Marine to fluvial transition: Proterozoic Upper Rewa Sandstone, Maihar, India

SEDIMENTARY GEOLOGY ELSEVIER

Sedimentary Geology 89 (1994) 285-302

Marine to fluvial transition: Proterozoic Upper Rewa Sandstone, Maihar, India Pradip K. Bose *, Partha Pratim Chakraborty Department of Geological Sciences, Jadavpur University, Calcutta-700032, India (Received September 16, 1992; revised version accepted June 15, 1993)

Abstract The Proterozoic Upper Rewa Sandstone in central India, is generally interpreted as entirely marine. A detailed study in Maihar, Madhya Pradesh, however, reveals upward transition from marine to fluvial through a mixed facies association also bearing eolian imprints. Facies associations differ in stratal geometry and arrangement, palaeocurrent direction and pattern as well as in grain size and sorting. The fine-grained marine association is dominated by tidal sheet deposits with characteristic rhythmic changes in thickness and style of cross-stratification and their packages, as well as local evidence of current reversals. Subordinate beach deposits are also present. The major palaeocurrent direction is highly consistent and westward. The coarse-grained, often granule-rich, sandstones of the fluvial association are embodied by four facies correlatable with flow stages and relative bed shear. These braided stream deposits are preserved in vertical stacks of tabular sandbodies in response mainly to basin subsidence. The palaeocurrent pattern is unimodal and northwestward, but its dispersion covers 180°. The rocks are poorly sorted and at places incorporate sand clasts, although they are virtually mud-free and mineralogically mature as in two other associations. The medium- to fine-grained sandstones at the intermediate interval between the marine and fluvial associations are of distinct lensoid geometry. They incorporate representatives from the previous two associations as well as translatent strata, interdune erg deposits and possibly aeolian cross-strata. Consequently the palaeocurrent pattern is polymodal with a spread over 270 °. An overall coarsening-up trend in the fluvial part and thinning-up trend in the tidal part imply progradation in spite of evidence of basin subsidence. A high rate of sediment discharge from a perennial river system in a humid climate is suggested.

I. Introduction T h e difficulty in i n t e r p r e t i n g p r e - S i l u r i a n unfossiliferous, texturally m a t u r e , t a b u l a r q u a r t z a r e -

* Corresponding author.

nite b o d i e s has b e e n succinctly discussed by m a n y (e.g., Long, 1978; D o t t et al., 1986; F e d o a n d C o o p e r , 1990). T h e resulting c o n f u s i o n is a p p a r e n t in p a l a e o g e o g r a p h i c r e c o n s t r u c t i o n o f t h e U p p e r R e w a S a n d s t o n e of the P r o t e r o z o i c Vind h y a n S u p e r g r o u p d e p o s i t e d in an i n t r a c r a t o n i c setting in c e n t r a l I n d i a ( P r a s a d a n d V e r m a , 1991).

0037-0738/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0037-0738(93)E0092-T

286

P.K. Bose. P.P. Chakraborty / Sedimentao' Geology 89 (1994) 28.5 302

81' f

82" I

.Manikpur

-25"

Fig. t. Outcrop area of Rewa Formation in the Son Valley (outlined) indicated on map of India (inset).

in the form of pseudomatrix at places. The sorting is commonly good to very good except for a strikingly poor sorting in the upper part of the formation. Petrographically all the sandstones are quartz arenites, although certain facies contain a considerable amount of intraformational mud and sandclasts. Quartz cement is present abundantly in the form of overgrowths; ferruginous cement is also present in significant amounts in certain parts and calcite cement is present, in a restricted way, in the lowermost part.

3. Facies analysis In spite of broad lithologic uniformity, a number of distinct facies can be recognized in these

A general consensus has long prevailed that it is entirely of marine origin (e.g., Banerjee, 1974; Singh, 1976, 1980; Chanda and Bhattacharyya, 1982; Prasad and Verma, 1991). Recently, Chakraborty and Choudhuri (1990) while dealing with the stratigraphy of the Rewa Formation, intuitively suggested an upward transition to fluvial deposits. Except for an overall coarsening-up trend, this sandstone seems monotonous over hundreds of square kilometres in the Son valley, India (Fig. 1). The present detailed study around Maihar, Madhya Pradesh, however, reveals an interplay of multiple depositional agents and a wide spectrum of flow conditions (Fig. 2).

~fi._-S_=

2

~

5

=I07

50

2. Geologic setting In Maihar, Madhya Pradesh, the Upper Rewa Sandstone (about 400 m thick) is complete with bounding shales and as said before, shows an overall coarsening-up trend turning rich in granules. In the topmost 20 m it rapidly fines and becomes granule-free medium grained eventually giving way upward to the Ganurgarh Shale (Fig. 2). Cross-stratifications are ubiquitous, planar laminations are subordinate while ripples are strikingly rare in these sandstones. Sand grain populations vary from fine grained to granular and individual grains are rounded and spherical. The rocks are, in general, devoid of mud, except

A n=2 ~. n=82

1" - ~

0 ....

-J'.S.=

*- G r a i n s i z e increases

Fig. 2. Vertical organization in facies associations (indicated by numbers) and their constituent facies indicated by capital letters in representative sections in the Upper Rewa Sandstone bounded by the Jhiri Shale (J.S.) and Ganurgarh Shale (G.S.). Grain-size variation trend indicated. Palaeocurrent patterns in individual facies associations (right) and in individual facies within the mixed association (3) (left) are also given.

P.K. Bose, P.P. Chakraborty / Sedimentary Geology 89 (1994) 285-302

287

Fig. 3. Downcurrent transition from convex-up to concave-up cross-stratifications within the marginal marine facies associations (1). Note successive packages bounded by ~trong erosion surfaces (thicknesses marked).

mineralogically mature, mud-free sandstones. They differ in structural organization, palaeocurrent directions and patterns, and texture. These facies have been grouped into three different facies associations, viz., (1) marginal marine, (2) fluvial, and (3) mixed, forming three vertical zones in succession.

3.1. Marginal marine association This association is represented by fine, very well sorted sheet sandstones with an abundance of mudclasts. The dominant palaeocurrent is highly consistent and directed westward (Fig. 2).

Facies .4, rhythmic cross-stratified facies This facies is characterized by unidirectional, large-scale (80 cm-2 m set thickness) planar cross-stratifications, their pattern changing laterally in a rhythmic fashion from (sigmoidal) convex- to concave-up (Fig. 3). The cross-laminae are arranged in discrete packages demarcated by strong erosion surfaces. Within the packages, lamina thickness shows overall decreasing and then increasing trends and the number of laminae constituting the packages varies from 12 to 14 or 22 to 28 (Fig. 4). In many sets thick and thin laminae alternate (Fig. 4). Though rare, some foresets are mantled by mud-draped small (am-

09 U~I, t,U Z

10

~.0

30

40 LAMINA NUMBER

50

60

70

Fig. 4. Lamina-thickness variations within successive packages of cross-stratifications. Numbers of laminae constituting individual packages given above.

288

P.K. Bose, P.P. Chakraborty / Sedimentao, Geology 89 (1994) 285--302

N=28

N=16

n=2o

rl 18

Fig. 5. Paleocurrent roses in the marginal marine facies association (1): large scale cross-stratifications (a), current crescents (b) and parting lineations (c) in facies A and, current roses as well as parting lineation in facies B (d).

plitude 5 - 8 cm) ripples moving upslope (exact orientation indeterminable). The sigmoidal foresets bear parting lineations above the brink points and these extend onto the topsets where they are associated with current crescents. The crossstrata, parting lineations and current crescents are orientated more or less in the same direction (Figs. 5a-5c). At places there are water recession marks and exhumed small pools with ripples confined within them (Fig. 6). The thickness of the facies varies from 1.5 to 3 m.

Facies B, plane-laminated facies The plane-laminated units are subordinate in occurrence in close alternations with facies A and

their thicknesses vary from 60 to 90 cm. The bedding plane surface is carved with parting lineations and current crescents (Fig. 7). Orientations of these bed-surface structures are the same as those present in facies A (Fig. 5d).

In terpreta tion The unidirectional large-scale cross-stratified facies A formed presumably by migration of sandwaves. The reverse orientations between the large-scale cross-strata and the rare mud-draped ripples on them record intermittent flow reversals suggesting strong tidal asymmetry. Similarly, the rhythmic changes in the cross-stratification pattern closely resemble those reported by Kreisa

Fig. 6. Shallow pool exhumed on bedding plane surface in the marginal marine facies association. Note termination and slight downcurrent curving of the crests of the ripples along the margin of the pool (left). Pen is 14 cm.

P.K. Bose, P.P. Chakraborty/ Sedimentary Geology 89 (1994) 285-302

and Moila (1986) from a tidal sequence. The alternating thick-thin laminae and laminae packages defined by most conspicuous erosional surfaces clearly resemble diurnal/bidiurnal (Visser, 1980; De Boer et al., 1989; Williams, 1991) and spring-neap-spring tidal cycles (Kreisa and Moiola, 1986; see also Terwindt, 1981; Kohshiek and Terwindt, 1981), respectively. The associated plane-laminated facies with parting lineations and current crescents closely resemble a high-energy beach. That some parts of facies A also were in the intertidal zone is evident from the presence of water recession marks and the shallow pools. The cross-sets in which the foresets are often parting-lineated, very probably formed in the same intertidal zone. The unidirectional foresets devoid of mud drape (in contrast to mud drape on reversely oriented ripples) were in that case generated during dominant ebb tides (see Kohshiek and Terwindt, 1981). The dominant current being that of ebb in this case, the spring-neap-spring packages are defined by strong erosion surfaces, instead of mud drapes, and the sea opened westward (Chanda and Bhattacharya, 1982; Sarkar et al., 1991). Tidal shear

289

abruptly increases during the final recession of water from the intertidal zone. Swash and backwash during this end phase of ebb is likely to generate parting lineations on the lee face of ebb-oriented intertidal sandwaves. 3.2. Flucial association

The sandstones comprising this association are fine to coarse grained, even granule-rich. Some contain rip-up sandclasts. Petrographically the rocks of this association differ from those of the other two associations by commonly being strikingly less sorted, although similarly devoid of mud and mineralogically mature. The association is characterized thoroughly by cross-stratifications of varied geometries to be described later. The palaeocurrent direction measured from these cross-stratifications spans over 180° and its mean is towards the northwest (Fig. 2); the pattern as well as the direction is strikingly different from those obtained from the marginal marine facies association. The facies constituting this fluvial association differ from each other in type, scale and arrangement of cross-stratifications as well as in grain size.

Fig. 7. Parting-lineatedsurface superimposedwith current crescentsin the marginal marine faciesassociation.

291)

P.I( Bose, P.P. Chakraboro,/ Sedimentary Geolo,~' 89 (1994) 285 302

Facies C, granule-rich sandstone T h i s is t h e c o a r s e s t n o t o n l y w i t h i n this associa t i o n but a m o n g s t all t h e f a c i e s d e s c r i b e d , t h e granule content being more than 30%. The granules c o m m o n l y c o n c e n t r a t e at t h e b a s e s o f t h e

f o r e s e t s w h i c h a r e n o r m a l l y g r a d e d (Fig. 8a). T h e c r o s s - s e t s a v e r a g i n g a b o u t 15 c m in t h i c k n e s s f o r m l e n s o i d b o d i e s with an a v e r a g e l e n g t h o f a b o u t 25 c m a n d w i t h t h e i r b a s e s c o n c a v e - u p .

Fig. 8. (a) Granule-filled scour marked by pen (length 15 cm) and arrow. (b) Large-scale cross-stratified bed resting on a master erosion surface (arrow) and overlain by overturned cross-stratifications in the fluvial facies association (2). (c) Facies D overlain by a coset of trough cross-stratifications (facies E) (length of the pen at top 14 cm). (d) Downdipping cross-stratifications within the fluvial facies association. (e) Very low-angle granule ripple laminae within a broad scour marked by a match (length 4 cm). Note very poor sorting of the sediments forming the cross-laminae.

P.K Bose, P.P. Chakraborty / Sedimentary Geology 89 (1994) 285-302

291

Fig. 8 (continued).

Facies D, planar cross-stratified granular sandstone This granular sandstone facies is characterized commonly by planar cross-stratified solitary lenticular bodies of relief up to 80-90 cm on planar or broadly undulating master erosion surfaces (Fig. 8b). Their length ranges from 6 to 9 m, but their width is indeterminable due to outcrop constraints. On the bedding plane surfaces the traces of the foresets are often very broadly curved. The foresets are commonly low-dipping ( ~ 6-9°). When granules are present, they de-

marcate the foresets bases, and the foresets are normally graded. No rhythmic change in the pattern of the cross-strata, as in facies A, is discernible.

Facies E, trough cross-stratified sandstone This sandstone is granule-free, mediumgrained, trough cross-stratified, and forms lenticular bodies. The cross-set thickness is about 10-15 cm and the foresets dip at an average angle of 8°. The morphology of the basal surface of this facies

P.K. Bose, P.P. Chakraborty / Sedimentary Geology 89 (1994) 285-302

292

Fig. 8 (continued).

is either planar and extensive or convex-up and scalloped. T h e r e can be stacks of cosets separated by laterally nonpersistent planar or slightly

curved erosional surfaces (Fig. 8c). These cosets laterally give way to, as well as overlie facies D (Fig. 8c). The coset thickness is on an average 80

NW

.

SE

.

.

.

~t~J~

,~,

~

~

,~"

~ _ _ _ _ _ _ _ _ _

,

|

;° Fig. 9. Vertical stacking of fluvial sandbodies of roughly tabular shape bounded above and below by master erosion surfaces. The smaller sections depict facies successions within individual sandstone bodies at different points. Note lateral variations in facies organisation within individual bodies.

P.K. Bose, P.P. Chakraborty/ Sedimentary Geology89 (1994) 285-302

293

Fig. 10. Two orders of bounding surfaces: planar persistent master erosion surfaces (larger arrows) and internal curved and planar, but laterally impersistent erosion surfaces (smaller arrows) in the fluvial facies association.

c m - 1 m. On bedding plane exposures the troughs are often found to be controlled by the broad curvature of the foresets of facies D.

Facies F, down-dipping cross-stratified sandstone This is characterized by compound crossstratification (Banks, 1973), major cross-strata dipping at about 2 - 3 °, and the smaller internal planar cross-laminae make an angle of about 17-19 ° with them in the same direction (Fig. 8d). The basal surface of such down-dipping crossstrata may be planar or broadly convex-up. These overlie either the master erosion surfaces or facies D or E. On the master erosion surfaces this facies often assume a lensoid geometry having length and height in the range of 3 - 4 m and 60-70 cm, respectively; their width is indeterminable. Some other important structures Significant erosion surfaces. Several orders of major erosion surfaces can be identified within the

association: first-order, planar or broadly undulated master erosion surfaces which confine successive tabular, broadly lensoid bodies embodied by various combinations of the recognised facies (Fig. 9); second-order, interfacies curved, or less commonly flat and locally scalloped erosion surfaces (Fig. 10); third-order, reactivation surfaces within individual cross-sets.

Deformation structures. This association is replete with slump structures (Fig. 11). Besides, overturned cross-stratifications are also present in selected levels (Fig. 12), irrespective of facies. Interpretation Sandstones of this association having a contrasting low degree of textural maturity in terms of grain sorting, and quite often incorporating granules, even small pebbles locally, are not likely products of tide or wind. But in vertical and lateral contiguity with the mixed facies association bearing eolian imprints (see later), they can be of a fluvial origin. The more than 180 ° span in

294

P,K. Bose, P.P. Chakraborty /Sedimentary (;eok)gy 89 (1994) 28.5 302

Fig, 11. Slump structure in the fluvial facies association.

palaeocurrent, the mean being at a considerable angle to the tidal path, also encourages such an interpretation. As vegetation and overbank mud are absent, this Proterozoic river should have had a braided character (Schumm, 1968; Cotter, 1978; Fuller, 1985), and the stacks of extensive tabular or broad lensoidal sandbodies bounded by planar or broadly undulated master erosion surfaces are likely products of a sandy braided river system (Long, 1978, his fig. 22). Scarcity of channel forms, fining-up short sequences and complete absence of muddy overbank deposits are consistent with this idea. The slump folds may, at places, have been engendered by rapid shifting of the river channels undercutting the unstable banks. However, the presumed minimal relief variations between the broad and shallow river channels and

banks cannot explain the observed profusion of slump folding solely by the undercutting of banks. Among the different constituent facies, the cross-stratified granule-rich facies is obviously a product of scour-filling, some manifest side-filling with dip of cross-stratifications decreasing in downcurrent direction. Some rare scours with preserved length exceeding 0.5 m house very low angle (ca. 4-5 °) cross-stratifications in very poorly sorted sediments (Fig. 8e) and these are possible equivalents of granule ripples of aeolian origin (within yardangs) described by Fryberger et al. (1992). The large solitary planar cross-sets presumably represent large unitary bedforms with gently inclined lee faces. These are closely comparable to sinuous-to-lobate linguoid bars described by Smith (1970) from the Platte River

Fig. 12. Overturned crossbedding in successive sandstone beds in the fluvial facies association.

P.IC Bose, P.P. Chakraborty/ Sedimentary Geology 89 (1994) 285-302

system. Their invariable presence on the planar master erosion surfaces suggests their formation on the channel floor during a rising moderately high water stage when the bedshear was high. In contrast, the smaller trough sets (facies E) present in stacks on similar planar master erosion surfaces (cf. Rust and Jones, 1987) and giving way laterally to facies D, possibly formed through migration of small lunate bedforms in relatively finer-grained sediment when the flow was rising but at a comparatively lower stage. Similar channel-fills have been described in the South Saskatchewan River and at Battery Point (Cant and Walker, 1976; Cant, 1978). However, the cosets overlying the convex-up top of facies D bedforms were possibly generated also at a rising but comparatively higher water stage when bed shear diminished on the sandwave surface (cf. Blodgett and Stanley, 1980). The downdipping cross-stratifications which constitute large bedforms on the planar master erosion surfaces are equivalent to the D A macroforms of Miall (1985, 1988, 1991) on the channel floor. There are only a few descriptions of similar downstream accreting macroforms from ancient fluvial sequences (Allen, 1983; Kirk, 1983; Haszeldine, 1983). Kirk (1983) suggested their formation during a high flow stage as from the material already thrown in suspension during a low flow stage. This is corroborated in the present case by the common occurrence of downdipping cross-stratifications above facies D (Fig. 9) or facies E overlying facies D. These occurrences are similar t~ the products of sandflats on top of linguoid bars or on the overbank in the South Saskatchewan River (Cant, 1978). These sandflats in modern settings are active when water spills over the channels. Accordingly downdipping cross-strata are interpreted here as products of a flow stage higher than that suitable for facies E and also D. The grit-filled scours incised on top of all other fluvial facies presumably represent a stage of strongest turbulence and bedshear. Side-filling of the scours suggests emergence, thus we are inclined to take these scours as ultimate products of sheet flow during a falling water stage when bed shear and grain entrainment capacity were at a maximum. This contention and the presumed origin of the

295

associated granule ripples are mutually corroborative. This discussion makes apparent the danger of an a-priori correlation of certain structures with certain microgeomorphic settings or elements within a pre-Silurian sandy river system where grain-size variation in the entire system is limited. Small lunate bedforms may form both on the channel floor and also on top of the sandwaves. The downdipping cross-stratification occurs on the channel floor, on top of the sandwaves, as well as on the sandflat beyond the channel margin. Similarly the grit-filled scours can form anywhere. In the present example, similar structural elements formed at different geomorphic locales during different stages of the flood events, rise or fall. The factors which control the growth of various bedforms most dominantly are perhaps depth of the flow above the sediment substrate (substrate depth) and the bed shear exerted by the flow. The bed shear is in turn dependent on flow velocity and bed roughness. The position of the substrate with respect to any datum plane, such as the channel floor or bank level, is certainly of much less importance in this respect. Ordered successions of various structural elements and fining-up sequences are scarce in this part and this can possibly be explained by the great variability in discharge of sediment and water in such a sandy braided river system (Schumm, 1967; Gregory and Walling, 1973; Rust, 1978). Rapid variability in discharge is clearly documented in the scallops and the scours, as well as in the abundantly present reactivation surfaces. The planar master erosion surfaces confining individual tabular fluvial units are presumably products of successive depositional phases vertically aggraded. Their preservation suggests a rise of the base level of erosion through time which might be either due to basin subsidence or rise in eustatic sea level. We intend to consider basin subsidence as of greater importance for the common occurrence of the overturned cross-beddings; the attribution of this phenomenon to seismic activity by Allen and Banks (1972; but for alternative interpretations see Jones and Rust, 1983) is consistent with our view. Many of the

P.K. Bose, P.P. Chakraborty / Sedimentary Geology 89 (1994) 285~-302

296

more inconsistent, covering about 300 °, and the current rose is polymodal (Fig. 2). The exposure lengths and thickness of the lensoid bodies are also widely variable, ranging from 1.5 to 4 m and from 60 cm to 1 m, respectively. Thick-thin lamina alternations and packages consisting of 10-13 or 21-26 laminae and showing convex- to concave-up transitions similar to facies A crossstratifications, characterize some of these lenses (Fig. 13). Thus these can be attributed to the dominant tidal current. Some other lenticular

slump folds occurring in profusion might also have their origin in basin subsidence. This would mean intermittent enhancement of the rate of basin subsidence within the overall regressive background.

3.3. Mixed facies association This is characterized by intertwined lensoid bodies with variable styles of cross-stratifications. The orientation of the cross-stratifications is still

(b)

11

12

1-

(Crr "8

q ~

I I

i

--

"6-

~'4t"~_-

i

'5

'10 -----='15 Lamina Number

'20

Fig. 13. (a) Thick-thin cross-lamina alternaUons within the mixed facies association (3). (b) Thickness variations in successive cross-laminae within facies I of the mixed facies association. Note thick-thin alternations and number of laminae between successive peaks.

297

P.K. Bose, P.P. Chakraborty / Sedimentary Geology 89 (1994) 285-302

bodies are characterized by trough cross-sets essentially similar t o those in faces E and can likewise be interpreted as fluvial products. With these are associated three new facies described below. Facies G, thinly laminated medium- to fine-grained well-sorted sandstone

This facies of well-sorted, medium- to finegrained and thinly laminated sandstone has a tabular geometry erosionally terminated by some other facies. The internal laminae characteristically tend to be horizontal, maximum dip being about 10° and planar or slightly curved (Fig. 14). Internally there are preserved rippleforms or cross-laminations at places. The evidence of erosion under the laminae is generally minimal. However, within individual tabular bodies of maximum thicknesses of 1 m and exposure lengths of 7.5 m several distinct relatively high-angle erosional planes often marked by lags of slightly coarser grains are present. The laminae overlying them, however, disregard the inclinations of the scour planes but tend to be horizontal. Between successive erosional planes the sets of laminae differ in dip and direction.

e~

e~

o t~

0

0

o .=.

Interpretation. In all details these deposits resem-

ble aeolian interdune erg deposits described by Clemmenson and Hegner (1991; see also Wilson, 1973; Kocurek, 1988). The growth of these inferred erg deposits in the Rewa Formation might have been controlled by climate, tectonism and sand supply (Kocurek 1988, 1991). However, the more immediate cause for the short-term growth pattern might have been the rise and fall of the water table induced by climate a n d / o r tectonism, and that possibly largely controlled the rate of sand supply by means of deflation and weak saltation (Clemmensen and Dam, 1993). Facies H, fine-grained sandstone with translatent strata

This fine-grained, well sorted sandstone is characterized by low-angle planar strata with a general thickness of about 1.5-2 cm (Fig. 15). Significantly these strata show a distinct concen-

o~

e~