Cant 1976

Development of a braided-fluvial facies model for the Devonian
Battery Point Sandstone, Qubbec

DOUGLAS J. CANTAND ROGERG . WALKER
Department of Geology, McMaster Universily, Hamilton, Ontario L8S 4MI
Received 29 May 1975
Can. J. Earth Sci. Downloaded from www.nrcresearchpress.com by University of Sydney on 08/20/13
Revision accepted for publication 15 September 1975
Eight distinct facies have been defined in a 110-mthick section of the Lower Devonian Battery
Point Sandstone near Gaspe, Quebec. The first is a scoured surface overlain by massive sand-
stone with mudstone intraclasts. Facies A and B are trough cross-bedded sandstones, with
poorly- and well-defined stratification, respectively. Facies C and D consist of large isolated, and
smaller multiple, sets of planar cross-stratified sandstones, respectively. Facies E comprises
large sandstone-filled scours, facies F comprises ripple cross stratified fine sandstones with
interbedded mudstones, and facies G comprises sets of very low angle cross-stratified sand-
stones.
The overall context of the Battery Point Sandstone, the presence of rootlets, and the abun-
dance of trough and planar-tabular cross bedding, all suggest a generally fluvial environment of
deposition. Analysis of the facies sequence and interpretation of the primary sedimentary
structures suggest that channel development began by scouring, and deposition of an intraclast
lag. Above this, the two trough cross bedded facies indicate unidirectional dune migration
downchannel (vector mean direction 291"). The large planar tabular sets are associated with the
trough cross bedded facies, but always show a large (almost 90") paleoflow divergence, suggesting
For personal use only.
lateral movement of in-channel transverse bars. The smaller planar tabular sets occur higher
topographically in the fluvial system, and the rippled silts and muds indicate vertical accretion.
Because of the very high ratio of in-channel sandy facies to fine-grained vertical accretion facies,
and because of the evidence of lateral migration of large in-channel bars, the Battery Point River
appears to resemble modern braided systems more than meandering ones.
Huit facies distincts ont ete definis dans une section de 110 m d'kpaisseur de gres du Devonien
inferieur de Battery Point, prts de Gaspe, Quebec. Le premier est une surface d'erosion
sumontee de gres massif afragments de mudstone. Les facies A et B sont des grts a stratifications
entrecroisees du type 'fosse'; ces stratifications sont peu evidentes dans le premier de ces facies,
tres nettement marquees dans le second. Les facies C et D consistent en series de gres a
stratifications entrecroisees du type 'plan', tpaisses et isolees pour C, minces mais multiplespour
D. Le f a d s E comprend des chenaux a remplissage de gres, le facies F des gres fins a
ripple-marks, et le facies G des ensembles greseux a stratifications entrecroisees sous un angle
trks faible.
Le contexte general du gres de Battery Point, la presence de radicelles et l'abondance de
stratifications entrecroisees de type 'fosse' et de type 'plan', tout suggere un milieu de
sedimentation fluviatile. L'analyse sequentielle des facies et des structures sedimentaires
syngenetiques montrent que la formation du chenal dkbuta par une forte erosion, laissant sur
place des fragments intraclastiques. Au-dessus, les deux facies a stratification entrecroisee du
type 'fosse' indiquent la migration vers I'aval d'une dune unidirectionnelle (direction moyenne:
2919.Les grands ensembles a stratificationentrecroisee de type 'plan' sont associds avec ceux du
type 'fosse', mais montrent toujours une grande divergence de direction d'ecoulement (pres de
90") ce qui suggtre un mouvement lateral de barres transversales.
Les ensembles de lits entrecroises plans se presentent plus haut dans latopographie du systeme
fluviatile, et les silts et boues a ripple-marks indiquent une accretion verticale.
La proportion tres elevke des facits sableux par rapport aux facies fins d'accrktion verticale et
l'evidence d'une migration laterale de grandes barres interieures font que la Rivitre de Battery
Point Cvoque plus un systeme anastomose actuel qu'une riviere 2 meandres.
[Traduit par le journal]
Problems of Fluvial Depositional Models of the area (McGerrigle 1950), and the known
During reconnaissance work in 1971, it ap- occurrences of in situ plant remains and fresh
peared that the Battery Point Formation (Fig. water fish fossils. It was stengthened by our
1) was dominantly fluvial in origin. This im- finding classical "fining-upward sequences" of
pression was based upon previous descriptions the alluvial type (Allen 1965) in the upper part of
Can. J. Earth Sci., 13, 102-1 19 (1976)
CANT AND WALKER: BRAIDED F L U V I A L FACIES MODEL 103
Regional Geology
The Devonian rocks of Gasp6 are structurally
LOWER DEVONIAN part of the GaspbConnecticut Synclinorium.
CARBONATES The spectacular vertical cliffs at the tip of the
BATTERY POINT Gasp6 Peninsula in Forillon National Park are
composed of Lower Devonian marine limestones
about 2000 m thick, and they are followed by
three formations collectively known as the Gasp6

Sandstones. The York River Formation (Lower
Devonian) is about 330 m thick and contains
many marine fossils. It has been omitted from
the map (Fig. 1) for the sake of clarity. The
-
0 MILES 6
Battery Point Formation (Lower Devonian) is
mostly nonmarine, about 2000 m thick, and is
composed mainly of sandstones and mudstones,
0 3 6KM with subordinate conglomerate. The highest unit,
the Malbaie Formation (Middle Devonian), is
mostly conglomerate, with some sandstone and
FIG.1. Sketch map of geology of Gasp6 area. Letters shale. It is about 830 m thick, and is nonmarine.
C.d.R. indicate Cap des Rosiers; CO, Cap aux 0 s ; P,
Penouille. Studied section is 7 km southeast of Cap aux It crops out south and west of the area shown in
0s. within Forillon National Park. the map.
No regional mapping has been undertaken
since the study of McGerrigle (1950). He set up

the formation (at Penouille, P on Fig. l), formed the present stratigraphic terminology, and de-
by lateral and vertical accretion in a meandering scribed each formation, but no interpretation was
river environment. attempted. The age of the Battery Point Forma-
By contrast, the very well exposed coastal tion was given as Middle Devonian, but this has
section in Forillon National Park (arrowed on been revised to Lower Devonian by Boucot et al.
Fig. 1) appeared rather different. The vertical- (1967). A regional guidebook to the Gasp6 area
accretion siltstones and mudstones noted a t has been prepared by St. Julien et al. (1972).
Penouille were almost entirely absent; ripple There is no published paleocurrent data on the
cross-laminated sandstones were volumetrically Battery Point Formation-the southwesterly
unimportant; and the bulk of the section was pointing arrow in Allen and Friend (1968, fig. 11)
composed of various types of cross-bedded sand- appears to refer to Upper Devonian rocks of the
stones. It therefore seemed important to define Escuminac Bay area.
the different facies present at Forillon, and to The Battery Point Formation is part of the
see whether they occurred in a specific sequence North Atlantic 'Old Red Sandstone', forming
that could be compared with the classical mean- part of the clastic wedge shed from the rising
dering river "fining-upward" sequence. Our ulti- Acadian (Caledonian) mountains. In North
mate aim was to extend the understanding of America, the nearest desposits similar to those of
fluvial depositional models to account for the the Gasp6 and Chaleur Bay areas occur over 650 mi
Forillon deposits. Also, we hoped to isolate from (> 1046 km) away in New York State, where the
the local model those features that might have dominant transport direction is westward, away
general significance, and hence a broader appli- from the Acadian mountains. In the pre-Con-
cation to the interpretation of other fluvial tinental drift best-fit maps of Bullard et al. (1965),
deposits. the Gasp6 area is almost exactly halfway between
-Because the Battery Point formation is com- the New York State area and southwest Ireland,
posed mainly of sandstones, we have restricted where very great thicknesses of Devonian rocks
our discussions and comparisons here to modern are preserved. A better understanding of the
and ancient sandy systems. Coarser braided facies and paleocurrent directions of the Gasp6
streams have been reviewed recently by Eynon Devonian is therefore important in an overall
and Walker (1974), where a full bibliography of understanding of North Atlantic 'Old Red
pebbly and gravelly streams is given. Sandstone' deposition and continental evolution.
104 CAN. J. EARTH SCI . VOL. 13, 1976
This study is restricted to the 110 m section

exposed along the shoreline in Forillon National
Park. The park can be reached from Qutbec
Highway 6 (Fig. 1); after entering the Park gates,
follow the highway to the end, and then walk
westward along the pebble beach to the bottom
of the section. The dip is 35" SW. The fairly
narrow outcrop at the foot of the cliffs gives a

vertical section through the beds, that is more or
less continuous, although broken by a few small
faults (Fig. 2). At the top of our 110 m section,
the outcrop becomes discontinuous and is un-
suitable for the type of facies analysis discussed
below.
Facies Descriptions
To attempt the description and interpretation
of the 110 m thick sequence (Fig. 2) in one piece
is clearly crude and imprecise. It is necessary to
subdivide the sequence, if possible, into smaller,
homogeneous 'pieces' that might occur only
once, or may be repeated many times within the
110 m sequence. Subdivision into 'pieces', or

facies, is based upon careful measurement of the
section and documentation of all physical and
biological features. Facies were only defined
formally after we were very familiar with the
section, and hence with the variability associated
with each of our proposed facies-that is, after
we had some appreciation for how different, or
how intergradational the facies were. Following FIG.2. Measured section through part of the Battery
de Raaf et al. 1965, we use the term 'facies' for a Point Formation. Zero datum is about 50 to 100 m above
unit of rock distinguished in the field by litho- the base of the formation. Facies lettering scheme is ex-
plained in the text. Blank portions of the column repre-
logy, sedimentary structures, and organic fea- sent small faults or covered intervals; we do not believe
tures from adjacent facies, both vertically and that there are any major dislocations in this part of the
laterally in the section. The facies will be inter- area, but the indicated thicknesses where faults or cover
preted after the facies sequence has been estab- are present may be inaccurate. Ten generally fining-up-
lished, so that the interpretation can be based ward sequences, defined by scoured surfaces, are indica-
ted; above about 86 m the outcrop becomes too dis-
upon their specific characteristics and their over- continuous to define more sequences. The sketched pro-
all context. jection of the beds to the right is an approximate measure
of mean grain size; maximum projection indicates coarse
Scoured Surfaces ( S S ) sandstone (about 1 mm).
The scoured surface facies (Fig. 3) consists of
erosion surfaces overlain by layers of massive, ited. The scoured surfaces are not facies in quite
coarse sandstone with abundant mudstone clasts. the same sense as the facies described below, but
The maximum depth of erosion observed was they are recognizable 'states' within the sedi-
about 5 m. Overlying the erosion surface, the mentation sequence, and will be analyzed as such
mixture of mudstone clasts and coarse massive below.
sandstone reaches a maximum thickness of about
25 cm. As well as the major scoured surfaces, Poorly Dejined Trough Cross-Bedded Facies ( A )
there are other intraclast-strewn layers (com- This facies (Fig. 4) rests upon scoured, intra-
monly within facies A or B) that represent minor clast-strewn surfaces (SS, Fig. 3), and is com-
breaks in sedimentation as the clasts are depos- posed of poorly defined sets of trough cross
CANT AND WALKER: BRAIDED-FLUVIAL FACIES MODEL 105
FIG. 3. Scoured surface (SS), cutting down about 5 rn stratigraphically from top left to person's
hand, lower right. Backpack rests on trough cross bedding (facies B); above and to the left is a prornin-
ent isolated scour (facies E) overlain by rippled sandstone and siltstone, with interbedded mudstone
(facies F). Photo from 5 to 10 rn on Fig. 2.
FIG. 4. Poorly defined trough cross bedding, facies A. Crude troughs can be seen, particularly
below the 30 crn scale. This is an 'end-member' example of facies A, and with improving definition of
the troughs, and with better sorting, facies A grades into facies B. Photo at 62 rn, Fig. 2.
106 CAN. J. EARTH SCI. VOL. 13, 1976
FIG.5 . Well defined trough cross bedding, facies B. Scale is 16 cm long. Photo at 96 m, Fig. 2.
bedding, with trough depths averaging about 30 troughs are regularly stacked on top of each
cm (range 15 to 60 cm). In most occurrences of other, but in some individual occurrences of the
the facies, the troughs are regularly stacked on facies, trough depths decrease upward. Trough
one another, with uniform depth of scour. widths varv from 30 cm to 1.5 m. Individual
Trough width is variable (up to 3 m), and indi- troughs ha;e been traced downcurrent for up to
vidual troughs have been traced downcurrent for 6.2 m, and in plan view, the internal cross strata
up to 10 m. In plan view, the very wide troughs are almost semi-circular. In vertical section~aral-
have straight internal strata with curvature only lel to flow, the cross strata curve only slightly at
at trough margins-the smaller troughs have the toe, giving an appearance similar to planar-
more arcuate internal strata. tabular cross bedding. In at least one occurrence,
The internal strata are poorly defined because the sets climb downcurrent at about 5" to re-
of the poor sorting and lack of fine material gional bedding.
necessary to show up the cross strata. Pebbles up The sets are composed of well sorted medium
to 5 cm in diameter are scattered through the sand, and the well defined cross strata are shown
troughs, rather than concentrated at their bases, up by fine, darker colored layers. Also, many
or else form discontinuous inclined strata, which individual cross strata are graded, further em-
rarely persist through the total thickness of the phasizing the layering. The mean grain size is
set. Mudstone intraclasts up to 5 cm diameter about 0.4 mm (medium sand), ranging from
occur at the bases of some troughs, rendering the about 0.25 mm to 0.8 mm (coarse sand) in
internal organization of the troughs even poorer. different occurrences of the facies. A few of the
The mean grain size of the sand is about 0.75 coarser sets have granules and pebbles concen-
mm (coarse), but the mean ranges from about trated at their bases.
0.5 to 1.0 mm in different occurrences of the We believe there is a gradation from the poorly
facies. Individual occurrences of the facies vary to well defined trough cross bedded facies, and
in thickness from about 60 cm to 4.3 m. some occurrences were difficult to assign posi-
tively to either one.
Well Defined Trough Cross-Bedded Facies (B)
This facies is composed of well defined sets of Large Scale, Planar-Tabular Cross-Bedded Facies
trough cross bedding (Fig. 5), with trough depths (C)
averaging 15 to 20 cm (range 10 to 45 cm). The The planar-tabular sets comprising this facies
FIG. 6. Isolated set of planar-tabular cross bedding, facies C. Note straight foresets and angular
bottom contacts. Scale is 16 cm long, and set is about 1 m thick. Photo at 19 m, Fig. 2.
occur singly, or in cosets of up to 4 sets (Fig. 6). mm), and in the rare sets where pebbles are
Sets average about 60 cm in thickness (range 30 present, they are scattered within the set rather
cm to 3 m). The sets are extensive, with planar than being concentrated at the base.
foresets having been observed up to 18 m wide,
and sets persisting in the downcurrent direction Small Scale Planar-Tabular Cross-Bedded Facies
for at least 20 m. Every occurrence of this facies (D)
(except one) is developed on a very flat surface This facies is composed of small scale, planar-
that truncates one of the two troughed facies tabular sets of cross-bedding (Fig. 8), with indi-
(A or B) (Fig. 7). The exception rests on an vidual sets ranging in thickness from 12 to 30
intraclast-strewn scour surface (SS). The tops of cm. Cosets are composed of between 2 and 10
the planar-tabular sets are also flat truncation individual sets, averaging 5 or 6 sets. Each set
surfaces, commonly with large scours cut into is very persistent in a downcurrent direction, and
them both parallel and perpendicular to the most can be traced for at least 10 m. Within
paleocurrent direction of the planar-tabular set cosets, paleocurrent directions tend to be fairly
(paleocurrents are discussed later). In vertical constant, with a maximum range of about 20" on
section parallel to flow, the sets have straight either side of the mean.
foresets with sharply angular bottom contacts. The small scale cross-bed cosets are developed
The foreset geometry remains constant in the on flat truncation surfaces on top of facies B or
downstream direction in any one set, except C. They are normally truncated above by flat
where reactivation surfaces truncate the fore- surfaces, and more rarely by irregular scours.
sets. In plan view, the foresets are straight across Cross-bedding within sets is normally sharply
the entire width of outcrop. angular without reactivation surfaces, and the in-
The inclined foresets are shown up either by clined strata are shown up by thin, darker,
thin partings of finer sand, or because of normal slightly finer layers.
grading from bottom to top of individual in- The sets are composed of well sorted, fine to
clined layers. The layers vary from I to 8 cm in medium sand, with mean sizes ranging from
thickness. Grain size is generally in the coarse about 0.25-0.5 mm in different cosets. As indi-
sand range (average 0.75 mm, range 0.5 to 1.0 vidual sets become thicker, there is a transition
108 CAN. J. EARTH SCI. VOL. 13. 1976
FIG. 7. Isolated set of planar-tabular cross bedding (facies C, top of photo) overlying broad
troughs (facies B). Paleoflow for trough cross bedding is toward top left, almost exactly perpendicular
to paleoflow of the planar tabular set (toward right). Scale is foreground 25 cm long. Photo at 25.5 m,
Fig. 2.
FIG. 8. Multiple sets of small scale planar-tabular cross bedding, facies D. Scale is 15 cm long.
Photo at 26.5 m, Fig. 2.
FIG.9. Large isolated scour, facies E, with stratification filling the scour conforming to the shape
of the lower surface. Scour cuts into well defined trough cross bedding (facies B). Dark bars on note-
book each 5 cm, numbers are tenths of feet. Photo at 6 m, Fig. 2.
into facies C, the large scale planar-tabular cross- (mean diameters range from 0.25 to 0.75 mm),
bedding, and sets about 30 cm in thickness could but some have pebbly sand fills. One scour fill is
probably be assigned to either facies. coarse and pebbly at the upstream end, fining to
well sorted medium sand at the downstream knd
Asymmetrical Scour Facies (E)
of the scour.
This facies consists of large, asymmetrical
scours and scour fillings, up to 45 cm deep and Rippled Sandstone and Mudstone Facies (F)
3 m wide (Fig. 9). The scours cut into each other, This facies includes cross-laminated sand-
and into underlying troughed facies (A and B), stones (Fig. lo), and alternating cross-laminated
and occurrences of the asymmetrical scour facies sandstones and mudstones (Fig. 11).
have a flat, erosionally truncated top. The shape The two occurrences of ripple cross-laminated
of the scours ranges from spoon shaped to flat sandstones without mudstones are 45 and 90 cm
bottomed, with some very shallow, almost flat thick, respectively. One consists of tabluar sets
scours. of cross-lamination, and the other consists of
The main difference between the scour fillings trough cross-lamination with abundant organic
and the two troughed facies (A and B) lies in the material.
geometry of the infilling strata. In the asymmetri- There are two occurrences of the interbedded
cal scours, the layers are not at the angle of fine sandstones (mean size < 0.25 mm) and
repose, but are parallel to the lower bounding mudstones. The first is 1.5 m thick and consists
surface. The layers show up by darker, finer of three coarsening-upward sequences, which
partings, and within some scours there are minor grade from basal mudstones into trough cross-
erosion surfaces and 2-3 cm sets of cross-lamin- laminated fine sandstone, and finally into granule
ation. In some places where the layers can be sandstone. The sandstones capping each coarsen-
seen in plan view, a moderately well developed ing-upward sequence have sharp, bioturbated
parting lineation is oriented parallel to the in- tops. The second occurrence is also 1.5 m thick,
ferred axes of spoon-shaped scours. but consists of at least five sandstone-to-mud-
Most scours are filled with well sorted sand stone fining-upward sequences. The sandstones
110 CAN. J. EARTH SCI. VOL. 13. 1976
FIG.10. Small scale cross laminated sandstones, facies F. Scale is 30 cm long. Photo at 27 m,
Fig. 2.
have sharp bases and commonly contain climb- are truncated by erosional intraclast-strewn sur-
ing ripples with eroded stoss sides. Other sand- faces.
stones are structureless, probably due to bio- Within each set, the sweeping slightly curved
turbation. layers are very continuous laterally, and in
almost all cases have an original dip of less than
Low Angle Strat$ed Sandstone Facies ( G )
10", flattening to horizontal at the base of each
This facies sets 2, up set. Individual layers are about 1 cm thick, and
8) of low angle (< 10") stratified sandstones are well defined by alternating finer layers.
ranging in thickness 30 90 cm (Fig. 12). In places, the top surfaces of the layers show
The sets are commonly continuous across the
a well developed parting lineation, which is
width of Outcrop (minimum lo m), but in parallel to the depositional strike of the inclined
places truncate each other vertically and later- layers. The grain size is fine sand, with mean
ally. In one occurrence the sets change laterally grain sizes in the range O.10 to 0.25 mm.
to more trough-like shapes, and in a different minuted plant material is abundant.
occurrence the single set resembles a very shallow,
very wide, trough-cross-bed. Each occbrrence is Petrology of the Sandstones
developed on a surface that may be flat or The objectives of our work did not include a
eroded into shallow scours, and all occurrences detailed study of petrology. The average com-
FIG.11. Alternating layers of fine siltstone and ripple cross laminated fine sandstone (facies F),
truncated by a massive sandstone with mudstone intraclasts (scoured surface, SS). Dark bars on note-
book each 5 cm, numbers are tenths of feet. Photo at 9 m, Fig. 2.
position of five point-counted thin sections from TABLE1. Average composition of five
different levels in the 120 m sequence is shown in sandstone specimens
Table 1; this composition falls into the sub-
arkose class of Folk (1968). Constituent Percentage
The cement is dominantly clay, probably kao- Quartz 66
linite, but there is also some quartz cementation Orthoclase 10
that is younger than the clay. Hematite is dis- Volcanic rock fragments 7
persed throughout the cement. The analyses sug- Sedimentary rock fragments 5
Chert 3
gest a 'granite'-rich source area, with some vol- Hematite 3
canic contribution: this is consistent with an Perthite 2
Acadian-Caledonian mountain source. Plagioclase 2
Cement 1
Facies Sequence: Analysis and Distillation Others* 1
Having defined eight facies, we are faced with 'Includes metamorphic rock fragments, zircon, mus-
covite, biotite, chlorite, pyrite, and epidote.
the problem of analyzing their sequence and, if
possible, distilling their sequence into a general- judgements, and where our conclusions can be
ized or idealized form that can be used as a basis substantiated statistically.
for a facies model. Many models have been pre- There are two main ways that the raw facies
sented in the literature without any real discus- sequence (Fig. 13) can be visualized. The up-
sion of their formulation. We feel that it is im- ward transitions can either be shown with arrows
portant to discuss how our model is formulated, in a 'Facies Relationship Diagram' (Fig. 14), or
and to show where we are making subjective they can be tallied and converted into probabili-
FIG.12. Below notebook, interbedded fine siltstones and sandstones of facies F. Immediately
above notebook, a fine sandstone (60 cm) shows low angle inclined stratification(facies G),and is trun-
cated irregularly by coarser sandstones showing poorly and well defined trough cross stratification
(facies A and B). Photo at 37 m, Fig. 2.
ties, shown in Table 2. This transition probabil- D.A,B,E.F.B.A,G.B.C,A,C,B,C,B.D,F/

ity matrix is concise, and has been used by many A,B.D,F,G. A,C,A/ A.A/C.B.G. A,B,
workers; the most recent application to fluvial A,B,G. A..B. A,E.C,B,F. A , B . A,C,D/
sediments was by Mia11 (1973), who cites most of A/B.A. A/G.C
the earlier literature. The main objection to the
use of such a matrix is that it does not show the . SS, SCOURED SURFACE
nature of the transitions (gradational, sharp, , SHARP CONTACT
channelled), but in the Battery Point Formation ,, GRADATIONAL CONTACT
where sharp and channelled contacts predomin- / FAULT OR COVERED INTERVAL
ate, this is not a serious objection (see de Raaf

et al. 1965, fig. 22). FIG. 13. Sequence of facies observed in the study
area. From stratigraphic base, upward, the sequence reads
In order to comprehend the facies relationship D, A, B, E, etc.
diagram (Fig. 14), some simplification is neces-
sary. There are at least three possible approaches: was first suggested by Selley (1970; see also
(1) arbitrary rejection from Figure 14 of those Harms et al. 1975), and has been refined by Miall
transitions that occur only once; (2) arbitrary re- (1973). From the raw data, a transition proba-
jection from Table 2 of those transitions that bility matrix was calculated for the observed
have a low probability of occurring; and (3) transitions (Table 2). Then a second matrix was
examination of the data to see which transitions calculated assuming the same abundance of
occur more, and less, frequently than if the facies facies but in a random sequence; thus, the
were in a random sequence. The third approach probability of transition from facies A to any
CANT AND WALKER: BRAIDED-FLUVIAL FACIES MODEL
TABLE2. Observed transition probabilities

TABLE3. Transition probabilities for random sequence

OBSERVED FACIES RELATIONSHIPS -- monly certain transitions occur (Fig. 15). By

trial and error, it appears for the Battery Point
data that a reasonable amount of simplification
occurs if only the entries in the difference matrix
(Table 4) greater than +0.05 are shown in
Fig. 15. Lower difference values indicate transi-
tions little different from random. Because our
purpose is simplification, we have not shown
such transitions in Fig. 15, although we are
FIG. 14. Facies relationship diagram showing all
observed transitions. Numbers indicate how many times --
a particular upward transition occurred. PREFERRED RELATIONSHIPS IF FACIES ARE NOT RANDOM
VERTICAL
other facies depends only upon the observed ACCSIUIOH
relative abundances of the other facies (Table 3). -

I' ss "'P, #'
A chi-squared test showed that the observed

facies transitions were statistically significantly
different from random. A difference matrix was
04-
nw
\ --.
B----+ /D
-
I-
then calculated (observed minus random), which ---------
IN-CHANNEL
,
'
highlights those transitions that have a higher or e m - TOP*
lower probability of occurring than if the se-

FIG. 15. Facies relationship diagram, showing only
quence were random kcomputational those transitions that occur more commonly than ran-
details in Miall 1973). dom. Diagram is derived from data in Table 4, and only
We can now reconstruct a facies relationship shows transitions whose observed-minus-predicted tran-
diagram showing the transitions that occur more sition probabilities exceed 0.05. Dotted arrows show
frequently than random. ~h~ nature of the tran- probabilities in the range 0.05 to 0.10; light solid arrows,
0.10 to 0.30; heavy solid arrows, probabilities > 0.30.
sitions can be shown by reference to the raw Four facies associations have been interpreted and cir-
data, and different weights of arrows can be cled on the diagram: the asterisk against Bar Top is
used to show subjectively how much more corn- explained in theiext.
TABLE4. Observed minus random transition probabilities
SS A R C D E F G
SS +0.480 -0.112 -0.084 -0.075 -0.038 -0.075 -0.094
A -0.126 +0.202 +0.071 -0.080 +0.037 -0.080 -0.023
B +0.049 -0.238 +0.006 +0.080 +0.040 +0.003 +0.061
C -0.237 -0.002 +0.351 +0.075 -0.034 -0.068 -0.085
D +0.111 -0.270 -0.206 -0.127 -0.032 +0.604 -0.079

E -0.215 -0.262 -0.200 +0.377 -0.062 +0.438 -0.077
F +0.445 -0.270 -0.206 -0.127 -0.063 -0.032 +0.254
G t0.774 -0.274 -0.210 -0.129 -0.063 -0.032 -0.065
aware (and will continue to be aware in our in- BATTERY POINT

terpretation) that we have arbitrarily rejected MODEL SEQUENCE 5 "MEANDERING"
some of the data.

Facies and Facies Sequence: Interpretation and
Model Building
We consider here the interpretation of Fig. 15,
and the way in which this interpretation, along
with interpretations of individual facies, can be
presented as a local model.
First, it is clear in Fig. 15 that, above the

scoured surface, facies A, B, and C (trough
cross-bedded and planar-tabular cross-bedded
sandstones) tend to occur in close mutual associ-
ation. The trough cross-bedded sandstones
(facies A and B) represent unidirectional migra-
FIG. 16. The Battery Point model sequence is a
tion of dunes, and the planar-tabular sets prob- sketch of the data in Fig. 15, showing average facies
ably represent movement of larger bed forms. thicknesses as well as facies sequence. Facies E is omitted
The occurrence of all three facies above intra- because it does not occur a sufficient number of times to
clast-strewn scoured surfaces suggests deposition allow its typical position to be sketched. Paleocurrent
arrows are averages of all the data applicable to each
within a channel. This interpretation will be facies: the grand vector mean for facies A and B together
elaborated below. is 291" (standard deviation 40"). The 'meandering' fining-
The smaller scale of the cross bedding in facies upward sequence is drawn to the same scale, using the
D, together with its tendency to occur above data on thicknesses and sequence of structures given by
facies B and C (Fig. 15), suggests deposition in Allen (1970). V.A. indicates vertical accretion, and within
the V.A. deposits, calcareous concretions, desiccation
topographically higher parts of the fluvial sys- cracks, and rootlets are shown symbolically.
tem, perhaps as small bed forms migrating on
top of larger sand flats or bars in fairly shallow
water, or in the uppermost parts of channels frequently than random. This relationship of E
after most of the channel has been filled. For with A, B, and C suggests that the formation of
brevity, both of these topographically higher en- the asymmetrical scours takes place in-channel
vironments have been combined as bar top* in rather than in topographically higher parts of the
Figs. 15 and 16. The asterisk is intended to re- system (scouring on the tops of point bars, for
mind readers that more than just bar top is example, has been described by Harms et al.
implied. 1963).
The particular usefulness of this type of facies Finally, Fig. 15 shows that facies F and G
analysis shows up in the interpretation of facies consistently occur stratigraphically above all
E, the large asymmetrical scours. In Fig. 15, facies other facies, and are consistently followed by
E is related to facies C and F ; a re-examination scoured surfaces. The alternation of cross-
of the difference matrix (Table 4) also shows that laminated fine sandstones and siltstones with
the A + E and B + E transitions occur more mudstones (facies F) suggests vertical accretion
CANT AND WALKER: BRAIDE:D-FLUVIAL FACIES MODEL 115
in disused parts of the floodplain, with subse-

quent erosion and re-establishment of an active
channel. Facies G has enigmatic sedimentologi- Care must be taken with azimuths close to
cal characteristics, but because of its consistent north: with hypothetical readings of 350" and
position in the facies sequence and its upper flow 010, A@ is 20, not 340!
regime features, it must represent vertical flood The average value of AOij (written a i j ) for
stage accretion of sand in an overbank or non- all facies pairs (i directly followed by j) can be
calculated from
channelled area.
The transitions shown in Fig. 15 can be ex-
pressed as a stratigraphic column (Fig. 16). This
figure attempts to show the characteristic thick-
nesses of the various facies, and the number of where n is the number of times facies i is followed
times they can be interbedded in a typical Battery directly by facies j, where paleocurrent data is
Point facies sequence. Not all sequences are so available for both facies. The matrix for the
complete (see Fig. 2); for example, the sequence Battery Point facies is shown in Table 5. The
SS -r A -+ B SS implies channel cutting, dune
-+ highest angular divergences (excluding single
migration, and then renewed cutting before any observations) are B -+ C, C A, and C B.
-+ -+
of the topographically higher facies could be This indicates a very strong tendency for the
formed or preserved. Facies E has been omitted single planar-tabular sets of facies C to occur
from the stratigraphic column in Fig. l b t h e above B at a high angle, and to be followed in
two occurrences of the facies are in different turn by facies A or B at a high angle.
contexts in the facies sequence, and hence, we do These relationships can be interpreted to imply
not have enough consistent data to idealize its that the planar-tabular cross-bedding represents
position in Fig. 16. the deposits of some form of bar or 'microdelta'
that grows laterally in an active channel (B -+
Paleocurrent Directions, and the Use of A@ C). After construction of the planar-tabular set,
In the field, paleocurrent directions were small dunes resume migration down the channel
measured directly on such features as plan views in their original direction (C -+ B). The planar-
of trough axes, and direction of dip of foresets of tabular sets probably represent lateral growth
planar-tabular cross-bedding. For any one of an existing mid-channel transverse bar. This
occurrence of an individual facies, individual type of bar growth has been documented in de-
readings were averaged to give a single direction tail for the Platte River by Smith (1971, 1972).
associated with the deposition of that facies. These paleocurrent data are interesting, because
The bulk of the data (Fig. 16) is from the they again suggest that braided streams have
trough and planar cross-bedded facies. Com- relatively high paleocurrent variances. This point
bining facies A and B, 32 occurrences of the was emphasized by Smith (1971, 1972), but was
facies show a grand vector mean flow direction criticized on theoretical grounds by Banks and
of 291" (standard deviation, 40"). Because these Collinson (1974), who discussed the preserv-
are the basal in-channel facies, their flow direc- ability of many of the features observed by Smith.
tions are probably close to that of the regional However, deposits formed by lateral bar growth
paleoslope, which therefore had a west-north- appear to be preserved in the Battery Point
west dip. formation, and hence we agree with Smith (1972,
Facies C (planar-tabular cross-bedded sand- 1974) that lateral bar growth is important, and
stones) characteristically has a paleoflow direc- that it gives rise to relatively high paleocurrent
tion that is at a high angle to the immediately variances.
underlying and overlying facies (commonly facies
B). This angular difference (Fig. 16) can best be Nature of the Battery Point River: Comparison
demonstrated by using a new way of presenting With Other Fluvial Systems
paleocurrent data-that is, the matrix. If an The combination of the overall stratigraphic
occurrence of facies i (flow direction Oi) is setting, the presence of rare rootlets, and the
directly followed above by facies j (flow direction scoured surfaces overlain by trough cross bed-
Oj), then AOij is the absolute value of the ding (with some preservation of vertical accre-
difference in directions : tion deposits) indicate very strongly that the
TABLE
5. AOmatrix, showing average paleocurrent divergences between different facies
(see text for details)
NOTE: 50 facies transitions are represented in this table; readings marked * represent single observations, all
others are averages.
section of Battery Point studied represents a sandy braided deposits with which the Battery
fluvial environment. Fortunately, there exists a Point can be compared. The most important
well developed generalized facies model for studies of modern sandy braided streams are those
shallow meandering rivers that can be used as a of Doeglas (1962), Collinson (1970), and Smith
basis for comparison. The latest and most (1 970, 1971, 1972). All give excellent descriptions
general statement of this meandering sequence of the various morphological elements in the
has been given by Allen (1970), and is repro- river, and discuss their formation, but the three
duced to scale in Fig. 16. Deposition takes place authors do not emphasize the stratigraphic se-
above a channel floor lag by-lateral accretion of quence that might form if the deposits were pre-
a point bar. The upper part of the bar is rippled, served. In ancient rocks, braided (or low sinu-
and is overlain by vertical accretion deposits. osity) channel interpretations have been sug-
It can be seen that the Battery point local gested for Devonian rocks in Spitsbergen
model differs considerably from the meandering (Moody-Stuart 1966) and part of the Late
fluvial model. The major differences are: Pennsylvanian Coal Measures of South Wales
(1) The Battery Point is characterized by pres- (Kelling 1968). Kelling emphasizes fining-up-
ervation of a thick sandstone sequence, the bulk ward sequences, with a typical one beginning
of which represents the deposits of sinuous- with a lag gravel, and passing u p into cross-
crested dunes moving on the channel floor. stratified sandstones, ripple-laminated sand-
(2) Along with the abundant trough cross- stones, flat laminated siltstones, mudstone,
bedding, the Battery Point contains thick, isola- underclay, and coal (Kelling 1968, fig. 4). The
ted, planar-tabular cross-bed sets. They always paleocurrent vectors for the cross-stratified
have high paleocurrent divergences from the sandstones agree fairly closely with the axial
channel floor trough cross-bedding. Thick planar trend of the channels, but higher in the cycle, there
tabular sets do not appear to be an important is considerable paleocurrent variance. This is
part of the meandering river model. comparable with the Battery Point (Fig. 16),
(3) There is relatively little preservation of especially with respect to the alignment of the
vertical accretion deposits in the Battery Point trough cross bedding with the assumed channel
river system. trend, but the up-to-180" divergences in ripple
We believe that these differences are suffi- orientation are greater in the Coal Measures than
ciently great that the Battery Point local model in the Battery Point.
cannot be considered as a mere variant of the Because of the limited number of studies, we
general meandering model. The abundance of are unable to say how the Battery Point com-
trough cross-bedding, the evidence of mid-chan- pares with other sandy braided deposits. We are
nel transverse bars with lateral growth of planar- also unable to locate the Battery Point in the
tabular sets of cross-bedding, and the meagre spectrum between highly sinuous meandering
preservation of vertical accretion deposits all river types, and highly anastomosing braided
suggest a braided, or low sinuosity pattern for types characterized by many active channels and
the Battery Point River. rapidly migrating bars.
Unfortunately, there is no existing model for In suggesting the existence of bars within the
CANT AND WALKER: BRAIDEDFLUVIAL FACIES MODEL 117
Battery Point channel, we have relied on the evi- straight crested dunes or sand waves up to
dence of the single, thick, planar-tabular sets of about 30 cm high migrated across the sand flats
cross-bedding, with their anomalous paleocur- or shallow channels now occupying this part of
rent directions. These sets overlie trough cross- the river. As the main flow became diverted
bedding, implying aggradation of the channel elsewhere, the area was in places covered by
floor before bar formation. The planar-tabular rippled silts; occasional overbank flooding con-
sets are themselves overlain by trough cross- tributed vertical accretion muds (facies F). More
bedding that shows the original paleocurrent severe flooding probably gave rise to the low
direction (Fig. 16), implying yet more channel angle, cross stratified sandstones (facies G); this
filling to, and above, the level of the bars. We particular interpretation is suggested by the posi-
therefore suggest that aggradation of the channel tion of facies G in the overall facies sequence,
floors was an important process leading to pres- and by the presence of upper flow regime sedi-
ervation of sand bodies in the geological record. mentary structures. With overall subsidence of
This was earlier suggested by Moody-Stuart the basin, the area just described was later re-
(1966) for low sinuosity (braided) sandy streams, occupied, and another overall fining-upward se-
but was emphatically denied by Allen (1970) in quence was deposited.
his discussion of Moody-Stuart. Allen (1970,
pp. 31 1-314) contends that all preservation of Discharge and Width of the Battery Point River
coarse members in alluvial fining-upward se- The deepest scoured channel that we observed
quences (presumably of any type) is due to was about 5 m; it was filled entirely with trough
lateral accretion, as in the well understood classi- cross bedding. If we make the reasonable as-
cal meandering model. We have shown above sumption that at some time (bankfull discharge)
(and in Fig. 16) that the Battery Point model is dunes were moving on the bed in a flow depth of
quite unlike the classical meandering model, and 5 m, then we can estimate the discharge of the
our present understanding of the evidence seems Battery Point River.
to suggest that vertical aggradation was more The mean flow velocities required for dunes
important than Allen would imply ("vertical (composed of medium sand) to migrate in
deposition of the coarse members as envisaged shallow flows (up to 40 cm) lies in the range
by Moody-Stuart (1966) can therefore be ruled 50-100 cm/s (Harms et al. 1975, fig. 2-3). For
out"; Allen 1970, p. 313). Vertical aggradation flows a few metres deep, the velocities are up-
has been implied for some modern braided ward of 100 cm/s (Harms et al. 1975, figs. 2-10,
rivers, for example, the Lower Yellow River of and 2-11); we shall use a value of 120 cm/s in
China (Chien 1961, p. 749, and is presently being the calculations below.
studied by Cant (1975) in the South Saskatche- Also, for dunes we can estimate an average
wan River. resistance to flow. The dimensionless Chezy dis-
We envisage the initial formation of a new charge coefficient, c / & , ranges from 8 to 12
scour, and the lining of the scour with mud clasts for dunes; we will use a value of 10.
and massive sandstone (facies SS). Rapid migra- The Chezy equation is
tion of sinuous crested dunes, with coarse bed
load, gave rise to the poorly defined trough cross
bedding (A), but as some aggradation took place, where U = mean velocity (cmls), C is the Chezy
the coarser material ceased to move, the dunes discharge coefficient, D is mean flow depth
advanced more slowly, and the well defined (cm), and S is slope.
trough cross bedding (B) formed. At places Substituting in [3], we have
within the stream, large transverse bars with
foreset slopes up to 3 m high were formed; they
migrated both downstream and laterally, to form
planar-tabular sets of cross bedding (C). When
these bars became inactive, further aggradation It has been shown that in a general way, braided
led to their burial by more trough cross bedding and meandering streams can be separated em-
(B). At places within the channels, unusually pirically by their slope (S) and discharge (Q)
deep isolated scours were cut and filled (facies relationships (Leopold and Wolman 1957), the
E), and after extensive aggradation, smaller equation being
Extrapolation from the meandering model to

braided situations is clearly not possible, and
Thus, for a given slope the discharge is generally this emphasizes the importance of establishing
greater than that given by [5] for braiding, and some form of preliminary braided model. The
less for meandering. If the Battery Point River model ideally will be descriptive, as is our local
were braided, on a slope of 2.9 x lop4, the model, but the more examples that can be in-
bankfull discharge may have exceeded 100 000 corporated and distilled into a general model,
ft3/s (2840 m3/s). This is about one sixth of the the better it will serve as a basis for interpreta-
mean annual discharge of the Mississippi, and tion. The braided model is presently not suffi-
implies a fairly large river draining the Acadian ciently well formulated to resolve the interpre-
mountains. tive disagreement between Allen and Moody-
At a mean velocity of about 1 m/s, the cross Stuart. Input into the model must come from
sectional area of the river must have been detailed studies of ancient rocks believed to be
2840 sq. m to accommodate bankfull flow. braided in origin, and from recent sediment
Assuming an average depth at bankfull to be 5 studies of the type done on sandy braided
m, the channel width must have been about 567 streams by Collinson (1970), Smith (1970), and
m. Adding in the various bars and islands within Cant (1975). As more studies are made, the
the river, the original width may have been of 'generalized' braided model will gain in its pre-
the order of 1-2 km. dictive power, and we will have a better geologi-
Status of Sandy Fluvial Facies Models cal understanding of the spectrum of braided
river types.
A well-established facies model should serve
four important functions (Walker, in Harms et al. Acknowledgments
1975): (1) because it is a distillation of many Many of the ideas in this paper have been
examples, the model is a norm with which other stimulated by discussions with G . V. Middleton,
examples can be compared and contrasted; (2) J. C. Harms, J. B. Southard, and D. B. Spearing;
because the model embodies the descriptive some of the ideas were presented at an SEPM
essence of many examples, it acts as a framework Short Course in Dallas in April, 1975. We are
and guide for future observations; (3) because the indebted to the Society of Economic Paleontol-
model emphasizes the important genetic features ogists and Mineralogists for permission to re-
of many examples, it can act as a generalized produce Figs. 13 and 14. The text has been im-
basis for hydrodynamic interpretation; and (4) proved by the comments of P. McCabe. The
because the model emphasizes the genetic re- work was supported by the National Research
lationships between its components, it can act as Council of Canada, and the field expenses of the
a predictor in new situations where only parts of first author were covered by a grant from the
the system may be exposed. Geological Society of America. We thank these
The generalized meandering model of Allen colleagues and agencies for their interest in, and
(1970; our Fig. 16) fulfills all of the above func- support of, our work; we are solely responsible
tions about as well as any existing facies model. for the facts and interpretations presented herein.
A possible objection to the model is that in its
statistical formulation, Allen (1970) may have
ALLEN,J. R. L. 1965. A review of the origin and charac-
blurred together the deposits of (1) active lateral teristics of recent alluvial sediments. Sedimentology, 5,
accretion of point bars, (2) chute cutoffs, and pp. 89-191.
(3) neck cutoffs. These three types of alluvial 1970. Studies in fluviatile sedimentation: a com-
deposition give rise to three recognizably distinct parison of fining-upwards cyclothems, with special re-
ference to coarse-member composition and interpreta-
fining-upward sequences. Nevertheless, with the tion. J. Sediment. Petrol. 40, pp. 298-323.
existence of the norm, departures from it, in the ALLEN,J. R. L. and FRIEND,P. F. 1968. Deposition of the
form of neck or chute cutoff deposits, are more Catskill facies, Appalachian region: with notes on some
easily identified. Thus Allen's model is a norm; other Old Red Sandstone basins. In: Late Paleozoic
it has also guided other observations, and has and Mesozoic continental sedimentation, northeastern
North America. (G. deV. Klein, Ed.) Geol. Soc. Am.
acted as a basis for hydrodynamic interpretation Spec. Pap. 106, pp. 21-74.
(Allen 1970, pp. 314-317). It can also be used BANKS,N. L. and COLLINSON, J. D. 1974. Discussion of
successfully as a predictor in new situations. "some sedimentological aspects of planar cross-strati-
fication in a sandy braided river". J. Sediment. Petrol. KELLING,G. 1968. Patterns of sedimentation in Rhondda
44, pp. 265-267. Beds of South Wales. Am. Assoc. Pet. Geol., Bull. 52,
B o u c o ~ A.
, J., CUMMING, L. M. and JAEGER,H. 1967. pp. 2369-2386.
Contributions to the age of the Gaspt Sandstone and LEOPOLD,L. B. and WOLMAN, M. G. 1957. River channel
Gaspt Limestone. Geol. Surv. Can. Pap. 67-25,27 p. patterns: braided meandering and straight. U.S. Geol.
BULLARD, E. C. EVERETT,J. E. and SMITH,A. G. 1965. Surv. Prof. Pap. 282-B.
The fit of the continents around the Atlantic. In: A MCGERRIGLE, H. W. 1950. The geology of eastern GaspC.
symposiumoncontinentaldrift. (P. M. S. Blackett, E. C. Quebec Dep. Mines, Geol. Rep. 3 5 , 1 6 8 ~ .
Bullard, and S. K. Runcorn, Eds.) R. Soc. London Phil. MIALL,A. D. 1973. Markov chain analysis applied to an
Trans. 258A, pp. 41-51. ancient alluvial plain succession. Sedimentology, 20, pp.
CANT,D. J. 1975. Sandy braided stream deposits in the 347-364.
South Saskatchewan River. Geol. Soc. Am. Abst. with MOODY-STUART, M. 1966. High- and low-sinuosity stream
Prog. 7 (no. 6), p. 731. deposits with examples from the Devonian of Spits-
CHIEN,NING. 1961. The braided stream of the lower Yel- bergen. J. Sediment. Petrol. 36, pp. 1102-1 117.
low River. Sci. Sin. 10, pp. 734-754. SELLEY,R. C. 1970. Studies of sequence in sediments
COLLINSON, J. D. 1970. Bed forms of the Tana River, using a simple mathematical device. Geol. Soc. London
Norway. Geog. Ann. 52A, pp. 31-56. Quart. J. 125, pp. 557-581.
DE RAAF,J. F. M., READING,H. G., and WALKER, R. G. SMITH,N. D. 1970. The braided stream depositional envi-
1965. Cyclic sedimentation in the lower Westphalian of ronment: comparison of the Platte River with some
North Devon, England. Sedimentology, 4, pp. 1-52. Silurian clastic rocks, North-Central Appalachians.
DOEGLAS,D. J. 1962. The structure of sedimentary de- Geol. Soc. Am. Bull. 81, pp. 2993-3014.
posits of braided rivers. Sedimentology 1, pp. 167-190. 1971. Transverse bars and braiding in the lower
EYNON,G. and WALKER,R. G. 1974. Facies relationships Platte River, Nebraska. Geol. Soc. Am. Bull. 82, pp.
in Pleistocene outwash gravels, southern Ontario: a 3407-3420.
model for bar growth in braided rivers. Sedimentology, -1972. Some sedimentological aspects of planar
21, pp. 43-70. cross-stratification in a sandy braided river. J. Sediment.
FOLK,R. L. 1968. Petrology of sedimentary rocks. Hem- Petrol. 42, pp. 624-634.
phills, Austin, Texas. 170 p. 1974. Some sedimentological aspects of planar

HARMS,J. C., MACKENZIE, D. B., and MCCUBBIN, D. G. cross-stratification in a sandy braided river: areply to N.
1963. Stratification in modern sands of the Red River, L. Banks and J. D. Collinson. J. Sediment. Petrol. 44,
Louisiana. J. Geol. 71, pp. 566-580. pp. 267-269.
HARMS,J. C., SOUTHARD, J. B. SPEARING, D. R., and ST. JULIEN,P., HUBERT,C., SKIDMORE, W. B., and
WALKER,R. G. 1975. Depositional environments as in- BELAND,J. 1972. Appalachian structure and stratig-
terpreted from primary sedimentary structures and raphy, Quebec. Int. Geol. Cong., Montreal, Guidebook
stratification sequences. Soc. Econ. Paleontol. Mineral, for field excursion A56-C56,99 p.
Short Course 2, 161 p.
This article has been cited by:
1. Luca Colombera, Nigel P. Mountney, William D. McCaffrey. 2013. A quantitative approach to fluvial
facies models: Methods and example results. Sedimentology n/a-n/a. [CrossRef]
2. P.L. Corcoran, A.J. Bumby, D.W. Davis. 2013. The Paleoproterozoic Waterberg Group, South Africa:
Provenance and its relation to the timing of the Limpopo orogeny. Precambrian Research 230, 45-60.
[CrossRef]
3. References 335-363. [CrossRef]

4. Mrinal Kanti Roy, Syed Samsuddin Ahmed, Tapas Kumar Bhattacharjee, Sultan Mahmud, Md.
Moniruzzaman, Md. Masidul Haque, Sudip Saha, Md. Ismail Molla, Pulin Chandra Roy. 2012.
Paleoenvironment of deposition of the Dupi Tila Formation, Lalmai Hills, Comilla, Bangladesh.
Journal of the Geological Society of India 80:3, 409-419. [CrossRef]
5. PHILIP J. ASHWORTH, GREGORY H. SAMBROOK SMITH, JAMES L. BEST, JOHN S.
BRIDGE, STUART N. LANE, IAN. A. LUNT, ARNOLD J. H. REESINK, CHRISTOPHER
J. SIMPSON, ROBERT E. THOMAS. 2011. Evolution and sedimentology of a channel fill in the
sandy braided South Saskatchewan River and its comparison to the deposits of an adjacent compound
bar. Sedimentology 58:7, 1860-1883. [CrossRef]
6. Piotr Weckwerth, Krzysztof Przegitka, Alicja Chruciska, Barbara Woronko, Hubert L. Oczkowski.
2011. Age and sedimentological features of fluvial series in the Toru Basin and the Drwca Valley
(Poland). Geochronometria 38:4, 397-412. [CrossRef]
7. Neil S. Davies, Martin R. Gibling. 2010. Cambrian to Devonian evolution of alluvial systems: The
sedimentological impact of the earliest land plants. Earth-Science Reviews 98:3-4, 171-200. [CrossRef]
8. M. Kh. Khalifa, C. R. Ward. 2010. Sedimentological analysis of the subsurface Mulga Downs Group
in the central part of the Darling Basin, western New South Wales. Australian Journal of Earth Sciences
57:1, 111-139. [CrossRef]
9. Shahid Ghazi, Nigel P. Mountney. 2009. Facies and architectural element analysis of a meandering
fluvial succession: The Permian Warchha Sandstone, Salt Range, Pakistan. Sedimentary Geology
221:1-4, 99-126. [CrossRef]
10. A. Wolela. 2009. Sedimentation and depositional environments of the Barremian-Cenomanian
Debre Libanose Sandstone, Blue Nile (Abay) Basin, Ethiopia. Cretaceous Research 30:5, 1133-1145.
[CrossRef]
11. Bryan Krape, Mark E. Barley, Stuart J.A. Brown. 2008. Late Archaean synorogenic basins of the
Eastern Goldfields Superterrane, Yilgarn Craton, Western Australia. Precambrian Research 161:1-2,
135-153. [CrossRef]
12. Denis LavoieChapter 3 Appalachian Foreland Basin of Canada 5, 65-103. [CrossRef]
13. Franois Therrien. 2006. Depositional environments and fluvial system changes in the dinosaur-
bearing Snpetru Formation (Late Cretaceous, Romania): Post-orogenic sedimentation in an active
extensional basin. Sedimentary Geology 192:3-4, 183-205. [CrossRef]
14. G. H. SAMBROOK SMITH, P. J. ASHWORTH, J. L. BEST, J. WOODWARD, C. J. SIMPSON.
2006. The sedimentology and alluvial architecture of the sandy braided South Saskatchewan River,
Canada. Sedimentology 53:2, 413-434. [CrossRef]
15. Tsutomu Nakazawa, Rei Nakashima, Takeyuki Ueki, Susumu Tanabe, Hideaki Oshima, Seiji
Horiuchi. 2006. Sequence stratigraphy of the Pleistocene Kioroshi Formation, Shimosa Group beneath
the Omiya Upland, central Kanto Plain, central Japan. The Journal of the Geological Society of Japan
112:5, 349-368. [CrossRef]
16. Franois Therrien. 2005. Palaeoenvironments of the latest Cretaceous (Maastrichtian) dinosaurs
of Romania: insights from fluvial deposits and paleosols of the Transylvanian and Haeg basins.
Palaeogeography, Palaeoclimatology, Palaeoecology 218:1-2, 15-56. [CrossRef]
17. Rajat Mazumder, Subir Sarkar. 2004. Sedimentation history of the Palaeoproterozoic Dhanjori
Formation, Singhbhum, eastern India. Precambrian Research 130:1-4, 267-287. [CrossRef]
18. C. Gouramanis, J. A. Webb, A. A. Warren. 2003. Fluviodeltaic sedimentology and ichnology of part
of the Silurian Grampians Group, western Victoria. Australian Journal of Earth Sciences 50:5, 811-825.
[CrossRef]
19. Catherine A Rigsby, Paul A Baker, Mark S Aldenderfer. 2003. Fluvial history of the Rio Ilave
valley, Peru, and its relationship to climate and human history. Palaeogeography, Palaeoclimatology,
Palaeoecology 194:1-3, 165-185. [CrossRef]
20. Maneesh Sharma, Shikha Sharma, Uma Kant Shukla, Indra Bir Singh. 2002. Sandstone body
architecture and stratigraphic trends in the Middle Siwalik Succession of the Jammu area, India.
Journal of Asian Earth Sciences 20:7, 817-828. [CrossRef]
21. Mike Pole. 2001. Repeated flood events and fossil forests at Curio Bay (Middle Jurassic), New Zealand.
Sedimentary Geology 144:3-4, 223-242. [CrossRef]
22. Finlay M. Stuart, Brian J. Bluck, Malcolm S. Pringle. 2001. Detrital muscovite 40 Ar/ 39 Ar ages
from Carboniferous sandstones of the British Isles: Provenance and implications for the uplift history
of orogenic belts. Tectonics 20:2, 255-267. [CrossRef]
23. M Huisink. 2000. Changing river styles in response to Weichselian climate changes in the Vecht
valley, eastern Netherlands. Sedimentary Geology 133:1-2, 115-134. [CrossRef]

24. J Vandenberghe, R.A van Overmeeren. 1999. Ground penetrating radar images of selected fluvial
deposits in the Netherlands. Sedimentary Geology 128:3-4, 245-270. [CrossRef]
25. OLE J. MARTINSEN, ALF RYSETH, WILLIAM HELLAND-HANSEN, HARALD
FLESCHE, GEIR TORKILDSEN, SAHIRE IDIL. 1999. Stratigraphic base level and fluvial
architecture: Ericson Sandstone (Campanian), Rock Springs Uplift, SW Wyoming, USA.
Sedimentology 46:2, 235-263. [CrossRef]
26. David J. Cantrill, G.J. Nichols. 1996. Taxonomy and palaeoecology of Early Cretaceous (Late Albian)
angiosperm leaves from Alexander Island, Antarctica. Review of Palaeobotany and Palynology 92:1-2,
1-28. [CrossRef]
27. Christine E. Koltermann, Steven M. Gorelick. 1996. Heterogeneity in Sedimentary Deposits: A
Review of Structure-Imitating, Process-Imitating, and Descriptive Approaches. Water Resources
Research 32:9, 2617-2658. [CrossRef]
28. Erik K. Webb, Mary P. Anderson. 1996. Simulation of Preferential Flow in Three-Dimensional,
Heterogeneous Conductivity Fields with Realistic Internal Architecture. Water Resources Research
32:3, 533-545. [CrossRef]
29. Patrick G. Eriksson, Boris F.F. Reczko, A. Jaco Boshoff, Ute M. Schreiber, Markus Van der Neut,
Carel P. Snyman. 1995. Architectural elements from Lower Proterozoic braid-delta and high-energy
tidal flat deposits in the Magaliesberg Formation, Transvaal Supergroup, South Africa. Sedimentary
Geology 97:1-2, 99-117. [CrossRef]
30. Pradip K. Bose, Partha Pratim Chakraborty. 1994. Marine to fluvial transition: Proterozoic Upper
Rewa Sandstone, Maihar, India. Sedimentary Geology 89:3-4, 285-302. [CrossRef]
31. Sylvain Desbiens. 1994. Les Bivalves Cyrtodontacea Ulrich 1894 de la Formation de York River,
Emsien de Gaspsie, Qubec. Geobios 27:4, 449-458. [CrossRef]
32. Brian Willis. 1993. Ancient river systems in the Himalayan foredeep, Chinji Village area, northern
Pakistan. Sedimentary Geology 88:1-2, 1-76. [CrossRef]
33. Ajit Bhattacharyya, Sadoon Morad. 1993. Proterozoic braided ephemeral fluvial deposits: an example
from the Dhandraul Sandstone Formation of the Kaimur Group, Son Valley, central India. Sedimentary
34. G. H. Browne, M. B. Reay. 1993. The Warder Formation: Cyclic fluvial sedimentation during the
Ngaterian (late AlbianCenomanian) of Marlborough, New Zealand. New Zealand Journal of Geology
and Geophysics 36:1, 27-35. [CrossRef]
35. Anthony J. Martin. 1993. Semiquantitative and statistical analysis of bioturbate textures, sequatchie
formation (upper ordovician), Georgia and Tennessee, USA. Ichnos 2:2, 117-136. [CrossRef]
36. Michael C. Wizevich. 1992. Sedimentology of pennsylvanian quartzose sandstones of the lee
formation, central appalachian basin: fluvial interpretation based on lateral profile analysis.
37. Tamrat Worku, Timothy R. Astin. 1992. The Karoo sediments (Late Palaeozoic to Early Jurassic) of
the Ogaden Basin, Ethiopia. Sedimentary Geology 76:1-2, 7-21. [CrossRef]
38. G.J. Orton. 1991. Emergence of subaqueous depositional environments in advance of a major
ignimbrite eruption, Capel Curig Volcanic Formation, Ordovician, North Wales an example of
regional volcanotectonic uplift?. Sedimentary Geology 74:1-4, 251-286. [CrossRef]
39. Pradip K. Bose, Subir Sarkar. 1991. Basinal autoclastic mass flow regime in the Precambrian Chanda
Limestone Formation, Adilabad, India. Sedimentary Geology 73:3-4, 299-315. [CrossRef]
40. I.M. Makhlouf, B.R. Turner, A.M. Abed. 1991. Depositional facies and environments in the Permian
Umm Irna formation, Dead Sea area, Jordan. Sedimentary Geology 73:1-2, 117-139. [CrossRef]
41. Michael J. Ranger, S.George Pemberton. 1991. Multivariate analysis of ichnofossil associations in
the subsurface Bluesky Formation (Albian, Alberta, Canada). Palaeogeography, Palaeoclimatology,
Palaeoecology 85:3-4, 169-187. [CrossRef]
42. FRANCES J. HEIN, GREGORY A. ROBB, ANDREW C. WOLBERG, FREDERICK J.
LONGSTAFFE. 1991. Facies descriptions and associations in ancient reworked (?transgressive) shelf
sandstones: Cambrian and Cretaceous examples. Sedimentology 38:3, 405-431. [CrossRef]
43. Adal Singh, B.D. Bhardwaj. 1991. Fluvial facies model of the Ganga River sediments, India.
44. I.M. Makhlouf, A.M. Abed. 1991. Depositional facies and environments in the Umm Ishrin Sandstone
Formation, Dead Sea area, Jordan. Sedimentary Geology 71:3-4, 177-187. [CrossRef]
45. P.D. Godin. 1991. Fining-upward cycles in the sandy braided-river deposits of the Westwater Canyon
Member (Upper Jurassic), Morrison Formation, New Mexico. Sedimentary Geology 70:1, 61-82.
[CrossRef]
46. Dilip Saha, Tapan Chakraborty. 1990. TILTVEC: A fortran-77 program for the tilt correction
of paleocurrent data with resolution of incongruities. Computers & Geosciences 16:8, 1193-1207.
[CrossRef]
47. References 49, 561-667. [CrossRef]
48. Romeo M. Flores. 1989. Rocky Mountain Tertiary coal-basin models and their applicability to some
world basins. International Journal of Coal Geology 12:1-4, 767-798. [CrossRef]
49. J.L. Deluca, K.A. Eriksson. 1989. Controls on synchronous ephemeral- and perennial-river
sedimentation in the middle sandstone member of the Triassic Chinle Formation, northeastern New
Mexico, U.S.A. Sedimentary Geology 61:3-4, 155-175. [CrossRef]
50. Neil A. Wells. 1989. A program in BASIC for facies-by-facies Markov chain analysis. Computers &
Geosciences 15:1, 143-155. [CrossRef]
51. GARY G. LASH. 1988. Sedimentology and evolution of the Martinsburg Formation (Upper
Ordovician) fine-grained turbidite depositional system, central Appalachians. Sedimentology 35:3,
429-447. [CrossRef]
52. Volkmar Stingl. 1987. Die fazielle Entwicklung des Alpinen Buntsandsteins (Skyth) im Westabschnitt
der Nrdlichen Kalkalpen (Tirol/Salzburg, sterreich). Geologische Rundschau 76:2, 647-664.
[CrossRef]
53. Douglas I. Benn, Martin Dawson. 1987. Something for all spectators? Postgraduate training, self
paced undergraduate fieldwork, reducing staff boredom. Journal of Geography in Higher Education
11:1, 72-78. [CrossRef]
54. H. G. READING. 1987. Fashions and models in sedimentology: a personal perspective. Sedimentology
34:1, 3-9. [CrossRef]
55. S.M. Casshyap, Ajay Kumar. 1987. Fluvial architecture of the Upper Permian Raniganj coal measure
in the Damodar basin, Eastern India. Sedimentary Geology 51:3-4, 181-213. [CrossRef]
56. A. Tevfik Arguden, Kelvin S. Rodolfo. 1986. Sedimentary facies and tectonic implications of Lower
Mesozoic alluvial-fan conglomerates of the Newark Basin, northeastern United States. Sedimentary
57. Pradip K. Bose, Sabyasachi Shome, Shubhendu Bardhan, Gautam Ghosh. 1986. Facies mosaic in the
Ghuneri Member (Jurassic) of the Bhuj Formation, western Kutch, India. Sedimentary Geology 46:3-4,
293-309. [CrossRef]
58. James W. Collinson, Deana Chapman Pennington, Noel R. KempStratigraphy and petrology of
Permian and Triassic fluvial deposits in northern Victoria Land, Antarctica 46, 211-242. [CrossRef]
59. Andrew D. Miall. 1985. Architectural-element analysis: A new method of facies analysis applied to
fluvial deposits. Earth-Science Reviews 22:4, 261-308. [CrossRef]
60. Willo M. Stear. 1985. Comparison of the bedform distribution and dynamics of modern and ancient
sandy ephemeral flood deposits in the southwestern Karoo region, South Africa. Sedimentary Geology
45:3-4, 209-230. [CrossRef]
61. ROBERT V. DEMICCO. 1985. Platform and off-platform carbonates of the Upper Cambrian of
western Maryland, U.S.A. Sedimentology 32:1, 1-22. [CrossRef]
62. M. Hoque, C. S. Nwajide. 1985. Application of Markov chain and entropy analysis to lithologic
successions: An example from the Cretaceous of the Benue trough (Nigeria). Geologische Rundschau
74:1, 165-177. [CrossRef]
63. BRIAN R. RUST. 1984. Proximal braidplain deposits in the Middle Devonian Malbaie Formation of
Eastern Gaspe, Quebec, Canada. Sedimentology 31:5, 675-695. [CrossRef]
64. Detlef Mader, Thomas Teyssen. 1984. Palaeoenvironmental interpretation of fluvial red beds by
statistical analysis of palaeocurrent data: Examples from the buntsandstein (lower triassic) of the eifel
and bavaria in the German basin (middle Europe). Sedimentary Geology 41:1, 1-74. [CrossRef]
65. PHILIP R. HILL. 1984. Sedimentary facies of the Nova Scotian upper and middle continental slope,
offshore eastern Canada. Sedimentology 31:3, 293-309. [CrossRef]
66. Barry R Gager. 1984. The tiger formation: Paleogene fluvial sediments deposited adjacent to a
deforming Cordilleran metamorphic core complex, northeastern Washington. Sedimentary Geology
38:1-4, 393-420. [CrossRef]
67. Samuel Y Johnson. 1984. Cyclic fluvial sedimentation in a rapidly subsiding basin, northwest
Washington. Sedimentary Geology 38:1-4, 361-391. [CrossRef]
68. MARTIN KIRK. 1983. Bar development in a fluvial sandstone (Westphalian 'A'), Scotland.
69. Ram C. Tewari, Satyendra M. Casshyap. 1983. Cyclicity in early Permian fluviatile Gondwana coal
measures: An example from Giridih and Saharjuri basins, Bihar, India. Sedimentary Geology 35:4,
297-312. [CrossRef]
70. BRIAN R. TURNER. 1983. Braidplain deposition of the Upper Triassic Molteno Formation in the
main Karoo (Gondwana) Basin, South Africa. Sedimentology 30:1, 77-89. [CrossRef]
71. J.R.L. Allen. 1983. Studies in fluviatile sedimentation: Bars, bar-complexes and sandstone sheets
(low-sinuosity braided streams) in the brownstones (L. devonian), welsh borders. Sedimentary Geology
33:4, 237-293. [CrossRef]
72. Zahid A. Khan, Satyendra M. Casshyap. 1982. Sedimentological synthesis of permian fluviatile
sediments of East Bokaro basin, Bihar, India. Sedimentary Geology 33:2, 111-128. [CrossRef]
73. R. A. Glen. 1982. Nature of lateEarly to Middle Devonian tectonism in the Buckambool area, Cobar,
New South Wales. Journal of the Geological Society of Australia 29:1-2, 127-138. [CrossRef]
74. John R. Graham. 1981. Fluvial sedimentation in the lower carboniferous of Clew Bay, County Mayo,
Ireland. Sedimentary Geology 30:3, 195-211. [CrossRef]
75. Richard G. Vos, Anthony J. Tankard. 1981. Braided fluvial sedimentation in the lower paleozoic cape
basin, South Africa. Sedimentary Geology 29:2-3, 171-193. [CrossRef]

76. D. J. STEWART. 1981. A meander-belt sandstone of the Lower Cretaceous of Southern England.
77. Ian P. Tunbridge. 1981. Sandy high-energy flood sedimentation Some criteria for recognition,
with an example from the devonian of S.W. England. Sedimentary Geology 28:2, 79-95. [CrossRef]
78. J.R.L. Allen, R.G. Thomas, B.P.J. Williams. 1981. Field meeting: The facies of the Lower Old Red
Sandstone, north of Milford Haven, southwest Dyfed, Wales. Proceedings of the Geologists' Association
92:4, 251-267. [CrossRef]
79. References 29, 493-536. [CrossRef]
80. B. J. Bluck. 1980. Structure, generation and preservation of upward fining, braided stream cycles
in the Old Red Sandstone of Scotland. Transactions of the Royal Society of Edinburgh: Earth Sciences
71:01, 29-46. [CrossRef]
81. Colin M. Jones. 1979. Tabular cross-bedding in Upper Carboniferous fluvial channel sediments in
the southern Pennines, England. Sedimentary Geology 24:1-2, 85-104. [CrossRef]
82. David K. Hobday, Victor von Brunn. 1979. Fluvial sedimentation and paleogeography of an early
paleozoic failed rift, southeastern margin of Africa. Palaeogeography, Palaeoclimatology, Palaeoecology
28, 169-184. [CrossRef]
83. DOUGLAS J. CANT, ROGER G. WALKER. 1978. Fluvial processes and facies sequences in the
sandy braided South Saskatchewan River, Canada. Sedimentology 25:5, 625-648. [CrossRef]
84. Andrew D. Miall, Martin R. Gibling. 1978. The Siluro-Devonian clastic wedge of Somerset Island,
Arctic Canada, and some regional paleogeographic implications. Sedimentary Geology 21:2, 85-127.
[CrossRef]
85. Andrew D. Miall. 1977. A review of the braided-river depositional environment. Earth-Science Reviews
13:1, 1-62. [CrossRef]

Cant 1976

Uploaded by

Copyright:

Available Formats

Cant 1976

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Cant 1976

Uploaded by

Copyright:

Available Formats

Development of a braided-fluvial facies model for the Devonian

Battery Point Sandstone, Qubbec

Revision accepted for publication 15 September 1975

three formations collectively known as the Gasp6

since the study of McGerrigle (1950). He set up

This study is restricted to the 110 m section

narrow outcrop at the foot of the cliffs gives a

110 m sequence. Subdivision into 'pieces', or

ties, shown in Table 2. This transition probabil- D.A,B,E.F.B.A,G.B.C,A,C,B,C,B.D,F/

where sharp and channelled contacts predomin- / FAULT OR COVERED INTERVAL

ate, this is not a serious objection (see de Raaf

TABLE2. Observed transition probabilities

TABLE3. Transition probabilities for random sequence

OBSERVED FACIES RELATIONSHIPS -- monly certain transitions occur (Fig. 15). By

relative abundances of the other facies (Table 3). -

A chi-squared test showed that the observed

lower probability of occurring than if the se-

TABLE4. Observed minus random transition probabilities

D +0.111 -0.270 -0.206 -0.127 -0.032 +0.604 -0.079

aware (and will continue to be aware in our in- BATTERY POINT

some of the data.

First, it is clear in Fig. 15 that, above the

in disused parts of the floodplain, with subse-

Extrapolation from the meandering model to

phills, Austin, Texas. 170 p. 1974. Some sedimentological aspects of planar

3. References 335-363. [CrossRef]

valley, eastern Netherlands. Sedimentary Geology 133:1-2, 115-134. [CrossRef]

basin, South Africa. Sedimentary Geology 29:2-3, 171-193. [CrossRef]

You might also like