Sedimentology and Stratigraphy of the Eocene

Home , Gauja Formation, Ophiomorpha, Planolites

FACIES ARCHITECTURE AND STRATIGRAPHY OF TIDAL RIDGES

IN THE EOCENE RODA FORMATION, NORTHERN SPAIN

KAIN J. MICHAUD

A THESIS SUBMITTED TO THE DEPARTMENT OF GEOLOGICAL SCIENCES & GEOLOGICAL ENGINEERING IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF

MASTER OF SCIENCE

QUEEN’S UNIVERSITY KINGSTON, ONTARIO, CANADA APRIL, 2011

It doesn't matter if the water is cold or warm if you're going to have to wade through it anyway.

~Teilhard de Chardin

i ABSTRACT

The Eocene Roda Formation in northern Spain documents the deposits from a range of coastal depositional environments. These include alluvial plains, distributary channels, mouth bars, upper to lower-shorefaces, and tidal shelf ridges. Eighteen progradational sand tongues that are interpreted as parasequences compose two third-order sequences. Sequence 1 accumulated in an environment with strong tidal currents and high rates of progradation, while

Sequence 2 was deposited under relatively weaker currents and higher rates of aggradation, which produced a higher mudstone:sandstone ratio.

The stratigraphy highlights the transgressive origin of six tidal shelf ridges, three in each sequence, that overlie regressive deltaic tongues. Sequence 1 shelf ridges are composed almost entirely of cross-bedded sandstones, whereas Sequence 2 ridges are composed of a mixture of cross-bedded and ripple-laminated deposits. Ridges in both sequences contain bioturbation that is typical of the Cruziana Ichnofacies, and that indicates a marine origin.

The tidal ridges are stratigraphically located at or near the point of maximum third-order regression, and are not found within early highstand or late transgressive deposits― times of high relative sea level when the deltaic shoreline did not protrude significantly. Tidal currents were accentuated at the coast when the delta complex had prograded several kilometres into the basin, while during times of high relative sea level, the basin was wider and tidal currents were weaker, consequently leading to a lack of tidal deposits. The tidal ridges are, thus, interpreted as being headland-associated deposits.

ii ACKNOWLEDGEMENTS

My time at Queen’s University has been both exciting and rewarding, and it gives me great pleasure to express my appreciation to all of you who have supported me.

Dr. Robert Dalrymple provided insightful guidance while patiently awaiting the completion of the project. I thank him for developing my geologic skills, our pleasant insightful discussions and for sending me to Spain, Bermuda, New York and Quebec (Ft.

McMurray excluded). Bob let me find my own way through scientific problems, encouraging out-of-the-box thinking, which inevitably led to numerous fruitful debates. While he won most of those debates (or all depending on who you ask), I like to think I came out the victor with a strong education and a resilient thesis.

Fellow graduate students provided much needed distractions and discussions, resulting in the improvement of the thesis content alongside my foosball and ping pong skills. Few days went by without discussions with Duncan Mackay, Aitor Ichaso, Cody Miller, Geoff Reith,

Meg Seibal, Pim Van Geffen, Nick Riordan and Laura O’Connell. I would like to thank all of you for the great company and insightful conversations. Bryce Jablonski deserves special mention for his hard work in Spain. Thanks again for not complaining about the long hikes, carrying the heavy pack, and constantly reminding me to take what I had forgotten.

I would also like to thank Allard Martinius, Cornel Olariu, and Ron Steel for introducing me to the study area. My time spent there would have been exponentially more confusing without your guidance. Local Spaniard, Jose Antonio Naval, treated me like a family member during my stay and I am especially grateful to his family for sharing their home and culture.

iii Also my family’s contributions must be recognized. Continued support and guidance from my parents over the last 7 years of my education made this thesis possible and I cannot thank them enough for that. My wife deserves pages upon pages of credit for her patience and motivation; however, for the sanity of readers I will conclude with a simple— thank you Linda.

This project would not have been possible without funding which was graciously provided by NSERC grants to Robert Dalrymple and by Queen’s University.

STATEMENT OF ORIGINALITY

The following thesis is a compilation of my own field work, literature review and insight.

Nevertheless, Dr. Robert Dalrymple’s contributions must be recognized as he has provided invaluable guidance and revisions that greatly improved the scientific and grammatical quality of the thesis.

This study is unique as it 1) is the first to document the detailed sedimentology and stratigraphy of the Eocene Esdolomada Member, a 180 m thick mixed siliciclastic/carbonate succession, and also 2) illustrates the range of facies present within tidal shelf ridges and documenting the distribution of facies, stratigraphic location and variability of 6 ridges.

TABLE OF CONTENTS

Abstract ...... ii

Acknowledgments ...... iii

Statement of Originality ...... iv

Table of Contents ...... v

List of Figures ...... vii

List of Tables ...... viii

CHAPTER 1

INTRODUCTION

1.1: General Introduction ...... 1

1.2: Geologic Setting ...... 3

1.3: Stratigraphic Background ...... 4

1.4: Methods ...... 6

CHAPTER 2

THE FACIES , SEQUENCE-STRATIGRAPHIC FRAMEWORK AND COMPARISON OF

SHELF RIDGES IN THE RODA FORMATION, NORTHERN SPAIN

2.1: Introduction ...... 8

2.2: Facies Analysis ...... 8

2.3: Stratigraphic Framework ...... 21

2.4: A Comparison between the Upper and Lower Sequences ...... 25

2.5: Conclusions ...... 29

CHAPTER 3

FACIES ARCHITECTURE OF THE TRANSGRESSIVE TIDAL RIDGES

3.1: Introduction ...... 31

3.2: Facies Analysis ...... 33

3.3: Facies Distributions, Isopach Patterns and Paleocurrents ...... 45

3.4: Ridge B ...... 48

3.5: Ridge F ...... 53

3.6: Origin of Tidal Ridges ...... 55

3.7: Comparison between Roda and Esdolomada members ...... 59

3.8: Variation between Ridges ...... 60

3.9: Conclusions ...... 62

CHAPTER 4

CONCLUSIONS

4.1 Concluding Remarks ...... 65

REFERENCES ...... 68

APPENDIX

A.1 Measured Sections ...... 89

A.2 Outcrop Photographs and Uninterpreted Figures from Previous Chapters ...... 112

LIST OF FIGURES

CHAPTER 1

Figure 1.1: Regional Geology ...... 2

Figure 1.2: Stratigraphic Section ...... 5

Figure 1.3: Outcrop Outline ...... 6

Figure 1.4: Study Location ...... 7

CHAPTER 2

Figure 2.1: Facies A Channels and Facies B Alluvial Plain ...... 11

Figure 2.2: Facies C Delta Mouth Bar ...... 14

Figure 2.3: Facies D Tidal Ridge ...... 15

Figure 2.4: Facies E and F Shelf Mud and Carbonate Facies ...... 18

Figure 2.5: Facies G: Thin Shelf Tempestites and Turbidites ...... 20

Figure 2.6: Stratigraphic Cross-Section ...... 22

Figure 2.7: Eustatic Sea-level Curve ...... 27

CHAPTER 3

Figure 3.1: Stratigraphic Cross-Section with Tidal Ridges A-F Identified ...... 35

Figure 3.2: Sandy Tidal Ridge Facies ...... 37

Figure 3.3: Evidence of Tidal Influence ...... 40

Figure 3.4: Tidal Sand Ridge Ichnology ...... 41

Figure 3.5: Muddy Tidal Ridge Facies ...... 44

Figure 3.6: Isopach and Paleocurrent Maps ...... 46

Figure 3.7: Facies Percentage and Bioturbation Index of Tidal Ridges ...... 47

vii

Figure 3.8: Oblique Cross-section of Ridge B ...... 49

Figure 3.9: Oblique Cross-section of Ridge F ...... 54

Figure 3.10: Paleogeographic Reconstruction ...... 57

Figure 3.11: Modern Analogue ...... 58

Figure 3.12: Variation Between Tidal Ridges ...... 61

LIST OF TABLES

CHAPTER 2

Table 2.1: Facies of the Roda Formation ...... 9

CHAPTER 3

Figure 3.1: Tidal Ridge Facies ...... 34

viii

CHAPTER 1

INTRODUCTION

1.1 General Introduction

One of the most influential and frequently cited examples of ancient tidal sedimentation is the Eocene Roda Formation in northern Spain (Fig. 1.1). Among the first studies in the area,

Nio (1976) documented transgressive tide-generated compound dunes and well-exposed tidal bundles. The tidal bundles were further investigated by Yang and Nio (1985) who documented the cyclic variation in thickness of bundles through neap-spring cycles. Nio and Yang (1991) then investigated the entire Roda Member, which constitutes the lower half of the Roda

Formation, interpreting a range of sandy deposits including elongate sand bars, tidal point bars and deltas. More recent work has highlighted the well exposed Gilbert-style delta deposits

(Lopez-Blanco, 2003; Tinterri et al., 2007). These studies typically depict tidal sandstones as lying on top and seaward of deltaic wedges, but do not provide any detailed description or interpretation of their origin. Drawing on some decades of literature, further studies (Tinterri et al., 2007; Leren, 2010) have reviewed and updated the stratigraphy of the lower Roda Member of the Roda Formation.

The steep-fronted coarse-grained deltas within the lower Roda Member have received the most attention, while tidal deposits have remained poorly documented, despite the early interest by Nio (1976) and Nio and Yang (1991). Tide-generated compound dunes have been illustrated as developing during deltaic progradation (Lopez-Blanco et al., 2003; Tinterri et al.,

2007), but several problems exist with this interpretation. Firstly, the tidal deposits have a diverse marine ichnofacies which is inconsistent with a deltaic origin. Second, there is one tidal deposit that is disconnected by over a kilometre from its respective delta. Finally, the tidal deposits overlie and often erode into underlying deltaic deposits. All of these observations

indicate the former interpretation― that the tidal deposits are regressive in nature― is incorrect.

Figure 1.1: (A) General location map and (B) more detailed geologic map showing the main sedimentary basins formed in the southern foreland area of the Pyrenean Fold and Thrust Belt. The deposits under investigation—marked by a star in inset box-- were deposited in the Eocene Tremp-Graus Basin (4) which was closed to the east as a result of uplift associated with the transverse Segre Fault, and opened westward into deeper water in the Basque Basin (1) and the proto-Atlantic Ocean.

The goals of this thesis are, thus, as follows: 1) to document the stratigraphic framework of the entire Roda Formation, including the origin and controlling processes of various sandstone bodies in both the Roda Member and the formerly undescribed Esdolomada Member;

2) record and evaluate the causes of the differences between the two members of the Roda

Formation; and 3) describe the three-dimensional facies organization within the tidal deposits in the Roda and Esdolomada members, and discuss the controls on their occurrence.

1.2 Geologic Setting

The Roda Formation is an Eocene (Ypresian) mixed clastic/carbonate unit that prograded into and infilled a piggy-back basin on the southern flank of the Pyrenean fold and thrust belt (Fig. 1.1). The fold and thrust belt formed due to the collision of the Iberian and

Eurasian plates, beginning in the late Cretaceous (Puigdefàbregas et al., 1992; Anadón and

Roca, 1996). North-south compression led to crustal thickening along an E-W trending fold- and-thrust belt and the development of compressive structures in the Catalan Coastal Range

(early Eocene). As the Iberian plate subducted beneath Eurasia, southerly-propagating thrusts caused subsidence of the northern part of the Iberian plate, creating a foreland basin, parts of which lay on top of the moving thrust sheets.

The early Eocene foreland basin is segmented into three depozones by the Pamplona and Segre tectonic structures (Fig. 1.1): (1) an eastern basin including the Terrades and Cadi basins; (2) a central zone including the Tremp-Graus, Ainsa, Jaca, and Ager basins; and (3) a western zone including the Basque Basin (Puigdefàbregas, 1992; Verges and Burbank, 1996).

The central basin is further segmented into three main sub-basins, (1) the shallow-marine

Tremp-Graus Basin in the east, which passes westward into the deeper (2) Ainsa Basin that is bounded by the Mediano anticline in the east and the Boltana anticline in the west, and finally into (3) the Jaca Basin in the west, which opened westward into the Basque Basin, the Bay of

Biscay and the Proto-Atlantic Ocean. A small satellite basin, the Ager Basin opened to the southeast off the southern side of the Tremp-Graus Basin.

The deposits that are the subject of this study, crop out in an erosional window along the northern flank of the Tremp-Graus Basin. This basin originated as a thrust-top or piggyback basin situated between two southward directed thrusts—the Boixols thrust to the north and the

Montsec thrust to the south. The basin was asymmetrical with a steeper northern margin, from

which most of the sediment infill was sourced, and a gentler southern flank. Deposition within the basin documents some of the initial stages of foreland-basin sedimentation in the Pyrenees.

1.3 Stratigraphic Background

The Tremp-Graus Basin contains a thick succession of Late Cretaceous to Oligocene deposits (Fig. 1.2). Initial subsidence during the Cretaceous created accommodation that was filled with strata deposited in fluvial and lacustrine environments of the Tremp Formation (Nio,

1976). As the rate of base-level rise increased into the Paleocene and Eocene, the basin was flooded, creating a shallow warm sea approximately 25 km in width (narrowing to the east) and

90-100 km in length (Plaziat, 1981; Verges and Burbank, 1996). Past studies focusing on paleo-fauna reveal that in the Roda area, water depth reached approximately 80 m, but was typically only 10-30 m, with water temperature around 20 °C (Martinius, 1995; Molenaar and

Martinius, 1996; Torricelli et al., 2006). Clear, shallow, warm marine water allowed shallow- marine organisms to thrive, resulting in thick calcareous deposits. Early limestones (the

Alveolina and La Puebla limestones; Fig. 1.2) are composed of red algae-coral reefs in the north, with carbonate ramp facies to the south. They interfinger with basinal marls of the

Serraduy Formation that accumulated in deeper water (but still < 60-80 m deep) to the south and west (Martinius, 1995; Torricelli et al., 2006).

Periodic clastic deposition occurred along the northern margin of the basin as rivers transported sediment from rising highlands in the north east. The Roda Formation accumulated over approximately 1 My, from ~53-52 Ma according to biostratigraphic data (Torricelli, 2006) or from ~53.4-52.4 Ma according to magnetopolarity (Lopez-Blanco et al., 2003) (Figure 1.2).

It is composed of interbedded fossil-rich mudstones, carbonates, and several ~20 m-thick sandstone tongues that stack to form two larger sandstone wedges, each 150 m thick, separated

by a shale succession. The formation is well-exposed along both margins of the Isabena River valley (Fig 1.3) and is locally exposed in smaller channels cutting into the valley wall, thus allowing for partial 3D coverage.

Figure 1.2: Stratigraphic terminology and representative measured section (see location of measured section in Figure 1.4) through the study interval. The generalized terminology of Mutti (1988) works well at a basin-wide scale, whereas the more detailed subdivisions proposed by Nio and Yang (1991) match the depositional succession in the study area best and are used here. Ages from Lopez-Blanco et al. (2003).

The Roda Formation is overlain by the 10-30 metre-thick Morillo Limestone, which is in turn overlain by sandstones and conglomerates of the San Esteban Formation. In the study

area the San Esteban Formation is composed of amalgamated channel sandstones and conglomerates of fluvial origin; equivalent deposits to the south have been interpreted as deltaic

(Van Eden, 1970).

Figure 1.3: Isabena River valley and outcrop of the Roda Formation. Local towns are circled and identified. Photograph is taken looking toward the southwest from above the town of Serraduy (see Fig. 1.4 for town locations and scale.)

1.4 Methods

Outcrop investigation of the Roda Formation in the Isabena River valley took place during two field seasons totalling over two months, with the work concentrated in the vicinity of Roda de Isabena (Figs. 1.3 and 1.4). Twenty-one stratigraphic sections totalling more than 3 kilometres in length were logged at a centimetre scale (Appendix A1). The vertical sections have horizontal separations ranging from under 100 metres to 2 kilometres. The area between sections was investigated, photographed (Appendix A2), and physically walked. Special attention was given to tracing significant bounding surfaces in an effort to correlate the vertical logs with confidence and to understand horizontal stratal transitions. The data acquired

includes high-resolution photographs of outcrops, and detailed observations of sedimentary structures, paleocurrent orientations, thickness of units, bioturbation diversity and intensity

(sensu Bann et al., 2004), fossilized organisms (sensu Martinius, 1995), grain-size trends, sedimentary structures and the attitude of bedding planes that define the architecture of sediment packages.

Figure 1.4: Study area in the Isabena River valley. See regional location in Figure 1.1 and panoramic photo in Figure 1.3. Locations of measured sections are marked 1-21. The strata under investigation are the Roda (ro) and Esdolomada (es) members of the Roda Formation (see Figure 1.2).

CHAPTER 2

THE FACIES, SEQUENCE-STRATIGRAPHIC FRAMEWORK AND COMPARISON

OF SHELF RIDGES IN THE RODA FORMATION, NORTHERN SPAIN

2.1 Introduction

The Roda Formation is composed of interbedded carbonate and siliciclastic sediments that can be subdivided into 6 genetically distinct facies (Table 2.1), each interpreted as having formed in a specific depositional environment. Grain size, bioturbation, fossil content, and sedimentary structures differentiate them. The following section will document the composition and architecture of these facies and interpret their origin. Following this analysis, the deposits are placed in a stratigraphic framework. The stratigraphic location of six shelf tidal ridges is documented and discussed. These tidal ridges are shown to form during high- frequency flooding events that punctuate two third-order sequences, occurring in third-order highstand, lowstand and transgressive systems tracts. Since this is the first time the sedimentology of the entire Roda Formation has been documented, the differences between the members are also discussed and interpreted.

2.2 Facies Analysis

Facies A: Distributary Channel

Description: Facies A (Fig. 2.1; Table 2.1) is composed of upward-fining arkosic sandstone successions ranging from 1-9 m thick that can be internally cross-bedded.

Sandstones are limited to laterally discontinuous coarse lenses, commonly pinching out from 2-

3 m in thickness to 0 m over a distance of 20 m. The base of each succession is composed of pebble- to boulder-sized clasts, becoming increasingly sandy upward. These basal lags are

Facies Description Ichnology Interpretation A Upward-fining 1-9 m thick discontinuous sandstone Bioturbation intensity commonly 0-2 Fresh to fully marine lenses with common pebbles and boulders. Contains: infrequently 5-6. Contains: channel deposits likely cross beds, red staining, peaty horizons, common plant Ophiomorpha, Thalassinoides, influenced by tides and and wood fossils, rhizolith traces, rare shelly debris Planolites, Gastrochaenolites, and variable river discharge. . Teredolites. B Poorly exposed dark purple to brown, planar bedded No distinguishable burrows. Alluvial floodplain mudstone interbedded with Facies A deposits. Common deposits frequently plant debris, peat and coal laminations. Rare rhizoliths. covered by vegetation and periodically influenced by river floods.

C i 5-20 m thick gradationally based coarsening-upward Bioturbation intensity is typically 0-2 Delta mouth bar deposit arkosic sandstone successions composed of large-scale although the top of the facies often with a very high rate of angle-of-repose cross-beds. Rare gravel and trace reaches 3-6. Contains: Ophiomorpha sedimentation relative to cobbles. and Thalassinoides. basin energies.

C ii 5-20 m thick gradationally based coarsening-upward Bioturbation intensity is typically 0-2 Delta mouth bar deposit arkosic sandstone successions composed of compound although the top of the facies often with a high rate of cross-bedded sandstones. Master-bedding surfaces dip at reaches 3-6. Contains: Ophiomorpha sedimentation relative to 5-15° and separate cross-bed sets 10-30 cm thick. Cross- and Thalassinoides. basin energies. beds dip in the same direction as master-bedding surfaces.

D Typically coarsening upward, sharply based, 25 m thick Bioturbation intensity ranges from 0- Elongate tidal sand ridges sandstone succession composed of cross-bed sets 10-300 5. Contains: Rosselia, Cylindrichnus, that accumulated in a m thick and cross-laminations which dip at 90° relative Asterosoma, Planolites, fully marine setting. to master-bedding surfaces. Common herringbone cross- Thalassinoides, Ophiomorpha, Deposition was controlled stratification, mudstone-sandstone rhythmites, mudstone Teichichnus, cryptic bioturbation, by tidal currents and drapes, and stacked tidal bundles. Palaeophycus, Chondrites, Skolithos, water was turbid. Rusophycus, Arenicolites, Scolicia, Bergaueria, Protichnites, Macaronichnus, Polykladichnus, Cruziana, Teichichnus, Asteriacites, Psammichnites, and Phycosiphon.

E Typically 1 m thick and locally cross-bedded sandstone Bioturbation intensity is typically 5- Tide and wave influenced composed of a mix of arkosic sandstone and carbonate 6. Contains: Thalassinoides, shallow-marine carbonate grains. Carbonate grains typically include foraminifera, Rosselia, Ophiomorpha, and Scolicia. shelf. echinoderms, gastropods, and bivalves with rare hermatypic coral. Arkosic sandstone grains are typically coarse grained.

F Fossiliferous planar parallel laminated mudstones Bioturbation intensity NA (obscured Distal hemipelagic- containing abundant gastropods and large benthic by mudstone compaction). Contains dominated prodelta and foraminifera with common bryozoan and trace crab, local Thalassinoides. lower-shoreface. stingray, echinoderm, nautiloids, coral and bivalve.

G Typically 10-50 cm thick very fine to very coarse Bioturbation intensity typically 0-2. Turbidites and sandstones encased in facies F muds. Contain: fossilized Wood clasts contain Teredolites. tempestites on a shallow- wood, shells, plant debris, tool marks, flute casts, current marine shelf. ripples, wave ripples, and hummocky cross-stratification.

Table 2.1: Facies of the Roda Formation.

typically less than 10 cm thick, and sometimes contain mud rip-up clasts. Above the lag, sands are very poorly sorted with the matrix being very coarse sand, with out-sized clasts ranging from pebble to boulder size. Cross-beds are present but are difficult to distinguish in the coarse poorly-sorted sands. Red staining of sandstones is common. Peaty horizons are present near

the top of the successions and can be thinly interbedded on the millimetre scale with mudstones. Plant and wood fossils are abundant and some layers contain rhizoliths.

Occasionally, shelly debris and trace fossils (Ophiomorpha, Thalassinoides, and

Planolites) can be found within Facies A. Bioturbation intensity (quantified by the bioturbation index or BI sensu Bann et al., 2004) ranges from 0-2 (common) to 5-6 (infrequent). Clasts often contain the borings Teredolites or Gastrochaenolites. Teredolites is found in localized pockets with fossilized wood between borings.

Interpretation: These upward-fining lenses of coarse, poorly-sorted sandstone are interpreted as channel deposits. Conditions were periodically favourable to bioturbation, as evidenced by Ophiomorpha, Thalassinoides, Planolites and Gastrochaenolites, but were typically inhospitable as indicated by the general paucity of burrows. The channelized architecture, large grainsize and ichnological signature suggests shifts in salinity from fresh water to brackish as a result of variations in river discharge. Fine inter-bedding of coffee- ground organic debris and muds may indicate periodic tidal influence as indicated by tidal facies discussed later in this section.

Facies B: Alluvial Floodplain

Description: Facies B (Fig. 2.1; Table 2.1) is composed of laminated dark purple to brown mudstone interbedded with Facies A channel deposits. Shelly fossils are generally absent and plant debris is common, forming centimetre-thick peat and coal laminations with abundant rhizoliths. Facies B is typically poorly exposed, and the crumbly nature of the outcrops makes the differentiation of sedimentary structures difficult, although planar lamination can be observed locally.

Figure 2.1: A and B) Facies A channels consist of lenses of sandstone several metres thick that completely pinch out within metres. Lenses fine upwards and are typically composed of the coarsest sediment in the Roda Formation. Muddy material overlying or adjacent to the channels is rich in plant fossils and often contains coalified horizons. Marine fossils are typically absent. A) highlights the edge of a sandy point-bar that is draped by a muddy abandoned channel-fill deposit. B) outlines a highly aggradation channel with numerous internal lags. Scale bar is 150 cm long and is marked at 10 cm intervals.

Interpretation: The absence of marine fossils and abundance of plant fossils combined with the stratigraphic occurrence alongside Facies A channel deposits, indicates that Facies B was deposited on alluvial floodplains. Frequent alternation between sandstone and mudstone likely represents periodic deposition of crevasse splays or sheet-flood deposits. Abundant plant fossils and rooting indicates vegetative cover.

Facies C: Delta Mouth Bar

Description: Facies C is typified by unbioturbated, coarsening-upward successions, 5-20 m thick. It is composed of upper medium to upper very coarse sandstone (Fig. 2.2; Table 2.1).

The arkosic sands (approximately 50% quartz and 50% feldspar) are very poorly sorted and angular. Gravel can be found along inclined master-bedding, and locally in topsets. Facies C forms extensive, laterally continuous tongues that can be traced for 1-4 km downdip. The internal architecture generally manifests in one of two ways.

Architecture i: These deposits are composed of large-scale coarse-grained angle-of- repose cross-beds that attain thicknesses of up to 20 m (Fig. 2.2 A and B). These large foresets reach from the top of the facies to the bottom with dips of approximately 25-30°. The toes of the foresets contain a high percentage of silt and clay, dipping at much lower angles (<1°-5°) and gradually become finer with increasing distance from the correlative foreset. The base of the deposit is gradational, passing upward from entirely muddy low-angle toesets at the base into siltstones followed by a relatively rapid transition into coarse-grained sandstones.

Architecture ii: The deposits are composed of compound cross-bedding with coset thicknesses that are equivalent to the height of adjacent high-angle foresets (Fig. 2.2 C). Large- scale master-bedding surfaces dipping 5-15° separate cross-bed sets that are 10-30 cm thick.

These cross-bed sets have a 180° spread of dips, but, on average, dip in the same direction as the master-bedding surfaces. The muddy toes of the foresets dip at angles of <1°-4°. The lower angle of master-bedding within Facies C ii (5-15° compared with 25-30° of Facies C i) is associated with a thicker heterolithic base and a more gradual transition into the bottomset beds than is seen in Facies C i.

Internally, Facies C has limited bioturbation. The trace fossils present are Ophiomorpha and Thalassinoides with a bioturbation index of 0-2. However, the top 20-150 cm of the facies can have a significantly higher bioturbation index of 3-6. The eroded tops of Facies C can also contain larger (10 cm in diameter) crustacean domiciles and abundant echinoderm spines.

Interpretation: These sandstone tongues are interpreted as delta mouth bars that prograded periodically into the marine basin. High rates of sedimentation, abundant coarse- grained sediment and sufficiently deep water are responsible for the archetypal Gilbert-style delta architecture seen in Facies C i (Postma, 1990). Gilbert-style architecture forms when river energy is strong relative to basin processes (i.e., waves and tidal currents) that are unable to redistribute the sediment. The compound cross-bedded architecture seen in Facies C ii forms when relative river energy is somewhat lower, although still higher than basinal energies.

Similar architectural differences are noted in the Frontier Formation of central Wyoming

(Willis et al., 1999). Those authors interpret similar large angle of repose beds as being the deposits from strongly asymmetric currents on a tide-influenced delta bar, whereas compound cross-bedded facies on the delta bar accumulated where flood and ebb currents were more equally matched.

Topsets are rarely preserved, and are interpreted to either have not been formed or as having been eroded during the ensuing transgression. Evidence of transgression is provided by the intensely burrowed top and the marked increase in marine fossils in the immediately overlying deposits, which represent a period of sediment starvation and reworking.

Figure 2.2: Facies C delta mouth-bar deposits typically have steeply inclined foresets of subfacies C i (A and B), although some locations show lower angle foresets that are internally cross-bedded forming subfacies C ii (C). Deltaic deposits coarsen upward from prodeltaic mudstones of Facies F (see bottom of A), through interbedded sandstones and mudstones, and then into large coarse delta-front foresets. Scale bar is 150 cm long and is marked at 10 cm intervals.

Facies D: Tidal Ridge

Facies D is complex and heterogeneous, warranting an in-depth analysis. This is done in the following chapter. However, a condensed description and interpretation is presented here, in order to illustrate the general nature of this facies and to place them into their stratigraphic context.

Description: Facies D (Fig. 2.3; Table 2.1) is composed of cross-bed sets 10-300 cm thick and cross-laminations which dip at ~90° to the master-bedding planes that separate the sets, in contrast to the cross beds in Facies C ii that dip down the master-bedding planes. Sand grain size generally coarsens upward as the mud percentage decreases, although this is not a universal trend. The base can be sharp and erosive, or vertically gradational from the underlying mudstones over 1-3 m. The sands are well sorted and sub-angular. Herringbone cross-stratification, stacked tidal bundles, mudstone-sandstone rhythmites and double mudstone drapes are present, although the identification of rhythmites and mudstone drapes is generally limited to overhangs where mudstones have not been removed by weathering. Facies D forms elongate ridges of sandstone, with maximum thicknesses of 15-25 m.

Figure 2.3: Facies D sandy tidal-ridge deposit (outcrop is 20 m high) composed of cross-beds dipping along master-bedding (i.e., parallel to their strike). In this annotated image, most cross-beds dip toward the northwest, whereas master-bedding surfaces dip at a low angle toward the southwest. See Appendix A.2 for unannotated version.

The trace-fossil assemblage present within Facies D belongs to a fully marine Cruziana

Ichnofacies (unlike Facies C). Trace fossils include Rosselia, Cylindrichnus, Asterosoma,

Planolites, Thalassinoides, Ophiomorpha, Teichichnus, cryptic bioturbation, Palaeophycus,

Chondrites, Skolithos, Rusophycus, Arenicolites, Scolicia, Bergaueria, Protichnites,

Macaronichnus, Polykladichnus, Cruziana, Teichichnus, Asteriacites, Psammichnites, and

Phycosiphon.

Interpretation: This facies represents the deposits of elongate ridges that accumulated in a fully marine setting as evidenced by the trace-fossil assemblage. Deposition was controlled by tidal currents, as indicated by the widespread occurrence of tidal bundles, mud drapes and herringbone cross-bedding. Water would have been turbid as indicated by the abundance of mudstone drapes which would have settled out during slack-water periods.

Facies E: Shallow Carbonate Shelf

Description: Facies E consists of typically 1 m-thick, intensely bioturbated carbonate- rich beds that lie within the Roda Formation and commonly cap Facies C mouth-bar deposits

(Fig. 2.4; Table 2.1). Composition ranges from 90% sandstone with minor carbonate grains to

90% carbonate with minor admixed sandstone, with an upward-increase in carbonate content.

Beds of this facies are not present in the far north of the study area and generally thicken and rapidly become more carbonate rich to the south, where they reaches 2 m in maximum thickness. Grain size is largely a function of the carbonate grains present. It ranges from coarse sand to pebble packestone/grainstone; the sand component is typically coarse sand.

Martinius (1995) has identified the different biofacies of the Roda Formation, noting that the typical fauna found in this facies consists of multiple species of foraminifera, fragmented echinoderms, numerous gastropods (Turritella sp., Melanatria, Sycostoma, Naticidae), and

bivalves (Crasatella depressa, Mytilus inflatus, Crassostrea rarilamella, and Ostrea multicostata). There are localized patches with abundant hermatypic coral (Goniaraea elegans) that have been interpreted by Molenaar et al. (1988) as developing on marine hardgrounds.

Facies E typically has a bioturbation index of 5-6, with Thalassinoides, Rosselia,

Ophiomorpha and Scolicia forming a Cruziana Ichnofacies. The high bioturbation index obscures bedding, but cross-bedding is found locally.

Interpretation: Facies E was deposited in marine waters within the photic zone, during a time of negligible siliciclastic and fresh-water input. This allowed the establishment of a high-diversity assemblage of benthic organisms. These conditions occurred on the top of abandoned deltas, which formed bathymetric highs, and in interlobe bays (Martinius and

Molenaar, 1991). The sorting of carbonate grains and local cross-bedding indicates some of these beds were winnowed, likely by marine currents and waves.

Facies F: Shallow Muddy Marine Shelf

Description: Moving further to the south into more distal settings, coarse mouth-bar sandstones give way to increasingly silty and muddy deposits containing up to 50% fossil content, but typically 5-20%. Fossils are similar to those found in Facies E but are mud supported. Adjacent to mouth-bar sandstones, the facies contains abundant Turritella gastropods (Fig. 2.4 A). As the deposits become muddier, Turritella gastropods decrease in abundance and large benthic foraminifera become more abundant. In addition to the dominant gastropods and foraminifera (Fig. 2.4 D and E), other fossils present include bryozoa, crabs, stingray barbs, echinoderm spines, nautiloid shells, corals, and numerous types of bivalves (Fig.

2.4 B, C, D and E; Table 2.1). Sedimentary structures are limited to extensive planar parallel

Figure 2.4: A) Photograph in the southwest at measured section 4 illustrating cyclic deposition between Facies E carbonate grainstones (resistant) and Facies F shelf muds (recessive) with thicker event deposits of Facies G. These mudstones contain abundant gastropods (E) and large benthic foraminifera (L.B.F.; D); however, marine fossils like crabs (B) and stingray barbs (C) are more commonly found in Facies E.

laminations millimetres to centimetres thick. Bioturbation is typically difficult to assess due to the unimodal mud grain size, although some burrows contain abundant foraminifera and are identifiable within mudstone as Thalassinoides. Those burrows are typically oval in shape and are five times wider than high.

Interpretation: Facies F represents the resultant deposit from a distal hemipelagic- dominated prodelta and lower-shoreface setting. Water depth was 60-80 m (Martinius, 1995) and muddy sedimentation occurred near the delta. The sedimentation rate was relatively low away from the delta, allowing for benthic organisms to thrive. Compaction has flattened burrows making their distinction difficult. Facies E beds lying within Facies F mudstones are interpreted to have reworked and winnowed Facies F.

Facies G: Shallow-Marine Shelf Clastic Tempestites and Turbidites

Description: In the northeast, the muddy marine shelf deposits inter-finger with coarse grained deposits of Facies G (Fig. 2.5; Table 2.1). These deposits occur in beds that are typically 10-50 cm thick, but can range from a few centimetres to 1 m in thickness. Stacking of packages is uncommon, and individual beds are typically encased in Facies F mudstones.

Grain size ranges from very fine sand to very coarse sand. Sedimentary structures include tool marks, flute casts, current ripples, wave ripples and hummocky cross-stratification. Shells and fossilized plant debris are present and abundant. Teredolites is found at numerous intervals within these units, and is typically found in boulder sized clasts of fossilized wood. Bedding planes are marked by plant debris, often giving the outcrop a flaggy appearance.

Interpretation: Facies G is interpreted as shallow shelf sands. Encased in Facies F, they were clearly deposited in a marine environment. The presence of hummocky cross-

stratification indicates deposition by waves. Not all deposition was wave generated, however, as thin sandy deposits often contain flute casts, current ripples and tool marks with similar orientations to direction of deltaic progradation (Facies C). Because those thin sandstones are surrounded by marine muds and contain unidirectional current-generated structures they are interpreted as turbidites. Boulder-sized clasts of Teredolites colonies indicate large wood fragments were present. The abundance of plant fossils within turbidites indicates wood fragments were transported within the turbidity currents into a marine basin during floods and storms.

Figure 2.5: Shallow-marine sandstones of Facies G display (A) flaggy hummocky cross-stratified sandstone, (B) current ripples, (C) abundant plant debris on bedding planes, and (D) tool and drag marks.

2.3 Stratigraphic Framework

A cross-section of the study area (Fig. 2.6) illustrates the distribution of depositional environments within the Roda Formation. In general, channelized sandstones and alluvial-plain mudstones are abundant in the northeast, delta mouth bar and tidal ridge sandstones are located in the middle of the study area, and shelf mudstones are abundant in the southwest. Shallow shelf sandstones are most abundant in the upper sequence and are found at various levels throughout it.

High-frequency 4th to 5th-Order Parasequences

Within the Roda Formation, 18 progradational tongues of coarse sediment extend to the southwest into the Tremp-Graus Basin, and are overlain by flooding surfaces (Fig. 2.6). As the formation was deposited over 1 My, each sandstone tongue and adjacent mudstone represents a time-span of approximately 55 ky. In the northeast, tongues are composed of coarse-grained channel and alluvial-plain deposits that pass southwest-ward into upward-coarsening, gradationally based delta mouth-bar deposits. Mouth-bar deposits can be traced up to 4 km in the dip direction (to the southwest), and are exposed for up to 3 km along a coast-parallel, northwest-southeast trend (see also Lopez-Blanco et al., 2003). Mouth-bar sandstone tongues, which are typically 5 m thick in the far northeast, thicken to a maximum of 20 m towards the southwest and then pinch out into prodeltaic muds 2-5 m thick. The outcrop allows the western, southwestern and southern extents of the tongues to be confidently delineated; however, they subcrop towards the southeast so their extent in that direction is unknown.

Within a single progradational cycle, multiple mouth-bar deposits can be differentiated, showing higher-frequency cyclicity.

Figure 2.6: Dip-oriented cross-section of study area with labelled measured sections from Figure 1.4. Fluvial to marginal-marine channels in the northeast can have metres-scale dimensions and are thus simplified in this sketch. Deltaic mouth bars pinch out to the southwest and are often onlapped and overlain by tidal ridges. Datum chosen is the base of the Morillo Limesone which is an easily recognizable basin-wide flooding surface present in all areas except the far northeast. For sections lower in the succession, other well developed flooding surfaces are utilized for correlation; for 19 and 15 the Plateau Limestone is used, and for section 18 the El Villar Limestone is used. Subcropped formation is indicated by lack of color.

The outcrop exposure of alluvial-plain and channelized facies is limited to the northeast corner of the study area. Paleocurrent measurements in that area recorded in this study, as well as those of Lopez-Blanco et al. (2003), indicate that the channels flowed predominantly to the south, with some variation in flow to the south-southwest and south-southeast. In a typical lobate delta, a southerly directed fluvial source would produce deltaic mouth bars with a 180° spread of dips ranging from west to south and east. However, in the study area mouth bars have a preserved dip range from west to south, indicating the outcrop cross-cuts the western half of a south-directed delta complex.

Tidal ridges are present at six different levels within the Roda Formation, three in the

Roda Member, and three in the Esdolomada Member. The stratigraphic relationship between the tidal ridges and deltaic mouth bars within the Roda Member can be locally ambiguous, because it can be difficult to determine whether the ridges are interbedded with the progradational mouth bar as inferred by Tinterri et al. (2007), or whether they are transgressive and lie above the underlying delta. However, there are several locations within the Roda

Formation where ridges erosively overlie delta sandstone, and are capped by shelfal mudstone making them clearly transgressive. The most compelling observations come from the

Esdolomada Member and show that the uppermost two ridges erosively overlie the underlying deltaic parasequence (Fig. 2.6), placing them confidently in the transgressive systems tract.

3rd Order Cycles

The Roda and Esdolomada members encompass two third-order relative sea-level fluctuations interpreted as sequences. In this study the Genetic-Sequence model (Frazier, 1974;

Galloway, 1989) is utilized for several reasons. First, the subaerial unconformity marking the typical sequence boundary for Depositional-Sequence models (see Catuneanu et al., 2009 and

- 23 -

references therein for sequence-model discussion) is difficult to distinguish, whereas the maximum flooding surface is easy to identify. Furthermore, each sequence as defined by the

Genetic-Sequence model contains very different deposits. Sequence 1 is typified by Gilbert- style deltas and an abundance of sand, whereas Sequence 2 contains compound cross-bedded deltas and an abundance of mud. Since the sequences can be confidently classified as genetically distinct, and the subaerial unconformity is not visible, the Genetic-Sequence model is ideal for this study. As defined in this way, Sequence 1 contains all of the Roda Member and the basal part of the Esdolomada Member, whereas Sequence 2 is equivalent to the remainder of the Esdolomada Member.

Each sequence is composed of multiple sets of parasequences with defined stacking patterns including: 1) basinward-stepping parasequences (parasequences 1-8 and 13-14 in Fig.

2.6), 2) aggrading parasequences (9-10 and 15-16) and 3) back-stepping parasequences (11-12 and 17-18). Each genetically related set represents a systems tract in a sequence (i.e., the highstand (HST), lowstand (LST), and transgressive systems tracts (TST), respectively). All three Sequence 2 parasequence sets have a more aggradational component than those in

Sequence 1, whereas Sequence 1 contains a single parasequence that might have a downward trajectory (Parasequence 10). A downward trajectory of the shoreline is important because it implies a fall in relative sea level, and could, therefore, suggest the existence of lowstand deposits in a more basinward location.

Sequences often record complete base-level cycles, but in rapidly subsiding basins like the thrust-top Tremp-Graus Basin, subsidence can outpace eustatic sea-level fall and the falling- stage systems tract (FSST) is not always developed (Posamentier et al., 1988; Mitchum and

Van Wagoner, 1991; Posamentier and Allen, 1999). Nio and Yang (1991) estimated subsidence rates of the Roda Formation to be much higher than rates of eustatic sea-level fall,

- 24 -

estimating rates of 100 m/Myr. With recent dating and updated stratigraphy, the Roda

Formation is found to be typically 300 m thick, and with deposition occurring over 1 Myr; thus, it would have an average subsidence rate closer to 300 m/Myr. Subsidence rates are higher than reconstructed rates of eustatic fall, which are thought to have been 34-100 m/Myr (Haq et al., 1987; Miller et al., 2005).

Rather than representing cycles where relative sea level rose and then fell, the third- order sequences represent variations in the rate of relative sea-level rise. During times when the rate slowed, progradational parasequences of the HST were formed. When the rate accelerated, an aggradational LST parasequence set was developed, followed by a backstepping TST parasequence set. As a result, no falling-stage deposits are documented. In most situations subsidence is not uniform across the basin, and sequences without falling-stage deposits can correlate laterally with sequences that contain such deposits. This implies relative sea-level falls may have occurred in some parts of the basin, but not in the study area due to rapid local subsidence close to the thrust front.

2.4 A Comparison between the Upper and Lower Sequences

Sequence 2 contains a coarser sand component than Sequence 1 and a lower bioturbation index, on average. In addition, there is an obvious progradation of proximal

Sequence 2 alluvial-plain and channel facies over Sequence 1 subaqueous delta mouth bars

(Fig. 2.6). These observations suggest an overall more proximal depositional realm for

Sequence 2.

Sequence 2 is typically 15-30% thicker than Sequence 1. The interpretation for the increase in thickness is that the rate of accommodation creation was higher during the deposition of Sequence 2. Higher subsidence rates across the basin should decrease the amount

- 25 -

of coarse-grained sediment making it to the basin as more sediment is trapped in the alluvial plain. The more extensive alluvial plain, and the lower proportion of deltaic sandstone relative to mudstone in Sequence 2 correlates with this interpretation. The rate of sediment supply also must have increased into the second sequence, because even though the basin was subsiding much faster than in Sequence 1, parasequences still prograded just as far into the basin during

Sequence 2.

An alternative interpretation for the cause of the increase in thickness, considered unlikely, is that Sequence 2 accumulated over a longer period of time than Sequence 1. This is considered improbable because Sequence 2 appears to contain fewer parasequences than

Sequence 1, which is the opposite to what would be expected if each parasequence represents approximately the same length of time (i.e., perhaps controlled by Milankovitch cyclicity). If, however, parasequences do represent longer periods of time in Sequence 1, and perhaps twice as long because there are half as many parasequences in Sequence 1, then there might be more extensive progradation of each parasequence in Sequence 2 than in Sequence 1. However, this study’s observations show that the upper sequence is predominantly aggradational with much less progradation than is seen in Sequence 1. Thus, it is unlikely that parasequences represent twice as much time during the Esdolomada Member and reinforces the hypothesis that accommodation creation was actually greater.

Mechanisms for the increase in the rate of accommodation creation include eustatic sea- level rise and a combination of tectonic and compaction-related subsidence. Eustatic sea-level curves for the time of deposition (Fig. 2.7) are inconsistent: some reconstructions depict a change from rise to a fall around 52.5 Ma, correlating to the middle of the Roda Formation, while others show a continuous fall.

- 26 -

Figure 2.7: Eustatic sea-level reconstructions from multiple sources (after Müller et al. (2008) and references therein) with the proposed age ranges for the Roda Formation (Lopez-Blanco et al., 2003; Torricelli et al., 2006) show a slowly decreasing eustatic sea level.

None of the reconstructions project a significant increase in eustatic sea-level from 55-50 ma so a eustatic rise is unlikely to have contributed to the increase in accommodation creation. Thus, the increase in accommodation is probably related to increased subsidence. However, the eustatic sea-level fluctuations (proposed by Haq et al., 1987; Miller et al., 2005) do indicate the presence of ~500 k.y. cycles which may coincide with Sequence 1 and 2 periods of regression.

Migration of Depocentre

There are two potential explanations for the smaller number of parasequences in

Sequence 2: 1) the upper cycle accumulated in a shorter period of time than the lower cycle; and 2) the depocentre migrated laterally away from the location of the study area, such that not all parasequences present in Sequence 2 are discernible in the outcrop belt.

If each parasequence represents approximately the same time interval, then the sedimentation rate must have been higher during Sequence 2 as there are fewer parasequences but a thicker succession. An increase in sequence thickness of 15% over a shorter time period

(50% shorter than for Sequence 1, based on the presence of 6, as opposed to 12, parasequences, each of which are assumed to represent the same time interval) would require sedimentation

- 27 -

rates to be > 2 times higher during Sequence 2. Such an increase of clastic sedimentation rate would likely maintain the river-dominated Gilbert-Delta architecture within Sequence 2, since river influence would easily overpower basin processes. This prediction is not born out by this study’s observations. Though Sequence 2 contains two Gilbert-style deltas comprised of Facies

C i (parasequences 13 and 17), four are the lower-angle type (Facies C ii), indicating that the ratio of river to basin energies was lower.

The lower number of parasequences in Sequence 2 is most likely caused by the migration of the delta’s depocentre a few kilometres toward the east, so that the Esdolomada outcrop only cuts the western edge of that complex. East is the likeliest direction for depocentre migration because the deltaic deposits thicken in that direction before subcropping along the Isabena valley wall (Lopez-Blanco et al., 2003). There may be additional sandy parasequences to the east, but they are composed of mudstone in the study area and are difficult to differentiate. Similar lateral changes have been documented by Mitchum and Van Wagoner

(1991) in the upper Wilcox and Queen City formations. These authors found that as distance from a river mouth increased, sandy parasequences graded into indistinguishable prodeltaic muds. In the Roda Formation, lateral relationships are limited by outcrop constraints, but changes can be inferred with distance from the mouth of the delta by looking at basinward changes in the parasequences. The tips of the sandy tongues are thoroughly bioturbated, and are overlain by a well-cemented carbonate grainstone extending basinward of the most distal sandstones, further correlating with cyclic deposits of silts and fossiliferous material within distal shelf muds (Fig. 2.4 A). If grain size decreases along strike in the same way that it does in the dip direction, it is probable that additional parasequences may be present but difficult to differentiate, each parasequence being represented by a single carbonate grainstone or shallow shelf sandstone in an otherwise continuous shale succession.

- 28 -

2.5 Conclusion

The Roda Formation is composed of at least 18 parasequences, each of which is composed of alluvial-plain and channel deposits (Facies A and B) that grade into coarsening- upward, deltaic mouth-bar deposits (Facies C) that are capped by a flooding surface and mantled by shallow-marine carbonates (Facies E). Thick tidal sand ridges (Facies D) are present locally on the distal tips of parasequences. These parasequences grade laterally into shallow shelf mudstones and sandstones (Facies F and G).

Stacked parasequences that share a trajectory (progradational, aggradational or backstepping) are grouped into sets that represent systems tracts (highstand, lowstand and transgressive, respectively). The Roda Formation is composed of two third-order sequences that lack falling-stage deposits, because the high rate of subsidence outpaced eustatic sea-level fall.

The lower sequence, Sequence 1, had a depocentre located approximately within the study area. It is characterized by abundant sandy deltaic mouth bars with thin intervening shelf mudstones and rare shelf sandstones. Due to the migration of the deltaic depocentre a few kilometres to the east and increased subsidence, the upper sequence, Sequence 2, contains a higher percentage of mudstone, more shelf sandstones and fewer outcropping parasequences than Sequence 1. It is probable that additional parasequences exist within the upper sequence, but basinward of the alluvial channels and flood-plain deposits they are composed primarily of recessive mudstone in the outcrop belt, making the identification of flooding surfaces difficult.

High rates of subsidence caused by tectonic loading and sediment compaction likely caused a relative sea-level rise for the entirety of the Roda Formation. During the deposition of the Roda Member, the deltaic depocentre was located in the study area, so the sedimentation rate was locally very high. The lower rate of subsidence allowed extensive progradation of the

- 29 -

deltas. Gilbert-type architecture is typical within the Roda Member because an abundance of coarse-grained sediment was being delivered to the coast. The Esdolomada Member marks a time of higher subsidence, so coarse-grained sediment was predominantly trapped in the alluvial plain, making it difficult for Gilbert-style deltas to form.

In some parasequences, after the delta lobe was abandoned, its tip was reworked to create a transgressive tidal ridge (see parasequences 7-11, 16, and 17 in Fig. 2.6). In the Roda

Member, another delta lobe would rapidly prograde and bury the tidal ridge. This contrasts the stratigraphy of ridges in the Esdolomada Member, where there was a longer period of time between mouth-bar progradation, ridges became moribund and were draped by marine carbonates.

Without studying the Esdolomada Member or the detailed ichnological assemblages the task of differentiating whether ridges were deposited above or below mouth bars is difficult. In the Roda Member, successive delta lobes are often stacked with tidal ridges wedged between them making it hard to determine whether the ridges are related to deltaic progradation or the ensuing transgression. The three Esdolomada ridges, on the other hand, overlie and incise into underlying deltas that are separated by thick shelf muds (Fig. 2.6). It is clear that the deposition of these three tidal ridges occurred after the flooding of deltaic mouth bars. Unlike relatively unbioturbated, poorly-sorted, coarse mouth bars, the tidal ridges were deposited in marine waters as evidenced by a diverse marine trace-fossil assemblage, marine fossils and well- developed tidal signatures― features that would not be present within a river-dominated delta.

Therefore, the tidal shelf ridges in the Roda Formation represent the local thickening of the deposits formed during the high-frequency transgressions.

- 30 -

CHAPTER 3

FACIES ARCHITECTURE OF THE TRANSGRESSIVE TIDAL RIDGES

3.1 Introduction

In many modern shallow-marine settings, tidal currents are an effective agent of sea- floor reworking, especially during transgressions when the shelf receives little new sediment because of trapping in estuaries. This is especially true in areas around the British Isles and the

East China and Yellow seas. The transgressive tidal ridges produced by this process are being documented in increasing detail as geophysical techniques improve. Many excellent studies have investigated the large-scale architecture, grain-size trends, surface bedforms and flow around such ridges, examples including: the Shambles Bank, southern U.K. (Pingree, 1978;

Pingree and Maddock, 1979; Bastos et al., 2002; Berthot and Pattiaratchi, 2006); Kwinte Bank and Middelkerke banks, southern North Sea (Trentesaux et al., 1999; Vanwesenbeek and

Lankneus, 2000); southwest Florida Inner Shelf (Davis et al., 1993); Milne Bank, Prince

Edward Island, Canada (Shaw et al., 2008); Helwick Sandbank, Southern Wales (Schmitt et al.,

2007); Kaiser Bank, Celtic Sea (Reynaud et al., 1999); Longe de Boyard Sandbank, French

Atlantic coast (Chaumillon et al., 2008); Snouw and Braek banks, Dunkerque area Northern

France (Tessier et al., 1999); Dutch Coast (Van de Meen, 1996); East Bank, North Sea (Davis and Balson, 1991); Norfolk Banks, North Sea (Robinson, 1965; Stride, 1978; McCave and

Langhorne, 1982; Park and Vincent, 2007; Horrillo-Caraballo and Reeve, 2008); and East

China Sea (Yang, 1989; Chough et al., 2002; Jin and Chough, 2002; Chen et al. 2003). Despite this improved database, the detailed nature of the facies comprising such banks remains poorly constrained because of the difficulty in sampling using cores and grab-samples (Reynaud and

Dalrymple, in press).

- 31 -

Although transgressive tidal ridges are abundant in the modern, ancient examples are comparably few [Miocene, offshore northwest Java (Posamentier, 2002)]. This is due to the scarcity of detailed facies information for tidal shelf ridges, as the potential for preservation has been clearly documented (Berné et al., 2002). The facies models for these features are based largely on early work on the tidal shelf seas around Great Britain (Houbolt, 1968; Swift, 1975;

Belderson, 1982; Stride, 1982; Dalrymple, 1992; Dalrymple and Rhodes, 1995). These models predict that the deposits of tidal ridges should be sandy and composed of simple and compound dunes that accumulated on one or both sides of the ridge, generating a lateral-accretion architecture. However, numerous modern studies report a range of internal facies and architectures, some of which are inconsistent with the early models (Pingree, 1978; Davis et al.,

1993; Berné et al., 1999; Reynaud et al., 1999; Jin and Chough, 2002; Schmitt et al., 2007;

Shaw et al., 2008). The increasingly heterogeneous sample group highlights the need for a larger number of detailed descriptions in order to encompass the full range of variability.

To address the need for more examples of tidal shelf ridges from the ancient, detailed documentation of six headland-associated tidal sand ridges within the Eocene Roda Formation is presented here. The Roda Formation has been the subject of numerous sedimentological, palynological and structural-geology studies (Nio, 1976; Puigdefàbregas, 1992; Torricelli,

2006), but all such studies focus on the lower member. In recent studies the primary attention has been directed at the steep-fronted, coarse-grained deltas that are well exposed along the side of the Isabena River valley. However, tidal deposits have long been known to occur within the lower ‘Roda Member’ (Nio, 1976). Since then, numerous studies have documented the presence of tidal sands overlying or inter-fingering with deltaic deposits (Lopez-Blanco et al.,

2003; Tinterri et al., 2007), but none focus on the tidal deposits with the exception of Olariu et al. (in press), which examined one of the tidal bars in detail.

- 32 -

The Roda Formation is composed of two third-order cycles composed of 18 stacked, high-frequency parasequences. Here, we document the presence of tidal ridges in six of these parasequences. By analyzing all of these tidal ridges, our aim is to document: 1) the range of preserved facies within such ridges; 2) the internal architecture of the ridges; and 3) the stratigraphic occurrence of these tidal ridges. Because several ridges are documented and analyzed, it is also possible evaluate the degree and causes of differences between ridges.

3.2 Facies Analysis

Within the Roda Formation, the sandstone bodies that are interpreted to be tidal sand ridges overlie the distal tip of six of the 18 progradational delta lobes sharply (Fig 3.1). These are in turn overlain by marly shales. There are three tidal ridges in the Roda Member with a possible fourth ridge in the southwest where the member subcrops, whereas the Esdolomada

Member contains another three tidal ridges (Fig. 3.1). The stratigraphic position of these sandstone bodies combined with the nature of the bioturbation and evidence of tidal action shows that each occupies a transgressive position overlying a progradational delta tongue. The following section describes the facies that constitute these ridges. Facies are differentiated by grain size, bedding-plane geometry, sedimentary structures, bioturbation, and body fossils

(Table 3.1).

Facies 1

Description: This facies consists of 100-500 cm thick, high-angle cross-beds composed of fine to lower coarse, well-sorted, arkosic sandstone (Table 3.1). Foreset beds are 2-20 cm thick and dip at up to 30o (Fig. 3.2 A). Small sets can stack to form packages up to 500 cm thick. Semi-cyclic thickening and thinning of foresets suggests the presence of crude tidal bundles, but they are poorly developed. Some mudstone drapes are present, concentrated in the

- 33 -

Facies Description Ichnology Interpretation 1 Well sorted fine to lower coarse arkosic Bioturbation intensity ranges from 1-5 but Lee-side deposits of either large arenite composing 1-5 m thick cross- is lower (2) in the foresets and higher in to very large dunes or the steep beds forming packages up to 5m thick. the bottomsets (3-5). Foresets contain: side of an elongate tidal bar Contains crude tidal bundles and rare fugichnia, Rosselia, Cylindrichnus, with strongly asymmetric mudstone drapes, Asterosoma, Planolites, Thalassinoides, currents. Cruziana Ichnofacies Ophiomorpha, Teichichnus, and cryptic indicates normal-marine bioturbation. Bottomsets commonly conditions and a rapid contain the preceding burrows as well as sedimentation rate decreasing less commonly found: Chondrites, away from the bedform. Skolithos, Rusophycus, Arenicolites, Scolicia, Bergaueria, Protichnites, and Macaronichnus.

2 Well sorted upper fine to medium, and Bioturbation intensity of 1-5 with lower Formed by dunes migrating up rarely lower coarse angular arkosic values in foresets (1-3). Contains the stoss side of compound sandstone compose 10-300 cm thick abundant Rosselia, Asterosoma, dunes or tidal ridges. Bedform cross-bed sets dipping mostly up (but at Thalassinoides, Planolites, Palaeophycus, migration was slower than a slight angle to) master-bedding Polykladichnus, Rusophycus, Cruziana, Facies 1. Currents were tidal in surfaces. Contains cyclic bundle sets, Ophiomorpha, Teichichnus, Asteriacites, origin, and the Cruziana double mud drapes, and cyclic Chondrites, and cryptic bioturbation. Ichnofacies indicates normal- mudstone/sandstone alternations. Total marine conditions. Dune thickness of this facies reaches 20 m. height indicates a potential water depth of 12-18 m.

3 Well sorted upper fine to medium- Bioturbation intensity ranges from 0-5 Formed by dunes migrating grained angular sandstone in cross-bed with no difference between foresets and down the lee face of a sets dipping in the same direction as bottomsets. Contains Rosselia, compound dune or tidal ridge. master-bedding surfaces. Contains Arenicolites, Teichichnus, Scolicia, Currents were tidal in origin. herring-bone cross-stratification, tidal Planolites, Skolithos, Cruziana, Higher stress than Facies 1 and bundles, mud laminations, and grain- Ophiomorpha, Thalassinoides, 2 possibly related to higher size striping. Psammichnites, and cryptic bioturbation. suspended-sediment concentration and/or lower salinity due to river influence.

4 Poorly sorted, angular, fine to medium- Bioturbation intensity ranges from 3-6. Formed by current ripples grained sandstone compose 1-3 cm Contains Thalassinoides, Planolites, migrating in the troughs of thick cross-laminated sets stacking to Phycosiphon, Rusophycus, Asteroides, compound dunes or tidal form packages up to several meters in Chondrites, Teichichnus, Skolithos, ridges. Current speeds slower thickness. Interbedded with Facies 5. Psammichnites, Palaeophycus, and cryptic than Facies 1-3, but still tidal in bioturbation. origin.

5 Well sorted, planar laminated, fine Bioturbation intensity ranges from 2-4. Muddy sediment that grained grey siltstones and mudstones Contains orange-red oxidized accumulated in the troughs of forming packages from 10-300 cm Thalassinoides, Rosselia, and Asterosoma. dunes and tidal sand ridges. thick. Contains flame structures, Shell lags within thicker mudstone-sandstone rhythmites, paired accumulations indicate mud drapes, and carbonate lags prolonged periods of weak composed of bivalves, gastropods and currents punctuated by periods large benthic foraminifera. of accelerated currents.

Table 3.1: Tidal ridge facies (1-5) within the Roda Formation.

- 34 -

Figure 3.1: Dip-oriented cross-section of the Roda Formation showing the location of Ridges A-F in the Roda Formation. Sequence-stratigraphic subdivision of the succession is shown in the right-hand column. Flooding surfaces highlight parasequence tops. The tidal ridges overlie flooding surfaces and formed during high-frequency transgressions.

- 35 -

toesets and bottomsets of the cross-beds, but mudstone layers are thin (a few mm) and sporadic in their occurrence. In the foresets bioturbation consists predominantly of fugichnia, Rosselia,

Cylindrichnus, Asterosoma, Planolites, Thalassinoides, Ophiomorpha, Teichichnus, and cryptic bioturbation (Fig. 3.3). Bottomsets commonly contain the preceding list of burrows, but also contain the less commonly found trace fossils Chondrites, Skolithos, Rusophycus, Arenicolites,

Scolicia, Bergaueria, Protichnites, and Macaronichnus. The bioturbation index ranges from 1-

5, but is lower (BI of 2) within the foresets, with higher bioturbation indices in the toesets and bottomsets of the cross-bed sets. Skolithos and Rosselia burrows reach lengths of up to 2 m.

Interpretation: These high-angle cross-beds were formed by avalanching on the lee side of either large to very large dunes or the steep side of an elongate tidal bar with strongly asymmetric tidal currents. The paucity of tidal sedimentary structures in Facies 1 appears to be a result of bedform size, as Facies 1 bedforms are large and would have migrated very slowly.

As a result, the tidal bundles that could be present in Facies 1 would be very thin, and difficult to recognize. Tidal evidence is further exemplified in Facies 2 and 3, which are formed by smaller dunes and preserve distinct tidal bundles.

The diverse suite of trace fossils belonging to the Cruziana Ichnofacies indicates the presence of normal-marine conditions, with a rapid sedimentation rate as indicated by the low bioturbation index and angle-of-repose foresets.

Two ichnocoenoses are present. The primary suite (consisting of fugichnia, Rosselia,

Cylindrichnus, Asterosoma, Planolites, Thalassinoides, Ophiomorpha, Teichichnus and cryptic bioturbation) occurs in the foresets (Fig. 3.4 B) and represents behaviours that are adapted to high depositional rates (MacEachern et al., 2005c), and has an inherently low BI of 1-3. The second ichnocoenose (consisting of Planolites, Chondrites, Skolithos, Rusophycus, Arenicolites,

Scolicia, Protichnites and Bergaueria) occurs in the toeset and bottomset strata, and reflects

- 36 -

F Figure 3.2: Sandy ridge facies with 1.5 m stick marked at 10 cm intervals (circled) for scale in A-C. (A) Facies 1 composed of large, angle-of-repose foresets. (B) Facies 2 composed of cross-beds migrating obliquely up gently inclined, erosive set boundaries. (C) Facies 3 composed of cross-beds migrating obliquely down gently inclined, erosive set boundaries. Facies 1 and 3 indicate deposition on a lee/down-current side while Facies 2 indicates deposition on the stoss/upcurrent side of the tidal ridge. (D) Coarsening-upward packages of ripple cross-lamination (Facies 4) passing upward into increasingly cross-bedded sands (Facies 3). (E) Facies 4 sandstones are composed of bioturbated current ripples dipping obliquely down shallowly dipping master bedding surfaces.

- 37 -

slower rates of sedimentation as evidenced by a higher level of bioturbation (BI 3-5). In the second ichnocoenose a lower sedimentation rate and current speed likely allowed a more diverse assemblage of organisms to occupy this region.

Facies 2

Description: This facies consists of upper fine to medium, and rarely lower coarse, well sorted angular arkosic sandstone in which 10-300 cm thick cross-bed sets climb mostly up (but at a slight angle to) gently inclined (1-10o) master-bedding surfaces (Fig. 3.2B; Table 3.1).

Individual foresets are 1-2 cm thick, 50-300 cm long, and dip at 30-35o. The master-bedding surfaces bounding cross-beds (erosional set boundaries) can be traced for over 100 m up-dip.

Cross-beds indicate migration along and slightly up these surfaces. Cyclic bundle sets, double mud drapes, and cyclic mudstone/sandstone alternations are present (Fig. 3.4 A). Packages of this facies can reach 20 m in thickness.

The trace-fossil suite includes abundant Rosselia, Asterosoma, Thalassinoides,

Planolites, Palaeophycus, Polykladichnus, Rusophycus, Cruziana, Ophiomorpha, Teichichnus,

Asteriacites, Chondrites, and cryptobioturbation (Fig. 3.3 and Fig. 3.4 B). Rusophycus,

Thalassinoides, and Palaeophycus are dominant on the foresets and at the erosive set boundaries between cross-beds, whereas Rosselia is the dominant vertical trace, reaching lengths exceeding one metre, cross cutting up to 3 or 4 cross-bed sets. Bioturbation intensity partly obscures bedding along set-boundaries (BI 1-5), whereas foresets are less intensely bioturbated (BI 1-3).

Interpretation: This facies was formed by dunes that migrated obliquely up an inclined surface, forming upstream-accreting architectural elements. The inclined master bedding

- 38 -

planes can represent the stoss side of either compound dunes or tidal bars. Deposition was rapid, but the bedforms migrated less rapidly than those that formed Facies 1 as indicated by the higher degree of bioturbation. As in Facies 1, the ichnogenera present within Facies 2 constitute 2 ichnocoenose: an archetypal Cruziana Ichnofacies that occurs in the toeset and bottomset deposits, and an impoverished, high sedimentation rate Cruziana Ichnofacies that occurs in the foresets. Asteriacites (a five armed echinoderm trace) suggests marine conditions because most echinoderms cannot handle reduced (i.e., brackish) salinity (Turner and Meyer,

1980).

Well-developed tidal bundling and cyclic mud draping indicate that tidal currents were responsible for deposition. The bioturbation rate was highest when sedimentation rate was at its lowest, as evidenced by relatively unbioturbated spring-tide sandy tidal bundles and intensely bioturbated neap-spring muddier tidal bundles (Fig. 3.4 B).

If the Leclair and Bridge (2001) approach to depth reconstruction is used, cross-bed sets are, supposedly, approximately 1/3 of the height of their ancestral dune. Average cross-bed thickness of approximately 2 meters indicates that dunes ~6 m in height formed the deposits.

Water depth is typically 6-10 times the height of subaqueous dunes (Bridge and Tye, 2000;

Leclair and Bridge, 2001), which would indicate a plausible water depth of 36-60 meters. This analysis likely overestimates depths though, as dune climbing is obviously occurring so that the preserved sets are probably much closer to actual dune height. If the thickest cross-bed sets (2-

3 m) are taken as being nearly the height of the dune that formed them, then water depth would be closer to 12-18 m.

- 39 -

Figure 3.3: Trace fossils common to the sandy portion of tidal ridges (Facies 1-4). (A and C) Rosselia (ro) and Asterosoma (as) are very abundant and well preserved. (B) Echinoid traces Scolicia (sc) is abundant on bedding planes and can reach BI6 in some locations). (C-D) Asterosoma (as), Arenicolites (ar), Planolites (pl), and Ophiomorpha (oph) are also quite common. (E-F) Large Ophiomorpha (oph), Palaeophycus (pa), Chondrites (ch), Asteroides (at), and Planolites (pl) are also commonly found. (A-E) are taken from cross- bedded Facies 1-3 whereas (F) is from current-rippled Facies 5. The diverse suite of trace fossils indicates the sandstones were deposited in a fully marine setting.

- 40 -

Figure 3.4: A) Photo of Facies 2 in Ridge C: Spring (s) and neap (n) deposition is indicated by cyclic sandstone-mudstone development. B) Intense and diverse bioturbation occurs within muddy sandstones, less bioturbation within sandstones (Rosselia (ro), Ophiomorpha (oph), Asterosoma (as), Skolithos, (sk), Chondrites (ch) and crytobioturbation (cry)). C) Taken from Ridge F Facies 3 and 4: Cyclic mud drapes and current ripples migrating back up dune foresets. D) Herringbone cross-stratification indicates flow reversal in Ridge F.

- 41 -

Facies 3

Description: Facies 3 is composed of compound cross-bedded upper fine to medium- grained sandstone that is well sorted and angular (Table 3.1). Cross-bed foresets dip in the same direction as their set boundaries that themselves have dips of 1-10o (Fig. 3.2 C), thus defining forward-accretion architectural elements. Individual foresets range in height from 10-

100 cm and typically dip at 30 o. Dip orientation is generally unidirectional, but occasional reversals exist in the form of ripple cross-laminations dipping in the opposite direction to the adjacent cross-beds. The total thickness of this facies is 2-10 m. Mud laminations are present on foresets and in bottomsets, and grain-size striping was observed in some locations. Tidal bundles are present. The trace-fossil suite is uniform with no difference between foresets and bottomsets, the common traces being Rosselia, Arenicolites, Teichichnus, Scolicia, Planolites,

Skolithos, Cruziana, Ophiomorpha, Thalassinoides, Psammichnites, and cryptobioturbation

(Fig. 3.3). Bioturbation index ranges from 0 to 5.

Interpretation: These deposits were formed by simple subtidal dunes that migrated down the lee face of a compound dune or tidal ridge. Tidal bundles and mud drapes (Fig. 3.4) indicate that tidal currents were the predominant sediment-transporting agent. Grain size is typically markedly finer than the mouth-bar deposits that can show a similar architecture.

While the architecture is similar to mouth-bar deposits, the trace fossils are contrastingly more marine in origin. The trace-fossil suite is similar to Facies 1 and 2

(proximal Cruziana Ichnofacies), but is slightly less diverse, indicating the presence of a stress.

The presence of volumetrically more mud than Facies 1 suggests that this stress might be related to a higher suspended-sediment concentration and/or lower salinity due to river influence. The single ichnocoenose differs from Facies 1 and 2 that had high-energy foresets and low-energy bottomsets. It appears that Facies 3 had relatively equal stresses on the foresets

- 42 -

and bottomsets of the dunes. Cross-beds are found on both low-angle master-bedding surfaces and on steeper reactivation surfaces. This indicates that simple dunes were active on the compound dunes and in the adjacent troughs.

Facies 4

Description: Facies 4 is composed of 1-3 cm thick cross-laminated sets of poorly sorted, angular, fine- to medium-grained sandstone (Fig. 3.3 D; 3.4 D; Table 3.1). Facies 4 has a higher mudstone percentage than Facies 1-3 and weathers recessively. While an individual set is 1-2 cm thick, multiple sets can stack to form cosets that reach several metres in thickness.

Facies 4 occurs between occurrences of Facies 1-3 and is interbedded with the muddy Facies 5 in the lower part of the ridges or compound dunes.

The bioturbation signature includes Thalassinoides, Planolites, Phycosiphon,

Rusophycus, Asteroides, Chondrites, Teichichnus, Skolithos, Psammichnites, Palaeophycus, and cryptobioturbation (Fig. 3.3 F). The bioturbation index ranges from 3-6.

Interpretation: These deposits were formed by current ripples that migrated in the troughs of compound dunes or tidal ridges. The finer grain size than Facies 1-3, combined with the presence of ripples instead of dunes, indicates current speeds were slower. The trace-fossil suite is similar to the sandy facies; however, the bioturbation index is typically higher, often partially to completely obliterating physical sedimentary structures (BI 4-6).

Facies 5

Description: Facies 5 is composed of well sorted, planar-laminated, fine grained grey siltstones and mudstones (Table 3.1). The thickness of Facies 5 ranges from 10 cm to about 2-3 m. Due to poor exposure, the structures within Facies 5 are difficult to determine. However,

- 43 -

flame structures, centimetre-thick mudstone–sandstone rhythmites and paired mud drapes are visible in some heterolithic occurrences. Carbonate lags composed of Arcidae bivalves, various

Turritella gastropods and large benthic foraminifera are common within the muds and are typically a few centimetres thick (Fig. 3.5 D). Facies 2 cross-beds can often be traced downdip into Facies 4 and then into the increasingly muddy Facies 5 (Fig. 3.5); elsewhere, Facies 2 erodes into Facies 5. Where the contact is gradational, mud percentage typically increases outward from the toe of Facies 2 cross-beds from ca. 10% to100% over a distance of approximately 10 m.

Figure 3.5: Muddy Facies 5. Within sandy Facies 1-3, muds can be quite thin with sparse bioturbation (A). Moving downward into the underlying mudstones (from A through to D), muddy layers between the sandstones thicken (B) and bioturbation intensity increases, although low-diversity, oxidized sand-filled tubular burrows are abundant. Below sandy Facies 1-3, mudstones are thick (C) and carbonate lags composed of large benthic foraminifera, gastropods and shells from the family Arcidae are found locally (D).

- 44 -

The bioturbation signature includes Thalassinoides, Rosselia, and Asterosoma, all of which are oxidized orange-red in the dark grey-blue shale. Burrows are difficult to discern, as is bioturbation intensity. The degree of continuity of horizontal shelly beds indicates the bioturbation index is within the range of 2-4.

Interpretation: These muddy sediments accumulated in the troughs of large simple and compound dunes and tidal sand ridges. Thinner occurrences (2-50 cm thick) appear to be associated with the troughs of simple and compound dunes, whereas mudstones several metres in thickness were deposited alongside a larger tidal sand ridge. Shell lags within these thicker ridge-associated mudstones indicate periods of colonization by marine bivalves, punctuated by periods of accelerated currents which eroded the muds and concentrated the shells.

3.3 Facies Distributions, Isopach Patterns and Paleocurrents

All of the tidal sandstone bodies have a large-scale lenticular geometry. Isopach maps of these bodies were created by mapping measured thicknesses along outcrop exposures and extrapolating using paleocurrents and master-bedding trends (Fig. 3.6). Dimensions are best constrained to the northwest. The southeasterly extent is less well known because the deposits subcrop. Outcrop orientation is nearly perpendicular to ridge length, allowing the width of the bodies to be determined with reasonable confidence, but the lengths are less certain. Thus, we estimated total ridge length and southeasterly extent using the typical dimensions and length/width ratios of offshore ridges reported by Off (1963; see also Wood, 2003).

Isopach data from the 6 tidal sandstone bodies indicate maximum thicknesses of 25 m and widths up to 2 km, which implies that lengths are a minimum of 8-10 km (Figure 3.6).

Ridges B, D, E and F are noticeably asymmetric with the steep side facing southwest (i.e., offshore). The other ridges either subcrop or are covered by vegetation, such that the degree of

- 45 -

asymmetry cannot be determined accurately. The crests of all ridges are oriented northwest- southeast, approximately at right angles to the progradation direction of the delta lobes.

Paleocurrent directions, as measured using dune cross-bedding, are typically nearly parallel to the crest.

Figure 3.6: Interpreted isopach maps (thickness in metres) with paleocurrent rose diagrams for tidal ridges A-F (see Figure 3.1 for stratigraphic location of ridges): A-C: ridges in the Roda Member; D-F: ridges in the Esdolomada Member. Grey and colored portions indicate where the member outcrops and circles represent measurement locations. Crests are typically oriented NW-SE with preserved paleocurrents towards the northwest. Deltaic deposits which lie to the northeast of these ridges have relatively uniform paleocurrents towards the southwest making them distinct from the ridges. Ridge lengths are drawn to fall within typical dimensions of tidal ridges (Off, 1963; Wood, 2003), but the southeasterly extent is unconstrained because the ridges subcrop.

- 46 -

Ridges A-C in Sequence 1 contain equal proportions of Facies 1-3, whereas Ridges D-F in Sequence 2 have sandy components composed of Facies 2-4 and an elevated proportion of

Facies 4 (Fig. 3.7).

In order to illustrate the internal organization of facies within the sandstone bodies, a particularly well-exposed example (ridge B; Fig. 3.1) is described in detail. It contains the same facies as Ridges A and C, but is different from Ridges D-F which lack Facies 1 and contain an increased percentage of Facies 4. Ridge F is then described in order to document the variability between Sequence 1 Ridges A-C and Sequence 2 ridges D-F.

Figure 3.7: Internal composition by facies of Ridges A-F. Bioturbation index is indicated by numbers 0-6 within each column. Facies 1 is only present in Ridges A-C, whereas elevated percentages of Facies 4 and 5 are found in Ridges D-F.

- 47 -

3.4 Ridge B

Three genetically distinct units can be recognized within Ridge B (Fig. 3.8 A). They are differentiated on the basis of facies, paleocurrent orientation and the dip direction of the master bedding surfaces.

Unit 1

Description: Unit 1 constitutes the nucleus of the ridge and is composed of Facies 1-5 that show master-bedding dips to the north-northeast (Fig. 3.8 A, B and C). The base is sharp to the north and truncates delta-front sandstones, whereas to the south it is increasingly gradational, with the mudstone percentage decreasing upward rapidly, over a vertical distance of 2 m. The typical succession is as follows: 0-1 m of Facies 5 mudstones that grade upward into 1-2 m of Facies 4 rippled sandstone, which is then overlain by 10-20 m of Facies 1-3 sandstones. Unit 1 thickens northward toward the ridge crest where it reaches 20 m in thickness. Facies 1 is volumetrically dominant, followed by Facies 3 with subordinate amounts of Facies 2, 4, and 5.

Facies 1 deposits comprise approximately 50% of Unit 1. Large 3-10 m-high sandy foresets dip to the north, commonly at 30o. The size of the large foresets decreases to the north, transverse to paleocurrent interpreted from the dip direction of Facies 2 and 3 cross beds.

Reactivation surfaces are present within Facies 1 and are mantled by Facies 2 or 3 deposits.

Facies 3 deposits comprise approximately 30% of Unit 1. Cross-bed sets range in thickness from 10-200 cm and lie on north-dipping surfaces, which are inclined at 1-15 o. Facies 2 deposits comprise approximately 10% of Unit 1. Small 10-50 cm cross-beds climb obliquely up steep 10-15 o master-bedding surfaces. Facies 2 is overlain by Facies 3 and is underlain by either Facies 1 or 3 (Fig. 3.8 C).

- 48 -

- 49 -

Figure 3.8 Previous Page: (A) Oblique cross-section of Ridge B with observed (black) and inferred (gray) master-bedding surfaces (see Figure 3.1 for stratigraphic location). Outcrop photographs are overlain on the sketch in A. Zones without photographs are either at an angle to the section or are vegetated. (B) Facies distribution within the ridge and detailed outcrop sketches. (C, D) Panoramic photos with vertical logs (measured on the left edge of photos) and scales on left to illustrate grain-size trends within the ridge. The photographs and logs can be correlated with the facies illustrated in (A). See Appendix A.2 for untraced versions of these photos.

Interpretation: NW-SE tidal currents flowed nearly parallel to a WNW-ESE sand-ridge crest. The dominant northwest-directed current preferentially flowed along and obliquely up the southern edge of the bar, while subordinate southeast-directed currents were directed along and obliquely up the northern flank (Fig. 3.6 B). During the dominant tidal-current phase, sediment was eroded from the gently dipping stoss side of the bar, moved over the crest and deposited on the steeply dipping lee side as Facies 1 and 3. The subordinate current tidal phase reworked sand on the northern face of the bar into southeast migrating dunes (Facies 2), which are periodically preserved on the steep northern side. Variability in the preservation of Facies 2 on reactivation surfaces may reflect short-term, yearly to decadal, fluctuations in current speed.

Unit 2

Description: Unit 2 is found on the southern flank of the ridge and is composed of

Facies 2, 3, 4 and 5 with master-bedding planes that dip consistently toward the southwest (Fig.

3.8). Like Unit 1 it thickens toward WNW-ESE oriented ridge crest where it reaches 25-30 m in thickness. The base of Unit 2 is marked by 1-4 m of mudstone with content of siltstone and sandstone (reaching upper-medium grain size) increasing upward. Immediately above the first sandstone, bioturbated sandstones interfinger with centimetre-thick mudstones which thin and decrease in abundance upward. The sandy component appears to be relatively uniform medium-grained sand. The internal architecture is dominated by Facies 2. Unlike the rare occurrences of Facies 2 in Unit 1 in which the cross beds were relatively small (10-50 cm thick

- 50 -

sets), the sets in Facies 2 in Unit 2 are 2-3 m thick and make up at least 90% of the unit. Two orders of master-bedding surface are present within Unit 2. Large master-bedding surfaces dipping 1-3 o to the south-southwest bound 1-3 m-thick packages within which smaller-scale master-bedding dips either to the northwest or the southeast. The cross-bedding within these smaller packages dips to the northwest at 30 o.

Interpretation: Tidal currents maintained the same orientation as in Unit 1 with a dominant current directed toward the northwest. Instead of sediment avalanching down a north-dipping lee face, it accumulated on the obliquely up-current (south) side of the ridge. The thickness of the cross-beds along the crest of the ridge indicates a possible water depth of 12-18 m (see description and interpretation of Facies 2). The change in the ridge’s accretion direction, from northward in Unit 1 to southward in Unit 2, is attributed to the ridge reaching a height (~25-30 m) where the ridge began to impede tidal currents. This caused deceleration of the dominant current as it flowed up the southern ridge flank. Compound dunes migrating northwestward along the southern flank of the ridge deposited sediment, and the troughs of the compound dunes formed the large-scale master-bedding planes which represent the general dip of the southern flank of the ridge.

Unit 3

Description: Unit 3 is a relatively thin, 2-4 m-thick, sandy deposit composed of Facies

3. It overlies the northern flank of Unit 1 (Fig. 3.8). The base of Unit 3 is sharp, eroding into

Unit 2. Mudstone percentage decreases upward as cross-bed size increases. Cross-beds are 50-

100 cm thick and dip mostly to the northwest, approximately in the same direction as the gently dipping (1-4 o) master-bedding planes within this unit forming a forward-accretion architecture.

- 51 -

Unit 3 is different from Unit 1 in that it is thinner, the erosive set boundaries are more gently dipping, and the unit lacks Facies 1 and 2.

Interpretation: Unit 3 likely represents the deposit of a compound dune migrating along and slightly down the northern margin of a larger ridge. Paleocurrent directions are the same as in Units 1 and 2, with a NW-SE orientation. Small simple dune-scale foresets dipping down the master-bedding planes indicate a downstream accretion direction that is along the ridge axis, and is consistent with the interpretation of a compound dune (Dalrymple and Rhodes,

1995; Reynaud et al., 1999).

Summary of Ridge B Evolution

Ridge B formed after the southwestward progradation of a river-dominated delta ceased and transgression had begun. The cessation of deltaic progradation coincided with the end of river influence, as indicated by the increase in the degree and diversity of bioturbation in the tidal sand ridge, relative to the underlying deltaic unit, and a relative increase in the influence of tidal currents. Tidal currents that impinged on the delta front reworked deltaic sediment into a shore-parallel elongate ridge constituted of compound dunes, simple dunes, and current ripples.

Bedform stability diagrams typically denote simple dunes as occurring within the 50-100 cm/s range, whereas compound dunes documented in the modern typically occur within the 50-200 cm/s range, the increase in velocity likely relating to an increase in water depth (Boguchwal and Southard;1990a, 1990b, 1990c; Dalrymple and Rhodes, 1995). Thus, the northwest- directed current likely reached speeds of 50-200 cm/s, as evidenced by the presence of large compound dunes, whereas the subordinate, southeasterly current reached speeds of 50-100 cm/s as evidenced by the presence of simple dunes.

- 52 -

Initially the ridge grew by lateral accretion to the northeast as dunes migrated up and over the ridge crest and deposited sand along that margin of the ridge. The ridge eventually reached a height where it began to impede cross-ridge currents. At this point, lateral accretion occurred on the southern side, preserving compound dunes climbing at a low angle up the stoss side of the ridge. Some mud was still being deposited in the trough adjacent to the ridge, and the amount of mud within the ridge continued to be less at higher elevations. While the ridge accreted to the southwest it may have also accreted on the northern margin. Unit 3 documents a northwesterly migrating compound dune on the northern flank of the ridge, and may have been coeval with Unit 2

3.5 Ridge F

Description: Ridge F (Fig. 3.9) has a relatively simple internal architecture compared with Ridge B. Master-bedding surfaces dip toward the southwest, while ripple cross-lamination and dune cross-bedding dip toward the WNW and ESE (Fig. 3.6 F); however, there is a thin layer of Facies 4 on the top of the ridge which has relatively flat master-bedding surfaces. The ridge is composed predominantly of rippled Facies 4 (50%) and cross-bedded Facies 3 (35%), with lesser amounts of muddy Facies 5 (10%) and cross-bedded Facies 2 (5%).

Muddy Facies 5 typically occurs at the base of the ridge, which coarsens upward into cross-laminated fine sandstone of Facies 4, and then into 10-100 cm thick cross-bedded sets of

Facies 3. Facies 2 is found locally as metre-scale lenses surrounded by Facies 4. Along its northeastern flank, the ridge erosively overlies deltaic sandstones. Here the ridge typically lacks significant muddy Facies 5 deposits at its base. Further to the south, the ridge overlies deeper-water marls, and mud is more abundant at the toe of the accretional flank.

- 53 -

Interpretation: Ridge F formed after the cessation of south-southwestward progradation of another river-dominated delta. The ridge is interpreted as transgressive because it erodes into deltaic deposits and contains a fully marine trace-fossil assemblage. The interpretation of

Figure 3.9: (A) Oblique cross-section of Ridge F (see Figure 3.1 for stratigraphic location). The black lines in the boxed area were traced from high-resolution photographs, while the light-grey surfaces are based on a mixture of observation and interpolation, due either to vegetative cover or outcrop being at an angle to the section. (B) Panoramic photo with vertical log (measured 5 meters to the left/north of this photo) and scale on left illustrate the grain-size trend within the ridge. See Appendix A.2 for untraced version.

the underlying deposits as regressive coincides with an upward increase in grain-size (to cobbles and small boulders) within distributary channels further to the north.

The ridge is composed dominantly of ripples and small cross-beds with paleocurrents indicating WNW-ESE directed currents. Master-bedding planes indicate that the direction of ridge migration was towards the southwest, approximately perpendicular to the direction of the

- 54 -

currents, forming lateral-accretion deposits. Most of the deposit (all but the small percentage of

Facies 2) consists of cross-beds and cross-laminations dipping along but slightly down master- bedding.

Current ripples are present in sands ranging from fine to upper medium. Dunes are capable of forming within this grain-size range; thus, it is likely that the currents were less than

~40 cm/s (Boguchwal and Southard; 1990a, 1990b, 1990c) along the flanks of the ridge. The occurrence of dunes at the top of the deposit indicates that the currents were fastest along the crest of the ridge. Although water depth can also dictate whether dunes or ripples will form, the presence of dunes higher up on the ridge indicates that water depth was not the limiting factor.

Faster currents on the crest of the ridge show that the ridge had not aggraded to shallow friction-dominated depths (Huthnance, 1982).

3.6 Origin of Tidal Ridges

Irregularities of the coastline or on the sea floor influence the preferential pathway of tidal currents (Pingree, 1978; Williams, 2000), and cause flood and ebb currents to become unequal in strength and duration. Tidal ridges accumulate in the zone between these flood- and ebb-dominant pathways. Time-averaged maximum tidal currents (over a single cycle or longer) move in opposite directions on either side of the ridge, forming a residual circulation cell

(Pingree and Maddock, 1979; Huthnance, 1982a.,b; McCave and Langhorne, 1982; Lanckneus et al., 1992; Hulscher et al., 1993; Bastos et al., 2004, Berthot and Pattriaratchi, 2005).

Observations from the Roda Formation indicate that currents moved in opposite directions on either side of the ridge, clearly matching flow processes acting in modern examples.

Transgressive tidal ridges occur in 3 main settings: in the mouth of estuaries, on the open shelf away from shorelines, or adjacent to headlands (Dyer and Huntley, 1999). Tidal

- 55 -

ridges in the Roda Formation are likely headland associated. Such ridges differ from the other ridge types in several ways. First, open-shelf ridges are typically larger than the ridges documented herein which are near the small end of the range of ridges sizes (Wood, 2003).

Open-shelf ridges typically reach several tens of kilometres in length and several kilometres in width, dimensions which are significantly greater than the measured size of the Roda ridges.

Second, shelf ridges typically occur in clusters, but the ridges in the Roda Formation appear to be solitary features, although this might be due to outcrop limitation, in which case, additional ridges might be present further to the south. Similar-sized lone ridges can occur in estuaries; however, the consistent NW-SE orientation of the ridges (Fig. 3.6) is parallel to the paleo- coastline and perpendicular to what would be expected within an estuarine funnel at the mouth of the river that fed the deltas. A fully marine suite of trace fossils would not be expected in estuary-mouth ridges as such locations typically have brackish water. The location of the ridges on the seaward tip of its ancestral delta does not fit with the inner estuarine model or the open shelf model, but it is consistent with a headland origin, the protruding delta acting as the headland (Fig. 3.10).

It is hypothesized that each progradational delta formed a headland. After the delta had prograded a significant distance, tidal currents were accelerated as they impinged on the delta.

Following intermittent periods of deltaic progradation, delta abandonment and transgression allowed tidal currents to rework the deltaic sediment into a tidal ridge (Fig. 3.10). The cross- section (Fig 3.1) shows that not every delta lobe is associated with a tidal ridge. In both sequences, tidal ridges are observed to only occur on those deltas that prograded a minimum of

2 km to the southwest (i.e., basinward) of the first early-highstand parasequence. This means that the delta complex had to prograde approximately 2 kilometres into the basin before it created a protrusion large enough to influence the tidal currents sufficiently. It took several

- 56 -

progradational episodes to reach this distance, so tidal deposits are not found in the initial HST parasequences. Once the headland was large enough, each time a delta was abandoned, tidal currents could rework sediment into the documented ridges A-F.

Figure 3.10: Interpreted paleogeographic setting and flow conditions around the tidal ridges under investigation. The ridges are interpreted to represent headland-associated sand bodies that lay slightly to the west of the underlying deltaic headland because of the general dominance of the ebb (westerly flowing) currents.

The tidal sand ridge at Economy Point within the Minas Basin, Nova Scotia (Klein,

1970), presents a modern analogue to the tidal ridges of the Roda Formation. These modern and ancient ridges have similar orientations, sizes, flow patterns and sedimentary structures

(Fig. 3.11). Indeed, the flow pattern, most sedimentary structures, ridge asymmetry and ridge size match almost perfectly between the two, although it must be noted that the headland at

Economy Point is bedrock rather than a delta lobe. The only significant difference in the deposits is the presence of wave-generated lags within the Economy Point ridge, versus the near absence of wave-generated structures in the ridges in the Roda Formation.

- 57 -

- 58 -

Figure 3.11 Previous Page: A) A view of the Minas Basin and Cobequid Bay, Nova Scotia with Economy point inset (Satellite Images courtesy of Google Maps). B) The Economy Point tidal ridge (Klein, 1970) is a similar size and has paleocurrents and bedforms remarkably similar to those reconstructed for the Roda Formation (C). Its top is visible at low tide and it is completely submerged during high tides― tidal range of up to 16.3 meters in the Minas Basin (Dalrymple et al., 1990).

3.7 Comparison between the Roda and Esdolomada members

An abundance of compound dunes and cross-bedded sandstone, the lack of mudstone, and the scarcity of Facies 4 rippled sandstone imply that tidal currents were fast during the deposition of Ridges A-C in the Roda Member. Esdolomada Member Ridges D-F, on the other hand, have a higher mudstone percentage, smaller cross-beds, and a higher proportion of Facies

4 rippled sandstone (Fig. 3.7) implying slower currents. The decrease in current velocity could be due to regional changes in basin shape influencing tidal resonance or could be caused by local coastal irregularities. Syndepositional deformation is documented to have occurred during deposition of the Roda Member and had generally ceased prior to the deposition of the

Esdolomada Member (Lopez-Blanco et al., 2003). This tectonic activity created an elongate topographic high (the Turbon anticline; Fig. 1.1), oriented NNW-SSE and located near the southwestern margin of the study area. It is believed to have influenced tidal-current strength and direction by funnelling the currents between the anticlinal high and the delta to the northeast. The antiform was buried and essentially inactive by the beginning of the

Esdolomada Member.

The topographic high formed by the Turbon anticline would have influenced flow paths of the tidal currents. Ridges A-D appear to have a relatively consistent orientation, whereas

Ridges E and F are rotated slightly clockwise from the earlier ridges. Thus, the Turbon paleo- high appears to have influenced the location and orientation of the ridges in the Roda Member, but not in the Esdolomada Member.

- 59 -

3.8 Variation between Ridges

Sedimentary Structures

The specific bedforms present on a given ridge are dependent on current speed and water depth, with secondary influence from sediment availability and grain size (Fig. 3.12).

Where there is an abundance of medium-grained sand, low current velocities (ca. 20-50 cm/s) will generate current ripples, whereas higher current velocities will form simple and compound dunes (50-200 cm/s; Boguchwal and Southard; 1990a, 1990b, 1990c; Dalrymple and Rhodes,

1995).

Water depth influences the circulation pattern on and around tidal ridges. Shallow ridges (Fig. 3.12 A-C) typically have slower velocities at the crest with faster currents in the trough due to frictional slowing in the shallow water on the ridge crest. As a result, the lowest- energy bedforms (i.e., ripples) can be expected to occur most commonly on crest of such ridges, although the slower currents along the crest can still be high enough for dune formation

(Schmitt et al., 2007). Shallow ridges are also prone to the development of swatchways

(Dalrymple and Rhodes, 1995). Ridges forming in deeper water (Fig. 3.12 D-E) will typically have the highest velocities at the crest. Thus, the lowest-energy bedforms will be found along the lower parts of the ridge flanks, with increasingly higher-energy bedforms towards the crest.

The Roda Formation is interesting and unique in that it preserves nearly the full gamut of ridge types, from low-energy ridges composed of ripples and dunes (e.g., Ridge F) to high- energy, compound-dune-dominated ridges (e.g., Ridge B). Ridge B further documents the transition from deep to shallow as aggradation occurred over the life of the ridge.

- 60 -

Figure 3.12: Variation between tidal ridges due to changes in current velocity and water depth. Current speeds are displayed as mean maximum current speeds for the dominant and subordinate directions. Currents of 50-200 cm/s are capable of forming dunes (A, B, D, E), while slower currents (20-50 cm/s) typically form ripples (C). If current speeds are periodically very high and the ridge significantly alters flow, swatchway channels can be incised into the top of the ridge. Deep ridges (D and E) have faster currents along the crest, so bedforms decrease in energy down the flanks. Fast currents are more effective at transporting sediment, and ridges are likely to be inherently larger than those forming under weak currents.

Vertical Grain-size Changes

Vertical grain-size changes through the ridges are variable. Coarsening-upward with decreasing mud percentage upward should be typical for ridges that occur on the shelf away from the influence of estuarine or deltaic channels, and that is what is observed (Fig. 3.8 C and

Fig. 3.9 B). This trend occurs because deep-water currents accelerate up onto the crest of

- 61 -

ridges, moving coarse sediment up and onto the crest. As currents decrease away from the crest, capacity and competence decreases and coarse grains cannot be transported away. These conditions should occur in the initial stages of ridge growth, continuing until the ridge is buried or the ridge reaches a shallow water depth where friction becomes significant and current speeds at the crest begin to decrease (Huthnance, 1982). Every ridge investigated predominantly showed an upward-coarsening grain-size trend.

Once the ridge builds high enough, however, currents will become slower at the crest, while the current in the surrounding lower areas will be faster. This situation would likely produce a fining-upward trend with a sharp base, similar to what is seen in estuarine and deltaic channels (Olariu and Bhattacharya, 2006; Dalrymple et al., 1992). Swatchway channels incising into ridge deposits can complicate grain-size trends, but typically fine upward.

3.9 Conclusions

The facies variation within, and the stratigraphic and geographic distribution of, six tidal sand ridges within the Eocene Roda Formation has been documented. The ridges are interpreted as headland associated because they have flow patterns, grain-size trends, and size that matches models and modern examples. They are situated immediately to the west of the deltaic headland, implying the overall dominance of the westerly flowing ebb tide: the southwestern (offshore) side of the ridge was dominated by this current, whereas the sheltered landward side of the ridge was dominated by the flood tide. Because the ridges are located in a consistent geographic location within the basin, are not found in groups, and are oriented approximately parallel to the paleo-shoreline, they are neither estuarine nor open-shelf ridges.

A progradational deltaic origin for the tidal deposits is ruled out by three observations: the ridges are not perpendicular to the shoreline; they incise into underlying deltaic deposits, and

- 62 -

they contain a marine ichnofacies that contrasts with the severely impoverished delta mouth-bar facies.

In the Tremp-Graus Basin, deltaic complexes had to prograde at least two kilometres into the basin to form a headland large enough to significantly influence tidal currents. In this case, this only began to occur in late highstand time of the 3rd-order cycles. The fact that the feeder river appears to have been locked in place structurally may have aided in allowing sufficient protrusion to occur. When deltas sourced from static feeder systems stop depositing sediment at the coast, tidal processes can take over and rework the seaward edge of the delta into a headland-attached tidal ridge.

The deposits comprising tidal ridges in the Roda Formation can be subdivided into five facies including; large 3-10 metre-thick sandy foresets (Facies 1), compound cross-bedded facies with cross-beds dipping up (Facies 2) or down master bedding (Facies 3), rippled sands

(Facies 4), and muddy deposits (Facies 5). Two types of ridges are documented, high energy and low energy. The high-energy ridges are composed predominantly of cross-bedded Facies

1-3, whereas the low-energy ridge is composed predominantly of cross-laminated Facies 4.

Ridge F records relatively constant flow processes with the fastest currents occurring on the ridge crest and decreasing into the adjacent troughs, whereas Ridge B records a change in flow processes: initially the fastest currents were along the crest of the ridge forming Unit 1 on the northern flank of the ridge, but, once the ridge reached a height of 25 metres, it impeded cross-ridge currents, causing sediment to accrete mostly on the southern flank forming Unit 2, although some deposition also occurred on the northern margin (Unit 3). The preferential preservation of the initial ‘deep’ phase within the ridges appears to indicate that the ridges did not migrate significantly, possibly because they were anchored to the adjacent deltaic headland.

The suite of facies that can occur within tidal ridges is dependant on current speed, water depth

- 63 -

and grain size, whereas the organization of those facies is determined by the pattern of flow around the ridge. If currents are fastest on the crest of the ridge, the highest energy bedforms are found there, with lower-energy bedforms on the lower flanks. If currents are slower on the crest of the ridge, the highest energy bedforms are typically found on the flanks of the ridge.

If conditions remain stable over the life of a ridge, it will likely contain a simple facies architecture, which conforms to a given combination of current speed, water depth and flow pattern (e.g., Ridge F; Fig. 3.9). Changes in any of these variables during the ridge’s life can significantly complicate the preserved architecture (e.g., Ridge B; Fig. 3.8).

- 64 -

CHAPTER 4

CONCLUSION

4.1 Concluding Remarks

The Eocene Roda Formation in northern Spain contains a total of 18 sandy parasequences, each of which records the progradation of a coarse-grained delta. Alluvial-plain mudstones (Facies A), channel deposits (Facies B) and deltaic mouth-bar tongues (Facies C) are capped by carbonate grainstones (Facies E) that were developed during a period of transgression and marine mudstones (Facies F). Six of the progradational deltaic tongues are overlain by tidal ridges (Facies D). The 18 parasequences comprise two third-order cycles that were deposited from ~53-52 Ma. A maximum flooding surface within the formation separates these two genetically distinct sequences. Gilbert-style delta deposits typify the lower of the two, Sequence 1, whereas low-angle compound cross-bedded deltaic deposits with thicker and muddier parasequences are common in Sequence 2. These differences reflect higher rates of subsidence within Sequence 2. Also, there are half as many deltaic parasequences in Sequence

2. Sandy shelf deposits, Facies G, likely represent distal or lateral equivalents to deltaic deposits, whereas carbonate shelf deposits, Facies E, can, but does not always, represent the distal and lateral equivalent of flooding surfaces. Thus, the basinward equivalent of a full deltaic parasequence may be only represented by thin shelf sandstones (Facies G) that may or may not be capped by carbonate grainstones (Facies E). Facies E and G are more abundant in

Sequence 2, implying that the depocentre migrated away from the study area, likely towards the east.

There is a larger time gap after Sequence 2 ridges than after Sequence 1 ridges as evidenced by well developed carbonate lags on Sequence 2 ridges. This is likely a direct effect of the higher subsidence in Sequence 2. During Sequence 2, abundant accommodation

- 65 -

inhibited deltaic progradation by trapping coarse grained sediment in the proximal areas of the basin. Trapping of sediment in the proximal areas allowed carbonate organisms, susceptible to siliciclastic sedimentation, to thrive and accumulate. During Sequence 1, there was relatively little accommodation available, and sediment reached deltas quickly allowing for rapid progradation. As a result, there was not enough time in between parasequences for carbonate drapes to form on the ridges.

Sequence 1 ridges are composed of higher-energy deposits than those in Sequence 2, which indicates the presence of faster tidal currents. The change in tidal-current speed may have been due to the cessation of synsedimentary deformation and the growth of an anticline that constricted and accelerated the flow across the front of the delta, or it may have been related to a basinwide decrease in tidal resonance due to changes in basin shape. Tidal ridges

(generalized as Facie D in Chapter 2) are further subdivided into Facies 1-5 in Chapter 3.

Higher energy ridges are composed mostly of 1-5 m thick, angle-of-repose foresets (Facies 1), compound cross-beds with internal cross-beds migrating up (Facies 2) or down master-bedding planes (Facies 3), and subordinate amounts of current-rippled sandstones (Facies 4) and laminated siltstones and mudstones (Facies 5). Lower energy ridges contain no Facies 1 deposits and are composed mostly of Facies 2, 3, 4, and 5. Ridges are documented as having a

Cruziana ichnofacies, ranging within ridges from archetypal to impoverished Cruziana ichnocoenoses with differences reflecting variations in tidal-current speed and sedimentation rate.

Headlands created by deltaic progradation can locally enhance tidal currents and concentrate sediment into headland-associated tidal ridges. A minimum amount of protrusion appears to be necessary to produce a large enough headland for this process to operate. Within the Roda Formation, early HST parasequences are not capped by tidal ridges. This is because

- 66 -

those deltas had only produced a small bulge on the coast, and were ineffective at influencing tidal currents. In the Roda Formation, tidal ridges were only produced after a two kilometre progradation of the delta-complex. In order to generate a sufficiently large bulge, the delta needs to remain in one location for a prolonged period of time so that it does not avulse frequently and produce a uniform fringe along the basin margin. In the study area, this is accomplished by the structural control that locks the river into a fixed location.

Parasequences typically coarsen and shallow upwards, but localized thickening of the transgressive systems tract (which typically manifests as a simple flooding surface) can and does occur. Thicknesses of several metres have been documented in former studies (e.g.,

Arnott, 1995), but we have shown that deposits tens of metres thick and kilometres in extent

(e.g., tidal sand ridges) can be formed during high-frequency transgressions. The presence of well-exposed tidal shelf ridges in the Roda Formation provides a unique opportunity to study the facies and the internal architecture of those facies. As a result, the ability to predict the location and character of ancient tidal ridges has been improved significantly.

- 67 -

REFERENCES

ALLEN, J.R.L., 1980, Sand waves: a model of origin and internal structure: Sedimentary

Geology, v. 26, p. 281-328.

ARENAS, C., MILLA, H., PARDO, G. and POCOVI, A., 2001, Ebro Basin continental

sedimentation associated with late compressional Pyrenean tectonics (north-

eastern Iberia): Controls on basin margin fans and fluvial Systems: Basin

Research, v. 13, p. 65-89.

ARNOTT, R.W.C., 1995, The Parasequence Definition- Are transgressive deposits

inadequately addressed?: Journal of Sedimentary Research, v. B65, No.1, p. 1-6.

ANADÓN, P., ROCA, E., 1996, Geological setting of the tertiary basins of northeast

Spain, in FRIEND, P.F., DABRIO, C.J. (Eds.) Tertiary Basins of Spain: Cambridge

University Press, Cambridge, p. 43-48.

BANN, K.L., FIELDING, C.R.,MACEACHERN, J.A., and TYE, S.C., 2004, Differentiation

of estuarine and offshore marine deposits using integrated ichnology and

sedimentology: Permian Pebbley Beach Formation, Sydney Basin, Australia: in

MCILROY, D., ed., The Application of Ichnology to Palaeoenvironmental and

Stratigraphic Analysis: Geological Society of London Special Publication, v.

228, p. 179–211.

BANN, K.L., TYE, S.C., MACEACHERN, J.A., Fielding, C.R., and Jones, B.G., 2008,

Wave- and storm-dominated deltaic deposits: examples from the early Permian

Sydney Basin, Australia: SEPM Special Publication, v. 90, p. 293-332.

BASTOS, A.C., KENYON, N.H., and COLLINS, M., 2002, Sedimentary Processes,

Bedforms and Facies, Associated with a Coastal Headland: Portland Bill,

Southern UK: Marine Geology, v. 187, no. 3-4, p. 235-258, doi:10.1016/S0025-

3227(02)00380-8.

BASTOS, A.C., PAPHITIS, D., and COLLINS, M.B., 2004. Short-term dynamics and

maintenance processes of headland-associated sandbanks: Shambles Bank,

English Channel, UK: Estuarine, Coastal and Shelf Science, v. 59, p. 33-47,

doi:10.1016/j.ecss.2003.07.008.

BEAUMONT, C., MUNOZ, J.A., HAMILTON, J., and FULLSACK, P., 2000, Factors

controlling the alpine evolution of the central Pyrenees inferred from a

comparison of observations and geodynamical: Journal of Geophysical Research,

v. 105, no. B4, p. 8121-8145.

BELDERSON, R H., 1986, Offshore tidal and non-tidal sand ridges and sheets:

differences in morphology and hydrodynamic setting, in KNIGHT, R.J., and J.R.

MCLEAN, J.R., (Eds.) Shelf sands and sandstones: Canadian Society of Petroleum

Geologists, Memoir II, p. 293-301.

BENNETT, M.R., CASSIDY, N.J., and PILE, J., 2009, Geomorphology internal structure of

a barrier beach as revealed by ground penetrating radar (GPR): Chesil beach,

UK: Geomorphology, v. 104, no. 3-4, p. 218-229,

doi:10.1016/j.geomorph.2008.08.015.

BERNÉ, S., LERICOLAIS, G., MARSSET, T., BOURILLET, J.F., and BATIST, M. DE, 1991,

Erosional offshore sand ridges and lowstand shorefaces: examples from tide- and

wave-dominated environments of France: Journal of Sedimentary Research, v.

68, no. 4, p. 540-555.

BERNÉ, S., TRENTESAUX, A., STOLK, A., MISSIAEN, T., and BATIST, M. DE. 1994,

Architecture and long term evolution of a tidal sandbank: The Middelkerke Bank

(southern North Sea): Marine Geology, v. 121, p. 57-72.

BERNÉ, S., VAGNER, P., GUICHARD, F., LERICOLAIS, G., LIU, Z., TRENTESAUX, A., YIN,

P., and YI, H.I., 2002, Pleistocene forced regressions and tidal sand ridges in the

East China Sea: Marine Geology v. 188, p. 293-315.

BERTHOT, A., and PATTIARATCHI, C., 2005, Maintenance of headland-associated linear

sandbanks: modelling the secondary flows and sediment transport: Ocean

Dynamics, v. 55, p. 526-540, doi:10.1007/s10236-005-0010-9.

BERTHOT, A., and PATTIARATCHI, C., 2006, Field measurements of the three-

dimensional current structure in the vicinity of a headland-associated linear

sandbank: Continental Shelf Research, v. 26, p. 295-317,

doi:10.1016/j.csr.2005.11.010.

BERTHOT, A., and PATTIARATCHI, C., 2006, Mechanisms for the formation of headland-

associated linear sandbanks: Continental Shelf Research, v. 26, p. 987-1004,

doi:10.1016/j.csr.2006.03.004.

BOER, P.L. DE, PRAGT, J.S.J., and OOST, A.P., 1991, Vertically persistent sedimentary

facies boundaries along growth anticlines and climate-controlled sedimentation

in the thrust-sheet-top South Pyrenean Tremp-Graus Foreland Basin: Basin

Research, v. 3, p. 63-78.

BRENNER, R.L., and MARTINSEN, O.J., 1990, The Fossil Sandstone—A shallow marine

sand wave complex in the Namurian of Cumbria and North Yorkshire, England:

Cretaceous Research, v. 48, p. 149-162.

BOYD, R., DALRYMPLE, R.W., and ZAITLIN, B.A., 1992, Classification of clastic coastal

depositional environments: Sedimentary Geology, v. 80, p. 139-150.

BUATOIS, L.A., GINGRAS, M.K., MACEACHERN, J.A., MANGANO, M.G., ZONNEVELD,

J.P., PEMBERTON, S.G., NETTO, R.G., and MARTIN, A., 2005, Colonization of

brackish-water systems through time: evidence from the trace-fossil record:

Palaios v. 20, p. 321-347, doi:10.2110/palo.2004.p04-32.

BURBANK, D.W., PUIGDEFÀBREGAS, C., and MUNOZ, J.A., 1992, The chronology of the

eocene tectonic and stratigraphic development of the Eastern Pyrenean Foreland

Basin, Northeast Spain: Geological Society of America Bulletin, v. 104, p. 1101-

1120.

CATUNEANU, O., 2006, Principals of sequence stratigraphy: Burlington: Elsevier, 375 p.

CHAUMILLON, E., BERTIN, X., FALCHETTO, H., ALLARD, J., WEBER, N., WALKER, P.,

POUVREAU, N., and WOPPELMANN, G., 2008, Multi time-scale evolution of a

wide estuary linear sandbank, the Longe de Boyard, on the French Atlantic coast:

Marine Geology, v. 251, p. 209-223.

CHEN, Z., SAITO, Y., HORI, K., and ZHAO, Y., 2003, Early Holocene mud-ridge

formation in the Yangtze offshore, China: A tidal-controlled estuarine pattern and

sea-level implications: Marine Geology, v. 198, p. 245-257, doi:10.1016/S0025-

3227(03)00119-1.

CHOUGH, S.K., KIM, J.W., LEE, S.H., SHINN, Y.J., JIN, J.H., SUH, M.C., and LEE, J.S.,

2002, High-resolution acoustic characteristics of epicontinental sea deposits,

central-eastern Yellow Sea: Marine Geology, v. 188, p. 317-331.

CHRISTOPHER G. St. C., POSAMENTIER, H.W., ROSS, C.A., and VAN WAGONER, J.C.,

1988, Sea level changes: An integrated approach: Society of Economic

Paleontologists and Mineralogists, Special Publication v. 42, p. 125-154.

COLLINS, M.B., SHIMWELL, S.J., GAO, S., POWELL, H., HEWITSON, C., and TAYLOR,

J.A., 1995, Water and sediment movement in the vicinity of linear sandbanks: the

Norfolk Banks, Southern North Sea: Marine Geology, v. 123, p. 125-142.

CURNELLE, R., DUBOIS, P., and SEGUIN, J.C., 2008, The Mesozoic-Tertiary evolution of

the Aquitaine Basin [and Discussion]: Philosophical Transactions of the Royal

Society of London, Series A, Mathematical and Physical Sciences, v. 305, no.

1489, p. 63-84.

DALRYMPLE, R.W., 1984, Morphology and internal structure of sandwaves in the Bay

of Fundy: Sedimentology, v. 31, p. 365-382.

DALRYMPLE, R.W., KNIGHT, R.J., ZAITLIN, B.A., and MIDDLETON, G.V., 1990,

Dynamics and facies model of a macrotidal sand-bar complex, Cobequid Bay-

Salmon River Estuary (Bay of Fundy): Sedimentology, v. 37, p. 577-612.

DALRYMPLE, R.W., and HOOGENDOORN, E.L., 1986, Erosion and deposition on

migrating shoreface-attached ridges, Sable Island, Eastern Canada: Geoscience

Canada, v. 24, no. 1: 25-36.

DALRYMPLE, R.W., MAKINO, Y., and ZAITLIN, B.A., 1991, Temporal and spatial

patterns of rhythmite deposition on mud flats in the macrotidal Cobequid Bay-

Salmon River Estuary, Bay of Fundy, Canada, in Smith, D.G., Reinson, G.E.,

Zaitlin, B.A., and Rahmani, R.A., Clastic Tidal Sedimentology: Canadian Society

of Petroleum Geologists, Memoir 16, p. 137-160.

DALRYMPLE, R.W., ZAITLIN, B.A., and BOYD, R., 1992, Estuarine facies models;

conceptual basis and stratigraphic implications: Journal of Sedimentary Research,

v. 62, no. 6, p. 1130-1146.

DALRYMPLE, R.W., and RHODES, R.N., 1995, Estuarine dunes and bars, in PERILLO,

G.M.E., Geomorphology and Sedimentology of Estuaries: Developments in

Sedimentology, v. 53, p. 359-422.

DAVIS JR., R.A., and BALSON, P.S., 1991, Stratigraphy of a North Sea tidal sand ridge:

Journal of Sedimentary Petrology, v. 62, no. 1, p. 116-121.

DAVIS JR., R.A., KLAY, J., and JEWELL, P.IV, 1993, Sedimentology and stratigraphy of

tidal sand ridges Southwest Florida Inner Shelf: Journal of Sedimentary

Petrology, v. 63, no. 1, p. 91-104.

DECELLES, P.G., and GILES, K.A., 1996. Foreland basin systems: Basin Research, v. 8,

p. 105-123.

DERAMOND, J., SOUQUET, P., and SPECHT, M., 1993, Relationships between thrust

tectonics and sequence stratigraphy surfaces in foredeeps; model and examples

from the Pyreneees (Cretaceous-Eocene, France, Spain): Geological Society

Special Publication, v. 71, p. 193-219, doi:10.1144/GSL.SP.1993.071.01.09.

De Vries Klein, G., 1970, Depositional and dispersal dynamics of intertidal sand bars.

Journal of Sedimentary Petrology, v. 40, no. 4, p. 1095-1127.

Dyer, K.R., Huntley, D.A., 1999, The origin, classification and modelling of sand

banks and ridges: Continental Shelf Research, v. 19, no. 10, p. 1285-1330, doi:

10.1016/S0278-4343(99)00028-X.

FISHER, J.A., NICHOLS, G.J., and WALTHAM, D.A., 2007, Unconfined flow deposits in

distal sectors of fluvial distributary systems: Examples from the Miocene Luna

and Huesca systems, northern Spain: Sedimentary Geology 195: 55 - 73.

doi:10.1016/j.sedgeo.2006.07.005.

FITZGERALD, P.G., J.A. MUNOZ, and P.J. CONEY. 1999, Asymmetric exhumation across

the Pyrenean orogen: implications for the tectonic evolution of a collisional

orogen: Earth and Planetary Science Letters, v. 173, p. 157-170.

GAWENDA, P., WINKLER, W., SCHMITZ, B., and ADATTE, T., 1999, Climate and

bioproductivity control on carbonate turbidite sedimentation (Paleocene to

earliest Eocene, Gulf of Biscay, Zumaia, Spain: Climatic Change, v. 69, p. 1253-

1261.

GRASMUECK, M., and WEGER, R., 2002, 3d GPR reveals complex internal structure of

Pleistocene oolitic sandbar: Geophysics, v. 21, no. 7, p. 634-639.

HAQ, B.U., HARDENBOL, J. and VAIL, P.R., 1987, Chronology of fluctuating sea levels

since the Triassic: Science, v. 235, p. 1156–1167

HARRIS, P.T., and COLLINS, M.B., 1991, Sand transport in the Bristol Channel: bedload

parting zone or mutually evasive transport pathways?: Marine Geology 101: 209-

216.

HARRIS, P.T., PATTIARATCHI, C.B., COLE, A.R., and KEEN, J.B., 1992, Evolution of

subtidal sandbanks in Moreton Bay, eastern Australia: Marine Geology, v. 103,

p. 225-247.

HORRILLO-CARABALLO, J.M., and REEVE, D.E., 2008, Morphodynamic behaviour of a

nearshore sandbank system: The Great Yarmouth: Marine Geology, v. 254, p. 91-

106, doi:10.1016/j.margeo.2008.05.014.

HULSCHER, S.J.M.H, SWART De, H.E., and VRIEND DE, H.J., 1993, The generation of

offshore tidal sand banks and sand waves: Continental Shelf Research, v. 13, no.

11, p. 1183-1204.

HUNTLEY, D.A., HUTHNANCE, J.M., COLLINS, M.B., LIU, C., NICHOLLS, R.J.,

HEWITSON, C., DYER, K.R., and JAGO, C.F., 1993, Hydrodynamics and sediment

dynamics of North Sea sand waves and sand banks [and discussion]:

Philosophical Transactions, Physical Sciences and Engineering, v. 343, no. 1669,

p. 461-474.

HUTHNANCE, J.M., 1982a, On one mechanism forming linear sand banks: Estuarine,

Coastal and Shelf Science, v. 14, no.1, p. 79-99.

HUTHNANCE, J.M., 1982b, On the formation of sand banks of finite extent: Estuarine,

Coastal and Shelf Science, v. 15, no. 3, p. 277-299.

JIN, H., and CHOUGH, K., 2002, Erosional shelf ridges in the mid-eastern Yellow Sea:

Geo-Marine Letters, v. 21, p. 219-225, doi:10.1007/s00367-001-0082-6.

JUNG, W.Y., SUK, B.C., MIN, G.H., and LEE, Y.K., 1998, Sedimentary structure and

origin of a mud-cored pseudo-tidal sand ridge, eastern Yellow Sea, Korea:

Marine Geology, v. 151, p. 73-88.

KANAZAWA, K., 1992, Adaptation of test shape for burrowing and locomotion in

spatangoid echinoids: Palaeontology, v. 35, Part 4, p. 733-750.

KENYON, P.M., and TURCOTTE, D.L., 1985, Morphology of a delta prograding by bulk

sediment transport: Geological Society of America Bulletin, v. 96, p. 1457-1465.

KLEIN, G.D., and RYER, T.A., 1978, Tidal circulation patterns in Precambrian,

Paleozoic, and Cretaceous epeiric and mioclinal shelf seas: Bulletin of the

Geological Society of America, v. 89, no. 7, p. 1050-1058.

KOSTASCHUK, R.A., 1985, River mouth processes in a fjord-delta, British Columbia,

Canada: Marine Geology, v. 69, p. 1-23.

KOWALIK, Z., and POLYAKOV, I., 1998, Tides in the Sea of Okhotsk: Journal of

Physical Oceanography, v. 28, p. 1389-1408.

KREISA, R.D., and MOIOLA, R.J., 1986, Sigmoidal tidal bundles and other tide-

generated sedimentary structures of the Curtis Formation, Utah: Geological

Society of America Bulletin, v. 97, p. 381-387.

KVALE, E.P., and MASTALERZ, M., 1998, Evidence of ancient freshwater tidal deposits,

in ALEXANDER, C.R., DAVIS, R.A., and HENRY, V.J., Tidalites: Processes and

products: SEPM Special Publication, v. 61, p. 95–107.

KVALE, E.P., 2006, The origin of neap – spring tidal cycles: Marine Geology, v. 235, p.

5 – 18, doi:10.1016/j.margeo.2006.10.001.

LECLAIR, S.F., and BRIDGE, J.S., 2001, Quantitative interpretation of sedimentary

structures formed by river dunes: Journal of Sedimentary Research, V. 71, no. 5,

p. 713-716.

LEE, H.J., 2008, Sediment transport processes over a sand bank in macrotidal Garolim

Bay, west coast of Korea: Geosciences Journal, v. 12, no. 3, p. 243-253,

doi:10.1007/s12303-008-0025-6.

LEREN, B.L.S., HOWELL, J., ENGE, H., and MARTINIUS, A.W., 2010, Controls on

stratigraphic architecture in contemporaneous delta systems from the Eocene

Roda Sandstone, Tremp-Graus Basin, northern Spain: Sedimentary Geology, v.

229, no. 1-2, p. 9-40. doi:10.1016/j.sedgeo.2010.03.013.

LOPEZ-BLANCO, M., MARZO, M., and MUNOZ, J.A., 2003, Low-amplitude,

synsedimentary folding of a deltaic complex: Roda Sandstone (lower Eocene),

South-Pyrenean Foreland Basin: Basin Research, v. 15, p. 73-95.

MACEACHERN, J.A., BANN, K.L., BHATTACHARYA, J.P., and HOWELL, C.D., 2005a,

Ichnology of deltas: Organism responses to the dynamic interplay of rivers,

waves, storms and tides, in GIOSON, L., and BHATTACHARYA, J.P., River deltas-

Concepts models and examples: SEPM Special Publication, v. 83, p. 49–85.

MACEACHERN, J.A., BANN, K.L., PEMBERTON, S.G., and GINGRAS, M.K., 2005b, The

ichnofacies paradigm: high-resolution paleoenvironmental interpretation of the

rock record, in MACEACHERN, J.A., BANN, GINGRAS, M.K., PEMBERTON, S.G.,

Applied ichnology: SEPM Short Course Notes, v. 52, p. 27–64.

MACEACHERN, J.A., PEMBERTON, S.G., BANN, K.L., and GINGRAS, M.K., 2005c,

Departures from the archetypal ichnofacies: Effective recognition of physico-

chemical stresses in the rock record, in MACEACHERN, J.A., BANN, GINGRAS,

M.K., PEMBERTON, S.G., Applied ichnology: SEPM Short Course Notes, v. 52, p.

65–93.

MCPHERSON, J.G., SHANMUGAM, G., and MOIOLA, R.J., 1987, Fan-deltas and braid

deltas: Varieties of coarse-grained deltas: Geological Society of America

Bulletin, v.99, p. 331-340.

MADDOCK, L., and PINGREE, R.D., 1978, Numerical simulation of the Portland tidal

eddies: Estuarine And Coastal Marine Science, v. 6, p. 353-363.

MARTINIUS, A.W., 1995, Macrofauna associations and formation of shell

concentrations in the early Eocene Roda Formation (southern Pyrenees, Spain):

Scripta Geologica, v. 108, p.1-39.

MCCAVE, I.N., and LANGHORNE, D.N., 1982, Sand waves and sediment transport

around the end of a tidal sand bank: Sedimentology, v. 29, p. 95-110.

MIALL, A.D., 1997, The geology of stratigraphic sequences: Springer-Verleg Berlin

Heidelberg, New York, 433 p.

MITCHUM Jr., R.M., VAN WAGONER, J.C., 1991, High-frequency sequences and their

stacking patterns: Sequence-stratigraphic evidence of high-frequency eustatic

cycles: Sedimentary Geology, v. 70, p. 131-160.

MOLENAAR, N., and MARTINIUS, A.W., 1990, Origin of nodules in mixed siliciclastic-

carbonate sandstones, the lower Eocene Roda Sandstone Member, southern

Pyrenees, Spain: Sedimentary Geology, v. 66, p. 277-293.

MOLENAAR, N., and MARTINIUS, A.W., 1996, Fossiliferous intervals and sequence

boundaries in shallow marine, fan-deltaic deposits (early Eocene, Southern

Pyrenees, Spain): Palaeogeography, Palaeoclimatology, Palaeoecology, v. 121,

no. 3-4, p. 147–168.

MÜLLER, R.D., SDROLIAS, M., GAINA, C., STEINBERGER, B., and HEINE, C., 2008,

Long-term sea-level fluctuations driven by ocean basin dynamics: Science, v.

319 (5868): 1357-1362.

MUTTI, E., TINTERRI, R., BIASE DI, D., FAVA, L., MAVILLA, N., ANGELLA, S., and

CALABRESE, L., 2000, Delta-front facies associations of ancient flood-dominated

fluvio-deltaic systems: Reviews of the Society of Geology Espana, v. 13, no. 2, p.

165–190.

MUTTI, E., TINTERRI, R., BENEVELLI, G., BIASE DI, D., and CAVANNA, G., 2003,

Deltaic, mixed and turbidite sedimentation of ancient foreland basins: Marine and

Petroleum Geology, v. 20, p. 733-755, doi:10.1016/j.marpetgeo.2003.09.001.

NEILL, S.P., HASHEMI, M.R., and ELLIOTT, A.J., 2007, An enhanced depth-averaged

tidal model for morphological studies in the presence of rotary currents:

Continental Shelf Research, v. 27, p. 82-102, doi:10.1016/j.csr.2006.09.002.

NEILL, S.P., 2008, The role of Coriolis in sandbank formation due to a headland/island

system: Estuarine, Coastal and Shelf Science, v. 79, p. 419-428,

doi:10.1016/j.ecss.2008.04.015.

NIO, S.D., and YANG, C.S., 1991, Sea-level fluctuations and the geometric variability of

tide-dominated sand bodies: Sedimentary Geology, v. 70, p. 161-193.

NIO, S.D., 1976, Marine Transgressions as a Factor in the Formation of Sandwave

Complexes: Geologie En Mijnbouw, v. 55, p. 18-40.

Yang, C.S., and Nio, S.D., 1985, The estimation of palaeohydrodynamic processes

from subtidal deposits using time series analysis methods: Sedimentology, v. 32.

p. 41-57.

OFF, T., 1963, Rhythmic linear sand bodies caused by tidal currents: AAPG bulletin, v.

47, no. 2, p. 324-341.

OLARIU, C., and BHATTACHARYA, J.P., 2006, Terminal distributary channels and delta

front architecture of river-dominated delta systems: Journal of Sedimentary

Research, v. 76, p. 212-233, doi:10.2110/jsr.2006.026.

OLARIU, I., OLARIU, C., and STEEL, R.J., (2011), Anatomy of a laterally migrating tidal

bar in front of a delta system: Esdolomada Member, Roda Formation, Graus-

Tremp Basin, Spain.

ORTON, G.J. and READING, H.G., 1993, Variability of deltaic processes in terms of

sediment supply, with particular emphasis on grain size: Sedimentology, v. 40, p.

475-512.

PARDO, A., KELLER, G., and OBERHAENSLI, H., 1999, Paleoecologic and

paleoceanographic evolution of the Tethyan realm during the Paleocene-Eocene

transition: The Journal of Foraminiferal Research, v. 29, no. 1, p. 37.

PARK, H., and VINCENT, C.E., 2007, Evolution of Scroby Sands in the East Anglian

coast, UK: Journal of Coastal Research, Special Issue v. 50, p. 868-873.

PAYROS, A., PUJALTE, V., and ORUE-ETXEBARRIA, X., 1999, The South Pyrenean

Eocene carbonate megabreccias revisited: New interpretation based on evidence

from the Pamplona Basin: Sedimentary Geology, v. 125, p. 165-194.

PENLAND, S., BOYD, R., and SUTER, J.R., 1988, Transgressive depositional systems of

the Mississippi Delta Plain: A model for barrier shoreline and shelf sand

development: Journal of Sedimentary Petrology, v. 58, no. 6, p. 932-949.

PINGREE, R.D., 1978, The formation of the Shambles and other banks by tidal stirring

of the seas: Journal of the Marine Biological Association of the United Kingdom,

v. 58, no. 1, p. 211-226.

PINGREE, R.D., and GRIFFITHS, D.K., 1978, Tidal fronts on the shelf seas around the

British Isles: Journal of Geophysical Research, v. 83, no. C9, p. 4615-4622.

PINGREE, R.D., and MADDOCK, L., 1979, The tidal physics of headland flows and

offshore tidal bank formation: Marine Geology, v. 32, p. 269-289.

PLAZIAT, J. C., 1981, Late Cretaceous to Late Eocene palaeogeographic evolution of

southwest Europe: Palaeogeography, palaeoclimatology, palaeoecology, v. 36 no.

3-4, p. 263-320.

POREBSKI, S.J., and STEEL, R.J., 2006, Deltas and sea-level change: Journal of

Sedimentary Research, v. 76, p. 390-403, doi:10.2110/jsr.2006.034.

POSAMENTIER, H.W., and VAIL, P.R., 1988a, Eustatic controls on clastic deposition I―

conceptual framework, in WILGUS, C.K., HASTINGS, B.S., KENDALL, C.G. St. C.,

POSAMENTIER, H.W., ROSS, C.A., VAN WAGONER, J.C., Sea level changes: An

integrated approach: Society of Economic Paleontologists and Mineralogists

Special Publication, v. 42, p. 109-124.

POSAMENTIER, H.W., and VAIL, P.R., 1988b, Eustatic controls on clastic deposition

II― sequence and systems tract models, in WILGUS, C.K., HASTINGS, B.S.,

KENDALL, C.G. St. C., POSAMENTIER, H.W., ROSS, C.A., and VAN WAGONER,

J.C., Sea level changes: An integrated approach: Society of Economic

Paleontologists and Mineralogists Special Publication, v. 42, p. 109-124.

POSAMENTIER, H.W., and ALLEN, G.O., 1999, Siliciclastic sequence stratigraphy—

concepts and applications: SEPM Concepts in Sedimentology and Paleontology,

v. 7, p. 210.

POSAMENTIER, H.W., 2001, Ancient shelf ridges―a potentially significant component

of the transgressive systems tract: Case study from offshore northwest Java:

AAPG Bulletin, v. 86, no.1, p. 75-106.

PONTÉN, A., and PLINK-BJÖRKLUND, P., 2007, Depositional environments in an

extensive tide-influenced delta plain, Middle Devonian Gauja Formation,

Devonian Baltic Basin: Sedimentology, v. 54, p. 969-1006.

PONTÉN, A., and PLINK-BJÖRKLUND, P., 2009, Regressive to transgressive transits

reflected in tidal bars, Middle Devonian Baltic Basin: Sedimentary Geology, v.

218, p. 48-60.

POSTMA, G., 1990, Depositional architecture and facies of river and fan deltas: a

synthesis: Special Publication, International Associations of Sedimentologists, v.

10, p. 13-27.

POSTMA, G., 1995, Sea-level-related architectural trends in coarse-grained delta

complexes: Sedimentary Geology, v. 98, p. 3-12.

PUIGDEFÀBREGAS, C., MUÑOZ, J.A. and MARZO, M., 1986. Thrust belt development in

the eastern Pyrenees and related depositional sequences in the southern foreland

basin, in ALLEN, P.A. and HOMEWOOD, P., Foreland Basins Special Publication

of the International Association of Sedimentologists: v.8, p. 229–246.

PUIGDEFÀBREGAS, C., MUÑOZ, J.A., and VERGÉS, J., 1992, Thrusting and foreland

basin evolution in the southern Pyrenees, in McClay, K.R., Thrust Tectonics:

Chapman and Hall, London, p. 247–254.

REYNAUD, J.Y., TESSIER, B., PROUST, J.N., DALRYMPLE, R.W., MARSSET, T., BATIST,

M. DE, BOURILLET, J.F., and LERICOLAIS, G., 1999, Eustatic and hydrodynamic

controls on the architecture of a deep shelf sand bank (Celtic Sea):

Sedimentology, v. 46, p. 703-721.

REYNAUD, J.Y., TESSIER, B., AUFFRET, J.P., BERNÉ, S., MARSSET, T., and WALKER, P.,

2003, The offshore quaternary sediment bodies of the English channel and its

western approaches: Journal of Quaternary Science, v. 18, p. 361-371,

doi:10.1002/jqs.758.

REYNAUD, J.Y., and DALRYMPLE, R.W., (2011), Tidal deposits on continental shelves

and shallow-water seas, in Davis, R.A., Jr. and Dalrymple, R.W., eds., Principles

of Tidal Sedimentology. Springer, New York, in press.

ROBINSON, A.H.W., 1965, Residual currents in relation to shoreline evolution of the

East Anglian Coast: Marine Geology, v. 4, p. 57-84.

ROBINSON, I.S., 1980, Tidal vorticity and residual circulation: Deep-Sea Research, v.

28, no. 3, p. 195-212.

SCHEIBNER, C., RASSER, M.W., and MUTTI, M., 2007, The Campo section (Pyrenees ,

Spain) revisited: Implications for changing benthic carbonate assemblages across

the Paleocene – Eocene boundary: Palaeogeography, Palaeoclimatology,

Palaoecology, v. 248, p. 145-168, doi:10.1016/j.palaeo.2006.12.007.

SCHMITT, T., MITCHELL, N.C., and RAMSAY, T.S., 2007, Use of swath bathymetry in

the investigation of sand dune geometry and migration around a near shore

‘banner’ tidal sandbank, in BALSON, P.S., COLLINS, M.B., Coastal and Shelf

Sediment Transport, Geological Society of London, Special Publications, v. 274,

p. 53-64.

SCHMITZ, B., PUJALTE V., and NUNEZ-BETELU, K., 2001, Climate and sea-level

perturbations during the initial Eocene Thermal Maximum: evidence from

siliciclastic units in the Basque Basin (Ermua, Zumaia and Trabakua Pass),

northern Spain: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 165, p.

299-320.

SCHMITZ, B., and PUJALTE, V., 2003, Sea-level, humidity, and land-erosion records

across the initial Eocene thermal maximum from a continental-marine transect in

northern Spain: Geology 31, v. 8, p. 689-692.

SCHWENTKE, W., and KUHNT, W., 1992, Subsidence history and continental margin

evolution of the Western Pyrenean and Basque Basins: Paleogeography,

Palaeoclimatology, Palaeoecology v. 95, p. 297-318.

SHAW, J., DUFFY, G., TAYLOR, R.B., CHASSE, J., and FROBEL, D., 2008, Role of a

submarine bank in the long-term evolution of the northeast coast of Prince

Edward Island, Canada: Evolution, v. 24, no. 5, p. 1249-1259, doi:10.2112/07-

08607.1.

SNEDDEN, J.W., TILLMAN, R.W., KREISA, R.D., SCHWELLER, W.J., CULVER, S.J., and

WINN, R.D. JR., 1994, Stratigraphy and genesis of a modern shoreface-attached

sand ridge, Peahala Ridge, New Jersey: Journal of Sedimentary Research, v. B64,

no. 4, p. 560-581.

SNEDDEN, J.W., and DALRYMPLE, R.W., 1999, Modern shelf sand ridges: From

historical perspective to a unified hydrodynamic and evolutionary model, in

BERGMAN, K. M., Isolated shallow marine sand bodies: Sequence stratigraphic

analysis and sedimentologic interpretation: SEPM Special Publication, v. 64, p.

13–28.

STRIDE, A.H., BELDERSON, R.H., KENYON, N.H., and JOHNSON, M.A., 1982, Offshore

tidal deposits: sand sheet and sand bank facies, in STRIDE, A.H., Offshore tidal

sands: Processes and deposits, p. 58–94.

STRIDE, A H., 1988, Indications of long term episodic suspension transport of sand

across the Norfolk Banks, North Sea: Marine Geology, v. 79, p. 55-64.

SWIFT, D.J.P., 1975, Tidal sand ridges and shoal-retreat massifs: Marine Geology, v.

18, p. 105-134.

SWIFT, D.J.P., SEARS, P.C., BOHLKE, B., and HUNT, R., 1978, Evolution of a shoal

retreat massif, North Carolina Shelf: Inferences from areal geology: Marine

Geology, v. 27, p. 19-42.

TESSIER, B., CORBAU, C., CHAMLEY, H., and AUFFRET, J.P., 1999, Internal structure of

shoreface banks revealed by high-resolution seismic reflection in a macrotidal

environment (Dunkerque Area, northern France): Journal of Coastal Research, v.

15, no. 3, p. 593-606.

TRENTESAUX, A., STOLK, A., and BERNÉ, S., 1999. Sedimentology and stratigraphy of a

tidal sand bank in the southern North Sea: Marine Geology, v. 159, p. 253-272.

TINTERRI, R., 2007, The lower Eocene Roda Sandstone (south-central Pyrenees): An

example of a flood-dominated river-delta system in a tectonically controlled

basin: Rivista Italiana di Paleontologia e Stratigrapfia, v. 113, no. 2, p. 223-255.

TORRICELLI, S., KNEZAUREK, G., and BIFFI, U., 2006, Sequence biostratigraphy and

paleoenvironmental reconstruction in the Early Eocene Figols Group of the

Tremp-Graus Basin (south-central Pyrenees, Spain): Palaeogeography,

palaeoclimatology, palaeoecology, v. 232, no.1, p. 1-35,

doi:10.1016/j.palaeo.2005.08.009.

TURNER, J.P., 1990, Structural and stratigraphic evolution of the West Jaca thrust-top

basin, Spanish Pyrenees: Journal of the Geological Society, London, v. 147, p.

177-184.

TURNER, J.P., 1992, Evolving alluvial stratigraphy and thrust front development in the

West Jaca piggyback basin, Spanish Pyrenees: Journal of the Geological Society,

London, v. 149, p. 51-63.

VAN EDEN, J. G., 1970, A reconnaissance of deltaic environment in the Middle Eocene

of the south-central Pyrenees, Spain: Geologie & mijnbouw: v. 49, no. 2, p. 145.

VAN DE MEENE, J.W.H., BOERSMA, J.R., and TERWINDT, J.H.J., 1996, Sedimentary

structures of combined flow deposits from the shoreface-connected ridges along

the central Dutch coast: Marine Geology, v. 131, no. 3-4, p. 151–175.

VAN WAGONER, J.C., MITCHUM, R.M., CAMPION, K.M., and RAHMANIAN, V.D., 1990,

Siliciclastic sequence stratigraphy in well logs, cores, and outcrops: AAPG

Methods in Exploration, v. 7, 55 p.

VANWESENBEECK, V., and LANKNEUS, J., 2000, Residual sediment transport paths on a

tidal sand bank: A comparison between the modified McLaren Model and

Bedform Analysis: Journal of Sedimentary Research, v. 70, no. 3, p. 470-477.

VERGÉS, J, MILLAN, H., ROCA, E., MUFIOZ, J.A., MARZO, M., CITES, J., BEZEMER,

T.D., ZOETEMEIJER, R., and CLOETINGH S., 1995, Eastern Pyrenees and related

foreland basins: Marine and Petroleum Geology, v. 12, no. 8, p. 893-915.

VERGÉS, J., and MUNOZ, J.A., 1990, Thrust sequences in the southern central Pyrenees:

Bulletin de la Société géologique de France, v. 8, no. 2, p. 265–271.

VERGÉS, J., and Burbank, D.W., 1996, Eocene-Oligocene thrusting and basin

configuration in the eastern and central Pyrenees (Spain) in FRIEND, P.F.,

DABRIO, C.J. (Eds.) Tertiary Basins of Spain: Cambridge University Press,

Cambridge, p. 120-133.

VINCENT, S.J., 1999, The role of sediment supply in controlling alluvial architecture: an

example from the Spanish Pyrenees: Journal of the Geological Society, London,

v. 156, p. 749-759.

VINCENT, S.J., 2001, The Sis palaeovalley: A record of proximal fluvial sedimentation

and drainage basin development in response to Pyrenean mountain building:

Sedimentology, v. 48, no. 6, p. 1235–1276.

WILLIS, B.J., Bhattacharya, J.P., GABEL, S.L., and WHITE, C.D., 1999, Architecture of a

tide-influenced river delta in the Frontier Formation of central Wyoming, USA:

Sedimentology, v. 46, p. 667-688.

WILLIS, B.J., and GABEL, S.L., 2000, Formation of deep incisions into tide-dominated

river deltas: Implications for the stratigraphy of the Sego Sandstone, Book Cliffs,

Utah, U.S.A.: Journal of Sedimentary Research, v. 73, p. 246-263.

WILLIS, B.J., and GABEL, S.L., 2001, Sharp-based, tide-dominated deltas of the Sego

Sandstone, Book Cliffs, Utah, USA: Sedimentology, v. 48, no. 3, p. 479-506.

WILLIS, B.J., 2005, Deposits of Tide-influenced river deltas, in GIOSAN, L., and

BHATTACHARYA, J.P., River deltas―Concepts, models, and examples: SEPM

Special Publication, v. 83, p. 87-129.

WOOD, L.J., 2003, Predicting tidal sand reservoir architecture using data from modern

and ancient depositional systems: AAPG Memoir, v. 80, p. 1 - 22.

YANG, C.S., and NIO, S.D., 1985, The estimation of palaeohydrodynamic processes

from subtidal deposits using time series analysis methods: Sedimentology, v. 32,

no. 1, p. 41–57.

YANG, C.S., and NIO, S.D., 1989, An ebb-tide delta depositional model―A comparison

between the modern Eastern Scheldt tidal basin (southwest Netherlands) and the

lower Eocene Roda Sandstone in the southern Pyrenees (Spain): Sedimentary

Geology, v. 64, p. 175-196.

YANG, C-S., 1989, Active, moribund and buried tidal sand ridges in the East China Sea

and the southern Yellow Sea: Marine Geology, v. 88, p. 97-116.

YOSHIDA, S., STEEL, R., and DALRYMPLE, R.W., 2007, Changes in depositional

processes — An ingredient in a new generation of sequence-stratigraphic models:

Journal of Sedimentary Research: Perspectives, v. 77, p. 447-460,

doi:10.2110/jsr.2007.048.

ZOETEMEIJER, R.S., CLOETINGH, SASSI, W., and ROURE, F., 1993, Modelling of

piggyback-basin stratigraphy: Record of tectonic evolution: Tectonophysics p.

226: 253-269.

APPENDIX

A.1 Measured Sections

Data from the measured sections in the study area is presented here. Locations of the measured sections are documented in Figure A1. Within the measured sections dip orientations are documented (dip-direction usually followed by dip angle) for master- bedding surfaces (m.b.), cross-beds (c.b.), and cross-laminations (c.l.). Section headings include Facies A-G from Chapter 2 and a general observations column (Obs.). Although measured section data is documented here, many observations noted within this study came from high resolution photographs and notes of section in-between logs. A sample of those photographs is provided in Appendix A.2.

Figure A.1.1: Locations of measured sections 1-21 within the study area.

100

101

102

103

104

105

106

107

108

109

110

111

A.2 Outcrop Photos

High-resolution photographs were fundamental to this paper; however, due to the number and size of those photographs only a small sample is chosen here to illustrate the nature of the outcrop. See Figure 1.4 for the location of sections mentioned here.

Figure A.2.1 View of La Puebla de Roda from the east, including outcrop measured sections 12, 13 and 14. The town is built upon the resistant La Puebla Limestone.

112

Figure A.2.2 Section 21 (left) to section 20.

Figure A.2.3 Section 20 (left) to section 18.

113

Figure A.2.4 Photograph of the Roda Member taken from section 20 (left) towards the SW.

Figure A.2.5 View to the north showing sections 18 (left) and 19 containing deltaic and channelized facies.

114

Figure A.2.6 Section 18 (left) to section 16 looking toward the east.

Figure A.2.7 View toward the northeast showing an alternate view of section 18 (left) to section 16.

115

Figure A.2.8 View toward the NE of section 16 (left) to section 8.

Figure A.2.9 View toward the NE of section 16 (left) to section 8

116

Figure A.2.10 View toward the NE of section 8 (left) to Section 7.

Figure A.2.11 View toward the NE of section 7 (left) to southern faulted zone.

117

Figure A.2.12 View toward the NE of section 7 (left) to section 8.

118

Figure A.2.13 View toward the south of section 13 (right) to section 14.

Figure A.2.14 View toward the west of section 14 (right) to section 3.

119

Figure A.2.15 Untraced photographs used in Figure 3.8.

120

Figure A.2.16 Untraced photograph used in Figure 3.9.

121

Figure A.2.17 Untraced photograph used in Figure 2.3

122

Figure A.2.18 Ridge A.

Figure A.2.19 Ridge A.

123

Figure A.2.20 Ridge B.

124

Figure A.2.21 Ridge B.

125

Figure A.2.22 Ridge B.

Figure A.2.23 Ridge C (level with river) is draped by deltaic deposits from parasequence 9.

126

Figure A.2.24 Deltaic deposits of parasequence 5.

Figure A.2.25 Deltaic deposits of parasequence 8 dip (outlined) to the southwest (left).

127

Figure A.3.7 Deltaic deposits of parasequence 9.

Figure A.3.8 Deltaic deposits of parasequence 10.

128