change from supercritical to subcritical flow.Additionally, the rapid development of hyperconcentrated dispersions re-sulted in deposition from surge and/or traction carpets producing diffuselygraded sandy rhythmites. Downflow the transition to the ZEF is markedby accreting foresets of planar cross-bedded sand deposited predominantlyby 2-D dunes that grade downflow into small-scale cross-laminated sand.The fan complex is overlain by a shallow-channel-fill association of cross-bedded medium sand and small-scale cross-laminated fine sand interpretedto represent a subaqueous braided system that marked the waning stagesof jet discharge. The final episode of fan development was marked by aperiod of basinal suspension sedimentation that draped the fan with a layerof silt and clay.

Discussion of fan facies and architecture within a framework of jet-efflux


dynamics provides an improved understanding of subaqueous fan sedi-mentation. This fan succession suggests that during the later stages of de-velopment the Oak Ridges Moraine was influenced strongly by local con-duits along the margins of a glacilacustrine basin. At least locally, fanswere constructed by flood discharges that represent only a single meltwaterseason. Recognition of flood deposits within morainal subaqueous fan suc-cessions has important implication for interpreting the origin of continentalmoraines. For example, some moraines may be constructed over short pe-riods of time by episodic meltwater discharges that are unrelated to long-term climate-modulated ablation and seasonal meltwater production.

ACKNOWLEDGMENTS

Access to the study site was provided by Dufferin Aggregates. Assistance in thefield was provided by D. Cummings, S. Dumas, and A. Grignon. The artistic ren-dering by J. Glew of figures 2 and 17 is much appreciated. Reviews of this paperby D. Sharpe, R. Rainbird, and R. Cheel are gratefully acknowledged. An insightfulreview by R. Gilbert helped to significantly clarify parts of the paper. N. Rajaratnamis thanked for his patient e-mail correspondence concerning plane-wall jets. Fundingfor this study was provided by the Geological Survey of Canada National Mappingand Hydrogeology programs and a Natural Science and Engineering Research Coun-cil research grant to R.W.C. Arnott. This is Geological Survey of Canada contri-bution 2002052.

REFERENCES

ALEXANDER, J., BRIDGE, J.S., CHEEL, R.J., AND LECLAIR, S.F., 2001, Bedforms and associatedsedimentary structures formed under supercritical water flows over aggrading sand beds:Sedimentology, v. 48, p. 133–152.

ARNOTT, R.W., AND HEIN, F.J., 1986, Submarine canyon fills of the Hector Formation, LakeLouise, Alberta: Late Precambrian syn-rift deposits of the proto-Pacific miogeocline: Bulletinof Canadian Petroleum Geology, v. 34, p. 395–407.

ARNOTT, R.W.C., AND HAND, B.M., 1989, Bedforms, primary structures and grain fabric in thepresence of suspended sediment rain: Journal of Sedimentary Petrology, v. 59, p. 1062–1069.

ASHLEY, G.M., SOUTHARD, J.B., AND BOOTHROYD, J.C., 1982, Deposition of climbing-ripple beds;a flume simulation: Sedimentology, v. 29, p. 67–79.

BANERJEE, I., 1973, Part A: Sedimentology of Pleistocene glacial varves in Ontario, Canada;Part B: Nature of grain-size distribution of some Pleistocene glacial varves in Ontario,Canada: Ottawa, Department of Energy Mines and Resources, Geological Survey of Canada,Bulletin, v. 226, 60 p.

BANERJEE, I., AND MCDONALD, B.C., 1975, Nature of esker sedimentation, in Jopling, A.V., andMcDonald, B.C., eds., Glaciofluvial and Glaciolacustrine Sedimentation: Society of Eco-nomic Paleontologists and Mineralogists, Special Publication 23, p. 132–154.

BARNETT, P.J., SHARPE, D.R., RUSSELL, H.A.J., BRENNAND, T.A., KENNY, F.M., AND PUGIN, A.,1998, On the origins of the Oak Ridges Moraine: Canadian Journal of Earth Sciences, v.35, p. 1152–1167.

BATES, C.C., 1953, Rational theory of delta formation: American Association of PetroleumGeologists, Bulletin, v. 37, p. 2119–2162.

BLAIR, T.C., 1987, Sedimentary processes, vertical stratification sequences, and geomorphologyof the Roaring River alluvial fan, Rocky Mountain National Park, Colorado: Journal ofSedimentary Petrology, v. 57, p. 1–18.

BLAIR, T.C., 1999, Sedimentology of the debris-flow-dominated Warm Spring Canyon alluvialfan, Death Valley, California: Sedimentology, v. 46, p. 941–965.

BLAIR, T., AND MCPHERSON, J.G., 1994, Alluvial fans and their natural distinction from riversbased on morphology, hydraulic processes, sedimentary processes, and facies assemblages:Journal of Sedimentary Research, v. A64, p. 450–489.

BOOTHROYD, J.C., AND ASHLEY, G.M., 1975, Processes, bar morphology, and sedimentary struc-tures on braided outwash fans, northeastern Gulf of Alaska, in Jopling, A.V., and McDonald,B.C., eds., Glaciofluvial and Glaciolacustrine Sedimentation: Society of Economic Paleon-tologists and Mineralogists, Special Publication 23, p. 193–222.

BRENNAND, T.A., 1994, Macroforms, large bedforms and rhythmic sedimentary sequences insubglacial eskers, south-central Ontario: implications for esker genesis and meltwater re-gime: Sedimentary Geology, v. 91, p. 9–55.

BRENNAND, T.A., AND SHAW, J., 1996, The Harricana glacifluvial complex, Abitibi region, Que-bec: its genesis and implications for meltwater regime and ice-sheet dynamics: SedimentaryGeology, v. 102, p. 221–262.

BURBIDGE, G.H., AND RUST, B.R., 1988, A Champlain sea subwash fan at St Lazare, Quebec,in Gadd, N.R., ed., The Late Quaternary Development of the Champlain Sea Basin: St.John’s, Newfoundland, Geological Association of Canada, Special Paper, 35, p. 47–61.

CARLING, P.A., 1995, Flow-separation berms downstream of a hydraulic jump in a bedrockchannel: Geomorphology, v. 11, p. 245–253.

CHAPMAN, L.J., 1985, On the origin of the Oak Ridges Moraine, southern Ontario: CanadianJournal of Earth Sciences, v. 22, p. 300–303.

CHAPMAN, L.J., AND PUTNAM, D.F., 1984, The Physiography of Southern Ontario: Ontario Geo-logical Survey, Special Volume 2, 270 p.

CHEEL, R.J., AND RUST, B.R., 1982, Coarse grained facies of glacio-marine deposits near Ottawa,

Canada, in Davidson, A.R., Nickling, W., and Fahey, B.D., eds., Research in Glacial, Gla-cio-fluvial, and Glacio-lacustrine Systems: Guelph, Ontario, Canada, University of Guelph,Guelph Symposium on Geomorphology, Proceedings, p. 279–295.

CHEEL, R.J., AND RUST, B.R., 1986, A sequence of soft-sediment deformation (dewatering)structures in late Quaternary subaqueous outwash near Ottawa, Canada: Sedimentary Ge-ology, v. 47, p. 77–93.

CHEEL, R.J., BRIDGE, J.S., AND BEST, J.L., 1990, Flow, sediment transport and bedform dynamicsover the transition from dunes to upper-stage plane beds: implications for the formation ofplanar laminae: discussion and reply: Sedimentology, v. 37, p. 549–553.

CHOW, V.T., 1959, Open-Channel Hydraulics: New York, McGraw-Hill Book Company, Mc-Graw-Hill Civil Engineering Series, 680 p.

CHURCH, M., AND GILBERT, R., 1975, Proglacial fluvial and lacustrine environments, in Jopling,A.V., and McDonald, B.C., eds., Glaciofluvial and Glaciolacustrine Sedimentation: Societyof Economic Paleontologists and Mineralogists, Special Publication 23, p. 22–100.

COLLINSON, J.D., AND THOMPSON, D.B., 1989, Sedimentary Structures: London, Unwin, 207 p.COSTA, J.E., 1988, Rheologic, geomorphic, and sedimentologic differentiation of water floods,

hyperconcentrated flows, and debris flows, in Baker, V.R., Kochel, R.C., and Patton, P.C.,eds., Flood Geomorphology: New York, John Wiley & Sons, p. 113–122.

DAUB, G.A., 1996, Experimental study of deep-sea fan evolution and sedimentology [unpub-lished MSc thesis]: Colorado State University, Fort Collins, Colorado, 165 p.

DIEMER, J.A., 1988, Subaqueous outwash deposits in the Ingraham ridge, Chazy, New York:Canadian Journal of Earth Sciences, v. 25, p. 1384–1396.

DORSEY, R.J., AND FALK, P.D., 1998, Rapid development of gravelly high-density turbiditycurrents in marine Gilbert-type fan deltas, Loreto Basin, Baja California Sur, Mexico: Sed-imentology, v. 45, p. 331–349.

DUCKWORTH, P.B., 1979, The late depositional history of the western end of the Oak RidgesMoraine, Ontario: Canadian Journal Earth Sciences, v. 16, p. 1094–1107.

FYFE, G.J., 1990, The effect of water depth on ice-proximal glacilacustrine sedimentation:Salpausslka I, southern Finland: Boreas, v. 19, p. 147–164.

GARCIA, M.H., 1993, Hydraulic jumps in sediment-driven bottom currents: Journal of HydraulicEngineering, v. 119, p. 1094–1117.

GILBERT, R., 1997, Glacilacustrine sedimentation in part of the Oak Ridges Moraine: Geogra-phie Physique et Quaternaire, v. 7, p. 55–66.

GORRELL, G., AND SHAW, J., 1991, Deposition in an esker, bead and fan complex, Lanark,Ontario, Canada: Sedimentary Geology, v. 72, p. 285–314.

HAND, B.M., 1974, Supercritical flow in density currents: Journal Sedimentary Petrology, v.44, p. 637–648.

HENDERSON, P.J., 1988, Sedimentation in an esker system influenced by bedrock topographynear Kingston, Ontario: Canadian of Journal Earth Sciences, v. 25, p. 987–999.

HARMS, J.C., SOUTHARD, J.B., SPEARING, D.R., AND WALKER, R.G., 1975, Depositional environ-ments as interpreted from primary sedimentary structures and stratification sequences: So-ciety of Economic Paleontologists and Mineralogists, Lecture notes, Short Course no. 2,161 p.

HEIN, F.J., 1984, Deep-sea and fluvial braided channel conglomerates: a comparison of twocase studies, in Koster, E.H., and Steel, R.J., eds., Sedimentology of Gravels and Conglom-erates: Calgary, Canadian Society of Petroleum Geologists, Memoir 10, p. 33–50.

HEIN, F.J., AND WALKER, R.G., 1982, The Cambro–Ordovician Cap Enrage Formation, Quebec,Canada: conglomeratic deposits of a braided submarine channel with terraces: Sedimentol-ogy, v. 29, p. 309–329.

HILLAIRE-MARCEL, C., OCCHIETTI, S., AND VINCENT, J.S., 1981, Sakami moraine, Quebec: A 500-km-long moraine without climatic control: Geology, v. 9, p. 210–214.

HUNTER, L.E., POWELL, R.D., AND SMITH, G.W., 1996, Facies architecture and grounding-linefan processes of morainal banks during deglaciation of coastal Maine: Geological Societyof America, Bulletin, v. 108, p. 1022–1038.

ISEYA, F., AND IKEDA, H., 1987, Pulsation in bedload transport rates induced by a longitudinalsediment sorting: a flume study using sand and gravel mixtures: Geografiska Annaler, v.69, p. 15–27.

JOPLING, A.V., AND WALKER, R.G., 1968, Morphology and origin of ripple-drift cross-lamination,with examples from the Pleistocene of Massachusetts: Journal of Sedimentary Petrology, v.38, p. 971–984.

KNELLER, B.C., AND BRANNEY, M.J., 1995, Sustained high-density turbidity currents and thedeposition of thick massive sands: Sedimentology, v. 42, p. 607–616.

KOMAR, P.D., 1971, Hydraulic jumps in turbidity currents: Geological Society of America,Bulletin, v. 82, p. 1477–1488.

KOSTER, E.H., 1978, Transverse ribs: their characteristics, origin and paleohydraulic signifi-cance, in Miall, A.D., ed., Fluvial Sedimentology: Canadian Society of Petroleum Geolo-gists, Memoir 5 p. 161–186.

KUBO, Y., AND YOKOKAWA, M., 2001, Theoretical study on breaking of waves on antidunes, inPeakall, J., McCaffrey, B., and Kneller, B., eds., Sediment Transport and Deposition byParticulate Gravity Currents: International Association of Sedimentologists, Special Publi-cation 31, p. 65–70.

LAJEUNESSE, P., AND MICHEL, A., 2002, Sedimentology of an ice-contact glacimarine fan com-plex, Nastapoka Hills, eastern Hudson Bay, northern Quebec: Sedimentary Geology, v. 152,p. 201–220.

LAUNDER, B.E., AND RODI, W., 1983, The turbulent wall jet—measuring and modeling: AnnualReview of Fluid Mechanics, v. 15, p. 429–459.

LONG, D., STEFFLER, P.M., AND RAJARATNAM, N., 1990, LDA study of flow structure in sub-merged hydraulic jump: Journal of Hydraulic Research, v. 28, p. 437–460.

LONG, D., STEFFLER, P.M., RAJARATNAM, N., AND SMY, P., 1991, Structure of flow in hydraulicjumps: Journal of Hydraulic Research, v. 29, p. 293–308.

LONNE, I., 1995, Sedimentary facies and depositional architecture of ice-contact glacimarinesystems: Sedimentary Geology, v. 98, p. 13–45.


LONNE, I., 2001, Dynamics of marine glacier termini read from moraine architecture: Geology,v. 29, p. 199–202.

LOWE, D.R., 1976, Grain flow and grain flow deposits: Journal of Sedimentary Petrology, v.46, p. 188–199.

LOWE, D.R., 1982, Sedimentary gravity flows: II. Depositional models with special referenceto the deposits of high density turbidity currents: Journal of Sedimentary Petrology, v. 52,p. 279–297.

MAIZELS, J., 1989, Sedimentology, paleoflow dynamics and flood history of jokulhlaup depos-its: paleohydrology of Holocene sediment sequences in southern Iceland sandur deposits:Journal of Sedimentary Petrology, v. 59, p. 204–223.

MAIZELS, J., 1993, Lithofacies variations within sandur deposits: the role of runoff regime, flowdynamics and sediment supply characteristics: Sedimentary Geology, v. 85, p. 299–325.

MARTIN, C.A.L., AND TURNER, B.R., 1998, Origins of massive-type sandstones in braided riversystems: Earth-Science Reviews, v. 44, p. 15–38.

MCDONALD, B.C., AND DAY, T.J., 1978, An experimental flume study on the formation oftransverse ribs: Geological Survey of Canada, Paper 78–1A, Current Research, p. 441–451.

MCDONALD, B.C., AND VINCENT, J.S., 1972, Fluvial sedimentary structures formed experimen-tally in a pipe, and their implications for interpretation of subglacial sedimentary environ-ments: Geological Survey of Canada, paper 72–27, 30 p.

MIALL, A.D., 1996, The Geology of Fluvial Deposits: Sedimentary Facies, Basin Analysis, andPetroleum Geology: Berlin; Springer, 582 p.

MIDDLETON, G.V., 1965, Antidune cross-bedding in a large flume: Journal of SedimentaryPetrology, v. 35, p. 922–927.

MULDER, T., AND ALEXANDER, J., 2001, The physical character of subaqueous sedimentary den-sity flows and their deposits: Sedimentology, v. 48, p. 269–299.

PATERSON, J.T., AND CHEEL, R.J., 1997, The depositional history of the Bloomington Complex,an ice-contact deposit in the Oak Ridges Moraine, southern Ontario, Canada: QuaternaryScience Reviews, v. 16, p. 705–719.

PICKERING, K.T., HISCOTT, R.N., AND HEIN, F.J., 1989, Deep Marine Environments; Clastic Sed-imentation and Tectonics: London, Unwin Hyman, 416 p.

PLINK-BJORKLUND, P., AND RONNERT, L., 1999, Depositional processes and internal architectureof Late Weichselian ice-margin submarine fan and delta settings, Swedish west coast: Sed-imentology, v. 46, p. 215–234.

POSTMA, G., NEMEC, W., AND KLEINSPEHN, K.L., 1988, Large floating clasts in turbidites: amechanism for their emplacement: Sedimentary Geology, v. 58, p. 47–61.

POSTMA, G., ROEP, T.B., AND RUEGG, G.H.J., 1983, Sandy-gravelly mass-flow deposits in anice-marginal lake (Saalian, Leuvenumsche Beek valley, Veluwe, the Netherlands), with em-phasis on plug-flow deposits: Sedimentary Geology, v. 34, p. 59–82.

POWELL, R.D., 1981, A model for sedimentation by tidewater glaciers: Annals of Glaciology,v. 2, p. 129–134.

POWELL, R.D., 1990, Glacimarine processes at grounding-line fans and their growth to ice-contact deltas, in Dowdeswell, J.A., and Scourse, J.D., eds., Glacimarine Environments:Processes and Sediments: Geological Society of London, Special Publication, 53, p. 53–73.

RAJARATNAM, N., 1976, Turbulent Jets: Amsterdam, Elsevier, Developments in Water Science5, 304 p.

RAJARATNAM, N., AND SUBRAMANYAN, S., 1986, Plane turbulent denser wall jets and jumps:Journal of Hydraulic Research, v. 24, p. 281–296.

READING, H.G., ed., 1996, Sedimentary Environments; Processes, Facies and Stratigraphy: Ox-ford, U.K., Blackwell Science, 688 p.

RUSSELL, A.J., AND KNUDSEN, O., 1999, An ice-contact rhythmite (turbidite) succession depositedduring the November 1996 catastrophic outburst flood (jokulhlaup), Skeiararjokull, Ice-land: Sedimentary Geology, v. 127, p. 1–10.

RUSSELL, H.A.J., 2001, The record of high-energy episodic sedimentation in the western OakRidges Moraine, southern Ontario [unpublished Ph.D. thesis]: University of Ottawa, Ottawa,Ontario, 297 p.

RUST, B.R., 1977, Mass flow deposits in a Quaternary succession near Ottawa, Canada: Ca-nadian Journal of Earth Sciences, v. 14, p. 175–184.

RUST, B.R., 1988, Ice-proximal deposits of the Champlain Sea at south Gloucester, near Ot-tawa, Canada, in Gadd, N.R., ed., The Late Quaternary Development of the Champlain SeaBasin: Geological Association of Canada, Special Paper 35, p. 37–45.

RUST, B.R., AND ROMANELLI, R., 1975, Late Quaternary subaqueous outwash deposits nearOttawa, Canada, in Jopling, A.V., and McDonald, B.C., eds., Glaciofluvial and Glaciola-custrine Sedimentation: Society of Economic Paleontologists and Mineralogists, SpecialPublication 23, p. 177–192.

SAUNDERSON, H.C., AND LOCKETT, F.P.J., 1983, Flume experiments on bedform and structuresat the dune–plane bed transition, in Collinson, J.D., and Lewin, L., eds., Modern and AncientFluvial Systems: International Association of Sedimentologists, Special Publication 6, p.49–58.

SHANMUGAM, G., 1997, The Bouma Sequence and the turbidite mind set: Earth-Science Re-views, v. 42, p. 201–229.

SHARPE, D.R., 1988, Glacimarine fan deposition in the Champlain Sea, in Gadd, N.R., ed., TheLate Quaternary Development of the Champlain Sea Basin: Geological Association of Can-ada, Special Paper 35, p. 63–82.

SHARPE, D.R., AND COWAN, W.R., 1990, Moraine formation in northwestern Ontario: productof subglacial fluvial and glacilacustrine sedimentation: Canadian Journal of Earth Sciences,v. 27, p. 1478–1486.

SIMONS, D.B., RICHARDSON, E.V., AND NORDIN, C.F., 1965, Sedimentary structures generated byflow in alluvial channels, in Middleton, G.V., ed., Primary Sedimentary Structures and theirHydrodynamic Interpretation: Society of Economic Paleontologists and Mineralogists, Spe-cial Publication 12, p. 34–52.

SKIPPER, K., AND BHATTACHARJEE, S.B., 1978, Backset bedding in turbidites: a further examplefrom the Cloridorme formation (Middle Ordovician), Gaspe, Quebec: Journal of SedimentaryPetrology, v. 48, p. 193–202.

SMITH, G.A., 1986, Coarse-grained nonmarine volcaniclastic sediment: terminology and de-positional process: Geological Society of America, Bulletin, v. 97, p. 1–10.

SMITH, G.A., AND LOWE, D.R., 1991, Lahars: volcano-hydrologic events and deposition in thedebris flow–hyperconcentrated flow continuum, in Fisher, R.V., and Smith, G.A., eds., Sed-imentation in Volcanic Settings: SEPM, Special Publication 45, p. 59–70.

SOHN, Y.K., 1997, On traction-carpet sedimentation: Journal of Sedimentary Research, v. 67,p. 502–509.

SPOONER, I.S., AND DALRYMPLE, R.W., 1994, Sedimentary facies relationships in esker-ridge/esker-fan complexes, southeastern Ontario, Canada: Application to the exploration for as-phalt blending sand: Quaternary International, v. 20, p. 81–92.

VROLIJK, P.J., AND SOUTHARD, J.B., 1997, Experiments on rapid deposition of sand from high-velocity flows: Geoscience Canada, v. 24, p. 45–54.

WALKER, R.G., 1992, Turbidites and submarine fans, in Walker, R.G., and James, N.P., eds.,Facies Models—Response to Sea Level Change: St John’s, Newfoundland, Geological As-sociation of Canada, p. 239–264.

WAN, Z., AND WANG, Z., 1994, Hyperconcentrated Flow: Rotterdam, A.A. Balkema, 290 p.WARBURTON, J., AND FENN, C.R., 1994, Unusual flood events from an Alpine glacier: observa-

tions and deductions on generating mechanisms: Journal of Glaciology, v. 40, p. 176–186.WARREN, W.P., AND ASHLEY, G.M., 1994, Origins of the ice-contact stratified ridges (eskers) of

Ireland: Journal of Sedimentary Research, v. A64, p. 433–449.WEIRICH, F.H., 1989, The generation of turbidity currents by subaerial debris flows, California:

Geological Society of America, Bulletin, v. 101, p. 278–291.WRIGHT, L.D., 1977, Sediment transport and deposition at river mouths: A synthesis: Geolog-

ical Society of America, Bulletin, v. 88, p. 857–868.WU, S., AND RAJARATNAM, N., 1995, Free jumps, submerged jumps and wall jets: Journal of

Hydraulic Research, v. 33, p. 197–212.

Received 10 June 2002; accepted 11 April 2003.

Documents

Hydraulic-Jump and Hyperconcentrated-Flow Deposits of a